<%BANNER%>

Diversity of Structural Types and Molecular Nanomagnetism in Iron and Manganese Clusters

Permanent Link: http://ufdc.ufl.edu/UFE0021981/00001

Material Information

Title: Diversity of Structural Types and Molecular Nanomagnetism in Iron and Manganese Clusters
Physical Description: 1 online resource (284 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: hysteresis, iron, lanthanide, magnet, magnetism, manganese, molecule
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The primary reason for the current interest in high nuclearity manganese and iron-oxo clusters is because of their relevance in molecular magnetism and bio-inorganic chemistry. A particularly appealing area in molecular magnetism is that of molecules which show slow relaxation of magnetization at low temperatures, behaving as tiny magnets, and thus known as single molecule magnets (SMMs). One of the first SMM to be synthesized was Mn12O12(O2CMe)16(H2O)4 (Mn12), which now serves as the 'drosophila' of molecular magnetism. The various modifications of the Mn12 family of SMMs have permitted advances in our knowledge and understanding of Mn12 complexes and the SMM phenomenon in general. For the first time, Mn12 family of SMMs has been extended to a fourth isolated member by the successful isolation and characterization of Mn123- complexes with S = 17/2 ground-state spin and D = - 0.24(1) cm-1. When studied by ac susceptibility techniques, the Mn123- complexes exhibit frequency-dependent out-of-phase signals indicating them to be SMMs, albeit with smaller barriers than the other Mn12 oxidation levels. The Mn123- complexes represent a fourth isolated oxidation level of the Mn12 family of SMMs, by far the largest range of oxidation levels yet encountered within single-molecule magnetism. Towards the synthesis of polynuclear molecular clusters, various alcohol-based ligands have been explored. Among them is a family of ligands incorporating one, two and four hydroxyethyl arms on ethylenediamine backbone. Use of dmemH (2-{2-(dimethylamino)ethyl-methylamino}ethanol) has led to two new Fe7 clusters and one Fe6 cluster, depending on the identity of the carboxylate employed. Unlike dmemH, use of hmbpH (6-hydroxymethyl-2,2'-bipyridine), one that amalgamates the chelating property of bipyridine and hmpH (hydroxymethyl pyridine), resulted in Fe6 cluster irrespective of the carboxylate employed. These contrasting results from flexible dmemH and rigid hmbpH underline the exquisite sensitivity of the reaction product on a variety of reaction conditions and reagents used. The magnetochemical characterization of these clusters emphasize how ground state spin values of significant magnitude can result from spin-frustration effects even though all the pair wise exchange interactions are antiferromagnetic. The use of heenH2 (N,N'-bis(2-hydroxyethyl)ethylenediamine) has provided an entry into new cluster types, including a discrete Fe18 molecular chain with an unusual double-headed serpentine structure and a Fe9 SMM, both having unprecedented structures in Fe chemistry. Fe18 represents the highest nuclearity, chain-like metal-containing molecule to be yet discovered, and Fe9 SMM contains a mixture of ON and OFF dimers with respect to the quantum-mechanical coupling through the hydrogen-bond. The initial use of edteH4 (N,N,N',N'-tetrakis(2-hydroxyethyl)ethylenediamine) in Mn and Fe chemistry has resulted in novel complexes of high nuclearity and architectural beauty ranging from Mn8 to Mn20 and Fe5 to Fe12. The complexes all possess rare or novel core topologies. The combined results demonstrate the ligating flexibility of alkoxide containing chelates and their usefulness in the synthesis of a variety of Fex and Mnx molecular clusters. A family of tetranuclear Mn clusters has been synthesized using Schiff-base ligand, salproH3 (1,3-bis(salicylideneamino)-2-propanol). The structure of these is much more closed than the previously reported butterfly-like complexes as a result of the alkoxide oxygen of salpro bridging the two wingtip Mn atoms. Fitting of the dc magnetic susceptibility data revealed that the various exchange parameters are all antiferromagnetic, and the core thus experiences spin frustration effects. Use of hmpH and pdmH2 (2,6-pyridine dimethanol) has resulted in an aesthetically pleasing Mn25 with a ground-state spin of 65/2, which is the second highest in Mn chemistry. Achieving high spin ground state is one of the elusive goals in the search for obtaining superior SMMs. A family of isostructural heterometallic Mn-Ln clusters with a MnIII10LnIII2 core (Ln = Pr, Nd, Sm, Gd, Tb, Dy, Ho and Er) has been synthesized as well as the MnIII10YIII2 analog with diamagnetic YIII to assist the magnetic studies of the nature of Mn-Ln exchange interactions. Complexes containing Tb, Dy and Ho exhibit strong frequency-dependent out-of-phase ac susceptibility signals characteristic of SMMs which was confirmed for the Dy complex by the observation of magnetization hysteresis.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Christou, George.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021981:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021981/00001

Material Information

Title: Diversity of Structural Types and Molecular Nanomagnetism in Iron and Manganese Clusters
Physical Description: 1 online resource (284 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: hysteresis, iron, lanthanide, magnet, magnetism, manganese, molecule
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The primary reason for the current interest in high nuclearity manganese and iron-oxo clusters is because of their relevance in molecular magnetism and bio-inorganic chemistry. A particularly appealing area in molecular magnetism is that of molecules which show slow relaxation of magnetization at low temperatures, behaving as tiny magnets, and thus known as single molecule magnets (SMMs). One of the first SMM to be synthesized was Mn12O12(O2CMe)16(H2O)4 (Mn12), which now serves as the 'drosophila' of molecular magnetism. The various modifications of the Mn12 family of SMMs have permitted advances in our knowledge and understanding of Mn12 complexes and the SMM phenomenon in general. For the first time, Mn12 family of SMMs has been extended to a fourth isolated member by the successful isolation and characterization of Mn123- complexes with S = 17/2 ground-state spin and D = - 0.24(1) cm-1. When studied by ac susceptibility techniques, the Mn123- complexes exhibit frequency-dependent out-of-phase signals indicating them to be SMMs, albeit with smaller barriers than the other Mn12 oxidation levels. The Mn123- complexes represent a fourth isolated oxidation level of the Mn12 family of SMMs, by far the largest range of oxidation levels yet encountered within single-molecule magnetism. Towards the synthesis of polynuclear molecular clusters, various alcohol-based ligands have been explored. Among them is a family of ligands incorporating one, two and four hydroxyethyl arms on ethylenediamine backbone. Use of dmemH (2-{2-(dimethylamino)ethyl-methylamino}ethanol) has led to two new Fe7 clusters and one Fe6 cluster, depending on the identity of the carboxylate employed. Unlike dmemH, use of hmbpH (6-hydroxymethyl-2,2'-bipyridine), one that amalgamates the chelating property of bipyridine and hmpH (hydroxymethyl pyridine), resulted in Fe6 cluster irrespective of the carboxylate employed. These contrasting results from flexible dmemH and rigid hmbpH underline the exquisite sensitivity of the reaction product on a variety of reaction conditions and reagents used. The magnetochemical characterization of these clusters emphasize how ground state spin values of significant magnitude can result from spin-frustration effects even though all the pair wise exchange interactions are antiferromagnetic. The use of heenH2 (N,N'-bis(2-hydroxyethyl)ethylenediamine) has provided an entry into new cluster types, including a discrete Fe18 molecular chain with an unusual double-headed serpentine structure and a Fe9 SMM, both having unprecedented structures in Fe chemistry. Fe18 represents the highest nuclearity, chain-like metal-containing molecule to be yet discovered, and Fe9 SMM contains a mixture of ON and OFF dimers with respect to the quantum-mechanical coupling through the hydrogen-bond. The initial use of edteH4 (N,N,N',N'-tetrakis(2-hydroxyethyl)ethylenediamine) in Mn and Fe chemistry has resulted in novel complexes of high nuclearity and architectural beauty ranging from Mn8 to Mn20 and Fe5 to Fe12. The complexes all possess rare or novel core topologies. The combined results demonstrate the ligating flexibility of alkoxide containing chelates and their usefulness in the synthesis of a variety of Fex and Mnx molecular clusters. A family of tetranuclear Mn clusters has been synthesized using Schiff-base ligand, salproH3 (1,3-bis(salicylideneamino)-2-propanol). The structure of these is much more closed than the previously reported butterfly-like complexes as a result of the alkoxide oxygen of salpro bridging the two wingtip Mn atoms. Fitting of the dc magnetic susceptibility data revealed that the various exchange parameters are all antiferromagnetic, and the core thus experiences spin frustration effects. Use of hmpH and pdmH2 (2,6-pyridine dimethanol) has resulted in an aesthetically pleasing Mn25 with a ground-state spin of 65/2, which is the second highest in Mn chemistry. Achieving high spin ground state is one of the elusive goals in the search for obtaining superior SMMs. A family of isostructural heterometallic Mn-Ln clusters with a MnIII10LnIII2 core (Ln = Pr, Nd, Sm, Gd, Tb, Dy, Ho and Er) has been synthesized as well as the MnIII10YIII2 analog with diamagnetic YIII to assist the magnetic studies of the nature of Mn-Ln exchange interactions. Complexes containing Tb, Dy and Ho exhibit strong frequency-dependent out-of-phase ac susceptibility signals characteristic of SMMs which was confirmed for the Dy complex by the observation of magnetization hysteresis.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Christou, George.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021981:00001


This item has the following downloads:


Full Text





DIVERSITY OF STRUCTURAL TYPES AND MOLECULAR NANOMAGNETISM IN
IRON AND MANGANESE CLUSTERS




















By

RASHMI BAGAI


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2008
































O 2008 Rashmi Bagai


































To mummi, papa, and Shammi, for their infinite unconditional love









ACKNOWLEDGMENTS

Undertaking and completing this dissertation has been an extraordinary journey for me. It

is my pleasure to thank the many people who have encouraged and supported me and made the

completion of this work possible.

First and foremost, I would like to thank my mentor and research advisor, Prof. George

Christou, for his guidance and help over the last 5 years. His constant encouragement, insightful

advice and confidence in my abilities as a chemist have been invaluable in attaining my goal. His

impact on my scientific writing skills is enormous. I deeply appreciate the freedom that I enjoyed

while working under him, nicely balanced by much-needed guidance and support. Most

importantly I would like to thank him for always having the time to sit and talk about research.

He was always open to my ideas, yet quick to point out weaknesses and incompleteness in my

arguments. Without his support, this dissertation would have been impossible. I walk away from

his lab infinitely more knowledgeable about crystal growth, magnetochemistry, and a lifelong

experience. I wear the title of "Christou group Alumnus" with pride.

I would also like to thank the other members of my committee, Prof. Stephen Hill, Prof.

Lisa McElwee-White, Prof. Michael J. Scott and Prof. Daniel R. Talham for their insightful

discussions and comments. I would like to express my gratitude to Dr. Khalil A. Abboud and his

staff at the UFCXC for not only solving my crystal structures but most importantly for giving me

the rough structures asap. I would also like to acknowledge Dr. Wolfgang Wernsdorfer for

providing essential single crystal measurements on Fe9 and MnloDy2 COmpounds below 1.8 K

using his micro-SQUID apparatus and answering all my questions to help bring these results

together. Many thanks to Prof. Stephen Hill, and Saiti Datta at the UF Physics Department for

the HFEPR measurements. Their patience as they taught me the fundamentals, from









understanding a HFEPR experiment to interpreting the terms in a spin Hamiltonian of a SMM,

was exceptional.

I would like to thank the entire Christou group, past and present, for their friendship,

laughs and all the positive and negative interactions which have helped me in some way or the

other. I appreciate the friendship and help of all my seniors Jon, Nicole, Abhu, Dolos and Alina

who helped me get started in the lab. Special thanks go to Nicole for her sincere help and for

always cheering me up. My juniors Chris, Antonio, Taketo, Arpita, Shreya, Jennifer also deserve

special mention for the all fun talking. I would also like to acknowledge our postdoc Tasos (Dr.

A. J. Tasiopoulos) for helping me in my first proj ect and Muralee (Dr. M. Murugesu) for

introducing me to madras curry paste. Watching Harris (Dr. T. C. Stamatatos) getting new

clusters everyday has been very inspiring. During the last five years I had the pleasure of

working with two great undergraduates Sarah Nam and Matthew Daniels. Sincere thanks go

for the efficient secretaries of Prof. Christou, Sondra and Melinda. Without them, the little things

that make everything run smoothly would never get done. A special thanks goes out to Soma,

Rumi, Parul, Arpita, Shreya and Ozge for their warm friendship.

Finally, I must thank my family, which has been an integral part of my existence. I am

forever indebted to my parents and my brother Shammi for their constant pride, encouragement,

never-ending love, and unwavering confidence in me and for always being there for me. Papa

and mummi's philosophy of "work is worship" and "time is money" have helped me become

who I am today and where I am today. Thanks shammi for being my greatest critique and best

brother in the world. I would like to thank my in-laws for accepting me like a daughter and for

all their warmth, love and affection. "Thanks" just seems too insignificant of a word to express









my gratefulness. I feel that my gratitude can not be expressed in words alone. I am grateful for

having such a wonderful, loving family and cannot wait to go back.

Last but not least, I would like to thank one very special person, my husband Ranjan. The

understanding ear with which he listens to me day and night is commendable. His patience and

perfectionism in chemistry and cooking has helped me grow as a person. His love and support

carried me through the roughest times. Without him, life would be meaningless. Thanks Ranj an

for helping me finish this journey so that we may begin another. I thank you for the daily

experience of helping, sharing, complaining, j oking and all the things that make life what it is.

The chain of my gratitude would be incomplete if I forget to thank the first cause of this

chain, 'Thou Formless One'. My deepest and sincere gratitude for inspiring and guiding this

humble being.












TABLE OF CONTENTS


page

ACKNOWLEDGMENTS .............. ...............4.....


LIST OF TABLES ............_...... ._ ...............11...


LIST OF FIGURES .............. ...............13....


LI ST OF AB BREVIAT IONS ............ ..... ..__ ............... 18..


AB S TRAC T ............._. .......... ..............._ 19...


CHAPTER


1 GENERAL INTRODUCTION .............. ...............22....


2 LIGAND-INDUCED DISTORTION OF A TETRANUCLEAR MANGANESE
BUTTERFLY COMPLEX ................. ...............34.......... ......


2.1 Introduction ................. ...............34.................
2.2 Experimental Section .............. ...............35....
2.2.1 Syntheses............... .. ..............3
2.2.2 X-ray Crystallography............... ............3
2.3 Results and Discussion............... ...............3
2.3.1 Syntheses............... ... ..............3
2.3.2 Description of Structures ......................... .... ..... ....................4
2.3.2.1 Structure of [Mn402(O2CR)sasalpro) (R Me (2-1), But (2-3)) .......40
2.3.2.2 Structure of NMe4 [Mn(O2 CPh)2(S alproH)] (2- 4) .............. ..... ........._.42
2.3.3 Magnetochemi stry ................. ...............43........... ....
2.4 Conclusions ................. ...............47.................


3 ROLE OF MIXED-LIGAND AND MIXED-SOLVENT SYSTEM: ROUTE TO Mn4
AND M n25 .............. ...............54....

3.1 Introduction ....._.................. ...............54.......
3.2 Experimental Section .............. ...............55....
3.2.1 Syntheses............... .. ..............5
3.2.2 X-ray Crystallography............... ............5
3.3 Results and Discussion............... ...............5
3.3.1 Syntheses............... ... ..............5
3.3.2 Description of Structures ................................................ ... .......5
3.3.2.1 Structure of [Mn4(hmp)4(pdmH)2(MeCN)4](CIO4)4 (3-1) ..................59
3.3.2.2 Structure of [Mn2501s(OH)2(hmp)6(pdm)s(pdmH) 2(L)2](C IO1
(3-2) .............. ...............60....
3.3.3 Magnetochemi stry ................. ...............61........... ....
3.3.3.1 Dc Studies............... ...............61











3.3.3.2 Ac Studies .............. ...............63....
3.4 Conclusions ................. ...............64.................


4 DIVERSITY OF STRUCTURAL TYPES IN POLYNUCLEAR IRON CHEMISTRY
WITH A (N,N,O)-TRIDENTATE LIGAND ................... ...............7

4.1 Introduction ................. ...............70.................
4.2 Experimental Section .............. ...............71....
4.2.1 Syntheses............... .. ..............7
4.2.2 X-ray Crystallography............... ............7
4.3 Results and Discussion............... ...............7
4.3.1 Syntheses............... ... ..............7
4.3.2 Description of Structures .................... ... .. .... ...... .........7
4.3.2.1 Structure of [Fe7O4(O2CPh)ll(dmem)2] (4-1)............... .................7
4.3.2.2 Structure of [FeO4(O2CMe) 11 (dmem)2] (4-2) ................. ................78
4.3.2.3 Structure of [Fe602(OH)4(O2C~ur~s~dmem)2] (4-3) ...........................79
4.3.2.4 Structure of [Fe30(O2CBut)2(N3)3(dmem)2] (4-4)..................... ........._79
4.3.3 Magnetochemistry of Complexes 4-1 to 4-4 ........................... ...............80
4.3.3.1 D c Studies.................. ...... .. .. ...................8
4.3.3.2 Rationalization of the Ground State Spin of 4-1 and 4-3 .................. .82
4.3.3.3 Determination of the Exchange Interactions in 4-4 ................... .........83
4.3.4 High-Frequency EPR Spectroscopy ................. ...............85................
4.4 Conclusions ................. ...............89.................


5 A NEW N, N, O CHELATE FOR TRANSITION METAL CLUSTER CHEMISTRY:
Fes AND Fe6 CLUSTERS FROM THE USE OF 6-HYDROXYMETHYL-2, 2'
BIPYRIDINE ............ ..... ._ ...............99....


5.1 Introduction ............ ..... ._ ...............99....
5.2 Experimental Section ............ ..... .__ ...............100..
5.2.1 Syntheses............... .. ..............10
5.2.2 X-ray Crystallography............... ............10
5.3 Re sults and Di scussi on ................. ...............104........... ..
5.3.1 Syntheses............... ... .............10
5.3.2 Description of Structures .......................... .................. ... ............0
5.3.2.1 Structure of [FesO2(OH)(O2CMe)s(hmbp)3](CIO4)2 (5-1) ...............105
5.3.2.2 Structure of [Fe602(OH)2(O2CPh)6(hmbp)4] (NO3)2 (5-2) ................106
5.3.3 agnetochemistry of Complexes 5-1 to 5-4 .............. .....................107
5.3.3.1 D c Studies................. ... ..... .................10
5.3.3.2 Rationalization of the Ground State Spin ................. ................ ..109
5.4 Conclusions ................. ...............111................

6 NEW STRUCTURAL TYPES IN POLYNUCLEAR IRON CLUSTERS
INCORPORATING O,N,N,O LIGAND: A SNAKE LIKE CHAIN AND A
SUPRAMOLECULAR DIMER OF SMMs............... ...............1...17


6.1 Introduction ................. ...............117................











6.2 Experimental Section ................. ...............119._.._. .....
6.2.1 Syntheses............... .. ..............11
6.2.2. X-ray Crystallography............... ............12
6.3 Re sults and Di scussi on ................. ...............123......... ...
6.3.1 Syntheses............... ... .............12
6.3.2 Description of Structures .................................... .. .............12
6.3.2.1 Structure of [FelsOs(OH)2(O2CBut)28(heen)4] (6-1) ................... ......124
6.3.2.2 Structure of [Fe904(OH)4(O2CPh)13(heenH)2] (6-2) ........................125
6.3.2.3 Structure of
[Fe7O3(OMe)3(MeOH)1(heen)3 14.5(H20)1.sCl.5 F1l (6-3) ....126
6.3.2.4 Structure of [Fe602(O2CPh)s(heen)3(heenH)](CIO4)2 (6-4).............. 127
6.3.3 Magnetochemistry of complexes 6-1 to 6-4............... ...................128
6.3.3.1 Dc Studies............... ...............128
6.3.3.2 Ac Studies............... .. ........ .........3
6.3.3.3 Single-Crystal Hysteresis Studies............... ...............131
6.4 Conclusions ........._.___..... .___ ...............132....

7 UNUSUAL STRUCTURAL TYPES IN Mn AND Fe CHEMISTRY FROM THE USE
OF N,N,N',N' TETRAKIS (2-HYDROXYETHYL)ETHYLENEDIAMINE ........._._.........142

7.1 Introduction ........._.___..... .___ ...............142....
7.2 Experimental Section ........._.___..... .__. ...............143...
7.2.1 Syntheses............... .. ..............14
7.2.2 X-ray Crystallography............... ............14
7.3 Re sults and Di scussi on ................. ...............149........... ..
7.3.1 Syntheses............... ... .............14
7.3.2 Description of Structures ............... ............................................15
7.3.2.1 Structure of [MnsO3(OH)(OMe)(O2CPh)((de)(edteedteH2)(O2CPh)
(7-1) .................................................. ............15
7.3.2.2 Structure of [Mnl204(OH)2(edte)4C 6(H20)2] (7-2) ................... .......155
7.3.2.3 Structure of [Mn200s(OH)4(O2CMe)6(edte)6](CIO4)2 (7-3).............. 157
7.3.2.4 Structure of [F e 5O2(O2CPh)7(edte)(H2 0)] (7-4)..........._.._.. ..............15 8
7.3.2.5 Structure of [F e602(O2 CBu? s(e dteH)2]i (7-5) ............... ... .................15 9
7.3.2.6 Structure of [Fel1204(OH)2(O2CMe)6(edte)4(H20)2] (CIO4)4 (7-6) ....160
7.3.3 Magnetochemi stry .................. ............ ...............161 .....
7.3.3.1 Dc Studies on 7-1 to 7-3 ................. ...............161........... ..
7.3.3.2 Dc Studies on 7-4 to 7-8 ................. ...............163........... ..
7.3.3.3 Ac Studies on 7-1 to 7-5 .................. .. .. ........ .. ...... .......... .....16
7.3.3.4 Rationalization of the Ground State Spin of 7-4 and 7-5 ................. 167
7.4 Conclusions ................. ...............168................

8 SINGLE-MOLECULE MAGNETISM AND MAGNETOSTRUCTURAL
CORRELATION WITHIN A FAMILY OF [Mn" 10Ln 2z] COMPLEXES ........................184

8.1 Introduction ................. ...............184................
8.2 Experimental Section ................ ...............186................
8.2.1 Syntheses............... ...............18












8.2.2 X-ray Crystallography............... ............18
8.3 Re sults and Di scussi on ................. ...............190........... ..
8.3.1 Syntheses............... ... .............19
8.3.2 Description of Structures .............. ...............191....
8.3.3 Magnetochemi stry ................. ...... .. ........ ..... ........ ...... .........19
8.3.3.1 Complexes 8-9 (MnloY2) and 8-4 (MnloGd2)................................19
8.3.3.2 Comparison of 8-9 (MnloY2) with 8-4 (MnloGd2), 8-5 (MnloTb2),
8-6 (MnloDy2), 8-7 (MnloHo2), and 8-8 (MnloEr2) ..........................195
8.3.3.3 Comparison of 8-9 (MnloY2) with 8-1 (MnloPr2), 8-2 (MnloNd2),
8-3 (M nloSm 2) ........................ .... .... .. .. .. .. .. .......19
8.3.3.4 Out-of-Phase ac Susceptibility Signals and Magnetization
Hysteresis Loops ................. .......__. ..........199.........
8.4 Conclusions ................. ...............202...............


9 A FOURTH ISOLATED OXIDATION LEVEL OF THE [Mnl2012(O2CR)16(H20)4]
FAMILY OF SINGLE MOLECULE MAGNETS............... ...............212


9.1 Introduction ................. ...............212................
9.2 Experimental Section .............. ...............214....
9.2.1 Syntheses............... ...............21
9.3 Results and Discussion............... ..............21
9.3.1 Syntheses............... ...............21
9.3.2 Magnetochemi stry ................. ...............218................
9.3.2.1 D c Studies................. .... ..... .. .. .. ... .. ........ 1
9.3.2.2 Comparison of the Magnetic Properties of the [Mnl2 z- (z = 0 3)
Fam ily ............ ..... .._ ...............219...
9.3.2.3 Ac Studies............... ...............222
9.4 Conclusions ............ ..... .._ ...............224...


APPENDIX


A BOND DISTANCES AND ANGLES................ ...............233


B LIST OF COMPOUNDS............... ...............25


C PHYSICAL MEASUREMENT S .............. ...............258....


D VAN VLECK EQUATIONS .............. ...............259....

LIST OF REFERENCES ............ ..... ._ ...............266...


BIOGRAPHICAL SKETCH .............. ...............284....












10










LIST OF TABLES


Table page

2-1 Crystallographic Data for 2-1-MeCN, 2-3-MeOH-2CHzC2 27H16 and 2-4-CH2 12 .........47

2-2 Bond-valence sums for the Mn atoms of complexes 2-1, 2-3 and 2-4 .............. ................48

2-3 Comparison of core parameters of selected [Mn402 8+ COmplexes (A+, o) ................... ...........48

2-4 Comparison of exchange parameters in [Mn402 8+ COmplexes .............. ....................48

2-5 Spin states of 2-1-CH21 2in the |ST, SA, SB> format ....._____ .... ... .__ ................49

3-1 Crystallographic Data for 3-1 and 3-2-8MeCN-4MeOH ................ ........................65

3-2 Bond-valence sums for the Mn atoms of complex 3-2 ....._____ ........._ ........._.....65

3-3 Bond-valence sums for the O atoms of complex 3-2 ....._____ .... ... ..__ ................. 65

4-1 Crystallographic data for 4-1-4MeCN, 4-2-MeCN, 4-3-2MeCN and 4-4-CH21 2.............91

5-1 Crystallographic Data for 5-1-5MeCN and 5-2-3MeCN-H20. ................ ................. .112

6-1 Crystallographic data for 6-1-4CSH12-4CH2 12, 6-2-9MeCN, 6-3-2MeOH-%H2H0 and
6-4-2EtOH -2H 20 ................. ...............134................

7-1 Crystallographic Data for 7-1-2CH2 2-MeOH, 7-2-6MeCN-%H2H0 and 7-3-10MeOH..169

7-2 Crystallographic Data for 7-4-CH2 12, 7-5-2CHCl3, 7-6-MeCN and 7-7 ................... .....170

7-3 Bond-valence sums for the Mn atoms of complex 7-1 and 7-3 ................. ................. .170

7-4 Bond-valence sums for the O atoms of complex 7-1 ................... ............... 171

7-5 Bond-valence sums for the Mn and O atoms of complex 7-2. ............. ......................171

7-6 Bond-valence sums for the O atoms of complex 7-5 and 7-6............... ...................172

7-7 Selected Fe-O distances and Fe-O-Fe angles for 7-5............... ...............172..

8-1 Crystallographic data for 8-4-3MeCN-MeOH, 8-6-3MeCN-MeOH and 8-9-4MeCN.....204

8-2 Bond-valence sums for the Mn atoms of complex 8-4, 8-6 and 8-9............... ................204

9-1 Magnetism Data for [Mnl2z-z (z = 0 3) Complexes 9-1 to 9-5............... ...................2

A-1 Selected interatomic distances (A+) and angles (o) for 2-1-MeCN................. ................3









A-2 Selected interatomic distances (A+) and angles (o) for 2-3-MeOH-2CH2 2' C7H16 ..........234

A-3 Selected interatomic distances (A+) and angles (o) for 2-4-CHzC2 2................ ................235

A-4 Selected interatomic distances (A+) and angles (o) for 3-1............... ..................23

A-5 Selected interatomic distances (A+) and angles (o) for 3-2-8MeCN-4MeOH....................237

A-6 Selected interatomic distances (A+) and angles (o) for 4-1-4MeCN............... ...............23

A-7 Selected interatomic distances (A+) and angles (o) for 4-2-MeCN............... .................3

A-8 Selected interatomic distances (A+) and angles (o) for 4-3-2MeCN............... ...............24

A-9 Selected interatomic distances (A+) and angles (o) for 4-4-2CHzC2 2............... ... .........._..241

A-10 Selected interatomic distances (A+) and angles (o) for 5-1-5MeCN................. ...............4

A-11 Selected interatomic distances (A+) and angles (o) for 5-2-3MeCN-H20 ................... .......243

A-12 Selected interatomic distances (A+) and angles (o) for 6-1-4CsH12-4CH2 12 ....................244

A-13 Selected interatomic distances (A+) and angles (o) for 6-2-9MeCN................ ...............24

A-14 Selected interatomic distances (A+) and angles (o) for 6-3-6MeOH-H20 .........................246

A-15 Selected interatomic distances (A+) and angles (o) for 6-4-2EtOH-1.5H20 ................... ...247

A-16 Selected interatomic distances (A+) and angles (o) for 7-1-2CH2 2-MeOH......................248

A-17 Selected interatomic distances (A+) and angles (o) for 7-2-6MeCN-V2H20 ....................249

A-18 Selected interatomic distances (A+) and angles (o) for 7-3-10MeOH ............... ..............250

A-19 Selected interatomic distances (A+) and angles (o) for 7-5-2CHCl3 ................ ...............251

A-20 Selected interatomic distances (A+) and angles (o) for 7-6-4MeCN................ ...............25

A-21 Selected interatomic distances (A+) and angles (o) for 8-4-3MeCN-MeOH......................253

A-22 Selected interatomic distances (A+) and angles (o) for 8-6-3MeCN-MeOH......................254

A-23 Selected interatomic distances (A+) and angles (o) for 8-9-4MeCN................ ...............25










LIST OF FIGURES


Figure page

1-1 Representations of magnetic dipole arrangements in paramagnetic, ferromagnetic,
antiferromagnetic, and ferrimagnetic materials ................ ...............31........... ...

1-2 Schematic diagram of a hysteresis curve for a typical ferromagnet ................ ...............3 1

1-3 Schematic representation of a multidomain ferromagnetic particle in the
unmagnetized state ................. ...............3.. 1..............

1-4 Representation of the [Mnl2012 16+ COre and [Mnl2012(O2CMe)16(H20)4] COmplex
with peripheral ligation. .............. ...............32....

1-5 Representative plots of the potential energy versus orientation of the ms vector along
the z axis and the ms sublevels for a Mnl2 COmplex. ............. ...............32.....

1-6 Magnetization hysteresis loops for a typical [Mnl2012(O2CR)16(H20)4] COmplex. ..........33

1-7 Representation of the change in energy of the ms sublevels as a function of the
applied magnetic field ................. ...............33........... ....

2-1 Structure of SalproH3 ................. ...............50...___ .....

2-2 Labeled representation of the structure of 2-1 and 2-3 ................ .......... ...............50

2-3 Comparison of the cores of 2-1 and 2-3 with that of the normal butterfly complexes......5 1

2-4 Labeled representation of the structure of 2-4 ................. ...............51..............

2-5 Plots oflu~Tys Tfor complexes 2-1-CH2 12, 2-2-CH2 12 and 2-3-M/CH2 2....................52

2-6 The core of 2-1 defining the pairwise exchange interactions and rationalization of
ground state spin of 2-1 ........... __..... ._ ...............53...

2-7 3s;Tyvs. T and reduced magnetization (M~NpB) VS H/ Tplots for 2-4-M2CHC2. 2..........._.....53

2-8 Two-dimensional contour plot of the fitting-error surface vs D and g for 2-4-
'/CH2C 2.................. .............5

3-1 Structure of ligands: 2-hydroxymethyl pyridine (hmpH), 2,6-pyridine dimethanol
(pdmH2), 6-hydroxymethyl 2-pyridine carboxylic acid (L). ............. .....................6

3-2 Labeled representation of the structure of 3-1 ................ .......... ............. ......6

3-3 Structure of the cation of 3-2. ................ ...............66.......... ...

3-4 Centrosymmetric core of 3-2 and its three types of constituent layers ................... ...........67










3-5 Plots of XnnTys T for complex 3-1 ................ ............... ......... ........ ...._..67

3-6 The core of 3-1 defining the pairwise exchange interactions. ................ .....................68

3-7 Plot of reduced magnetization (M~~B IB) VS T and two-dimensional contour plot of
the fitting error surface vs D and g for complex 3-1............... ...............68..

3-8 Plot of in-phase (gs,'T) and out-of-phase (gs,") ac susceptibility data for complex 3-1.....69

3-9 3s;Tyvs T and reduced magnetization (M I~B) VS H/ Tplots for complex 3-2-3H20. .........69

3-10 Plot of3~u'Tys T (in-phase) ac susceptibility data for 3-2-3H20. .................. ...............69

4-1 Structure of ligands: mdaH2 and dmemH. ....__ ....._.___ ........__. ................91

4-2 Labeled representation of the structure of 4-1 and 4-2 ................ .......... ...............92

4-3 Comparison of cores of 4-1, 4-2 and 4-3 .............. ...............92....

4-4 Labeled representation of the structures of 4-3 and 4-4. ................ ........................93

4-5 Plots of XnTys T for complexes 4-1, 4-2, 4-3 and 4-4 ......____ ........._ ..............93

4-6 Plot of reduced magnetization (M I~B) VS HIT for complex 4-1- MMeCN. ................... ....94

4-7 Two-dimensional contour plot of the fitting error surface vs D and g for 4-1*
M2M eCN ................. ...............94.................

4-8 Plot of reduced magnetization (M~~B IB) VS T and two-dimensional contour plot of
the fitting error surface vs D and g for 4-2-2MeCN .............. ...............95....

4-9 Plot of reduced magnetization (M~~B IB) VS T and two-dimensional contour plot of
the fitting error surface vs D and g for 4-3. ............. ...............95.....

4-10 Rationalization of the ground state spin of 4-1 and 4-3. ....._____ ........._ ..............96

4-11 Plot of reduced magnetization (M/NuB IB) VST and two-dimensional contour plot of
the fitting error surface vs D and g for 4-4-M2CH2C 2. ............. ...............96.....

4-12 Core of 4-4 defining the pairwise exchange interactions and the rationalization of its
ground state spin .............. ...............97....

4-13 HFEPR peak positions for 4-3-2MeCN from angle-dependent studies and frequency
dependence for 4-3-2MeCN ................. ...............97........... ....

4-14 Simulated Zeeman diagram for a spin S = 5 and 5/2 system with D < 0 and D > 0..........98

4-15 Temperature dependent spectra and easy-plane peak positions for 4-1-4MeCN
plotted versus frequency .............. ...............98....










5-1 Structure of ligands: dmemH and hmbpH. ......_._.__ ......___. ......_ ...........1

5-2 Synthetic scheme for hmbpH ........._...... ...............112.._........

5-3 Labeled representation of the structure of 5-1 ................ ...............113.............

5-4 Labeled representation of the structure of 5-2 ................. ...............113.............

5-5 Plots of XnTys T for complexes 5-1, 5-2-H20, 5-3-H20 and 5-4-H20..............._._. ........ 114

5-6 Plot of reduced magnetization (M/NpB)) VS H/T and two-dimensional contour plot of
the fitting error surface vs D and g for 5-1 ....._ ......___ ........__ ........14

5-7 Plot of reduced magnetization (M/NpB)) VS H/T and two-dimensional contour plot of
the fitting error surface vs D and g for 5-2*H20, 5-3-H20 and 5-4-H20. .......................115

5-8 Rationalization of spin ground state of complex 5-1 and 5-2............. ............__ ...1 16

6-1 Structure of chelates: dmemH, heenH2 ...._.._.._ .... ... ..._. ....._.._.........13

6-2 Labeled representation of the structure of 6-1 ................ ...............135.............

6-3 Labeled representation of the structure of 6-2 ................. ...............135.............

6-4 [Fe9 2 dimer showing intermolecular and intramolecular hydrogen-bonding, and the
resultant ON and OFF state with respect to the coupling of the two molecules. .............136

6-5 Labeled representation of the structure of 6-3. ................ ...............137.............

6-6 Labeled representation of the cation of 6-4 ................. ...............137......._._..

6-7 Plot of ynTys Tfor complexes 6-1 to 6-4. ................ ....__ ...._._ ...............138

6-8 Spin alignments at the Fe atoms of 6-1 rationalizing its ground state spin .....................138

6-9 Plot of reduced magnetization (M~NpB) VS H/ Tand two-dimensional contour plot of
the r.m.s. error vs D and g for the fit for 6-2............... ...............139..

6-10 Plot of reduced magnetization (M~NpB) VS H/ Tfor 6-3 and 6-4..........._._... ...............1 39

6-11 Two-dimensional contour plot of the r.m.s. error vs D and g for the fit for complexes
6-3 and 6-4 ................ ...............140....._ _.....

6-12 Plot of the in-phase (p~'T) and out-of-phase (p")1 ac susceptibility data for 6-2............140

6-13 Plot of in-phase ac susceptibility data for 6-3 and 6-4 ...._._ ................ ................141

6-14 Single-crystal magnetization (M)1 vs dc field (H) hysteresis loops for 6-2. .....................141











7-1 Structure of ligands. ................. ......._.__........._... ...._._. ...._.__.......172

7-2 Labeled representation of the cation of 7-1 ................. ...............173......._._..

7-3 Crystallographically established coordination modes of edte4- and edteH22- found in
complexes 7-1 to 7-3............... ...............173..

7-4 Labeled representation of the structure of 7-2 ................. ...............174.............

7-5 The structure of the cation of 7-3 ................. ...............175.............

7-6 Labeled representation of the core of 7-3. ................ .................. .................176

7-7 Labeled representation of the structure of 7-4 ................. ...............177.............

7-8 Crystallographically established coordination modes of edte4- and edteH3- found in
complexes 7-4 to 7-6............... ...............177..

7-9 Labeled representation of the structure of 7-5 ................. ...............177.............

7-10 Labeled representation of the cation of 7-6 ................. ...............178.............

7-11 Labeled representation of the cation of 7-7 ................. ...............179.............

7-12 Plots of XnTys T for complexes 7-1-2H20, 7-2 and 7-3-5H20............ .. ........._.._..179

7-13 Plots of reduced magnetization (M I~B) VS HIT for 7-1-2H20 and 7-2 ........._..............180

7-14 Two-dimensional contour plot of the r.m. s. error surface vs D and g for the
magnetization fit for 7-1-2H20 and 7-2 ....._.._._ .... ...... ....._.............8

7-15 Plots of reduced magnetization (M~~B IB) VS T and two-dimensional contour plot of
the r.m.s. error surface vs D and g for the magnetization fit for 7-3-5H20.................__181

7-16 Plots of XnTys T for complexes 7-4 to 7-8. ......____ ........__..... ..............18 1

7-17 Plot of reduced magnetization (M IBu) VS HIT for 7-4 and 7-5-2CHCl3-4H20 ........._....182

7-18 Two-dimensional contour plot of the fitting error surface vs D and g for complexes
7-4 and 7-5-2CHC l3-4H 20 .............. ...............182....

7-19 Plot of XM'T vs. T for complexes 7-1 7-5. ........._.__......__ ...._._ ...............1 83

7-20 Spin alignments at the Fe atoms of 7-5 rationalizing its S = 5 ground state ................... .183

8-1 Labeled structures of 8-4, 8-6, and 8-9. ............. ...............205....

8-2 Centrosymmetri c core of 8-6 emphasizing the ABCBA layer structure............._.._.. ......206











8-3 Plot of3&Tys Tand Xh,'Tys Tfor complexes 8-4 and 8-9 ................. ......................206

8-4 Plot of reduced magnetization (M~NpB) VS H/ Tfor 8-9 and 8-4 ............... ..............207

8-5 Two-dimensional contour plot of the error surface for the D vs g fit for 8-9 and 8-4.....207

8-6 Plots of dc XhT ys T for Gd(NO3)3 and Dy(NO3)3 ................. ...............208...........

8-7 Plots of3&Tys Tfor 8-9, 8-5 and 8-6 ................. ...............208........... .

8-8 Plots of3C&Tys Tfor 8-9, 8-7 and 8-8 ................. ...............208........... .

8-9 Plots of3&Tys Tfor 8-1, 8-2, 8-3 and 8-9 .............. ...............209....

8-10 Plots of out-of-phase 3C"l vs T ac susceptibility data for 8-9 and 8-4 ........._...................209

8-11 Plots of out-of-phase 3y"l vs T ac susceptibility data for 8-1 and 8-2 ........._...................209

8-12 Plots of in-phase, 3Cy'Tys T, and out-of-phase Xhllvs T ac susceptibility data for 8-5,
8-6 and 8-7 .............. ...............210....

8-13 Plot of relaxation rate vs reciprocal temperature for 8-5 8-7 ................. ........._..._... .210

8-14 Magnetization vs. time decay plots in zero dc field and z vs 1/T plot for 8-6. ..............21 1

8-15 Single-crystal magnetization vs dc field hysteresi s loops for 8-6-3MeCN-MeOH. .........21 1

9-1 Cyclic voltammogram and differential pulse voltammogram for 9-1. ............................226

9-2 Proposed structural core of 9-2 9-4 ........._._._ ....__. ...............227.

9-3 Plot of3&Tys T for 9-2 9-5. ............... ......... ........ ......... ................ 228

9-4 Plot of reduced magnetization (M/NpuB) VS H/T for 9-4 and 9-5 ............. ...............228

9-5 Plot of reduced magnetization (M/NpuB) VS H/T for 9-2 and 9-3 ............. ...............229

9-6 Two-dimensional contour plot of the error surface for the D vs g fit and D vs E fit for
com plex 9-4. ............. ...............229....

9-7 Plot of the in-phase (p'nT) and out-of-phase (p")' ac susceptibility data for 9-4............230

9-8 Plot of the in-phase (p~'T) and out-of-phase (p")1 ac susceptibility data for 9-5..........230

9-9 3Cn Ivs Tplots for vacuum-dried complexes [Mnl2 z- (z = 0-3) ................. ................ ..231

9-10 Comparison of the 3yllvs T plots for vacuum-dried complexes [Mnl2 z- (z =0-3)..........232









LIST OF ABBREVIATIONS


But tertiary butyl

BVS bond valence sun

CV cyclic voltammogram

dmemH: 2- {[2-(dimethyl amino)ethyl] -methyl amino) ethanol

DPV differential pulse voltammogram

edteH4 N, N, N',N'-tetraki s(2 -hy droxy ethyl)ethyl enedi ami ne

heenH2 N,N~r-b is(2 -hy droxy ethyl)ethyl enedi ami ne

HFEPR high Frequency electron paramagnetic resonance

hmbpH 6-hydroxymethyl-2,2 '-bipyridine

hmpH: 2-hydroxymethyl pyridine

L 6-hydroxymethyl-2-pyridine carboxylic acid

pdmH2: 2,6-pyridine dimethanol

PS II photosystem II

Py pyridine

salproHI3 1 ,3-bis(salicylideneamino)-2-propanol

TIP temperature independent paramagnetism

ZFS zero-field splitting









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

DIVERSITY OF STRUCTURAL TYPES AND MOLECULAR NANOMAGNETISM IN
IRON AND MANGANESE CLUSTERS


By

Rashmi Bagai

May 2008

Chair: George Christou
Major: Chemistry

The primary reason for the current interest in high nuclearity manganese and iron-oxo

clusters is because of their relevance in molecular magnetism and bio-inorganic chemistry. A

particularly appealing area in molecular magnetism is that of molecules which show slow

relaxation of magnetization at low temperatures, behaving as tiny magnets, and thus known as

single molecule magnets (SMMs). One of the first SMM to be synthesized was

[Mnl2012(O2CMe)16(H20)4] (Mnl2), which now serves as the "drosophila" of molecular

magnetism. The various modifications of the Mnl2 family of SMMs have permitted advances in

our knowledge and understanding of Mnl2 COmplexes and the SMM phenomenon in general. For

the first time, Mnl2 family of SMMs has been extended to a fourth isolated member by the

successful isolation and characterization of [Mnl2 3- COmplexes with S = 17/2 ground-state spin

and D = 0.24(1) cm- When studied by ac susceptibility techniques, the [Mnl2 3- COmplexes

exhibit frequency-dependent out-of-phase signals indicating them to be SMMs, albeit with

smaller barriers than the other Mnl2 Oxidation levels. The [Mnl2 3- COmplexes represent a fourth

isolated oxidation level of the Mnl2 family of SMMs, by far the largest range of oxidation levels

yet encountered within single-molecule magnetism.









Towards the synthesis of polynuclear molecular clusters, various alcohol-based ligands

have been explored. Among them is a family of ligands incorporating one, two and four

hydroxyethyl arms on ethylenediamine backbone. Use of dmemH (2- { [2-(dmethyrla~minnoethl]-

methylamino: ethanol) has ledl to +two new~ Fey clusters~ andl oneFe CnIUrltelr, dependi~ng on the


identity of the carboxylate employed. Unlike dmemH, use of hmbpH (6-hydroxymethyl-2,2'-

bipyridine), one that amalgamates the chelating property of bipyridine and hmpH

hydroxymethyll pyridine), resulted in Fe6 ClUSter irrespective of the carboxylate employed.

These contrasting results from flexible dmemH and rigid hmbpH underline the exquisite

sensitivity of the reaction product on a variety of reaction conditions and reagents used. The

magnetochemical characterization of these clusters emphasize how ground state spin values of

significant magnitude can result from spin-frustration effects even though all the pair wise

exchange interactions are antiferromagnetic.

The use of heenH2 (N,N~-bis(2-hydroxyethyl)ethylenediamine has provided an entry into

new cluster types, including a discrete Fels molecular chain with an unusual double-headed

serpentine structure and a Fe9 SMM, both having unprecedented structures in Fe chemistry. Fels

represents the highest nuclearity, chain-like metal-containing molecule to be yet discovered, and

Fe9 SMM contains a mixture of ON and OFF dimers with respect to the quantum-mechanical

coupling through the hydrogen-bond.

The initial use of edteH4 (N,N,N',N'-tetrakis(2-hydroxyethyl)ethylendaie in Mn and

Fe chemistry has resulted in novel complexes of high nuclearity and architectural beauty ranging

from Mns to Mn20 and Fes to Felt. The complexes all possess rare or novel core topologies. The

combined results demonstrate the lighting flexibility of alkoxide containing chelates and their

usefulness in the synthesis of a variety of Fex and Mnx molecular clusters.









A family of tetranuclear Mn clusters has been synthesized using Schiff-base ligand,

salproH3 (1 ,3-bis(salicylideneamino)-2-propanol). The structure of these is much more closed

than the previously reported butterfly-like complexes as a result of the alkoxide oxygen of salpro

bridging the two wingtip Mn atoms. Fitting of the dc magnetic susceptibility data revealed that

the various exchange parameters are all antiferromagnetic, and the core thus experiences spin

frustration effects.

Use of hmpH and pdmH2 (2,6-pyridine dimethanol) has resulted in an aesthetically

pleasing Mn25 with a ground-state spin of 65/2, which is the second highest in Mn chemistry.

Achieving high spin ground state is one of the elusive goals in the search for obtaining superior

SMMs. A family of isostructural heterometallic Mn-Ln clusters with a [Mn" 10Ln 2z] COre (Ln =

Pr, Nd, Sm, Gd, Tb, Dy, Ho and Er) has been synthesized as well as the [Mn 'loY 2z] analOg with

diamagnetic Y"' to assist the magnetic studies of the nature of Mn---Ln exchange interactions.

Complexes containing Tb, Dy and Ho exhibit strong frequency-dependent out-of-phase ac

susceptibility signals characteristic of SMMs which was confirmed for the Dy complex by the

observation of magnetization hysteresis.









CHAPTER 1
GENERAL INTRODUCTION

Magnetism has been known to humans for millennia. What is magnetism? This question

has fascinated people ever since Thales of Miletus (about 634-546 BC) first described the

phenomenon as the attraction of iron by "lodestone."l Over the last 2500 years, magnetism has

played an important role in the development of civilization, we have not only extensively used

the phenomenon for navigation, power production and "high tech" applications but we have also

come a long way in exploring its origin. Today we more specifically associate lodestone with the

spinel magnetite, Fe304, which is magnetically aligned in nature, most likely by the earth's

magnetic field during the cooling process of hot lava.

One of the most fundamental ideas in magnetism is the concept of a magnetic field. A

magnetic field is produced whenever there is an electrical charge in motion, specifically the spin

and orbital angular moment of electrons within atoms of a material. Thus the essential

component of any magnetic material is the presence of an unpaired electron and how they

interact with each other determines the magnetic behavior of all magnets.2 The different types of

magnetic materials are usually classified on the basis of their susceptibility (X). A magnetic

susceptibility is merely the quantitative measure of the response of a material to an applied

magnetic field. Substances were first classified as diamagnetic or paramagnetic by Michael

Faraday in 1845, but it was not until many years later that these phenomena came to be

understood in terms of electronic structure.3 Diamagnetism is an underlying property of all

matter and arises from the interaction of paired electrons with the magnetic field while

paramagnetism is a property exhibited by substances containing unpaired electrons. Diamagnets

are slightly repelled by a magnetic field and X for these is small negative (-10- ) and paramagnets

are attracted into an applied field and X for these is small positive (10-3 to 10-'). Diamagnetic









susceptibilities are independent of field and temperature while paramagnetic susceptibility varies

inversely with temperature, X = C/T, where C is Curie constant.4 Upon bringing the spins closer

together, spin-coupling enables a tendency toward parallel (ff ) or antiparallel (T1) alignment.

This behavior can be modeled as a function of temperature by the Curie-Weiss expression, X =

C/(T-6), where 6 is proportional to the strength of coupling between adj acent spins. Pairwise

ferromagnetic coupling (ff ) can lead to long-range ferromagnetic order, whereas

antiferromagnetic order may arise from pairwise antiferromagentic coupling (T1) as shown in

Figure 1-1. The ferromagnetic solids are the most widely recognized magnetic materials with X

ranging from 50 to 10,000. Examples of these materials are iron, cobalt, nickel and several rare

earth metals and their alloys.' Ferrimagnets (e.g. magnetite, Fe304) ariSe from antiferromagnetic

coupling, which does not lead to complete cancellation, and thus they have a net magnetic

moment (Figure 1-1). Ferro-, antiferro- and ferrimagnetic ordering occurs below a critical

temperature, Te. Below Tc, the magnetic moments for ferro- and ferrimagnets align in small

domains. In the absence an applied magnetic field, despite the nature of interactions, a net zero

magnetization is thermodynamically favored, as different domains have their net magnetizations

randomly oriented. A useful property of these materials is that when an external field is applied

to them, the magnetic fields of the individual domains tend to line up in the direction of the

external field. At some particular value of the field, no domains are present as all the spins are

aligned and net magnetization is at its saturation point. The system will remain in that situation

unless enough energy is given to overcome the energy barrier for domain formation. This

property can be monitored in a plot of magnetization vs applied field called hysteresis loop

(Figure 1-2).2 Because an additional field is required to reverse the direction of magnetization,

magnetic storage of information is possible in ferromagnetic and ferrimagnetic materials.









In addition to the ferri- and ferromagnetic behavior, other magnetic-ordering phenomena,

such as metamagnetism, canted ferromagnetism, and spin-glass behavior may occur.6 The

transformation from an antiferromagnetic state to a high moment state is called metamagnetism.

A canted antiferromagnet (or weak ferromagnet) results from the relative canting of

antiferromagnetically coupled spins that lead to a net moment. A spin glass occurs when local

spatial correlations with neighboring spins exist, but long-range order does not. The spin

alignment for a spin glass is that of paramagnet; however, unlike paramagnets, for which spin

directions vary with time, the spin orientations of a spin glass remain fixed or vary only very

slowly with time.'

As throughout history, today's magnetism research remains closely tied to applications. It

is therefore no surprise that some of the forefront research areas in magnetism today are driven

by the "smaller and faster" mantra of advanced technology. Future of the magnetic data storage

and memory technology is concerned with cramming information into smaller and smaller bits

and manipulating these bits faster and faster. Thus, the need to develop magnetic particles of

nanoscale dimensions is unavoidable. The synthesis of such nanomagnets can be accomplished

by fragmentation of bulk ferromagnets or ferrimagnets, for example, crystals of magnetite can be

broken down such that each fragment is smaller in size than a single domain (20-200 nm); these

subdomain nanoscale magnetic particles with varying sizes are called superparamagnets (Figure

1-3).8 The magnetic moments within one superparamagnetic particle are ferromagnetically

aligned due to short range order. These clusters are thermally unstable, i.e., their magnetic

moments (represented by moment vectors) experience thermal fluctuations with great ease, as is

the case with paramagnetic species due to lack of long-range ordering. In other words, even

though the temperature is below the Curie or Neel temperature and the thermal energy is not









sufficient to overcome the coupling forces between neighboring atoms, the thermal energy is

sufficient to change the direction of magnetization of the entire crystallite. The resulting

fluctuations in the direction of magnetization cause the magnetic moment to average to zero.

When cooled below a critical blocking temperature (TI3), superparamagnetic systems experience

a very slow relaxation time, their net magnetic moments align parallel to the applied field and

appear to behave as if they had an apparent "bulk-like" ferromagnetic behavior.8 This aspect will

result in hysteresis of "apparent" ferromagnetic behavior. But this approach unfortunately gives a

distribution of particle sizes, and this complicates detailed study of these systems, making

difficult, for example, an accurate assessment of variation of properties as a function of particle

size.g

One approach being investigated for new magnets is based on molecules as building blocks

also called bottom-up approach. Molecule-based magnets present several attributes unavailable

in conventional metal and metal-oxide magnets. These properties include low density,

mechanical flexibility, low temperature processability, high magnetic susceptibility,

biocompatibility and several other desirable characteristics.10 This strategy has the advantage that

a single particle size can be ensured, that organic ligands on the periphery can be chosen or

systematically altered to ensure crystallinity and solubility in a variety of solvents, and that X-ray

crystallographic and various spectroscopic and physicochemical studies can be readily performed

in solution and/or solid state.6,11

Worldwide interest in molecule-based magnets has arisen for both fundamental scientific

and technological reasons. Molecular magnetic materials have been added to the library of

magnetism only at the end of twentieth century. The first molecule-based magnet, reported in

1967 was that of [Fe '(S2CNEt2)2C1] mOlecules, which orders at 2.46 K.4 Subsequently, there









was very little published activity in this area until 1987 when Miller et al. characterized a

ferromagnetic transition in the organomettalic donor acceptor salt, [Fe(CsMes)2][TCNE]. The V/2

spins associated with both donor and acceptor units are strongly coupled along the chains in a

ferromagnetic fashion resulting in bulk ferromagnetic properties with a spontaneous

magnetization below To = 4.8 K.7 Since then, plethora of molecule-based magnets exhibiting a

wide variety of bonding and structural motifs have been synthesized. These include molecules

with extended bonding within chains (lD), within layers (2D), and within 3D network structures.

Molecule-based magnets include materials with spins only in organic moieties (in p orbitals), and

those with spins both on metal ions and organic moieties and those materials with spins on metal

ions with exchange pathways provided by organic moieties that do not contain spin.10

A big breakthrough in molecule-based systems was the discovery of zero-dimensional

(isolated molecules) nanoscale magnets now called Single Molecule Magmets (SMMs). One of

the first SMM to be synthesized was [Mnl2012(O2CMe)16(H20)4] (Mnl2),12 which by now serves

as the "drosophila" of molecular magnetism (Figure 1-4).13 The ease of preparation, stability,

high ground state spin, high magnetic anisotropy coupled with its highly crystalline nature and

high symmetry space group, which simplifies the spin Hamiltonian by precluding second-order

transverse (rhombic) terms, has made Mnl2 the favorite for detailed study within the chemistry

and physics communities by a myriad of techniques.14 Structurally, the family of Mnl2 mOleCUleS

contain an external crown of eight Mn ions (S = 2), which are ferromagnetically coupled and an

inner core of four Mn ions (S = 3/2), also ferromagnetically coupled. The crown and core are

antiferromagnetically coupled to produce a total spin of S = 10.1 Remarkably, a negative axial

zero-field splitting (ZFS), D, leads to a loss in the degeneracy of the associated ms levels, such

that ms = +10 and ms = -10 are lowest in energy (Figure 1-5). Thus there exists an energy barrier









for the conversion of "spin up" to "spin down". The slow relaxation of magnetization, which is at

the origin of the interesting behavior, is due to the presence of an energy barrier to be overcome

in the reversal of the magnetic moment. The energy barrier is proportional to S2| D | for integer

spin and (S2-1/4)| D | for half-integer spin system.13

A unique feature of Mnl2 is that slow relaxation of magnetization gives rise to hysteresis

cycle, similar to that observed in bulk magnets, but of molecular origin (Figure 1-6). The

magnet-like behavior of Mnl2 has sparked the idea that information might one day be stored as

the direction of magnetization in individual molecules.16 The most information that can be stored

on hard drives and other devices currently is 3 billion bits, or 3 gigabits, in 1 cm2 area of a

cobalt-based magnetic material. The much smaller size of SMMs means that one can get 30

trillion of them into 1 cm2, and thus a storage density of 30 trillion bits, or 30 terabits, is feasible.

This is 10,000 times greater than the current best by computer manufacturers." One of the

research challenges now is to find better SMMs that function at higher temperatures. The second

appealing feature of Mnl2 is that relaxation of its magnetization shows clear quantum effects,

which is manifested in the step-like hysteresis loop (Figure 1-6).xs The observed steps

correspond to an increase in the relaxation rate of magnetization that occurs when there is an

energy coincidence of ms sublevels on the opposite sides of the potential energy barrier. For

these critical field values, H = nD/gCLB, at which steps occur, quantum tunneling of magnetization

(QTM)19 is allowed, resulting in an increase in the relaxation rate of the molecule. Thus the

relaxation of the magnetization of an SMM occurs not just by thermal activation over the energy

barrier, but also by QTM through the energy barrier (Figure 1-7). A transverse component

contained in the Hamiltonian of the molecule must be present to promote tunneling through the

energy barrier; such transverse components can be provided in three ways 1) by low symmetry









components of the crystal field 2) by a magnetic field provided by magnetic nuclei 3) by a

magnetic field provided by neighboring molecules.19 Although tunneling provides a route for

rapid reversal of magnetization and, hence, a less attractive memory storage device but, these

features can be used to develop new classes of quantum computers in which quantum coherence

is used to store information. The SMM phenomenon is not unique to Mnl2, the family of SMM

compounds has been extended to various other metals including Fe, V, Co, Ni and combinations

of 3d with 4d, 5d, 4f paramagnetic ions, and homometallic Ln species.9,13,20-23 In Order to be

considered for real applications, the challenge that falls within the domain of inorganic chemistry

is to synthesize molecules with higher spin-reversal barriers, capable of storing information at

more practical temperatures.

In order to observe SMM behavior from a molecular compound, it is necessary to have a

high spin ground state, S, and a large negative ZFS of the ground state and negligible interactions

between molecules. The combination of these properties can lead to energy barrier so that, at low

temperatures, the system can be trapped in one of the high-spin energy wells. In principle, large

spin can be achieved using a small number of ions if the individual components have a large spin

and ZFS of clusters is determined by single ion anisotropy and spin-spin interactions which can

be both magnetic dipolar and exchange in nature.20 Two ions with S = 5/2 ferromagnetically

coupled can have a ground state with S = 5, while it is necessary to assemble a cluster of ten S =

V/2 ions in order to achieve the same result. This is the reason why ions with large spin such as

high spin Fe"' and Mn have been largely used for the preparation of SMMs.13,22

In addition to their relevance in magnetochemistry, Mn and Fe are two of the most

important elements in biochemistry.24-27 The ability to exist in different oxidations states makes

manganese well suited as the active site for redox reactions in a number of metalloproteins and









enzymes.28 The most important of which is the water-oxidizing complex of PS II. This reaction

is responsible for the generation of almost all the oxygen for this planet.29 The relevance of Fe in

geology and biochemistry is even larger than that of Mn. Iron is the fourth most abundant

element in the earth's crust. In addition to its availability, iron also possesses chemical properties

such as Lewis acidity and redox capability that allows it to perform diverse set of metabolic

functions. Iron storage protein ferritin can store upto 4500 iron atoms.30 In addition, ferritin is

also considered as a nanosize magnetic particle and has been investigated for quantum tunneling

effects of magnetization.31 The work presented in this thesis focuses on the synthesis and

characterization of new complexes of Mn and Fe, stimulated by the search for inorganic models

of important metalloproteins and by the remarkable magnetic properties that these complexes

display.

In order to design clusters, it is necessary to have available both connecting blocks, which

provide efficient bridges and determine the growth of the cluster, and terminal blocks, which

stop the growth of the cluster at a finite size. The bridging blocks must not only provide the right

connection between metal ions but also provide efficient exchange pathways thus assuring strong

magnetic coupling. In the design and synthesis of polynuclear molecular clusters, the choice of

appropriate ligand is the most important step. The ligand (from the latin word ligare, to bind) is

any molecular moiety that has at least one donor atom i.e. an atom with a non-binding electron

pair. A ligand is called mono-, bi-, tri- etc, dentate if it possesses one, two, three etc donor atoms.

Commonly employed are the polydentate ligands that possess more than one donor atom. If the

different donor atoms coordinate the same metal ion it is called a chelating ligand. An important

class of polydentate ligands is constituted by polyalcohols since this functionality is an excellent

bridging group and fosters higher nuclearity product formation on deprotonation.32 In this work









we decided to explore the possibility of creating new structural types using a variety of alcohol

based ligands:

1) a pentadentate Schiff-base ligand,

2) bi- and tridentate ligands containing one and two hydroxymethyl arms on pyridine backbone,

3) tri-, tetra- and hexadentate ligands incorporating one, two and four hydroxyethyl arms on

ethylenediamine backbone.

The layout of the thesis is as follows. Chapters 2 to 7 are organized according to the types

of ligands used. In Chapter 2, a Schiff-based ligand (salproH3) has been used to synthesize

tetranuclear Mn clusters and magnetic interactions between different spin centers have been

determined. Chapter 3 is devoted to the 'mixed-ligand approach' where synthesis of a high

nuclearity, high spin complex has been achieved using hmpH and pdmH2. In Chapter 4 and 5, a

flexible and a rigid N,N,O based ligand, dmemH and hmbpH respectively, is employed for the

synthesis of Fes, Fe6 and Fe7 clusters. Chapter 6 explores the potential of O,N,N,O based ligand,

heenH2, which has led to the largest molecular chain complex, Fels, and also provides insight

into the micro SQUID measurements of a supramolecular Fe9 dimer. Denticity of ligand is

further extended in Chapter 7, which employs O,O,N,N,O,O based ligand, edteH4, for the

preparation of novel Mnx and Fex clusters. Synthesis and magnetostructural correlation of a

family of heterometallic Mn-Ln SMMs is the theme of Chapter 8. The chemical reduction of

[Mnl2012(O2CCHCl2)16(H20)4] forming one-, two- and three- electron reduced complexes with

identical peripheral ligation is detailed in Chapter 9.



























H


Paramagnetic


Figure 1-1. Representations of magnetic dipole arrangements in 1) paramagnetic, 2)
ferromagnetic, 3) antiferromagnetic, and 4) ferrimagnetic materials.


Figure 1-2. Schematic diagram of a hysteresis curve for a typical ferromagnet showing
magnetization (M) as a function of the applied magnetic field (H). Saturation
magnetization is indicated by M,.


Figure 1-3. Schematic representation of a multidomain ferromagnetic particle in the
unmagnetized state. Each of the three domains with net moments remain randomized
in this state. Circles of varying sizes represent the subdomain superparamagnetic
clusters.


Femantic Atifrrmantic Ferimantic


















































Orientation of ms vector (0)


Figure 1-4. (left) Representation of the [Mnl2012 16+ COre. (right) [Mnl2012(O2CMe)16(H20)4]
complex with peripheral ligation. Color Code: Mn'V, green; Mn"', blue; O, red; C,
grey.


m. = 0


m. = 0


ms -


ms


-10t


Figure 1-5. Representative plots of the potential energy versus a) the orientation of the ms vector
(0) along the z axis and b) the ms sublevels for a Mnl2 COmplex with an S = 10 ground
state, experiencing zero-field splitting.






















-1 -0.5 0 0.5 1
tLHo(T)


Figure 1-6. Magnetization hysteresis loops for a typical [Mnl2012(O2CR)16(H20)4] COmplex in
the 1.3-3.6 K temperature range at 4 mT/s field sweep rate, M is normalized to its
saturation value, Ms.



f// H=0







r/ H#0












Figure 1-7. Representation of the change in energy of the ms sublevels as the magnetic field is
swept from zero to a non-zero value. Resonant magnetization tunneling occurs when
the ms sublevels are aligned between the two halves of the diagram.









CHAPTER 2
LIGAND-INDUCED DISTORTION OF A TETRANUCLEAR MANGANESE BUTTERFLY
COMPLEX

2.1 Introduction

Manganese cluster chemistry has been receiving a great deal of attention for two main

reasons: (i) the occurrence of this metal in a variety of manganese-containing biomolecules, the

most important of which is the water oxidizing complex (WOC) in the photosynthetic apparatus

of green plants and cyanobacteria. This contains an oxide-bridged Mn4 unit and is responsible for

essentially all the oxygen gas on this planet.33 This has stimulated the search for tetranuclear Mn

complexes with oxide bridges that can serve as models for the WOC.29 (ii) High nuclearity Mn

clusters often display large ground state spin (S) states as a result of ferromagnetic exchange

interactions and/or spin frustration effects.34,35 If Such molecules with large S values also possess

significant magnetoanisotropy of the Ising (easy-axis) type, then they have the potential to be

single-molecule magnets (SMMs).13 These are individual molecules that possess a significant

barrier (vs kT) to magnetization relaxation and thus exhibit the ability to function as magnets

below their blocking temperature (TB)

Our group has had a strong interest over many years in the development of synthesis

methodologies to oxide-bridged Mn clusters, primarily with carboxylate ligands. One synthetic

strategy that has proven particularly useful has been the use of the preformed clusters of general

formula [Mn30(O2CR)6 97)3 0,+ as starting materials in reactions with a variety of co-reagents.36-

38 A wide range of the latter have been employed, almost always bidentate or higher denticity

chelates, and often ones that also contain potentially bridging alkoxide groups. Such reactions

have often caused higher-nuclearity products to form, both homo- and mixed-valent.36-38 The

present work represents an extension of this approach. As part of our continuing search for new

preparative routes to high nuclearity Mn clusters, we have investigated the reactivity of the









pentadentate Schiff-base ligand 1 ,3-bis(salicylideneamino)-2-propanol (salproH3, Figure 2-1).

This group has been used previously by others in Mn chemistry and had afforded mononuclear,

dinuclear and polymeric complexes.39-42 With this precedent, we believed that salproH3 might

prove a route to more new Mn compounds under appropriate reaction conditions, and decided to

investigate its reactions with the [Mn30(O2CR)6 97)3] COmplexes. It was obvious that

pentadentate salproH3 CannOt bind to these Mn3 Species without resulting in a serious structural

perturbation, and a possible nuclearity change. Indeed, as will be described below, these

reactions have yielded new types of Mn4 COmplexes with a core structure that is distinctly

different from those seen before. Additionally, a mononuclear complex has also been obtained.

The syntheses, structures and magnetochemical properties of these complexes are the subj ect of

this chapter.43

2.2 Experimental Section

2.2.1 Syntheses

All preparations were performed under aerobic conditions using reagents and solvents as

received. The compound salproH3 WAS synthesized using the reported procedure.44

[Mn30(O2CMe)6 Py)3] -py, [Mnl2012(O2CMe) 16(H20)4], [Mn30(O2CEt)6 97)3] .py,

[Mn30(O2CBut)6 97)3] and [Mn30(O2CPh)6 py)2(H20)] were synthesized as reported

elsewhere. 12,45

[Mn402(O2CMe)5(alpro)] (2-1). Method A. To a stirred solution of salproH3 (0.05 g,

0.17 mmol) in CH2C 2/MeOH (3/2 mL) was added triethylamine (0.08 mL, 0.55 mmol) followed

by the addition of a solution of complex [Mn30(O2CMe)6 py)3]-py (0.22 g, 0.25 mmol) in

CH2C 2 (10 mL). This solution was left under magnetic stirring for 30 minutes and then filtered

through a medium frit. The brown filtrate was left undisturbed to evaporate slowly, giving X-ray

quality crystals that grew slowly over five days. These were collected by filtration, washed with









CH2C 2 and dried in vacuo. Yield 56%. Anal. Called (Found) for 2-1-CH2C 2

(C28H32Mn4N2015Cl2): C, 36.27 (36.82); H, 3.48 (3.64); N, 3.02 (2.97). IR (KBr, cm )~: 3446br,

1625s, 1568s, 1445s, 1394s, 1297m, 1153m, 1092w, 1026m, 759m, 677s, 595s, 468m.

Method B. To a stirred solution of salproH3 (0.20 g, 0.67 mmol) in MeCN/MeOH (3/2

mL) was added triethylamine (0.08 mL, 0.55 mmol) followed by the addition of a solution of

complex [Mnl2012(O2CMe)16(H20)4] (0.32 g, 0.17 mmol) in MeCN (10 mL). This solution was

left under magnetic stirring for one hour and then filtered through a medium frit. The

homogeneous brown solution was left undisturbed for slow evaporation, giving X- ray quality

crystals that grew slowly over the course of one week. These were collected by filtration, washed

with acetonitrile, and dried in vacuo. Yield 16%. The product was identified as 2-1 by IR

spectral comparison with material from method A.

Method C. To a stirred solution of salproH3 (0.05 g, 0. 17 mmol) in MeCN/MeOH (3/2

mL) was added triethylamine (0.08 mL, 0.55 mmol) followed by the addition of a solution of

Mn(O2CMe)3-2H20 (0.09 g, 0.34 mmol) in MeCN (10 mL).This solution was left under

magnetic stirring for 30 minutes and then worked up as for Method B. Yield 28%. The product

was identified as 2-1 by IR spectral comparison with material from method A.

[Mn402(O2CEt)s(salpro)] (2-2). To a stirred solution of salproH3 (0.05 g, 0. 17 mmol) in

CH2C 2/MeOH (3/2 mL) was added triethylamine (0.08 mL, 0.55 mmol) followed by the

addition of a solution of complex [Mn30(O2CEt)6 py)3]-py (0.31 g, 0.34 mmol) in CH2C 2 (10

mL). This solution was left under magnetic stirring for 30 minutes and then filtered through a

medium frit. X-ray quality crystals were obtained over the course of three days by vapour-

diffusing diethyl ether into the filtrate. The resulting crystals were collected by filtration, washed

with ether, and dried in vacuo. Yield 22%. Anal. Called (Found) for 2-2-CH2C 2









(C33H42Mn4N2015Cl2): C, 39.74 (39.66); H, 4.24 (4.21); N, 2.80 (2.68). IR (KBr, cm )~: 3441br,

2879m, 1626s,1572s, 1446m, 1404m, 1299m, 1150w 00,13w 5w 7m 9s 6m

[Mn402(O2CBu')s(salpro)] (2-3). To a stirred solution of salproH3 (0.05 g, 0.17 mmol) in

CH2C 2/MeOH (3/2 mL) was added triethylamine (0.08 mL, 0.55 mmol) followed by the

addition of solution of complex [Mn30(O2CBut)6 97)3] (0.29 g, 0.28 mmol) in CH2C 2 (10 mL).

This solution was left under magnetic stirring for 30 minutes and then filtered through a medium

frit. X-ray quality crystals were obtained during the course of five days by layering the filtrate

with heptane and allowing the solvents to slowly mix. The resulting crystals were collected by

filtration, washed with heptane and dried in vacuo. Yield 25%. Anal. Called (Found) for 2-3-%/

CH2C 2 (C42.5H61Mn4N2015C1): C, 47.58 (47.44); H, 5.83 (5.86); N, 2.49 (2.74). IR (KBr, cm )~:

3442br,2959m, 1627m,1 560s142 1482m, 1447,10m 38,10m,12m 10,12w

895w, 757w, 678m, 599s, 439m.

NMe4[Mn(O2CPh)2(SalproH)] (2-4). To a stirred solution of salproH3 (0.05 g, 0.17

mmol) in MeCN was added 25 wt % solution of NMe40H in MeOH (0.02 mL, 0.49 mmol)

followed by the addition of a solution of complex [Mn30(O2CPh)6 py)2(H20)] (0.08 g, 0.07

mmol) in CH2C 2 (10 mL). This solution was left under magnetic stirring for one hour and then

filtered through a medium frit. X-ray quality crystals were obtained during the course of five

days by layering the filtrate with diethyl ether:hexane (1:1 v/v). Yield 25%. Anal. Called (Found)

for 2-4-M2CH2C 2 (C35.5H39MnN307C1): C, 60.04 (60.34); H, 5.54 (5.90); N, 5.92 (6.02). IR (KBr,

cm )~: 1623s, 1599s, 1541m, 1530m, 1467m, 1446s, 1393w, 1399m, 1304m, 1193m, 1149m,

1089w, 1070w, 1008w, 905m, 861w, 758s, 614 s,599m57m49.

2.2.2 X-ray Crystallography

Data were collected by Dr. Khalil A. Abboud using a Siemens SMART PLATFORM

equipped with a CCD area detector and a graphite monochromator utilizing Mo-Ku radiation (h









= 0.71073 A+). Suitable crystals of 2-1-MeCN, 2-3-MeOH-2CH2C 2C7H16 and 2-4-CH2C 2 were

attached to glass fibers using silicone grease and transferred to a goniostat where they were

cooled to 173 K for data collection. Cell parameters were refined using up to 8192 reflections. A

full sphere of data (1850 frames) was collected using the co-scan method (0.30 frame width). The

first 50 frames were remeasured at the end of data collection to monitor the instrument and

crystal stability (maximum correction on I was < 1 %). Absorption corrections by integration

were applied based on measured indexed crystal faces.

The structures were solved by direct methods in SHELXTL6,46 and refined using full-matrix

least squares. The non-HI atoms were treated anisotropically, whereas the hydrogen atoms were

placed in ideal, calculated positions and refined as riding on their respective carbon atoms.

Refinement was done using F2. In 2-1-MeCN, the asymmetric unit consists of the Mn4 ClUSter

and a disordered MeCN molecule. A total of 465 parameters were refined in the final cycle of

refinement using 5581 reflections withlI> 20(I to yield R1 and wR2 Of 4. 13 and 8.28%,

respectively. In 2-3-MeOH-2CH2C 2C7H16, the asymmetric unit consists of the cluster and one

heptane, one methanol, and two dichloromethane molecules. The three methyl groups on C17 are

disordered, and each dichloromethane molecule has one chlorine atom disordered. In each case,

two disorder sites were included, and their site occupation factors independently refined. A total

of 723 parameters were included in the final cycle of refinement using 9336 reflections with I >

20(I to yield R1 and wR2 Of 4.73% and 12.20%, respectively. In 2-4-CH2C 2, A total of 920

parameters were refined in the final cycle of refinement using 7287 reflections withlI> 20(1 to

yield R1 and wR2 Of 4.58% and 8.44%, respectively.

The crystallographic data and structure refinement details for 2-1-MeCN, 2-3 MeOH-

-2CH2C 2C7H16 and 2-4-CH2C 2 are liSted in Table 2-1.









2.3 Results and Discussion


2.3.1 Syntheses

Trinuclear [Mn30(O2CR)6 97)3 0,+ ClUSters have proven to be very useful starting materials

for the synthesis of higher nuclearity products, affording complexes of nuclearity 4 18. For

example, reaction of [Mn30(O2CMe)6 97)3]+ with 2,2' bipyridine (bpy)36 Or picolinate (pic-)37

gave [Mn402(O2CMe)7(bpy)2]+ and [Mn402(O2CMe)7(pic)2]~ Salts, respectively. In addition, the

reaction of a mixture of [Mn30(O2CMe)6 97)3]' and [Mn30(O2CMe)6 py)3]-py with 2-

(hydroxyethyl)pyridine (hepH) gave [Mnls0l4(O2CMe)ls(hep)4(hepH)2(H20)2 2+ 3 However, a

chelating reagent is not always necessary: Treatment of [Mn30(O2CPh)6 py)2(H20)] with phenol

give s [Mn6 02(O2 CPh) 10(py)2(Me CN)2], the phenol m erely acti ng as a reduci ng agent and

trigerring dimerization.47 Thus, the choice of chelating or other co-reagent and the reaction

conditions have significant effect not only on the nuclearity of the product but also on its metal

topology.

Along the same lines, the reaction of [Mn30(O2CMe)6 py)3]-py with salproH3 and NEt3 18

a roughly 3:2:6 molar ratio in CH2 12:MeOH gave the novel tetranuclear complex

[Mn402(O2CMe)s(salpro)](2-1). Its formation can be summarized in eq. 2-1.The reaction is

sensitive to the Mn3:SalproH3 ratio, and complex 2-1 is obtained only when 0.5 0.7 equivalents

of salproH3 per Mn3 is employed. We also found that the mixed CH2 12:MeOH solvent system is

very important; no reaction was observed when the reaction was performed in CH2 12 alOne, and

only starting material was recovered. Presumably, the more polar MeOH facilitates the necessary

proton transfer steps. However, the yield of the product decreases as the concentration of MeOH

increases beyond that described in the Experimental Section, presumably due to the solubility of

the product. Thus, a controlled amount of MeOH is essential for a high yield reaction. However,

the same product was obtained using a CH2 12:EtOH solvent system. Further investigation









showed that the same complex 2-1 was also obtained from MeCN:MeOH. Also the same product

is obtained in the absence of base but in lower yield.

2 [Mn30(O2CMe)6 97)3] + SalproH3 [Mn402(O2CMe)s(salpro)] + 7MeCO2~ + 2Mn2+

6py + 3H' (2-1)

Since complex 2-1 contains 4 Mn ions, we wondered if a higher oxidation state product

might form if we employed a higher oxidation state reagent, and thus explored the reaction of

salproH3 with [Mnl2012(O2CMe) 16(H20)4], which contains 8Mn"',4Mn'V. Thus, an

MeCN:MeOH solution of [Mnl2012(O2CMe)16(H20)4] WAS treated with 4 equivalents of

salproH3, but the same product, complex 2-1, was again obtained, but only in poor yield (16%).

We also obtained complex 2-1 when the [Mn30(O2CMe)6 py)3]py Starting material was

replaced with "Mn(O2CMe)3-2H20". This "Mn" acetate" is really a polymer of Mn3 trinuclear

units similar to those in [Mn30(O2CMe)6 py)3]-py, and so it was perhaps not surprising that its

reaction with salproH3 gave the same product.

Access to other carboxylate derivatives of 2-1 is possible using the described procedure of

Method A. Thus, the reaction using the R = Et propionatee) or But (pivalate) derivatives of

[Mn30(O2CR)6 97)3] gave the corresponding [Mn402(O2CR)(salpro)] complexes 2-2 and 2-3,

respectively. This provides access to more soluble versions of this new structural type of Mn4

complex. However, all our attempts to prepare benzoate derivative of 2-1 resulted in the

formation of the monomeric complex NMe4[Mn(O2CPh)2(SalproH)] (2-4).

2.3.2 Description of Structures

2.3.2.1 Structure of [Mn402(o2CR)s(salpro)] (R = Me (2-1), But (2-3))

The structures of 2-1 and 2-3 are shown in Figure 2-2. Selected interatomic distances and

angles for 2-1 are listed in Table A-1. Complex 2-1 crystallizes in monoclinic space group P21/n.

It contains a [Mn402 8+ COre with peripheral ligation provided by five doubly-bridging acetate









groups and one pentadentate salpro3- ligand (Figure 2-2). The core can be described as derived

from two triangular, oxide-bridged [Mn30] units sharing an edge and thus giving a [Mn402]

butterfly-like core as found in several other Mn4 COmplexes (as well as with other transition

metals), for example, [Mn402(O2CMe)7(bpy)2]+ (Mn4-bpy).36 Atoms Mn1 and Mn3 occupy the

'body' positions of the butterfly and are five-coordinate (square pyramidal), and Mn2 and Mn4

occupy the 'wingtip' positions and are six-coordinate (octahedral). However, the [Mn402] in 2-1

is much more closed up, i.e. a more acute V-shape, than normally found in such butterfly units

(Figure 2-3) This is reflected in the dihedral angle between the two Mn3 planOS, which is 79.20 in

2-1 compared with 134.30 in Mn4-bpy,36 which is typical of previous butterfly complexes. This

can clearly be assigned to the fact that the wingtip Mn atoms of the butterfly topology, Mn2 and

Mn4, are mono-atomically bridged by the salpro3- Oxygen atom 01 (Figure 2-2, top). In fact, this

drastic closing up of the butterfly makes the core of 2-1 intermediate between a butterfly and a

cubane (i.e. tetrahedral) Mn4 topology. There is a carboxylate group bridging each body-wingtip

Mn pair, and a fifth carboxylate bridges the body-body Mn pair. The pentadentate salpro3- ligand

completes the peripheral ligation, chelating each wingtip Mn atom and bridging them via Ol.

All the Mn atoms are in the +3 oxidation state. This was established by qualitative

consideration of the bond distances at each Mn, and confirmed quantitatively by bond valence

sum (BVS) calculations (Table 2-2).48,49 This also agreed with charge considerations and the

overall neutrality of the molecule, as well as the clear presence of a Jahn-Teller (JT) distortion at

near-octahedral Mn2 and Mn4, the JT axes being along the 09, 07 and 015,013 vectors,

respectively.

Complex 2-3 crystallizes in monoclinic space group P21/n. Selected interatomic distances

and angles are listed in Table A-2. Complex 2-3 is isostructural with complex 2-1 except for the









difference in the carboxylate R groups. In particular, the dihedral angle between the two Mn3

planes is 79.80, and the Mn JT axes have the same relative orientation. The bulky But groups

thus have only a minimal effect on the structure, as expected from the lack of any steric

interactions.

As discussed above, the structures of 2-1 and 2-3 can be considered closed-up versions of

the familiar butterfly structures observed on several previous occasions in Mn"' chemistry. It is

thus of interest to structurally compare the two types, and this is done in Table 2-3. The metric

parameters are fairly similar, as expected given that they are all Mn species, but some overall

conclusions can nevertheless be drawn. The closing up of the core of 2-1 and 2-3, which is

effectively a pivoting of the wingtip Mn atoms (Mnw) about the CL3-02- ions, has the effect of

greatly decreasing the Mnw-O-Mnb as expected (by ~10-150), but also slightly decreasing the

Mnb-O-Mnb angles (by ~1-20) as the central [Mn202] rhombus buckles into a non-planar

conformation. These angle changes are also reflected in the Mn---Mn separations, which all

decrease by ~ 0.1-0.2 A+, except for the Mnw---Mnw separation which is much shorter in 2-1 and

2-3.

2.3.2.2 Structure of NMe4[Mn(O2CPh)2(SalproH)] (2-4)

Complex 2-4 crystallizes in monoclinic space group P21/c. Selected interatomic distances

and angles are listed in Table A-3. The asymmetric unit contains two monomers, the structures

of which are essentially superimposable, therefore structure of one of these will be discussed

here. It contains a Mn" "ion in octahedral geometry as shown in Figure 2-4. The four equatorial

sites are occupied by two N and two alkoxide oxygen atoms of salproH molecule and two rl

benzoate groups disposed trans to each other along the axial direction forming the JT axis. The

protonation level of the central alcohol arm (07) of salproH was established by a BVS of 0.9.









07 also forms an intramolecular hydrogen bond with the 03 of the benzoate anion; 07---03 =

2.65 A+.

2.3.3 Magnetochemistry

Variable temperature dc magnetic susceptibility data were collected in the 5.0 300 K

range in a 0. 1 T magnetic field on powdered microcrystalline samples of 2-1-CH2 12,

2-2-CH2 12, 2-3-M2CH2 1 2 and 2-4-M2CHzC 2 2 TStrained in eicosane to prevent torquing. The 3&T

vs Tfor 2-1 to 2-3 are shown in Figure 2-5. For 2-1-CH2 12, 3CnT Smoothly decreases from 9. 1

cm3Kmoll at 300 K to 0.4 cm3Kmol-l at 5.0 K. The 300 K value is much less than the spin-only

value of 12.0 cm3Kmol-l (g = 2.0) expected for four Mn ions with non-interacting metal

centers, indicating the presence of appreciable intramolecular antiferromagnetic interactions

between the Mn ions, with the low temperature data suggesting a spin S = 0 ground state. Similar

data were obtained for 2-2-CH2 12 and 2-3-M2CH2 12, COnsistent with their isostructural nature

and a minimal influence of the different ligands in the three complexes (Figure 2-5).

The isotropic Heisenberg-Dirac-VanVleck (HDVV) spin Hamiltonian describing the

exchange interactions within these Mn4 COmplexes with virtual C2V Symmetry is given by eq. 2-

2, where b = body, w = wingtip, Si (i = 1 4) is the spin operator for metal atom Mni, and J is the

exchange parameter. The exchange and atom labeling are summarized in the Figure 2-6.

K= 2Jbb 1 ~3 2Jbw ~1 2 + 1 4 + 2. 3 + 3 4) 2Jww $2. 4 (2-2)

The eigenvalues of the spin Hamiltonian of eq. 2-2 can be determined analytically using the

Kambe vector coupling method,' as described elsewhere for the more common butterfly

complexes such as Mn4-bpy, which also have C2V Symmetry.36 Thus, use of the coupling scheme

RA = 1 +3, ~B = 2 +4, and AT = A +B allOws the spin Hamiltonian to be transformed into

the equivalent form given by eq. 2-3, where ST is the total spin of the molecule. The eigenvalues

of eq. 2-3 can be determined using the relationship Ri2uy = Si(Si+1)Ur, and are given in eq. 2-4,









where E ST, SA, SB> is the energy of state ST, SA, SB>, and constant terms contributing equally

to all states have been omitted. The overall multiplicity of the spin system is 625, made up of 85

individual spin states ranging from ST = 0 8 .

K= Jbb ~A2 12 32) bw T2 A2 2 A A 1w 2 22 2 (2-3)

E ST, SA, SB> = Jbb[SA(SA 1 bw[ST(ST+1)- SA(SA+1)- SB(SB+) J- ww[SB(SB3+1)] (2-4)

An expression for the molar paramagnetic susceptibility, 3y, was derived using the above

and the Van Vleck equation," and assuming an isotropic g tensor (Appendix D-1). This equation

was then used to fit the experimental 3yTyvs T data in Figure 2-5 as a function of the three

exchange parameters Jbb, Jbw and Jww and the g factor. A contribution from temperature

independent paramagnetism (TIP) was held constant at 400 x 10-6 CM13O1-1. The obtained fits are

shown as the solid lines in Figure 2-5. The fitting parameters were: For 2-1, Jbb = -6.37 cm- Jbw

= -5.72 cm- Jww = -1.78 cm-l and g = 2.00; for 2-2, Jbb = -7.64 cm- Jbw = -6.73 cm- Jww = -

2.49 cm-l and g = 2.00; and for 2-3, Jbb = -7.37 cm- Jbw = -6.57 cm l, Jww = -1.79 cm-l and g =

1.99. The obtained values indicate that the ground state of the molecules is ST, SA\, SB> = 0, 4,

4>, as anticipated from the low temperature data in Figure 2-5.

The exchange interactions within the Mn4 COres of 2-1 to 2-3 are thus all antiferromagnetic

and weak. The weakest is the Jww between the wingtip Mn" "ions, but note that this is

nevertheless a significant interaction relative to the others, unlike the more common types of

butterfly species where this Jww interaction is not a maj or contributor since the wingtip Mn atoms

are not directly (monoatomically) bridged. Since the wingtip Mn atoms are bridged by an

alkoxide O atom whereas the other Mn pairs are all bridged by either one or two oxide O atoms,

it is qualitatively reasonable for Jww to be the weakest interaction in the molecule, although the









precise values of all the J parameters are the net sum of contributions from ferro- and

antiferromagnetic pathways and thus it is difficult to rationalize their differences precisely.

It is, however, of interest to compare the exchange parameters for 2-1 to 2-3 with those for

the the more common type of Mn butterfly complexes and see if any observed differences can

be correlated with the str-uctural differences mentioned earlier. In Table 2-4 are compared the

exchange parameters for 2-1 to 2-3 with those for [Mn402(O2CMe)7(bpy)2]+ (Mn4-bpy),36

[Mn402(O2CMe)7(pic)2]- (Mn4-piC 37 and [Mn402(O2CEt)7(bpya)2]+ (Mn4-bpya).52 Although the

Jbw interaction in 2-1 to 2-3 is within the range found for the previous complexes, the Jbb

interaction in the former is distinctly weaker than in the latter. This is consistent with the

significantly more acute angles at the CL3-02- ions, since these will presumably weaken the

anti ferrom agneti c contribute on s to the Jbb i nteracti on by we akeni ng the Mn(dxn)-O(pn) -Mn(dxn)

overlap that would be stronger when mediated by an essentially trigonal planar O atom as in

Mn4-bpy, Mn4-pic and Mn4-bpya. The buckling of the central [Mn202] HO doubt also contributes

to the change in Jbb by affecting the orbital overlap.

The fact that Jbw ~ Jbb in 2-1 to 2-3 is expected to have a clear impact on the ground state

because the butterfly [Mn402 8+ COre in Mn4-bpy, Mn4-pic and Mn4-bpya has been well

established from previous work to experience spin frustration effects as a result of the presence

of triangular Mn3 within its structure.36,37,52 Since the interactions are all antiferromagnetic, they

are competing and the precise ground state spin alignment is thus very sensitive to the Jbw:Jbb

ratio, with Jww not being a factor in Mn4-bpy, Mn4-pic and Mn4-bpya because of its weakness.

For example, the typical butterfly complexes Mn4-bpy and Mn4-pic have an ST = 3 ground state

spin, the ST, SA, SB> = 3, 1, 4> state, which results from the dominating Jbb interaction

aligning the Mnb spins almost perfectly antiparallel, but not quite (i.e. SA = 1 HOt 0). The Jbw









interactions are individually weaker than Jbb, but there are four of them, and as a result prevent

SA being 0. An intermediate resultant spin is thus obtained in the ground state. In Mn4-bpya, Jbb

>> Jwb, and the Mnb spins are now aligned antiparallel, i.e. SA = 0, with the weak Jww serving to

couple the Mnw spins antiparallel and giving an overall ST = 0 ground state, the ST, SA\, SB,

S0, 0, 0> state. The ground state of 2-1 to 2-3 can now be satisfactorily rationalized within this

description. In fact, it represents the situation at the other extreme compared to Mn4-bpya, i.e. the

Jbb is HOw weakened relative to Jbw and the two interactions are comparable in magnitude.

However, there are four of the latter, and thus Jbw now dominates the spin alignments, aligning

the spins antiparallel to their neighbors along the outer edges of the [Mn402 8+ butterfly. The Jbb

interaction is antiferromagnetic but nevertheless completely frustrated, as is Jww, with the two

Mnb spins and the two Mnw spins both being aligned parallel by the Jbw interactions. Thus, the

ground state is again ST = 0, as in Mvn4-bpya, but now it is the ST, SA, SB> = 0, 4, 4> state as

depicted in Figure 2-6. Table 2-5 calculates the spin states of 2-1-CH2 2zin the |ST, SA, SB>

format arranged as a function of their energy calculated using the calculated exchange

parameters, Jbb, Jbw and Jww, and the Van Vleck equation

The &Tys T for 2-4 is shown in Figure 2-7. For 2-4-V2CH2 12, the value of yT decreases

very smoothly from 3.19 cm3Kmol-l at 300 K to 2.57 cm3Kmol-l at 15.0 K and then drops

sharply to 2.21 cm3Kmol-l at 5.0 K. Complexes 2-4 exhibits behavior expected for high-spin

Mn (S = 2) center exhibiting zero-field splitting (ZFS). To characterize the ZFS parameter D

further, magnetization vs field data were collected in the 2.0 10.0 K range in fields of 0. 1 7 T

(Figure 2-7). Fitting of the data using the programM4~rGNET,53 described elsewhere,36,54-56 that

involves diagnolization of the spin Hamiltonian matrix, assuming that only the ground state spin

is populated at these temperatures, includes axial ZFS and Zeeman interactions and carries out a









full powder average, gave S = 2, g = 1.86, and D = -4.09 cm l. To ensure that the global

minimum had been located, we calculated the root-mean-square error surface for the fit as a

function of D and g (Figure 2-8). The plot clearly shows only the above mentioned minima.

2.4 Conclusions

SalproH3 has proved an effective route to a novel type of tetranuclear Mn complex whose

core can be described as a more closed up version of the butterfly-like core that is relatively

common. Three isostructural complexes of this new family have been synthesized and

characterized. These complexes also complement and extend the currently rich area of Mn"

schiff-base species. Complexes 2-1 to 2-3 extend the type of spin frustration effects observed

within the Mn4 butterfly core, giving an ST = 0 ground state due to domination of the spin

alignments by Jbw-

Table 2-1. Crystallographic Data for 2-1-MeCN, 2-3-MeOH-2CHzC2 27H16 and 2-4-CH2 22
2-1 2-3 2-4
Formula C29H33Mn4N3015 C52H84 14Mn4N2016 C36H40 12MnN3 Oz
Fw, g/mola 883.34 1354.77 752.55
Space group P21/n P21/n P21/c
a, A+ 9.3368(6) 17.7518(13) 24.2193(12)
b, A+ 22.5058(15) 17.3654(12) 17.4454(9)
c, A+ 16.5079(11) 21.4029(15) 18.3088(9)
ao 90 90 90
P, 0 90.945(1) 100.856(1) 110.569(1)
Y, 0 90 90 90
7, A3 3468.4(4) 6479.7(8) 7242.6(6)
Z 4 4 8
T, K 173(2) 173(2) 173(2)
n, Ab 0.71073 0.71073 0.71073
Pealc, g/cm3 1.692 1.389 1.380
pmm-l 1.097 0.988 0.563
R1 c"d 0.0413 0.0473 0.0458
wR2 e 0.0828 0.1220 0.0844
a Including solvate molecules. b Graphite monochromator. I> 20(I. d R1 = C(||Fo| |FI|)
c|Fol. e wR2 [C[w(Fo2 Fc2 2] CWFo2 2 11/2, 1[2 Fo2) [P2 +bp], where p = [max
(Fo2, O) + 2Fc2]/3.










Table 2-2. Bond-valence sums for the Mn atoms of complexes 2-1, 2-3 and 2-4 a
Atom 2-1-MeCN 2-3-MeOH-2CHzC2 27H16 2-4-CH2 22
Mn" Mn Mn" Mn" Mn" Mn" Mn" Mn" Mn"
Mn1 3.07 2.81 2.95 2.98 2.73 2.87 3.29 2.84 3.12
Mn2 3.28 3.04 3.13 3.30 3.06 3.14 3.30 2.88 3.14
Mn3 3.04 2.78 2.92 3.33 3.26 3.18
Mn4 3.23 3.16 3.08 3.05 2.79 2.93
a The underlined value is the one closest to the charge for which it was calculated. The oxidation state of a particular
atom can be taken as the nearest whole number to the underlined value


Table 2-3. Comparison of core parameters of selected [Mn402 8+ COmplexes (A+, o)


Mnw-O

909, 1.914a


Mnb-O- Mnb-O- Mnw-O-
Ref
Mnb Mnw Mnw
94.30, 112.7-
133.88 43
93.87 114.5
94.97, 110.6-
134.1 43
94.74 114.6
123.3-
5.7, 96.8 36
131.3
123.2-
96.9 37
129.7
97.07, 125.3-
52
97.25 131.4


ComplexMnb---MnbMnb---MnwMnw--Mnw Mnb-O


3.171-
2-1 2.770
3.196
3.136-
2-3 2.794
3.193
3.299-
Mn4-bpy 2.848
3.385
3.308-
Mn4-pic 2.842
3.406
Mn4- 3.307-
2.871
bpya 3.344


1.886- 1.!


3651


1.906
1.887-
3.640
1.910
1.889-
5.593
1.930
1.888-
1.910
1.873-
1.957


1.977, 1.991
1.907, 1.910a
1.970, 1.984

1.804, 1.844 9

1.840, 1.847

1.833, 1.838


a Top and bottom entries refer to distances to oxide and alkoxide O atoms, respectively.


Table 2-4. Comparison of exchange parameters in [Mn402 8+ COmplexes
Complex Jbb a Jbw a Jww a g Ref
2-1 -6.37 -5.72 -1.78 2.00 43
2-2 -7.64 -6.73 -2.49 2.00 43
2-3 -7.37 -6.57 -1.79 1.99 43
Mn4-bpy -23.5 -7.8 2.0 36
Mn4-pic -24.6 -5.3 1.96 37
Mn4-bpya -25.7 -3.3 -0.77 1.99 52
cml









Table 2-5. Spin states of 2-1-CH2C 2 in the | ST, SA, SB> format arranged as a function of their
energy calculated using the calculated exchange parameters, Jbb, Jbw and Jr~, and the
Van Vleck equation
| ST, SA, SB> E (cm- ) | ST, SA, SB> E (cm- ) | ST, SA, SB> E (cm )
|0,4,4> 0.00 |1,1,0> 95.00 |4, 1,3> 159.20
|1,4,4> 12.40 |2,2,2> 95.30 |4,4,2> 179.30
|1.3.4> 14.00 |2, 1,2> 96.10 |4,3,2> 180.90
|0,3,3> 33.20 |2,0,2> 96.50 |4,2,2> 182.10
|2,4,4> 37.20 |3,4,3> 106.00 |5,4,4> 186.00
|2,3,4> 38.80 |3,3,3> 107.60 |5,3,4> 187.60
|2,2,4> 40.00 |3,2,3> 108.80 |5,2,4> 188.80
|1,4,3> 44.00 |3,1,3> 109.60 |5, 1,4> 189.60
|1,3,3> 45.60 |2,3,1> 109.90 |4,4,1> 195.10
|1,2,3> 46.80 |3,0,3> 110.00 |4,3,1> 196.70
|0,2,2> 58.10 |2,2,1> 111.10 |4,4,0> 203.00
|2,4,3> 68.80 |2, 1,1> 111.90 |5,4,3> 217.60
|1,3,2> 69.30 |2,2,0> 119.00 |5,3,3> 219.20
|2,3,3> 70.40 |4,4,4> 124.00 |5,2,3> 220.40
|1,2,2> 70.50 |4,3,4> 125.60 |5,4,2> 241.30
|1,2,2> 71.30 |4,2,4> 126.80 |5,3,2> 242.90
|2,2,3> 71.60 |4, 1,4> 127.60 |5,4,1> 257.10
|2, 1,3> 72.40 |4,0,4> 128.00 |6,4,4> 260.40
|3,4,4> 74.40 |3,4,2> 129.70 |6,3,4> 262.00
|0, 1,1> 74.70 |3,3,2> 131.30 |6,2,4> 263 .20
|3,3,4> 76.00 |3,2,2> 132.50 |6,4,3> 292.00
|3,2,4> 77.20 |3,1,2> 133.30 |6,3,3> 293.60
|3,1,4> 78.00 |3,4,1> 145.50 |6,4,2> 315.70
|0,0,0> 83.00 |3,3,1> 147.10 |7,4,4> 347.20
|1,2,1> 86.30 |3,2,1> 148.30 |7,3,4> 348.80
|1,1,1> 87.10 |3,3,0> 155.00 |7,4,3> 378.80
|1,0,1> 87.50 |4,4,3> 155.60 |8,4,4> 446.40
|2,4,2> 92.50 |4,3,3> 157.20
|2,3,2> 94.10 |4,2,3> 158.40

















Figure 2-1. Structure of SalproH3


Figure 2-2. Labeled representation of the structure of 2-1 (top) and 2-3 (bottom). Hydrogen
atoms have been omitted for clarity. JT axis on Mn are shown in black. Color code:
Mn ', green; O, red; N, blue; C, grey.





















Figure 2-3. Comparison of the cores of 2-1 and 2-3 (left) with that of the normal butterfly
complexes (right). Color code: Mn ", green; O, red.














05

02
-101







Figure 2-4. Labeled representation of the structure of 2-4. Hydrogen atoms have been omitted for
clarity. JT axis on Mn"' are shown in black. Color code: Mn ", green; O, red; N, blue;
C, grey.




























0 SO 100 150 200 250 300
TK


0 SO 100 150
TK


200 250 300


0 SO 100 150
T.K


200 250 300


Figure 2-5. Plots of gynTys Tfor complexes 2-1-CH2 12 (top), 2-2-CH2 12 (middle) and 2-
3-V2CH2 12 (bottom). The solid line is the fit of the data; see the text for the fit
parameters.













t~ot
Mnz Mn4
Mn,
O O
Mn,


Mn~l.- Mni Mn

O bb; O

Mn


Figure 2-6. (left) The core of 2-1 defining the pairwise exchange interactions. (right)
Rationalization of ground state spin of 2-1.


3.5

3.0

2.5

~ z.o
z


i.o

0.5

0.0


a *


0 50 100 150) 2[00
T/K


250) 3(00


40 50


0 10 20 30
HT~'lkGK'"


Figure 2-7. (left) Plots of gy ys. T for complex 2-4-M2CH2 2z. (right) Plot of reduced
magnetization (M/NpuB) VS H/T for complex 2-4-M2CH2 12. The solid lines are the fit of
the data; see the text for the fit parameters.


4



2


E
O
Y
O


-2


1.70 1.75 1.80 1.85 1.90


1.95


Figure 2-8. Two-dimensional contour plot of the fitting-error surface vs D and g for complex 2-
4-M2CH2 12.









CHAPTER 3
ROLE OF MIXED-LIGAND AND MIXED-SOLVENT SYSTEM: ROUTE TO Mn4 AND Mn25

3.1 Introduction

For a number of years, we have been engaged in developing manganese cluster chemistry

with oxide bridges. The motivation for this is the relevance of manganese in bio-inorganic

chemistry and molecular magnetism, not least their aesthetic qualities.27,35,57 In the field of

molecular magnetism, single-molecule magnets (SMMs) hold great current interest. SMMs are

molecular species that can function as nanoscale magnets as a result of their intrinsic properties

rather than as a result of inter-unit interactions and long-range ordering as would be found in

traditional magnetic materials (metals, metal oxides, etc).13 Thus each SMM is a single-domain

magnetic particle, and this arises from the combination of a large ground state spin (S) and an

Ising type (easy-axis) magnetoanisotropy (-ve zero field splitting parameter, D).13

As part of the search of new SMMs, it is important to build high spin and high anisotropy

molecules. It is not possible to achieve a rational synthesis of a molecule with a high D, but there

are certain approaches to high spin molecules.22 One of them being replacement of hydroxides

by end-on azides and second is the use of ligands which generally promote ferromagnetic

interactions.58-6 More recently, our group has demonstrated "spin tweaking', the conversion of

an already high-spin Mn25 SMM with S = 51/2 into a structurally similar one with S = 61/2 by

modification of the peripheral ligands and conversion of a low-spin Mn30 triangular cluster to a

high spin Mn3 ClUSter by the use of appropriate chelate.34,61

We and others have extensively investigated chelating agent 2-hydroxymethyl pyridine

(hmpH) and 2,6-pyridinedimethanol (pdmH2) in manganese cluster chemistry and these have

proven to be very useful in affording high nuclearity and high spin complexes in Mn cluster

chemistry.34,62,63 Examples of these include Mnlo (S = 22)63 and Mn25 (S = 51/2, 61/2).34,62 We









have been continuing our investigation into the usefulness of the mixed-chelate system and

therefore we decided to use mixture of hmpH and pdmH2, which have already been proved to be

very promising in high nuclearity Mn cluster chemistry, with [Mn30(O2CMe)6 97)3]+ COmplexes.

As mentioned in Chapter 2, Mn30 trinuclear complexes have proven to be extremely useful in

producing polynuclear clusters with high ground state spin and possible SMM behavior.64-67

Indeed, as will be described below, reaction of mixed chelate system (hmpH and pdmH2) with

[Mn30(O2CMe)6 97)3]+ has led to a ferromagnetically coupled Mn4 ClUSter and a Mn25 ClUSter

with a ground state spin of 65/2.

3.2 Experimental Section

3.2.1 Syntheses

All manipulations were performed under aerobic conditions and all chemicals were used as

received unless otherwise noted. [Mn30(O2CMe)6 py)3]CIO4 WAS synthesized as reported

elsewhere.45

[Mn4(hmp)4(pdmH)2 Me N)4] (CIO4)4 (3-1). Method A. To a stirred solution of hmpH

(0.05 mL, 0.50 mmol) and pdmH2(0.033 g, 0.25 mmol) in MeCN (15 mL) was added MeCO2Na

(0.08 g, 1.0 mmol) and Mn(CIO4)2 ( 0.40 g, 1.1 mmol) followed by the addition of NEt3 (0.07

mL, 0.5 mmol). The resulting reddish-brown solution was stirred for two hours and then filtered

to remove any undissolved solid and then left for slow evaporation. X-ray quality crystals of 3-1

grew over a period of two weeks in 20% yield. These were collected by filteration, washed with

MeCN and dried in vacuo. Anal. Called (Found) for 3-1(C46H52NloMn4024C 4): C, 37.07 (36.85);

H, 3.52 (3.46); N 9.40 (9.07)%. IR (v, cm )~: 1607m, 1578w, 1440m, 1369w, 1281w, 1145s,

1088s,768w, 720w, 673w, 626m, 575w.

Method B. To a stirred solution of hmpH (0.02 mL, 0.21 mmol) and pdmH2 (0.06 g, 0.45

mmol) in MeCN (15 mL) was added [Mn30(O2CMe)6 py)3]CIO4 (0.22 g, 0.25 mmol). The









resulting solution was stirred for 30 minutes and then filtered and left for slow evaporaton.

Reddish brown crystals of 3-1 were obtained after 15 days in 5% yield. The product was

identifield as 3-1 by IR spectral comparison with material from method A.

[Mn2501s(OH)2(hm p)6(p dm)s(p dmH)2(L)2] (CIO4)6 (3-2). T o a sti rred s oluti on of

[Mn30(O2CMe)6Py3]CIO4 (0.20 g, 0.23 mmol) in MeOH/MeCN (5/15 mL) was added hmpH

(0.02 mL, 0.23 mmol) and pdmH2 (0.06 g, 0.46 mmol). This was followed by the addition of

NEt3 (0.03 mL, 0.23 mmol). The resulting reddish brown solution was left under magnetic

stirring for 30 minutes and then filtered through a medium frit. The resulting brown filterate was

left undisturbed to evaporate slowly at room temperature. The X-ray quality crystals of

3-2-8MeCN-4MeOH were obtained over a period of two weeks in 10% yield. These were

collected by filteration, washed with MeCN and MeOH and dried in vacuo. Anal. Called (Found)

for 3-2-3H20 (C120H128NlsMn25079C 6): C, 30.85 (30.70); H, 2.76 (2.79); N 5.39 (5.05)%. IR (v,

cm )~: 1600s, 1577s, 1440s, 1388m, 1276w, 1102s, 1074s, 774w, 676m, 635s, 559m, 503w,

438w.

3.2.2 X-ray Crystallography

Data were collected by Dr. Khalil A. Abboud on a Siemens SMART PLATFORM equipped

with a CCD area detector and a graphite monochromator utilizing MoKu radiation (h = 0.71073

A+). Suitable crystals of 3-1 and 3-2-8MeCN-4MeOH were attached to glass fibres using silicone

grease and transferred to a goniostat where they were cooled to 173 K for data collection. Cell

parameters were refined using up to 8192 reflections. A full sphere of data (1850 frames) was

collected using the co-scan method (0.3o frame width). The first 50 frames were re-measured at

the end of data collection to monitor instrument and crystal stability (maximum correction on I

was < 1 %). Absorption corrections by integration were applied based on measured indexed

crystal faces. The structure was solved by the Direct Methods in SHELXTL6,46 and refined using









full-matrix least squares. The non-HI atoms were treated anisotropically, whereas the hydrogen

atoms were calculated in ideal positions and were riding on their respective carbon atoms.

Refinement was done using F2

In 3-1, the asymmetric unit consists of a half Mn4 ClUSter and two perchlorate anions. A

total of 403 parameters were refined in the final cycle of refinement using 10860 reflections with

I > 20(1 to yield R1 and wR2 Of 4.97% and 1 1.82%, respectively.

In 3-2-8MeCN-4MeOH, the asymmetric unit consists of a half Mn25 ClUSter, three

perchlorate anions, four acetonitrile molecules and two methanol molecules. All solvent

molecules and the anions were disordered and could not be modeled properly, thus program

SQUEEZE,68 a part of the PLATON package of crystallographic software, was used to calculate

the solvent disorder area and remove its contribution to the overall intensity data. In particular,

the perchlorate anions were extensively disordered and each was refined in three parts before

their contributions were removed from the intensity data by SQUEEZE. The cluster also has a

couple of disorders. Two of the pdmH2 ligands seem to have been oxidized during synthesis to

produce, in each case, a carboxylate instead of one of the hydroxyl group. Both parts of each

disorder were fixed at 50% occupancy after the refinement produced value close to 50%. A total

of 979 parameters were refined in the final cycle of refinement using 55761 reflections with I > 2

o(I to yield R1 and wR2 Of 6.70% and 15.77%, respectively. The crystallographic data and

structure refinement details for 3-1 and 3-2-8MeCN-4MeOH are listed in Table 3-1.

3.3 Results and Discussion

3.3.1 Syntheses

The reaction of preformed trinuclear complexes [Mn30(O2CMe)6 97)3 0,+ with chelating

ligands has been used extensively in the past to trigger structural rearrangements leading to

higher nuclearity products. The choice of chelate and the reaction conditions have significant









effect not only on the nuclearity of the product but also on its metal topology. Along the same

lines, the reaction of [Mn30(O2CMe)6Py3]CIO4 with hmpH and pdmH2 in 1:1:1 molar ratio in

MeCN gave [Mn4(hmp)4(pdmH)2(MeCN)4 4+ (3-1). Its formation is summarized in eq. 3-1. The

yield of the reaction is extremely low and the attempts to increase the yield gave a mixture of

products and were not further pursued. Instead, we sought its synthesis from a more convenient

procedure that doesn't employ preformed triangular complex but only simple starting materials.

This was successfully accomplished from the reaction of Mn(CIO4)2, hmpH and pdmH2 in the

presence of MeCO2Na. Use of sodium acetate is found to be essential in this reaction, it does not

appear in the final product but presumably acts as a base.

2 [Mn30(O2CMe)6Py3]' + 4hmpH + 2pdmH2 + 4MeCN + 2e -

[Mn4(hmp)4(pdmH)2(MeCN)4 4+ + 2Mn3+ + 12MeCO2- + 6Py +2H20 + 2H+ (3-1)

On the other hand, 3-2-8MeCN-4MeOH is obtained upon the reaction of

[Mn30(O2CMe)6Py3]CIO4 with hmpH and pdmH2 in MeCN/MeOH system. This reaction is very

sensitive to the ratio of Mn3:hmpH:pdmH2 and the clean product is obtained only when ratio

mentioned in the experimental section is used. Using higher concentration of hmpH or pdmH2

beyond that described in the experimental section gives a mixture or orange and red crystals.

And the presence of MeOH is absolutely essential for the formation of product. Presumably, the

more polar MeOH facilitates the necessary proton transfer steps. In addition, 3:1 v/v

MeCN:MeOH is found to be the best for clean reaction. However, the yield of the product

decreases as the concentration of MeOH increases, most likely due to the solubility of the

product, while reducing the amount of MeOH gives white crystals along with the desired

product. Also, the same product is obtained in the absence of base but in extremely low yield.









One remarkable feature of this reaction is the in situ formation of the oxidized pdmH2

ligand. In 3-2-8MeCN-4MeOH, oxidation of the chelate provides the reducing equivalents to

lower the metal oxidation state of the product. The reduction of the Mn" centers to Mn" is

associated with oxidation of some of the pdmH2, pOssibly to the corresponding aldehyde, 6-

(hydroxymethyl)-2-pyridine carboxaldehyde followed by the further oxidation to 6-

(hydroxymethyl)-2-pyridine carboxylic acid (L, Figure 3-1). This has been seen before with

hmpH69 and pdmH270 ligands in high-oxidation state Mn chemistry. The redox chemistry is also

accompanied by the fragmentation and structural rearrangement of the Mn3 ClUSter to yield Mn25.

Also filterate after the isolation of the product is still dark colored, thus there is a possibility that

disproportionation of Mn is occurring during the reaction and filterate contains both Mn" and

Mn'V ions.

For both compounds, the final yields are low, nevertheless the preparations are

reproducible. As is usually the case for such reactions, several species in equilibrium are likely

present in solution, and the low solubility of one of them is undoubtedly the reason a pure

product can be obtained.

3.3.2 Description of Structures

3.3.2.1 Structure of [Mn4(hmp)4(pdmH)2 Me N)4] (CIO4)4 (3-1)

The labeled structure of complex 3-1 is shown in Figure 3-2 and selected interatomic

distances and angles are listed in Table A-4. Complex 3-1 crystallizes in monoclinic space group

P21/c with the cation lying on an inversion center. The centrosymmetric structure of 3-1 can be

described as a planar mixed-valence Mn4 rhombus consisting of 2Mn" and 2Mn ions. Mn1 and

Mn2 are assigned as Mn and Mn" respectively on the basis of structural parameters and Bond

Valence Sum (BVS)48,49 calculations that gave a value of 2.99 and 1.88 for Mn1 and Mn2

respectively and the presence at Mn1 of a Jahn Teller (JT) elongation axis (O4-Mnl-N1), as









expected for a high spin Mn"'. The rhombus is composed of two Mn3 triangular faces each

bridged by a pU3- Oxygen, 04 from a bidentate chelating pdmH- whose other arm 03 is protonated

as evident from the BVS of 1.09. Hmp- groups are also bidentate with 01 bridging Mn1 and

Mn2 and 02 bridging Mn1 and Mn2'. Peripheral ligation is provided by four MeCN molecules at

Mn2 and Mn2'; which are seven coordinated, whereas Mn1 is distorted octahedral. Structure of

complex 3-1 is similar to the Mn4 COmplexes seen with hmp- and pdmH- alone.

3.3.2.2 Structure of [Mn250ls(OH)2(hmp)6(pdm)s(pdmH)2(L)2] (CIO4)6 (3-2)

The structure of complex 3-2 is shown in Figure 3-3 and selected interatomic distances and

angles are listed in Table A-5. Complex 3-2 crystallizes in triclinic space group Pi with the

cation lying on an inversion center. It can be described as having a barrel-like cage structure

consisting of 6Mn", 18Mn"' and Mnlv ions. The 12pu4-O2- 6pU3-O2- and 2pu3-OH- ions hold the

core together as well as chelating bridging hmp- and pdmH- and L groups, where L = 6-

hydroxymethyl-2-pyridine carboxylic acid (Figure 3-1). The manganese oxidation states and

protonation levels of OH- hmp-, pdm2-, pdmH- and L oxygen atoms were established by Mn and

O BVS calculations (Table 3-2 and 3-3), inspection of metric parameters and detection of Mn"'

JT elongation axes. The core can be dissected into five parallel layers of three types with an

ABCBA arrangement (Figure 3-4). Layer A is a Mn"3 triangular unit (Mn1, Mnl1 and Mnl2)

with a capping pu3-OH- ions; layer B is a Mn '6 triangle (Mn2, Mn3, Mn5, Mn7, Mnl0 and

Mnl3) comprising three corner-sharing Mn 3 triangles; and layer C is a Mn '6 hexagon (Mn4,

Mn6, Mn8 and their symmetry equivalents) with a central Mn'V ion (Mn9). Layer C has the

Anderson type structure seen before in some Mn complexes. Each layer is held together and

linked to its neighboring layers by a combination of oxide, alkoxide bridges. The outer

coordination shell is occupied by hmp- pdm2-, pdmH- and L ligands. There are two types of









Mn ions; those in layer B are nearly octahedral while those in layer C are pentagonal

bipyramidal.

There are two Mn25 COmplexes reported in literature, [Mn2501s(OH)2(N3)12(pdm)6-

(pdmH)6]Cz 2 MR25+2) and [Mn2501s(OH)(OMe)(hmp)6(pdm)6(pdmH)6](N32CO) (MR25 s).

All these Mn25 COmplexes have the same layered arrangement of Mn ions, however the precise

means by which layers are connected to each other are different. The main difference between

Mn25+2 with S = 51/2 and Mn25 with S = 61/2 is that 12 bound azides in Mn25+2 are replaced by

six rl:rl :p2 hmp- ligands in Mn25 ". As a result, all intra- and inter layer bridges as well as all

M2 pairWISe exchange interactions are now through oxide anions. Complex 3-2 is very much

similar to Mn25 except for the arrangement of layer A, where one alcohol arm of the pdmH-

group has been oxidized and the three Mn" ions are bridged by an OH- instead of OMe- in

Mn25+

3.3.3 Magnetochemistry

3.3.3.1 De Studies

Variable-temperatue dc magnetic susceptibility measurements were performed on dried micro

crystalline sample of 3-1 and 3-2-3H20 at 0. 1 T and 5.0-300 K range. The data for 3-1 is shown

as&nTys Tplot in Figure 3-5. For 3-1, the value ofnT increases from 16.44 cm3Kmol-l at 300

K to 35.3 cm3Kmol-l at 5 K. The &Tvalue at 300 K is higher than the spin only value of 14.75

cm3Kmol-1 expected for two Mn and two Mn" ions suggesting predominant ferromagnetic

interactions within the molecule. The value ofnT at 5 K suggests S = 9 ground state spin. To

determine the individual pair-wise exchange interactions Jij between MniMnj pairs within the

molecule, the ynTys T data for complex 3-1 was fit to a simulation curve deduced from the

Heisenberg-Dirac-VanVleck spin Hamiltonian given in eq. 3-2 where RA = 1 +3, ~B 2 z 4,

and ST = SA + SB. The exchange and the atom labeling scheme are summarized in Figure 3-6.









The energies of the resultant total spin states ST, which are eigen functions of the Hamiltonian in

this coupling scheme, are given by eq. 3-3. The overall multiplicity of the spin system is 900,

made up of 110 individual spins states ranging from ST = 1 to 9. An expression for the molar

paramagnetic susceptibility was derived for this complex using the Van Vleck equation" (see

Appendix D-2, Refer Chapter 2 for details of vector coupling)

K= Jwb ~T2 ~A2 ~B2) bb ~A2 ~12 ~32) (3 -2)

E|ST,SA, SB> = wb[ST(ST+1)-SA(SA+1)-SB(SB 1)-Jbb[SA(S+) (3-3)

Excellent fits were achieved when data below 10 K was omitted and TIP fixed to 6 X 10-4

cm3mOl-1. The obtained fit is shown as solid lines in Figure 3-5 and was obtained with Jbb = 6.67

cm- Jbw = 0.36 cm-l and g = 2.06. These values identify the ground state as |ST,SA,SB>

|9,4,5>. The values of Jbb, Jbw and g are similar to those reported in literature.54,65

To confirm the above ground state spin estimates and to determine the magnitude of zero

field splitting parameter (D), variable field (H) and temperature-magnetization (M)1 data were

collected in the 1.8-10 K and 2-6 T ranges. The resulting data for 3-1 are plotted in Figure 3-7 as

reduced magnetization (M/NpuB) VS H/T, where Nis Avogadro's number and puB is the Bohr

magneton. The data were fit, using the programM4~rGNET,53 by diagonalization of the spin

Hamiltonian matrix assuming that only the ground state is populated, incorporating axial

anisotropy (Di2) and Zeeman terms, and employing a full powder average.36,54-56 The

corresponding spin Hamiltonian is given by eq 3-4, where Sz is the easy-axis spin operator, g is

the Lande' g factor, puo is the vacuum permeability, and H is the applied field. The last term in eq

3-3 is the Zeeman energy associated with an applied magnetic field. The best fit for 3-1 is shown

as the solid lines in Figure 3-7 and was obtained with S = 9, g = 2.00 and D = -0.27 cml

K = Di2 + gB UOS-H (3 -4)









To ensure that the true global minimum had been located and to ensure the hardness of the

fit, root-mean-square error surface for the fit was generated using the program GRID,n which

calculates the relative difference between the experimental M/NpuB data and those calculated for

various combinations of D and g. This is shown as two-dimensional contour plot in Figure 3-7,

the plot shows that fit minimum is a soft one, consistent with significant uncertainty in the

precision of the obtained g and D fit values, which we estimate as & 0.04 on g and & 10 % on D.

For 3-2-3H20, the value ofnT steadily increases from 1 17.5 cm3Kmol-l at 300 K to a

maximum of 540.9 cm3Kmol-l at 15 K before dropping to 489.4 cm3Kmol-l at 5 K (Figure 3-9).

The value ofnT at 15 K suggests a very large ground-state spin (S) value, with the sharp

decrease at the lowest temperature, assigned to Zeeman effects, zero-Hield splitting, and/or weak

intermolecular interactions. To determine the ground state, magnetization (M)1 data were

collected in 0. 1-0.8 T and 1.8-10.0 K ranges and are plotted as (M/NpuB) VS H/T in Figure 3-9. We

used only the low-Hield data (< 0.8 T) to avoid problems with the low-lying excited states. The

best fit (solid-lines in figure 3-9) gave S = 65/2, g = 1.99 and D =- 0.0082 cm- The fits for 63/2,

67/2 were also good, with best fit parameters ofg =2.08/D = -0.01 and g = 1.93/D = -0.0077 cml

respectively. We conclude that 3-2-3H20 has a ground state of 65/2 & 1.

3.3.3.2 Ac Studies

Ac susceptibility measurements were performed on polycrystalline sample of 3-1 and

3-2-3H20 under 3.5 G oscillating ac field and zero dc field as a function of temperature and

frequency. The obtained data for 3-1 is shown as 3y'Tyvs T and 3y"l vs T plots in Figure 3-8.

Extrapolation of the 3y'T plot to 0 K from temperatures above ~ 3.5 K gives value of ~ 43

cm3Kmol l, corresponding to S = 9 with g = 1.95 in very satisfying agreement with the

conclusions from the fits of dc magnetization data. Complex 3-1 shows frequency-dependent in-









phase (a)n and out-of-phase (3p") signals, which are a signature of slow relaxation of

magnetization i.e. SMM behavior by analogy to related [Mn4] COmpounds.

The in-phase ac susceptibility (3a) signal for 3-2-3H20 is shown as y'Tys Tin Figure 3-

10, and extrapolation of the 3y'T signal to 0 K from above about 8 K (to avoid the effect of

intermolecular interactions at lower temperatures) gave 550 cm3Kmol- consistent with the dc

data. The value of 550 cm3Kmol-l is consistent with 1) S = 65/2 / g = 2.01, 2) S = 63/2 / g = 2.07

and 3) S = 67/2 / g = 1.95. The ac data thus confirms a ground state of 65/2 & 1. No out-of-phase

signal was observed down to 1.8 K. The very small D value seen for 3-2-3H20 is consistent with

the nearly perpendicular disposition of the Mn anisotropy axes. We have not done with micro-

SQUID hysteresis measurements on single crystals of 3-2 but on the basis of previously reported

Mn25 COmplexes (Mn25+2 and Mn25 "), which show hysteresis loops and are SMMs, we believe

that 3-2-3H20 is also an SMM, albeit at very low temperature.

3.4 Conclusions

The combination of mixed-chelate system (hmpH and pdmH2) in Mn cluster chemistry

has yielded a ferromagnetic coupled Mn4 rhombus and an unusually high spin Mn25 COmplex

depending on the identity of the solvent employed. The latter point emphasizes the exquisite

sensitivity of the reaction product on reaction conditions. Dc and ac studies have established that

Mn25 pOssesses an S = 65/211 ground state spin. Such a high spin value is extremely rare; in fact,

it is the second highest in Mn chemistry. Achieving high spin ground state is one of the elusive

goals in the search for obtaining superior SMMs. Once again manganese cluster chemistry

continues to surprise and astound with the remarkable variety and aesthetic beauty of its

molecular offspring.










Table 3-1. Crystallographic Data for 3-1 and 3-2-8MeCN-4MeOH
3-1 3-2
Formula C46H52C 4Mn4N1oO24 C140H162C 6Mn25N260so
Fw, g/mola 1490.54 5075.16
Space group P21/n C2/c
a, A+ 11.2401(9) 30.710(2)
b, A+ 21.2408(15) 30.818(2)
c, A+ 12.6654(2) 18.0521(13)
ao 90 90
P, a 101.312(2) 91.415(2)
Y, 0 90 90
i: A3 2965.1(4) 17080(2)
Z 2 4
T, K 173(2) 173(2)
n, Ab 0.71073 0.71073
Pealc, g/cm3 1.669 1.974
pu, mm-l 1.101 1.976
R1 c"d 0.0497 0.0670
wR2 e 0.1182 0.1577
including solvate molecules. b Graphite monochromator. I> 20(1. R1 = C(||Fo| |Fc||)l /l|Fol.
w'R2 = [C[w'(Fo2 F, ) ] / C[w'(Fo ) ]]'" 2. M 1/[o (Fo ) + [(ap)) +bp)], where p = [max (Fo O) + 2F, ]/3.

Table 3-2. Bond-valence sums for the Mn atoms of complex 3-2a
Atom Mnn' Mn"' MnIV Atom Mnn' Mn MnIV
Mn1 1.97 1.82 1.88 Mn8 3.21 2.97 3.07
Mn2 3.23 2.99 3.08 Mn9 4.29 3.92 4.12
Mn3 3.25 2.98 3.12 Mnl0 3.19 2.96 3.05
Mn4 3.17 2.92 3.03 Mnl1 1.92 1.77 1.83
Mn5 3.26 2.98 3.13 Mnl2 1.96 1.81 1.87
Mn6 3.22 2.97 3.07 Mnl3 3.19 2.95 3.04
Mn7 3.24 2.97 3.11
aThe underlined value is the one closest to the charge for which it was calculated. The oxidation state of a
particular atom can be taken as the nearest whole number to the underlined value

Table 3-3. Bond-valence sums for the O atoms of complex 3-2a
Atom BVS Assignment Atom BVS Assignment
Ol 1.94 O2 013 1.85 O2
03 2.06 O2 015 1.16 OH~
04 1.85 O2 017 1.85 O2
011 1.81 O2 026 1.05 OH
aThe BVS values for O atoms of O -, OH- and HzO groups are typically
1.8-2.0, 1.0-1.2 and 0.2-0.4, respectively, but can be affected somewhat
by hydrogen-bonding.















OH OH
hmpH pdmH2


OH OH


Figure 3-1. Structure of ligands: 2-hydroxymethyl pyridine (hmpH), 2,6-pyridine dimethanol
(pdmH2), 6-hydroxymethyl 2-pyridine carboxylic acid (L).


Figure 3-2. Labeled representation of the structure of 3-1. JT axis on Mn are shown in black.
Color code: Mn ", green; Mn", purple; O, red; N, blue; C, grey.


Figure 3-3. Structure of the cation of 3-2. Color code: Mn V, yellow; Mn ", green; Mn", purple;
O, red; N, blue; C, grey.












































A B -C
Figure 3-4. Centrosymmetric core of 3-2 (top) and its three types of constituent layers (bottom).
JT axis on Mn are shown in bold. Color code: Mn V, yellow; Mn"', green; Mn",
purple; O, red; N, blue; C, grey.

40 ,


1 I I I I I 1


N6
1,'


35 e
e


0 50 100 150

T(K)


200 250 300


Figure 3-5. Plots of XnnTys T for complex 3-1. The solid line is the fit of the data; see the text for
the fit parameters.


eMn,




































_ ,







v 3T
o 4 T
a 5T
S 6T
Fitting


Mn1


Mn2


Mn2'


Mn1'


Figure 3-6. The core of 3-1 defining the pairwise exchange interactions.


18

16

14

S12
Z

10

8


-0.15

-0 20

-0.25




-0.35

-0.40

-0.45


0 5 10 15 20 25

H/T(ktGlK)


30 35


1.90 1 95 2.00


Figure 3-7. (left) Plot of reduced magnetization (M/NpuB) VS H/T for complex 3-1. The solid lines
are the fit of the data; see the text for the fit parameters. (right) Two-dimensional
contour plot of the fitting error surface vs D and g for complex 3-1.











































*0.1 T
a 0.2T


v 0.5T
07T

-- Fining


40 -







15 -

--


0 2 4 6 8 10 12 0 2 4 6 8 10 12

T(K) T(K)

Figure 3-8. Plot of in-phase (gs,'T) and out-of-phase (gs,") ac susceptibility data for complex 3-1.


600 -

5 00 -

S400

300 -

t2 200 -

100

0-


70

6 0 i

50

,m 40 i
Z


20 -

10 i


r ,






**


O 50 100 150 200 250 300 0 1 2 3 4 5

T(K) H17(kGlK)

Figure 3-9. (left) Plot of;0uTys Tfor 3-2-3H20 at 0.1 T. (right) Plot of reduced magnetization
(M/NpB) VS H/T for complex 3-2-3H20. The solid lines are the fit of the data; see the
text for the fit parameters.

600





S400


E 300

1-
s 200


100



0 2 4 6 8 10 12 14 16

T(K)

Figure 3- 10. Plot of I')T vs T (in-phase) ac susceptibility data for 3-2-3H20.


-c1000 Hz
-o-t 250 Hz
-A- 50 Hz


-* 1000 Hz
-a- 250 Hz
-A 50 Hz









CHAPTER 4
DIVERSITY OF STRUCTURAL TYPES IN POLYNUCLEAR IRON CHEMISTRY WITH A
(N, N, O)- TRIDENTATE LIGAND

4.1 Introduction

Polynuclear iron compounds with oxygen-based ligation are relevant to a variety of fields

such as bioinorganic chemistry and magnetic materials. Iron-oxo centers are found in several

non-heme metalloproteins and metalloenzymes; for example, in mammals iron is stored as

ferritin, a protein that sequesters Fe"' as a polymeric oxo-hydroxo complex.26 A number of

polynuclear iron complexes have thus been synthesized and studied as possible models for

ferritin in order to gain insights into the biomineralization process involved in the formation of

its metal core.31,72,73 On the other hand, the paramagnetic nature of Fe in its common oxidation

states can often lead to interesting magnetic properties for polynuclear Fe clusters, such as high

ground state spin values and even single-molecule magnetism.13

Although the exchange interactions between Fe"' centers are almost always

antiferromagnetic, certain Fex topologies can nevertheless possess large ground state spin values

as a result of spin frustration. The latter is here defined in its more general sense of competing

exchange interactions of comparable magnitude, preventing (frustrating) the preferred

antiparallel alignment of all spins, and thus giving larger ground state spin values than might be

predicted.74-7 In favorable cases, where these large ground state spins are coupled to a

significant magnetic anisotropy, the compounds can behave as single-molecule magnets

(SMMs).20 Thi s i s the case for SMMs such as [FesO2(OH)12z(tacn)6] 8+ and [Fe4(OMe)6(dpm)6]

ec79-82

The above considerations and others continue to stimulate groups around the world to

develop new synthetic methods that can yield new polynuclear Fe/O clusters. A common

approach has been to use chelates in order to encourage aggregation while ensuring discrete









products. Examples include 2,2'-bipyridine (bpy),7 1,4,7-trialzacyclononane (tacn),s and the

anion of dibenzoylmethane (dbm~).83 When the chelate also contains potentially bridging groups

such as alkoxides, new high-nuclearity products can be obtained. Examples of this include the

deprotonated, tridentate (N,O,O) form of N-methyldiethanolamine (mdaH2) and (O,O,0) form of

tri s(hydroxymethyl)ethane (thmeH3), and others.84-89 We decided to extend thi s approach to the

potentially tridentate (N,N,O) chelate 2- { [2-(dimethylamino)ethyl] -methylamino) ethanol

(dmemH, Figure 4-1). This has some similarity to mdaH2, but it only has one alcohol group and

it was thus anticipated to give new structural types of products. We were unable to locate

previous examples in the literature of transition metal complexes with this chelate.

Our first investigations with dmemH have been in Fe chemistry using the triangular

[Fe30(O2CR)6(H20)3]+ COmplexes as reagents, a common strategy in both Fe"' and Mn"

chemi stry.37,38,90-92 We have found from these reactions that dmemH i s indeed a good route to a

variety of interesting new cluster types. These results are described in this chapter, which reports

the syntheses, structures, and magnetochemical characterization of four new Fe clusters
93
containmng dmem .

4.2 Experimental Section

4.2.1 Syntheses

All preparations were performed under aerobic conditions using reagents and solvents as

received. [Fe30(O2CPh)6(H20)3](NO3), [Fe30(O2CBut)6(H20)3](NO3),

[Fe30(O2CMe)6(H20)3](NO3) and (NEt4)2(Fe20Cl6) WeTO Synthesized as reported elsewhere.94-97

[Fe7O4(O2CPh)nl(dmem)2] (4-1). Method A. An orange red solution of

[Fe3(O2CPh)6(H20)3](NO3) (0.20 g, 0.19 mmol) in MeCN (20 mL) was treated with dmemH

(0.06 mL, 0.38 mmol) and the solution stirred overnight at room temperature. It was then filtered

to remove undissolved starting material, and the filtrate was allowed to stand undisturbed at









room temperature. X-ray quality orange crystals of 4-1-4MeCN slowly formed over 5 days in

45% yield. These were collected by filtration, washed with MeCN, and dried under vacuum.

Anal. Called (Found) for 4-1-V2MeCN (C92H90.5N4.5Fe7O28): C, 52.66 (52.55); H, 4.35 (4.38); N,

3.00 (3.05). Selected IR data (cm )~: 1598(m), 1567(m), 1539(m), 1413(vs), 1175(w), 1069(w),

1025(w), 717(m), 675(w), 644(m), 461(m).

Method B. A solution of FeCl3-6H20 (0.20 g, 0.74 mmol) and NaO2CPh (0.21 g, 1.48

mmol) in MeCN (15 mL) was treated with dmemH (0.06 mL, 0.37 mmol) and stirred for 3

hours. The resultant red brown solution was filtered to remove NaC1, and the filtrate was left

undisturbed at room temperature for slow evaporation. Orange crystals slowly formed over 5

days in 30% yield; the product was identified by IR spectral comparison with material from

Method A.

Method C. A solution of (NEt4)2(Fe20Cl6) (0.20 g, 0.33 mmol) and NaO2CPh (0. 14 g,

0.99 mmol) in MeCN (15 mL) was treated with dmemH (0. 11 mL, 0.66 mmol) and stirred for

few hours. The resultant red-brown solution was filtered and kept undisturbed at room

temperature for slow evaporation. Orange crystals slowly formed over 3 days in 40% yield; the

product was identified by IR spectral comparison with material from Method A.

[Fe7O4(O2 Me)11(dmem)2] (4-2). Method A. A solution of FeCl3-6H20 (0.20 g, 0.74

mmol) and NaO2CMe-3H20 (0.25 g, 1.85 mmol) in MeCN (15 mL) was treated with dmemH

(0.06 mL, 0.37 mmol) and stirred for 3 hours. The resultant red-brown solution was filtered to

remove NaC1, and the filtrate was left undisturbed at room temperature for slow evaporation. X-

ray quality, dark orange crystals appeared over 20 days in 15% yield. These were collected by

filtration, washed with MeCN, and dried under vacuum. Anal. Called (Found) for 4-2-2MeCN

(C40H73N6Fe7O28): C, 32.53 (32.66); H, 4.98 (5.26); N, 5.69 (5.46). Selected IR data (cm )~:









3431(br), 2985(w), 2875(w), 1565(vs), 1426(vs), 1088(w), 1052(w), 1033(w), 886(w), 709(w),

668(m), 637(m), 615(m), 539(m), 487(m).

Method B. An orange-red solution of [Fe30(O2CMe)6(H20)3](NO3) (0.20 g, 0.03 mmol)

in MeCN (15 mL) was treated with dmemH (0.10 mL, 0.06 mmol) and the solution stirred

overnight at room temperature. It was then filtered and the filtrate allowed to stand undisturbed

at room temperature. Orange crystals of the product formed over 25 days in 10% yield; the

product was identified by IR spectral comparison with material from Method A.

[Fe6O2(OH)4(O2CCBu')s(dmem)2] (4-3). A solution of dmemH (0.03 ml, 0. 19 mmol) in

MeCN (5 mL) was treated with pyridine (15 CLL, 0.19 mmol), followed by the addition of a

solution of [Fe30(O2CBut)6(H20)3](NO3) (0.18 g, 0.19 mmol) in MeCN (12 mL). The resultant

solution was filtered and the filtrate left undisturbed at room temperature. X-ray quality orange

needles of 4-3-2MeCN grew over 10 days in 20 % yield. These were collected by filtration,

washed with MeCN, and dried under vacuum. Dried solid analyzed as solvent-free. Anal. called

(Found) for 4-3 (C54HlloN4Fe6024): C, 42.27 (42.53); H, 7.23 (7.40); N, 3.65 (3.68). Selected IR

data (cm )~: 2960(m), 2925(w), 2866(w), 1558(vs), 1484(s), 1427(vs), 1376(w), 1332(w),

1228(m), 1073(w), 903(w), 787(w), 662(m), 608(m), 530(m), 427(m).

[Fe30(O2CBut)2 N3)3(dmlem1)2] (4-4). A solution of [Fe30(O2CBut)6(H20)3](NO3) (0.10 g,

0.11 mmol) in EtOH (15 mL) was treated with dmemH (34 CLL, 0.20 mmol) and solid sodium

azide (0.03 g, 0.46 mmol), and then stirred overnight at room temperature to give an orange

precipitate. The solid was collected by filtration, washed with a little EtOH. It was recrystallized

from a CH2C 2/hexanes layering to give X-ray quality orange crystals of 4-4-CH2C 2 OVer 3 days

in 25 % yield. Anal called (Found) for 4-4-V2CH2C 2 (C24.5H53N13Fe307C1): C, 34.83 (34.77); H,

6.32 (6.30); N, 21.55 (21.16). Selected IR data (cm )~: 3390(br), 2959(w), 2870(w), 2066(vs),









1543(m), 1480(w), 1418(m), 1342(w), 1225(w), 1087(m), 986(w), 720(m), 633(w), 606(w),

429(w).

4.2.2 X-ray Crystallography

Data were collected by Dr. Khalil A. Abboud on a Siemens SMART PLATFORM

equipped with a CCD area detector and a graphite monochromator utilizing Mo-Ku radiation (h

= 0.71073 A+). Suitable crystals of 4-1-4MeCN, 4-2-MeCN, 4-3-2MeCN, and 4-4-CHzC2 2 Of

attached to glass fibers using silicone grease and transferred to a goniostat where they were

cooled to 173 K for data collection. Cell parameters were refined using up to 8192 reflections. A

full sphere of data (1850 frames) was collected using the co-scan method (0.3o frame width). The

first 50 frames were remeasured at the end of data collection to monitor instrument and crystal

stability (maximum correction on I was < 1 %). Absorption corrections by integration were

applied based on measured indexed crystal faces. The structure was solved by the Direct

Methods in SHELXTL6, 46 and refined using full-matrix least squares. The non-HI atoms were

treated anisotropically, whereas the hydrogen atoms were calculated in ideal positions and were

riding on their respective carbon atoms. Refinement was done using F2

In 4-1-4MeCN, the asymmetric unit consists of half the Fe7 cluster and two MeCN

molecules of crystallization. A total of 644 parameters were refined in the final cycle of

refinement using 32986reflections with I > 20(1 to yield R1 and wR2 Of 4.57 and 8.99 %,

respectively. In 4-2-MeCN, a total of 721 parameters were refined in the final cycle of

refinement using 18637 reflections with I > 20(I to yield R1 and wR2 Of 3.53 and 8.64 %,

respectively. In 4-3-2MeCN, a total of 924 parameters were refined in the final cycle of

refinement using 17808 reflections with I > 20(I to yield R1 and wR2 Of 4. 15 and 10.0 %,

respectively. In 4-4-CH2C 2, the azide ligand at N11 was disordered and it was refined in two

positions with the site occupation factors dependently refined. A total of 472 parameters were









refined in the final cycle of refinement using 9205 reflections with I > 20(1 to yield R1 and wR2

of 4.97 and 10.23 %, respectively. Unit cell data and details of the structure refinements for the

four complexes are listed in Table 4-1.

4.3 Results and Discussion

4.3.1 Syntheses

Many synthetic procedures to polynuclear iron clusters rely on the reaction of

[Fe30(O2CR)6(H20)3]' Species with a potentially chelating ligand, and this was one of the

procedures chosen in the present work. In such reactions, the [Fe30]7 core of the trinuclear iron

complex serves as a building block for higher nuclearity species, but the exact nuclearity and

structure of the product depends on several factors; in the present work, we have found that the

identity of the carboxylate group is one of these.

Reaction of [Fe30(O2CPh)6(H20)3](NO3) with 1 3 equiv of dmemH in MeCN gave the

heptanuclear complex [FeO4(O2CPh)ll(dmem)2] (4-1) with a core topology not previously

encountered (eq. 4-1) The same product was obtained from an EtOH reaction solvent, and also

from the treatment of an MeCN solution of FeCl3-6H20 with sodium benzoate and dmemH in a

2:4: 1 ratio. Increasing the amount of sodium benzoate or dmemH still gave complex 4-1, but the

reaction was not so clean. Reactions in which the MeCN was replaced by EtOH, and the

FeCl3-6H20 by Fe(CIO4)3-xH20 or (NEt4)2(Fe20Cl6), alSo gave the same product, for Fe:dmemH

ratios of both 1:1 and 1:2. Clearly, complex 4-1 is a preferred product of these components and

particular carboxylate group.

7[Fe30(O2CPh)6(H20)3]' + 2dmemH + 4e 3[FeO4(O2CPh)ll(dmem)2] + 9PhCO2~

16H20 + 12H+ (4-1)

If the carboxylate employed was acetate instead of benzoate, then the product from the

FeCl3/MeCO2Na/dmemH (2:5:1) reaction system in MeCN (Method A of Experimental Section)









was another heptanuclear complex, [FeO4(O2CMe)ll(dmem)2] (4-2) (eq. 4-2). Its formula is the

same as that of 4-1, except for the carboxylate identity, but structurally the two complexes are

very different videe infra). The same product 4-2 was obtained using [Fe30(O2CMe)6(H20)3]+ aS

the starting material in a reaction with two equiv. of dmemH in MeCN (Method B). The yields of

4-2 were much lower than 4-1, although they could be improved somewhat by addition of some

NEt3 base to the reaction.

7FeCl3 + 11MeCO2- + 2dmemH + 4H20 [FeO4(O2CMe)ll(dmem)2] + 21Cl- + 10H+ (4-2)

In contrast to the heptanuclear products from the use of benzoate and acetate reagents, the

use of pivalate ones led to a hexanuclear product. Treatment of [Fe30(O2CBut)6(H20)3](NO3)

with dmemH in MeCN led to subsequent isolation of [Fe602(OH)4(O2CBur~s~dmem)2] (4-3) (eq.

4-3). The addition of one equivalent of NEt3 Or pyridine as base improves the yield from 10 to

20 %. The same product is obtained on increasing the amount of dmemH to three equivalents.

2[Fe30(O2CBut)6(H20) 3] + 2dmemH -[Fe602(OH)4(O2CBut)s(dmem)2] + 4ButCO2~

6H1 + 2H20 (4-3)

It is clear that the reactions that lead to 4-1 to 4-3 are very complicated, and the reaction

solutions likely contain a complicated mixture of several species in equilibrium. In such cases,

factors such as relative solubility, lattice energies, crystallization kinetics, and others determine

the identity of the isolated products, and one (or more) of these factors is undoubtedly the reason

that the reaction product is so dependent on the exact carboxylate employed.

Since complex 4-3 contains bridging hydroxide groups, a similar reaction was explored in

the presence of sodium azide. Perlepes and coworkers have demonstrated that replacement of

bridging hydroxide groups (which almost always mediate antiferromagnetic exchange

interactions) with end-on bridging azide groups (which mediate ferromagnetic exchange) in









cobalt, nickel and iron clusters leads to products with much higher ground state spin values.58-6

Thus, we explored a variety of reaction conditions differing in the azide amount and/or solvent,

and it was found that reaction of [Fe30(O2CBut)6(H20)3](NO3), dmemH and azide in a 1:2:4

ratio gave the new trinuclear complex [Fe30(O2CBut)2(N3)3(dmem)2] (4-4) (eq. 4-4). The

complex has its azide groups all in terminal sites, but it nevertheless has an interesting core

structure. Complex 4-4 was also obtained in lower yield from the reaction of preformed complex

4-3 with four equivalents of sodium azide in EtOH.

[Fe30(O2CBut)6(H20)3] + 2dmemH + 3N3 [Fe30(O2CBut)2(N3)3(dmem)2] + 3H20 +

4ButCO2- + 2H+ (4-4)

4.3.2 Description of Structures

4.3.2.1 Structure of [FeO4(O2CPh)nl(dmem)2] (4-1)

A labeled representation of complex 4-1 is shown in Figure 4-2(a). Selected interatomic

distances and angles are summarized in Table A-6. Complex 4-1-4MeCN crystallizes in the

monoclinic space group C2/c with the Fe7 molecule lying on a crystallographic C2 aXiS passing

through the central Fe4 atom. The core can be described as two [Fe4 u3-O)2] planar-butterfly

units fused at body atom Fe4, one butterfly unit being atoms Fel', Fe2, Fe3, Fe4, 09, 010'.

Further, each butterfly unit can be considered as two edge-sharing Fe30 triangular units, with the

pu3-O2- bridging atoms 09 and 010 slightly above and below their Fe3 planOS. These O atoms

bridge somewhat asymmetrically; the bonds to the wingtip Fe atoms (Fel---O0, 1.828 A+ and

Fe3---9, 1.844 A+) are shorter than the bonds to the body Fe atoms (Fe2---O0', 1.941 A+ and

Fe2---9, 1.923 A+). The two dmem- groups bind as tridentate chelates to Fel and its symmetry

partner Fel', with their alkoxide O atoms bridging wingtip atom Fel in one Fe4 unit with body

atom Fe2 in the other. The remaining peripheral ligation about the [FeO4] COre is provided by









eleven benzoate groups, nine in their common rl : rl : p bridging mode and the other two in the

rare r12 chelating mode on Fe3 and Fe3'.

4.3.2.2 Structure of [Fe7O4(O2 Me)11(dmem)2] (4-2)

A labeled representation of complex 4-2 is provided in Figure 4-2(b). Selected interatomic

distances and angles are given in Table A-7. Complex 4-2-MeCN crystallizes in the triclinic

space group Pi. The molecule contains a remarkable [Fe7Cp3-O)4] COre. It can be described as

consisting of a central [Fe303] Ting COntaining Fe2, Fe3 and Fe5, with each of the doubly-

bridging 02- ions of this hexagon becoming Cl3 by also bridging to a third, external Fe atom (Fel,

Fe4, Fe6). The fourth 02- ion bridges ring atom Fe5, Fe6 and a seventh Fe atom (Fe7) on the

periphery of the molecule. The two dmem- groups bind one each to the external atoms Fel and

Fe4 in a tridentate chelating manner with their alkoxide O atoms also bridging to ring atoms Fe2

and Fe3, respectively. Peripheral ligation is completed by eleven acetate groups, ten in rl : r : pu-

bridging modes and one r12 chelating to Fe7.

The molecular structures of 4-1 and 4-2 can be said to represent two different ways of

linking a number of Fe30 triangular units, as is clear in Figure 4-3 where the cores of 4-1 to 4-3

are compared. The core topologies of complexes 4-1 and 4-2 are unprecedented within Fe"

chemistry. Indeed, there are only a few Fe7 complexes in the literature, and they are all mixed-

valent except for the one reported by Winpenny and coworkers containing phenylphosphonate

ligand and Zheng and coworkers containing cyclohexenephosphonate ligand.98,99 In addition,

very recently few disklike and domelike heptanuclear Fe"' clusters were also reported,100-102 but

they are significantly different from the ones obtained with dmemH. Complexes 4-1 and 4-2 are

thus the novel heptanuclear Fe"' carboxylate complexes.









4.3.2.3 Structure of [Fe6O2(OH)4(O2CBut~s(dmlem1)2] (4-3)

A labeled representation of complex 4-3 is shown in Figure 4-4(a). Selected interatomic

distances and angles are given in Table A-8. Complex 4-3-2MeCN crystallizes in the triclinic

space group Pi with the asymmetric unit containing two independent Fe6 ClUSters, both lying on

inversion centers; since the two molecules are essentially superimposable, we show and discuss

the structure of only one of them here. The core consists of an [Fe4 u3-O)2] unit (Fel, Fel', Fe2

and Fe2'] on either side of which is attached a [Fe~p-OH)2(p-OR)] unit containing Fe3; the OH-

ions are 09 and 010 on one side, and their symmetry partners on the other. One OH- bridge

(010) connects Fe3 to central Fel whereas the other (09) connects to Fe2. The OH- nature of 09

and 010 was confirmed by BVS calculations,103 which gave values of 1.14 for 09 and 1.05 for

010. Peripheral ligation is provided by two dmem- and eight pivalate groups. There are three

types of pivalate binding modes: four are in the common rl : rl : p bridging mode, two are in the

rare r12 chelating mode, and the remaining two are in an rl terminal mode.

A number of other Fe6 COmplexes have been previously reported possessing a variety of

metal topologies, such as planar, twisted boat, chair, parallel triangles, octahedral, ladder-like,

cyclic, etc.104 However, the only previous compounds structurally similar to 4-3 are

[Fe602(OMe)12(tren)z2 2 and [Fe602(OR)s(O2CPh)6] -105,106 Both of the latter complexes contain a

central [Fe4 u3-O)2 8+ COre with an additional Fe atom on each side, as in 4-3, but the precise

means of attachment are different.

4.3.2.4 Structure of [Fe30(O2CBut)2 N3)3(dmem1)2] (4-4)

A labeled representation of 4-4 is provided in Figure 4-4(b). Selected interatomic distances

and angles are given in Table A-9. Complex 4-4-CH2~l 2CryStallizes in the monoclinic space

group P21/n. The structure consists of an Fe3 isosceles triangle bridged by a pu3-O2- ion (Ol) with

a rare T-shaped geometry, rather than the common trigonal planar geometry usually seen in









triangular metal carboxylates.107 The Fel---Fe2 and Fe2---Fe3 edges are each additionally bridged

by an alkoxide O atom of the dmem- ligand and a rl : rl : p pivalate group. As a result, Fe2---Fel

(2.997(1) A+) and Fe2---Fe3 (2.980(1) A+) are much shorter than Fel---Fe3 (3.694(2) A+). Similarly,

Fe2-01 (2.070(19) A+) is noticeably longer than Fel-01 (1.872(19) A+) and Fe3-01 (1.865(19)

A+). Fel, Fe2, Fe3 and 01 are co-planar, and 02 and 03 are slightly above and below this plane.

A chelating dmem- and a terminal azide on each Fe atom complete the ligation at the metal

atoms, which are all near-octahedral. The overall asymmetric Fe30 is with little precedent in iron

chemistry, the only previous discrete example being [Fe30(TIEO)2(O2CPh)2 13], where TIEO is

1 1,2-tri s(1 -methylimi dazol -2-yl)ethoxi de.73,108

4.3.3 Magnetochemistry of Complexes 4-1 to 4-4

4.3.3.1 De Studies

Solid-state, variable-temperature dc magnetic susceptibility data in a 0. 1 T field and in the

5.0-300 K range were collected on powdered crystalline samples of 4-1 to 4-4 restrained in

eicosane. The obtained data are plotted as 3&Tys Tin Figure 4-5. For 4-1-M2MeCN, 3&T steadily

decreases from 6.95 cm3Kmol-l at 300 K to 4.07 cm3Kmol-l at 5.0 K. The 300 K value is much

less than the spin-only (g = 2.0) value of 30.62 cm3Kmoll for seven non-interacting Fe"' ions,

indicating the presence of strong antiferromagnetic interactions. The 5.0 K value suggests an S =

5/2 ground state. For 4-2-2MeCN, 3&T steadily decreases from 8. 19 cm3Kmol-l at 300 K to 4. 14

cm3Kmol-l at 34 K, stays essentially constant until 10 K, and then decreases slightly to 3.85

cm3Kmol-l at 5.0 K. As for 4-1-M2MeCN, this behavior is indicative of antiferromagnetic

interactions and an S = 5/2 ground state. For 4-3, 3&T increases from 9.73 cm3Kmol-l at 300 K to

a maximum of 14. 10 cm3Kmol-l at 20K and then drops to 12.92 at 5.0 K. The 300 K value is

again much less than the spin-only value of 26.25 cm3Kmoll expected for six non-interacting

Fe"' ions, indicating predominantly antiferromagnetic interactions. The increase in &T as the









temperature then decreases suggests the lowest lying states are of high spin values, and the

maximum at 20 K of 14.10 cm3Kmoll is very close to the spin-only value of 15.00 cm3Kmoll

for an S = 5 ground state. The decrease in &T at the lowest temperatures is very likely due to

zero-field splitting (ZFS) within the S = 5 ground state and perhaps some weak intermolecular

interactions. For 4-4-M2CH2 12, 3nT Steadily decreases from 5.74 cm3Kmol-l at 300 K to 4.07

cm3Kmol-l at 50 K, and then stays approximately constant until 15 K, below which it decreases

slightly to 3.75 cm3Kmol-l at 5.0 K. The latter value suggests an S = 5/2 ground state.

To confirm the above ground state spin estimates, variable-field (H) and -temperature

magnetization (M)1 data were collected in the 0.1 to 7.0 T and 1.8-10 K ranges. The resulting data

for 4-1-M2MeCN are plotted in Figure 4-6 as reduced magnetization (M/NpuB) VS. H/T, where Nis

Avogadro's number and puB is the Bohr magneton. The saturation value at the highest fields and

lowest temperatures is ~4.8, as expected for an S = 5/2 and g slightly less than 2; the saturation

value should be gS. The data were fit, using the programM4~rGNE753 described elsewhere.54-5

The best-fit for 4-1-M2MeCN is shown as the solid lines in Figure 4-6, and was obtained with S =

5/2 and either of two sets of parameters, g = 1.94 and D = -0.56 cm- or g = 1.95 and D = 0.77

cm- Alternative fits with S = 3/2 or 7/2 were rej ected because they gave unreasonable values of

g and D. It should be noted that it is common to obtain two acceptable fits of magnetization data

for a given S value, one with D > 0 and the other with D < 0. This was indeed the case for the

magnetization fits for all of the complexes 4-1 to 4-4 in this work. In order to assess which is the

superior fit in these cases, and also to ensure that the true global minimum had been located for

each compound, we calculated the root-mean-square error surface for the fits as a function of D

and g, and have plotted them as two-dimensional contour plots in Figure 4-7. For 4-1-M2MeCN,









the plot clearly shows only the above-mentioned minima with positive and negative D values,

with both fits being of comparable quality as shown in Figure 4-7.

For 4-2-2MeCN, the reduced magnetization plot saturates at ~4.5, again suggesting an S =

5/2 ground state. The fit, shown as the solid lines in Figure 4-8(a), gave S = 5/2 with either g =

1.91 and D = -0.76 cm- or g = 1.91 and D = 0.98 cm- The error surface contour plot is shown

in Figure 4-8(b) and shows the above minima, with the one with negative D clearly the superior

fit since it has a lower (deeper) minimum. Figure 4-8(b) also clearly shows that the fit minimum

is a soft one, consistent with a significant uncertainty in the precision of the obtained g and D fit

values, which we estimate as & 0.01 on g and a 5 10 % on D.

For 4-3, the reduced magnetization plot saturates at ~9.5, suggesting an S = 5 state with g <

2 (Figure 4-9(a)). A satisfactory fit could only be obtained if data collected at fields above 5 T

were excluded, suggesting that some low-lying excited states with S > 5 are being stabilized by

the applied field to the point that they are significantly populated at these temperatures. To avoid

this, the data at 6 and 7 T were excluded, and now a good fit was obtained (solid lines in Figure

4-9(a)) with S = 5 and either g = 1.95 and D = -0.28 cm- or g = 1.92 and D = 0.33 cm- The

error surface for the fit shows again that the fit with negative D is far superior (Figure 4-9(b)),

suggesting this is the true sign of D.

4.3.3.2 Rationalization of the Ground State Spin of 4-1 and 4-3

It is of interest to try to rationalize the observed ground state spin values of 4-1 to 4-3. It is

assumed that all Fe2 pairWISe exchange interactions are antiferromagnetic, as is essentially

always the case for high-spin Fe"', and there will thus be competing antiferromagnetic exchange

interactions and spin frustration effects within the many Fe3 triangular units in these complexes.

The ground state of 4-1 is the easiest to rationalize: the discrete Fe4 butterfly (planar or bent

rhombus) topology is known to usually give an S = 0 ground state as a result of the









antiferromagnetic interactions along the four outer edges overcoming, and thus frustrating, the

diagonal interaction.77,109-1 11 In 4-1, two such Fe4 units are fused at body (central) Fe4 of the two

butterfly units, and assuming the same spin alignments as in the discrete Fe4 mOlecules, then the

ground state spin alignments are predicted to be those shown in Figure 4-10(a), giving an S = 5/2

ground state, as observed experimentally. It is worth mentioning here that disklike and domelike

heptanuclear amino alkoxo clusters reported in literature have also been shown to possess S = 5/2

ground state spin.101102

The ground state for 4-2 is not so easy to rationalize convincingly because of its high

content of triangular units. For 4-3, the recognizable Fe4 unit as in 4-1 suggests that the spin of

this sub-unit is zero, and then the two Fe atoms Fe3 and Fe3' above and below would have their

spins parallel to each other by both being antiparallel to the spins of Fel and Fel' as shown in

Figure 4-10(b). This would thus rationalize an overall S = 5 ground state for 4-3.

For 4-4-V2CH2C 2, the reduced magnetization saturates at ~4.7, suggesting an S= 5/2

ground state and g < 2 (Figure 4-11(a)). The fit of the data (solid lines in Figure 4-11l(a)) gave S

= 5/2 with either g = 1.92 and D = -0.69 cm- or g = 1.92 and D = 0.82 cm- The D vs. g error

surface (Figure 4-11(b)) shows that the fit with negative D is again superior suggesting this may

be the true sign of D. Since complex 4-4 is only trinuclear, we determined its pairwise Fe2

exchange interactions by fitting the variable temperature susceptibility data to the appropriate

theoretical expression.

4.3.3.3 Determination of the Exchange Interactions in 4-4

The Heisenberg spin Hamiltonian describing the isotropic exchange interactions within an

isosceles Fe3 triangle of Ct, symmetry (Figure 4-12(a)) is given by eq. 4-5, where J, refers to the

interactions between Fe2---Fe3 and Fel---Fe2, and Jb refers to the Fe3---Fel interaction; S,refers









to the spin of atom Fei. The energies of the resultant total spin states ST, which are eigenfunctions

of the Hamiltonian in this coupling scheme, are given by eq. 4-6, where & = S; + S3. The overall

multiplicity of the spin system is 216 made up of 27 individual spin states ranging from ST = 1/2

to 15/2.

K = -2a(2 ~3 2 1) Wb ~3 1) (4-5)

E|ST, SA > = JL[Sr(Sy+1) SA(S+1)] Jb[L(SA+1)] (4-6)

An expression for the molar paramagnetic susceptibility was derived for this complex

using the Van Vleck equation." This was then used to fit the experimental y;Tys T data, with fit

parameters J,, Jb and an isotropic g value (see Appendix D-3). The fit is shown as the solid line

in Figure 4-5, which gave J, = 3.6 cm l, Jb = 45.9 cml and g = 1.93. These values identify the

ground state as the | ST, SA > = | 5/2, O > state shown in Figure 4-12(b), which is in agreement

with the reduced magnetization fit.

The marked inequality in the exchange constants, I A >> I 4 is as expected on the basis

of the iron-oxo bond lengths, where Fe3-01 = Fel-01 < Fe2-Ol. A similar situation was also

observed in the previous Fe3 COmplex with a similar core, [Fe30(TIEO)2(O2CPh)2 13], for which

J,= -8(4) cm-l and Jb = -55(6) cm- It has been established that the magnitude of the exchange

coupling constant J for an oxo-bridged Fe" 2 unit can be approximately correlated with a single

structural parameter P by the equation in eq. 4-7, if the Fe-O-Fe does not alter too much. In

this relationship, A = 8.763 x 1011 B = -12.663 and P is the shortest superexchange pathway.112

Applying this relationship to complex 4-4 gives J,= -12.9 cm-l and Jb = 46.4 cm- which are in

reasonable overall agreement with the experimental values obtained from fitting the

susceptibility data, and given that an angular dependence is of lesser importance than the radial

one, and is ignored by eq. 4-7. In particular, the acute values of angles of Fe2-O2-Fe3 and Fe2-









03-Fel (96.07(8) and 96.77(8) A+, respectively), which lead to the weak J, coupling, are

significantly smaller than those found in dinuclear Fe"' 2COmplexes on which the relationship of

eq. 4-7 was based and probably reflect a greater angular dependence. The value of A is stronger

than the magnitude of the antiferromagnetic coupling constant found for the triangular Fe"'

carboxylate complexes with an approximately equilateral [Fe30]"+ core (~30cm )~,113 but weaker

than the 80-130 cm-l values observed for the [Fe20]4+ and [Fe20(O2CR)2 2+ dinuclear cores.114-



-I = A eBP (4-7)

None of the compounds 4-1 to 4-4 exhibited an out-of-phase AC magnetic susceptibility

signal down to 1.8 K in an AC Hield of 3.5 Oe oscillating with frequencies up to 997 Hz. This

indicates that they do not exhibit a large enough barrier (vs kT) to exhibit the characteristic

signature of slow magnetization relaxation characteristic of single-molecule magnets (SMMs), at

least down to 1.8 K.

As discussed above, fits of variable-temperature and variable-Hield magnetization data are

not the most reliable way to obtain the most precise and accurate values of D, or its sign. The

magnetization fits suggested D to be negative for 4-2 and 4-3, but they could not suggest the sign

ofD for 4-1. Since the sign and magnitude ofD are crucial to the potential ability of a complex

to function as a SMM, we desired to better characterize D for these new and relatively rare

examples of Fex clusters with significant ground state spin values. The perfect technique for this

is high-frequency electron paramagnetic resonance spectroscopy.

4.3.4 High-Frequency EPR Spectroscopy

A detailed single-crystal study of representative complexes 4-1-4MeCN and 4-3-2MeCN

has been carried out by HFEPR spectroscopy. The main overall obj ective was to measure the

ZFS parameters in the spin Hamiltonian of eq. 4-8, which is the same as that in eq. 3-4 except









that it also now includes the rhombic ZFS term, E(12 2), where E is the rhombic ZFS

parameter, and Sx and S, are the x and y components of the total spin operator S. EPR is a high

resolution spectroscopic technique that can be used to investigate the more complete spin

Hamiltonian of eq. 4-8, whereas fits of bulk magnetization data are essentially insensitive to

inclusion of the rhombic E term.

K = Di2 + E(1x2 2) + ~B 09-H (4-8)

Single-axis angle dependence studies were first performed to roughly determine the

orientation of each crystal in the magnetic field. Both complexes 4-1-4MeCN and 4-3-2MeCN

possess low symmetry structures. Thus, determining the precise symmetry directions represents a

highly complex task requiring detailed two-axis rotation studies. However, one can readily

obtain basic information from single-axis studies; in particular, the sign of D, which is the crucial

factor in whether a particular complex is a SMM.100,117

Figure 4-13(a) displays the angle-dependence of the field positions of the strongest EPR

transitions determined from field-swept spectra recorded at 116 GHz and 1.4 K for complex 4-3

-2MeCN; given the low temperature, these data points must correspond to transitions from the

lowest-lying ms levels. Two series of resonances are observed (black and red data points) which

shift significantly upon rotating the field, thus providing the clearest evidence for a significant

magnetoanisotropy. Both series exhibit 180o periodicity, with virtually identical amplitudes. The

source of the two series has a natural explanation for complex 4-3-2MeCN for which there are

two differently oriented molecules in the unit cell. Thus, one naturally expects two distinct EPR

signatures, one from each species. The solid curves represent phenomenological fits to the two

sets of data, and are intended to capture the qualitative nature of the angle dependence. The

phase shift between the two data sets is 75+2o









In order to determine the sign of D, frequency- and temperature-dependent data were

collected on complex 4-3-2MeCN with the field oriented along one of the minima in Figure 4-

13(a) (1910). Figure 4-13(b) displays the frequency dependence of the angle-dependent peak

from Figure 4-13(a), and the inset displays representative spectra taken at higher temperatures. A

remarkable feature of the frequency-dependent data is that all peaks lie on a straight line which

extrapolates to a finite frequency on the vertical axis, i.e. there is no evidence for curvature in the

data. Assuming |DS| 1.5 cml (from reduced magnetization measurements), one realizes that at

least a 3 T magnetic field would be required to overcome the axial term in eq. 4-8. This suggests

that the Zeeman interaction commutes with the dominant axial term in eq. 4-8 across the entire

range of fields for which data were collected (0.6 to 2 T). In other words, the minima in Figure 4-

13(a) and the data in Figure 4-13(b) correspond to field orientations parallel to, or very close to

the z axes of the two species. This is quite coincidental, as the sample orientation was not

previously known.

Figure 4-14(a) displays a simulation of the Zeeman diagram for a SMM with S = 5, i.e.

with D < 0. As can clearly be seen, the transition from the lowest-lying ms level occurs at the

lowest field; the excited state transitions all occur at higher field. This agrees qualitatively with

the data in Figure 4-13(b). Therefore, we can conclude that D is negative, and that 4-3 is a SMM.

The intercept on the frequency axis in Figure 4-13(b) (66.4 GHz) then corresponds to the ZFS

between the ground and first excited state. If one assumes that S = 5, then D = -0.25(1) cm l,

which is in reasonable agreement with the value from the magnetization fits (D = -0.28(3) cm )~.

Because of the uncertainty in the precise orientation of the field relative to the easy-axis, we

cannot quote a precise value for g; the main purpose of the HFEPR measurements was to

unambiguously determine the sign of D, which was successfully achieved.










Single-axis rotation experiments for complex 4-1-4MeCN were not able to locate the axial

direction (presumably the rotation plane was inclined significantly with respect to the magnetic

:-axis of the molecule). Nevertheless, we were able to locate the plane perpendicular to the axial

direction (xy plane) from measurements similar to those shown in Figure 4-13(a). Thus, all of the

temperature- and frequency- dependent studies were carried out with the Hield aligned within the

magnetic xy plane of the Fe7 molecule. Only a single molecular species was anticipated for

complex 4-1-4MeCN, making interpretation of the data more straightforward. Furthermore, this

complex exhibits sharper EPR peaks, as evident from Figure 4-15(a), which shows the high-Hield

xy-plane spectra obtained at different temperatures and a frequency of 197 GHz. Comparison of

the data in Figure 4-14(a) with the simulated Zeeman diagram in Figure 4-13(b) reveals that

complex 4-1 cannot be a SMM because its D value is positive. As can be seen from Figure 4-

14(b), upon reducing the temperature, the stronger EPR peaks should be observed at the lowest

Shields for an easy-plane magnet (D > 0) when the Hield is applied parallel to the easy (xy) plane;

this is exactly what is seen in the data. If 4-1-4MeCN were a SMM, the intensities of the Hyve

transitions (labeled a to e in the Figure 4-15(a)) would be reversed.

Figure 4-15(b) displays the results of a multi-frequency study for complex 4-1-4MeCN,

with the Hield applied within the easy plane; the temperature was 20 K. Fits (solid curves) to the

positions of the EPR peaks were performed via exact diagonalization of eq. 4-8. It is very clear

from Figure 4-15(b) that the data lie on a series of lines that are not evenly spaced, and exhibit

significant curvature at low frequencies and fields. These trends are a characteristic of xy-plane

spectra obtained for a system with a significant uniaxial anisotropy (both positive and negative

D), due to the competition between the orthogonal Zeeman and ZFS (DSz2) interactions. In other

words, the data displayed in Figure 4-15(b) provide further confirmation that the field is in the









xy-plane and, when combined with the temperature dependence in Figure 4-15(a), also confirm

the positive sign of D. The fit assumes an S = 5/2 ground state, and yields g = 2.0 and D = +0.62

cm- This value again agrees reasonably well with that from the reduced magnetization studies

(D = +0.77(7) cm- ). The low value of g obtained from the reduced magnetization fits can be

explained as the limitation of the fitting program Magnet which assumes axial anisotropy;

therefore HFEPR data is more reliable. The best fit to the data required the inclusion of a

rhombic ZFS anisotropy, |E| > 0.067 cm- This is not unexpected, given the low symmetry of the

molecule. Our estimate of E represents a lower bound, as the orientation of the field within the

easy plane was not known. Low-temperature EPR measurements on domelike Fe7 cluster yield a

D value of 0.28 cm l,lot which is considerably lower than calculated for 4-1-4MeCN. This can be

attributed to their different structural arrangements leading to the differences in single-ion

anisotropy and spin-spin anisotropy.

4.4 Conclusions

The tridentate N, N, O ligand dmem- has proven to be a very fruitful new route to a variety

of new Fe"' clusters comprising two Fe7 and one Fe6 Species, depending on the identity of the

carboxylate employed. The latter point emphasizes the exquisite sensitivity of the reaction

product on a variety of reaction conditions and reagents employed. For example, even though

complexes 4-1 and 4-2 have the same formula except for the identity of the carboxylate, the

structures of the two complexes are very different. It was interesting that the azide ligands in 4-4

were only terminal rather than bridging, but yet nevertheless fostered formation of a product very

different from that of the non-azide product 4-3.

Fitting of the reduced magnetization vs H T data established that 4-1, 4-2 and 4-4 each

possess an S= 5/2 ground state spin, whereas 4-3 has an S= 5 ground state. The complexes all

serve to clearly emphasize again how ground state spin values of significant magnitude can result









from spin frustration effects even though all the pairwise exchange interaction constants are

antiferromagnetic. The magnetization fits of 4-1 to 4-4 serve to emphasize, however, the

difficulty of determining the sign of D for Fe"' clusters from such measurements, making it thus

difficult to predict whether a given cluster might be a new example of a SMM. Representative

complexes 4-1 and 4-3 were therefore studied by HFEPR spectroscopy, a tremendously powerful

and sensitive technique, not least for obtaining accurate and precise values for spin Hamiltonian

parameters such as D, including an unequivocal determination of its sign. From these

measurements, we concluded that complex 4-3 has D < 0 and thus is a potential SMM, whereas

complex 4-1 has D > 0 and is not. In fact, none of the compounds 4-1 to 4-4 exhibited an out-of-

phase ac magnetic susceptibility signal down to 1.8 K in ac frequencies up to 997Hz. Even for 4-

3, which was confirmed by HFEPR spectroscopy to have a negative D value, its S = 5 and D = -

0.25 cm-l gives a barrier (U) to magnetization relaxation with an upper value of U= S2|D| = 6.3

cml (= 9.0 K). Remembering that the true or effective barrier (Ugf) is less that U due to quantum

tunneling of the magnetization (QTM) through the barrier, it is not surprising that no sign of

slow relaxation is seen at temperatures above 1.8 K. Studies significantly below 1 K will be

required in order to better investigate the potential SMM behavior. Nevertheless, the present

work does establish interesting new examples ofFex clusters with significant ground state S

values and negative D values.

Finally, the preparation of complexes 4-1 to 4-4 again serves to emphasize the utility of

alkoxide-containing chelates in polynuclear metal cluster chemistry.









Table 4-1. Crystallouraphic data for 4-1-4MeCN, 4-2-MeCN, 4-3-2MeCN and 4-4-CH2C 2


4-1
C99Hlo1Fe7NsO28
2241.83
C2/c
18.6028(14)
26.8523(14)
20.8083(13)
90
103.879(2)
90
10090.9(11)


4-2
C38H70Fe7N5O28
1435.94
Pi
12.4586(8)
13.5495(9)
18.690(12)
70.636(2)
79.731(2)
73.099(2)
2836.2(3)
2


4-3
C58H116Fe6N6024
1616.67
Pi
12.9769(10)
14.4142(11)
23.9082(18)
87.6240(10)
88.5620(10)
66.0920(10)
4084.7(5)


4-4
C25H54C 2Fe3N1307
887.26
P2 l/n
12.3260(8)
25.3961(17)
13.1400(9)
90
99.1490(10)
90
4060.9(5)


Formula
Fw, g/mola
Space group
a, A
b, A
c, A
a, a
P, a
Y, a
V', A3


T, K 173(2) 173(2) 173(2) 173(2)
Radiation, Ab 0.71073 0.71073 0.71073 0.71073
Peaic, g/cm3 1.476 1.681 1.311 1.451
pu, mm-l 1.058 1.828 1.105 1.244
R1 c~d 0.0457 0.0463 0.0415 0.0497
wR2 e 0.0899 0.0927 0. 1009 0. 1023
aIncluding solvate molecules. bGraphite monochromator. cI> 2o(I). dR1 = C(||Fo| |8||l)l /LF'ol. ewR2 =
[C[w(Fo2 F, ) ] / C[w(Fo ) ]]'", w 1/l[o- (Fo ) + [(ap)) +bp)], where p = [max (Fo O) + 2F, ]/3.


NN"



HO
dmemH


OH HO

m daH,


Figure 4-1. Structure of ligands: mdaH2 and dmemH.




























Figure 4-2. (a) Labeled representation of the structure of 4-1. Hydrogen atoms and phenyl rings
(except for the ipso carbon atoms) have been omitted for clarity. The C2 Symmetry
axis is approximately vertical. Color code: Fe"', green; O, red; N, blue; C, grey. (b)
Labeled representation of the structure of 4-2. Hydrogen atoms have been omitted for
clarity. Color code: Fe"', green; O, red; C, grey; N, blue

(a) (b) C (c)












Figure 4-3. Comparison of cores of 4-1 (a), 4-2 (b), and 4-3 (c). Color code: Fe"', green; O, red

































































UI~
0 50 100 150 200 250 300
T/K


Figure 4-5. Plots of guTys Tfor complexes 4-1 (0), 4-2 (m), 4-3 (A) and 4-4 (+). The solid line
is the fit of the data for 4-4; see the text for the fit parameters.


Figure 4-4. (a) Labeled representation of the centrosymmetric structure of 4-3. Hydrogen atoms
and methyl groups on pivalate groups have been omitted for clarity. Color code: Fe"',
green; O, red; C, grey; N, blue. (b) Labeled representation of the structure of 4-4.
Hydrogen atoms have been omitted for clarity. Color code: Fe"', green; O, red; C,
grey; N, blue.


~~ m 4-2
A 4-3
S4-4
itn





































0 10 20 30 40 50

H/T(kGlK)


Figure 4-6. Plot of reduced magnetization (M/NpuB) VS H/T for complex 4-1-M2MeCN. The solid
lines are the fit of the data; see the text for the fit parameters.


1.0





0.5











-0.5


1.80


1.85 1.90 1.95 2.00 2.05


Figure 4-7. Two-dimensional contour plot of the fitting error surface vs D and g for 4-1*%2MeCN.
























































*1T
O 2T

* 4T
v 5T


1.0


0.5

E
8 0.0


-0.5


-1.0


5 10D 15

H/T(kG/K)


20 25 30


1.84 1.88 1.92 1.96

9


Figure 4-8. (a) Plot of reduced magnetization (M/NpuB) VS H/T for complex 4-2-2MeCN. The
solid lines are the fit of the data; see the text for the fit parameters. (b) Two-
dimensional contour plot of the fitting error surface vs D and g for 4-2-2MeCN.


0.5



E 0.0



-0.5


I-1.0 P>
20 25 30 1.80


0 5 10 15

H/T(kG/K)


1.85 1.90 1.95 2.00 2.05


Figure 4-9. (a) Plot of reduced magnetization (M/NpuB) VS H/T for complex 4-3. The solid lines
are the fit of the data; see the text for the fit parameters. (b) Two-dimensional contour
plot of the fitting error surface vs D and g for 4-3.






































Figure 4-10. Rationalization of the ground state spin of (a) 4-1 and (b) 4-3.


1.0



0.5



Y 0.0

-0.


e 1T
0 2T
n 3T
4 4T
1 BT
o BT
7T
-- Fiitfrg


-1.0 F
1.84


0 10 20 30 40

HTT(kG/K)


1.88 1.92 1.96 2.00


Figure 4-11i. (a) Plot of reduced magnetization (M/NpuB) VS H/T for 4-4-M2CH2 12. The solid lines
are the fit of the data; see the text for the fit parameters (b) Two-dimensional contour

plot of the fitting error surface vs D and g for 4-4-M2CH2 12.










Fe2


02




Fe3


03




Fe'l


Figure 4-12. (left) The core of 4-4 defining the pairwise exchange interactions. (right)
Rationalization of the ground state spin of 4-4.


~ 3.0
a
c~i 2.5
o
~ z.o


f= 116 GHz; T = 1.4 K
50 100 150 200 250 300

Angle (Dcgrces)


0.5 1.0 1.5

Magnetic Ficld (Tcsla)


Figure 4-13. (a) Plot of the HFEPR peak positions for 4-3-2MeCN obtained from angle-
dependent studies at 116 GHz and 1.4 K. (b) Frequency dependence for 4-3-2MeCN
with the field oriented along one of the minima in Figure 4-13(a) (1910); the inset
displays temperature-dependent spectra obtained at 106 GHz.

















-10


0-


-10 B//xy a

-S = 5/2, D > 0


Magnetic Field (Tesla)


I 2
Magnetic Field (Tesla)


6 7


Figure 4-14. (a) Simulated Zeeman diagram for a spin S = 5 system with D < 0 with the magnetic
field parallel to the : axis. The red lines (labeled a to d) correspond to the transitions
shown in the inset of Figure 4-13(b). (b) Simulated Zeeman diagram for a spin S =
5/2 system with D > 0 with the magnetic field parallel to the xy plane. The red lines
(labeled a to c) correspond to the transitions shown in Figure 4-15(a).


2oo
T = 20 K oa





S1006 on




0 2 4 6
Magnetic Field (Tesla)


56d78
Magnetic Field (Tesla)


Figure 4-15. (a) Temperature dependent spectra for 4-1-4MeCN at 197 GHz with the DC
magnetic field applied within the easy (xy) plane. (b) Easy-plane peak positions for
4-1-4MeCN plotted versus frequency at 20 K. The solid lines are simulations using
the ZFS parameters given in the main text









CHAPTER 5
A NEW N, N, O CHELATE FOR TRANSITION METAL CLUSTER CHEMISTRY: FeS AND
Fe6 CLUSTERS FROM THE USE OF 6-HYDROXYMETHYL-2, 2' BIPYRIDINE

5.1 Introduction

There continues to be a great interest by many groups around the world in the synthesis

and study of 3d transition metal cluster compounds, not least for the structural aesthetics

possessed by such species. Other reasons for this interest are varied. For Fe chemistry, for

example, there are bioinorganic areas of relevance such as the great desire to understand and

model the assembly of the polynuclear iron core of the iron storage protein ferritin.26,30,31 There

is also a materials interest in that high nuclearity iron compounds can sometimes exhibit unusual

and occasionally novel magnetic properties, with some of them even being examples of single-

molecule magnets (SMMs);13,20,82 the latter are molecules with a combination of a relatively

large ground state spin and a significant magnetoanisotropy of the easy-axis (Ising) type.20,118

That Fe"' is one area where high nuclearity species are often encountered is as expected from the

high charge-to-size ratio of this oxidation state and the resulting propensity to favor oxide-

bridged multinuclear products. Indeed, the formation of the Fe/O/OH core of ferritin that was

mentioned as a bioinorganic area of interest is merely an extreme example of such polynuclear

chemistry. As a result, many large Fe"' clusters have been reported to date with nuclearities up to

22.72,75,87,119

Although the exchange interactions between Fe"' centers are almost always

antiferromagnetic, certain Fex topologies can nevertheless possess large ground state spin values

as a result of spin frustration. The latter is here defined in its more general sense of competing

exchange interactions of comparable magnitude, preventing (frustrating) the preferred

antiparallel alignment of all spins, and thus giving larger ground state spin values than might be

expected.74-78 In Some cases, as mentioned above, the compounds can behave as SMMs. This is









the case for clusters such as [FesO2(OH)12z(tacn)6 8+, 81s,82,12 and [Fe4(OMe)6(dpm)6 79 foT

example.

For the above reasons and more, there is a continuing search for new synthetic methods

that can yield new polynuclear Fe/O clusters. One approach that has proven successful in this

regard is the use of alcohol-containing chelate groups that, on deprotonation, can provide

alkoxide groups that are excellent bridging units and thus foster formation of high nuclearity

products.74,86,88 In chapter 4, we reported the use of deprotonated 2- { [2-(dmethyrla~minnoethl]-

methylamino~ethanol (dlmemHT ; Fgre~ 51) asI a newr andl flexible~ NN, chelate for the


synthesis of Fe3, Fe6 and two Fe7 complexes, some of which possess novel Fex topologies. As

part of these continuing efforts to synthesize new Fex clusters, we have now turned to another

potential chelating group that has also never before been employed, to our knowledge, in

transition metal chemistry. This is 6-hydroxymethyl-2,2'-bipyridine (hmbpH; Figure 5-1), whose

deprotonated form, like dmem-, would be a potential N,N,O-chelate, but a more rigid one than

dmem-. In fact, hmbpH was selected as a 'fusion' of two chelates that have each proven a rich

source of Mx, and particularly Fex, species in the past, 2,2'-bipyridine and the anion of 2-

(hydroxymethyl)pyridine (hmpH). 77,109,121-126 We thus considered it a potentially viable route to

new clusters, and probably of a different type than previously encountered with bpy or hmp~

separately. Our first investigations with hmbpH have been in Fe chemistry and we have indeed

found it to lead to new structural types of products. We herein report the syntheses, crystal

structures and magnetochemical characterization of new Fes and Fe6 mOlecular species.127

5.2 Experimental Section

5.2.1 Syntheses

All preparations were performed under aerobic conditions using reagents and solvents as

received. [Fe30(O2CPh)6(H20)3](NO3), [Fe30(O2CBut)6(H20)3](NO3) and









[Fe30(O2CMe)6(H20)3](NO3) WeTO Synthesized as reported elsewhere.96 The known organic

molecule hmbpH was synthesized, according to Figure 5-2, using previously reported

proce ures. 2-3

[FesO2(OH)(O2 Me)5(hmbp)3] (CIO4)2 (5-1). To ai stirred solution of Fe(CIO4)3-xH20

(0.19 g, 0.54 mmol) and sodium acetate (0.18 g, 2.0 mmol) in EtOH (15 mL) was added hmbpH

(0.10 g, 0.54 mmol). The resulting orange solution was stirred for 3 hours at room temperature,

during which time precipitated an orange solid. The precipitate was collected by filtration,

washed with EtOH, and dried. It was then dissolved in MeCN (15 mL), filtered, and the filtrate

layered with Et20. X-ray quality crystals of 5-1-5MeCN slowly grew over five days in 18%

yield. These were collected by filtration, washed with MeCN and dried in vacuo; dried solid

analyzed as solvent-free. Anal. Called (Found) for 5-1 (C43H43N6C 2Fe5O24): C, 37.48 (37.05); H,

3.14 (3.12); N, 6.09 (5.89). Selected IR data (cm )~: 1599(s), 15429s), 1490(m), 1402(s),

1175(m), 1068(m), 1025(m), 937(w), 820(w), 777(m), 719(s), 663(s), 600(w), 547(w), 463(m).

[Fe6O2(OH)2(O2CPh)6(hhmbp)4] (NO3)2 (5-2). An orange-red solution of

[Fe3(O2CPh)6(H20)3](NO3) (0.14 g, 0.14 mmol) in MeCN (20 mL) was treated with hmbpH

(0.05 g, 0.27 mmol). The solution was stirred for 2 hours, filtered, and the orange-red filtrate left

undisturbed to concentrate slowly by evaporation. X-ray quality, orange-red crystals of 5-2

-3MeCN-H20 formed over five days in 25% yield. These were collected by filtration, washed

with MeCN and dried in vacuo. Anal. Called (Found) for 5-2-H20 (C86H70N1oFe6027): C, 51.37

(51.34); H, 3.51 (3.37); N, 6.97 (6.82). Selected IR data (cm )~: 3426(br), 1602(m), 1539(s),

1490(m), 1433(s), 1350(m), 1299(w), 1254(w), 1226(w), 1088(s), 1044(s), 779(m), 690(m),

662(m), 623(m), 556(m), 429(w).









[Fe6O2(OH)2(O2 Me)6(hmbpb)4] (NO3)2 (5-3). An orange-red solution of

[Fe30(O2CMe)6(H20)3](NO3) (0. 11 g, 0. 17 mmol) in MeCN (20 mL) was treated with hmbpH

(0.065 g, 0.35 mmol). The solution was stirred for 2 hours, filtered, and the orange-red filtrate

left undisturbed to concentrate slowly by evaporation. Orange-red crystals formed over five days

in 15% yield. These were collected by filtration, washed with MeCN and dried in vacuo. Anal.

Called (Found) for 5-3-H20 (C56H58N1oFe6027): C, 41.06 (41.13); H, 3.57 (3.61); N, 8.55 (8.33).

Selected IR data (cm )~: 3399(br), 1601(m), 1546(s), 1491(m), 1438(s), 1384(s), 1256(w),

1225(w), 1166(w), 1090(w), 1037(s), 905(w), 833(w), 781(m), 664(s), 645(m), 556(m), 434(w),

413(w).

[Fe6O2(OH)2(O2CBu?6)s~hmbp)4](NO3)2 (5-4). An orange-red solution of

[Fe30(O2CBut)6(H20)3](NO3) (0.13 g, 0.14 mmol) in MeCN (20 ml) was treated with hmbpH

(0.05 g, 0.27 mmol). The solution was stirred for 2 hours, filtered, and the orange-red filtrate

layered with Et20. Orange crystals slowly grew over four days in 20% yield. Anal. Called

(Found) for 5-4-H20 (C74H94N1oFe6027): C, 47.01 (46.64); H, 5.01 (4.76); N, 7.41 (7.50).

Selected IR data (cm )~: 3408(br), 3077(w), 2962(m), 1601(w), 1539(s), 1484(m), 1459(m),

1425(s), 1383(s), 1361(s), 1300(w), 1227(m), 1163(w), 1091(w), 1036(m), 900(w), 832(w),

785(m), 663(s), 600(m), 553(m), 434(m).

5.2.2 X-ray Crystallography

Data were collected by Dr. Khalil A. Abboud on a Siemens SMART PLATFORM

equipped with a CCD area detector and a graphite monochromator utilizing Mo-Ku radiation (h

= 0.71073 A+). Suitable crystals of 5-1-5MeCN and 5-2-3MeCN-H20 were attached to glass fibers

using silicone grease and transferred to a goniostat where they were cooled to 173 K for data

collection. Cell parameters were refined using up to 8192 reflections. A full sphere of data (1850

frames) was collected using the co-scan method (0.3o frame width). The first 50 frames were re-









measured at the end of the data collection to monitor instrument and crystal stability (maximum

correction on I was <1 %). Absorption corrections by integration were applied based on

measured indexed crystal faces. The structure was solved by the direct methods in SHELXTL6,46

and refined on F2 USing full-matrix least-squares. The non-HI atoms were treated anisotropically,

whereas the hydrogen atoms were calculated in ideal positions and refined as riding on their

respective carbon atoms.

In 5-1-5MeCN, the asymmetric unit consists of the Fes cluster, two CIO4~ aniOns, and five

MeCN solvent molecules. The latter molecules were disordered and could not be modeled

properly, thus program SQUEEZE,68 a part of the PLATON package of crystallographic

software, was used to calculate the solvent disorder area and remove its contribution to the

overall intensity data. One of the CIO4~ aniOns was disordered and was refined in two positions

with their site occupation factors refined dependently. Both disorder components are HI-bonded

to the hydroxyl proton on 04. A total of 720 parameters were refined in the final cycle of

refinement using 42942 reflections with I > 20(I to yield R1 and wR2 Of 6. 11 and 17.05%,

respectively .

In 5-2-3MeCN-H20, the asymmetric unit consists of a half Fe6 ClUSter, one and a half

MeCN molecules, one NO3~ aniOn disordered over three positions, and a half water molecule

which exists counter to one of the half NO3~ aniOns. One nitrate exists with 50% occupancy,

while the other has all but one O atom disordered. The latter was refined in two parts with

occupation factors fixed at 20 and 30%. A total of 633 parameters were refined in the final cycle

of refinement using 15661 reflections with I > 20(I to yield R1 and wR2 Of 4.85 and 12.76%,

respectively. Unit cell data and structure refinement details for the two compounds are listed in

Table 5-1.









5.3 Results and Discussion


5.3.1 Syntheses

The reaction of Fe(CIO4)3 with hmbpH and sodium acetate in a 1:1:4 ratio in EtOH gave

an orange precipitate that after recrystallization from MeCN/Et20 gave crystals of the novel

pentanuclear cluster [FesO2(OH)(O2CMe)s(hmbp)3](CIO4)2 (5-1). The acetate acts as the proton

acceptor in this reaction, as well as providing ligand groups. The same product was also obtained

using MeCN as the reaction solvent, but the precipitate was found to be contaminated with some

other solid products. The formation of 5-1 is summarized in eq. 5-1. Decreasing the amount of

acetate from 4 to 2 equiv reduces the reaction yield, as expected from eq. 5-1. Other reactions

with small variations in the Fe3+:hmbpH:MeCO2- ratio also gave compound 5-1.

5Fe3+ + 3hmbpH + 13MeCO2- + 3H20 [FesO2(OH)(O2CMe)s(hmbp)3 2+ + 8MeCO2H (5-1)

Many synthetic procedures to polynuclear iron clusters rely on the reaction of

[Fe30(O2CR)6(H20)3]+ Species with a potentially chelating ligand,84,91,92,98,11919,125,2 and we thus

also explored this starting material for reactions with hmbpH. In such reactions, the [Fe30]7

core of the trinuclear iron complex serves as a useful building block for higher nuclearity

species, but we have occasionally found that the exact nuclearity and structure of the product is

sensitive to the identity of the carboxylate employed. An example of this is the reaction of

[Fe30(O2CR)6-(H20)3]+ Species with dmemH.93 Thus, we have also studied the product of

reactions with hmbpH as a function of the carboxylate, but in this case have found that we obtain

the same structural type in each case. Thus, the reaction of [Fe30(O2CR)6(H20)3]+ (R = Ph, Me,

Bur) with 1 3 equiv of hmbpH in MeCN led to the isolation of the corresponding hexanuclear

cluster [Fe602(OH)2(O2CR)6(hmbp)4 2+ (R = Ph (5-2), Me (5-3), But (5-4)). The formation of this

family is summarized in eq. 5-2.









2[Fe30(O2CR)6(H20)3]+ + 4hmbpH [Fe602(OH)2(O2CR)6(hmbp)4 2++ 6RCO2H +

4H20 (5-2)

5.3.2 Description of Structures

5.3.2.1 Structure of [FesO2(OH)(O2 Me)5(hmbp)3] (CIO4)2 (5-1)

A labeled representation of complex 5-1 is shown in Figure 5-3. Selected interatomic

distances and angles are listed in Table A-10. Complex 5-1 crystallizes in monoclinic space

group P21/c. The core can be described as consisting of a [Fe4 u3-O)2 8+ butterfly-like subunit

(Fel1, Fe3, Fe4 and Fe5), on the top of which is attached a [Fe~p-OH)(p-OR)3] unit containing

Fe2. There is an O atom monoatomically bridging Fe2 to each of the four Fe atoms of the

butterfly. Three of these O atoms (OS, OS, 06) are the alkoxide arms of the three hmbp- groups,

and the fourth is the OH- ion (04). The protonated OH- nature of 04 was confirmed by bond

valence sum (BVS) calculations,1l03,131 which gave a value of 1.09. The bipyridyl portions of two

hmbp- groups chelate one each to the two wingtip Fe atoms, Fel and Fe3, while the third

chelates Fe2. Peripheral ligation about the core is then completed by Hyve acetate groups in the

common rl : rl : pu bridging mode. It is interesting to note that bpy itself will react with Fe"' in

the presence of carboxylate groups to give the Fe4 butterfly complexes of formula

[Fe402(O2CR)7(bpy)2]+ with the two bpy groups attached at the wingtip Fe atoms.77 Thus, the

bpy 'fragments' of the hmbp- chelates are giving the analogous Fe4 butterfly unit, but the

alkoxide arms also then foster attachment of the fifth Fe atom. The core of complex 5-1 is

unprecedented in pentanuclear Fe"' chemistry. Indeed, there are relatively few pentanuclear Fe"'

complexes in the literature, and these have Fes topologies such as a square pyramid, a centered

tetrahedron, and a partial cubane extended at one face by a partial admantane unit.132-141









5.3.2.2 Structure of [Fe6O2(OH)2(o2CPh),(hmbp)4] (NO3)2 (5-2)

A labeled representation of complex 5-2 is shown in Figure 5-4, and selected interatomic

distances and angles are listed in Table A-11. Complex 5-2 crystallizes in triclinic space group

Pi. The core can be described as a modification of the structure of 5-1, and consists of a central

[Fe4 u3-O)2 8+ flattened-butterfly unit (Fe2, Fe2', Fe3 and Fe3') on either side of which is

attached a [Fe~p-OH)(p-OR)2] unit containing Fel and Fel'. There are now four hmbp- groups,

two again on the wingtip positions of the butterfly unit, and one each on Fel and Fel'. Unlike 28,

there are now only three monoatomically-bridging O atoms linking Fel to the butterfly unit, two

hmbp- alkoxide arms (03', 09) and the OH- group (07). The OH- nature of 07 was again

confirmed by BVS calculations, which gave a value of 1.16. The peripheral ligation about the

[Fe602(OH)2(hmbp)4] 8+ COTO is completed by six benzoate groups, four in the rl : rl : p -

bridging mode and two in an rl terminal mode. The main overall difference between the core

structures of 5-1 and 5-2 is that one of the wingtip hmbp- groups of 5-1 has rotated by 1800,

bringing its alkoxide arm to the opposite side of the molecule from the fifth Fe atom and thus

allowing attachment of a sixth Fe atom.

A number of other Fe6 COmplexes have been reported in the literature, and these possess a

variety of metal topologies such as planar, twisted boat, chair, parallel triangles, octahedral,

ladder-like, cyclic, etc.104 However, the only previous compounds somewhat structurally similar

to 5-2 are [Fe602(OMe)12(tren)z2 2-105 [Fe602(OR)8(O2CPh)6] 106 and

[Fe602(OH)2(O2CBut)s(dmem)2 -93 In these compounds there is again a central [Fe4 13-O)2 8+

core with an additional Fe atom on each side (as in 5-2), but the precise means by which the

latter are connected to the Fe4 unit are different from the situation in 5-2.









5.3.3 Magnetochemistry of Complexes 5-1 to 5-4

5.3.3.1 De Studies

Solid-state, variable-temperature dc magnetic susceptibility data in a 0. 1 T field and in the

5.0-300 K range were collected on powdered crystalline samples of 5-1 to 5-4 restrained in

eicosane. The obtained data are plotted as &Tys Tin Figure 5-5. For 5-1, 3&T steadily decreases

from 6.69 cm3mOl-1K at 300 K to 4.01 cm3mOl-1K at 5.0 K. The 300 K value is much less than

the spin-only (g = 2) value of 21.87 cm3mOl-1K for five non-interacting Fe"' ions, indicating the

presence of strong antiferromagnetic interactions, as expected for oxo-bridged Fe"' systems. The

5.0 K value of 4.01 cm3mOl-1K suggests a spin S = 5/2 ground state.

The&nTys Tplots for the three complexes 5-2 to 5-4 in Figure 5-5 are very similar,

indicating a minimal influence of the different carboxylate groups and supporting the conclusion

that not just their formulations are identical but also their structures. 3&T for 5-2-H20, 5-3-H20

and 5-4-H20 increases from 11.88, 11.11 and 11.51 cm3mOl-1K at 300 K to a maximum of 14.95,

14.52 and 15.07 cm3mOl-1K at 30 K, and then decreases very slightly to 14.65, 14.35 and 14.91

cm3mOl-1K respectively at 5.0 K. The &T at 300 K is again much less than the spin-only value of

26.25 cm3mOl-1K expected for six non interacting Fe"' ions indicating the presence of strong

antiferromagnetic interactions. However, the increase in&nTwith decreasing temperature

suggests that the lowest lying spin states are of high spin values, and the near-plateau value of

14.6 14.9 cm3mOl-1K at low temperatures is very close to the spin- only value of 15.0 cm3mOl

1K for an S = 5 ground state. The small decrease in &T at the lowest temperatures is very likely

due to the zero-field splitting (ZFS) within the S = 5 ground state and perhaps some weak

intermolecular interactions. The differences in 3&Tys T for the three complexes are almost

certainly just reflecting small differences in g values, intramolecular exchange coupling

constants (J), zero-field splitting parameters (D), and intermolecular antiferromagnetic









interactions, but the overall almost identicalkTys Tplots indicates these factors are

nevertheless almost identical for 5-2 to 5-4.

To confirm the above ground state spin estimates, variable-field (H) and -temperature

magnetization (M)1 data were collected in the 0.1 to 7.0 T and 1.8 to 10 K ranges. The resulting

data for 5-1 are plotted in Figure 5-6(a) as reduced magnetization (M~NflB) VS. H/T, where Nis

Avogadro's number and flB is the Bohr magneton. The saturation value at the highest fields and

lowest temperatures is ~4.86, as expected for an S = 5/2 ground state and g slightly less than 2;

the saturation value should be gS in the absence of complications from low-lying excited states.

The data were fit, using the program M~AGNET,53 by diagonalization of the spin Hamiltonian

matrix assuming only the ground state is populated, incorporating axial anisotropy (Di- ) and

Zeeman terms, and employing a full powder average. The corresponding spin Hamiltonian is

given by eq. 5-3, where S-is the easy-axis spin operator, g is the electronic g factor, flo is the

vacuum permeability, and His the applied field. The last term in eq. 5-3 is the Zeeman energy

associated with an applied magnetic field.

K= Di- + gfBfl0S-H (5-3)

The best-fit for 5-1 is shown as the solid lines in Figure 5-6(a), and was obtained with S =

5/2 and either of the two sets of parameters: g = 1.96 and D = 0.75 cm- or g = 1.95 and D = -

0.59 cm- Alternative fits with S = 3/2 or 7/2 were rej ected because they gave unreasonable

values of g and D. It is common to obtain two acceptable fits of magnetization data for a given S

value, one with D > 0 and the other with D < 0, since magnetization fits are not very sensitive to

the sign of D. This was indeed the case for the magnetization fits for all complexes 5-1 to 5-4. In

order to assess which is the superior fit in all these cases, and also to ensure that the true global

minimum had been located for each compound, we calculated the root-mean-square error surface









for the fits as a function ofD and g. For 5-1, the error surface (Figure 5-6(b)) clearly shows only

two minima with positive and negative D values, with both fits being of comparable quality.

The reduced magnetization plots saturate at 9.78 for 5-2-H20, 9.64 for 5-3-H20 and 9.88

for 5-4-H20, suggesting an S = 5 ground state and g < 2. The best fit for 5-2-H20 is shown as the

solid lines in Figure 5-7(a, top), and was obtained with S = 5 and either g = 2.00 and D = 0.36

cm- or g = 1.97 and D = -0.20 cm- In this case, the fit error surface (Figure 5-7b, top) clearly

shows that the fit with positive D is far superior, suggesting that this is the true sign ofD. The

best fit for 5-3-H20 was obtained with S = 5 and either g = 1.98 and D = 0.46 cm- or g = 1.94

and D = -0.21 cm l. For 5-4-H20, the best fit was with S = 5 and either g = 2.02 and D = 0.36

cm- or g = 1.99 and D = -0. 19 cm- The corresponding figures and the two-dimensional D vs g

error plots for 5-3-H20 and 5-4-H20 are provided in Figure 5-7 (middle and bottom

respectively).

5.3.3.2 Rationalization of the Ground State Spin

It is interesting to try to rationalize the observed ground state spin values of 5-1 and 5-2. It

is assumed that all Fe2 painWise exchange interactions are antiferromagnetic, as is essentially

always the case for high-spin Fe"', and there will thus be competing antiferromagnetic exchange

interactions and spin frustration effects within the many Fe3 triangular units in these complexes.

The ground state of 5-1 is the easiest to rationalize: the discrete Fe4 butterfly (rhombus) topology

is known to usually give an S = 0 ground state as a result of the antiferromagnetic interactions

along the four outer (wingtip-body) edges overcoming, and thus frustrating, the diagonal (body-

body) interaction.77,109-111,214213 The structure of 5-2 comprises such an Fe4 unit with an

additional Fe above it, and assuming the same spin alignments as in the discrete Fe4 mOleCUleS,

then the ground state spin alignments are predicted to be those shown in Figure 5-8(a), giving the

S = 5/2 ground state observed experimentally. Note that whether the spin of the fifth Fe atom is










aligned parallel to the wingtip spins (as shown) or parallel to the body spins, an S= 5/2 ground

state will still result as long as the interactions within the butterfly are stronger than those

between it and the fifth Fe atom. This seems reasonable given that the interactions within the

butterfly involve monoatomically bridging oxide ions. For 5-2, we can again rationalize the

ground state, using a simple extension of the argument for 5-1, on the basis of a central S = 0

planar-butterfly unit coupling with the fifth and sixth Fe atoms as shown in Figure 5-8(b). This

will give an overall S = 5 ground state for 5-2 as observed experimentally. It should be noted that

we have sought to rationalize the ground states of 5-1 and 5-2 on the basis of previous

observations for the Fe4 butterfly units, and with as straightforward a description as possible.

Thus, we have not invoked intermediate spin alignments of individual spins. In reality, the spin

alignments leading to the observed S = 5/2 and 5 ground states could be more complicated than

shown in Figure 5-8.

None of the compounds exhibited an out-of-phase ac magnetic susceptibility signal down

to 1.8 K in an ac field of 3.5 Oe oscillating with frequencies up to 997 Hz, indicating that they do

not exhibit a barrier large enough vs kT, down to 1.8 K at least, to exhibit slow relaxation of their

magnetization vectors, i.e. they are not single-molecule magnets. This is not surprising that the D

values for 5-2 to 5-4 were concluded to be positive, whereas negative D values are required to

yield the easy-axis (Ising) anisotropy necessary for SMMs. For 5-1, we could not conclude the

sign of D: assuming it is negative, the combination of S= 5/2 and D = -0.59 cm-l would yield an

upper limit to the magnetization relaxation barrier (U) of U= (S2-1/4)|D| = 3.5 cml = 5.1 K.

Remembering that the actual or effective barrier (Ugf) is significantly less than U, it is not

surprising that even with a negative D complex 5-1 does not display slow relaxation down to 1.8









K. Studies at much lower temperatures would be required to search for what would at best be a

tiny barrier.

5.4 Conclusions

We have reported the initial use of a new N,N,O based tridentate chelate in coordination

chemistry, one that amalgamates the chelating property of 2,2'-bipyridine (bpy) with the

chelating/bridging properties of the anion of 2-(hydroxymethyl)pyridine (hmpH). The resulting

hmbp- has been employed in Fe"' chemistry, and it has provided clean access to four new

polynuclear Fe clusters 5-1 to 5-4. The structures of 5-2 to 5-4 are concluded to be the same,

given their identical formulation and almost superimposable magnetic properties. Note that

identical formulation by itself does not mean identical structure: we reported two compounds in

chapter 4, [Fe7O4(O2CPh)ll(dmem)2] (4-1) and [Fe7O4(O2CMe)ll(dmem)2] (4-2) that have the

same formula (except for the carboxylate) but very different structures.

The structures of 5-1 and 5-2 show the manifestation of the 'hybrid' bpy/hmp- nature of

hmbp- in that the bpy portion gives an Fe4 butterfly subunit, as does bpy itself, while the alkoxide

arm acts as an additional bridging group and raises the nuclearity to five or six. As a result, the

complexes have significant ground state spin values of S= 5/2 and 5, respectively.










Table 5-1. Crystallographic Data for 5-1-5MeCN and 5-2-3MeCN-H20.
5-1 5-2
Formulaa C53H58Cl:Fe5N11O24 C92H79Fe6N13027
Fw, g/mola 1583.25 2133.77
Space group P21/c Pi
a, Ai 21.6352(2) 13.8233(6)
b, Ai 13.4154(6) 14.0671(6)
c, A 23.1971(11) 14.2856(6)
a, O 90 65.175(2)
P, a 102.456(2) 70.147(2)
Y, o 90 89.561(2)
yq A, 6574.4(5) 2341.55(17)
Z 41
T, K 173(2) 173(2)
1. Ab 0.71073 0.71073
Asial, g/cm3 1.600 1.512
pu, mm' 1.244 0.990
R1 cd 0.0611 0.0487
wR2 e 0.1705 0.1276
a Including solvate molecules. b Graphite monochromator. "l> 20(1. dR1 = C(||Fo| |F,||)l / EFo. e wR2 = [C[w(Fo2
Fe ) ] / C[w(Fo ) ]] ,2 W = 1/[o (Fo ) + [(ap)2 +bp], where p = [max (Fo2 O) + 2Fc ]/3.






HOO
dmemH hmbpH

Figure 5-1. Structure of ligands.









O C





Figure 5-2. Synthetic scheme for hmbpH (i) H202, CF3CO2H (Ref. 117) (ii) Me3SiCN, PhCOC1,
CH2 12 (Ref. 118) (iii) NaOMe, MeOH (Ref. 118) (iv) (a) NaBH4, EtOH(b) H2SO4,
H20 (Ref. 119)





























Figure 5-3. Labeled representation of the structure of 5-1. Hydrogen atoms have been omitted for
clarity. Color code: Fe"', green; O, red; N, blue; C, grey.


Figure 5-4. Labeled representation of the structure of 5-2. Hydrogen atoms have been omitted for
clarity. Color code: Fe"', green; O, red; C, grey; N, blue.














14 -




10







4 -( Messe goe 5-1
0 5-2
r 5-3
S5-4


0 50 100 150 200 250 300

T/K


Figure 5-5. Plots of guTys Tfor complexes 5-1 (*), 5-2-H20 (m), 5-3-H20 (A) and 5-4-H20 (+).


Z 3 m 0.5T o 0.






HTT /kG g



plot ofthe fiting errr surfae vsD ndg fr51





















# 2T2


1 T

0 4T -0.2
m T

S7T
-Fitting -0.4


40 50 1.90 1.95 2.00


(b)O.6 ,,,


0 10 20 30

HT '/kGK


2.056 2."10


0.4


0.2

E
B 0.0


-0 2


-0.4


10 20 30

HT /lkGK


40 1.90 1.95 2.00

(b) s


2.05 2.10


0.4


0.2



_o0.0


-0.2


-0.4


40 50 1.90 1.95 2.00

9


2.05 2.10


0 10 20 30

HT ikGK 1


Figure 5-7 (a) Plot of reduced magnetization (M/NpuB) VS H/T for complexes 5-2*H20 (top), 5-
3-H20 (middle), 5-4-H20 (bottom). The solid lines are the fit of the data (b) Two-
dimensional contour plot of the fitting error surface vs D and g for complexes
5-2*H20 (top), 5-3-H20 (middle), 5-4-H20 (bottom).









(a) (b) -



o oe~





Fe


Figure 5-8. Rationalization of spin ground state of complex (a) 5-1 and (b) 5-2.









CHAPTER 6
NEW STRUCTURAL TYPES IN POLYNUCLEAR IRON CLUSTERS INCORPORATING
O,N,N,O LIGAND: A SNAKE LIKE CHAINT AND A SUPRAMOLECULAR DIMER OF
SMMs

6.1 Introduction

Molecular clusters of magnetic transition metal ions have been generating great interest

since the discovery that these molecules can behave as nanoscale magnets and show magnetic

bistability of pure molecular origin which can be used for information storage.13 The first

molecule to show this behaviour was [Mnl2012(O2CMe)16(H20)4, 15 which led to the field of

single molecule magnetism. Since then much effort has been put into finding new systems with

interesting magnetic properties. Although a variety of Mn containing SMMs have been reported

in recent years but to date, SMMs based on Fe"' are still rare.22

In addition, oxo-bridged Fe"' clusters of various nuclearities have been studied as models

of Fe sites in proteins and enzymes, as well as models of intermediate stages of the growth of the

giant Fe/O core of the Fe storage protein ferritin.144 In fact, the biological and magnetic areas

essentially involve the same Fe/O chemistry as emphasized by the fact that Fe/O core of ferritin

can be considered a nanoscale magnetic particle and has been investigated for quantum tunneling

effects of magnetization.31

SMMs derive their properties from the combination of a large ground-state spin quantum

number (S) and a magnetoanisotropy of the easy-axis (Ising)-type (negative zero-field splitting

parameter, D), rather than from intermolecular interactions and long-range ordering as in

traditional magnets.13 This combination leads to a significant barrier (U) to relaxation

(reorientation) of the magnetization vector, whose maximum value is given by S2|D| or (S2 1/4)

|D| for integer and half-integer spin, respectively. The use of high spin Fe"' can theoretically lead

to large spin values even for quite small number of paramagnetic centers.20 Although the









interactions between Fe"' are generally antiferromagnetic, some topological arrangements can

result in large ground state spin due to the phenomenon of spin frustration. In general, molecules

with high spin ground state have many spin states which are thermally populated at room

temperature; therefore the S1VM properties manifest itself at low temperature when excited states

are depopulated.

Fe"' clusters do not provide S1VINs by design because the conditions for large Ising type

anisotropy depend very unpredictably on minor features of the coordination environment of the

metal ion.20 The above considerations and others continue to stimulate groups around the world

to develop new synthetic methods that can yield new polynuclear Fe/O clusters. However there

is no obvious and guaranteed route to such species. Different approaches to the synthesis of iron-

oxo clusters are controlled hydrolysis or alcoholysis reactions of simple iron salts or more

complicated starting materials.30,122 Also, trinuclear iron oxo clusters have also proved to be very

useful building blocks for the preparation of higher nuclearity clusters in the presence of

appropriately chosen chelate.145 One successful strategy in producing polynuclear clusters is the

use of alkoxide-based ligands,32 Since this functionality is an excellent bridging group that fosters

higher nuclearity product formation. In chapter 4, we reported our results in polynuclear iron
clusters with N,N,O based chelate 2- { [2-(dimethyla~minno+ethl] methyrla~mi\no- ethnol


(dmemH).93 In the present work, we have been investigating the use of a new O,N,N,O-based

chelate N,N~-bis(2-hydroxyethyl)ethylenediamine (heenH2; Figure 6-1) for transition metal

cluster chemistry. It has been used before in the literature to make mononuclear Pt and Cu

molecules in its fully protonated form but there are no previous use of it in polynuclear metal

(i.e. cluster) chemistry, protonated or otherwise. Our first investigations with heenH2 have been

in Fe chemistry, and we have indeed found it to lead to new structural types of products. We









herein report the syntheses, crystal structures and magentochemical characterization of four new

iron complexes Fe6, Fe7, Fe9 and Fels.145,146

6.2 Experimental Section

6.2.1 Syntheses

All preparations were performed under aerobic conditions using reagents and solvents as

received. [Fe30(O2CBut)6(H20)3]NO3 and [Fe30(O2CPh)6(H20)3]NO3, WeTO Synthesized as

reported elsewhere.96

[FelsOs(OH)2(O2CBu')28(heen)4] (6-1). To a stirred dichloromethane solution (20 mL) of

[Fe30(O2CBut)6(H20)3]NO3 (0.20 g, 0.21 mmol) was added heenH2 (0.05 g, 0.31 mmol) and the

reaction mixture was stirred for 3 hours and then filtered to remove undissolved solid and

layered with pentane. Orange crystals of 6-1-4CsH12-4CHC2C 2Suitable for X-ray crystallography

formed over a week in 30% yield. Anal. called (Found) for 6-1 (C164H308N8Fe1sO74) : C, 42.98

(43.25); H, 6.82 (6.99); N, 2.44 (2.43). Selected IR data (cm )~: 2961(s), 2928(m), 2871(w),

1580(s), 1551(s), 1484(vs), 1421(vs), 1376(m), 1228(m), 1090(w), 906(w), 787(w), 663(w),

602(m), 508(w), 438(m).

[Fe904(OH)4(O2CPh)13(heenH)2] (6-2). Ligand heenH2 (0. 11 g, 0.74 mmol) was added to

a stirred MeCN solution (15 mL) of FeCl3-6H20 (0.20 g, 0.74 mmol) and NaO2CPh (0.21 g, 0.15

mmol). The reaction mixture was stirred for 2 hours and then filtered. Slow evaporation of the

filtrate gave X-ray quality crystals of 6-2-9MeCN in three weeks days in 15% yield. Anal. called

(Found) for 6-2 (C103H99N4Fe9038): C, 49.41 (48.99); H, 3.98 (3.90); N, 2.24 (2.57). Selected IR

data (cm )~: 3421(br), 1599(m), 1558(s), 1597(s), 1176(w), 1025(w), 827(w), 715(m), 676(m),

592(w), 469(m).

[Fe7O3(OMe)3(MeOH)1(heen)3Cl4.5 H20)1.s] Cll.25 Fe 14 % (6-3). To a stirred solution of

FeCl2-4H20 (0.20 g, 1.0 mmol) in methanol (10 mL) was added heenH2 (0.15 g, 1.9 mmol) and









the resulting solution was refluxed for 2 hours. It was filtered hot and kept for slow evaporation.

Orange needle like crystals of 6-3-2MeOH-%2H20 appeared in seven days in 10 % yield. Anal.

Called (Found) for 6-3 (C22H58N6Fe7.25014.5 16.75): C, 20.60 (21.09); H, 4.56 (4.93); N, 6.56

(6.08). Selected IR data (cm )~: 3224(s,br), 2952(m), 2868(m), 1634(m), 1455(m), 1344(w),

1236(w), 1090(s), 1066(s), 963(s), 886(w), 810(w), 670(s), 633(m), 558(m), 505(m), 407(m).

[Fe6O2(O2CPh)s(heen)3(heenH)] (CIO4)2 (6-4) Method A. A solution of Fe(CIO4)3 (0.20

g, 0.56 mmol) and NaO2CPh (0. 16 g, 0. 11 mmol) in EtOH (15 mL) was treated with heenH2

(0.08 g, 0.57 mmol) and stirred for 3 hours. The resultant red brown solution was filtered to

remove NaCl and the filterate was left undisturbed for slow evaporation. X-ray quality crystals

of 6-4-2EtOH-1.5H20 appeared in 5 days in 25% yield. Anal. called. (Found) for 6-4-H20

(C59H84N8Fe6029 12): C, 39.92 (39.84); H, 4.77 (4.62); N, 6.31 (6.19). Selected IR data (cm )~:

2968(m), 1596(m), 1557(m), 1400(s), 1341(w), 1302(w), 1176(w), 1190(s), 1024(m), 981(m),

829(w), 724(s), 624(m), 533(w), 468(m).

Method B. An orange red solution of [Fe30(O2CPh)6(H20)3]NO3 (0.20 g, 0.19 mmol) in

EtOH (15 mL) was treated with heenH2 (0.06 g, 0.39 mmol) and NaCIO4 (0.05 g, 0.41 mmol)

and the resulting solution stirred for 2 hours at room temperature. Next, it was filtered to remove

undissolved starting material and the filterate was allowed to stand undisturbed at room

temperature. Orange crystals of the product formed over 5 days in 15 % yield.

6.2.2. X-ray Crystallography

Data were collected by Dr. Khalil A. Abboud on a Siemens SMART PLATFORM

equipped with a CCD area detector and a graphite monochromator utilizing Mo-Ka radiation (il

= 0.71073 A+). Suitable crystals of 6-1-4CSH12-4CH2 12, 6-2-9MeCN, 6-3-2MeOH-%2H20 and 6-

4-2EtOH-1.5H20 were attached to glass fibers using silicone grease and transferred to a goniostat









where they were cooled to 173 K for data collection. Cell parameters were refined using up to

8192 reflections. A full sphere of data (1850 frames) was collected using the co-scan method

(0.30 frame width). The first 50 frames were remeasured at the end of data collection to monitor

instrument and crystal stability (maximum correction on I was < 1 %). Absorption corrections by

integration were applied based on measured indexed crystal faces. The structure was solved by

the Direct Methods in SHELXTL6,46 and refined using full-matrix least squares. The non-HI

atoms were treated anisotropically, whereas the hydrogen atoms were calculated in ideal

positions and were riding on their respective carbon atoms. Refinement was done using F2

In 6-1, the asymmetric unit consists of a half Fels cluster, two pentane molecules and two

dichloromethane molecules. The latter molecules were disordered and could not be modeled

properly, thus program SQUEEZE68 was used to calculate the solvent disorder area and remove

its contribution to the overall intensity data. A total of 1183 parameters were included in the final

cycle of refinement 42915 reflections with I > 20(1 to yield R1 and wR2 Of 4.67 % and 12.52 %,

respectively.

In 6-2, the asymmetric unit consists of the cluster and 9 acetonitrile molecules. The latter

were disordered and could not be modeled properly, thus program SQUEEZE68 was used to

calculate the solvent disorder area and remove its contribution to the overall intensity data. There

are three phenyl groups that are disordered (at C81, C91 and C121) and each was refined as a

rigid group and in two parts with their site occupation factors dependently refined. The last two

of them cause a disorder in an uncoordinated EtOH arm of the N4 diamine. Both hydroxy group

protons of the uncoordinated diamines were calculated in idealized positions and refined riding

on their O atoms. A total of 1375 parameters were included in the final cycle of refinement

using 42636 reflections with I > 20(I to yield R1 and wR2 Of 7.41 % and 18.03 %, respectively.









In 6-3, the asymmetric unit consists of one Fe7 cluster and and FeCl4 lOcated on a 222

symmetry site (thus '/ is in the asymmetric unit). The structure has several disorders one of

which is in the Nl ligand where two CH2 units are disordered and were refined in two parts with

their site occupation factors dependently refined. The 04 position is a disorder between a

coordinated water molecule and a methanol solvent (018) in one parts against a coordinated

methanol ligand and a water solvent (019) molecule. Another disorder has a coordinated

methanol and a solvent methanol in one part (0 11 and 015) against coordinated water and a

solvent methanol molecule (011 and 016). The coordinated Cl4 ligand is disordered with a

water (04") ligand while a lattice Cl7 anion is disordered against a lattice water (07") molecule.

The last two disorders lead to balanced charges in the asymmetric unit. A total of 566 parameters

were refined in the final cycle of refinement using 1 1605 reflections withlI> 20(I to yield R1

and wR2 Of 4.64 % and 10.44 %, respectively.

In 6-4, the asymmetric unit consists of a V/2 ClUSter, a perchlorate anion, disordered ethanol

molecules and three partial water molecules. The cluster has one arm of the diamine ligand

disordered against a benzoate anion. Due to symmetry, the two parts of the disorder were 0.5

occupation factors. The perchlorate anion is disordered and was refined in two parts with Cl and

Cl'. Further, the Cl' part was also disordered along a three fold rotation axis. The ethanol solvent

molecule was also disordered and refined in two parts with their site occupation factors

dependently refined. A total of 586 parameters were included in the final cycle of refinement

using 8642 reflections with I > 20(I to yield R1 and wR2 Of 6.22 % and 15.51 %, respectively.

Unit cell data and details of the structure refinements for 6-1 to 6-4 are listed in Table 6-1.









6.3 Results and Discussion


6.3.1 Syntheses

Reaction of [Fe30(O2CBut)6(H20)3]+ and heenH2 in a 2 : 3 molar ratio in CH2 12, followed

by layering with pentanes, gave orange crystals of an octadecanuclear product

[FelsOs(OH)2(O2CBut)28(heen)4] (6-1) with the core topology not previously encountered. Its

formation is summarized in eq. 6-1. Use of methylene chloride as the reaction medium was

found to be very essential for obtaining a clean product. Using acetonitrile as the reaction

medium gives a mixture of white and orange crystals, which were difficult to separate, and hence

could not be characterized. Further we decided to use polar alcoholic solvents seeking formation

of higher nuclerity products via methanolysis or ethanolysis, but always insoluble yellow

precipitates were obtained.

6[Fe30(O2CBut)6(H20)3]+ + 4heenH2 -[FelsOs(OH)2(O2CBut)28(heen)4] + 14H20 + 8ButCO2H

+ 6H' (6-1)

A similar reaction involving FeCl3, NaO2CPh and heenH2 in 1:2:1 molar ratio in MeCN,

gave [Fe904(OH)4(O2CPh)13(heenH)2] (6-2) (eq. 6-2). On the other hand, reaction of Fe(CIO4)3,

O2CPh- and heenH2 in 1:.2:1 rati o in ethanol gave [Fe602(O2CPh)s (heen)3 (heenH)](CIO4)2 (6-4)

(eq. 6-3). The reaction procedure employed contains an excess of ligand over that required by

stoichiometric reaction and this might be beneficial in providing H+ acceptors.

9FeCl3 + 13PhCO2- + 2heenH2 + 8H20 [Fe90s(O2CPh)9(HO2C~)4hen)4(enH2 + 14H1 + 27Cl-

(6-2)

6Fe(CIO4)3 + 5PhCO2- + 4heenH2 +2H20 -[Fe602(O2CPh)s(heen)3(heenH)]2+ + 18CIO4- + 1 1H

(6-3)

In both these reactions (eq. 6-2 and 6-3) choice of solvent and ratio of starting materials is

very crucial as other ratios do not give clean reaction. In the former reaction, using lower









concentration of sodium benzoate and heenH2 giVCS the crystals of trinuclear iron cluster and

using higher concentration of sodium benzoate and ligand gives messy reactions. Additionally,

nothing crashes out of reaction mixture if ethanol is used as the reaction medium. In the latter

reaction also, 1:2:1 ratio is very important in getting single crystals. In other ratios, different

products are obtained, whose IR is different from 6-4 but these could not be characterized as

single crystals could not be obtained. Further, the reaction mixture gives oily droplets if

acetonitrile is used as the solvent, making ethanol absolutely essential for the formation of 6-4.

Refluxing FeCl2 and heenH2 in 1:1 ratio in MeOH gave 6-3, while nothing comes out of

reaction mixture if EtOH is used instead of MeOH. Using lower or higher concentration of

ligand also gives the same product but very bad quality crystals are obtained.

It is clear that the reactions that lead to 6-1 to 6-4 are very complicated, and the reaction

solutions likely contain a complicated mixture of several species in equilibrium. In such cases,

factors such as relative solubility, lattice energies, crystallization kinetics, and others determine

the identity of the isolated products, and one (or more) of these factors is undoubtedly the reason

that the reaction product changes from an octadecanuclear complex to a nonanuclear complex on

changing the carboxylate employed from pivalate to benzoate.

6.3.2 Description of Structures

6.3.2.1 Structure of [FelsOs(OH)2(O2CBu')28(heen)4] 6-1)

A labeled representation of 6-1 is shown in Figure 6-2. Selected interatomic distances and

angles are given in Table A-12. Complex 6-1 crystallizes in triclinic Pi space group. The

structure of centrosymmetric 6-1 comprises a remarkable Fels cluster that can be described as

seven [Fe202] rhombs linked into a chain, and attached to four end Fe atoms (Fe8, Fe8', Fe9,

Fe9'). Alternatively, it can be better described as the linkage by heen2- alkoxide arms 05 and OS'

of two central [Fe4 13-O)2 8+ butterfly units (Fel, Fe2, Fe3 and Fe4, and its symmetry partner),









and then connected to additional [Fe4 u3-O)2 8+ butterfly units at each end via intermediate Fe

atoms Fe5 and Fe5', the bridges being heen2- alkoxide arms 016 and 017 on one side of Fe5, and

heen2- alkoxide arm 018 and a hydroxide ion 019 on the other side. Peripheral ligation is

provided by a total of four chelating heen2- and twenty-eight pivalate groups, twenty-two of the

latter in their common rl :rl :p-mode, four in a r12 chelating mode on Fe8, Fe8',Fe9 and Fe9', and

two in a rl terminal mode on Fe4 and Fe4'. Two of the four heen2- grOups are in a rl2:lrl 1 1 2:i3-

mode (N3 and N4 are bound to Fe5 while 017 is bridging Fe4 and Fe5, and 018 is bridging Fe5

and Fe6) and the other two are in r12 1 l~r2:i 3- mOde (Nl, N2 are chelating Fe3 and 05 is

bridging Fel and Fel', and 016 is bridging Fe4 and Fe5).

The overall topology of 6-1 is chain-like and resembles a double-headed serpent or

alligator with both sets of jaws wide open. Such a molecule is not only unprecedented in Fe

chemistry, it represents the highest-nuclearity, chain-like metal-containing molecule to be yet

discovered, and can reasonably be called a 'molecular chain'. The next longest such molecular

chain is a Crl2Ni species.147 There are only two previous Fels clusters known, and they are both

wheel complexes, i.e. closed molecular chains.89,148

6.3.2.2 Structure of [Fe904(OH)4(O2CPh)13(heenH)2] (6-2)

A labeled representation of 6-2 is shown in Figure 6-3 with selected geometric parameters

listed in Table A-13. Complex 6-2 crystallizes in the triclinic Pi space group. The structure

consists of pairs of Fe9 ClUSters arranged as head-to-head dimers. Each Fe9 COntains two [Fe4 3s-

O)(p3-OH)]"+ butterfly-like sub-units (Fel, Fe5, Fe6, Fe7, OS, 027 and Fe4, Fe7, Fe8, Fe9, 021,

031) fused at body atom Fe7 (Figure 6-3), and attached to two additional Fe atoms Fe2 and Fe3

by two pu3-OH~ (O6 and 011) and two pu3-O2- ions (07 and 014). There are thirteen PhCO2

groups, nine rl :rl :p-bridging and four rl terminal on Fel, Fe4, Fe6 and Fe8, with their

noncoordinated O atom intramolecularly H-bonded to OH- ions (O6, 011, 05 and 021,









respectively). Each heenH- group is a tridentate chelate to an Fe atom (Fe5 and Fe9) and bridges

through its deprotonated arm to a neighboring Fe atom. Each remaining heenH- protonated arm

(030 and 03 8) is unbound and involved in hydrogen-bonding. The one on Fe9 forms a HI-bond

to benzoate O atom 032 (03 8...032 = 2.788 A+) with no disorder. The one on Fe5, however, is

disordered, forming intramolecular HI-bonds to benzoate O atom 028 (030...028 = 2.882 A+) or

to the same heenH- arm of the corresponding chelate on the neighboring Fe9 molecule

(030...039 = 2.753 A+) (Figure 6-4). Refinement of the disorder components gave an essentially

statistical 36:64% mixture of the intramolecular and intermolecular HI-bonding situations, since

there are two equivalent forms of the latter, as shown in Figure 6-4. Thus, 2/3 of the molecules in

the crystal are linked within HI-bonded [Fe9 2 dimers, whereas the other third of the molecules are

within non-HI-bonded [Fe9 2 dimers.

The core of 6-2 is unprecedented in nonanuclear Fe(III) chemistry. Indeed, there are

relatively few Fe9 ClUSters known in literature and they are described as ferric "Triple-Decker",

nonanuclear ring, rhomb-like array, two distorted Fe4 tetrahedra linked to another Fe atom via

oxide and alkoxides, central Fe atom surrounded by four dinuclear Fe2 units etc.58,80,149-152

6.3.2.3 Structure of [Fe7O3(OMe)3(MeOH)1(heen)3Cl4.5 H20)1.s] Cll.25 Fe 1]4 % 6-3)

A labeled representation of 6-3 is shown in Figure 6-5. Selected interatomic distances and

angles are given in Table A-14. Complex 6-3 crystallizes in the high symmetry orthorhombic I222

space group and it consists of a planar arrangement of six Fe"' ions with a seventh central Fe"'

ion 1.437 A+ below the Fe6 plane. The central Fe atom, Fel, is tetracoordinated. It establishes

bonds with three pu3-O2- ions and with a terminal Cl ligand. Every peripheral iron is connected to

one of its adj acent Fe atoms through one pu-OMe- and one alkoxide group from heen2- and to the

neighboring iron atom through the second alkoxide arm of heen2- and a pu3-O2- bridge, which in

turn are bridging the alternate peripheral iron atoms to the central tetrahedral iron atom. N atoms









of the heen2- grOup complete the octahedral coordination of alternating iron atoms, Fe3, Fe5 and

Fe7 while terminal Cl ligand and H20/MeOH molecules complete the octahedral coordination of

alternating iron atoms Fe2, Fe4 and Fe6.

There is one molecule of FeCl4~ aniOn per four formula units of Fe7 for charge balance.

Additionally there are lattice Cl- ions for charge balance and molecules of water and methanol as

solvents of crystallization. There are several features which makes this structure especially

remarkable. First, it is one of very few polynuclear iron-oxo complexes with an odd number of

metal ions. Second striking feature of this compound is rather unusual coordination of central

iron atom. Fel has an almost regular tetrahedral geometry with O-Fe-O angles ranging between

108.1-111.40

The core topology of 6-3 is new within Fe chemistry. There are only a few Fe7

complexes in the literature and they are described as cagelike, disklike and domelike.98,101,102

Also, we reported two new heptanuclear clusters in chapter 4, [FeO4(O2CPh)ll(dmem)2] and

[Fe7O4(O2CMe)l l(dmem)2]. These complexes had the same formula except for the identity of the

carboxylate but the structures were very different. The former had two [Fe4 u3-O)2 8+ butterfly

units sharing a common body iron atom while in the other a number of Fe30 triangular units

were linked in an unusual way.93

6.3.2.4 Structure of [Fe6O2(O2CPh)s(ee)(heen)3(eeH)(CIO4)2 (6-4)

A labeled representation of 6-4 is shown in Figure 6-6. Selected interatomic distances and

angles are given in Table A-15. Complex 6-4 crystallizes in triclinic Pi space group. The

asymmetric unit consists of half of the cluster, a perchlorate ion and an ethanol molecule and one

and a half water molecule as the solvents of crystallization. The structure of the centrosymmetric

cation of 6-4 can be described as a central [Fe4 u3-O)2 8+ butterfly-like unit (Fel, Fel', Fe2 and

Fe2') connected through both its body (Fel, Fel') and wingtip atoms (Fe2, Fe2') to two









additional Fe atoms Fe3 and Fe3' by heen2- (OS, 09, 010) alkoxide arms. At one end of the

molecule (Fe3), the remaining heen2- alkoxide arm (011) binds terminally; at the other end

(Fe3'), this arm (020) is protonated and unbound (i.e. a heenH- group) and there is instead a

PhCO2- bound terminally (Figure 6-6; only benzoate ipso C atoms shown). These two situations

at the two ends are, of course, disordered by the centre of symmetry which makes Fe3 and Fe3'

equivalent; only one of the disorder components is shown in Figure 6-6. Ligation is completed

by four benzoate groups in the common rl :rl :p-mode bridging the body and wingtip iron atoms

of the central butterfly unit, only the ipso C atoms of benzoate rings have been shown for clarity

in Figure 6-6. A number of Fe6 COmplexes have been previously reported possessing a variety of

metal topologies such as planar, twisted boat, chair, parallel triangles, linked triangles, fused

butterflies, octahedral, ladder-like, cyclic etc.104

6.3.3 Magnetochemistry of complexes 6-1 to 6-4

6.3.3.1 De Studies

Solid-state, variable temperature dc magnetic susceptibility data in a 0.1 T and 5.0-300 K

range were collected on powdered crystalline samples of 6-1 to 6-4 restrained in eicosane. The

obtained data are plotted as 3guTys T in Figure 6-7. For 6-1, 30\T steadily decreases from 22.83

cm3Kmol-l at 300 K to 1.16 cm3Kmol-l at 5.0 K indicating predominant antiferromagnetic

interactions between Fe"' magnetic centers. Value of gs;Tat 300 K is much less than spin-only (g

= 2.0) value of 78.75 cm3Kmoll for eighteen non-interacting Fe"' ions, the behavior with

decreasing temperature and the low value of gs;Tat 5 K are indicative of S = 0 ground state spin.

This is not unexpected given that this is the most common ground state for Fe"' clusters where x

is even. There are common exceptions to this rule, however, when the topology is such so as to

introduce competing antiferromagnetic exchange interactions and spin frustration effects that

result in often significant ground state S values. For complex 6-1, the S = 0 ground state can be









rationalized as shown in Figure 6-8: the central butterfly units are known to exhibit spin

frustration effects within their triangular subunits and to possess an S = 0 ground state as a result

of the four 'wingtip-body' interactions of the four edges overcome (frustrating) the 'body-body'

interaction." The antiferromagnetic interactions between separate butterfly units and between

them and Fe5/Fe5' then lead to an expected S = 0 ground state, as found experimentally. Of

course, since 6-1 has an S= 0 ground state, the depicted spin alignments in Figure 6-8 represent

only one of the component wave functions of the ground state eigenstate of the molecule.

For 6-2, the value of the 3&T gradually decreases from 13.3 5 cm3Kmol-l at 300 K to 1 1.65

cm3Kmol-l at 150 K, stays essentially constant until 60 K and then decreases to 8.73 cm3Kmoll

at 5.0 K. The 300 K value is again less than the spin-only value of 39.37 cm3Kmoll expected for

nine non-interacting Fe"' ions, indicating predominantly antiferromagnetic interactions. The 5.0

K value suggests S = 7/2 ground state. For 6-3, the value of&nT gradually increases from 29.92

cm3Kmol-l at 300 K to 48.16 cm3Kmol-l at 50 K before dropping slowly to 32.83 cm3Kmol-l at

5.0 K. The decrease ofnT at low temperature may be a consequence of zeeman effects from the

DC field and/or the presence of antiferromagnetic interactions between clusters. Subtracting out

the &T contribution for the [FeCl4]~ aniOn indicates an S = 15/2 ground state. For 6-4-H20, the

value ofnT steadily decreases from a value of 10.05 cm3Kmol-l at 300 K to 8. 13 cm3Kmol-l at

70 K and then rises to a value of 10.32 cm3Kmol-l at 5.0 K. The 300 K value is again much less

than the spin-only value of 26.25 cm3Kmoll expected for six non-interacting Fe"' ions,

indicating predominant antiferromagnetic interactions. The increase in 3&T as the temperature

then decreases suggests the lowest lying states are of high spin values, and the value at 5.0 K

suggests S = 4 ground state.









To confirm the above ground state spin estimates, variable Hield (H) and temperature

magnetization (M)1 data were collected in the 0.1 T to 7 T and 1.8 K-10 K ranges. The resulting

data for 6-2 to 6-4 are plotted as reduced magnetization (M/NpuB) VS H/Tin Figures 6-9 and 6-10.

Thes daa wre t uing he rogam 4GNT,53 described elsewhere.56 The best-fit for 6-2 is

shown as the solid lines in Figure 6-9. A satisfactory fit could only be obtained if data collected

at fields above 4 T were excluded, suggesting that some low-lying excited states with S > 7/2 are

being stabilized by the applied field to the point that they are significantly populated at these

temperatures. The best fit was obtained using only the low-field data (< 4 T) (solid lines in

Figure 6-9) gave S = 7/2, D = -0.8510.01 cm-l and g = 2.0610.01. Alternative fits with S = 5/2 or

9/2 were rej ected because they gave unreasonable values of g and D. In order to ensure that the

true global minimum had been located for each compound, we calculated the root-mean-square

error surface for the fits as a function of D and g, and have plotted them as two-dimensional

contour plots. For 6-2, the plot clearly shows only the above-mentioned minima with negative D

value (Figure 6-9).

For 6-3, the best fit (obtained using low-field data; 5 2T) is shown as solid lines in Figure

6-10 (magnetization data was corrected for the paramagnetic anion) and was obtained with S =

15/2 and either of the two sets of parameters, g = 2. 12/D = -0. 13 cm-l or g = 2.07/D = 0. 16 cml

The error surface contour plot is shown in Figure 6-11 and shows the above two minima, with

the one with negative D clearly the superior fit since it has a lower (deeper) minimum.

For 6-4-H20, a good fit was obtained using 0.1-4 T data (solid lines in Figure 6-10) with S

= 4 and either of the two sets of parameters, g = 2.06/D = -0.29 cm-l or g = 2.05/D = 0.3 5 cml

A rms error analysis shows that the one with D < 0 is superior, suggesting this to be the true sign

of D (Figure 6- 11).









6.3.3.2 Ac Studies

Ac magnetic susceptibility studies were performed on vacuum-dried microcrystalline

samples of 6-2 to 6-4 in the temperature range 1.8-10 K with a zero dc field and a 3.5 G ac field

oscillating at frequencies in the 5 1000 Hz range. The in-phase (a)n component of the ac

susceptibility for 6-2, plotted as 3Cy'Tys T, is shown in Figure 6-12. The in-phase 3Cy'T decreases

with decreasing Tbefore exhibiting a frequency dependent drop below ~3 K. Extrapolation from

above 3 to 0 K gives a value of~-7.9 cm3Kmol l, confirming an S = 7/2 ground state. The drop

below ~3 K and the concomitant frequency dependent out-of-phase (p")' signal suggest 6-2 to

possibly be a SMM. The 3Cy'Tys T plot for 6-3 and 6-4-H20 is shown in Figure 6-13.

Extrapolation from above ~4 to 0 K gives a value of ~30 and ~10 cm3Kmol-l for 6-3 and 6-

4-H20, confirming an S= 15/2 and 4 ground state respectively. No out-of-phase signals were

seen for both complexes.

6.3.3.3 Single-Crystal Hysteresis Studies

To probe the possible SMM behavior further, single-crystal hysteresis loop and relaxation

measurements were performed using a micro-SQUID153 Setup. It was found that complex 6-2

indeed behaves as an SMM as shown by the presence of magnetic hysteresis loops whose

coercivity is strongly temperature and sweep rate dependent, increasing with decreasing

temperature and increasing sweep rate as expected for the superparamagnetic like behavior of a

SMM, as shown in Figure 6-14. QTM steps were observed, the first appearing before zero-field

as expected for an exchange-bias effect from the neighbor within the [Fe9 2 dimer, and as seen

for [Mn4 2.154 However, a QTM step at zero field was also seen, and this is not expected for an

exchange biased dimer. However, this and the ~2: 1 ratio of the steps at 0. 11 and 0.0 T can be

explained with reference to Figure 6-4 as due to an antiferromagnetic exchange interaction

between the two Fe9 units mediated by the intermolecular H-bond. Thus, 64% of the [Fe9 2









dimers show an exchange-bias (ON), whereas the remaining 36% do not (OFF). The spin

Hamiltonian (Wi) for each Fe9 SMM with Ising-like anisotropy is given by K= KI + K2 -WS1 2

From the Hx = 0. 11 T (Hx = exchange-bias field) and the relationship J' = -gpsHex/(2S) (H= -2

J'd,-3 convention) "" can be calculated that the intermolecular interaction is J = -0.04 Kt, that is,

very weakly antiferromagnetic.

Thus, 64% of the dimers in the crystal are in an ON state and 36% are OFF (Figure 6-4).

Since the intermolecular interaction J will serve to quantum mechanically entangle the two

molecules and generate superposition states,154,156,157 the ON and OFF states with respect to the

interaction thus correspond to potential ON and OFFstates of a coupled two-qubit system for

quantum computation. Being able to have the interaction ON or OFF in some simple way is

important, and the present work shows that a super exchangepathway via a single H-bond will

suffice. Note that the [Fe9 2 head to-head dimer structure does not depend on the intermolecular

H-bond, unlike [Mn4 2 where a total of six equivalent C-H---Cl intermolecular H-bonds clearly

control the crystallization of the dimer.

6.4 Conclusions

The use of dialcohol based ligand heenH2 in Fe(III) chemistry has led to fascinating new

polynuclear clusters, Fe6, Fe7, Fe9 and Fels. Complex Fels is a new structural type with a

serpentine topology. The diamagnetic ground state of Fels is not unexpected given that this is the

most common ground state for Fe" x clusters where x is even. Some important exceptions

however include [FesO2(OH)12(tacn)6 8+ (S = 10) and Fe6 preSented in this paper, where the

nonzero ground state is caused by spin frustration effects induced by the architecture of these

complexes. A new Fe9 Single-molecule magnet has been synthesized and found to crystallize as

head-to-head dimers. Two-thirds of these are exchange-coupled through a hydrogen-bond









whereas the other third of the dimers are non-interacting, as monitored by magnetization

hysteresis measurements. The crystal thus contains a mixture of ON and OFF dimers with

respect to the quantum mechanical coupling through the hydrogen-bond.

Identification of these new clusters suggests that introduction of alcohol based ligands can

lead to new Fex topologies/high spin molecules/SMMs not seen or accessible with oxide and

carboxylate ligands alone using a simple bottom up approach. These results continue to

emphasize the utility of small nuclearity preformed clusters as stepping-stones to higher

nuclearity products, in this case through the use of alcohol based chelate. Overall we feel that

this system is fascinating both in terms of the aesthetic appeal of the structural concepts and its

potential to provide nanoscale magnets.









Table 6-1. Crystallographic data for 6-1-4CsH12-4CH2C 2, 6-2-9MeCN, 6-3-2MeOH-V2H20 and
6-4-2EtOH-2H20
6-1 6-2 6-3 6-4

Formula C188H364C 8Fe18N8 C103H99Fe9N4038 C24H67N6Fe7.25017 C63H97C 2Fe6Ns
074 C 6.75 031.5
fw, g/mola 5209.77 2503.546 1356.04 1876.49
Space group Pi Pi I222 P2 l/n
a, A+ 15.7923(12) 17.0412(15) 20.8859(14) 11.7319(8)
b, A+ 17.9984(13) 19.8622(16) 21.5851(15) 13.5034(9)
c, A+ 24.8805(19) 19.9519(17) 22.6380(16) 13.8033(10)
a, a 106.938(2) 89.227(2) 90 113.332(1)
/7, 104.711(2) 77.146(2) 90 92.028(1)
Y, a 90.867(2) 77.433(2) 90 99.130(1)

y, A3 6513.0(8) 6421.9(9) 10205.8(12) 1970.8(2)
Z 1 2 8 1
T, K 173(2) 173(2) 173(2) 173(2)
Radiation,Ab 0.71073 0.71073 0.71073 0.71073
Peaic, g/cm3 1.328 1.474 1.765 1.581
pu, mm-l 1.124 1.277 2.420 1.231
R1 c~d 0.0467 0.0741 0.0464 0.0622
wR2e 0.1252 0.1803 0. 1044 0.1551
a Including solvate molecules. b Graphite monochromator. I> 20(I. d R1 = C(||Fo| |F,||)l / EFol.
ewR2 [C[w(Fo2 Fe2 2] w CWFo2 2 11 2, 1[2 Fo2) [P2 +bp], where p = [max (Fo2, O) +
2Fe2]/3.



/N NNH N


HO OH H
dmemH heenH2
Figure 6-1. Structure of chelates: dmemH, heenH2












P-


Figure 6-2. Labeled representation of the structure of 6-1, with core Fe-O bonds shown as thicker
black lines and pivalate Me groups omitted. Color code: Fe, green; O, red; N, blue; C,
grey.


Figure 6-3. Labeled representation of the structure of 6-2 with only the ipso C atoms of benzoate
rings shown for clarity. The core is outlined in bold. Color code: Fe, green; O, red; N,
blue; C, grey.


















































Figure 6-4. The [Fe9 2 dimer showing intermolecular (top, bottom) or only intramolecular
(middle) O-H---O hydrogen-bonding, and the resultant ON or OFF state with respect
to the coupling of the two molecules. Color code: Fe, green; O, red; N, blue; C, grey;
H, sky blue.


OFF




































Figure 6-5. Labeled representation of the structure of 6-3, with core Fe-O bonds shown as thicker
black lines. Color code: Fe, green; Cl, cyan; O, red; N, blue; C, grey.


Figure 6-6. Labeled representation of the cation of 6-4 with only the ipso C atoms of benzoate
rings shown for clarity. Core Fe-O bonds are shown as thicker lines. Colour code: Fe,
green; O, red; N, blue; C, grey.















AA AAAA


40 -



30


E -
q 20




10

o-


A g


O
0
O
oo
mmmug 2 m m m m


0 50 100 150

T/K

Figure 6-7. Plot of guTys Tfor complexes 6-1 to 6-4.


200 250 300


Figure 6-8. Spin alignments at the eighteen S = 5/2 Fe(III) atoms of 6-1 rationalizing its S = 0
ground state, based on the arguments given in the text


O 6-1
m6-2
A 6-3
* 6-4






















*0.1T
n 0.2T
a 1T
2T
3T
0 4T
- Fitting


S3

2


0


S-0.



-1.0



-1.2


0 5 10 15 20 25 '\\'
H T-l/kG K-1 Z oo 2.ob 2 10 2.15 2 0 .Z


Figure 6-9. (left) Plot of reduced magnetization (M~NCB) VSH/ Tfor 6-2. (right) Two-dimensional
contour plot of the r.m. s. error vs D and g for the fit for 6-2. The asterisk marks the
best-fit position (error minimum).


O 2 4 6 8 10 12 0 5 10 15 20 25

HIT(kG K) H/T~lkG/K)

Figure 6-10. Plot of reduced magnetization (M~NpB) VS H/ Tfor 6-3 (left) and 6-4 (right).












1.0




0.5







- 0.5


0.4



0.2








-0 2



-0.4


1.90 1.95 2.00 2.05 2.10J 2.15 2.20


1.90 1.95 2.00 2.05 2.101 2.15 2.20


Figure 6-11i. Two-dimensional contour plot of the r.m. s. error vs D and g for the fit for 6-3 (left)
and 6-4 (right). The asterisk marks the best-fit position (error minimum).


1.0 X



0.5



0.0


2 3 4 5 6


T/K

Figure 6-12. Plot of the in-phase (gs,'T) and out-of-phase (gs,") ac susceptibility data for 6-2.















140













*6-3
v 6-



30 *


20 -




20




2 4 6 8 10

T/K


Figure 6-13. Plot of in-phase ac susceptibility data for 6-3 and 6-4.



0.004 Kill lilli


05 -I 05


0.41 TAr 1 2


-0.5 nna n 0.5 '/ K


-1 *1 ~ ~ m -nan -1 --~--
-1.2 -0. -OA 0 OA 0.8 1.2 -1.2 -0. -OA 0 OA 0.8 1.
co" (T) coM (T)

Figure 6-14. Single-crystal magnetization (M)1 vs dc field (H) hysteresis loops for 6-2 at different
scan rates (left) and at various temperatures (right).









CHAPTER 7
UNUSUAL STRUCTURAL TYPES IN Mn AND Fe CHEMISTRY FROM THE USE OF
N,N,N',N' TETRAKIS (2-HYDROXYETHYL)ETHYLENEDIAMINE

7.1 Introduction

Interest in the preparation of polynuclear Mn and Fe complexes has developed worldwide

for both fundamental scientific and technological reasons since the discovery that some of these

molecules can behave as zero-dimensional nanoscale magnets now called single-molecule

magnets (SMMs).13,22 Since then, many polynuclear clusters containing 3d transition metals have

been reported to be SMMs,20,158-160 the vast majority of bring Mn complexes.21,161-164 In addition,

polynuclear Mn and Fe compounds with O and N based ligation are of interest because of their

relevance in bioinorganic chemistry.26

For the above reasons and more, we continue to seek new synthetic methods to new Mnx

and Fex complexes. In the design of a potentially new synthetic route to a polynuclear cluster, the

choice of the ligands and bridging groups is vital. As part of our continuing search for such new

methods, we have begun exploring the use of chelating/bridging groups based on the

ethylenediamine backbone. In chapter 4 and 6, we reported the use of dmemH and heenH2

(Figure 7-1) as new and flexible N,N,O and O,N,N,O chelates, respectively, for the synthesis of

Fe3, Fe6, Fe7, Fe9 and Fels complexes, some of which possess novel Fex topologies.93,145,146 The

hydroxyethyl arms, on deprotonation, usually act as bridging groups and thus foster formation of

a high nuclearity product. In the present work, we have extended this study by exploring the use

in Mn and Fe cluster chemistry of the related, potentially hexadentate ligand N,N,N',N'-

tetraki s(2-hydroxyethyl)ethylenediamine (edteH4; Figure 7-1). The edteH4 mOleCUle HOW

provides four hydroxyethyl arms on an ethylenediamine backbone, and was considered an

attractive potential new route to high nuclearity products. Previous use of edteH4 in the literature

with other metals has been limited to the preparation only of mononuclear Ca and dinuclear Ba,









Cu and V complexes.165-167 We herein report that the use of edteH4 in a variety of reactions with

Mn reagents has yielded novel Mns, Mnl2 and Mn20 COmplexes with core structures that are

distinctly different from any seen previously.168 The use of edteH4 in Fe chemistry also leads to

interesting new structural types of products, Fes, Fe6 and Fel2 COmplexes. The syntheses,

structures and magnetochemical properties of these complexes will be described.169

7.2 Experimental Section

7.2.1 Syntheses

All preparations were performed under aerobic conditions using reagents and solvents as

received. [Mn30(O2CPh)6 py)2H20],45 [Fe30(O2CPh)6(H20)3]NO3 and

[Fe30(O2CBut)6(H20)3]OH was synthesized as reported elsewhere.94,95,170

[MnsO3(OH)(OMe)(O2CPh)7(edte)(edteH2)] (O2CPh) (7-1). Method A. To a stirred

solution of edteH4 (0.05 g, 0.21 mmol) in CH2C 2/MeOH (16/4 mL) was added

[Mn30(O2CPh)6 py)2H20] (0.23 g, 0.21 mmol). The mixture was stirred for 30 minutes, filtered,

and the filtrate layered with Et20. X-ray quality dark orange-brown crystals of 7-1-2CH2C 2

MeOH slowly formed over a week. They were collected by filtration, washed with Et20 and

dried in vacuo. The yield was 40%. Dried solid appeared to be hygroscopic, analyzing as the

dihydrate. Anal. Called (Found) for 7-1-2H20 (C77H90N4MnsO31): C, 46.08 (45.97); H, 4.52

(4.36); N, 2.79 (2.81). Selected IR data (cm )~: 2868(w), 1649(w), 1595(m), 1547(m), 1447(w),

1383(s), 1315(m), 1174(w), 1122(w), 1066(m), 911(w), 719(m), 676(w), 603(m), 536(m).

Method B. To a stirred solution of edteH4 (0. 10 g, 0.42 mmol) in MeCN/MeOH (10/5 mL)

was added NEt3 (0.18 mL, 1.28 mmol) followed by Mn(O2CPh)2 (0.42 g, 1.26 mmol). The

resulting mixture was stirred for one hour, filtered, and the fi1trate layered with Et20. Dark

orange-brown crystals of 7-1 slowly formed over 5 days and were then isolated as in Method A.

The yield was 20%. The product was identified by IR spectral comparison with material from









Method A and elemental analysis. Anal. Called (Found) for 7-1-2H20 (C77H90N4MnsO31): C,

46.08 (45.79); H, 4.52 (4.37); N, 2.79 (2.76).

[Mnl204(OH)2(edte)4 16 20)2] (7-2). To a stirred solution of edteH4 (0. 15 g, 0.64

mmol) in MeCN/MeOH (10/1 mL) was added NEt3 (0.09 mL, 0.64 mmol) followed by

MnCl2-4H20 (0.25 g, 1.26 mmol). The solution was stirred for 2 hours and then filtered, and the

brown fi1trate was left undisturbed to evaporate slowly, giving X-ray quality crystals of

7-2-6MeCN-V2H20 over Hyve days. These were collected by filtration, washed with MeCN, and

dried in vacuo. The yield was 25%. Dried solid analyzed as solvent-free. Anal. Called (Found) for

7-2 (C40H86NsMnl2024C 6): C, 24.83 (24.57); H, 4.48 (4.50); N, 5.79 (6.20); Cl, 10.99 (11.89).

Selected IR data (cm )~: 2854(m), 1631l(w), 1465(w), 1359(w), 1270(w), 1160(w), 1088(s),

1059(s), 926(m), 899(m), 741(w), 669(m), 619(m), 557(m).

[Mn200s(OH)4(O2 ME)6(edte)6]](CIO4)2 (7-3). To a stirred solution of edteH4 (0. 10 g,

0.42 mmol) in MeOH (12 mL) was added NEt3 (0.12 mL, 0.85 mmol) followed by

Mn(O2CMe)2-4H20 (0.21 g, 0.86 mmol) and then NaCIO4 (0.05 g, 0.41 mmol). The mixture was

stirred for one hour, filtered, and the fi1trate layered with Et20. X-ray quality dark orange-brown

crystals of 7-3-10MeOH slowly formed over a week. They were collected by filtration, washed

with a little Et20, and dried in vacuo. The yield was 20%. Dried solid appeared to be hygrosco-

pic, analyzing as the pentahydrate. Anal. Called (Found) for 7-3-5H20 (C72H152N12Mn20061C 2):

C, 25.96 (25.92); H, 4.60 (4.69); N, 5.04 (4.54). Selected IR data (cm )~: 2929(w), 1560(s),

1418(s), 1145(m), 1112(m), 1088(s), 910(m), 627(s), 563(m).

[FesO2(O2 CPh)7(e dte)(H20)] (7-4). To a sti rred soluti on of edteH4 (0.0 5 g, 0.2 1 mm ol) i n

CH2C 2 (15 mL) was added [Fe30(O2CPh)6(H20)3](NO3)(0.38 g, 0.37 mmol). The mixture was

stirred for 30 minutes, filtered to remove undissolved solid, and the fi1trate layered with a 1:1









(v/v) mixture of Et20 and hexanes. X-ray quality orange crystals of 7-4-CH2C 2 Slowly formed

over a period of five days. These were collected by filtration, washed with Et20, and dried in

vacuo. The yield was 20%. Anal. Called (Found) for 7-4 (C59H57N2FesO21): C, 50.28 (50.63); H,

4.07 (4.27); N, 1.99 (1.85). Selected IR data (cm )~: 2862(w), 1597(m), 1552(s), 1534(s),

1400(s), 1175(w), 1087(m), 1067(m), 1024(w), 928(w), 892(w), 863(w), 720(s), 653(m),

602(w), 529(w), 465(m).

[Fe6O2(O2CBu')s(edteH)2] (7-5). To a stirred solution of edteH4 (0. 10 g, 0.42 mmol) in

CHCl3 (15 mL) was added [Fe30(O2CBut)6(H20)3](OH) (0.18 g, 0.21 mmol). The mixture was

stirred for 30 minutes, filtered to remove undissolved solid, and the filtrate layered with

pentanes. X-ray quality orange crystals of 7-5-2CHCl3 Slowly formed over a week. These were

collected by filtration, washed with pentanes, and dried in vacuo. The yield was 10 %. Dried

solid appeared to be very hygroscopic, analyzing as the tetrahydrate. Anal. Called (Found) for

7-5-2CHCl3-4H20 (C62H124N4Fe6C 6030): C, 38.12 (37.98); H, 6.40 (6.33); N, 2.87 (3.24).

Selected IR data (cm )~: 2960(m), 2869(m), 1562(s), 1483(s), 1421(s), 1375(m), 1360(m),

1227(m), 1098(m), 1042(w), 909(w), 788(w), 694(m), 603(m), 554(m), 480(w), 429(m).

[Fel204(OH)2(O2 ME)6(edte)4 H20)2] (CIO4)4 (7-6). To a stirred solution of edteH4 (0. 10

g, 0.42 mmol) in MeCN (15 mL) was added NaO2CMe-3H20 (0.23 g, 1.69 mmol) followed by

Fe(CIO4)3-6H20 (0.39 g, 0.85 mmol). The mixture was stirred for 30 minutes, filtered to remove

undissolved solid, and the filtrate left to slowly concentrate by evaporation. X-ray quality orange

crystals of 7-6-4MeCN slowly formed over a week. These were collected by filtration, washed

with MeCN, and dried in vacuo. The yield was 40%. Anal. Called (Found) for 7-6

(C52H104NsFel2C 4052) : C, 25.13 (24.82); H, 4.22 (4.21); N, 4.51 (4.47). Selected IR data (cm l):









2884(m), 1559(s), 1455(s), 1336(w), 1271(w), 1086(s), 933(m), 904(m), 744(w), 624(s), 532(m),

466(w), 437(w).

[Fel204(OH)s(edte)4 H20)2] (CIO4)4 (7-7). To a stirred solution of edteH4 (0.05 g, 0.21

mmol) in EtOH (15 mL) was added Fe(CIO4)3-6H20 (0.39 g, 0.85 mmol). The mixture was

stirred for 30 minutes, filtered to remove undissolved solid, and the filtrate left to slowly

concentrate by evaporation. X-ray quality orange crystals of 7-7 slowly formed over a week.

These were collected by filtration, washed with EtOH, and dried in vacuo. The yield was 20%.

Anal. Called (Found) for 7-7 (C40H92N8Fel2C 4046) : C, 21.66 (21.51); H, 3.74 (4. 15); N, 5.31

(5.02). Selected IR data (cm )~: 2867(m), 1628(w), 1469(w), 1363(w), 1270(w), 1088(s), 935(m),

910(m), 740(w), 661(m), 627(s), 583(w), 490(m).

[Fel204(OH)s(edte)4H20)2] (NO3)4 (7-8). To a stirred solution of edteH4 (0. 10 g, 0.42

mmol) in MeOH (15 mL) was added NEt3 (0.12 mL, 0.85 mmol) followed by Fe(NO3)3-9H20

(0.34 g, 0.85 mmol). The mixture was stirred for 30 minutes, and filtered to remove undissolved

solid. Vapor diffusion of THF into the fi1trate gave needle-like orange crystals of 7-8. These

were collected by filtration, washed with THF, and dried in vacuo. The yield was 10%. Anal.

Called (Found) for 7-8 (C40H92N12Fel2042): C, 23.29 (23.06); H, 4.61 (4.45); N, 8.12 (8.07).

Selected IR data (cm )~: 2938(w), 2677 (m), 1650(w), 1385(s), 1171(w), 1057(m), 934(m),

909(m), 825(m), 636(m), 613(w), 525(w), 492(m).

7.2.2 X-ray Crystallography

Data were collected by Dr. Khalil A. Abboud on a Siemens SMART PLATFORM

equipped with a CCD area detector and a graphite monochromator utilizing Mo-K, radiation (Al

= 0.71073 A+). Suitable crystals of 7-1-2CH2C 2-MeOH, 7-2-6MeCN-V2H20, 7-3-10MeOH, 7-

4-CH2 12, 7-5-2CHCl3, 7-6 -4MeCN and 7-7 were attached to glass fibers using silicone grease and









transferred to a goniostat where they were cooled to 173 K for data collection. Cell parameters

were refined using up to 8192 reflections. A full sphere of data (1850 frames) was collected

using the co-scan method (0.30 frame width). The first 50 frames were remeasured at the end of

data collection to monitor instrument and crystal stability (maximum correction on I was < 1 %).

Absorption corrections by integration were applied based on measured indexed crystal faces. The

structure was solved by direct methods in SHELXTL6, 46 and refined on F2 USing full-matrix least

squares. The non-HI atoms were treated anisotropically, whereas the hydrogen atoms were placed

in ideal, calculated positions and were refined as riding on their respective C atoms.

For 7-1-2CH2 2-MeOH, the asymmetric unit consists of the Mns cation, a benzoate anion,

and one MeOH and two CH2 12 mOlecules. The solvent molecules were disordered and could

not be modeled properly, thus program SQUEEZE,68 a part of the PLATON package of

crystallographic software, was used to calculate the solvent disorder area and remove its

contribution to the overall intensity data. A total of 1068 parameters were refined in the Einal

cycle of refinement using 10920 reflections with l > 20(I to yield R1 and wR2 Of 3.34 and

8.12%, respectively.

For 7-2-6MeCN-%2H20, the asymmetric unit contains '/ of the Mnl2 ClUSter, two '/ MeCN

molecules, one MeCN in a general position, and a 1/8 water molecule. All solvent molecules

were disordered and could not be modeled properly, thus program SQUEEZE68 was used to

calculate the solvent disorder area and remove its contribution to the overall intensity data. A

total of 204 parameters were refined in the Einal cycle of refinement using 2799 reflections with I

> 20(1 to yield R1 and wR2 Of 4.64 and 10.46%, respectively.

For 7-3-10MeOH, the asymmetric unit consists of half the Mn20 ClUSter, one CIO4~ aniOn,

and five MeOH molecules. The latter were disordered and could not be modeled properly, thus









program SQUEEZE68 was again used to calculate the solvent disorder area and remove its

contribution to the overall intensity data. The cluster exhibits two disorders: A large part of the

N4 ligand is disordered and was refined in two parts; attempts to resolve the two parts of C17'-

C18' were not successful, and thus this part remained common to both. C28 is also disordered

and was also refined in two parts. The two parts of each disorder were dependently refined.

Three O atoms of the CIO4~ aniOn containing Cl2 were also disordered and were refined in two

parts related by rotation along the Cl2-O32 axis. A total of 687 parameters were refined in the

final cycle of refinement using 6689 reflections with I > 20(I to yield R1 and wR2 Of 8.12 and

21.74%, respectively.

For 7-5-2CHCl3, the asymmetric unit consists of half an Fe6 ClUSter and a CHCl3 mOleCUle.

Two But groups are disordered and were refined in two parts each. A total of 481 parameters

were refined in the final cycle of refinement using 6037 reflections withlI> 20(1 to yield R1 and

wR2 Of 5.75 and 13.64%, respectively.

For 7-6-4MeCN, the asymmetric unit consists of the Fel2 ClUSter, three whole and two half

perchlorate anions, which are all disordered, and four MeCN molecules, three of which are very

disordered. Program SQUEEZE,68 a part of the PLATON package of crystallographic software,

was used to calculate the solvent disorder area and remove its contribution to the overall

intensity data. The N9 MeCN molecule was not removed by SQUEEZE68 because it is hydrogen-

bonded to the 017-H17 hydroxyl group and not disordered. The Cll perchlorate is hydrogen-

bonded to the opposite hydroxyl group (0 17-H17) through 03 8. While each disordered

perchlorate anion was refined in two parts, the second part of the Cl3 (Cl3') was not complete,

only one O atom being located. The charges are balanced by the fact that the groups occupying

the OS and 027 positions represent a disorder between a water molecule and a carboxyl group.









The others could not be found due to the heavy disorder. Finally, the hydroxyl protons and the

coordinated water protons were obtained from a difference Fourier map and included as riding

on their parent O atoms. A total of 1 165 parameters were refined in the final cycle of refinement

using 10158 reflections withlI> 20(I to yield R1 and wR2 Of 7.88 and 22. 19%, respectively.

Severe disorder problems were encountered for 7-4-CH2 12 and 7-7. For 7-4-CH2 2, the

asymmetric unit consists of an FeS cluster and a dichloromethane molecule; the structure

exhibited much disorder in the benzoate phenyl rings and edte4- grOups, preventing satisfactory

refinement of the structure. However, the core was well observed and showed no disorder. For

7-7, the asymmetric unit consists of half a Fel2 ClUSter and two perchlorate anions; again, the

structure exhibited bad disorder among the peripheral ligands. Despite examination of many

crystals of both compounds, we could not find ones that diffracted well enough to allow data of

sufficient quantity and quality to be obtained for satisfactory structure refinement. Thus, the

structures were refined as far as possible so that we could at least identify the overall structure

and nuclearity of the complexes for comparison with 7-5 and 7-6, which we were able to

successfully do. Knowing the number and arrangement of the Fe atoms in the core was also

essential for the interpretation of the magnetic data of 7-4 and 7-7. We include and briefly

describe the structures of these two complexes in this chapter only for the mentioned purposes;

the metric parameters are unreliable and are not discussed. Unit cell data and details of the

structure refinements for complexes 7-1 to 7-3 are listed in Table 7-1 and 7-4 to 7-7 are listed in

Table 7-2.

7.3 Results and Discussion

7.3.1 Syntheses

In order to make clusters containing Mn ions, it is generally necessary to either oxidize

simple Mn" salts, or use preformed higher oxidation state Mnx clusters. Both of these strategies









have previously proved to be useful routes to a variety of higher nuclearity complexes with

chelating ligands ranging from bidentate to pentadentate.34,35,43,63, 171,172 Therefore, we decided to

employ them both with the potentially hexadentate ligand edteH4. Thus, a variety of reaction

ratios, reagents, and other conditions were investigated. The reaction of edteH4 with

Mn(O2CPh)2 and NEt3 in a 1:3:3 molar ratio in MeOH afforded a reddish-brown solution from

which was subsequently obtained the octanuclear complex

[MnsO3(OH)(OMe>o~)7(2Cth),(edtez)(ete2)(O2CPh) (7-1) in ~20% yield. Its formation is

summarized in eq. 7-1, where atmospheric oxygen gas is assumed to provide the oxidizing

equivalents required to form the mixed-valence 6Mn ", 2Mn" product.

8[Mn(O2CPh)2] + 2edteH4 + MeOH + H20 + 3/2 02 -8PhCO2H +

[MnsO03(OH)(OMe)(O2 CPh)i(edte) (edteH2)] (O2CPh) (7-1)

Complex 7-1 was also obtained, and in a higher yield of ~40%, from the reaction of edteH4

with [Mn30(O2CPh)6 py)2H20] in a 1:1 molar ratio in CH2 12/MeOH. Such trinuclear [Mn30]

clusters have often proved to be very useful starting materials for the synthesis of higher

nuclearity products, some of which have also been new SMMs.38,173 The formation of 7-1 via

this route is summarized in eq. 7-2. The mixed solvent system was needed to ensure adequate

solubility of all reagents, and it also led to methoxide incorporation; no isolable products were

obtained when only CH2 12 WAS used. Small variations in the Mn:edteH4:PhCO2- ratio also gave

complex 7-1, which clearly is a preferred product of these components, and with benzoate. When

[Mn30(O2CR)6 97)3 0,+ ClUSters with other R groups were employed, we were unable to isolate

pure, crystalline materials for satisfactory characterization.

3 [Mn30(O2CPh)6 py)2(H20)] + 2edteH4 + MeOH -Mn2+ + 6py + 8H+ + 2H120 +

10OPhCO2- + [MnsO3(OH)(OMe)(O2CPh)7(edte)(edteH2)] (O2CPh) (7-2)









The reaction of edteH4 with MnCl2-4H20, and NEt3 in 1:2: 1 molar ratio in MeCN/MeOH

gave a brown solution from which was isolated [Mnl204(OH)2(edte)4 16(H20)2] (7-2) in 25%

yield. As found for 7-1, complex 7-2 is mixed-valent, containing 8Mn and 4Mn" ions, and its

formation is summarized in eq. 7-3, again assuming the participation of atmospheric 02.

Increasing or decreasing the amount of edteH4 Or NEt3 alSo gave complex 7-2, but the product

was not as pure. We were also unable to isolate any clean products when we employed an

MeCN/EtOH solvent system.

12MnCl2 + 4edteH4 + 4H20 + 202 -[Mnl204(OH)2(edte)4 16(H20)2] + 18HCI (7-3)

Finally, the reaction of edteH4 with Mn(O2CMe)2 and NEt3 in a 1:2:2 ratio in MeOH,

followed by addition of NaCIO4, gave a dark orange-brown solution from which was

sub sequently i isolated [Mn200s(OH)4(O2CMe)6(edte)6] (CIO4)2 (7-3) in 20% yield. Thi s product i s

once again mixed-valence, containing 12Mn ", 8Mn" ions, and its formation is summarized in

eq. 7-4, with 02 again included as the oxidizing agent.

20Mn(O2CMe)2 + 6edteH4 + 6H20 + 302 -[Mn200s(OH)4(O2CMe)6(edte)62 2+

32MeCO2H + 2MeCO2- (7-4)

A variety of reactions of edteH4 were explored with a number of different Fe(III) starting

materials and under different reagent ratios, solvents and other conditions before the following

successful procedures were identified. The reaction of [Fe30(O2CPh)6(H20)3](NO3) with edteH4

in a ~3:2 molar ratio in CH2C~ followed by layering with Et20 hexanes (1:1 v/v) gave orange

needle-like crystals of [FesO2(O2CPh)7(edte)(H20)] (7-4). Its formation is summarized in eq. 7-5.

The benzoate groups clearly function as proton acceptors facilitating the deprotonation of edteH4

in the absence of added base.









5 [Fe30(O2CPh)6(H20)3]+ + 3edteH4 3 [FesO2(O2CPh)7(edte)(H20)] + 9PhCO2H +

11H20 + 5H' (7-5)

With other chelates such as dmemH,93 we have found that the identity of the Fex product

depends on the carboxylate employed,93 and thus we also explored reactions of edteH4 with other

[Fe30(O2CR)6(H20)3]+ reagents. With pivalate (R = Bur), a related reaction to that which gave

7-4, but with a [Fe30(O2CBut)6(H20)3] to edteH4 mOlar ratio of 1:2 in CHCl3, gave a brown

solution and subsequent isolation of [Fe602(O2CBur~s(edteH)2] (7-5) on layering with pentanes.

The proton acceptors in this reaction are the carboxylate groups and the OH- anions; the

formation of 7-5 is summarized in eq. 7-6.

2[Fe30(O2CBut)6(H20)3](OH) + 2edteH4 [Fe602(O2CBur~s(edteH)2] + 4BurCO2H +

8H20 (7-6)

The same product was also obtained using CH2C 2 aS the solvent, but in poor crystallinity

and decreased yield. We also explored the use of simple Fe(III) salts as reagents, in the presence

of added carboxylate groups as proton acceptors. The reaction of Fe(CIO4)3-6H20 with edteH4

and NaO2CMe-3H20 in a 2: 1:4 ratio in MeCN gave a brown solution from which was obtained

[Fel204(OH)2(edte)4(O2C~)6He)6(2)](CIO4)4 (7-6). Its preparation is summarized in eq. 7-7.

12Fe3+ + 32MeCO2- + 4edteH4 + 8H20 [Fel204(OH)2(edteoCe)4(O2e)6(20)2 4+

26MeCO2H (7-7)

Decreasing the the amount of acetate from 4 to 2 equiv drastically reduced the reaction

yield, as expected from eq. 7-7. Complex 7-6 was also obtained from a MeCN:MeOH solvent

system. However, when the reaction of Fe(CIO4)3-6H20 with edteH4 in a 4: 1 ratio was carried

out in neat EtOH in the absence of NaO2CMe, the product was [Fel204(OH)s(edte)4(H20)2 -

(CIO4)4 (7-7), Obtained as orange needles on layering the solution with CHCl3. COmplex 7-7 is









structurally very similar to 7-6, except that the acetate groups have been replaced by hydroxide

ions videe infra). In a related fashion, the reaction of Fe(NO3)3 with edteH4 and NEt3 in a 2: 1:2

ratio in MeOH gave [Fel1204(OH)s(edte)4(H20)2] (NO3)4 (7-8) OH Vapor diffusion with

tetrahydrofuran. The product was identified by elemental analysis, and IR and magnetic

comparisons with complexes 7-6 and 7-7 videe infra).

It is clear that the described reactions to complexes 7-1 to 7-8 are very complicated and

involve acid/base and redox chemistry, as well as structural fragmentations and rearrangements.

As a result, the reaction solutions likely contain a complicated mixture of several species in

equilibrium. For this reason, we were happy to settle for the relatively low yields of 7-1 to 7-8,

given that the products were reproducibly obtained in a pure, crystalline form from the described

procedures.

7.3.2 Description of Structures

7.3.2.1 Structure of [MnsO3(OH)(OMe)(O2CPh)7(edte)(edteH2)] (O2CPh) (7-1)

A labeled representation of the cation of 7-1 is shown in Figure 7-2, and selected

interatomic distances and angles are summarized in Table A-16. Complex 7-1 crystallizes in the

monoclinic space group P21/n. The core is mixed valence (6Mn"', 2Mnn), with Mn3 and Mn6

being the Mn" ions, and contains a [Mn707] subunit consisting of two distorted [Mn" 3Mn"(pS-

O)(p3-O)(p3-OR)2 5+ cubanes sharing the Mn6-01 edge. This double-cubane unit is additionally

bridged by a pu-OMe~ (021) between Mn7/Mn8 and a pu3-OH~ (O6) between Mn2/Mn4. The latter

CL3-OH~ additionally connects the double-cubane to the eighth Mn atom, Mn3. The edteH22- grOup

is hexadentate-chelating on Mn" atom Mn3, with the two deprotonated alkoxide O atoms, 07

and 010, bridging to Mn2 and Mn4, respectively, and thus the edteH22- grOup is overall pu3-

bridging; oxygen atoms 08 and 09 are protonated. The edte4- grOup is also hexadentate chelating

to a Mn" atom, Mn6, with the four deprotonated alkoxide O atoms all adopting pu3 bridging









modes within the double-cubane, and thus the edte4- grOup is overall py~i-bridging. The

chelating/bridging modes of the py7-edte4- and pu3-edteH22- grOups are shown in Figures 7-3(a) and

(b), respectively. The remaining ligation is provided by seven benzoate groups, five of which are

'I:yl :p-bridging and the remaining two are rl terminal on Mn1 and Mn5.

The oxidation states of the Mn atoms and the protonation levels of Ol-, OH-, OMe- and

OR- groups were determined from a combination of charge balance considerations, inspection of

bond lengths, and bond valence sum (BVS) calculations.48,49,103 BVS values for Mn and O atoms

are listed in Table 7-3 and 7-4, respectively. All the Mn atoms are six-coordinate and display a

Jahn-Teller (JT) elongation, as expected for high spin Mn"' in octahedral geometry, with the JT

axes (shown as thicker black bonds in Figure 7-2) not co-parallel. The Mn" atoms, Mn3 and

Mn6, are seven coordinate. The anion of complex 1 is a PhCO2~ grOup (not shown in Figure 7-2),

which forms an intimate ion-pair with the Mns cation by hydrogen-bonding with 06 of the pu3-

OH- ion (06---029 = 2.580 A+) and 08 of an edteH22- prOtonated alcohol arm (08---028 = 2.584

A+). This also has the effect of raising the BVS of pu3-OH- ion 06 to 1.43, higher than normally

expected for a OH- group (1.0-1.2). The BVS for 08 (1.21) is much less affected by the

hydrogen-bonding (compare with 1.17 for 09), no doubt due to the only monodentate binding of

OS, which thus retains greater basicity and a stronger O-H bond than the pu3-OH- ion.

A number of other Mns complexes have previously been reported. These possess a variety

of metal topologies such as rod-like, serpentine, rectangular, linked Mn4 butterfly units, linked

tetrahedral, etc,67,174-188 but none have possessed the core of complex 7-1, which is

unprecedented. In addition to this novel core structure, another unusual feature of 7-1 is the

presence of its puS-O2- ion, 01. There are only two previous structural types with a puS-O2- i08 in

molecular Mn chemistry, certain Mnl2189,190 and Mnl3191-193 COmplexes.









7.3.2.2 Str uctu re of [Mnl204(OH)2 (edte)4 16 20)2] (7-2)

The labeled structure of 7-2 is shown in Figure 7-4, and selected interatomic distances and

angles are listed in Table A-17. Complex 7-2 crystallizes in the tetragonal space group P4/ncc

with the Mnl2 mOlecule lying on an S4 Symmetry axis and thus only '/ of it in the asymmetric

unit. The structure consists of a [Mn" gMn"4 4-> C-OH)4(p-OH)2 4 3-OR)4pOR1 2+CO

consisting of two near-planar Mn6 layerS sandwiched between three near-planar layers of O

atoms (Figure 7-4, bottom). For the sake of brevity, reference to specific atoms in the following

discussion includes their symmetry-related partners. BVS calculations for the Mn atoms (Table

7-5) identified Mn1, Mn2 and Mn3 as Mn", Mn"' and Mn atoms, respectively. Mn1 and Mn2

are six coordinate while Mn3 is seven coordinate. The four pu4-O2- ions (Ol) together serve to

connect all twelve Mn atoms. Each edte4- grOup is hexadentate-chelating on a Mn"' atom, Mn3,

with each of its deprotonated alkoxide arms bridging to either one (OS, OS, 06) or two (02)

additional Mn atoms. Thus, the edte4- grOups are overall ps5-bridging, as shown in Figure 7-3(c).

Charge balance considerations require that, with eight Mn ", four Mn", four Ol- and four

edte4- grOups, there must be eight additional negatively charged ligands to give neutral complex

7-2. The simplest conclusion is that with eight apparent Cl- ions in the complex, the two pu-O

atoms 04 and 04' bridging Mn2 atoms belong to H20 groups. However, we were unhappy with

this conclusion, being unaware of any precedent in the literature for H20 groups bridging two

Mn atoms; the high Lewis acidity of two Mn would be expected to make the water molecule

in a [Mn" 2(p-OH2)] unit very (Bronsted) acidic (very low pKa) and unlikely to be stable. In

contrast, a water molecule bridging two Mn" atoms is known.194-19 We thus decided to

determine the protonation levels of all O atoms in 7-2 by BVS calculations, and the results are

listed in Table 7-5. The oxide and edte4- O atoms have BVS values of >1.89, confirming them as

completely deprotonated, as concluded above from their bridging modes. In contrast, 04 has a









BVS of only 0.99, as expected for an OH- group. In addition, the Mn2-04 bond length of

2.0098(15) A+ is typical of Mn "-OH- bond lengths in the literature.198-20 This is a more realistic

bridging group between two Mn atoms. Thus, we conclude that 04 and 04' are OH- groups.

This now requires six additional anionic ligands for a neutral molecule, and we suspected that the

S4 Symmetry was masking a disorder between the Cl- and a neutral ligand such as H20 at the

terminal positions (C11). Crystallographic refinement of these terminal Cl atoms was

inconclusive as to whether there was a Cl/H20 disorder, and so we investigated the Cl content of

the molecule more directly with a chlorine elemental analysis. This did indeed give a value less

than expected for eight Cl atoms, whose formula of [Mnl204(H20)2(edte)4 8]] would have a

calculated 14.40% Cl content, much higher than the experimental 11.89%. The latter is,

however, consistent with the expected six Cl- ions that are required for the observed neutrality of

complex 7-2 if 04 is a OH- ion. Thus, we conclude that the correct formula of 7-2 is

[Mnl204(OH)2(edte)4 16(H20)2]. Note that for the reasons already mentioned, we disfavor the

H20 groups being disordered with Cl at the pu-Cl- positions (Cl2) bridging Mn "Mn" pairs, but

this is not as unlikely as water bridging two Mn atoms, and thus cannot be completely ruled

out. Indeed, maybe the two water molecules are disordered amongst the eight bridging and

terminal positions, and this is why the crystallographic refinement is fine with eight Cl atoms.

Recently, a complex apparently identical to complex 7-2 was reported by Zhou et al., but who

instead formulated it as [Mnl204(H20)2(edte)4 8 -201l

There are many other structural types of Mnl2 COmplexes already in the literature, the most

well-studied being the [Mnl2012(O2CR)16(H20)4] (Mnl2) family, which has been extended over

the years to include one, two and three-electron reduced [Mnl2z-z (z = 0 3) versions.12,202-207

Another Mnl2 family of complexes was more recently obtained by reductive aggregation of









MnO4- in MeOH-containing media; this family differs from the previous one in having a central

Mn'V4 rhombus rather than a Mn'V4 tetrahedron.208,209 The remaining Mnl2 COmplexes cover a

variety of other structural types, including loops and more complicated face-sharing cuboidal

units, amongst others.189,210-215

7.3.2.3 Structure of [Mn2008(OH)4(O2 ME)6(edte)6]](CIO4)2 (7-3)

The structure of the cation of 7-3 is shown in Figures 7-5 and 7-6, the latter providing the

atom labeling. Selected interatomic distances and angles are listed in Table A-18. Complex 7-3

crystallizes in the monoclinic space group P21/c with the Mn20 cation lying on a crystallographic

inversion center; again, reference to a specific atom will include its symmetry-related partner.

The cation can be described as consisting of two sets of three edge-sharing [Mn404] cubanes

(Figure 7-5, middle), with the upper and lower sets connected by face-sharing to give a 3 x 2

arrangement of six cubanes. This central [Mnl4016] unit is then attached to three additional Mn

ions at each end by additional O atoms (Figure 7-5, bottom). This gives an overall tube-like

arrangement of twenty Mn atoms inside of which are four Ol- ions. Note that the Mn7 edge-

sharing double-cubane structure of complex 7-1 is a recognizable sub-fragment of the central

[Mnl4016] unit of 7-3, and thus 7-3 can be considered a more extended version of 7-1. The

overall core is thus [Mn 12zMn'sCp6-O)2 4-O)2 3-O)4(p-OH)4 3-R1-R1 +with the two

p6s-O (024), two pU4-O (019) and four pU3-O (020, 022) atoms being 02- ions. The four pu-OH-

groups, 021 and 023, (The BVS values of 021 and 023 are 1.19 and 1.20 respectively) bridge

Mn8/Mn9' and Mn5/Mn9, respectively, and thus provide additional linkages between the

cubanes. The ten pu3-OR and fourteen pu-OR oxygen atoms are provided by the alkoxide arms of

six edte4- grOups. As seen in 7-1 and 7-2, each edte4- grOup binds as a hexadentate chelate to one

Mn and then bridges through its deprotonated alkoxide arms to a number of additional Mn

atoms; four edte4- grOups are overall ps5-bridging, and the remaining two are py~i-bridging, and









these modes are shown in Figures 7-3(c) and (d), respectively. The remaining ligation in the

molecule is provided by six acetate groups, two of which are rl:rl :pu-bridging, two are r12

chelating on Mn1, and two are rl terminal on Mn7.

Inspection of metric parameters and BVS calculations (Table 7-3) indicate that there are

twelve Mn and eight Mn" atoms in the molecule. The BVS values for Mn5 and Mn8 are a little

higher than normally expected for Mn", and those for Mn4 and Mn9 are a little lower than

normally expected for Mn ", so it is possible there is some Mn"/Mn"' static disorder within the

core. All the Mn ions are six-coordinate except Mn5, which is seven coordinate. The JT

elongation axes on six-coordinate Mn atoms are shown as thicker black bonds in Figure 7-6.

There is only one other Mn20 ClUSter in the literature, a complex with benzylphosphonate ligands

reported by Winpenny and coworkers,173 which contains twelve Mn atoms in one plane.

Complex 7-3 is thus structurally very different from this previous example. In addition to the

novel overall structure, there is again, as for 7-1, another unusual feature, namely the presence of

6U 02- ions (024). There are only two previous examples of such a pu6 02- in Mn chemistry,

[MnloO2C 8(thme)6 2- and [Mnls0l4(OMe)14(O2CCurs(MeOH)6] .216-218

7.3.2.4 Str uctu re of [FesO2(O2CP h)7(e dte)(H20)] (7-4)

A labeled representation of 7-4 is shown in Figure 7-7. Complex 7-4 crystallizes in

monoclinic space group P21/c. The core can be described as consisting of a [Fe4 u3-O)2 8+

butterfly like subunit (Fe2, Fe3, Fe4 and Fe5) on the top of which is attached a [Fe~p-OR)4] unit

containing Fel. There is an O atom monoatomically bridging Fel to each of the four Fe atoms of

the butterfly. These four O atoms (04, 06, 010 and 01 1) are the alkoxide arms of edte4- grOup.

The edte4- grOup is hexadentate with the four deprotonated alkoxide O atoms all adopting pu-

bridging modes and thus edte4- grOup is overall ps5-bridging as shown in Figure 7-8(a). Peripheral

ligation about the core is provided by one water molecule on Fe5 and seven benzoates, out of









which five are in rl :rl :p- bridging mode and one is rl terminal on Fe5 and one is r12 chelating on

Fe2. There are relatively few Fe5 clusters reported in literature, and these have Fe5 topologies

such as square pyramid, a centered tetrahedron, and a partial cubane extended at one face by a

partial admantane unit.132-141 However, the only previous compound structurally similar to 7-4 is

[FesO2(OH)(O2CMe)s(hmbp)3 2+ (5-1), where hmbpH = 6-hydroxy methyl-2,2' bipyridine.127 In

5-1 also, there is a [Fe4 u3-O)2 8+ COre with an additional iron atom on top (as in 7-4) but the

precise means by which latter is connected to the Fe4 unit is different from the situation in 7-4.

Specifically, one of the bridging alkoxide in 7-4 is replaced by a hydroxide.

7.3.2.5 Structure of [Fe6O2(O2CBut)s(edteH)2] (7-5)

A labeled representation of 7-5 is shown in Figure 7-9. Selected interatomic distances and

angles are summarized in Table A-19. Complex 7-5 crystallized in monoclinic space group C2/c.

The structure comprises roughly planar arrangement of six Fe atoms and the core can be

described as consisting of two triangular [Fe3 U3-O)]7 units j oined together via six alkoxide edte

O atoms. Specifically Fe2 one triangular unit is bridged to Fe3' of next triangular unit by 04. In

addition, Fe3 and Fe3' are bridged via Ol3 and 013'. Each triangular unit is essentially isosceles,

(Fel---Fe2 = 2.986 A+, Fe2---Fe3 = 3.313 A+ and Fel---Fe3 = 3.344 A+) and essentially planar (the

oxide, 012, is only 0.359 A+ from the Fe3 plane). The two Fe3 triangular units are trans to each

other. The peripheral ligation is provided by 8 pivalates out of which 6 are rl:rl :p- bridging and

two are rl terminal on Fel and Fel'. All the Fe atoms are six-coordinate. Additionally, Fel and

Fe2 of each triangular unit are bridged by alkoxide arm of edteH3- (03), while fourth arm (OS) is

protonated. The BVS for the O atoms of edteH3- is provided in Table 7-6. The protonated oxygen

atom, OS, is involved in intermolecular H-bonding to 01 (pivalate) of next molecule forming

one dimensional chains that run in two directions in lattice. The edteH3- grOup is overall pu4-

bridging as shown in Figure 7-8(b). The core of complex 7-5 is unprecedented in hexanuclear









Fe(III) chemistry. A number of Fe6 ClUSters have been reported in the literature and a recent

listing of these, together with their structural types, is available elsewhere.104 Among these are a

family ofFe6 ClUSters whose cores comprise linked [Fe3(fl3-O)]"+ triangular subunits as in 7-5,

but the two units are bridged by multiple hydroxo or alkoxo groups and overall all these prior

complexes possess core structures different from that of the present complex 7-5.

7.3.2.6 Structure of [Fel204(OH)2(o2 Me)6(edte)4H20)2] (CIO4)4 (7-6)

The labeled structure of cation of 7-6 is shown in Figure 7-10, and selected interatomic

distances and angles are listed in Table A-20. Complex 7-6 crystallizes in the monoclinic space

group C2/c. The structure consists of a [Fe" 12(fl4-O)4(p-O H)2(u-O O2e)Cue)(3-OR)4(O)22

core consisting of two near-planar Fe6 layerS sandwiched between three near-planar layers of O

atoms (Figure 7-10, bottom). All the iron atoms are six coordinate except Fel, Fe3, Fe9 and Fel2

which are seven coordinate. The four pu4-O2- ions (07, 013, 029 and 037) together serve to

connect all twelve Fe atoms. Each edte4- grOup is hexadentate-chelating on a Fe"' atom, Fel,

Fe3, Fe9 and Fel2, with each of its deprotonated alkoxide arms bridging to either one or two

additional Fe atoms. Thus, the edte4- grOups are overall ps5-bridging, as shown in Figure 7-8(c).

The protonation levels of Ol-, OH-, and OR- groups were determined from a combination of

charge balance considerations, inspection of bond lengths, and BVS calculations (Table 7-6).

The edte4- O atoms have BVS values of>1.87, confirming them as completely deprotonated, as

concluded above from their bridging modes. In contrast, 017 and 018 have a BVS of 1.24 and

1.20 as expected for an OH- group. Peripheral ligation is provided by two terminal water

molecules and six acetate groups, out of which 4 are rl:rl :p- bridging and two are rl terminal on

Fe8 and Fel0.

Complex 7-6 is only one of a very few dodecanuclear Fe(III) clusters known in the

literature, of which the maj ority have a wheel or loop structure.219-221 Among the remainder, one









is composed of face-sharing defect cuboidal units in the central fragment of the core, and the

other consists of four edge sharing [Fe3 U3-O)]"+units.91,222 The strcture of complex 7-6 is thus

unprecedented in Fe chemistry, but is similar to Mn cluster 7-2 with the formula

[Mnl204(OH)2(edte)4 16(H20)2] and a mixed-valence Mn" sMn"4 description.168

The labeled structure of the cation of [Fel204(OH)s(edte)4(H20)2](CIO4)4 (7-7) is shown in

Figure 7-11. The core is essentially the same as that of 7-6 except that acetate groups have been

replaced by hydroxide ones. Complex 7-8 gave an elemental analysis consistent with it being the

NO3~ Salt of the same cation as 7-7, and is thus formulated as [Fel204(OH)s(edte)4(H20)2](NO3)4.

This conclusion is also supported by the very similar magnetic properties of 7-7 and 7-8 videe

infra), and indeed the very similar magnetic properties of all three complexes 7-6 to 7-8, which is

consistent with the conclusion that they all possess the same or very similar Fel2 COre structure.

7.3.3 Magnetochemistry

7.3.3.1 De Studies on 7-1 to 7-3

Solid-state, variable-temperature dc magnetic susceptibility data in a 0. 1 T field and in the

5.0-300 K range were collected on powdered microcrystalline samples of 7-1-2H20, 7-2 and 7-

3-5H20 restrained in eicosane. The obtained data are plotted as &Tys Tin Figure 7-12. The 3&T

at 300 K is 26.8, 37.5 and 50.4 cm3Kmol-l for 7-1 to 7-3, respectively. The 300 K value is equal

to or less than the spin-only (g = 2) value of 26.75, 41.5, and 71.0 cm3Kmoll expected for non-

interacting Mn" 6Mn"2, Mn "sMn"4, and Mn" 12Mn's mixed-valence situations of 7-1 to 7-3,

respectively. For 7-1-2H20, 3&T stays essentially constant with decreasing temperature until 25

K and then increases to 32.1 cm3Kmol-l at 8.0 K before dropping slightly to 31.4 cm3Kmol-l at

5.0 K. The &T at the lowest temperatures suggests S = 8 ground state spin with g < 2, as

expected for Mn. For 7-2, 3&T again stays essentially constant with decreasing temperature until

70 K and then decreases smoothly to 26.2 cm3Kmol-l at 5.0 K, which is suggestive of an S = 7









ground state. For 7-3-5H20, 3DT decreases smoothly with decreasing temperature to a minimum

of 28.2 cm3mOl-1K at 35 K and then increases to 36.0 cm3Kmol-l at 5.0 K, which again suggests

an S= 8 ground state.

To confirm the above initial estimates of the ground state spin of the three compounds,

variable-field (H) and -temperature magnetization (M)1 data were collected in the 0.1-7 T and 1.8-

10 K ranges. The resulting data for 7-1-2H20 are plotted in Figure 7-13 (left) as reduced

magnetization (M~NpB) VS. H/T, where Nis Avogadro's number and puB is the Bohr magneton.

The data were fit, using the program M~AGNET,53 described elsewhere.56 The best-fit for 7-

1-2H20 is shown as the solid lines in Figure 7-13 and was obtained with S = 8, g = 2.00 and D =

-0.30 cm- Alternative fits with S = 7 and S = 9 gave unreasonable values of g of 2.28 and 1.78,

respectively. In order to ensure that the true global minimum had been located and to assess the

hardness of the fit, a root-mean-square D vs g error surface for the fit was generated using the

program GRID," which calculates the relative difference between the experimentalM2NpuB data

and those calculated for various combinations ofD and g. This is shown as a 2-D contour plot in

Figure 7-14 (left) covering the D = -0. 10 to -0.50 cm-l and g =1.90 to 2. 10 ranges. Only one

minimum was observed, and this was a relatively soft minimum; we thus estimate the fit

uncertainties as D = -0.30 + 0.01 cm-l and g = 2.00 + 0.02.

For 7-2, we could not obtain a satisfactory fit if data collected at all field values were

employed. In our experience, the usual reason for this is the presence of low-lying excited states

because (i) the excited states are close enough to the ground state and they have a non-zero

Boltzmann population even at the low temperatures used in the magnetization data collection,

and/or (ii) even excited states that are more separated from the ground state but have an S value

greater than that of the ground-state become populated as their larger M~s levels rapidly decrease









in energy in the applied dc magnetic field and approach (or even cross) those of the ground state.

Either (or both) of these two effects will lead to poor fits because the fitting program assumes

population of only the ground state. A large density of low-lying excited states is expected for

higher nuclearity complexes and/or those with a significant content of Mn" atoms, which give

weak exchange couplings. Thus, it is reasonable that such problems are more likely for 7-2 than

for 7-1, given both the higher nuclearity and the higher relative Mn" content of 7-2 vs 7-1. As

described elsewhere,69,70,223,224 One way around effect (ii) is to use only data collected at low

fields. Indeed, a satisfactory fit (solid lines in Figure 7-13, right) was now obtained using data in

fields up to 0.8 T, with fit parameters S = 7, D = -0. 16 cm-l and g = 1.90. Alternative fits with S =

6 and S = 8 gave g = 2.20 and 1.67, respectively. The corresponding error surface vs D and g

(Figure 7-14, right) gives a harder minimum than that for 7-1, with estimated fit uncertainties of

D = -0.16 & 0.01 cm-l and g = 1.90 + 0.01.

For 7-3-5H20, the even higher metal nuclearity and Mn" content again necessitated using

data collected at lower fields in the fit, and in this case a satisfactory fit (solid lines in Figure

7-15) was obtained for data up to 1 T with fit parameters S = 8, g = 1.90 and D = -0. 16 cml

Alternative fits with S = 7 and S = 9 gave g = 2. 16 and 1.70, respectively. The corresponding

error surface vs D and g (Figure 7-15) is similar to that for 7-2 and gives estimated fit

uncertainties ofD = -0.16 & 0.01 cm-l and g = 1.90 + 0.01. The magnetization fits confirmed the

preliminary estimates of the ground state spin S of 7-1 to 7-3.

7.3.3.2 De Studies on 7-4 to 7-8

Solid-state, variable-temperature dc magnetic susceptibility data were collected in a 0. 1 T

field and in the 5.0-300 K range on powdered crystalline samples of 7-4 to 7-8 restrained in

eicosane. The obtained data are plotted as &Tys Tin Figure 7-16. For 7-4, 3&T steadily

decreases from 6.73 cm3Kmol-l at 300 K to 3.88 cm3Kmol-l at 40.0 K, then stays approximately









constant until 25.0 K, and increases slightly to 4.02 cm3Kmoll at 5.0 K. The 300 K value is

much less than the spin-only (g = 2) value of 21.87 cm3Kmoll for five non-interacting Fe(III)

atoms, indicating the presence of strong antiferromagnetic interactions, as expected for oxo-

bridged Fe(III) systems. The 5.0 K value of 4.02 cm3Kmol-l suggests an S = 5/2 ground state

spin. 3nT for 7-5-2CHCl3-4H20 is 11.03 cm3Kmol-l at 300 K, and stays approximately constant

with decreasing temperature to 100 K and then increases to 13.83 cm3Kmol-l at 5 K. &T at 300

K is again much less than the spin-only value of 26.25 cm3Kmoll expected for six non-

interacting Fe(III) ions indicating strong antiferromagnetic interactions. The 5.0 K value of 13.83

cm3Kmol-l suggests an S= 5 ground state spin.

The XhTyvs T plots for the three complexes 7-6 to 7-8 in Figure 7-16 are very similar,

indicating a minimal influence of the peripheral groups and supporting the conclusions above

that they possess similar core structures. For 7-6 to 7-8, XhT steadily decreases from 22.04,

23.37, 20.53 cm3Kmol-l at 300 K to 0.25, 0.51, 0.50 cm3Kmol-l at 5.0 K, respectively. The

change in lOyTwith decreasing temperature and the low value at 5 K are indicative of an S = 0

ground state. The differences in XhTys T for the three complexes are almost certainly just

reflecting small differences in intramolecular exchange coupling constants (J), and perhaps in

ZFS parameters (D) and any intermolecular interactions.

To confirm the initial ground state spin estimates above for 7-4 and 7-5, variable-field (H)

and -temperature magnetization (M)1 data were collected in the 0.1-7.0 T and 1.8-10 K ranges.

The resulting data for 7-4 are plotted in Figure 7-17 as reduced magnetization (M/NpuB) VS HIT,

where Nis Avogadro' s number and puB is the Bohr magneton. The saturation value at the highest

fields and lowest temperatures is ~4.90, as expected for an S = 5/2 ground state and g slightly

less than 2; the saturation value should be gS in the absence of complications from low-lying









excited states. The data were fit, using the programM2AGNET,53 described elsewhere.56 The best

fit for 7-4 is shown as the solid lines in Figure 7-17 (left) and was obtained with S = 5/2 and

either of the two sets of parameters: g = 1.96 and D = 0.58 cm- and g = 1.96 and D = -0.50 cml

Alternative fits with S = 3/2 or 7/2 were rej ected because they gave unreasonable values of g and

D. It is common to obtain two acceptable fits of magnetization data for a given S value, one with

D > 0 and the other with D < 0, since magnetization fits are not very sensitive to the sign ofD.

This was indeed the case for the magnetization fits for both the complexes 7-4 and 7-5. In order

to assess which is the superior fit for these complexes and also to ensure that the true global

minimum had been located in each case, we calculated the root-mean-square error surface for the

fits as a function of D and g using the program GRID.7 For 7-4, the error surface (Figure 7-18,

left) clearly shows the two minima with positive and negative D values, with the fit with

negative D being clearly superior and suggesting that this is the true sign of D. However, it

would require a more sensitive technique such as EPR spectroscopy to confirm this.

The obtained magnetization data for 7-5 are plotted in Figure 7-17 (right) as M/NpuB VS

H/T, and it can be seen to saturate at ~9.29, suggesting an S = 5 ground state and g < 2. The

resulting best fit of the data is shown as the solid lines in Figure 7-17 (right), and was obtained

with S = 5 and either g = 1.90, D = 0.45 cm-l or g = 1.89, D = -0.28 cm- In this case also, the fit

error surface (Figure 7-18, right) clearly shows that the fit with negative D is far superior,

suggesting this to be the true sign of D.

7.3.3.3 Ac Studies on 7-1 to 7-5

To independently confirm the ground state S values, ac susceptibility data was collected on

microcrystalline samples of 7-1 to 7-5 in a 3.5 G ac field. The in-phase (a)n ac susceptibility

signal is invaluable for assessing S without any complications from a dc field,62,69,224 and these

signals for complexes 7-1 to 7-3 at 997 Hz are plotted as y'Tys Tin Figure 7-19. Extrapolation









of the plots to 0 K, from temperatures above ~5 K to avoid the effect of weak intermolecular

interactions (dipolar and superexchange), gives values of ~ 33, ~27 and ~37 cm3Kmoll for 7-1

to 7-3, respectively, corresponding to S = 8, 7 and 8, respectively, with g ~ 1.91, 1.96 and 2.02,

in very satisfying agreement with the conclusions from the fits of the dc magnetization data.

None of the complexes displayed out-of-phase (p")l ac susceptibility peaks above 1.8 K.

There were some very weak signs of the beginning of signals whose peaks would lie well below

1.8 K, and these may correspond to the very small dips in the 3y'T plots of Figure 7-19 (left) at T

< 2 K. However, it is clear that if any of the complexes 7-1 to 7-3 were single-molecule magnets

(SMMs), they would at best have very small barriers to magnetization relaxation. In fact,

complex 7-1, with its combination of S = 8 and D = -0.30 cm- would be predicted to have the

largest barrier of the three complexes, with an upper limit (U) of U = S2|D| = 19.2 cml

However, the true or effective barrier (Uesf) is expected to be significantly less than U, especially

given the low symmetry of the molecule, and it is perhaps not surprising that even if 7-1 were a

SMM, it would be one only at very low temperatures <1 K and thus not a significant new

addition to the family of SMMs.

In-phase ac susceptibility signals for complexes 7-4 and 7-5 at 997 Hz are plotted as 3y'T

vs Tin Figure 7-19 (right). The 3y'Tis essentially temperature independent below 15 K until ~ 4

- 5 K where there is a small decrease that can be assigned to low temperature effects such as ZFS

of the ground state and/or very weak intermolecular interactions. The essentially constant values

at > 5 K of ~4 and ~14 cm3Kmoll for 7-4 and 7-5, respectively, confirm S = 5/2 and 5 ground

states with g < 2, whose spin-only (g = 2.0) values are 4.38 and 15.0 cm3Kmol l, respectively.

Neither complex displayed out-of-phase (p")l ac susceptibility peaks above 1.8 K.









7.3.3.4 Rationalization of the Ground State Spin of 7-4 and 7-5

It is of interest to attempt to rationalize the observed ground state spin values of 7-4 and 7-

5. It is assumed that all Fe2 painWise exchange interactions are antiferromagnetic, as is essentially

always the case for high-spin Fe(III), and there will thus be competing antiferromagnetic

exchange interactions and spin frustration effects within the many Fe3 triangular units in these

complexes. In fact, for complex 7-4, its S= 5/2 ground state can be rationalized in an identical

fashion based on spin frustration as we previously described for complex 5-1 in chapter 5, which

has a similar [Fe5O6] COre topology as 7-4, as stated earlier, and an identical S = 5/2 ground

state.12

For complex 7-5, the spin alignments giving rise to the S = 5 ground state are again not

obvious owing to spin frustration within the triangular units of the Fe6 COre. There are five

inequivalent types of exchange interactions, J12, J13, J23, J23' and J33', the subscripts referring to

the atom labels of Figure 7-9. In Table 7-7, we list the average Fe-O distances and the Fe-O-Fe

angles for bridged Fe2 pairS within the molecule. It is well known that short Fe-O bond distances

and large Fe-O-Fe angles lead to the larger J values.76,112,225,226 In COmplex 7-5, the Fel/Fe3 and

Fe2/Fe3 pairs, with only a single monoatomic bridge, have both the shortest Fe-O superexchange

pathways and the largest Fe-O-Fe angles in the molecule and are thus expected on the basis of

magnetostructural correlations76 to have the strongest J values, in the order of ~3 8 cm- The

Fe2/Fe3' pair, also with a single monoatomic bridge, has a slightly longer Fe-O pathway but still

a large Fe-O-Fe angle; thus, it would also be expected to have a relatively strong Jvalue, in the

~15 cml region. In contrast, the Fel/Fe2 and Fe3/Fe3' pairs, which are now bis-monoatomically

bridged, have Fe-O distances similar to that for Fe2/Fe3' but by far the smallest Fe-O-Fe angles

in the molecule, and would thus be expected to have the weakest J values, in the ~7 cml region.

The estimates given for all the pairs are based on the J values of Fe2 pairS with similar metric










parameters in other Fe(III) clusters. Thus, we conclude that the Fel/Fe2 and Fe3/Fe3' exchange

will be frustrated by the other, stronger interactions, and as a result, the ground state spin

alignments in the molecule will be as shown in Figure 7-20 (top). The spins within Fe2 pairS

monoatomically bridged by a single O atom are aligned antiparallel, whereas those within the

three bis-monoatomically bridged Fe2 pairS are spin frustrated by the other stronger interactions

and forced to align parallel even though their exchange interactions are intrinsically

antiferromagnetic. This situation predicts an S= 5 ground state for 7-5, as experimentally

obtained. Note that it is not easy to formulate reasonable alternative ways of getting an S = 5

ground state, and thus we feel confident on the above proposal. For example, if the Fe2/Fe3'

interaction were considered weak enough to be frustrated by the Fe3/Fe3' interaction, then this

situation would give the spin alignments of Figure 7-20 (bottom) and an S = 0 ground state.

It is difficult to rationalize the S = 0 ground state spin for 7-6 due to the high content of

triangular units and large number of non-zero exchange interactions.

7.4 Conclusions

The initial use of edteH4 has proven to be a useful new route to a variety of novel Mnx and

Fex clusters. It is a hexadentate chelate whose four alcohol groups offer, on deprotonation, the

possibility of each bridging to one or more additional Mn and Fe atoms and thus fostering

formation of high nuclearity products. In the present work, we have described the synthesis and

characterization of new Mns, Mnl2 and Mn20 prOducts, all with unprecedented structural features

and all with significant ground state spin values of S = 7 or 8. Although the core of complex 7-4

is overall similar to that of a previous Fes complex with hmbp-, those of Fe6 COmplex 7-5 and

particularly that of Fel2 COmplex 7-6 are unprecedented in Fe(III) chemistry. The structures of

the cations 7-7 and 7-8 are concluded to be the same given their identical formulations and

almost superimposable magnetic properties. We have also successfully rationalized the S = 5









ground state of 7-6. The combined results emphasize the usefulness of the poly-alcohol-based

chelate edteH4 aS a route to new high-nuclearity products.

Table 7-1. Crystallographic Data for 7-1-2CH2C 2-MeOH, 7-2-6MeCN-V2H20 and 7-3-10MeOH
7-1 7-2 7-3
Formula CsoH94C 4MnsN4030 C52H99C 8Mnl2N14022.5 C82H182C 2Mn20N12066
Fw, g/mola 2172.91 2223.33 3562.05
Space group P21/n P4/ncc P21/c
a, A+ 16.0450(16) 19.7797(12) 17.0570(16)
b, A+ 17.6428(17) 19.7797(12) 25.409(2)
c, A+ 31.896(3) 20.851(2) 15.9322(15)
to 90 90 90
flo 95.425(2) 90 100.463(2)
70 90 90 90
Tr A3 8988.6(15) 8157.7(12) 6790.3(11)
Z 4 4 2
T, K 173(2) 173(2) 173(2)
n, Ab 0.71073 0.71073 0.71073
Peaic, g/cm3 1.577 1.810 1.573
pu, mm-l 1.340 2.125 1.886
R1 c"d 0.0334 0.0464 0.0812
wR2 e 0.0812 0. 1046 0.2174
aIncluding solvate molecules. bGraphite monochromator. c > 20(1. dR1 = C(||8| ||| ) / CI|8|.
w'R2 = [C[w'(F, F,) ] / C[w'(F ) ]]'". w' 1/l[o ( ,) + [(ap)) +bp)], where p = [max (F, O) + 2F, ]/3.










Table 7-2. Crystallographic Data for 7-4-CH2C 2, 7-5-2CHCl3, 7-6-MeCN and 7-7
7-4 7-5 7-6 7-7
Formula C60H59C 2FesN2- C62H116C 6Fe6N4- C60H116C 4Fel2- C40H92C 4Fel2-


021
1494.27
P21/c
21.3735(10)
18.6612(9)
17.7842(8)
90
113.280(1)
90
6515.81


026
1881.39
C2/c
14.211(2)
24.297(2)
25.676(3)
90
94.783(3)
90
8988.6(15)


N12052
2649.61
C2/c
29.590(4)
29.641(4)
23.174(3)
90
104.088(2)
90
19714(5)


N8O46
2233.17
C2/c
30.502(3)
11.9702(11)
30.517(3)
90
111.404(1)
90
10373.7


Fw, g/mola
Space group
a, A
b, A
c, A
to
flo
70
p', A3


Z 4 4 8 4
T, K 173(2) 173(2) 173(2) 173(2)
n, Ab 0.71073 0.71073 0.71073 0.71073
Pealc, g/cm3 -1.414 1.764
pmm- 1.210 1.916
R1 c"d -0.0575 0.0788
wR2 e 0.1364 0.2219
aIncluding solvate molecules. bGraphite monochromator. cI> 2o(1). dR1 = C(||Fo| |F,||) / C|Fol.
wR2 = [C[w(Fo2 F, ) ] / C[w(Fo ) ]]'", w 1/l[o (Fo ) + [(ap)) +bp)], where p = [max (Fo O) + 2F, ]/3.


and 7-3a
7-3


Table 7-3. Bond-valence sums for the Mn atoms of complex 7-1


Atom

Mn 1

Mn2
Mn3
Mn4
Mn5

Mn6
Mn7
Mn8
Mn9

Mnl0


Mn

3.21

3.18
2.01
3.13
3.14

1.96
3.17
3.18


Mn" Mn Mn" Mn" Mn

2.93 3.08 1.85 1.70 1.78

2.91 3.05 3.12 2.90 2.98
1.87 1.92 3.21 2.93 3.08
2.87 3.01 2.99 2.77 2.85
2.87 3.01 2.48 2.26 2.38
1.83 1.86 3.17 2.90 3.04
2.90 3.04 3.22 2.94 3.09


2.91


3.05 2.45 2.24 2.35


3.05 2.79 2.93

1.88 1.75 1.79


a The underlined value is the one closest to the charge for which it was calculated. The oxidation state of a
particular atom can be taken as the nearest whole number to the underlined value.










Table 7-4. Bond-valence sums for the O atoms of complex 7-la
Atom BVS Assignment group
021 1.99 OR~ OMe-

06 1.43 OH~ OH-

015 1.95 OR~ edte4-

016 1.93 OR~ edte4-

017 1.94 OR~ edte4-

018 1.96 OR~ edte4-

07 2.02 OR~ edte4-

08 1.21 ROH edteH22-

09 1.17 ROH edteH22-
010 1.98 OR~ edte4-

aThe BVS values for O atoms of 02-, OH- and H20 groups are typically 1.8-2.0, 1.0-1.2 and 0.2-0.4,
respectively, but can be affected somewhat by hydrogen-bonding.


Table 7-5. Bond-valence sums for the Mn and O atoms of complex 7-2.


Manganese BVS
Mn
2.12
3.09
2.82

Oxygen BVS
assignment
02-
OH
OR
OR
OR
OR


Atom

Mn1
Mn2
Mn3

Atom

01
04
02
03
05
06


Mn"
2.23
3.34
3.05


BVS
1.98
0.99
1.89
1.97
1.97
2.03


Mn'V
2.17
3.22
2.91


group
02-
OH-
edte4-
edte4-
edte4-
edte4-


a The underlined value is the one closest to the charge for which it was calculated. The oxidation state of a
particular atom can be taken as the nearest whole number to the underlined value. aThe BVS values for O
atoms of 02-, OH- and H20 groups are typically 1.8-2.0, 1.0-1.2 and 0.2-0.4, respectively.










Table 7-6. Bond-valence sums for


the O atoms of complex 7-5 and 7-6a


7-5
Atom
BVS assignment group
03 1.82 OR~ edte4-
04 1.94 OR~ edte4-
05 0.90 ROH edteH
013 1.88 OR~ edte4-
7-6
Atom
BVS assignment group
01 1.95 OR~ edte4-
02 1.87 OR~ edte4-
03 1.97 OR~ edte4-
04 1.98 OR~ edte4-
05 0.50 H20 H20
016 0.41 H20 H20
017 1.24 OH~ OH-
018 1.20 OH~ OH-
aThe BVS values for O atoms of 02-, OH- and H20 groups are typically 1.8-2.0, 1.0-1.2 and 0.2-0.4,
respectively.

Table 7-7. Selected Fe-O distances and Fe-O-Fe angles for 7-5
Fe2 pair Avg. Fe-O (Al) Angle (deg)
Fel/Fe3 1.895 123.8

Fe2/Fe3 1.883 123.2

Fe2/Fe3' 2.000 118.7

Fe3/Fe3' 2.026 102.3

Fel/Fe2 1.980 98.1 (avg.)





/N NNH HNN


HOOH HOOHO
dmemH heenH2 OH edteH4 H

Figure 7-1. Structure of ligands.































Figure 7-2. (left) Labeled representation of the cation of 7-1. Hydrogen atoms and phenyl rings
(except for the ipso carbon atoms) have been omitted for clarity. JT axes are shown as
thicker black bonds. (right) The core of 7-1. Color code: Mn ', green; Mn", purple; O,
red; N, blue; C, grey.


(b)


Figure 7-3. Crystallographically established coordination modes of edte4- and edteH22- found in
complexes 7-1 to 7-3. Color code: Mn ", green; Mn", purple; O, red; N, blue; C, grey.


(c) (d)


(a)














































Figure 7-4. (top) Labeled representation of the structure of 7-2. (bottom) The core of 7-2 viewed
along (left) the c axis, and (right) the b axis. Color code: Mn"', green; Mn", purple;
Cl, cyan; O, red; N, blue; C, grey.














174



















































Figure 7-5. (top) The structure of the cation of 7-3. (middle and bottom) The core of 7-3 from
different viewpoints emphasizing the 3 x 2 cubane arrangement. Color code: Mn ',
green; Mn", purple; O, red; N, blue; C, grey.








175














Mnlr


N41



07 (


014


017


Figure 7-6. Labeled representation of the core of 7-3. JT axes are shown as thicker black bonds. Color code: Mn"', green; Mn ,
purple; O, red; N, blue; C, grey.




















Figure 7-7. Labeled representation of the structure of 7-4 with core Fe-O bonds as thick black
lines; only the ipso benzoate carbon atoms are shown. Color code: Fe, green; O, red;
N, blue; C, grey.


~i r-
C1


(a) (b) (c)
Figure 7-8. Crystallographically established coordination modes of edte4- and edteH3- found in
complexes 7-4 to 7-6. Color code: Fe, green; O, red; N, blue; C, grey.

05




03 03Fe3'
012


'13


Figure 7-9. Labeled representation of the structure of 7-5 with core Fe-O bonds as thick black
lines; methyl groups on pivalate groups have been omitted for clarity. Color code: Fe,
green; O, red; N, blue; C, grey.


6- 4L-*:T
c
c +


P-~- f
t ~cLi i7 "
C -JC


















T LFe3 ~ I rer 1

012 033 02
Feli
019


"03 /


Figure 7-10. (top) Labeled representation of the cation of 7-6 with core Fe-O bonds as thick
black lines. (bottom) Core of 7-6. Color code: Fe, green; O, red; N, blue; C, grey.


k/~"-

































Figure 7-11i. Labeled representation of the cation of 7-7. Color code: Fe, green; O, red; N, blue;
C, grey.


60


50




O 30





*7-1
*10 7-2
S7-3


O 50 100 150 200 250 300

T(K)
Figure 7-12. Plots of guTys Tfor complexes 7-1-2H20 (*), 7-2 (m) and 7-3-5H20 (A)
















14 8 -


12 6 +


10 4 -0* 0.4T
*2T v QI 0.5T
a 3T o 06BT
6 4T 087T
-Fitting Fittmng
61 .
5 10 15 20 25 30 0 1 2 3 4 5

HTT(kGlK) H/T(kG/K()

Figure 7-13. Plots of reduced magnetization (M/NpIB) VS H/T for complexes 7-1-2H20 (left) and
7-2 (right). The solid lines are the fits of the data; see the text for the fit parameters.


-0.10


-0.15


-0.20


-0.25


S-0.30


-0.35


-0.40


-0.45


0.00


-0.05


-0.10


-0.15


S-0.20


-0.25


-0.30


-0.35


-0.50 1
1.90


-0.40 b '\~~\\\\\\ ". "**. *' ''.
1 80 1 82 1 84 1 86 1 88 1 90 1 92 1 94 1 96 1 98


1.95 2.00 2.05


Figure. 7- 14.Two-dimensional contour plot of the r.m. s. error surface vs D and g for the
magnetization fit for 7-1-2H20 (left) and 7-2 (right).















-71


-_ =---- E
``



``

,~,`,~~-;'-'';`-;
~';(:,;`(=~~





.801.82 i.84 1.86 188 1.90 1.92 1.94 i 961.98


-0 05




-0.15

-0 20

-0.25

-030

-0 35

6 4 40


*01T
n 0 2T
a 03T
0 4 T
v 05T
D 06T
0 /
m 08T
D 9T
v 1T
Fitting

4 5


0 1 2 3

HIT~kGlK)


Figure 7-15. (left) Plots of reduced magnetization (M/NpBB) VS H/T for complex 7-3-5H20. (right)
Two-dimensional contour plot of the r.m. s. error surface vs D and g for the
magnetization fit for 7-3-5H20.


o
E

E


E
x


* e


. *


. .


250 300


*7-4
o 7-5
a 7-6
S7-7
r 7-8


a
~nr
O



0000


~~b o
is


0 50 100 150 200


T(K)



Figure 7-16. Plots of XnnTys T for complexes 7-4 to 7-8.


















*01T
a 0 5T
23 a 1T
Z T
v3T
21 o 4T

6T
1 vr 7T


* 0.1T
o 0.5 T
a 1T
4 2T
3T
D 4T
e 5T
m 6T
v 7T
- Fitting


0 10 20 30 40 50 0 10 20 30

HIT(kGlK) H/T(ktGlK)


40 50


Figure 7-17. Plot of reduced magnetization (M/NpBB) VS H/T for complexes 7-4 (left) and
7-5-2CHCl3-4H20 (right). The solid lines are the fit of the data; see the text for the fit

parameters.


0.4


E 0.2

0.0









1.92 1.94 1.96 1.98 2.00


1.90


1.86 1.88 1.90 1.92 1.94

9


Figure 7-18. Two-dimensional contour plot of the fitting error surface vs D and g for 7-4 (left)
and 7-5-2CHCl3-4H20 (right).
























0 2 4 6 8 10 12 14 16
T(K)


a 7-2
S7-3

D 2 4 6 8 10 12 14 16
T(K)


Figure 7-19. (left) Plot of XM'TVs. T for complexes 7-1-2H20 (*), 7-2 (m) and 7-3-5H20 (A) at
997 Hz. (right) Plot of XM'TVs. T for complexes 7-4 (A) and 7-5-2CHCl3-4H20 (*) at
997 Hz.


S=5


S=0


Figure 7-20. (top) Spin alignments at the six S = 5/2 Fe(III) atoms of 7-5 rationalizing its S = 5
ground state, based on the arguments given in the text. (bottom) Spin alignments if
the strengths of the Fe2/Fe3' and Fe3/Fe3' couplings were reversed, showing that the
wrong ground state would be obtained.









CHAPTER 8
SINGLE-MOLECULE MAGNETISM AND MAGNETOSTRUCTURAL CORRELATION
WITHIN A FAMILY OF [Mn "loLn"'2] COMPLEXES

8.1 Introduction

Single-molecule magnets (SMMs) are individual molecules that can function as nanoscale

magnetic particles below their blocking temperature (TB). They thus represent a molecular (or

bottom-up) approach to nanoscale magnetic materials, and one that retains all the advantages of

molecular chemistry, particularly monodispersity, solubility, crystallinity, and a periphery of

organic ligands. 13,16,154,227 The SMM property results from a combination of a high-spin ground

state (S) and an easy-axis type magnetoanisotropy (negative zero-field splitting parameter, D)

and can be determined experimentally by the observation of frequency-dependent out-of-phase

(7") signals in ac magnetic susceptibility measurements, and hysteresis loops in magnetization

vs. dc field scans.13,15,19,228 Since the initial discovery of the Mnl2 family of SMMs, a number of

different structural types have been synthesized, the maj ority of which have been Mn clusters

composed completely or partially of Mn atoms, since the Jahn-Teller distortion of high-spin

Mn in near-octahedral geometry is the main source of the molecular anisotropy.21'22' 171,213,228-232

In order to develop new synthetic routes for the synthesis of new SMMs, we and others

have recently been focusing on polynuclear 3d-4f complexes.233-24 The hope has been that the

large anisotropy of most of the Ln"' ions, as well as their often large number of unpaired

electrons, would enhance the SMM property by raising the barrier to magnetization relaxation.

The initial success in 2004 in the synthesis of a mixed-metal, ferromagnetically-coupled Tb2CU2

SMM complex with a slow magnetization relaxation rate,246 prOVided a proof-of-feasibility that

amalgamation of transition metals with anisotropic Ln ions can have a maj or impact on the

resulting relaxation barrier. Since then, this area has steadily built up momentum and there are

now several types of 3d-4f SMMs, most of which are in Mn-Ln chemistry: [Mnl 1Dy4 239,









[Mn2Dy2 240 [Mn6Dy6 241, [Mn6Dy4 234 and [MnllGd2 -247 All these complexes display

frequency-dependent out-of-phase ac susceptibility signals at low temperatures, but it was the

[Mn11Dy4 239 COmplex that first confirmed that such Mn-Ln SMMs can exhibit clear

magnetization hysteresis and quantum tunneling of magnetization (QTM) through the anisotropy

barrier. The first 3d-4f SMMs in Fe chemistry have also now been reported,

[Fe2Ho2(OH)2(teaH)2(O2CPh)4(NO3)2] and [Fe2Dy2(OH)2(teaH)2(O2CPh)6] (teaH3

triethanolamine).245 The total number of examples of 3d-4f SMMs is thus still limited, and new

synthetic methodologies to additional examples are of continuing interest, especially with the

hope that some may have increased magnetization relaxation barriers, i.e. higher blocking

temperatures, TB. These points represent a maj or stimulus of this work.

As part of our continuing interest in the synthesis of new Mn-Ln SMMs, we have explored

the use of 2-hydroxymethylpyridine (hmpH) in mixed-metal reactions. This alcohol containing

group on deprotonation has been a common route to various homometallic Mnx clusters in our

previous work,34,63,211 and we expected that it might also prove a useful route to new Mn-Ln

species. Indeed, we report in this paper our development of a synthetic procedure employing

hmpH that has successfully led to a family of Mn "loLn"'2 isostructural complexes spanning most

Ln ions: Ln = Pr (8-1), Nd (8-2), Sm (8-3), Gd (8-4), Tb (8-5), Dy (8-6), Ho (8-7) and Er (8-8).

In addition to showing that some of these are new 3d-4f SMMs, we also take advantage of the

availability of an isostructural series to map out the 3d-4f exchange coupling (and SMM

properties) as a function of the Ln"' ion employed. It is rare to have such an opportunity in high

nuclearity Ln-containing mixed-metal complexes.23 In general, the large orbital angular

momentum in Ln"' ions, except for Gd"', prevents the convenient use of an isotropic

(Heisenberg) spin Hamiltonian for interpretation of the magnetic properties of polynuclear 3 d-4f









clusters. Costes et al. and Kahn et al. introduced an empirical approach of comparing a 3d-4f

system with an isostructural one in which the Ln"' is replaced by a diamagnetic M"' ion, thus

allowing the effect of the Ln to be factored out.248-250 In 2003, Figuerola et al. extended this to

an Fe" -Ln species by comparison with the isostructural Co" -Ln and Fe" -La"' complexes,

thus factoring out both the individual Fe"' and Ln properties.251 Since then, a variety of such

studies have been carried out on dinuclear, trinuclear, tetranuclear, 1-D and 2-D cyanide-bridged

3d-4f assemblies.233,236,238,252-257 However, this strategy has not been employed as yet on

polynuclear Mn-Ln oxide clusters relevant to the SMM field, and to the search for increased

barriers to magnetization relaxation. In order to do so in the present work, we have also prepared

the isostructural Mn" 10Y"'2 with diamagnetic Y"' for comparison with the Mn-Ln complexes,

which has allowed characterization of the magnetic properties of the Mn tlo portion of the

structure. We herein describe the combined results of this comparative investigation of the

properties of this family of complexes.

8.2 Experimental Section

8.2.1 Syntheses

All preparations were performed under aerobic conditions using reagents and solvents as

received. (NBun4)[Mn402(O2 CPh)9(H20)] was prep ared as previ ou sly de scrib ed. s

[MnloDy20s(O2CPhblo~hhmpp)(NO3)4] (8-6). To a stirred solution of

(NBun4)[Mn402(O2CPh)9(H20)] (0.36 g, 0.23 mmol) in MeOH/MeCN (1/19 mL) was added

Dy(NO3)3-5H20 (0.10 g, 0.23 mmol) followed by hmpH (0.02 mL, 0.23 mmol). The mixture was

stirred for an hour, filtered to remove undissolved solid, and the filtrate layered with Et20. X-ray

quality red-brown crystals of 8-6-3MeCN-MeOH were obtained over a period of one week.

These were collected by filtration, washed with Et20, and dried in vacuo; the yield was 15%.

The dried solid analyzed as solvent free. Anal. Called. (Found) for 8-6 (C106H86Dy2Mn 1oN1oO46)









C, 40.93 (40.50); H, 2.79 (2.90); N, 4.50 (4.24)%. Selected IR data (cm )~: 3434(br), 3063(w),

1707(w), 1605(m), 1566(s), 1487(m), 1385(s), 1308(w), 1228(w), 1175(w), 1157(w), 1069(m),

1051(w), 1027(w), 817(w), 765(w), 718(m), 668(m), 614(w), 548(m), 465(w), 429(w).

[MnloPr20s(O2CPhboo~hhmp)6(NO3)4] (8-1). COmplex 8-1 was prepared following the

same procedure as for 8-6 but with Pr(NO3)3-5H20 (0.095 g, 0.23 mmol). The yield was 25%.

Anal. Called. (Found) for 8-1 (C 106H86N1oMn1oO46Pr2) : C, 41.51 (41.55); H, 2.83 (2.84); N, 4.57

(4.38)%. Selected IR data (cm )~: 3434(br), 3063(w), 1707(w), 1606(m), 1566(m), 1403(s),

1290(w), 1176(w) 1069(m), 1051(w), 1027(w), 820(w), 763(w), 718(m), 661(m), 549(m),

429(w).

[MnloNd20s(O2CPhboo(hm1p)6(NO3)4] (8-2). Complex 8-2 was prepared following the

same procedure as for 8-6 but with Nd(NO3)3-6H20 (0.10 g, 0.23 mmol). The yield was 20%.

Anal. Called. (Found) for 8-2 (C 106H86N1oMn1oO46Nd2) : C, 41.42 (41.77); H, 2.82 (2.88); N, 4.56

(4.76)%. Selected IR data (cm )~: 3446(br), 3063(w), 1698(w), 1607(m), 1566(s), 1473(m),

1401(s), 1315(w), 1291(w), 1175(w), 1157(w), 1069(m), 1050(w), 762(w), 718(m), 660(m),

612(w), 549~>,(m), 460(w), 2()

[MnloSm20s(O2CPhblo(hmp)6(NO3)4] (8-3). COmplex 8-3 was prepared following the

same procedure as for 8-6 but with Sm(NO3)3-6H20 (0.10 g, 0.23 mmol). The yield was 12%.

Anal. Called. (Found) for 8-3 (C 106H86N1oMn1oO46 Sm2): C, 4 1.26 (4 1.3 6); H, 2.8 1 (2.73); N, 4.54

(4.21)%. Selected IR data (cm )~: 3436(br), 3063(w), 1707(w), 1606(m), 1566(s), 1485(m),

1402(s), 1291(w), 1230(w), 1176(w), 1070(w), 1051(m), 1027(w), 819(w), 765(w), 718(m),

663(m), 614(w), 551(m), 468(w).

[MnloGd20s(O2CPhboo(hm1p)6(NO3)4] (8-4). Complex 8-4 was prepared following the

same procedure as for 8-6 but with Gd(NO3)3-6H20 (0.10 g, 0.23 mmol). X-ray crystallography









characterized the obtained crystals as 8-4-3MeCN-MeOH. The yield was 10%. Anal. Called.

(Found) for 8-4 (C 106H86Gd2MnloN1oO46) : C, 41.07 (41.35); H, 2.80 (2.81); N, 4.52 (4.22)%.

[MnloTb20s(O2CPhboo(hm1p)6(NO3)4] (8-5). Complex 8-5 was prepared following the

same procedure as for 8-6 but with Tb(NO3)3-5H20 (0.10 g, 0.23 mmol). The yield was 20%.

Anal. Called. (Found) for 8-5 (C106H86NloMnloO46Tb2): C, 41.03 (41.03); H, 2.79 (2.80); N, 4.51

(4.31)%. Selected IR data (cm )~: 3432(br), 3062(w), 1710(w), 1605(m), 1566(s), 1487(m),

1401(s), 1291(w), 1175(w), 1157(w), 1070(m), 1051(w), 1026(w), 818(w), 765(w), 718(m),

6 6 5(m), 61 6(w), 5 5 0(m), 4 6 5(w).

[MnloHo20s(O2CPhboo(hm1p)6(NO3)4] (8-7). COmplex 8-7 was prepared following the

same procedure as for 8-6 but with Ho(NO3)3-5H20 (0.10 g, 0.23 mmol). The yield was 15%.

Anal. Called. (Found) for 8-7-2H20 (C106H90NloMnloO48Ho2): C, 40.40 (40. 11); H, 2.88 (2.79);

N, 4.44 (4.23)%. Selected IR data (cm )~: 3432(br), 3063(w), 1602(m), 1565(s), 1488(m),

1385(s), 1309(w), 1175(w), 1157(w), 1069(m), 1051(w), 1025(w), 765(w), 718(m), 670(m),

547(m), 466(w).

[MnloEr20s(O2CPhboo~hhmp)6(NO3)4] (8-8). COmplex 8-8 was prepared following the

same procedure as for 8-6 but with Er(NO3)3-5H20 (0.10 g, 0.23 mmol). The yield was 10%.

Anal. Called. (Found) for 8-8-3H20 (C 106H92N1oMn1oO48Er2) : C, 40. 11 (3 9.85); H, 2.92 (2.84); N,

4.41 (4.23)%. Selected IR data (cm )~: 3430(br), 3063(w), 1709(w), 1603(m), 1565(s), 1488(m),

1400(s), 1306(w), 1175(w), 1157(w), 1070(m), 1051(w), 1027(w), 817(w), 765(w), 717(m),

673(m), 612(w), 549(m).

[MnloY20s(O2CPhblo(hmp)6(NO3)4] (s-9). COmplex 8-9 was prepared following the same

procedure as for 8-6 but with Y(NO3)3-6H20 (0.087 g, 0.23 mmol). X-ray crystallography

characterized the obtained crystals as 8-9-4MeCN. The yield was 15%. Anal. Called. (Found) for









8-9 (C106H86NloMnloO46 2) : C, 42.97 (42.69); H, 2.92 (2.83); N, 4.73 (5.17)%. Selected IR data

(cm )>: 3421(br), 3063(w), 1698(w), 1602(m), 1564(s), 1506(m), 1385(s), 1311(w), 1176(w),

1069(m), 1051(w), 765(w), 718(m), 667(m), 547(w), 467(w).

8.2.2 X-ray Crystallography

Data were collected by Dr. Khalil A. Abboud on a Siemens SMART PLATFORM

equipped with a CCD area detector and a graphite monochromator utilizing Mo-Ku radiation (h

= 0.71073 A+). Suitable crystals of 8-4-3MeCN-MeOH, 8-6-3MeCN-MeOH and 8-9-4MeCN were

attached to glass fibers using silicone grease and transferred to a goniostat where they were

cooled to 173 K for data collection. Cell parameters were refined using up to 8192 reflections. A

full sphere of data (1850 frames) was collected using the co-scan method (0.30 frame width). The

first 50 frames were re-measured at the end of the data collection to monitor instrument and

crystal stability (maximum correction on I was <1 %). Absorption corrections by integration

were applied based on measured indexed crystal faces. The structure was solved by the direct

methods in SHELXTL6,258 and refined on F2 USing full-matrix least-squares. The non-H atoms

were treated anisotropically, whereas the H atoms were placed in calculated, ideal positions and

refined as riding on their respective C atoms.

For 8-4-3MeCN-MeOH and 8-6-3MeCN-MeOH, the asymmetric unit consists of a half

MnloGd2 Or MnloDy2 ClUSter, a disordered MeCN in a general position, and a half MeCN

disordered against a half MeOH molecule about an inversion center. A total of 825 (8-4) or 817

(8-6) parameters were refined in the final cycle of refinement using 20616 (8-4) or 10817 (8-6)

reflections withlI> 2o-(I) to give Rl(wR2) of 6.43(17. 13) and 5.08(13.30)% for 8-4 and 8-6,

respectively.

For 8-9-4MeCN, the asymmetric unit consists of a half MnloY2 ClUSter and two MeCN

molecules. The latter are disordered and could not be modeled properly, thus program









SQUEEZE,68 a part of the PLATON package of crystallographic software, was used to calculate

the solvent disorder area and remove its contribution to the overall intensity data. A total of 778

parameters were refined in the final cycle of refinement using 6197 reflections withlI> 2o-(I) to

yield Rl(wR2) of 7. 10(16. 14)%, respectively.

Unit cell data and structure refinement details for 8-4-3MeCN-MeOH, 8-6-3MeCN-MeOH

and 8-9-4MeCN are listed in Table 8-1.

8.3 Results and Discussion

8.3.1 Syntheses

Many synthetic methods to high-nuclearity Mn "-containing clusters and SMMs have

involved the reaction of a chelate with preformed [Mn30(O2CR)6L3 0,+ Oxide-centered triangular

complexes.32,259-262 The chelate has the dual function of encouraging molecular products rather

than polymers, and fostering high-nuclearity products if good binding groups such as alkoxides

are present. In the present work, the preformed tetranuclear Mn '4 COmplex

(NBun4)[Mn402(O2CPh)9(H20)] was employed.ls This has also proven in the past to be a good

stepping-stone to high nuclearity products.179,263,264 Reaction of (NBun4)[Mn402(O2CPh)9(H20)]

with 1 equiv each of hmpH and Ln(NO3)3 (Ln = Pr (8-1), Nd (8-2), Sm (8-3), Gd (8-4), Tb (8-5),

Dy (8-6), Ho (8-7) and Er (8-8)) in MeCN:MeOH led to subsequent isolation of red-brown

crystals of [MnloLn20s(O2CPh)l0(hmp)6(NO3)4] in fair yields of 10-30%. The yields are not

optimized and we were happy to settle for lower yields of well-formed, pure material rather than

add more Et20 co-solvent, which would contaminate the products with by-products such as

NBun4NO3 and others. The reaction is summarized in eq. 8-1.

5 [Mn402(O2CPh)9(H20)]- + 4Ln3+ + 12hmpH + H20 2[MnloLn20s(O2C~hh)10(hmp)6 4+

+ 25PhCO2- + 24H' (8-1)









We had hoped to make the complete series for every Ln (except Pm). However, the later

lanthanides Tm, Yb and Lu gave products that were clearly not isostructural with 8-1 8-8, and

we assume this is related to their decreased size compared with the others. We were also unable

to get the Eu analogue. The Ce reaction suffered from reactions that we assume involve redox

reactions; we have seen elsewhere on multiple occasions that Ce favors the Ce'V oxidation state

in mixed-metal Mn-Ce chemistry involving high oxidation state Mn "/Mn'V.264,265

To benefit the magnetism studies, as stated in the Introduction, we also attempted and

succeeded in preparing the isostructural complex [MnloY20s(O2CPh)10(hmp)6(NO3)4] (-9),

confirmed crystallographically videe infra), by the same route. All these reactions were very

sensitive to the Mn4:hmpH:Ln ratio; other ratios were found to give poor crystallinity and/or

mixtures of products. The mixed MeCN:MeOH solvent system is also very important to give

clean products 8-1 8-9. The structures of representative Ln"' complexes 8-4 and 8-6, and the

Y"' complex 8-9, were determined by X-ray crystallography; all the complexes gave essentially

superimposable IR spectra, and elemental analyses in agreement with the given formulations,

and we conclude on these bases that the complexes are all isostructural. The compounds are air

stable, but interstitial solvent molecules are easily lost during vacuum drying and the solids are

slightly hygroscopic.

8.3.2 Description of Structures

The labeled structures of 8-4, 8-6 and 8-9 are shown in Figure 8-1, and selected

interatomic distances and angles for 8-4, 8-6 and 8-9 are provided in Tables A-21, A-22 and A-

23 respectively. The complexes all crystallize in triclinic space group Pi with the MnloLn2

molecules lying on inversion centers, and their structures are essentially identical except for the

identity of the Ln or Y atoms. Therefore, we will only describe the structure of complex 8-6

here. The complex contains a [MnloDy2 013-O2-)4 04-O2-)4 20+ COre that can be dissected into five









parallel layers of three types with an ABCBA arrangement (Figure 8-2). Layer A is the Dyl1

atom, layer B consists of a triangular Mn" 3 unit (Mn1, Mn2, Mn3), and layer C consists of a

Mn "4 rhombus (Mn4, Mn4', Mn5, Mn5'). Each layer is held together and linked to its

neighboring layers by a combination of four CL3-02-, tWO C04-02- and ten benzoate groups, six of

which are in rl :rl :C-bridging modes and four are in r12:r1 13-bridging modes. Peripheral ligation

is provided by four chelating NO3~ grOups, two on each Dyl atom, and six chelating hmp

groups, one each on the Dyl1 atoms, and Mn2, Mn2', Mn3 and Mn3' of layer B; the hmp~

alkoxide arms also bridge Dy atoms with Mn atoms of layer B, or vice-versa. The Mn and Dy

atoms are six and nine coordinate, respectively, and the Mn oxidation states were determined

using a combination of charge-balance considerations, inspection of metric parameters and bond

valence sum (BVS) calculations (Table 8-2).48 As expected, the Mn" centers exhibit a Jahn-

Teller (JT) distortion, as expected for a high-spin d4 ion in near-octahedral geometry, and takes

the form of the usual axial elongation. The JT elongation axes of the Mn"' atoms of layer C are

approximately parallel to each other (thicker black bonds in Figure 8-2, bottom right). In layer B,

the JT axes of Mn1 and Mn3 are approximately parallel but that of Mn2 is roughly perpendicular

to them (Figure 8-2, bottom left).

The overall structure of these MnloLn2 COmplexes is unprecedented in 3d-4f chemistry, but

that of the central BCB Mnlo subunit is similar to that in the homometallic complexes

[MnloOs(O2CPh)6Ls] (L is the anion of picolinic acid or dibenzoylmethane),266 which also

comprises two Mn3 triangular units above and below a central Mn4 planar unit; however, there

are significant differences in the exact disposition of the three units and in the resulting metric

parameters.









8.3.3 Magnetochemistry

Solid-state variable-temperature dc magnetic susceptibility data were collected on

powdered microcrystalline samples of complexes 8-1 8-9 in the 5.0 300 K range and in a 0. 1

T magnetic field. We will first discuss the data for [MnloY2] (8-9) and [MnloGd2] (8-4): the first

will allow characterization of the magnetic properties of the Mnlo sub-unit alone, and the second

will show the resultant of its exchange coupling with isotropic Gd3+ ions (S = 7/2, sS7/2 free-iOn

term). This will assist the interpretation of the data for the other complexes, which contain

strongly anisotropic Ln"' ions.

8.3.3.1 Complexes 8-9 (MnloY2) and 8-4 (MnloGd2)

The plots of3&Tys Tfor 8-4 and 8-9 are shown in Figure 8-3. For 8-9, the value of&nT

smoothly decreases from 28.4 cm3Kmol-l at 300 K to 10.7 cm3Kmol-l at 5 K. The 300 K value is

slightly less than the spin-only (g = 2) value of 30 cm3Kmol-l for ten Mn ions with non-

interacting metal centers and decreases with decrease in temperature, indicating the presence of

dominant intramolecular antiferromagnetic exchange interactions. The 5.0 K value suggests an S

= 4 ground state. For 8-4, the value of yT decreases from 37.9 cm3Kmol-l at 300 K to 32.4

cm3Kmol-l at 50 K, stays roughly constant down to 15 K, and then decreases rapidly to 27.4

cm3Kmol-l at 5 K. The 300 K value is much less than the spin-only value of 45.7 cm3Kmoll for

ten Mn and two Gd"' non-interacting ions. The value ofnT at 5 K suggests an S = 7 ground

state.

To confirm the above ground state spin estimates, magnetization (M)1 data were collected at

various fields up to 7 T and in the 1.8-10 K temperature range. The resulting data are plotted in

Figure 8-4 as reduced magnetization (M/NpuB) VS H/T, where N is Avogadro' s number and puB is

the Bohr magneton. The data were fit, using the program MAGNET,53 by diagonalization of the

spin Hamiltonian matrix assuming only the ground state is populated, incorporating axial









anisotropy (Di2) and Zeeman terms, and employing a full powder average. The corresponding

spin Hamiltonian is given by eq 8-2, where Sz is the z-axis spin operator, g is the electronic g

Hi = Di2 + ~B U OSH (8-2)

factor, puo is the vacuum permeability, and H is the applied field. The last term in eq 8-2 is the

Zeeman energy associated with the applied magnetic field. An acceptable fit for 8-9 was

obtained with S = 4, g = 2.01 and D = 0.89 cm-l using data collected in the 0. 1-0.8 T field

range; this fit is shown as the solid lines in Figure 8-4 (left). Alternative fits with S = 3 or 5 were

rej ected because they gave unreasonable values of g of 2.67 and 1.65, respectively. We could not

get a good fit when we included data collected at fields higher than 0.8 T, as is often the case for

such polynuclear complexes where there are low-lying excited states as a result of weak

interactions and/or spin frustration effects.62,267 To assess the hardness of the fit and the resulting

uncertainties in g and D, a root-mean square D vs g error surface for the fit was generated for 8-9

using the program GRID," which calculates the difference between the experimental M/NpuB data

and those calculated for various combinations ofD and g. The corresponding D vs g 2-D contour

plot for 8-9 is provided in Figure 8-5 and shows a soft minimum from which we estimate the

uncertainties in the fit parameters to be S = 4, g = 2.01(1), and D = 0.89(5) cml

For MnloGd2 COmplex 8-4, even more difficulty was encountered in obtaining a good fit of

the data, which makes sense given that exchange couplings involving lanthanide ions are very

weak as a result of the buried nature of f orbitals, and thus there will be an even greater number

of very low-lying excited states. The best fit was with S = 7, g = 2.00 and D = -0. 11 cm- and

this is shown as the solid lines in Figure 8-4 (right). The overestimation by the fit of the

experimental data at the lower fields of the plot suggests that the true ground state may be S = 6

with a very low-lying S = 7 excited state that is stabilized below the true ground state by the









applied field. Thus, the only safe conclusion to be drawn for the magnetization fit is that the

ground state of 8-4 is probably either S = 6 or S = 7. We shall return to this point below. The

corresponding D vs g 2-D contour plot for 8-4 is provided in Figure 8-5 (right) and also shows a

soft minimum from which we estimate the uncertainties in the fit parameters to be S = 7, g =

2.00(2), and D = -0. 11(1) cml

To probe the ground states of 8-4 and 8-9 further, and to assess their magnetization

relaxation dynamics, ac susceptibility data were collected in the 1.8-10 K range under a 3.5 G ac

field oscillating at frequencies in the 50-1000 Hz range. The obtained in-phase ac susceptibility

(a)n signals are plotted as y'Tys Tin Figure 8-3 (right) and are invaluable as an additional and

independent means to determine the ground state spin of a molecule without any complications

from a dc field.69,224 fn' fOr 8-9 steadily decreases with decreasing T consistent with

depopulation of low-lying excited states, and extrapolates to just under 10 cm3Kmol-l at 0 K; this

indicates an S = 4 ground state and g~- 2, in agreement with the dc magnetization fit. 3y'T for 8-4

again decreases steadily down to 8 K and then more rapidly, appearing to be heading for ~19

cm3Kmol- Assuming the latter decrease is due to depopulation of excited states, the

extrapolation indicates an S = 6 ground state and a very low-lying excited statess. The ac data

thus confirm the conclusions of the dc magnetization fit that 8-4 has a greater ground state S

value than 8-9; in any case, the precise ground state spin value of 8-4 is not essential for the

analyses described below.

8.3.3.2 Comparison of 8-9 (MnloY2) with 8-4 (MnloGd2), 8-5 (MnloTb2), 8-6 (MRloDy2), 8-7
(MnloHo2), and 8-8 (MnloEr2)

The study of 3 d-4f complexes has been of increased interest to magnetochemists since the

discovery of intrinsic ferromagnetic coupling in the Cu"-Gd"' pair.268,269 In the present work, we

have similarly sought to identify the nature of the interactions between the Mnlo unit and the









Ln atoms of 8-1 8-8. The availability of [MnloY2] (8-9) allOws an empirical approach

analogous to that introduced by Costas et al and Kahn et al.249,250 The magnetic properties of a

[MnloLn2] COmplex are governed by (i) the magnetic properties of the central Mnlo unit resulting

from its many Mn "---Mn"' interactions; (ii) Mn" ---Ln interactions between the Mnlo unit and

the Ln atoms; and (iii) the thermal population of the spin-orbit (Stark) components of the Ln"'

ion. Complex 8-9 has allowed point (i) to be separately elucidated. Insights into the nature of the

Mn "---Ln interactions can now be obtained by comparing the 3yTyvs T data for [MnloLn2] and

[MnloY2] (aSSuming that the three separate Mn" ---Ln interactions at each end of the molecule

are insignificantly different, which is reasonable given that they are all bridged by a CL4-02- and

an hmp- alkoxide arm (Figures 8-1 and 8-2). This is carried out by determining the difference

A~3pT) between &T for a [MnloLn2] ClUSter and [MnloY2] (eq 8-3). A(3yT) will thus reflect both

points (ii) and (iii), except for isotropic Gd"' (8-4) where it directly probes point (ii).

A~3pT) = (JT)MnLn 3nT MnY (8-3)

We do not have access to the analogous Co" 10Ln2 COmplexes, i.e. containing diamagnetic

Co"' in place of the Mn"' of 8-1 8-8, and thus cannot therefore separately determine the exact

Ln "' yTys Tbehavior in these complexes, i.e. point (iii). However, the latter is available in the

literature for Ln"' ions in a variety of coordination environments similar to 8-1 8-8, and thus eq

8-3 can be employed to obtain important insights into the nature of the Mn" ---Ln interactions in

these [MnloLn2] COmplexes.

Application of eq 8-3 to 8-4 and 8-9 gives the A~3pT) shown as a dashed line in Figure 8-3

(left). A~3pT) increases from 9.5 cm3Kmol-l at 300 K to 18.7 cm3Kmol-l at 10 K, and then drops

to 1 6.6 cm3Km ol- at 5 K. Si nc e 3\Tyvs T for Gd"' i s e s senti ally temp erature-i ndep endent4

(Figure 8-6), the steady increase in A~3pT) with decreasing T suggests the Mn" ---Ln" exchange









interactions to be weakly ferromagnetic. This is consistent with the increase in ground state spin

between 8-9 (S = 4) and 8-4 (S = 6 or 7). Note that it is not expected that an S = 11 ground state

should result for 8-4 from weak ferromagnetic Mn "---Gd interactions; there will be extensive

spin frustration within the many triangular units of the Mnlo core, and resulting intermediate spin

alignments, so weak ferromagnetic Mn "---Gd interactions (likely comparable in strength to

some Mn "---Mn"' interactions), are by no means expected to give a simple spin summation.

Indeed, it is even possible that the Mnlo portion of 8-4 may not still be effectively S = 4 once the

extra Mn "---Gd interactions are introduced.

The &T for 8-5 (MnloTb2) inCreaSCS from 46.7 cm3Kmol-l at 300 K to 64.3 cm3Kmol-l at 5

K (Figure 8-7, left). The 300 K value is slightly less than the spin only value of 53.6 cm3Kmoll

expected for ten Mn"' ions and two Tb"' (4fX, 7F6, 23.6 cm3Kmol- ) non-interacting ions (using

the free-ion approximation for Tb) owing to the antiferromagnetic interactions within the Mnlo

unit. The difference A(3\nT) increases sharply with decreasing temperature, from 18.2 cm3Kmoll

at 300 K to 53.5 cm3Kmol-l at 5 K. This indicates ferromagnetic Mn "---Tb interactions. The

3&T for 8-6 (MnloDy2) decreases slightly from 52.3 cm3Kmol-l at 300 K to 48.0 cm3Kmol-l at 50

K and then increases again reaching 51.9 cm3Kmol-l at 5 K (Figure 8-7, right). The 300 K value

is slightly less than the spin only value of 58.3 cm3Kmoll expected for ten Mn"' ions and two

Dy"' ions (4f9, 6H15/2, 28.3 cm3Kmol )~. The difference A(3aT) again increases sharply as the

temperature is lowered, from 23.8 cm3Kmol-l at 300 K to 41.1 cm3Kmol-l at 5 K, again

indicating ferromagnetic Mn" ---Dy"' interactions. Note that&nTys Tplots for isolated Dy"' (and

Tb ") complexes show decreases with decreasing T,236,246,249-25 1,270 and thus they by themselves

(i.e. point (iii) above) cannot be responsible for the increasing A~3pT). The &T for 8-7 decreases

from 51.9 cm3Kmol-l at 300 K to 40.5 cm3Kmol-l at 5 K (Figure 8-8, left). The 300 K value is









again slightly less than the spin only value of 58. 1 cm3Kmoll expected for ten Mn"' ions and two

Ho"' ions (4flo, 5Is, 28.1 cm3Kmol- ). The A~3yT) increases from 23.5 cm3Kmol-l at 300 K to

only 30.0 cm3Kmol-l at 5 K. Thus, the increase in A~3yT) is much smaller than those for 8-5 and

8-6, and this makes it difficult to conclude with safety the nature of the Mn" ---Ho"' interactions.

The 3yT for 8-8 decreases continuously from 49.5 cm3Kmol-l at 300 K to 21.3 cm3Kmol-l at 5 K

(Figure 6d). The 300 K value is less than the spin only value of 52.9 cm3Kmoll expected for ten

Mn ions and two Er ions (4fll, 4 15/2, 22.9 cm3Kmol- ). The A~3yT) stays approximately

constant at 21.1 cm3Kmoll from 300 K to 100 K and then drops to 10.5 cm3Kmol-l suggesting

weak antiferromagnetic Mn "---Er"' interactions in this complex.

8.3.3.3 Comparison of 8-9 (MnloY2) with 8-1 (MnloPr2), 8-2 (MnloNd2 8-3 (MR10Sm2)

For these three complexes with early lanthanides, the 3&Tys Tplots were essentially

identical (Figure 8-9). The 3&T at 300 K is 27.4, 28.4, and 26.8 cm3Kmoll for 8-1, 8-2 and 8-3,

respectively. In each case, 3C&Tthen decreases with decreasing temperature to 14.6, 15.2, and

16.0 cm3Kmol l, respectively, at 5 K. The 3CyTys Tbehaviors of 8-1 8-3 are thus essentially

parallel to that of 8-9, except at the lowest temperatures. Consequently, A(3yT) is small and

negative at higher temperatures, and then slightly positive at lower temperatures. Since we do not

know the exact 3&Tys T behaviors of the Ln ions in the absence of the Mn ', we cannot draw

any safe conclusions from these small magnitudes of A(y3\T) about the exact sign of Mn ---Ln"'

In other work, the interactions of these early lanthanide ions with transition metals have been

found to be typically antiferromagnetic. 248-251,254,271

The combined results suggest that complex 8-5 contains ferromagnetic Mn "---Tb"'

interactions, as is likely also the case in 8-4 and 8-6, but the data are not clear enough to come to

safe conclusions for the other complexes. It would require access to the corresponding MloLn2

complexes with M = Co"' or other diamagnetic ion to probe this point further. It should also be









reiterated that the exchange interactions within the Mnlo unit may be slightly altered as the Ln"

changes (as a result of small changes to bond distances and angles), and that this and the

introduction of Mn" ---Ln interactions will complicate further the spin frustration effects. Thus,

small changes to A(p3\T) are not considered reliable indicators of the sign of the Mn" ---Ln"

interactions.

8.3.3.4 Out-of-Phase ac Susceptibility Signals and Magnetization Hysteresis Loops

As stated in the Introduction, one of the obj ectives of the present work was to obtain new

SMMs. We have thus explored all complexes by ac susceptibility studies. The in-phase signals

for 8-4 and 8-9 were shown in Figure 8-3 (right); their out-of-phase (p~") signals are shown in

Figure 8-10. Complex 8-9 shows only the merest hint of a frequency-dependent signal down to

1.8 K, the operating minimum of our SQUID. Complex 8-4 shows a stronger but still very weak

signal. Since the upper limit (U) to the magnetization relaxation barrier is given by S2|D| and (S2_

%)|1D| for integer and half-integer spin, respectively, the difference between 8-4 and 8-9 must

reflect a bigger barrier as a result of the former' s increased ground state spin isotropic Gd is

not expected to significantly increase the anisotropy. Similarly very weak signals were also

observed for 8-1 8-3 (Figure 8-11). Thus, these complexes are at best only poor SMMs with

very small relaxation barriers. However, much more encouraging results were observed for 8-5 -

8-7, where the Ln ions bring both a large spin and large anisotropy to the molecules. In Figure

8-12 are shown the yn'Tys T and y" vs Tplots for these complexes, and two points should be

noted: (a) in each case, the 3y'T at 5K is essentially identical to that in the corresponding dc 3yT

vs Tplot (Figure 8-7) showing that the latter was not unduly affected by the dc field used; and

(b) in each case, below ~3 K there is a frequency-dependent drop in 3y'T and a strong and

frequency-dependent 3y" peak, approximately one order of magnitude larger than those in 8-4.









These data suggest that 8-5 8-7 are SMMs with significant relaxation barriers, and thus their

magnetization vector cannot relax fast enough to stay in phase with the oscillating ac field.

The ac 3yllvs T plots can be used as a source of relaxation rate vs T kinetic data for

determining the true or effective energy barrier (URe) to magnetization relaxation, because at the

3y" peak maximum the relaxation rate (1/z, where z is the relaxation time) is equal to the angular

frequency (2xnu) of the ac field. The obtained data for 8-5 8-7 are shown as Arrhenius plots in

Figure 8-13, based on the Arrhenius Law of eq 8-4, where k is the Boltzmann constant and to is

the pre-exponential factor. The fit of the data for 8-6 to eq 8-4 (solid line in Figure 8-13) gave

t = to exp (URe/kT) (8-4)

UAff ~ 39 K and to~- 1 x 10~1 s. A similar analysis for 8-7 gave UAs~ 41 K and to~- 3 x 10-12 S.

Because the ac 3y"l vs T data were over a small temperature range (~0.4 K), these values must be

considered only approximations. The 3y"l peaks for 8-5 are at lower temperatures and thus over

too small a range for a meaningful plot, but its UAs was comparable with those for 8-6 and 8-7.

Nevertheless, the to values appear smaller than those typically seen for transition metal SMMs

(10- -10-9 S) but are comparable with other 3 d-4f ones.247

Because we were worried that the above Ugffand to values were not very accurate owing to

the small Trange employed, we carried out a more accurate analysis for representative complex

8-6 by supplementing the ac data with additional relaxation rate vs T data down to 0.04 K

obtained from dc magnetization decay vs time measurements on a single crystal of 8-6-3MeCN

-MeOH. The sample's magnetization was first saturated in one direction at~- 5 K with a large

applied dc field, the Twas then decreased to a chosen value in the 0.04 1.6 K range, the field

removed, and the magnetization decay monitored with time. The obtained magnetization vs time

plots are characteristic of a distribution of relaxation barriers, and are shown in Figure 8-14









(left). The dc decay and ac 3y"l data were combined and used to construct an Arrhenius plot over

a wider T range (Figure 8-14, right). Fitting of the data in the thermally activated region gave f

= 30 K and to = 6 x 10-10 s. Below ~0.5K, the z vs 1/T plot deviates from linearity as the

thermally activated relaxation rate diminishes and the relaxation via quantum tunneling through

the anisotropy barrier becomes more dominant. Eventually at T< 0.2 K, it becomes essentially

temperature-independent, as expected for the relaxation now being exclusively by quantum

tunneling through the lowest lying M~s (MJ) levels. Such quantum tunneling was first observed

for a 3d-4f complex in a [MnllDy4] COmplex.21 The Uefr value of 30 K for 8-6 is the highest yet

reported for a Mn-Ln SMM. Previous examples of [Mn11Dy4 ,239 [Mn2Dy2 ,240 [Mn6Dy4 ,234 and

[MnllGd2 247 SMMs have been reported to have U~values of 9.3, 15, 16 and 18.4 K

respectively.

It should be noted that the U~value of 39 K obtained for 8-6 using just the ac 3y"l data is

very different from the more reliable 30 K obtained from the combined dc and ac data. These

results thus represent an important caveat that inaccuracies of a large magnitude can be

introduced by determining the barrier U~from too small a temperature range. Inversely, it

should be accepted that when there is indeed no choice but to determine U~values over a

limited Trange, then large inaccuracies are likely.

To confirm whether these complexes are truly SMMs, magnetization vs dc field sweeps

were carried out on single crystals of representative complex 8-6-3MeCN-MeOH.153 Hysteresis

loops were observed below 1.6 K (Figure 8-15), whose coercivity increases with decreasing

temperature and increasing scan rate, as expected for the superparamagnet-like properties of a

SMM. The loops are dominated by a large step at zero field due to quantum tunneling of

magnetization (QTM) through the anisotropy barrier. The large step is indicative of fast QTM









rates, as is typical of low symmetry molecules. Steps at other field positions are barely visible,

and such smearing out is typical of broadening effects from a distribution of molecular

environments, as already concluded to be present in the magnetization decay vs time plots

mentioned above. Such distributions are due to solvent and/or ligand disorder, and the presence

of low-lying excited states.

8.4 Conclusions

The reaction of a preformed Mn4 ClUSter, hmpH and simple Ln salts has provided entry

into a new family of 3d-4f [MnloLn2] ClUSters for most of the Ln ions. Three representative

crystal structures have shown the complexes to be isostructural, including the corresponding

[MnloLn2] analOgue with diamagnetic Y" Comparisons of the combined dc and ac magnetic

susceptibility studies have allowed important insights into the effect of the Ln"' ions on the

magnetic properties. Complex 8-4, containing isotropic Gd"' with S = 7/2 has demonstrated an

increase in the spin compared with [MnloY2] COmplex 8-9 as a result of net ferromagnetic

interactions with the Mnlo core. This is reflected in the appearance of out-of-phase ac signals

indicating the slow relaxation of a SMM. The result demonstrates that as long as the anisotropy

of a Mn x, unit is sufficient, the large spin of Gd"' can boost the spin S if couplings are

ferromagnetic and thus improve the SMM properties. The other Ln ions also bring significant

anisotropy to the table, but the couplings for early lanthanide ions Pr, Nd and Sm do not improve

the SMM properties above those of 8-9, presumably due to their small spins and possibly

antiferromagnetic couplings. The later lanthanide ions Tb, Dy and Ho give much more

encouraging results, however, with all three compounds exhibiting strong out-of-phase signals

above 1.8 K. This can be attributed to both their significant spin and anisotropy, and the

observance of ferromagnetic Mn---Ln coupling between them and the central Mnlo unit; the

similar ac behavior between 8-7 and 8-5/8-6 suggests the coupling really is ferromagnetic in each









case. The calculated Ugfvalue of 30 K for representative complex 8-6 is the highest yet for a

Mn-Ln SMM. Single-crystal studies using a micro-SQUID on representative complex 8-6

confirm the SMM property by the clear observation of hysteresis loops.

The present work thus emphasizes that synthesis of Mn" -Ln complexes incorporating

Tb, Dy and Ho is a somewhat promising approach to higher-barrier SMMs. We have reported

elsewhere preliminary results on a related family of MnllLn4 COmplexes, and showed there that

the Dy analogue again is an SMM exhibiting magnetization hysteresis. However, the barrier Ugf

in that case was only 9.3 K, ~30% of that in 8-6. This and related reports of Mn-Ln

complexes234,239-241 SeTVe to emphasize that merely incorporating one or more anisotropic Ln"'

ions into a Mn "-containing cluster, SMM or otherwise, will not automatically increase barriers

and thus switch on or improve the SMM property. The nature of the Mn-Ln coupling, the

presence of low-lying excited states, the QTM rates, the symmetry of the molecule, and other

factors, all impact the observed magnetization relaxation barrier. Nevertheless, the present work

does emphasize that it is possible to get barriers from Mn-Ln species that are fairly high, akin to

those of two-electron reduced versions of the prototypical SMM family,

[Mnl2012(O2CR)16(H20)4 2-, which also show out-of-phase ac peaks in the 2-4 K range.203,272









Table 8-1. Crystallographic data for 8-4-3MeCN-MeOH, 8-6-3MeCN-MeOH and 8-9-4MeCN.
8-4 8-6 8-9
Formula C1 13H99MnloGd2N13047 C1 13H99MnloDy2N13047 C114H98MnloY2N14046
FW, g/mola 3245.94 3265.46 3127.28
Space group Pi Pi Pi
a, A+ 14.7083(7) 14.7358(11) 14.737(3)
b, A+ 15.2173(7) 15.2150(12) 15.080(3)
c, A+ 16.7604(8) 16.6441(13) 16.569(3)
a, a 67.414(2) 67.6290(10) 67.175(4)
P, a 65.549(2) 65.6580(10) 65.668(4)
Y, a 87.627(2) 87.6360(10) 87.374(4)
I', K3 3122.0(3) 3114.3(4) 3063.9(10)
Z 1 1 1
T, K 173(2) 173(2) 173(2)
n, Ab 0.71073 0.71073 0.71073
Pcalc, g/cm3 1.726 1.741 1.712
pu, mm-l 2.112 2.252 2.024
R1 c~d 0.0643 0.0508 0.0710
wR2 e 0.1713 0.1330 0.1614
aIncluding solvate molecules. bGraphite monochromator. cI> 2o(1). dR1 = C(||Fo| |F'c||)l / Fol. ew'R2 =
[C[w'(Fo2 F, ) ] / C[w'(Fo ) ]]'" 2. M 1/[o (Fo ) + [(ap)) +bp)], where p = [max (Fo O) + 2F, ]/3.


Table 8-2. Bond-valence sums for the Mn atoms of complex 8-4, 8-6 and 8-9 a
8-4 8-6 8-9
Mn" Mn"' Mn' Mn" Mn"' Mn' Mn" Mn"' Mn'

Mn1 3.09 2.83 2.97 3.05 2.79 2.93 3.13 2.87 3.01
Mn2 3.33 3.08 3.18 3.32 3.07 3.17 3.33 3.08 3.18


Mn3 3.31


3.06 3.16 3.31 3.06 3.16 3.34 3.09 3.18


Mn4 3.15 2.88 3.03 3.16 2.89 3.03 3.22 2.95 3.09
Mn5 3.06 2.80 2.94 3.07 2.81 2.94 3.09 2.83 2.97

a The underlined value is the one closest to the charge for which it was calculated. The oxidation state of a
particular atom can be taken as the nearest whole number to the underlined value




























































Figure 8-1. The structures of 8-4 (top), 8-6 (middle), and 8-9 (bottom). Hydrogen atoms and
phenyl rings (except for the ipso carbon atoms) have been omitted for clarity. Color
code: Gd, cyan, Dy, yellow; Y, pink; Mn, green; O, red; N, blue; C, grey.


205












































Figure 8-2. (top) Centrosymmetric core of 8-6 emphasizing the ABCBA layer structure. (bottom)
The B (left) and C (right) layers showing the Jahn-Teller elongation axes as thicker
black bonds. Color code: Dy, yellow; Mn, green; O, red; N, blue; C, grey.


(Mn,,Y,1
(Mn,,,Gd,]
,


~''


~pppCf~L~d~L~PLYLY~d~~b~~~d~~~~PP~Pd~~~~~


r r

v [MnmGd ]





0 50 100 150 200 250 300


0 2 4 6 8 10 12 14 16


T/K UIK
Figure 8-3. (left) Plot of3&Tys Tfor complexes 8-4 and 8-9 at 0. 1 T. The difference, A 3~T), is
shown as the dashed line. (right) Plot of in-phase ac susceptibility )3n as 3y'Tyvs T
for complexes 8-4 and 8-9.





206




















* 0.1 T



* 0.5 T
o 0.6 T
* 0.7 T
v 0.8 T
- Fitting


4 5 0 1 2 3 4 5

HTllkGK1


0 1 2 3
HT /lkGK


Figure 8-4. Plot of reduced magnetization (M~NpB) VS H/ Tfor complexes (left) 8-9 and (right)
8-4. See the text for the fit parameters.


-0.30 -eP
1.90


1.95


2.00


2.05


2.10


1.96 1.98 2.00 2.02 2.04


Figure 8-5. Two-dimensional contour plot of the error surface for the D vs g fit for (left) 8-9 and

(right) 8-4. The asterisk indicates the point of minimum error (best fit).


* 0.4T
4 0.5T
v 0.6T
* c.nT
- Fitting








































i [Mn ,TD















0 50 200 150 200 250 300


0 50 100 180 200 250 300
T/K


*** a *


* d(NO ),


0 50 100 150 200 250
T/K


300 0 50 100 150 200 250
T/K


Figure 8-6. Plots of dc ynTys T for (left) Gd(NO3)3.6H20 and (right) Dy(NO3)3.5H20.






*****


.. 40

E
30 t


-| \


___--


[IBEl
~__a~pa__a ---


.~'~,;D?;]
B [MnlpYI


Figure 8-7. Plots of3&Tys Tfor 8-9 and (left) 8-5 and (right) 8-6. In each case, the difference,
A( ~T), is shown as a dashed line.


60 -


50 -


40

O 3





20
1-


60-


50
-
40
o





" 0



0-


11
~1
IYYY*



_C_


J' ~ [Mn,~oj [Mn,,Y,]
[Mn,,~Hoj-[Mn,,,Yj





,**

a


1


[MoEr ]-[Mn ,Y,]


___


0 50 100 150 200 250 300
T/K


0 50 100 150 200 250 300
T/K


Figure 8-8. Plots of3C&Tys Tfor 8-9 and (left) 8-7 and (right) 8-8. In each case, the difference,
A~3pT), is shown as a dashed line.




208










































-*- 997 Hz
-0- 250 Hz
-A- 50 Hz


da- .CqbL-ka~IC~ .


30








10


, [ MnloPJ I n ]Y


0 60 100 150 200 250 300
T/rK



Figure 8-9. Plots of3~uTys Tfor 8-1, 8-2, 8-3 and 8-9. The dashed line is the difference, A( ~T),
between 8-3 and 8-9; the others are similar.


10~
0-



S0.6 -
o
E
04
E 042 -



0.0 -


-0.2 '


0.6

0.5

0.4

E 03
E
& 0.2

0.1

0.0

-0 1


24 6 810
T/K


2 4 68 10


Figure 8-10. Plots of out-of-phase 3C,"lvs Tac susceptibility data for (left) 8-9 (MnloY2), and

(right) 8-4 (MnloGd2)


1.0 -


0.8


-0,6 -

E



H 0.2


0.0


-0.2


1.0


0.8


S0.6


mE 0.4




0.0


-0.2


2 4 6810


2 4 6810
T/K


Figure 8-11i. Plots of out-of-phase 3g,"lvs Tac
(right) 8-2 (MnloNd2).


susceptibility data for (left) 8-1 (MnloPr2), and


-e250 H:
-*- 50 Hz













[MnoTby

606


404

20


0-ir





[MlnoDy~
S404
-*- 997 Hz
30 a 250 Hz
50 Hz







40
[MnoHo~


302



20 -1~nI1


10


2 4 6 8 10
T/KC


Figure 8-12. Plots of in-phase, 10'Tys T, and out-of-phase jO"I vs T ac susceptibility data for
(top) 8-5 (MnloTb2), (middle) 8-6 (MnloDy2), and (bottom) 8-7 (MnloHo2).

9.5

9.0 1 2, I soHoal

8.5 A IrvI ]hn .vTb

8.0






6.0


0 40 0.42 0,44 0 46 0,48 0,50 0,52
T IK'

Figure 8-13. Plot of relaxation rate vs reciprocal temperature for 8-5 -8-7.













10 6

10"

10~

10-

103

10


0.3


S0.2


0.1


0


t (s)


1/T (UIK)


Figure 8-14. (left) Magnetization (M)1 vs. time decay plots in zero dc field for 8-6-3MeCN
MeOH. The magnetization is normalized to its saturation value, Ms (right) Plot of
relaxation time vs 1/T for 8-6 using combined ac 3,"l and magnetization decay data.


-1 -0.5 0 0.5 1 -1 -0.5 0 0.5 1
IPoH (T) PoH (T)


Figure 8-15. Single-crystal magnetization (M)1 vs dc field (H) hysteresis loops for a single crystal
of 8-6-3MeCN-MeOH at different scan rates (left) and temperatures (right).









CHAPTER 9
A FOURTH ISOLATED OXIDATION LEVEL OF THE [Mnl2012(O2CR)16(H20)4] FAMILY
OF SINGLE MOLECULE MAGNETS

9.1 Introduction

Single-molecule magnets (SMMs) are molecules that possess a significant barrier (vs kT)

to reorientation of their magnetization (magnetic moment) vector as a result of the combination

of a large ground state spin (S) and Ising (easy-axis) magnetoanisotropy (negative axial zero-

Hield splitting parameter (D).13 As such, they represent a molecular (bottom-up) approach to

nanomagnetism. The first SMM was [Mnl2012(O2CMe)16(H20)4] -2HO2CMe-4H2015,273,274

(Mnl2-Ac; 4Mn ',8Mn"') and many more have since been synthesized.21,22,231,275 Although

complexes displaying SMM behavior are known for a variety of 3d, 4d, 4f and mixed-metal

complexes,9,20,58,79,86,1 18,162, 163,173,190,271,276-282 manganese carboxylate clusters have proven to be

the most fruitful source of SMMs.21,22,32 Using only a limited palette of ligands and starting

materials, a wide range of Mn SMMs has been obtained with their nuclearities ranging from 2 to

84.164,172,283 Amongst the known Mn SMMs, the Mnl2 family continues to be attractive for study

as a result of its ease of preparation, stability, ready modification in a variety of ways, high

ground state spin (S = 10) and anisotropy, and the access to derivatives that crystallize in high

symmetry (tetragonal) space groups.14,284,285

The various modifications of the Mnl2 family of SMMs that have been accomplished to

date have proven extremely useful for a myriad of reasons and studies, and have permitted great

advances in our knowledge and understanding of Mnl2 COmplexes and the SMM phenomenon in

general. In this regard, carboxylate sub stitution273,286-288 represented a big step forward because it

provided an extremely useful and convenient means of accessing other carboxylate analogues,

which provided benefits such as isotopic labeling, tunability of redox properties and increased

solubility in a variety of organic solvents. One of the most informative impacts of the latter two









points was the observation of multiple, reversible redox processes and the subsequent generation

and isolation of one-electron reduced complexes i.e. salts of the [Mnl2012(O2CR)16(H20)4]

anion, abbreviated [Mnl2]-. The crystal structures of such salts revealed minimal change to the

structure on reduction, with the added electron localized on an outer, formerly Mn atom giving

a trapped-valence Mn'V4Mn "'Mn" situation.202 The [Mnl2]~ Salts allowed an assessment of the

structural, magnetic and spectroscopic consequences of changing the electron count, as well as

allowing the study of the differences in quantum properties due to the integer vs half-integer S

value, since [Mnl2]~ Salts have an S = 9V/2 ground state.202,289-291 The subsequent introduction of

carboxylates with more electron-withdrawing substituents into the Mnl2 COmplexes made two-

electron reduction easier and led to the successful generation and isolation of two-electron

reduced [Mnl2012(O2CR)16(H20)4 2- COmplexes, [Mnl2 2-, Such as salts of

[Mnl2012(O2CCHCl2)16(H20)4] .203 The [Mnl2 2- anion was again found to be trapped-valence,

with a Mn'V4Mn"'6Mn"2 Oxidation state description, and the spin was found to be S = 10, the

same as the Mnl2 parent compound.

The above efforts had thus provided the Mnl2 family of SMMs in three oxidation states,

providing a wealth of comparative chemical and physical data. So much so that it was clearly

desirable to extend this family to a fourth oxidation level if at all possible. The three-electron

reduction of Mnl2 COmplexes is in fact observable in the cyclic voltammetry,203 and so we

decided to pursue the generation and isolation of this oxidation state. Indeed, this effort has been

successful, and we herein report the synthesis and characterization of the [Mnl2 3- Salts

(N~Prn4)3 [Mnl12012(O2CCHCl2)1 6(H20)4] and (N\Me4)3[Mnl12012(O2CCHCl2)1 6(H20)4 -207









9.2 Experimental Section


9.2.1 Syntheses

All manipulations were performed under aerobic conditions using materials as received,

except if otherwise noted. [Mnl2012(O2CCHCl2)16(H20] (9-1) was prepared as described

elsewhere.203

(NPrn4) [MR12012(O2 CC2)16 H20)4] (9-2). Solid NPrn4I (0.03 g, 0. 1 mmol) was added to a

stirred dark brown solution of complex 9-1 (0.30 g, 0. 10 mmol) in MeCN (15 mL). The resulting

solution was stirred for 4 hours with no noticeable color change. After 4 hours, hexanes (20 mL)

were added causing the formation of two phases, and the mixture shaken to facilitate extraction

of 12 into the hexanes phase. The hexanes layer was then removed, and the extraction process

repeated a few more times until the hexanes layer was colorless. The two layers were then

separated and the MeCN solution evaporated to dryness. The residue was dissolved in MeCN (10

mL), and Et20/hexanes (1:1 v/v, 20 mL) added. The resulting microcrystalline product was

isolated and dried in vacuo. Yield, 70 %. Anal. Called (Found) for 9-2-MeCN

(C46HssN2Mnl2048C 32): C, 17.28 (17.60); H, 1.73 (1.62); N, 0.87 (0.53) %.

(NPr"4)2[MR12012(O2CC2 1(20)16204 (9-3). COmplex 9-3 was synthesized following the

same procedure as for complex 9-2, except that two equivalents of NPrn4I (0.06 g, 0.2 mmol)

were employed and the reaction mixture was stirred for 10 hours. Yield, 65 %. Anal. Called

(Found) for 9-3-MeCN (CasH83N3Mnl2048C 32): C, 20.59(20.82); H, 2.47(2.26); N, 1.24(0.88) %.

(NPr"4)3[MR12012(O2CC2 1(20)16204 (9-4). Complex 9-4 was synthesized following the

same procedure as for complex 9-2, except that three equivalents of NPrn4I (0.09 g, 0.3 mmol)

were employed. The reaction mixture was stirred for 40 hours and the product was not

recrystallized. Yield, 85 %. Anal. Called (Found) for 9-4 (C68HlosN3Mnl2048C 32): C, 23.14

(22.85); H, 3.08 (2.78); N, 1.19 (1.16)%/.









(NMe4)3 [MR12012(O2 CC 2)16 H 20)4] (9-5). Complex 9-5 was synthesized following the

same procedure as for complex 9-4, except that three equivalents of NMe4I (0.06 g, 0.3 mmol)

were employed and the reaction mixture was stirred for 48 hours. Yield, 80 %. Anal. Called

(Found) for 9-5-MeCN (C46H63N4Mnl2048 132): C, 17.08(16.92); H, 1.96(1.90); N, 1.73(1.78) %.

9.3 Results and Discussion

9.3.1 Syntheses

Electrochemical studies on various [Mnl2012(O2CR)16(H20] complexes have revealed a

rich redox chemistry involving several quasi-reversible oxidation and reduction

processes.202,203,292 In addition, the redox potentials are, as expected, very sensitive to the

electron-withdrawing and -donating ability of the carboxylate ligand. For example, E1/2 (VS

ferrrocene) for the first reduction varies by almost a volt from 0.91 V for the R = CHCl2 COmplex

to 0.00 V for the R = p-C6H40Me complex. The particularly high electron-withdrawi ng

capability of the R = CHCl2 grOup, as reflected in the very low pKa of 1.48 for CHCl2CO2H,

brought the second reduction potential to 0.61 V (Figure 9-1), well within the reducing capability

of our preferred reducing agent, iodide (0. 14 V vs ferrocene in MeCN),293 and this led to the

subsequent successful generation and isolation of (PPh4)2[Mnl2012(O2CCHCl2)16(H20)4]

reported elsewhere.203 Similarly for the R = C6F5 substituent, which has also been used for the

synthesi s of the two-electron reduced complex (NMe4)2 [Mnl2012z(O2 6sFs) 16(H20)4 272 fTOm the

reaction of [Mnl2012(O2 6sFs)16(H20)4] with two equivalents of T~.

For the present work, we chose to employ the R = CHCl2 carboxylate complex because it

has a particularly well resolved third one-electron reduction in the cyclic voltammogram (CV)

and differential pulse voltammogram (DPV) (Figure 9-1), and one that is still within the reducing

capability of T~. A fourth, clearly irreversible reduction at ~0. 1 V represented a potential problem,

so we avoided the use of an excess of reducing agent beyond the stoichiometric three









equivalents. Thus, complex 9-1 was treated with three equivalents of NMe4I, NPrn4I, NBun4I and

PPh4I in MeCN for different lengths of time; the formation of 12 WAS confirmed by its extraction

into a hexane phase. It was found that longer reaction time of >40 h were required to give

complete conversion of [Mnl2] to [Mnl2 3-, as established by subsequent characterization of the

product; these are much longer times than routinely employed for the [Mnl2]- and [Mnl2 2-

complexes.203,272 Samples of [Mnl2 3- Salts that were analytically pure (and subsequently shown

by magnetism studies to be pure [Mnl2 3-) were obtained with the NMe4+ and NPrn + cations, but

we were not satisfied with the purity of the NBun + and PPh4' Salts or their prolonged stability in

the solid state videe infra). Thus, we used the NMe4+ and NPrn4' Salts for the detailed studies

below. In addition, for better comparisons of [Mnl2z-z (z = 0 3) complexes with the same cation,

we also prepared the NPrn + salts of the [Mnl2]- and [Mnl2 2- COmplexes. The transformations of

9-1 into 9-2 to 9-4 are summarized by general eq. 9-1, where z = 1, 2 or 3.

[Mnl2012(O2CR)16(H20)4] + z T [Mnl2012(O2CR)16(H20)4 z- + z/2 It (9-1)


It soon became apparent that the [Mnl2 3- anion is far less stable in solution than [Mnl2]

and [Mnl2 2-. Numerous attempts to grow crystals of a [Mnl2 3- Salt with a variety of cations and

under various crystallization conditions were all unsuccessful, giving amorphous powders and/or

crystals that turned out to be the [Mnl2 2- Salt on analysis and magnetic examination. However, in

reality a crystal structure would not have told us anything that we did not feel we already knew

about the [Mnl2 3- anion from previous observations of what happens to the structure of a Mnl2

complex on one- and two-electron reduction. The most important structural question in these

other complexes had been where does(do) the added electron(s) go, and the answer was on the

outer ring of Mn atoms. This is summarized in Figure 9-2, which shows the distribution of Mn

oxidation states within the [Mnl2012] COres of these compounds. The neutral Mnl2 (Figure 9-2,









top) has four central Mn atoms within a non-planar ring of eight outer Mn atoms. The latter

divide by symmetry into two classes, and addition of one or two extra electrons leads to

localization of these electrons onto Mn atoms of only one class leading to their conversion to

Mn", giving Mn '4Mn"'7Mn" and Mn V4Mn"'6Mn"2 Oxidation state descriptions, respectively.

This was established from the crystal structures of multiple [Mnl2]- and [Mnl2 2-

complexes,202,203,272,29 1 and i s shown in the two central figures of Figure 9-2. Thi s counter-

intuitive preferential reduction of a Mn"' rather than a Mn was rationalized on the basis that

reduction of a central Mn would convert it into a Mn atom that would show a characteristic

Jahn-Teller distortion, as expected for a high-spin d4 COnfiguration (and exhibited by the outer

Mn atoms). This would introduce strain into the relatively rigid central Mn404 cubane, and so

reduction of an outer Mn becomes thermodynamically preferred since it causes no significant

structural perturbation. We are thus certain that the third added electron in the [Mnl2 3-

complexes also has added to a formerly Mn atom of the same symmetry class, giving the

Mn V4Mn"'SMn"3 Situation depicted in Figure 9-2, bottom.

On the basis of the above arguments, the decreased stability of the [Mnl2 3- anion in

solution compared with the [Mnl2z-z (z = 0 2) is perhaps not surprising given the now high

content of Mn" in a complex that still contains four Mn atoms. It is reasonable that such a

species would be unstable to structural degradation initiated perhaps by the liability of the Mn"

centers and/or intramolecular redox transitions. This would also rationalize that observation in

Figure 9-1 that the four-electron reduced species [Mnl2 4-, which would be expected to be

Mn V4Mn"'4Mn"4, rapidly degrades even on the electrochemical timescale and thus does not

show a well-formed peak in the DPV or even a reversible CV wave on the faster CV timescale.









9.3.2 Magnetochemistry

9.3.2.1 De Studies

Solid-state, variable-temperature dc magnetic susceptibility data in a 0. 1 T field and in the

5.0-300 K range were collected on powdered crystalline samples of complexes 9-2 to 9-5,

restrained in eicosane to prevent torquing. The obtained data are plotted as 3&Tys T in Figure 9-

3. The &Tvalues for 9-2, 9-3, 9-4 and 9-5 slowly increase from 22.7, 21.8, 21.5 and 22.3 cm3 K

mol-l at 300 K to a maximum of 46.2, 50.1i, 38.8, and 38. 1 cm3 K mol-l at 10 K, respectively, and

then decrease at lower temperatures due to Zeeman effects from the applied field, any weak

intermolecular interactions, etc. The &Tys Tprofiles of 9-2 and 9-3 are essentially identical to

those of previously reported [Mnl2]- and [Mnl2 2- COmplexes.203,272,291 Their maxima of 46.2 and

50. 1 cm3 K mol-l at 10 K are in agreement with S = 19/2 and S = 10 ground states, respectively,

and g < 2 as expected for Mn. This is in agreement with the ground states found in previous work

for [Mnl2]- and [Mnl2 2- COmplexes.203,272,291 The calculated, spin-only (g = 2) values are 49.9

and 55.0 cm3 K mol-l for S = 19/2 and S = 10, respectively.

The &Tys T profiles of 9-4 and 9-5 are essentially superimposable with each other

throughout the whole temperature range, and with those of 9-2 and 9-3 in the 100-300 K range.

Below 100 K, they diverge from those of the latter, and reach maxima significantly below those

of 9-2 and 9-3. This shows that 9-2 and 9-5 are indeed at a different oxidation level from either

9-2 or 9-3, and also that they have a smaller ground state S value than them. Remembering that a

Mn'V4Mn'5Mn"3 COmplex must have a half-integer ground state, then the 3&T maxima at 10 K

of 38.8 and 38.1 cm3 K mol-l suggest that 9-4 and 9-5 have an S = 17/2 ground state with g < 2;

the spin-only (g = 2) value is 40.4 cm3 K moll

Confirmation of the above preliminary conclusion was sought from fits of magnetization

(M)1 data collected on complexes 9-4 and 9-5 in the 0.1 4 T and 1.8 10 K ranges. The obtained









data are shown as reduced magnetization (M~NpB) VS H/ Tplots in Figure 9-4, where N is

Avogadro's number and puB is the Bohr magneton. The data were fit using the program

M~AGNET,53 described elsewhere.56 The best fits for 9-4 and 9-5 are shown as the solid lines in

Figure 9-4, and the fit parameters were S = 17/2, D = -0.25 cm- g = 1.91 for 9-4, and S = 17/2,

D = -0.23 cm- g = 1.90 for 9-5. Fits of the data with S = 15/2 or 19/2 gave unreasonable g

values of 2.21 and 1.72, respectively, and were therefore discounted. For a comparison of data

for complexes with different degrees of reduction but with the same cation, we also collected

variable-temperature and -field magnetization data for complexes 9-2 and 9-3; the corresponding

(M~NpB) VS H/ Tplots and fits are provided in Figure 9-5. The fit parameters were S = 19/2, D = -

0.35 cm- g = 1.95 for 9-2, and S = 10, D = -0.28 cm- g = 1.98 for 9-3. The obtained ground

state S values of 9-2 and 9-3 are the same as those previously found for several other [Mnl2]- and

[Mnl12 2- COmplexes. 203,272,291

To confirm that the obtained fit minima were the true global minima and to assess the

hardness of the fit, a root-mean square D vs g error surface for the fit was generated for

representative complex 9-4 using the program GRID," which calculates the relative difference

between the experimental (M~NpB) data and those calculated for various combinations of D and

g. This is shown as a 2-D contour plot in Figure 9-6 covering the D = -0. 10 to -0.50 cm-l and g =

1.86 to 1.98 ranges. Only one minimum was observed, and this was a relatively soft minimum;

we thus estimate the fitting uncertainties as D = -0.25 & 0.01 cm-l and g = 1.91 & 0.01.

9.3.2.2 Comparison of the Magnetic Properties of the [Mnl2 z- (z = 0 3) Family

The combined results for complexes 9-2 to 9-5, as well as those for neutral complex

9-1,203 are collected in Table 9-1. Considering first the S values, it is well known that the spin

ground state changes very little on one- and two-electron reduction, from S= 10 to S= 19/2 and

then back to S = 10 along the series 9-1 (z = 0), 9-2 (z = 1) and 9-3 (z = 2), respectively. Thus,









the Mnl2 COre acts almost as a 'spin buffer', picking up electrons with little change to the ground

state S value. However, on three-electron reduction to complexes 9-4 and 9-5, there is a more

significant change to S = 17/2. This is no doubt due to the increased Mn" content and the general

weakening of many of the exchange interactions in the core. However, the [Mnl2012] COre is a

complicated one with many symmetry inequivalent exchange interactions, many of them

competing and it is thus not easy to provide a rationalization of the S = 17/2 ground state, as

indeed it has not been possible in the past to rationalize those of the [Mnl2]- and [Mnl2 2-

complexes either.

The g values given in Table 9-1 are provided only for completeness and should not be

taken as particularly accurate. It is well known that fits of bulk magnetization data are not a good

way to obtain accurate g values. While we prefer to quote the actual values obtained by having

the g value as a free parameter, rather than fixing it at a more realistic value at or near 2.0, we do

not attempt to draw any conclusions from resulting differences in g. It would require studies with

a more sensitive technique such as EPR spectroscopy to provide more accurate g values.

In contrast to the S value, the axial zero-field splitting parameter D does exhibit a

monotonic change with the extent of reduction; there is a clear decrease in |D| with progressive

one-electron reduction. This is exactly as expected because the molecular anisotropy, as gauged

by the magnitude of |D|, is the proj section of the single-ion Mn anisotropies onto the molecular

anisotropy axis. Mn and Mn" are relatively isotropic ions, and the primary contributions to the

molecular D are thus the Mn ions, which are significantly Jahn-Teller distorted. Since

reduction involves addition of electrons onto formerly Mn" centers converting them to Mn", the

greater the extent of reduction, the fewer will be the remaining Mn ions, and the smaller will

thus be the molecular anisotropy |D|. This assumes other factors remain the same, such as the









overall structure of the Mnl2 COmplex, and the relative orientation of the Mn JT axes essentially

parallel to the molecular z axis. It should be added that D values in Table 9-1 have been obtained

by fitting the magnetization data with the rhombic (transverse) zero-field splitting parameter (E)

fixed at E = 0. In fact, these complexes do not have axial symmetry, and E is unlikely to be

exactly zero. In our experience, however, magnetization fits are usually not very sensitive to E,

and the D values in Table 9-1 are therefore expected to be reasonable, especially for assessment

of relative magnitudes within a series, as here. Nevertheless, for information purposes, we

provide in Figure 9-6, the fit of the magnetization data of 9-4 as a function of D and E, with g

held constant at 2.0; the fit is shown as a contour plot of the error surface. The best-fit parameters

are D = -0.24 cm-l and |E| = 0.065 cm- The D value changes very slightly (from 0.25 cml

obtained with E = 0), while the non-zero E is consistent with the low symmetry of a three-

electron reduced [Mnl2 3- COmplex.

The final entries in Table 9-1 for each compound are the values of the U, the anisotropy

barrier to magnetization relaxation, whose upper limit is given by S2|D|. In practice, the true or

effective barrier (Ugf) is smaller than this upper limit because the magnetization vector need not

go over the top of the barrier but can tunnel through its upper regions via higher-lying M~s levels.

This quantum tunneling of magnetization (QTM) is a characteristic of all SMMs. Since |D|

monotonically decreases with reduction, whereas the S stays roughly the same or decreases, then

it would be expected that Uwould decrease with reduction, and this is what is indeed seen. The

Ufor [Mnl2 3- COmplexes 9-4 and 9-5, coming from their S = 17/2 spin and a |D| value that has

been decreased but is still reasonable, is still relatively large, and even decreased by QTM might

still be sufficient for them to function as SMMs. In order to explore whether these [Mnl2 3-










complexes might indeed exhibit slow relaxation; we investigated their magnetization dynamics

using ac susceptibility.

9.3.2.3 Ac Studies

In ac studies, a weak field (1 5 G) oscillating at a particular frequency, typically up to

1500 Hz, is applied to a sample to probe the dynamics of the magnetization relaxation. If the

magnetization vector can relax fast enough to keep up with the oscillating field, then there is no

imaginary (out-of-phase) susceptibility signal (7,"), and the real (in-phase) susceptibility (yu') is

equal to the dc susceptibility. However, if the barrier to magnetization relaxation is significant

compared to thermal energy (kT), then there is a non-zero X," signal and the in-phase signal

decreases. In addition, the X," signal will be frequency-dependent. The ac susceptibilities of

[Mnl2z-" (z = 0 2) complexes 9-1 to 9-3 have been previously reported, but they were re-

measured here for better comparison with those of 9-4 and 9-5 under identical conditions.

The ac susceptibilities for complexes 9-1 to 9-5 were collected on microcrystalline samples

in a 3.5 G ac field, and the obtained data for complexes 9-4 and 9-5 at representative frequencies

of 50, 250 and 1000 Hz are shown in Figures 9-7 and 9-8, respectively, as yn'Tys T and 3y"l vs T

plots. The in-phase (3p'T) ac signal is invaluable as an additional and independent means to

determine the ground state spin of a molecule, without any complications from a dc field.62,69,224

Inspection of Figures 9-7 and 9-8 shows that the yn'Tvalues are essentially temperature

independent down to ~ 5 K, below which they show decreases due to slow relaxation videe

infra). The temperature independent 3y'T shows that only the spin ground state of the molecule is

populated at these temperatures, and can be used to calculate its S value. The 3y'T values of 40.5

cm3 K mol-l and 39.5 cm3 K moll for 9-4 and 9-5, respectively, correspond to S = 17/2, g = 2.00,

and S = 17/2, g = 1.98, in very satisfying agreement with the conclusions from the fits of the dc

magnetization data discussed above. Note that for S = 15/2 or 19/2 states, a yn'Tvalue of ~40









cm3 K mol-l would require g values of 2.24 and 1.79, which are unreasonable for Mn. We

conclude that [Mnl2 3- COmplexes 9-4 and 9-5 are confirmed to possess S= 17/2 ground states.

Below ~5K, the in-phase 3y'T signals for 9-4 and 9-5 in Figures 9-7 and 9-8 exhibit a

frequency-dependent decrease concomitant with the appearance of frequency-dependent out-of-

phase (p")' signals. This is indicative of the onset of slow magnetization relaxation relative to

the ac field, i.e. the magnetization vector can no longer relax fast enough to stay in-phase with

the oscillating field. This is the characteristic superparamagnet-like behavior of a SMM, and

parallels that previously observed for the other oxidation levels of the [Mnl2z-z (z = 0 2) family.

On the basis of the comparative data presented in Table 9-1, the appearance of the 3y"l signals at

very low temperatures of ~2.5 K and below are as expected for a barrier to magnetization

relaxation in [Mnl2 3- COmplexes 9-4 and 9-5 being smaller than those in [Mnl2 z- (z = 0 2)

complexes. This is emphasized by the comparative ac data presented in Figure 9-9, which shows

the 3y" signals for complexes 9-1 to 9-4 at equivalent frequencies of 50, 250 and 1000 Hz. In

each case, the 3y"l signals are frequency-dependent and exhibit a monotonic shift to lower

temperatures with increasing reduction: 6 8 K for 9-1 [Mnl2]; 4 6 K for 9-2 [Mnl2]-; 2 4 K

for 9-3 [Mnl2 2-; and I 2.5 K for 9-4 [Mnl2 3-. For better comparisons at identical ac frequencies,

the 3y" signals for 9-1 to 9-4 at 50 and 1000 Hz are plotted together in Figure 9-10, bottom and

top, respectively. A clear shift to lower temperature is seen with each reduction step. The

combined data in Figures 9-9 and 9-10 are thus perfectly consistent with the conclusions from

the data in Table 9-1 and the discussion above, since the barrier to magnetization relaxation

scales with S2|D|, and either one or both of these quantities decrease with each one-electron

increase in the extent of reduction.









Note that it is not expected that that there should be a linear decrease in barrier with

reduction, since there are so many factors that determine the actual magnitude of the true or

effective barrier, UEg, including S, D, the rhombic ZFS parameter (E), fourth order spin

Hamiltonian parameters, precise QTM rate and tunneling channel (i.e., which Ms levels are

involved), spin-phonon coupling strengths, and others. Thus, there are too many parameters that

contribute to the observed URe to permit a more quantitative comparison between different

oxidation levels. Since we have not been able to obtain single-crystals, micro-SQUID hysteresis

measurements could not be performed. Note also that the 3gs," signals for complex 9-1 in Figures

9-9 and 9-10 also exhibit a weaker signal at lower temperatures, which is due to a faster-relaxing

form arising from a different Jahn-Teller isomer, i.e. a form in which one of the Mn Jahn-Teller

isomers is abnormally oriented towards a bridging oxide ion in the molecule.205,294,295 These

isomeric forms are known to possess smaller barriers to magnetization relaxation and thus to

exhibit their 3s," signals at lower temperatures.

9.4 Conclusions

The Mnl2 family of SMMs has been successfully extended to four isolated oxidation states

by the three-electron reduction of [Mnl2012(O2CCHCl2)16(H20)4] to

(N43[Mnl201I2(O2CCHCl2) 16(H20)4] (R = Me, Prn) with NR4I. The [Mnl2 3- COmplexes are

unstable in solution, which has prevented us from obtaining crystals suitable for X-ray

crystallography, but this is not unduly disappointing, because it is clear on the basis of the

structural characterization of the three other Mnl2 Oxidation states that the third electron will

have added to an outer, formerly Mn"' ion giving a Mn V4Mn"'SMn"3 trapped-valence situation.

We do not believe it will be possible to extend the Mnl2 family of SMMs to five members by

four-electron reduction, given the instability demonstrated by the putative [Mnl2 4- Species in the

electrochemical studies.









The [Mnl2 3- COmplexes 9-4 and 9-5 both possess a half-integer S = 17/2 ground state, and

a |D| value smaller than that for the [Mnl2 3- COmplex 9-4 which supports the above assertion

that the third added electron is localized on a formerly Mn ion, since the Jahn-Teller distorted

Mn"' ions are the primary source of the molecular anisotropy. As a result of the decreased S and

D relative to the other Mnl2 Oxidation states, the barrier to magnetization relaxation Uis also

smaller than for the other oxidation states, but is still sufficient to yield out-of-phase (3p") ac

susceptibility signals indicative of slow magnetization relaxation. Thus, we conclude that the

[Mnl2 3- COmplexes 9-4 and 9-5 are SMMs. Note that the observation of y" signals is indicative

of a SMM but not normally sufficient proof of one. In this case, however, the well-established

fact that the 3y"l ac signals for the other Mnl2 Oxidation states are correctly identifying SMMs, as

proven by single-crystal hysteresis studies, leaves little doubt that these same signals for the

[Mnl2 3- COmplexes 9-4 and 9-5 are also due to SMMs. Thus, although we do not have single

crystals with which to carry out micro-SQUID studies down to 0.04 K in order to observe

magnetization hysteresis loops for 9-4 and 9-5, there seems little doubt that the available data are

indicating that the Mnl2 family of SMMs now spans four oxidation levels.










Table 9-1. Magnetism Data for [Mnl2z-z (z = 0 3) Complexes 9-1 to 9-5.
z = 0 (9-1) z = 1- (9-2) z = 2- (9-3) z = 3- (9-4) z = 3- (9-5)
S 10 19/2 10 17/2 17/2
g 1.86 1.95 1.98 1.91 1.90
D/cml -0.45 -0.35 -0.28 -0.25 -0.23
D/K -0.65 -0.50 -0.40 -0.35 -0.33
U/Ka 65 45 40 25 24
a Calculated as S2|D| for 9-1 and 9-3, and as (S2-W4)|D| for 9-2, 9-4 and 9-5.







0.03

0.24
0.56
0.86




0.34 15 *
I 0.6450A
0.95

0.91 0.61 0.29

~10 pA







1.6 1.2 0.8 0.4 0.0 -0.4
Potential (V)
Figure 9-1. Cyclic voltammogram at 100 mV/s (top) and differential pulse voltammogram at 20
mV/s (bottom) for complex 9-1 in MeCN containing 0.1 M NBun4PF6 aS supporting
electrolyte. The indicated potentials are vs ferrocene.





















[MnJ-









[Mni 2-
















Figure 9-2. Proposed structural core of 9-1, 9-2, 9-3 and 9-4. Color code: Mn'V, green; Mn ,
blue; Mn", yellow, O, red.


[MnJ


































I I I I I I I


o 0).1T

a 1T
* 2T
S3T
v 4T


e0.17
a 0.ST
n 1T
S2T
e 3T
v 4T


0 50 100 150

T(K)

Figure 9-3. Plot of guTys Tfor 9-2 to 9-5.


200 250 300


0 5 10 15 20 25 0 5 10 15

HITT(ktG/K) HIT(ktGIK)


20 25


Figure 9-4. Plot of reduced magnetization (M/NpuB) VS H/T for 9-4 (left) and 9-5 (right). The solid
lines are the fit of the data; see the text for the fit parameters.


r 9-2
S9-3

*9-5


i

or
or
or
or
PI
o



























6 4T

FTtn
a 6


I I
o o
~tt~

*r I



-oaa
p
~1T
() ZT
o 3T
I4T

O m
riT


16





10







s-


O 10 20 30 40 50 0 10 20 30 40 50

HIT(kGlK) HIT(kGlK)

Figure 9-5. Plot of reduced magnetization (M/NpuB) VS H/T for 9-2 (left) and 9-3 (right).


-0.10J


-0.15


-0.20


-0.25


- 0.30


-0.35


-0.40


-0.45


0).04


0a 00L
-0 40


-0.50 N
1.86


-0.35 -0.30 -0.25 -0.20 -0.15 -0.10


1.88 1.90 1.92 1.94 1.96


g D~cm-1)

Figure 9-6. (left)Two-dimensional contour plot of the error surface for the D vs g fit for complex

9-4. (right) Two-dimensional contour plot of the root-mean-square error surface for

the D vs E fit for complex 9-4.


S14





























-*- 1000Hz
-r-- 250Hz
-A- 50Hz


40











2 31


El


O



0 4 6


T(K


Fiur 97 Poto te nphse 'T ndou-f-hae


-*- 1000Hz
-E-- 25GHz
-A- 50Hz


8 10 12


")n' ac susceptibility data for 9-4.


40




20







O 3
E



1.0





0 .5

0.


0 2 4 6 8 10 12

T(K)

Figure 9-8. Plot of the in-phase 'MT) and out-of-phase ")Ml ac susceptibility data for 9-5.











230


-*- 1000Hz
-E-- 250Hz
-A- 50Hz




-*- 1000Hz
-0- 250Hz
-A-- 50Hz




































































8 10 12


[Mnl2z-~ (z = 0, 1, 2, 3) at the indicated


-* 1000Hlz
4 ~- -0H-- 250Hz t

3 -





3- j













E
















4 4











3-3


Z =0


Z=1


Z= 2


Z= 3





O





0 0 1
T(K
Fiue91.Cmprsno h g"v posfr aumdidcopee M l z z=0 ,2
3) t 100Hz tp n 0H bto)










APPENDIX A
BOND DISTANCES AND ANGLES


Table A-1. Selected interatomic distances (A+) and angles (o) for [Mn402(O2CMe)s(salpro)]
-MeCN (2-1-MeCN)
Mn1-O3 1.8859(19) Mn2-Mn3 3.1763(6)
Mn1-O2 1.8906(18) Mn3-O2 1.8884(19)
Mn1-04 1.946(2) Mn3-O3 1.9062(18)
Mn1-06 1.956(2) Mn3-012 1.9452(19)
Mn1-05 2.0768(19) Mn3-010 1.950(2)
Mn1-Mn3 2.7704(6) Mn3-011 2.085(2)
Mn1-Mn2 3.1713(6) Mn3-Mn4 3.1803(6)
Mn2-08 1.864(2) Mn4-014 1.8768(19)
Mn2-O2 1.9090(18) Mn4-O3 1.9145(18)
Mn2-N1 1.972(2) Mn4-N2 1.972(2)
Mn2-01 1.9905(18) Mn4-01 1.9776(18)
Mn2-07 2.206(2) Mn4-Ol5 2.222(2)
Mn2-09 2.238(2) Mn4-013 2.271(2)
03-Mn1-O2 81.26(8) 03-Mn4-01 94.64(8)
03-Mn1-04 168.79(9) N2-Mn4-01 82.55(9)
02-Mn1-04 93.60(8) 03-Mn4-Ol5 86.21(8)
02-Mn3-O3 80.79(8) N2-Mn4-Ol5 98.94(9)
02-Mn3-012 170.65(8) 01-Mn4-Ol5 88.57(8)
03-Mn3-012 91.79(8) Mn4-01-Mn2 133.88(9)
02-Mn3-010 94.91(8) Mn3-O2-Mn1 94.30(8)
03-Mn3-010 165.26(8) Mn3-O2-Mn2 113.53(9)
012-Mn3-010 90.87(8) Mn1-O2-Mn2 113.16(9)
014-Mn4-O3 92.81(8) Mn1-O3-Mn3 93.87(8)
014-Mn4-N2 90.08(9) Mn1-O3-Mn4 114.47(9)
03-Mn4-N2 174.03 (9) Mn3-O3-Mn4 112.69(9)










Table A-2. Selected interatomic distances (A+) and angles (o) for [Mn402(O2CBur)s(salpro)]
MeOH-2CHzC 2 27H16 (2-3-MeOH-2CH 2 2 7H16)
Mn1-O2 1.887(2) Mn2-Mn4 3.1932(7)
Mn1-05 1.910(2) Mn3-011 1.877(2)
Mn1-01 1.955(2) Mn3 -O2 1.9096(19)
Mn1-O3 1.956(2) Mn3-N1 1.966(3)
Mn1-04 2.110(2) Mn3-09 1.970(2)
Mn1-Mn4 2.7942(6) Mn3-012 2.181(2)
Mn1-Mn3 3.1688(6) Mn3-010 2.224(2)
Mn2-08 1.884(2) Mn3-Mn4 3.1361(7)
Mn2-05 1.907(2) Mn4-05 1.888(2)
Mn2-N2 1.978(3) Mn4-O2 1.904(2)
Mn2-09 1.984(2) Mn4-013 1.946(2)
Mn2-07 2.183(2) Mn4-014 1.949(2)
Mn2-06 2.196(2) Mn4-Ol5 2.081(2)
Mn1-O2-Mn4 94.97(9) Mn4-05-Mn1 94.74(9)
Mn1-O2-Mn3 113.17(10) Mn2-05-Mn1 111.16(10)
Mn4-O2-Mn3 110.64(9) Mn3 -09-Mn2 134.10(10)
Mn4-05-Mn2 114.58(10)













1.8955(19)
1.9062(19)
2.025(2)
2.040(2)
2.118(2)
2.1862(19)
91.10(8)
88.66(9)
177.79(9)
175.42(9)
88.51(9)
91.56(9)
98.21(8)
90.70(8)
91.51(9)
86.36(9)
88.57(8)
90.04(8)
87.76(8)
86.87(8)
173.16(8)


Table A-3. Selected interatomic distances (A+) and angles (o) for
NDu4[Mn(O2CPh)2(SalproH)]-CHCH22 (2-4-CH2 12)
Mn1-05 1.8927(18) Mn2-013
Mn1-06 1.9031(19) Mn2-012
Mn1-N1 2.019(2) Mn2-N3
Mn1-N2 2.043(2) Mn2-N4
Mn1-O2 2.1392(19) Mn2-09
Mn1-04 2.2005(19) Mn2-011
05-Mn1-06 91.53(8) 013-Mn2-012
05-Mn1-N1 88.77(9) 013-Mn2-N3
06-Mn1-N1 176.05(9) 012-Mn2-N3
05-Mn1-N2 174.10(9) 013-Mn2-N4
06-Mn1-N2 88.20(9) 012-Mn2-N4
N1-Mn1-N2 91.10(9) N3-Mn2-N4
05-Mn1-O2 98.88(8) 013-Mn2-09
06-Mn1-O2 92.15(8) 012-Mn2-09
N1-Mn1-O2 91.69(9) N3-Mn2-09
N2-Mn1-O2 87.02(8) N4-Mn2-09
05-Mn1-04 87.08(7) 013-Mn2-011
06-Mn1-04 88.89(8) 012-Mn2-011
N1-Mn1-04 87.19(8) N3-Mn2-011
N2-Mn1-04 87.02(8) N4-Mn2-011
02-Mn1-04 173.91(7) 09-Mn2-011










Table A-4. Selected interatomic distances (A+) and angles (o) for
[Mn4(hmp)2(pdmH)2(MeCN)4](CIO4)4 (3-1)
Mn1-O2 1.876(3) Mn2-01
Mn1-01 1.898(3) Mn2-O2'
Mn1-04' 1.968(2) Mn2-N3


Mn1-N2
Mn1-N1
Mn1-04
Mn1---Mnl'
Mn1-01-Mn2
Mn1-O2-Mn2'
Mn1'-04-Mn1


2.157(2)
2.215(2)
2.267(3)
2.272(3)
2.303(3)
2.310(4)
2.363(2)
100.45(10)
93.85(9)


2.064(3)
2.205(3)
2.254(2)
3.2092(10)
112.43(11)
109.12(11)
98.73(10)


Mn2-N4
Mn2-O3
Mn2-N5
Mn2-04
Mn1'-04-Mn2
Mn 1-04-Mn2










Table A-5. Selected interatomic distances (A+) and angles (o) for
[Mn25 01 s(OH)2(hmp)6(pdm)s(pdmH)2(L)2] (CIO4)6- 8MeCN- 4MeOH (3-2 -8MeCN- 4MeOH)
Mn(1)-O(26)' 2.130(5) Mn(6)-O(22) 2.133(4)
Mn(1)-O(18) 2.180(5) Mn(6)-O(7) 2.259(5)
Mn(1)-O(1) 2.215(7) Mn(6)-O(8) 2.297(5)
Mn(2)-O(4) 1.874(5) Mn(7)-O(25) 1.884(4)
Mn(2)-O(19) 1.904(5) Mn(7)-O(22) 1.889(4)
Mn(2)-O(18) 1.918(4) Mn(7)-O(14) 1.939(5)
Mn(3)-O(18) 1.887(5) Mn(8)-O(20) 1.874(4)
Mn(3)-O(21) 1.901(5) Mn(8)-O(19) 1.885(4)
Mn(3)-O(3) 1.928(5) Mn(8)-O(22) 2.105(5)
Mn(4)-O(20)' 1.887(4) Mn(9)-O(27) 1.879(5)
Mn(4)-O(23) 1.889(4) Mn(9)-O(27)' 1.879(5)
Mn(4)-O(27) 2.109(4) Mn(9)-O(22)' 1.892(4)
Mn(4)-O(21) 2.112(5) Mn(10)-O(11) 1.883(5)
Mn(4)-N(3) 2.140(6) Mn(10)-O(20) 1.908(5)
Mn(4)-O(6) 2.241(5) Mn(10)-O(25) 1.909(5)
Mn(4)-O(5) 2.331(5) Mn(11)-O(26) 2.116(5)
Mn(5)-O(24) 1.866(5) Mn(11)-O(25) 2.205(4)
Mn(5)-O(27) 1.890(5) Mn(11)-O(11) 2.223(6)
Mn(5)-O(16) 1.936(5) Mn(11)-O(14) 2.493(5)
Mn(6)-O(19) 1.875(5) Mn(12)-O(26) 2.112(5)
Mn(6)-O(23) 1.877(5) Mn(13)-O(17) 1.887(5)
Mn(6)-O(21) 2.109(4) Mn(13)-O(23) 1.903(5)
Mn(12)-O(17) 2.182(5) Mn(13)-O(24) 1.919(4)
Mn(12)-O(24) 2.207(5) Mn(13)-N(9) 2.027(6)
Mn(12)-N(8) 2.258(8) Mn(13)-O(6) 2.258(5)
Mn(3)-O(3)-Mn(1 1)' 99.8(2) Mn(13)-O(17)-Mn(12) 103.7(2)
Mn(2)-O(4)-Mn(1) 102.2(2) Mn(3)-O(18)-Mn(2) 108.5(2)
Mn(3)-O(5)-Mn(10)' 96.68(17) Mn(6)-O(19)-Mn(8) 110.7(2)
Mn(4)-O(6)-Mn(13) 88.63(19) Mn(4)'-O(20)-Mn(10) 115.8(2)
Mn(2)-O(7)-Mn(3) 86.92(16) Mn(9)-O(2 1)-Mn(3) 136.2(2)
Mn(7)-O(8)-Mn(13) 95.5(2) Mn(8)-O(22)-Mn(6) 93.69(19)
Mn(5)'-O(9)-Mn(8) 97.1(2) Mn(6)-O(23)-Mn(4) 109.6(2)
Mn(8)-O(10)-Mn(10) 88.12(16) Mn(7)-O(24)-Mn(12) 100.21(18)










Table A-6. Selected interatomic distances (A+) and angles (o) for
[Fe-rO4(O2CPh)ll1(dmem)2]-4MeC (4-1-4MeCN)
Fel-Ol0' 1.8276(18) Fe2-Fe4
Fel-O2 1.9966(18) Fe3 -09
Fel-04' 2.0424(18) Fe3-08
Fel-01 2.0519(19) Fe3 -06
Fel-N2 2.248(2) Fe3 -013
Fel-N1 2.269(3) Fe3 -011
Fe2-09 1.9234(17) Fe3 -012
Fe2-010 1.941(2) Fe3-C34
Fe2-05 2.051(2) Fe4-09
Fe2-O2 2.0534(17) Fe4-010
Fe2-O3 2.0537(18) Fe4-014'
Fe2-07 2.1053(18) Fe4-014
010'-Fel1-O2 98.25(8) 09-Fe3 -06
010'-Fel-04' 95.05(8) 09-Fe4-09'
02-Fel-04' 166.09(8) 09-Fe4-010
010'-Fel-01 103.89(8) 09'-Fe4-010
02-Fel-01 90.28(8) 09-Fe4-Ol0'
09-Fe2-010 83.84(8) 09'-Fe4-Ol0'
09-Fe2-05 94.27(8) 010-Fe4-Ol0'
010-Fe2-05 176.11(8) 09-Fe4-014'
09-Fe2-O2 94.30(7) 09-Fe4-014
010-Fe2-O2 97.13(7) 010-Fe4-014
09-Fe2-O3 174.58(8) 010'-Fe4-014
010-Fe2-O3 94.32(8) Fel-O2-Fe2
02-Fe2-O3 90.99(7) Fe3-09-Fe2
09-Fe2-07 94.97(7) Fe3-09-Fe4
010-Fe2-07 91.06(8) Fe2-09-Fe4
02-Fe2-07 168.24(8) Fel'-010-Fe2
03-Fe2-07 79.95(7) Fel1'-010-Fe4
09-Fe3-08 102.13(8) Fe2-010-Fe4


2.9287(5)
1.8436(18)
2.0092(19)
2.027(2)
2.038(2)
2.0547(19)
2.200(2)
2.470(3)
1.989(2)
1.9915(17)
2.0680(17)
2.0681(18)
95.81(9)
176.62(10)
80.88(7)
101.66(7)
101.66(7)
80.87(7)
84.82(10)
87.48(7)
90.19(8)
92.39(7)
167.19(8)
125.76(10)
120.44(9)
125.57(10)
96.92(8)
124.59(9)
134.38(10)
96.25(8)











Table A-7.


Fel-01
Fel-06

Fel-08
Fel-05

Fel-N2
Fel-N1

Fel-Fe2
Fe2-O3

Fe2-01
Fe2-012

Fe2-011
Fe2-05

Fe2-07
Fe3-01

Fe3-O2
Fe3-Ol6

Fe3-010
Fe3-014

Fe3-09
Fe3-Fe4

Fe4-O2
Fe3-01-Fel

Fe3-01-Fe2
Fel1-01-Fe2

Fe4-O2-Fe5
Fe4-O2-Fe3

Fe5-O2-Fe3
Fe2-O3 -Fe5


Selected interatomic distances (A+) and angles (o) for
[Fe7O4(O2CMe)11 (dmem)2]-MeCN (4-2-MeCN)
1.8783(17) Fe4-Ol6

1.9992(19) Fe4-Ol7
2.0106(18) Fe4-Ol5

2.0147(17) Fe4-N3
2.191(2) Fe5-O2

2.282(2) Fe5-O3
2.9585(5) Fe5-04

1.8672(17) Fe5-Ol9
1.9833(18) Fe5-O21

2.0291(18) Fe5-Ol8
2.0413(19) Fe5-Fe6

2.0541(18) Fe6-04
2.1647(18) Fe6-O3

1.8621(17) Fe6-013
1.9900(17) Fe6-O23

2.0380(18) Fe6-O22
2.0658(19) Fe6-O25

2.094(2) Fe7-04
2.0976(18) Fe7-O24

2.9540(5) Fe7-O20
1.8787(17) Fe7-O27

133.53(10) Fe2-O3-Fe6
123.22(9) Fe5-O3-Fe6

99.98(8) Fe7-04-Fe6
127.18(9) Fe7-04-Fe5

99.52(8) Fe6-04-Fe5
130.01(9) Fel-05-Fe2

128.04(9) Fe4-Ol6-Fe3


1.9894(18)

2.0107(18)
2.057(2)

2.246(2)
1.9402(17)

1.9568(17)
2.0141(17)

2.0554(19)
2.0590(19)

2.0922(18)
2.9066(5)

1.9053(18)
1.9579(17)

2.0416(18)
2.0525(19)

2.055(2)
2.0619(19)

1.8175(18)
2.011(2)

2.057(2)
2.0623(19)

129.56(9)
95.88(7)

121.04(9)
132.30(10)
95.69(7)

93.29(7)
94.35(7)










Table A-8. Selected interatomic distances (A+) and angles (o) for
[Fe602(OH)4(O2CBur~s~ddmem)2]-2MeCN (4-3-2MeCN)
Fel-05' 1.9366(16) Fe2-08'
Fel-05 1.9382(17) Fe2-06
Fel-01 2.0251(18) Fe2-07
Fel-O2 2.0457(17) Fe2-C 11
Fel-Ol0' 2.0471(17) Fe3-010
Fel-O3 2.0504(18) Fe3-09
Fel-Fel' 2.8651(7) Fe3-O2
Fe2-05 1.8441(17) Fe3-011
Fe2-09 1.9580(18) Fe3-N1
Fe2-04 2.0412(18) Fe3-N2
05'-Fel-05 84.64(7) 09-Fe2-08'
05'-Fel-01 94.52(7) 05-Fe2-06
05-Fel-01 178.33(8) 09-Fe2-06
05'-Fel-O2 95.30(7) 05-Fe2-07
05-Fel-O2 88.98(7) 09-Fe2-07
01-Fel1-O2 89.67(7) 010-Fe3-09
05'-Fel1-Ol0' 87.64(7) 010-Fe3-O2
05-Fel-Ol0' 91.72(7) 09-Fe3-O2
01-Fel-Ol0' 89.68(7) 010-Fe3-011
02-Fel-Ol0' 177.02(7) 09-Fe3-011
05'-Fel-O3 171.93(7) 02-Fe3-011
05-Fel-O3 93.52(7) Fe3-O2-Fel
02-Fel1-O3 92.51(7) Fe2-05-Fe1'
010'-Fel-O3 84.56(7) Fe2-05-Fel
05-Fe2-09 103.36(8) Fel'-05-Fel
05-Fe2-04 96.43(7) Fe2-09-Fe3
09-Fe2-04 91.06(8) Fe3-010-Fel'

05-Fe2-08' 96.06(7)


2.0448(18)
2.0662(19)
2.2189(19)
2.481(3)
1.9420(17)
1.9651(18)
2.0181(18)
2.0331(19)
2.231(2)
2.272(2)
89.35(8)
99.44(8)
157.16(8)
160.49(7)
96.06(7)
99.75(8)
96.45(7)
90.25(8)
93.63(8)
89.97(8)
169.73(7)
125.40(8)
128.19(9)
126.51(9)
95.36(7)
117.76(9)
129.75(9)











Table A-9.


Fel-01
Fel-N8

Fel-O3
Fel-07

Fel-N4
Fel-N3

Fe2-O3
Fe2-O2

Fe2-N11
01-Fel-O3

01-Fel-07
03-Fel-07

01-Fel-N4
03-Fel-N4

07-Fel-N4
01-Fel-N3

N8-Fel-N3
03-Fel-N3

07-Fel-N3
N4-Fel-N3

03-Fe2-O2
03-Fe2-04

02-Fe2-04
03-Fe2-06

02-Fe2-06
04-Fe2-06

03-Fe2-01


Selected interatomic distances (A+) and angles (o) for
[Fe30(O2CBut)2(N3)3(dmmem)2]-4CH22 (4-4-2CH2 12)
1.8716(19) Fe2-04

2.007(2) Fe2-06
2.0291(18) Fe2-01

2.0608(19) Fe3-01
2.220(2) Fe3-N5

2.243(2) Fe3-O2
1.9787(19) Fe3-05

1.9834(19) Fe3-N2
2.007(2) Fe3-N1

80.87(8) 02-Fe2-01
96.21(8) 04-Fe2-01

88.65(8) 06-Fe2-01
97.88(9) 01-Fe3-N5

95.48(8) 01-Fe3-O2
165.78(9) 01-Fe3-05

157.25(8) 02-Fe3-05
95.53(10) 01-Fe3-N2

76.70(8) 02-Fe3-N2
87.04(9) 05-Fe3-N2

80.71(9) 01-Fe3-N1
155.14(8) 02-Fe3-N1

92.88(8) 05-Fe3-N1
88.52(8) N2-Fe3-N1

89.19(8) Fe3-01-Fel
91.02(8) Fe3-01-Fe2

176.04(8) Fel-01-Fe2
77.41(7) Fe2-O2-Fe3
Fe2-O2-Fel


2.0469(19)

2.0605(19)
2.0700(19)

1.8647(19)
2.020(2)

2.0245(19)
2.066(2)

2.211(2)
2.241(2)
77.73(7)

93.04(8)
90.70(7)

106.84(10)

81.60(8)
95.92(8)

88.49(8)
98.10(9)

95.09(9)
165.89(9)

157.93(8)
76.51(8)

86.38(9)
81.21(9)

162.82(11)
98.34(8)

98.85(8)
96.07(8)

96.77(8)










Table A-10. Selected interatomic distances (A+) and angles (o) for
[Fe5O2(OH)(O2CMe)s(hmbp)3](CIO4)2-5MeCN (5-1-5MeCN)
Fel-Ol6 1.825(3) Fe3-09 2.063(3)
Fel-O3 2.021(3) Fe3-N5 2.094(4)
Fel-Ol5 2.058(3) Fe3-N6 2.189(4)
Fel-01 2.075(3) Fe4-Ol6 1.951(3)
Fel-N2 2.104(3) Fe4-011 1.956(3)
Fel-N1 2.190(3) Fe4-014 2.003(3)
Fe2-04 1.933(3) Fe4-010 2.017(3)
Fe2-05 1.980(3) Fe4-012 2.039(3)
Fe2-O3 2.023(3) Fe4-05 2.058(3)
Fe2-06 2.052(3) Fe5-011 1.951(3)
Fe2-N4 2.116(4) Fe5-Ol6 1.951(3)
Fe2-N3 2.155(5) Fe5-O2 2.017(3)
Fe3-011 1.826(3) Fe5-08 2.021(3)
Fe3-06 2.008(3) Fe5-04 2.031(3)
Fe3-07 2.048(3) Fe5-013 2.041(3)
016-Fel-O3 100.08(12) 011-Fe5-Ol6 84.45(11)
016-Fel-Ol5 93.21(12) 011-Fe5-08 91.98(12)
016-Fel-01 94.68(12) 016-Fe5-08 176.20(12)
04-Fe2-05 111.55(12) 011-Fe5-04 88.95(12)
04-Fe2-O3 90.87(13) 016-Fe5-04 87.62(11)
05-Fe2-O3 89.21(13) 02-Fe5-04 91.79(13)
04-Fe2-06 88.51(13) 08-Fe5-04 93.62(12)
05-Fe2-06 88.59(13) 011-Fe5-013 88.85(11)
011-Fe3-06 104.43(12) 016-Fe5-013 89.14(11)
011-Fe3-09 93.41(13) 04-Fe5-013 176.24(12)
016-Fe4-011 84.34(11) Fel-O3-Fe2 114.99(14)
016-Fe4-014 92.60(11) Fe2-04-Fe5 119.95(15)
011-Fe4-014 176.42(12) Fe2-05-Fe4 118.09(14)
016-Fe4-010 176.14(12) Fe3-06-Fe2 112.45(14)
011-Fe4-012 88.83(11) Fe3-011-Fe5 125.89(15)
016-Fe4-05 88.71(12) Fe3-011-Fe4 127.03(14)
011-Fe4-05 87.71(11) Fe5-011-Fe4 94.45
014-Fe4-05 90.36(12) 010-Fe4-05 91.81(13)










Table A-11. Selected interatomic distances (A+) and angles (o) for
[Fe602(OH)2(O2CPh)6(hmbp)4] (NO3)2-3MeCN-H20 (5-2-3MeCN-H20)
Fel-07' 1.912(2) Fe2-04 2.033(2)
Fel-O3 1.974(2) Fe2-O3 2.063(2)
Fel-09' 2.046(2) Fe2-Fe2' 2.9430(8)
Fel-01 2.059(2) Fe3-08 1.832(2)
Fel-N2 2.097(3) Fe3-09 2.015(2)
Fel-N1 2.186(3) Fe3-06 2.031(2)
Fe2-08 1.938(2) Fe3-010 2.049(2)
Fe2-08' 1.978(2) Fe3-N3 2.095(3)
Fe2-07 2.011(2) Fe3-N4 2.183(3)
Fe2-05 2.029(2) 04-Fe2-O3 90.14(9)
07'-Fel-O3 105.96(9) 08-Fe3-09 99.68(9)
07'-Fel-09' 94.49(9) 08-Fe3-06 98.96(9)
03-Fel-09' 91.67(10) 09-Fe3-06 93.29(10)
07'-Fel-01 89.85(10) 08-Fe3-010 93.08(9)
03-Fel-01 96.07(10) 09-Fe3-010 95.63(10)
09'-Fel-01 169.75(10) 06-Fe3-010 163.60(10)
08-Fe2-08' 82.56(9) 08-Fe3-N3 173.82(10)
08-Fe2-07 88.09(9) 09-Fe3-N3 77.43(10)
08'-Fe2-07 94.18(9) 06-Fe3-N3 86.72(10)
08-Fe2-05 97.21(9) 010-Fe3-N3 81.85(10)
08'-Fe2-05 176.80(9) 08-Fe3-N4 109.05(10)
07-Fe2-05 89.00(9) 09-Fe3-N4 151.19(10)
08-Fe2-04 175.77(9) 06-Fe3-N4 80.31(10)
08'-Fe2-04 94.14(9) 010-Fe3-N4 85.27(11)
07-Fe2-04 89.52(9) Fel'-07-Fe2 125.85(11)
05-Fe2-04 86.24(9) Fe3-08-Fe2 124.38(11)
08-Fe2-O3 92.59(9) Fe3-08-Fe2' 129.12(11)
08'-Fe2-O3 91.68(9) Fe2-08-Fe2' 97.44(9)
07-Fe2-O3 174.13(9) Fe3-09-Fel' 116.32(10)
05-Fe2-O3 85.13(9)










Table A-12. Selected interatomic distances (A+) and angles (o) for
[FelsOs(OH)2(O2CBut)28(heen)4] -4CsH12-4CH2 12 (6-1-4CsH12-4CH2 2))
Fel-06 1.848(2) Fe5-N4 2.175(3)
Fel-05' 1.981(3) Fe5-N3 2.202(3)
Fel-05 2.022(3) Fe6-O23 1.943(2)
Fel-01 2.046(3) Fe6-O30 1.978(2)
Fel-O3 2.055(3) Fe6-O31 2.030(3)
Fel-07 2.102(3) Fe6-O21 2.032(3)
Fe2-06 1.923(2) Fe6-Ol8 2.034(2)
Fe2-011 1.993(2) Fe6-O29 2.042(3)
Fe2-012 1.994(3) Fe6-Fe7 2.9008(8)
Fe2-04 2.019(3) Fe7-O23 1.941(2)
Fe2-O2 2.028(3) Fe7-O30 1.949(2)
Fe2-09 2.083(3) Fe7-O24 2.024(3)
Fe3-06 1.894(2) Fe7-O35 2.042(2)
Fe3-011 1.937(3) Fe7-Ol9 2.048(2)
Fe3-08 2.019(3) Fe7-O33 2.073(3)
Fe3-010 2.060(2) Fe8-O23 1.849(2)
Fe3-N2 2.163(3) Fe8-O28 1.994(3)
Fe3-N1 2.220(3) Fe8-O25 2.036(3)
Fe4-011 1.819(3) Fe8-O22 2.037(3)
Fe4-014 1.872(4) Fe8-O27 2.108(3)
Fe4-Ol7 1.974(3) Fe8-O26 2.140(3)
Fe4-013 2.013(3) Fe9-O30 1.843(2)
Fe4-Ol6 2.029(3) Fe9-O34 1.980(3)
Fe5-Ol8 1.954(2) Fe9-O32 2.018(3)
Fe5-Ol6 1.966(3) Fe9-O36 2.084(3)
Fe5-Ol9 2.000(2) Fe9-O37 2.114(3)
Fe5-Ol7 2.011(2) Fe9-O38 2.123(3)
Fel'-05-Fel 104.35(11) Fe5-Ol8-Fe6 127.19(13)
Fel-06-Fe3 132.53(13) Fe5-Ol9-Fe7 126.89(12)
Fel-06-Fe2 122.96(12) Fe8-O23-Fe7 126.05(13)
Fe3-06-Fe2 96.22(11) Fe8-O23-Fe6 120.88(12)
Fe4-011-Fe3 138.87(13) Fe7-O23-Fe6 96.64(11)
Fe4-011-Fe2 125.23(13) Fe9-O30-Fe7 120.39(13)
Fe3-011-Fe2 92.62(11) Fe9-O30-Fe6 126.41(13)










Table A-13. Selected interatomic distances (A+) and angles (o) for
[Fe904(OH)4(O2CPh) 13(heenH)2]-9MeCN (6-2-9MeCN)
Fel-07 1.939(4) Fe5-O23 2.098(6)
Fel-01 1.984(4) Fe5-N3 2.189(6)
Fel-08 1.985(4) Fe5-N4 2.205(6)
Fel-05 2.013(4) Fe6-014 1.905(4)
Fel-O3 2.088(4) Fe6-O27 1.961(4)
Fel-06 2.145(4) Fe6-O24 2.027(4)
Fe2-07 1.912(4) Fe6-Ol6 2.035(4)
Fe2-012 1.990(4) Fe6-05 2.058(4)
Fe2-04 2.017(4) Fe6-O28 2.093(4)
Fe2-09 2.045(4) Fe7-O27 1.960(4)
Fe2-011 2.057(4) Fe7-O31 1.963(4)
Fe2-06 2.076(4) Fe7-O36 2.010(4)
Fe3-014 1.887(4) Fe7-O26 2.012(4)
Fe3-Ol5 1.989(4) Fe7-05 2.086(4)
Fe3-Ol7 2.010(4) Fe7-O21 2.141(4)
Fe3-010 2.049(4) Fe8-07 1.889(4)
Fe3-06 2.069(4) Fe8-O31 1.968(4)
Fe3-011 2.093(4) Fe8-O34 2.008(4)
Fe4-014 1.964(4) Fe8-O32 2.044(4)
Fe4-O22 1.979(4) Fe8-O21 2.048(4)
Fe4-Ol9 2.000(4) Fe8-013 2.071(4)
Fe4-O21 2.048(4) Fe9-O31 1.850(4)
Fe4-Ol8 2.067(4) Fe9-08 2.016(4)
Fe4-011 2.115(4) Fe9-O37 2.032(4)
Fe5-O27 1.853(4) Fe9-O35 2.071(4)
Fe5-O22 1.986(4) Fe9-N1 2.195(5)
Fe5-O25 2.038(5) Fe9-N2 2.223(5)
Fel-05-Fe6 130.80(18) Fe3-014-Fe6 125.3(2)
Fel-05-Fe7 121.71(18) Fe3-014-Fe4 101.98(17)
Fe6-05-Fe7 92.87(14) Fe6-014-Fe4 129.7(2)
Fe3-06-Fe2 97.96(15) Fe8-O21-Fe4 127.88(19)
Fe3-06-Fel 133.2(2) Fe8-O21-Fe7 93.22(15)
Fe2-06-Fel 90.39(15) Fe4-O21-Fe7 122.76(19)
Fe8-07-Fe2 125.50(19) Fe4-O22-Fe5 120.5(2)










Table A-14. Selected interatomic distances (A+) and angles (o) for
4[Fe7O3(OMe)3(MeOH)1.S(heen)3 14.5(H20)]Cl[FeCl4] -6MeOH-H20
(6-3-6MeOH-H20)
Fel-O2 1.844(4) Fe4-Cl3 2.3995(15)
Fel-O3 1.848(3) Fe4-C IA 2.458(4)
Fel-01 1.856(4) Fe5-O3 1.944(3)
Fel-Cll 2.2564(15) Fe5-08 1.960(4)
Fe2-01 1.955(4) Fe5-010 2.000(4)
Fe2-06 1.973(4) Fe5-09 2.022(4)
Fe2-014 1.993(4) Fe5-N3 2.157(5)
Fe2-05 2.028(4) Fe5-N4 2.195(4)
Fe2-04 2.128(4) Fe6-O3 1.949(4)
Fe2-Cl2 2.3670(15) Fe6-013 1.972(4)
Fe3-O2 1.947(4) Fe6-010 1.981(4)
Fe3-05 1.970(4) Fe6-012 2.025(4)
Fe3-07 1.996(4) Fe6-011 2.131(4)
Fe3-06 2.015(4) Fe6-Cl5 2.3727(15)
Fe3-N1 2.175(6) Fe7-01 1.941(4)
Fe3-N2 2.180(6) Fe7-012 1.960(4)
Fe4-09 1.964(4) Fe7-014 1.998(4)
Fe4-07 1.987(4) Fe7-013 2.018(4)
Fe4-O2 1.987(4) Fe7-N6 2.169(5)
Fe4-08 2.037(4) Fe7-N5 2.183(5)
Fe4-04" 2.043(13) Fe8-Cl8 2.2078(15)
Cl8'-Fe8-Cl8 109.79(8) Fe5-O3-Fe6 102.64(15)
Cl8"-Fe8-Cl8 110.36(8) Fe3-05-Fe2 102.14(18)
Cl8-Fe8-Cl8'" 108.27(9) Fe2-06-Fe3 102.49(17)
Fel-01-Fe7 123.50(18) Fe4-07-Fe3 100.48(17)
Fel-01-Fe2 128.89(19) Fe5-08-Fe4 102.71(17)
Fe7-01-Fe2 102.33(17) Fe4-09-Fe5 103.09(17)
Fel-O2-Fe4 128.2(2) Fe6-010-Fe5 99.53(15)
Fe3-O2-Fe4 102.21(17) Fe7-012-Fe6 102.13(16)
Fel-O3-Fe5 124.31(19) Fe6-013-Fe7 101.98(16)
Fel-O3-Fe6 128.29(19) Fe2-Ol4-Fe7 99.01(16)










Table A-15. Selected interatomic distances (A+) and angles (o) for
[Fe602(O2CPh)s (heen)3(heenH)](CIO04)2-2EtOH-1.5H20 (6-4-2EtOH-1.5H20)
Fel-07' 1.918(3) Fe2-O3 2.053(3)
Fel-010 1.957(3) Fe2-N1 2.171(4)
Fel-07 1.980(3) Fe2-N2 2.200(4)
Fel-01 2.027(3) Fe3-011 1.853(10)
Fel-08 2.044(3) Fe3-08 1.984(4)
Fel-04' 2.116(3) Fe3-09 1.995(3)
Fel-Fel' 2.9113(11) Fe3-010 1.997(3)
Fe2-07 1.846(3) Fe3-05 2.025(6)
Fe2-09 2.022(3) Fe3-N3 2.187(6)
Fe2-O2' 2.048(3) Fe3-N4 2.212(5)
Fe2-07-Fel' 120.52(15) Fe3-08-Fel 101.62(16)
Fe2-07-Fel 130.32(16) Fe3-09-Fe2 121.73(16)
Fel'-07-Fel 96.63(12) Fel-010-Fe3 104.30(16)










Table A-16. Selected interatomic distances (A+) and angles (o) for
[Mns O3(OH)(OMe)(O2 CPh)i(edte) (edteH2)] (O2CPh) 2 CH2 2-z Me OH
(7-1-2CH2 2-MeOH)
Mn1-O3 1.8836(19) Mn5-014 1.8923(18)
Mn1-O25 1.8844(17) Mn5-O22 1.8941(17)
Mn1-04 1.9492(18) Mn5-Ol6 1.9467(18)
Mn1-Ol7 1.9591(17) Mn5-012 1.9517(19)
Mn1-O27 2.176(2) Mn5-Ol9 2.195(2)
Mn1-Ol8 2.3097(18) Mn5-Ol5 2.3662(18)
Mn2-07 1.8554(18) Mn6-01 2.2242(17)
Mn2-O25 1.9105(18) Mn6-Ol6 2.2495(18)
Mn2-Ol8 1.9424(17) Mn6-Ol8 2.2606(17)
Mn2-06 1.9566(17) Mn6-Ol5 2.2617(18)
Mn2-05 2.1596(18) Mn6-Ol7 2.2647(18)
Mn2-01 2.4368(17) Mn6-N4 2.279(2)
Mn3 -010 2.1431(17) Mn6-N3 2.285(2)
Mn3 -07 2.1543(18) Mn7-O22 1.8957(18)
Mn3 -08 2.180(2) Mn7-01 1.8997(17)
Mn3 -09 2.239(2) Mn7-O20 1.9505(19)
Mn3 -N2 2.330(2) Mn7-O21 1.9565(18)
Mn3-N1 2.359(2) Mn7-O23 2.1235(19)
Mn3 -06 2.4054(17) Mn7-Ol6 2.3692(18)
Mn4-010 1.8661(17) Mn8-O25 1.8937(17)
Mn4-O22 1.9136(18) Mn8-01 1.9079(17)
Mn4-Ol5 1.9297(17) Mn8-O21 1.9517(18)
Mn4-06 1.9652(17) Mn8-O26 1.9571(19)
Mn4-011 2.1725(19) Mn8-O24 2.1360(19)
Mn4-01 2.4695(18) Mn8-Ol7 2.3122(18)
Mn7-01-Mn8 97.12(8) Mn2-07-Mn3 109.95(9)
Mn8-01-Mn2 88.95(6) Mn4-010-Mn3 110o.16(8)
Mn6-01-Mn2 92.52(6) Mn6-Ol5-Mn5 97.48(7)
Mn7-01-Mn4 89.31(6) Mn5-Ol6-Mn6 111.83(8)
Mn6-01-Mn4 92.09(6) Mn8-O21-Mn7 93.83(8)
Mn2-01-Mn4 80.05(5) Mn5-O22-Mn7 105.86(8)
Mn2-06-Mn4 107.14(8) Mn1-O25-Mn8 104.91(8)
Mn2-06-Mn3 97.31(7) Mn4-06-Mn3 97.19(7)










Table A-17. Selected interatomic distances (A+) and angles (o) for
[Mnl1204(OH)2(edte)416(H20)2] -6MeCN-%2H20 (7-2-6MeCN-%2H20)
Mn1-06' 2.065(3) Mn2-01' 2.033(3)
Mn1-O3 2.088(3) Mn2-Cl2 2.5255(14)
Mn1-01 2.138(3) Mn3-06 1.882(3)
Mn1-O2' 2.306(3) Mn3-O3 1.905(3)
Mn1-Cll 2.4050(13) Mn3-05 2.107(2)
Mn1-Cl2 2.5813(14) Mn3-01 2.163(3)
Mn2-01 1.841(2) Mn3-O2 2.209(3)
Mn2-05' 1.910(2) Mn3-N2 2.275(3)
Mn2-04 2.0098(15) Mn3-N1 2.320(3)
Mn2-O2 2.015(3) 02-Mn2-Cl2 92.62(8)
Mn2-04-Mn2' 133.9(2) 02-Mn2-Cl2 92.62(8)
06'-Mn1-O3 168.16(11) 01'-Mn2-Cl2 171.93(8)
02'-Mn1-Cll 108.95(7) 06-Mn3-05 87.60(11)
01-Mn1-Cl2 82.17(7) 05-Mn3-01 72.73(9)
02'-Mn1-Cl2 147.38(7) 03-Mn3-O2 105.80(11)
01-Mn2-05' 172.08(12) 05-Mn3-N2 75.17(10)
01-Mn2-04 92.17(10) 06-Mn3-N1 100.89(12)










Table A-18. Selected interatomic distances (A+) and angles (o) for
[Mn200s(OH)4(O2Cee)6(ee))6](CIO4)2- 10MeOH (7-3-10MeOH)
Mn1-O2 2.111(7) Mn6-O22 1.890(6)
Mn1-07 2.116(7) Mn6-O20 1.900(5)
Mn1-014 2.148(11) Mn6-Ol0' 1.927(6)
Mn1-013 2.258(9) Mn6-012' 1.944(5)
Mn1-06 2.293(7) Mn6-Ol7 2.132(6)
Mn1-04 2.333(6) Mn6-O24 2.523(5)
Mn2-O2 1.874(6) Mn7-Ol9 1.856(6)
Mn2-O3 1.903(5) Mn7-Ol6 1.919(6)
Mn2-01 1.926(7) Mn7-01 1.979(7)
Mn2-N2 2.165(7) Mn7-08' 2.002(8)
Mn2-04 2.175(6) Mn7-011 2.142(8)
Mn2-N1 2.308(8) Mn7-09 2.177(7)
Mn3 -O20 1.882(6) Mn8-O21 1.990(6)
Mn3 -O22 1.895(6) Mn8-O20 1.999(6)
Mn3 -06 1.973(6) Mn8-011 2.039(7)
Mn3 -04 1.988(6) Mn8-05 2.121(6)
Mn3 -Ol9 2.227(6) Mn8-O24 2.498(5)
Mn4-05 1.892(7) Mn9-O23 1.916(6)
Mn4-07 1.910(7) Mn9-O21' 1.929(6)
Mn4-08' 1.934(8) Mn9-O24' 1.939(5)
Mn4-N3 2.159(9) Mn9-O24 1.946(5)
Mn4-06 2.228(6) Mn9-010 2.270(5)
Mn4-N4' 2.35(2) Mn9-012' 2.270(5)
Mn4-N4 2.356(16) Mn10-09 2.229(6)
Mn5-O22 1.986(6) Mn10-012 2.246(6)
Mn5-O23 1.994(5) Mn10-011 2.260(6)
Mn5-09 2.040(6) Mn10-010 2.264(6)
Mn5-Ol9 2.085(7) Mn10-N6 2.303(8)
Mn5-O3 2.145(6) Mn10-N5 2.304(8)
Mn2-01-Mn7 128.2(4) Mn10-010-Mn9 93.0(2)
Mn2-O2-Mn1 110.8(3) Mn8-011-Mn7 95.2(3)
Mn2-O3-Mn5 110.8(3) Mn6'-012-Mn9' 99.1(2)










Table A-19. Selected interatomic distances (A+) and angles (o) for
[Fe602(O2CBur~s~eedte2H)2-CH13 (7-5-2CHCl3)
Fel-012 1.931(2) Fe2-O2 2.050(3)
Fel-06 1.957(3) Fe2-N2 2.251(3)
Fel-010 2.025(3) Fe2-N1 2.289(3)
Fel-08 2.026(3) Fe3-012 1.858(2)
Fel-01 2.037(3) Fe3-013' 2.019(3)
Fel-O3 2.051(2) Fe3-013 2.033(2)
Fe2-012 1.907(2) Fe3-04' 2.034(2)
Fe2-04 1.965(3) Fe3-011 2.039(3)
Fe2-O3 2.030(3) Fe3-09 2.040(3)
Fe3-012-Fe2 123.25(13) Fe3'-013-Fe3 102.33(11)
Fe3-012-Fel 123.83(13) Fe2-O3-Fel 94.06(10)
Fe2-012-Fel 102.16(11) Fe2-04-Fe3' 118.77(12)










Table A-20. Selected interatomic distances (A+) and angles (o) for
[Fel204,(OH)2(O2CMe)6(edte)4(H20)2] (CIO4)4-4MeCN (7-6-4MeCN)
Fel-O2 1.977(6) Fe6-04 2.074(6)
Fel-O3 1.985(6) Fe7-Ol7 1.926(6)
Fel-01 2.040(6) Fe7-O33 1.969(6)
Fel-07 2.187(6) Fe7-013 1.982(6)
Fel-04 2.223(6) Fe7-O29 2.037(6)
Fe2-O2 1.964(6) Fe7-012 2.092(6)
Fe2-011 1.984(7) Fe8-Ol9 1.972(7)
Fe2-05 2.010(7) Fe8-010 1.976(7)
Fe2-013 2.053(6) Fe8-O31 2.001(7)
Fe2-04 2.104(6) Fe8-O29 2.068(6)
Fe3-010 1.972(7) Fe8-012 2.098(7)
Fe3-011 1.988(7) Fe9-O21 1.959(6)
Fe3-09 2.038(6) Fe9-O24 1.979(6)
Fe3-013 2.160(6) Fe9-O22 2.031(7)
Fe3-012 2.239(6) Fe9-O37 2.186(6)
Fe4-O3 1.975(6) Fe9-O23 2.237(7)
Fe4-O21 1.981(7) Fe10-O32 1.954(6)
Fe4-07 2.004(6) Fe10-O24 1.974(6)
Fe4-Ol6 2.080(6) Fe10-O26 1.995(8)
Fe4-O23 2.098(6) Fe10-O37 2.031(6)
Fe5-Ol7 1.944(6) Fe10-O27 2.032(7)
Fe5-01 1.968(7) Fe10-O30 2.103(7)
Fe5-O37 1.971(6) Fell-Ol8 1.952(6)
Fe5-07 2.032(6) Fell-O29 1.963(6)
Fe5-O23 2.104(6) Fell-O22 1.980(6)
Fe6-Ol8 1.939(6) Fell-O37 2.026(6)
Fe6-09 1.967(6) Fell-O30 2.093(6)
Fe6-07 1.980(6) Fel2-O32 1.962(6)
Fe6-013 2.013(6) Fel2-O31 1.976(7)
Fel2-O29 2.156(6) Fel2-O30 2.236(7)
Fe7-Ol7-Fe5 135.1(3) Fe6-Ol8-Fell 134.1(3)










Table A-21. Selected interatomic distances (A+) and angles (o) for
[MnloGd20s(O2C~~lohm)6(0(hm)6(eNO34-MeNMO (8-4-3MeCN-MeOH)
Gdl-013 2.312(5) Mn2-N5 2.030(6)
Gdl-011 2.325(5) Mn2-O22 2.190(5)
Gdl-012 2.360(5) Mn2-014 2.223(5)
Gdl-08 2.469(7) Mn2-Mn3 3.1327(13)
Gdl-N3 2.504(7) Mn3-O2 1.882(4)
Gdl-01 2.509(4) Mn3-012 1.905(4)
Gdl-06 2.512(9) Mn3-01 1.915(4)
Gdl-05 2.535(8) Mn3-N4 2.026(5)
Gdl-09 2.567(7) Mn3-Ol6 2.086(5)
Gdl-N2 2.949(8) Mn3-014 2.380(6)
Gdl-N1 2.951(10) Mn4-O2 1.912(4)
Gdl-Mn2 3.4185(10) Mn4-O3' 1.947(4)
Mn1-011 1.916(5) Mn4-Ol9' 1.954(5)
Mn1-O3 1.931(4) Mn4-Ol5 1.994(6)
Mn1-01 1.942(4) Mn4-04 2.125(4)
Mn1-O20 1.973(5) Mn4-O21' 2.151(6)
Mn1-Ol8 2.137(5) Mn5-04' 1.885(4)
Mn1-O22 2.308(5) Mn5-O2 1.907(4)
Mn1-Mn2 3.1104(14) Mn5-O3' 1.938(4)
Mn2-04 1.841(4) Mn5-Ol7 1.962(4)
Mn2-013 1.908(5) Mn5-O23' 2.211(5)
Mn3 -01-Mn1 133.2(2) Mn5-O3 2.476(4)
Mn3-01-Mn2 108.5(2) Mn1-O3-Mn5 124.90(19)
Mn1-01-Mn2 106.31(19) Mn5'-O3-Mn5 93.99(17)
Mn3-01-Gd1 101.54(16) Mn4'-O3-Mn5 90.99(16)
Mn1-01-Gd1 102.50(18) Mn2-04-Mn5' 124.6(2)
Mn2-01-Gd1 99.50(17) Mn2-04-Mn4 118.7(2)
Mn3-O2-Mn5 125.8(2) Mn5'-04-Mn4 104.62(19)
Mn3-O2-Mn4 125.3(2) Mn1-011-Gd1 110.4(2)
Mn5-O2-Mn4 98.54(18) Mn3-012-Gd1 107.37(19)
Mn1-O3-Mn5' 124.1(2) Mn2-013-Gd1 107.8(2)
Mn1-O3-Mn4' 118.5(2) Mn2-014-Mn3 85.71(18)
Mn5'-O3-Mn4' 96.31(18) Mn2-O22-Mn1 87.45(16)










Table A-22. Selected interatomic distances (A+) and angles (o) for
[MnloDy20s(O2C~~lohm)6(0(hm)6(eNO34-MeNMO (8-6-3MeCN-MeOH)
Dyl-013 2.301(4) Mn2-N5 2.039(5)
Dyl-011 2.315(4) Mn2-O22 2.186(4)
Dy l-012 2.341(4) Mn2-014 2.217(5)
Dy l-06 2.454(6) Mn3-O2 1.878(3)
Dy l-08 2.477(4) Mn3-012 1.904(3)
Dy l-01 2.478(3) Mn3-01 1.911(3)
Dy l-N3 2.495(5) Mn3-N4 2.027(5)
Dyl-05 2.500(5) Mn3-Ol6 2.092(4)
Dyl-09 2.515(5) Mn3-014 2.384(5)
Dyl-N1 2.901(6) Mn4-O2 1.919(3)
Dyl-N2 2.921(5) Mn4-Ol9' 1.949(4)
Mn1-011 1.915(4) Mn4-O3' 1.949(3)
Mn1-O3 1.940(3) Mn4-Ol5 1.995(5)
Mn1-01 1.952(3) Mn4-04 2.117(3)
Mn1-O20 1.977(4) Mn4-O21' 2.143(5)
Mn1-Ol8 2.139(4) Mn5-04' 1.887(3)
Mn1-O22 2.301(4) Mn5-O2 1.905(3)
Mn1-Mn2 3.1220(11) Mn5-O3' 1.939(3)
Mn2-04 1.842(3) Mn5-Ol7 1.961(3)
Mn2-013 1.902(4) Mn5-O23' 2.212(4)
Mn3 -01-Mn1 132.96(18) Mn5-O3 2.466(3)
Mn3 -01-Mn2 108.24(17) Mn5'-O3-Mn5 94.11(13)
Mn1-01-Mn2 106.15(15) Mn1-O3-Mn5 124.33(15)
Mn3-01-Dyl 101.87(13) Mn4'-O3-Mn5 91.20(13)
Mn1-01-Dy l 102.70(15) Mn2-04-Mn5' 124.77(19)
Mn2-01-Dy l 100.00(13) Mn2-04-Mn4 118.35(16)
Mn3-O2-Mn5 126.65(18) Mn5'-04-Mn4 104.75(15)
Mn3-O2-Mn4 124.95(18) Mn1-011-Dy l 110.13(18)
Mn5-O2-Mn4 98.51(14) Mn3-012-Dyl 107.18(14)
Mn5'-O3-Mn1 124.39(17) Mn2-013-Dy l 108.13(17)
Mn5'-O3-Mn4' 96.34(14) Mn2-014-Mn3 85.69(16)
Mn1-O3-Mn4' 118.47(17) Mn2-O22-Mn1 88.14(13)










Table A-23. Selected interatomic distances (A+) and angles (o) for
[MnloY20s(O2C~h)10(hmp)6(NO3)4]-4MeCN (8-9-4MeCN)
Y1-011 2.290(5) Mn2-014
Y1-013 2.296(6) Mn3 -O2
Y1-012 2.334(4) Mn3 -012
Y1-06 2.432(7) Mn3 1
Y1-08 2.450(5) Mn3 -N5
Y1-01 2.450(4) Mn3 -Ol6
Y1-N3 2.465(7) Mn3 -014
Y1-05 2.475(7) Mn4-O2
Y1-09 2.497(6) Mn4-Ol9'
Y1-N1 2.885(8) Mn4-O3'
Y1-N2 2.910(7) Mn4-Ol5
Mn1-011 1.908(5) Mn4-04
Mn1-O3 1.921(4) Mn4-O21'
Mn1-01 1.947(4) Mn5-04'
Mn1-O20 1.967(5) Mn5-O2
Mn1-Ol8 2.137(5) Mn5-O3'
Mn1-O22 2.291(5) Mn5-Ol7
Mn2-04 1.843(4) Mn5-O23'
Mn2-013 1.895(5) Mn5-O3
Mn2-01 1.947(5) Mn5-Mn4'
Mn2-N4 2.014(6) Mn5-Mn5'
Mn2-O22 2.181(5) 03-Mn5'
Mn3 -01-Mn2 107.8(2) Mn1-O3-Mn5
Mn3 -01-Mn1 132.9(2) Mn5'-O3-Mn5
Mn2-01-Mn1 105.72(18) Mn4'-O3-Mn5
Mn3-01-Y1 102.26(16) Mn2-04-Mn5'
Mn2-01-Y1 100.61(17) Mn2-04-Mn4
Mn1-01-Y1 103.05(19) Mn5'-04-Mn4
Mn3-O2-Mn5 126.5(2) Mn1-011-Y1
Mn3 -O2-Mn4 124.8(2) Mn3-012-Y1
Mn5-O2-Mn4 98.53(17) Mn2-013-Y1
Mn1-O3-Mn5' 124.1(2) Mn2-014-Mn3
Mn1-O3-Mn4' 118.8(2) Mn2-O22-Mn1


2.285(7)
1.869(4)
1.892(4)
1.911(4)
2.015(5)
2.091(5)
2.470(8)
1.917(4)
1.948(5)
1.962(4)
1.998(7)
2.081(4)
2.100(6)
1.895(4)
1.910(4)
1.933(4)
1.945(5)
2.213(5)
2.439(4)
3.1527(14)
3.2167(19)
1.933(4)
124.78(18)
94.02(17)
90.85(16)
124.5(2)
118.21(19)
104.82(19)
110.5(2)
107.22(17)
108.0(2)

81.8(3)
87.87(16)









APPENDIX B
LIST OF COMPOUNDS

[Mn402(O2CMe)s(salpro)] (2-1)

[Mn402(O2CEt>,(salpro)] (2-2)

[Mn402(O2CBur~s(salpro)] (2-3)

NMe4 [Mn(O2CPh)2(SalproH)] (2-4)

[Mn4(hmp)4(pdmH)2(MeCN)4](CIO4)4 (3-1)

[Mn2501s(OH)2(hmp)6(pdm)s(pdm H)2(L)2](C IO1 (3-2)

[Fe7O4(O2CPh)ll(dmem)2] (4-1)

[FeO4(O2CMe)ll(dmem)2] (4-2)

[Fe602(OH)4(O2CCBur~s(dmem)2] (4-3)

[Fe30(O2CBut)2(N3)3(dmem)2] (4-4)

[FesO2(OH)(O2CMe)s(hmbp)3](CIO4)2 (5-1)

[Fe602(OH)2(O2CPh)6(hmbp)4] (NO3)2 (5-2)

[Fe602(OH)2(O2CMe)6(hmbp)4](NO3)2 (5-3)

[F e602(OH)2(O2 CBut)6(hmb p)4](NO3)2 (5-4)

[FesOs(OH)2(O2C~ut)28(heen)4] (-1)

[Fe904(OH)4(O2CPh)13(heenH)2] (6-2)

4[Fe7O3(OMe)3(MeOH)1.s(heen)3 14.5(H20)]Cl[FeCl4] 6-3)

[F e602(O2 CPh)5 (heen)3(heenH)] (CIO 4)2 (6- 4)

[MnsO3(OH)(OMe)(O2CPh)7(edte)(edteH2)] (O2CPh) (7-1)

[Mnl204(OH)2(edte)4C 6(H20)2] (7-2)

[Mn200s(OH)4(O2CMe)6(edte)6](CIO4)2 (7-3)

[F e O2(O2 CPh)7(edte)(H2 0)] (7-4)









[Fe602(O2CBur~s(edteH)2] (7-5)

[F el2t04(OH)2(O2 CMe)6(e dte)4(H20)2 ](CIO4)4 (7-6)

[Fel204(OH)s(edte)4(H20)2](CIO4)4 (-7)

[Fel204(OH>,(edte)4(H20)2](NO3)4 (78)

[MnloLn20s(O2C~h)10(hmp)6(NO3)4] (Ln = Pr (8-1), Nd (8-2), Sm (8-3), Gd (8-4), Tb (8-5),

Dy (8-6), Ho (8-7), Er (8-8) Y (8-9))

[Mnl2012(O2CCHCl2)16(H20)4] 9-1)

(NPrn4)[Mnl2012(O2CCHCl2)16(H20)4] (9-2)

(NPrn4)2 [Mnl12012(O2CCHCl2)1 6(H20)4] 9-3)

(NPrn4)3 [Mnl2012(O2CCHCl2)16(H20)4] (9-4)

(NMe4)3 [Mnl2012(O2CCHCl2)16(H20)4] (9-5)









APPENDIX C
PHYSICAL MEASUREMENTS

Infrared spectra were recorded in the solid state (KBr pellets) on a Nicolet Nexus 670

FTIR spectrometer in the 400 4000 cml range. Elemental analyses (C, H and N) were

performed by the in-house facilities of the University of Florida, Chemistry Department. Cl

analysis was performed by Complete Analysis Laboratories, Inc. in Parsippany, New Jersey.

Variable-temperature dc and ac magnetic susceptibility data were collected at the University of

Florida using a Quantum Design MPMS-XL SQUID susceptometer equipped with a 7 T magnet

and operating in the 1.8 300 K range. Samples were embedded in solid eicosane to prevent

torquing. Magnetization vs. field and temperature data was fit using the programM2~AGNET.53

Pascal's constants were used to estimate the diamagnetic correction, which was subtracted from

the experimental susceptibility to give the molar paramagnetic susceptibility (X, r). Double-axis

angle- dependent high-frequency electron paramagnetic resonance (HFEPR) studies were

performed on single crystals using a rotating cavity296 and a 7 T transverse magnetic field, which

can be rotated about an axis perpendicular to the axis of the rotating cavity. In addition, a 17 T

axial magnet was employed for some single-axis measurements. The experiments were carried

out over a wide range of frequencies (50-200 GHz) and with the sample at temperatures in the

1.8-20 K range.









APPENDIX D
VAN VLECK EQUATIONS

p = paramagnetic impurity
c = NCLB2/3k
N= Avogadro's number
g = Lande's factor
k = Boltzmann constant
T= Temperature
TIP = Temperature independent paramagnetism

D-1 [Mn402(O2CMe)s(salpro)] (2-1)


M~rn2 4n


Mn

XM = (c g2)/T (Num/Den) + TIP

l= Jbb/k/T
m= Jbw/k/T
n= Jww/k/T

Num=+ 300.0000 *exp( 0.0000 *1+ 0.0000 *m+ 0.0000 *n)
+ 6.0000 *exp( 2.0000 *1+ 0.0000 *m+ 0.0000 *n)
+ 0.0000 *exp( 2.0000 *1+ -4.0000 *m+ 2.0000 *n)
+ 300.0000 *exp( 2.0000 *1+ -2.0000 *m+ 2.0000 *n)
+ 30.0000 *exp( 2.0000 *1+ 2.0000 *m+ 2.0000 *n)
+ 6.0000 *exp( 2.0000 *1+ -6.0000 *m+ 6.0000 *n)
+ 84.0000 *exp( 2.0000 *1+ 4.0000 *m+ 6.0000 *n)
+ 30.0000 *exp( 2.0000 *1+ -8.0000 *m+ 12.0000 *n)
+ 180.0000 *exp( 2.0000 *1+ 6.0000 *m+ 12.0000 *n)
+ 84.0000 *exp( 2.0000 *1+ -10.0000 *m+ 20.0000 *n)
+ 330.0000 *exp( 2.0000 *1+ 8.0000 *m+ 20.0000 *n)
+ 114.0000 *exp( 6.0000 *1+ 0.0000 *m+ 0.0000 *n)
+ 300.0000 *exp( 6.0000 *1+ -6.0000 *m+ 2.0000 *n)
+ 30.0000 *exp( 6.0000 *1+ -2.0000 *m+ 2.0000 *n)
+ 414.0000 *exp( 6.0000 *1+ 4.0000 *m+ 2.0000 *n)
+ 30.0000 *exp( 6.0000 *1+ -12.0000 *m+ 6.0000 *n)
+ 6.0000 *exp( 6.0000 *1+ -10.0000 *m+ 6.0000 *n)
+ 180.0000 *exp( 6.0000 *1+ 8.0000 *m+ 6.0000 *n)
+ 6.0000 *exp( 6.0000 *1+ -16.0000 *m+ 12.0000 *n)
+ 180.0000 *exp( 6.0000 *1+ 2.0000 *m+ 12.0000 *n)
+ 330.0000 *exp( 6.0000 *1+ 12.0000 *m+ 12.0000 *n)









+ 30.0000 *exp( 6.0000 *1+ -20.0000 *m+ 20.0000 *n)
+ 84.0000 *exp( 6.0000 *1+ -14.0000 *m+ 20.0000 *n)
+ 546.0000 *exp( 6.0000 *1+ 16.0000 *m+ 20.0000 *n)
+ 84.0000 *exp( 12.0000 *1+ 0.0000 *m+ 0.0000 *n)
+ 30.0000 *exp( 12.0000 *1+ -8.0000 *m+ 2.0000 *n)
+ 414.0000 *exp( 12.0000 *1+ -2.0000 *m+ 2.0000 *n)
+ 510.0000 *exp( 12.0000 *1+ 6.0000 *m+ 2.0000 *n)
+ 6.0000 *exp( 12.0000 *1+ -16.0000 *m+ 6.0000 *n)
+ 294.0000 *exp( 12.0000 *1+ -12.0000 *m+ 6.0000 *n)
+ 84.0000 *exp( 12.0000 *1+ -6.0000 *m+ 6.0000 *n)
+ 180.0000 *exp( 12.0000 *1+ 2.0000 *m+ 6.0000 *n)
+ 330.0000 *exp( 12.0000 *1+ 12.0000 *m+ 6.0000 *n)
+ 0.0000 *exp( 12.0000 *1+ -24.0000 *m+ 12.0000 *n)
+ 6.0000 *exp( 12.0000 *1+ -22.0000 *m+ 12.0000 *n)
+ 30.0000 *exp( 12.0000 *1+ -18.0000 *m+ 12.0000 *n)
+ 180.0000 *exp( 12.0000 *1+ -4.0000 *m+ 12.0000 *n)
+ 546.0000 *exp( 12.0000 *1+ 18.0000 *m+ 12.0000 *n)
+ 6.0000 *exp( 12.0000 *1+ -30.0000 *m+ 20.0000 *n)
+ 30.0000 *exp( 12.0000 *1+ -26.0000 *m+ 20.0000 *n)
+ 84.0000 *exp( 12.0000 *1+ -20.0000 *m+ 20.0000 *n)
+ 546.0000 *exp( 12.0000 *1+ 10.0000 *m+ 20.0000 *n)
+ 840.0000 *exp( 12.0000 *1+ 24.0000 *m+ 20.0000 *n)
+ 180.0000 *exp( 20.0000 *1+ 0.0000 *m+ 0.0000 *n)
+ 414.0000 *exp( 20.0000 *1+ -10.0000 *m+ 2.0000 *n)
+ 510.0000 *exp( 20.0000 *1+ -2.0000 *m+ 2.0000 *n)
+ 330.0000 *exp( 20.0000 *1+ 8.0000 *m+ 2.0000 *n)
+ 294.0000 *exp( 20.0000 *1+ -20.0000 *m+ 6.0000 *n)
+ 84.0000 *exp( 20.0000 *1+ -14.0000 *m+ 6.0000 *n)
+ 180.0000 *exp( 20.0000 *1+ -6.0000 *m+ 6.0000 *n)
+ 330.0000 *exp( 20.0000 *1+ 4.0000 *m+ 6.0000 *n)
+ 1386.0000 *exp( 20.0000 *1+ 16.0000 *m+ 6.0000 *n)
+ 6.0000 *exp( 20.0000 *1+ -30.0000 *m+ 12.0000 *n)
+ 30.0000 *exp( 20.0000 *1+ -26.0000 *m+ 12.0000 *n)
+ 180.0000 *exp( 20.0000 *1+ -12.0000 *m+ 12.0000 *n)
+ 546.0000 *exp( 20.0000 *1+ 10.0000 *m+ 12.0000 *n)
+ 840.0000 *exp( 20.0000 *1+ 24.0000 *m+ 12.0000 *n)
+ 0.0000 *exp( 20.0000 *1+ -40.0000 *m+ 20.0000 *n)
+ 6.0000 *exp( 20.0000 *1+ -38.0000 *m+ 20.0000 *n)
+ 30.0000 *exp( 20.0000 *1+ -34.0000 *m+ 20.0000 *n)
+ 84.0000 *exp( 20.0000 *1+ -28.0000 *m+ 20.0000 *n)
+ 546.0000 *exp( 20.0000 *1+ 2.0000 *m+ 20.0000 *n)
+ 1224.0000 *exp( 20.0000 *1+ 32.0000 *m+ 20.0000 *n)

Den=+ 25.0000 *exp( 0.0000 *1+ 0.0000 *m+ 0.0000 *n)
+ 3.0000 *exp( 2.0000 *1+ 0.0000 *m+ 0.0000 *n)
+ 1.0000 *exp( 2.0000 *1+ -4.0000 *m+ 2.0000 *n)









+ 24.0000 *exp( 2.0000 *1+ -2.0000 *m+ 2.0000 *n)
+ 5.0000 *exp( 2.0000 *1+ 2.0000 *m+ 2.0000 *n)
+ 3.0000 *exp( 2.0000 *1+ -6.0000 *m+ 6.0000 *n)
+ 7.0000 *exp( 2.0000 *1+ 4.0000 *m+ 6.0000 *n)
+ 5.0000 *exp( 2.0000 *1+ -8.0000 *m+ 12.0000 *n)
+ 9.0000 *exp( 2.0000 *1+ 6.0000 *m+ 12.0000 *n)
+ 7.0000 *exp( 2.0000 *1+ -10.0000 *m+ 20.0000 *n)
+ 11.0000 *exp( 2.0000 *1+ 8.0000 *m+ 20.0000 *n)
+ 12.0000 *exp( 6.0000 *1+ 0.0000 *m+ 0.0000 *n)
+ 24.0000 *exp( 6.0000 *1+ -6.0000 *m+ 2.0000 *n)
+ 5.0000 *exp( 6.0000 *1+ -2.0000 *m+ 2.0000 *n)
+ 18.0000 *exp( 6.0000 *1+ 4.0000 *m+ 2.0000 *n)
+ 6.0000 *exp( 6.0000 *1+ -12.0000 *m+ 6.0000 *n)
+ 3.0000 *exp( 6.0000 *1+ -10.0000 *m+ 6.0000 *n)
+ 9.0000 *exp( 6.0000 *1+ 8.0000 *m+ 6.0000 *n)
+ 3.0000 *exp( 6.0000 *1+ -16.0000 *m+ 12.0000 *n)
+ 9.0000 *exp( 6.0000 *1+ 2.0000 *m+ 12.0000 *n)
+ 11.0000 *exp( 6.0000 *1+ 12.0000 *m+ 12.0000 *n)
+ 5.0000 *exp( 6.0000 *1+ -20.0000 *m+ 20.0000 *n)
+ 7.0000 *exp( 6.0000 *1+ -14.0000 *m+ 20.0000 *n)
+ 13.0000 *exp( 6.0000 *1+ 16.0000 *m+ 20.0000 *n)
+ 7.0000 *exp( 12.0000 *1+ 0.0000 *m+ 0.0000 *n)
+ 5.0000 *exp( 12.0000 *1+ -8.0000 *m+ 2.0000 *n)
+ 18.0000 *exp( 12.0000 *1+ -2.0000 *m+ 2.0000 *n)
+ 20.0000 *exp( 12.0000 *1+ 6.0000 *m+ 2.0000 *n)
+ 3.0000 *exp( 12.0000 *1+ -16.0000 *m+ 6.0000 *n)
+ 21.0000 *exp( 12.0000 *1+ -12.0000 *m+ 6.0000 *n)
+ 7.0000 *exp( 12.0000 *1+ -6.0000 *m+ 6.0000 *n)
+ 9.0000 *exp( 12.0000 *1+ 2.0000 *m+ 6.0000 *n)
+ 11.0000 *exp( 12.0000 *1+ 12.0000 *m+ 6.0000 *n)
+ 1.0000 *exp( 12.0000 *1+ -24.0000 *m+ 12.0000 *n)
+ 3.0000 *exp( 12.0000 *1+ -22.0000 *m+ 12.0000 *n)
+ 5.0000 *exp( 12.0000 *1+ -18.0000 *m+ 12.0000 *n)
+ 9.0000 *exp( 12.0000 *1+ -4.0000 *m+ 12.0000 *n)
+ 13.0000 *exp( 12.0000 *1+ 18.0000 *m+ 12.0000 *n)
+ 3.0000 *exp( 12.0000 *1+ -30.0000 *m+ 20.0000 *n)
+ 5.0000 *exp( 12.0000 *1+ -26.0000 *m+ 20.0000 *n)
+ 7.0000 *exp( 12.0000 *1+ -20.0000 *m+ 20.0000 *n)
+ 13.0000 *exp( 12.0000 *1+ 10.0000 *m+ 20.0000 *n)
+ 15.0000 *exp( 12.0000 *1+ 24.0000 *m+ 20.0000 *n)
+ 9.0000 *exp( 20.0000 *1+ 0.0000 *m+ 0.0000 *n)
+ 18.0000 *exp( 20.0000 *1+ -10.0000 *m+ 2.0000 *n)
+ 20.0000 *exp( 20.0000 *1+ -2.0000 *m+ 2.0000 *n)
+ 11.0000 *exp( 20.0000 *1+ 8.0000 *m+ 2.0000 *n)
+ 21.0000 *exp( 20.0000 *1+ -20.0000 *m+ 6.0000 *n)
+ 7.0000 *exp( 20.0000 *1+ -14.0000 *m+ 6.0000 *n)









+ 9.0000 *exp( 20.0000 *1+ -6.0000 *m+ 6.0000 *n)
+ 11.0000 *exp( 20.0000 *1+ 4.0000 *m+ 6.0000 *n)
+ 28.0000 *exp( 20.0000 *1+ 16.0000 *m+ 6.0000 *n)
+ 3.0000 *exp( 20.0000 *1+ -30.0000 *m+ 12.0000 *n)
+ 5.0000 *exp( 20.0000 *1+ -26.0000 *m+ 12.0000 *n)
+ 9.0000 *exp( 20.0000 *1+ -12.0000 *m+ 12.0000 *n)
+ 13.0000 *exp( 20.0000 *1+ 10.0000 *m+ 12.0000 *n)
+ 15.0000 *exp( 20.0000 *1+ 24.0000 *m+ 12.0000 *n)
+ 1.0000 *exp( 20.0000 *1+ -40.0000 *m+ 20.0000 *n)
+ 3.0000 *exp( 20.0000 *1+ -38.0000 *m+ 20.0000 *n)
+ 5.0000 *exp( 20.0000 *1+ -34.0000 *m+ 20.0000 *n)
+ 7.0000 *exp( 20.0000 *1+ -28.0000 *m+ 20.0000 *n)
+ 13.0000 *exp( 20.0000 *1+ 2.0000 *m+ 20.0000 *n)
+ 17.0000 *exp( 20.0000 *1+ 32.0000 *m+ 20.0000 *n)


D-2 [Mn4(hmp)4(pdmH)2(MeCN)4] (CIO4)4 (3-1
Mn1



Mn2 Jbb Mn2'



MMl'

XM = (c g2)/T (Num/Den) + TIP

l= Jbb/k/T
m= Jbw/k/T

Num=+ 630.0000 *exp( 0.0000 *1+ 0.0000 *m)+ 6.0000 *exp( 2.0000 *1+ 0.0000 *m)
+ 0.0000 *exp( 2.0000 *1+ -4.0000 *m)+ 630.0000 *exp( 2.0000 *1+ -2.0000 *m)
+ 30.0000 *exp( 2.0000 *1+ 2.0000 *m)+ 6.0000 *exp( 2.0000 *1+ -6.0000 *m)
+ 84.0000 *exp( 2.0000 *1+ 4.0000 *m)+ 30.0000 *exp( 2.0000 *1+ -8.0000 *m)
+ 180.0000 *exp( 2.0000 *1+ 6.0000 *m)+ 84.0000 *exp( 2.0000 *1+ -10.0000 *m)
+ 330.0000 *exp( 2.0000 *1+ 8.0000 *m)+ 180.0000 *exp( 2.0000 *1+ -12.0000 *m)
+ 546.0000 *exp( 2.0000 *1+ 10.0000 *m)+ 114.0000 *exp( 6.0000 *1+ 0.0000 *m)
+ 630.0000 *exp( 6.0000 *1+ -6.0000 *m)+ 30.0000 *exp( 6.0000 *1+ -2.0000 *m)
+ 414.0000 *exp( 6.0000 *1+ 4.0000 *m)+ 30.0000 *exp( 6.0000 *1+ -12.0000 *m)
+ 6.0000 *exp( 6.0000 *1+ -10.0000 *m)+ 180.0000 *exp( 6.0000 *1+ 8.0000 *m)
+ 186.0000 *exp( 6.0000 *1+ -16.0000 *m)+ 180.0000 *exp( 6.0000 *1+ 2.0000 *m)
+ 330.0000 *exp( 6.0000 *1+ 12.0000 *m)+ 30.0000 *exp( 6.0000 *1+ -20.0000 *m)
+ 84.0000 *exp( 6.0000 *1+ -14.0000 *m)+ 546.0000 *exp( 6.0000 *1+ 16.0000 *m)
+ 84.0000 *exp( 6.0000 *1+ -24.0000 *m)+ 546.0000 *exp( 6.0000 *1+ 6.0000 *m)









+ 840.0000 *exp( 6.0000 *1+ 20.0000 *m)+ 630.0000 *exp( 12.0000 *1+ 0.0000 *m)
+ 30.0000 *exp( 12.0000 *1+ -8.0000 *m)+ 414.0000 *exp( 12.0000 *1+ -2.0000 *m)
+ 510.0000 *exp( 12.0000 *1+ 6.0000 *m)+ 6.0000 *exp( 12.0000 *1+ -16.0000 *m)
+ 624.0000 *exp( 12.0000 *1+ -12.0000 *m)+ 84.0000 *exp( 12.0000 *1+ -6.0000 *m)
+ 180.0000 *exp( 12.0000 *1+ 2.0000 *m)+ 330.0000 *exp( 12.0000 *1+ 12.0000 *m)
+ 0.0000 *exp( 12.0000 *1+ -24.0000 *m)+ 186.0000 *exp( 12.0000 *1+ -22.0000 *m)
+ 30.0000 *exp( 12.0000 *1+ -18.0000 *m)+ 180.0000 *exp( 12.0000 *1+ -4.0000 *m)
+ 546.0000 *exp( 12.0000 *1+ 18.0000 *m)+ 90.0000 *exp( 12.0000 *1+ -30.0000 *m)
+ 30.0000 *exp( 12.0000 *1+ -26.0000 *m)+ 84.0000 *exp( 12.0000 *1+ -20.0000 *m)
+ 546.0000 *exp( 12.0000 *1+ 10.0000 *m)+ 840.0000 *exp( 12.0000 *1+ 24.0000 *m)
+ 30.0000 *exp( 12.0000 *1+ -36.0000 *m)+ 840.0000 *exp( 12.0000 *1+ 14.0000 *m)
+ 1224.0000 *exp( 12.0000 *1+ 30.0000 *m)+ 180.0000 *exp( 20.0000 *1+ 0.0000 *m)
+ 414.0000 *exp( 20.0000 *1+ -10.0000 *m)+ 510.0000 *exp( 20.0000 *1+ -2.0000 *m)
+ 330.0000 *exp( 20.0000 *1+ 8.0000 *m)+ 624.0000 *exp( 20.0000 *1+ -20.0000 *m)
+ 84.0000 *exp( 20.0000 *1+ -14.0000 *m)+ 180.0000 *exp( 20.0000 *1+ -6.0000 *m)
+ 330.0000 *exp( 20.0000 *1+ 4.0000 *m)+ 1386.0000 *exp( 20.0000 *1+ 16.0000 *m)
+ 186.0000 *exp( 20.0000 *1+ -30.0000 *m)+ 30.0000 *exp( 20.0000 *1+ -26.0000 *m)
+ 180.0000 *exp( 20.0000 *1+ -12.0000 *m)+ 546.0000 *exp( 20.0000 *1+ 10.0000 *m)
+ 840.0000 *exp( 20.0000 *1+ 24.0000 *m)+ 0.0000 *exp( 20.0000 *1+ -40.0000 *m)
+ 90.0000 *exp( 20.0000 *1+ -38.0000 *m)+ 30.0000 *exp( 20.0000 *1+ -34.0000 *m)
+ 84.0000 *exp( 20.0000 *1+ -28.0000 *m)+ 546.0000 *exp( 20.0000 *1+ 2.0000 *m)
+ 1224.0000 *exp( 20.0000 *1+ 32.0000 *m)+ 6.0000 *exp( 20.0000 *1+ -48.0000 *m)
+ 30.0000 *exp( 20.0000 *1+ -44.0000 *m)+ 546.0000 *exp( 20.0000 *1+ -8.0000 *m)
+ 840.0000 *exp( 20.0000 *1+ 6.0000 *m)+ 1224.0000 *exp( 20.0000 *1+ 22.0000 *m)
+ 1710.0000 *exp( 20.0000 *1+ 40.0000 *m)

Den=+ 36.0000 *exp( 0.0000 *1+ 0.0000 *m)+ 3.0000 *exp( 2.0000 *1+ 0.0000 *m)
+ 1.0000 *exp( 2.0000 *1+ -4.0000 *m)+ 35.0000 *exp( 2.0000 *1+ -2.0000 *m)
+ 5.0000 *exp( 2.0000 *1+ 2.0000 *m)+ 3.0000 *exp( 2.0000 *1+ -6.0000 *m)
+ 7.0000 *exp( 2.0000 *1+ 4.0000 *m)+ 5.0000 *exp( 2.0000 *1+ -8.0000 *m)
+ 9.0000 *exp( 2.0000 *1+ 6.0000 *m)+ 7.0000 *exp( 2.0000 *1+ -10.0000 *m)
+ 11.0000 *exp( 2.0000 *1+ 8.0000 *m)+ 9.0000 *exp( 2.0000 *1+ -12.0000 *m)
+ 13.0000 *exp( 2.0000 *1+ 10.0000 *m)+ 12.0000 *exp( 6.0000 *1+ 0.0000 *m)
+ 35.0000 *exp( 6.0000 *1+ -6.0000 *m)+ 5.0000 *exp( 6.0000 *1+ -2.0000 *m)
+ 18.0000 *exp( 6.0000 *1+ 4.0000 *m)+ 6.0000 *exp( 6.0000 *1+ -12.0000 *m)
+ 3.0000 *exp( 6.0000 *1+ -10.0000 *m)+ 9.0000 *exp( 6.0000 *1+ 8.0000 *m)
+ 12.0000 *exp( 6.0000 *1+ -16.0000 *m)+ 9.0000 *exp( 6.0000 *1+ 2.0000 *m)
+ 11.0000 *exp( 6.0000 *1+ 12.0000 *m)+ 5.0000 *exp( 6.0000 *1+ -20.0000 *m)
+ 7.0000 *exp( 6.0000 *1+ -14.0000 *m)+ 13.0000 *exp( 6.0000 *1+ 16.0000 *m)
+ 7.0000 *exp( 6.0000 *1+ -24.0000 *m)+ 13.0000 *exp( 6.0000 *1+ 6.0000 *m)
+ 15.0000 *exp( 6.0000 *1+ 20.0000 *m)+ 20.0000 *exp( 12.0000 *1+ 0.0000 *m)
+ 5.0000 *exp( 12.0000 *1+ -8.0000 *m)+ 18.0000 *exp( 12.0000 *1+ -2.0000 *m)
+ 20.0000 *exp( 12.0000 *1+ 6.0000 *m)+ 3.0000 *exp( 12.0000 *1+ -16.0000 *m)
+ 32.0000 *exp( 12.0000 *1+ -12.0000 *m)+ 7.0000 *exp( 12.0000 *1+ -6.0000 *m)
+ 9.0000 *exp( 12.0000 *1+ 2.0000 *m)+ 11.0000 *exp( 12.0000 *1+ 12.0000 *m)
+ 1.0000 *exp( 12.0000 *1+ -24.0000 *m)+ 12.0000 *exp( 12.0000 *1+ -22.0000 *m)









+ 5.0000 *exp( 12.0000 *1+ -18.0000 *m)+ 9.0000 *exp( 12.0000 *1+ -4.0000 *m)
+ 13.0000 *exp( 12.0000 *1+ 18.0000 *m)+ 10.0000 *exp( 12.0000 *1+ -30.0000 *m)
+ 5.0000 *exp( 12.0000 *1+ -26.0000 *m)+ 7.0000 *exp( 12.0000 *1+ -20.0000 *m)
+ 13.0000 *exp( 12.0000 *1+ 10.0000 *m)+ 15.0000 *exp( 12.0000 *1+ 24.0000 *m)
+ 5.0000 *exp( 12.0000 *1+ -36.0000 *m)+ 15.0000 *exp( 12.0000 *1+ 14.0000 *m)
+ 17.0000 *exp( 12.0000 *1+ 30.0000 *m)+ 9.0000 *exp( 20.0000 *1+ 0.0000 *m)
+ 18.0000 *exp( 20.0000 *1+ -10.0000 *m)+ 20.0000 *exp( 20.0000 *1+ -2.0000 *m)
+ 11.0000 *exp( 20.0000 *1+ 8.0000 *m)+ 32.0000 *exp( 20.0000 *1+ -20.0000 *m)
+ 7.0000 *exp( 20.0000 *1+ -14.0000 *m)+ 9.0000 *exp( 20.0000 *1+ -6.0000 *m)
+ 11.0000 *exp( 20.0000 *1+ 4.0000 *m)+ 28.0000 *exp( 20.0000 *1+ 16.0000 *m)
+ 12.0000 *exp( 20.0000 *1+ -30.0000 *m)+ 5.0000 *exp( 20.0000 *1+ -26.0000 *m)
+ 9.0000 *exp( 20.0000 *1+ -12.0000 *m)+ 13.0000 *exp( 20.0000 *1+ 10.0000 *m)
+ 15.0000 *exp( 20.0000 *1+ 24.0000 *m)+ 1.0000 *exp( 20.0000 *1+ -40.0000 *m)
+ 10.0000 *exp( 20.0000 *1+ -38.0000 *m)+ 5.0000 *exp( 20.0000 *1+ -34.0000 *m)
+ 7.0000 *exp( 20.0000 *1+ -28.0000 *m)+ 13.0000 *exp( 20.0000 *1+ 2.0000 *m)
+ 17.0000 *exp( 20.0000 *1+ 32.0000 *m)+ 3.0000 *exp( 20.0000 *1+ -48.0000 *m)
+ 5.0000 *exp( 20.0000 *1+ -44.0000 *m)+ 13.0000 *exp( 20.0000 *1+ -8.0000 *m)
+ 15.0000 *exp( 20.0000 *1+ 6.0000 *m)+ 17.0000 *exp( 20.0000 *1+ 22.0000 *m)
+ 19.0000 *exp( 20.0000 *1+ 40.0000 *m)


D.3 [Fe30(O2CBut)2(N3)3(dmmem) (4-4)

Fe2

O 03



Fe3 C~Fel


XM = (c g2)/T (Num/Den)

m=Ja/k/T
n= Jb/k/T

Num = + 52.5000 exp( 8.7500 m+ 0.0000 n) + 15.0000 exp( 1.7500 m+ 2.0000 n)
+ 52.5000 exp( 6.7500 m+ 2.0000 n) + 126.0000 exp( 13.7500 m+ 2.0000 n)
+ 1.5000 exp( -5.2500 m+ 6.0000 n) + 15.0000 exp( -2.2500 m+ 6.0000 n)
+ 52.5000 exp( 2.7500 m+ 6.0000 n) + 126.0000 exp( 9.7500 m+ 6.0000 n)
+ 247.5000 exp( 18.7500 m+ 6.0000 n) + 1.5000 exp( -11.2500 m+ 12.0000 n)
+ 15.0000 exp( -8.2500 m+ 12.0000 n) + 52.5000 exp( -3.2500 m+ 12.0000 n)
+ 126.0000 exp( 3.7500 m+ 12.0000 n)+ 247.5000 exp( 12.7500 m+ 12.0000 n)
+ 429.0000 exp( 23.7500 m+ 12.0000 n)+ 15.0000 exp( -16.2500 m+ 20.0000 n)
+ 52.5000 exp( -11.2500 m+ 20.0000 n)+ 126.0000 exp( -4.2500 m+ 20.0000 n)









+ 247.5000 exp( 4.7500 m+ 20.0000 n)+ 429.0000 exp( 15.7500 m+ 20.0000 n)
+ 682.5000 exp( 28.7500 m+ 20.0000 n)+ 52.5000 exp( -21.2500 m+ 30.0000 n)
+ 126.0000 exp( -14.2500 m+ 30.0000 n)+ 247.5000 exp( -5.2500 m+ 30.0000 n)
+ 429.0000 exp( 5.7500 m+ 30.0000 n)+ 682.5000 exp( 18.7500 m+ 30.0000 n)
+ 1020.0000 exp( 33.7500 m+ 30.0000 n)

Den = + 6.0000 exp( 8.7500 m+ 0.0000 n)+ 4.0000 exp( 1.7500 m+ 2.0000 n)
+ 6.0000 exp( 6.7500 m+ 2.0000 n)+ 8.0000 exp( 13.7500 m+ 2.0000 n)
+ 2.0000 exp( -5.2500 m+ 6.0000 n)+ 4.0000 exp( -2.2500 m+ 6.0000 n)
+ 6.0000 exp( 2.7500 m+ 6.0000 n)+ 8.0000 exp( 9.7500 m+ 6.0000 n)
+ 10.0000 exp( 18.7500 m+ 6.0000 n)+ 2.0000 exp( -11.2500 m+ 12.0000 n)
+ 4.0000 exp( -8.2500 m+ 12.0000 n)+ 6.0000 exp( -3.2500 m+ 12.0000 n)
+ 8.0000 exp( 3.7500 m+ 12.0000 n)+ 10.0000 exp( 12.7500 m+ 12.0000 n)
+ 12.0000 exp( 23.7500 m+ 12.0000 n)+ 4.0000 exp( -16.2500 m+ 20.0000 n)
+ 6.0000 exp( -11.2500 m+ 20.0000 n)+ 8.0000 exp( -4.2500 m+ 20.0000 n)
+ 10.0000 exp( 4.7500 m+ 20.0000 n)+ 12.0000 exp( 15.7500 m+ 20.0000 n)
+ 14.0000 exp( 28.7500 m+ 20.0000 n)+ 6.0000 exp( -21.2500 m+ 30.0000 n)
+ 8.0000 exp( -14.2500 m+ 30.0000 n)+ 10.0000 exp( -5.2500 m+ 30.0000 n)
+ 12.0000 exp( 5.7500 m+ 30.0000 n)+ 14.0000 exp( 18.7500 m+ 30.0000 n)
+ 16.0000 exp( 33.7500 m+ 30.0000 n)










LIST OF REFERENCES

(1) Stohr, J.; Siegmann, H. C. Mapletisnz: From Fundamentalsd~~~dd~~~ddd~~ to Nanoscale Dynamnics;
Springer 2006; Vol. 152.

(2) Jiles, D. Introduction to Mapletisn; and Maletic Materals; Chapman & Hall:
NewYork, 1991.

(3) Huheey, J. E.; Keiter, E. A.; Keiter, R. L. hzorgan2ic Chentistry: Principles of Structure
and Reactivity; 4th ed.; HarperCollin College: New York, 1993.

(4) Kahn, O. Molecular Mapletisnz; VCH: Weinheim, Germany, 1993.

(5) Smart, L.; Moore, E. Solid State Chentistly; Chapman & Hall: New York, 1993.

(6) Miller, J. S.; Epstein, A. J. Angew. Chens. Int. Ed. 1994, 33, 385.

(7) Epstein, A. J. M~RS Bulletin 2000, 25, 33.

(8) Drago, R. S. Physical M~ethods for Chemists; 211d ed.; Surfside Scientific Publishers, 1992.

(9) Aromi, G.; Aubin, S. M. J.; Bolcar, M. A.; Christou, G.; Eppley, H. J.; Folting, K.;
Hendrickson, D. N.; Huffman, J. C.; Squire, R. C.; Tsai, H. L.; Wang, S.; Wemple, M. W.
Polyhedron 1998, 1 7, 3005.

(10) Miller, J. S.; Epstein, A. J. M~RS Bulletin 2000, 25, 21.

(11) Mapletisnz: A Supra~nolecular Function; Kahn, O., Ed.; Kluwer: Dordrecht, 1996.

(12) Lis, T. Acta Crystallogr., Sect. B: Struct. Sci. 1980, 36, 2042.

(13) Christou, G.; Gatteschi, D.; Hendrickson, D. N.; Sessoli, R. M~RS Bulletin 2000, 25, 66.

(14) Chakov, N. E.; Lee, S. C.; Harter, A. G.; Kuhns, P. L.; Reyes, A. P.; Hill, S. O.; Dalal, N.
S.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. J. Am. Chent. Soc. 2006, 128, 6975.

(15) Sessoli, R.; Gatteschi, D.; Caneschi, A.; Novak, M. A. Nature 1993, 365, 141.

(16) Leuenberger, M. N.; Loss, D. Nature 2001, 410, 789.

(17) Ritter, S. K. In Chens. Eng. News 2004; Vol. 82, p 29.

(18) Chudnovsky, E. M. Science 1996, 274, 938.

(19) Gatteschi, D.; Sessoli, R. Angew. Chenz. hIt. Ed. 2003, 42, 268.

(20) Gatteschi, D.; Sessoli, R.; Cornia, A. Chent. Conanun. 2000, 725.










(21) Christou, G. Polyhedron 2005, 24, 2065.

(22) Aromi, G.; Brechin, E. K. Struct. Bonding 2006, 122, 1.

(23) Benelli, C.; Gatteschi, D. Chens. Rev. 2002, 102, 2369.

(24) Christou, G. Acc. Chens. Res. 1989, 22, 328.

(25) Kurtz, D. M. Chens. Rev. 1990, 90, 585.

(26) Bertini, I.; Gray, H. B.; Lippard, S. J.; Valentine, J. S. Bioinorgan2ic Chentistly;
University Science Books: Mill Valley, CA, 1994.

(27) Manganese Redox Enzymes; Pecoraro, V. L., Ed.; VCH Publishers, Inc., 1992.

(28) Wu, A. J.; Penner-Hahn, J. E.; Pecoraro, V. L. Chens. Rev. 2004, 104, 903.

(29) Mukhopadhyay, S.; Mandal, S. K.; Bhaduri, S.; Armstrong, W. H. Chens. Rev. 2004, 104,
3981.

(30) Gorun, S. M.; Papaefthymiou, G. C.; Frankel, R. B.; Lippard, S. J. J. Am. Chent. Soc.
1987, 109, 3337.

(31) Taft, K. L.; Papaefthymiou, G. C.; Lippard, S. J. Science 1993, 259, 1302.

(32) Brechin, E. K. Chent. Conanun. 2005, 5141.

(33) Yachandra, V. K.; Sauer, K.; Klein, M. P. Chens. Rev. 1996, 96, 2927.

(34) Stamatatos, T. C.; Abboud, K. A.; Wernsdorfer, W.; Christou, G. Angew. Chens. Int. Ed.
2007, 46, 884.

(35) Ako, A. M.; Hewitt, I. J.; Mereacre, V.; Clerac, R.; Wernsdorfer, W.; Anson, C. E.;
Powell, A. K. Angew. Chens. Int. Ed. 2006, 45, 4926.

(36) Vincent, J. B.; Christmas, C.; Chang, H. R.; Li, Q. Y.; Boyd, P. D. W.; Huffman, J. C.;
Hendrickson, D. N.; Christou, G. J. Am. Chent. Soc. 1989, 111, 2086.

(37) Libby, E.; McCusker, J. K.; Schmitt, E. A.; Folting, K.; Hendrickson, D. N.; Christou, G.
Inorg. Chent. 1991, 30, 3486.

(38) Brechin, E. K.; Boskovic, C.; Wernsdorfer, W.; Yoo, J.; Yamaguchi, A.; Sanudo, E. C.;
Concolino, T. R.; Rheingold, A. L.; Ishimoto, H.; Hendrickson, D. N.; Christou, G. J. Am. Chent.
Soc. 2002, 124, 9710.










(39) Bertoncello, K.; Fallon, G. D.; Murray, K. S.; Tiekink, E. R. T. Inorg. Chem. 1991, 30,
3562.

(40) Bonadies, J. A.; Kirk, M. L.; Lah, M. S.; Kessissoglou, D. P.; Hatfield, W. E.; Pecoraro,
V. L. Inorg. Chem. 1989, 28, 2037.

(41) Gelasco, A.; Pecoraro, V. L. J. Am. Chem. Soc. 1993, 115, 7928.

(42) Mikuriya, M.; Yamato, Y.; Tokii, T. Bull. Chem. Soc. Jpn. 1992, 65, 1466.

(43) Bagai, R.; Abboud, K. A.; Christou, G. Dalton Trans. 2006, 3306.

(44) Nishida, Y.; Kida, S. J. Chem. Soc., Dalton Trans. 1986, 2633.

(45) Vincent, J. B.; Chang, H. R.; Folting, K.; Huffman, J. C.; Christou, G.; Hendrickson, D.
N. J. Am. Chem. Soc. 1987, 109, 5703.

(46) SHELXTL6; Bruker-AXS:Madison, WI -2000.

(47) Schake, A. R.; Vincent, J. B.; Li, Q. Y.; Boyd, P. D. W.; Folting, K.; Huffman, J. C.;
Hendrickson, D. N.; Christou, G. Inorg. Chem. 1989, 28, 1915.

(48) Liu, W. T.; Thorp, H. H. Inorg. Chem. 1993, 32, 4102.

(49) Palenik, G. J.1Inorg. Chem. 1997, 36, 4888.

(50) Kambe, K. 1. Phys. Soc. Jpn. 1950, 5, 48.

(51) VanVleck, J. H. The Theory of Electrical and Magnetic Susceptibilities; Oxford Press,
London, 1932.

(52) Aromi, G.; Bhaduri, S.; Artus, P.; Huffman, J. C.; Hendrickson, D. N.; Christou, G.
Polyhedron 2002, 21, 1779.

(53) Davidson, E. R. M4rGNET; Indiana University: Bloomington, IN. 1999.

(54) Yoo, J.; Yamaguchi, A.; Nakano, M.; Krzystek, J.; Streib, W. E.; Brunel, L. C.; Ishimoto,
H.; Christou, G.; Hendrickson, D. N. Inorg. Chem. 2001, 40, 4604.

(55) Hendrickson, D. N.; Christou, G.; Schmitt, E. A.; Libby, E.; Bashkin, J. S.; Wang, S. Y.;
Tsai, H. L.; Vincent, J. B.; Boyd, P. D. W.; Huffman, J. C.; Folting, K.; Li, Q. Y.; Streib, W. E.
J. Am. Chem. Soc. 1992, 114, 2455.

(56) Aliaga-Alcalde, N.; Edwards, R. S.; Hill, S. O.; Wernsdorfer, W.; Folting, K.; Christou,
G. J. Am. Chem. Soc. 2004, 126, 12503.










(57) Brockman, J. T.; Huffman, J. C.; Christou, G. Angew. Chem. Int. Ed. 2002, 41, 2506.

(58) Boudalis, A. K.; Donnadieu, B.; Nastopoulos, V.; Clemente-Juan, J. M.; Mari, A.;
Sanakis, Y.; Tuchagues, J. P.; Perlepes, S. P. Angew. Chem. Int. Ed. 2004, 43, 2266.

(59) Papaefstathiou, G. S.; Perlepes, S. P.; Escuer, A.; Vicente, R.; Font-Bardia, M.; Solans,
X. Angew. Chem. Int. Ed. 2001, 40, 884.

(60) Papaefstathiou, G. S.; Escuer, A.; Vicente, R.; Font-Bardia, M.; Solans, X.; Perlepes, S.
P. Chem. Commun. 2001, 2414.

(61) Stamatatos, T. C.; Foguet-Albiol, D.; Lee, S. C.; Stoumpos, C. C.; Raptopoulou, C. P.;
Terzis, A.; Wemnsdorfer, W.; Hill, S. O.; Perlepes, S. P.; Christou, G. J. Am. Chem. Soc. 2007,
129, 9484.

(62) Murugesu, M.; Habrych, M.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. J. Am.
Chem. Soc. 2004, 126, 4766.

(63) Stamatatos, T. C.; Abboud, K. A.; Wernsdorfer, W.; Christou, G. Angew. Chem. Int. Ed.
2006, 45, 4134.

(64) Brechin, E. K.; Yoo, J.; Nakano, M.; Huffman, J. C.; Hendrickson, D. N.; Christou, G.
Chem. Commun. 1999, 783.

(65) Yoo, J.; Brechin, E. K.; Yamaguchi, A.; Nakano, M.; Huffman, J. C.; Maniero, A. L.;
Brunel, L. C.; Awaga, K.; Ishimoto, H.; Christou, G.; Hendrickson, D. N. Inorg. Chem. 2000, 39,
3615.

(66) Brechin, E. K.; Soler, M.; Davidson, J.; Hendrickson, D. N.; Parsons, S.; Christou, G.
Chem. Commun. 2002, 2252.

(67) Brechin, E. K.; Soler, M.; Christou, G.; Helliwell, M.; Teat, S. J.; Wernsdorfer, W. Chem.
Commun. 2003, 1276.

(68) Vandersluis, P.; Spek, A. L. Acta Crystallogr.; Sect. A 1990, 46, 194.

(69) Sanudo, E. C.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. Inorg. Chem. 2004, 43,
4137.

(70) Boskovic, C.; Wemnsdorfer, W.; Folting, K.; Huffman, J. C.; Hendrickson, D. N.;
Christou, G.1Inorg. Chem. 2002, 41, 5107.

(71) Davidson, E. R. GRID; Indiana University: Bloomington, IN. 1999.

(72) Goodwin, J. C.; Sessoli, R.; Gatteschi, D.; Wemnsdorfer, W.; Powell, A. K.; Heath, S. L.
J. Chem. Soc., Dalton Trans. 2000, 1835.










(73) Gorun, S. M.; Papaefthymiou, G. C.; Frankel, R. B.; Lippard, S. J. J. Am. Chem. Soc.
1987, 109, 4244.

(74) Jones, L. F.; Brechin, E. K.; Collison, D.; Helliwell, M.; Mallah, T.; Piligkos, S.;
Rajaraman, G.; Wemnsdorfer, W.1Inorg. Chem. 2003, 42, 6601.

(75) Powell, A. K.; Heath, S. L.; Gatteschi, D.; Pardi, L.; Sessoli, R.; Spina, G.; Del Giallo, F.;
Pieralli, F. 1. Am. Chem. Soc. 1995, 117, 2491.

(76) Canada-Vilalta, C.; O'Brien, T. A.; Brechin, E. K.; Pink, M.; Davidson, E. R.; Christou,
G. Inorg. Chem. 2004, 43, 5505.

(77) McCusker, J. K.; Vincent, J. B.; Schmitt, E. A.; Mino, M. L.; Shin, K.; Coggin, D. K.;
Hagen, P. M.; Huffman, J. C.; Christou, G.; Hendrickson, D. N. J. Am. Chem. Soc. 1991, 113,
3012.

(78) Kahn, O. Chem. Phys. Lett. 1997, 265, 109.

(79) Barra, A. L.; Caneschi, A.; Cornia, A.; de Biani, F. F.; Gatteschi, D.; Sangregorio, C.;
Sessoli, R.; Sorace, L. J. Am. Chem. Soc. 1999, 121, 5302.

(80) Powell, G. W.; Lancashire, H. N.; Brechin, E. K.; Collison, D.; Health, S. L.; Malluh, T.;
Wernsdorfer, W. Angew. Chem. Int. Ed. 2004, 43, 5772.

(81) Delfs, C.; Gatteschi, D.; Pardi, L.; Sessoli, R.; Wieghardt, K.; Hanke, D.1Inorg. Chem.
1993, 32, 3099.

(82) Sangregorio, C.; Ohm, T.; Paulsen, C.; Sessoli, R.; Gatteschi, D. Phys. Rev. Lett. 1997,
78, 4645.

(83) Caneschi, A.; Comia, A.; Lippard, S. J. Angew. Chem. Int. Ed. 1995, 34, 467.

(84) Murugesu, M.; Abboud, K. A.; Christou, G. Dalton Trans. 2003, 4552.

(85) Jones, L. F.; Batsanov, A.; Brechin, E. K.; Collison, D.; Helliwell, M.; Mallah, T.;
McInnes, E. J. L.; Piligkos, S. Angew. Chem. Int. Ed. 2002, 41, 4318.

(86) Moragues-Canovas, M.; Riviere, P.; Ricard, L.; Paulsen, C.; Wernsdorfer, W.;
Rajaraman, G.; Brechin, E. K.; Mallah, T. Adv. Mater. 2004, 16, 1101.

(87) Foguet-Albiol, D.; Abboud, K. A.; Christou, G. Chem. Commun. 2005, 4282.

(88) Saalfrank, R. W.; Bemnt, I.; Uller, E.; Hampel, F. Angew. Chem. Int. Ed. 1997, 36, 2482.

(89) King, P.; Stamatatos, T. C.; Abboud, K. A.; Christou, G. Angew. Chem. Int. Ed. 2006, 45,
7379.










(90) Ammala, P.; Cashion, J. D.; Kepert, C. M.; Moubaraki, B.; Murray, K. S.; Spiccia, L.;
West, B. O. Angew. Chenz. hIt. Ed. 2000, 39, 1688.

(91) Murugesu, M.; Abboud, K. A.; Christou, G. Polyhedron 2004, 23, 2779.

(92) Taft, K. L.; Lippard, S. J. J. Am. Chent. Soc. 1990, 112, 9629.

(93) Bagai, R.; Datta, S.; Betancur-Rodriguez, A.; Abboud, K. A.; Hill, S.; Christou, G. hzorg.
Chent. 2007, 46, 4535.

(94) Duncan, J. F.; Kanekar, C. R.; Mok, K. F. J. Chent. Soc. A 1969, 480.

(95) Earnshaw, A.; Figgis, B. N.; Lewis, J. J. Chent. Soc. A 1966, 1656.

(96) Bond, A. M.; Clark, R. J. H.; Humphrey, D. G.; Panayiotopoulos, P.; Skelton, B. W.;
White, A. H. J. Chent. Soc., Dalton Trans. 1998, 1845 and references there in.

(97) Armstrong, W. H.; Lippard, S. J. hzorg. Chent. 1985, 24, 981.

(98) Tolis, E. I.; Helliwell, M.; Langley, S.; Raftery, J.; Winpenny, R. E. P. Angew. Chens. Int.
Ed. 2003, 42, 3804.

(99) Yao, H. C.; Li, Y. Z.; Zheng, L. M.; Xin, X. Q. hzorg. China. Acta 2005, 358, 2523.

(100) Datta, S.; Betancur-RodrigueZ, A.; Lee, S. C.; Hill, S.; Foguet-Albiol, D.; Bagai, R.;
Christou, G. Polyhedron 2007, 26, 2243.

(101) Jones, L. F.; Jensen, P.; Moubaraki, B.; Berry, K. J.; Boas, J. F.; Pilbrow, J. R.; Murray,
K. S. J. Mater. Chent. 2006, 16, 2690.

(102) Ako, A. M.; Waldmann, O.; Mereacre, V.; Klower, F.; Hewitt, I. J.; Anson, C. E.; Gudel,
H. U.; Powell, A. K. hzorg. Chent. 2007, 46, 756.

(103) Brown, I. D.; Wu, K. K. Acta Crystallogr., Sect. B: Struct. Sci. 1976, 32, 1957.

(104) Taguchi, T.; Stamatatos, T. C.; Abboud, K. A.; Jones, C. M.; O' Brien, T. A.; Christou,
G. 1norg. Chent. 2007, In Press and references there in.

(105) Nair, V. S.; Hagen, K. S. hnroly Chent. 1992, 31, 4048.

(106) Ammala, P. S.; Batten, S. R.; Cashion, J. D.; Kepert, C. M.; Moubaraki, B.; Murray, K.
S.; Spiccia, L.; West, B. O. hnroly China. Acta 2002, 331, 90.

(107) Cotton, F. A.; Wilkinson, G. Advancedlnorgan2ic Chentistly; Wiley: New York, 1980.

(108) Gorun, S. M.; Lippard, S. J. J. Am. Chent. Soc. 1985, 107, 4568.










(109) Wemple, M. W.; Coggin, D. K.; Vincent, J. B.; McCusker, J. K.; Streib, W. E.; Huffman,
J. C.; Hendrickson, D. N.; Christou, G. J. Chent. Soc., Dalton Trans. 1998, 719.

(110) Boudalis, A. K.; Tangoulis, V.; Raptopoulou, C. P.; Terzis, A.; Tuchagues, J. P.;
Perlepes, S. P. hzorg. China. Acta 2004, 357, 1345.

(111) Boudalis, A. K.; Lalioti, N.; Spyroulias, G. A.; Raptopoulou, C. P.; Terzis, A.;
Bousseksou, A.; Tangoulis, V.; Tuchagues, J. P.; Perlepes, S. P. hzorg. Chent. 2002, 41, 6474.

(112) Gorun, S. M.; Lippard, S. J. hzorg. Chent. 1991, 30, 1625.

(113) Dziobkowski, C. T.; Wrobleski, J. T.; Brown, D. B. hzorg. Chent. 1981, 20, 671.

(114) Armstrong, W. H.; Lippard, S. J. J. Am. Chent. Soc. 1983, 105, 4837.

(115) Armstrong, W. H.; Spool, A.; Papaefthymiou, G. C.; Frankel, R. B.; Lippard, S. J. J. Am.
Chent. Soc. 1984, 106, 3653.

(116) Armstrong, W. H.; Lippard, S. J. J. Am. Chent. Soc. 1985, 107, 3730.

(117) Lee, S. C.; Stamatatos, T. C.; Hill, S.; Perlepes, S. P.; Christou, G. Polyhedron 2007, 26,
2225.

(118) Oshio, H.; Hoshino, N.; Ito, T.; Nakano, M. J. Am. Chent. Soc. 2004, 126, 8805.

(119) Benelli, C.; Parsons, S.; Solan, G. A.; Winpenny, R. E. P. Angew. Chenz. hIt. Ed. 1996,
35, 1825.

(120) Barra, A. L.; Gatteschi, D.; Sessoli, R. Chens. Eur. J. 2000, 6, 1608.

(121) Bacsa, J.; Zhao, H.; Dunbar, K. R. Acta Crystallogr. Sect. C 2003, C59, m561.

(122) Canada-Vilalta, C.; O'Brien Ted, A.; Pink, M.; Davidson Ernest, R.; Christou, G. Inorg.
Chent. 2003, 42, 7819.

(123) Yan, B.; Chen, Z. D. hzorg. Chent. Conanun. 2001, 4, 138.

(124) Seddon, E. J.; Huffman, J. C.; Christou, G. Dalton 2000, 4446.

(125) Brechin, E. K.; Knapp, M. J.; Huffman, J. C.; Hendrickson, D. N.; Christou, G. Inorg.
China. Acta 2000, 297, 389.

(126) Christmas, C. A.; Tsai, H. L.; Pardi, L.; Kesselman, J. M.; Gantzel, P. K.; Chadha, R. K.;
Gatteschi, D.; Harvey, D. F.; Hendrickson, D. N. J. Am. Chent. Soc. 1993, 115, 12483.

(127) Bagai, R.; Abboud, K. A.; Christou, G. hnroly Chent. 2007, 46, 5567.










(128) Demnitz, F. W. J.; D'Heni, M. B. Org. Prep. Proced. Int. 1998, 30, 467.

(129) Heirtzler, F. R. Synlett 1999, 1203.

(130) Norrby, T.; Borje, A.; Zhang, L.; Akermark, B. Acta Chem. Scand. 1998, 52, 77.

(131) Brown, I. D.; Altermatt, D. Acta Crystallogr., Sect. B: Struct. Sci. 1985, B41, 244.

(132) Tabernor, J.; Jones, L. F.; Heath, S. L.; Muryn, C.; Aromi, G.; Ribas, J.; Brechin, E. K.;
Collison, D. Dalton Trans. 2004, 975.

(133) Boskovic, C.; Sieber, A.; Chaboussant, G.; Guedel, H. U.; Ensling, J.; Wernsdorfer, W.;
Neels, A.; Labat, G.; Stoeckli-Evans, H.; Janssen, S. Inorg. Chem. 2004, 43, 5053.

(134) Boskovic, C.; Labat, G.; Neels, A.; Gudel, H. U. Dalton Trans. 2003, 3671.

(135) Lachicotte, R. J.; Hagen, K. S. Inorg. Chim. Acta 1997, 263, 407.

(136) Reynolds, R. A.; Coucouvanis, D. Inorg. Chem. 1998, 37, 170.

(137) O'Keefe, B. J.; Monnier, S. M.; Hillmyer, M. A.; Tolman, W. B. J. Am. Chem. Soc. 2001,
123, 339.

(138) Mikuriya, M.; Nakadera, K. Chem. Lett. 1995, 213.

(139) Mikuriya, M.; Hashimoto, Y.; Nakashima, S. Chem. Commun. 1996, 295.

(140) Herold, S.; Lippard, S. J. Inorg. Chem. 1997, 36, 50.

(141) Krishnamurthy, D.; Sarjeant, A. N.; Goldberg, D. P.; Caneschi, A.; Totti, F.; Zakharov,
L. N.; Rheingold, A. L. Chem. Eur. J. 2005, 11, 7328.

(142) Overgaard, J.; Hibbs, D. E.; Rentschler, E.; Timco, G. A.; Larsen, F. K.1Inorg. Chem.
2003, 42, 7593.

(143) Chaudhuri, P.; Rentschler, E.; Birkelbach, F.; Krebs, C.; Bill, E.; Weyhermuller, T.;
Florke, U. Eur. J. Inorg. Chem. 2003, 541.

(144) Taft, K. L.; Papaefthymiou, G. C.; Lippard, S. J.1Inorg. Chem. 1994, 33, 1510.

(145) Bagai, R.; Abboud, K. A.; Christou, G. Chem. Commun. 2007, 3359.

(146) Bagai, R.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. J. Am. Chem. Soc. 2007, 129,
12918.










(147) Affronte, M.; Carretta, S.; Timco, G. A.; Winpenny, R. E. P. Chent. Conanun. 2007,
1789.

(148) Watton, S. P.; Fuhrmann, P.; Pence, L. E.; Caneschi, A.; Cornia, A.; Abbati, G. L.;
Lippard, S. J. Angew. Chens. Int. Ed. 1997, 36, 2774.

(149) Schmitt, W.; Anson, C. E.; Wernsdorfer, W.; Powell, A. K. Chent. Conanun. 2005, 2098.

(150) Jones, L. F.; Low, D. M.; Helliwell, M.; Raftery, J.; Collison, D.; Aromi, G.; Cano, J.;
Mallah, T.; Wemnsdorfer, W.; Brechin, E. K.; McInnes, E. J. L. Polyhedron 2006, 25, 325.

(151) Yao, H. C.; Wang, J. J.; Ma, Y. S.; Waldmann, O.; Du, W. X.; Song, Y.; Li, Y. Z.;
Zheng, L. M.; Decurtins, S.; Xin, X. Q. Chent. Conanun. 2006, 1745.

(152) Bino, A.; Shweky, I.; Cohen, S.; Bauminger, E. R.; Lippard, S. J. hzorg. Chent. 1998, 37,
5168.

(153) Wemnsdorfer, W. Adv. Chens. Phys. 2001, 118, 99.

(154) Wemnsdorfer, W.; Aliaga-Alcalde, N.; Hendrickson, D. N.; Christou, G. Nature 2002,
416, 406.

(155) Tiron, R.; Wemnsdorfer, W.; Aliaga-Alcalde, N.; Christou, G. Phys. Rev. B 2003, 68,
140407.

(156) Hill, S.; Edwards, R. S.; Aliaga-Alcalde, N.; Christou, G. Science 2003, 302, 1015.

(157) Tiron, R.; Wemnsdorfer, W.; Foguet-Albiol, D.; Aliaga-Alcalde, N.; Christou, G. Phys.
Rev. Lett. 2003, 91, 227203.

(158) Sun, Z. M.; Grant, C. M.; Castro, S. L.; Hendrickson, D. N.; Christou, G. Chent.
Conanun. 1998, 721.

(159) Yang, E. C.; Hendrickson, D. N.; Wernsdorfer, W.; Nakano, M.; Zakharov, L. N.;
Sommer, R. D.; Rheingold, A. L.; Ledezma-Gairaud, M.; Christou, G. J. Appl. Phys. 2002, 91,
7382.

(160) Andres, H.; Basler, R.; Blake, A. J.; Cadiou, C.; Chaboussant, G.; Grant, C. M.; Gudel,
H. U.; Murrie, M.; Parsons, S.; Paulsen, C.; Semadini, F.; Villar, V.; Wernsdorfer, W.;
Winpenny, R. E. P. Chens. Eur. J. 2002, 8, 4867.

(161) Aubin, S. M. J.; Wemple, M. W.; Adams, D. M.; Tsai, H. L.; Christou, G.; Hendrickson,
D. N. J. Am. Chent. Soc. 1996, 118, 7746.

(162) Dendrinou-Samara, C.; Alexiou, M.; Zaleski, C. M.; Kampf, J. W.; Kirk, M. L.;
Kessissoglou, D. P.; Pecoraro, V. L. Angew. Chens. Int. Ed. 2003, 42, 3763.










(163) Milios, C. J.; Raptopoulou, C. P.; Terzis, A.; Lloret, F.; Vicente, R.; Perlepes, S. P.;
Escuer, A. Angew. Chem. Int. Ed. 2004, 43, 210.

(164) Miyasaka, H.; Clerac, R.; Wernsdorfer, W.; Lecren, L.; Bonhomme, C.; Sugiura, K.;
Yamashita, M. Angew. Chem. Int. Ed. 2004, 43, 2801.

(165) Hundal, G.; Hundal, M. S.; Obrai, S.; Poonia, N. S.; Kumar, S. Inorg. Chem. 2002, 41,
2077.

(166) Plass, W. Eur. J. Inorg. Chem. 1998, 799.

(167) de Sousa, A. S.; Fernandes, M. A. Polyhedron 2002, 21, 1883.

(168) Bagai, R.; Abboud, K. A.; Christou, G. Inorg. Chem. 2007, In Press.

(169) Bagai, R.; Daniels, M. R.; Abboud, K. A.; Christou, G. Inorg. Chem., 2008, In Press.

(170) Wu, R. W.; Poyraz, M.; Sowrey, F. E.; Anson, C. E.; Wocadlo, S.; Powell, A. K.;
Jayasooriya, U. A.; Cannon, R. D.; Nakamoto, T.; Katada, M.; Sano, H. Inorg. Chem. 1998, 37,
1913.

(171) Milios, C. J.; Vinslava, A.; Wernsdorfer, W.; Prescimone, A.; Wood, P. A.; Parsons, S.;
Perlepes, S. P.; Christou, G.; Brechin, E. K. J. Am. Chem. Soc. 2007, 129, 6547.

(172) Tasiopoulos, A. J.; Vinslava, A.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. Angew.
Chem. Int. Ed. 2004, 43, 2117.

(173) Maheswaran, S.; Chastanet, G.; Teat, S. J.; Mallah, T.; Sessoli, R.; Wernsdorfer, W.;
Winpenny, R. E. P. Angew. Chem. Int. Ed. 2005, 44, 5044.

(174) Godbole, M. D.; Roubeau, O.; Clerac, R.; Kooijman, H.; Spek, A. L.; Bouwman, E.
Chem. Commun. 2005, 3715.

(175) Rajaraman, G.; Murugesu, M.; Sanudo, E. C.; Soler, M.; Wernsdorfer, W.; Helliwell, M.;
Muryn, C.; Raftery, J.; Teat, S. J.; Christou, G.; Brechin, E. K. J. Am. Chem. Soc. 2004, 126,
15445.

(176) Murugesu, M.; Wernsdorfer, F.; Abboud, K. A.; Christou, G. Angew. Chem. Int. Ed.
2005, 44, 892.

(177) Brechin, E. K.; Christou, G.; Soler, M.; Helliwell, M.; Teat, S. J. Dalton Trans. 2003,
513.

(178) Jones, L. F.; Brechin, E. K.; Collison, D.; Raftery, J.; Teat, S. J. Inorg. Chem. 2003, 42,
6971.










(179) Tsai, H. L.; Wang, S. Y.; Folting, K.; Streib, W. E.; Hendrickson, D. N.; Christou, G. J.
Am. Chent. Soc. 1995, 117, 2503.

(180) Tanase, S.; Aromi, G.; Bouwman, E.; Kooijman, H.; Spek, A. L.; Reedijk, J. Chent.
Conanun. 2005, 3147.

(181) Milios, C. J.; Kefalloniti, E.; Raptopoulou, C. P.; Terzis, A.; Vicente, R.; Lalioti, N.;
Escuer, A.; Perlepes, S. P. Chent. Conanun. 2003, 819.

(182) Alvareza, C. S.; Bond, A. D.; Cave, D.; Mosquera, M. E. G.; Harron, E. A.; Layfield, R.
A.; McPartlin, M.; Rawson, J. M.; Wood, P. T.; Wright, D. S. Chent. Conanun. 2002, 2980.

(183) Boskovic, C.; Huffman, J. C.; Christou, G. Chent. Conanun. 2002, 2502.

(184) Saalfrank, R. W.; Low, N.; Demleitner, B.; Stalke, D.; Teichert, M. Chens. Eur. J. 1998,
4, 1305.

(185) Wemple, M. W.; Tsai, H. L.; Wang, S. Y.; Claude, J. P.; Streib, W. E.; Huffman, J. C.;
Hendrickson, D. N.; Christou, G. hnroly Chent. 1996, 35, 6437.

(186) Tasiopoulos, A. J.; Abboud, K. A.; Christou, G. Chent. Conanun. 2003, 580.

(187) Milios, C. J.; Fabbiani, F. P. A.; Parsons, S.; Murugesu, M.; Christou, G.; Brechin, E. K.
Dalton Trans. 2006, 351.

(188) Saalfrank, R. W.; Low, N.; Trummer, S.; Sheldrick, G. M.; Teichert, M.; Stalke, D. Eur.
J. Inorg. Chent. 1998, 559.

(189) Murugesu, M.; Wernsdorfer, W.; Abboud, K. A.; Brechin, E. K.; Christou, G. Dalton
Trans. 2006, 2285.

(190) Li, Y. G.; Wernsdorfer, W.; Clerac, R.; Hewitt, I. J.; Anson, C. E.; Powell, A. K. hzorg.
Chent. 2006, 45, 2376.

(191) Masello, A.; Murugesu, M.; Abboud, K. A.; Christou, G. Polyhedron 2007, 26, 2276.

(192) Lampropoulos, C.; Murugesu, M.; Abboud, K. A.; Christou, G. Polyhedron 2007, 26,
2129.

(193) Sun, Z. M.; Gantzel, P. K.; Hendrickson, D. N. hzorg. Chent. 1996, 35, 6640.

(194) Caneschi, A.; Ferraro, F.; Gatteschi, D.; Melandri, M. C.; Rey, P.; Sessoli, R. Angew.
Chenz. hIt. Ed. 1989, 28, 1365.

(195) Yu, S. B.; Lippard, S. J.; Shweky, I.; Bino, A. hzorg. Chent. 1992, 31, 3502.










(196) Ye, B. H.; Mak, T.; Williams, I. D.; Li, X. Y. Chent. Conanun. 1997, 1813.

(197) Coucouvanis, D.; Reynolds, R. A.; Dunham, W. R. J. Am. Chent. Soc. 1995, 117, 7570.

(198) Cheng, B. S.; Fries, P. H.; Marchon, J. C.; Scheidt, W. R. hzorg. Chent. 1996, 35, 1024.

(199) Komatsuzaki, H.; Ichikawa, S.; Hikichi, S.; Akita, M.; Moro-oka, Y. hzorg. Chent. 1998,
37, 3652.

(200) Biswas, S.; Mitra, K.; Adhikary, B.; Lucas, C. R. Transition M~et. Chent. 2005, 30, 586.

(201) Zhou, A. J.; Qin, L. J.; Beedle, C. C.; Ding, S.; Nakano, M.; Leng, J. D.; Tong, M. L.;
Hendrickson, D. N. hzorg. Chent. 2007, 46, 8111.

(202) Eppley, H. J.; Tsai, H. L.; Devries, N.; Folting, K.; Christou, G.; Hendrickson, D. N. J.
Am. Chent. Soc. 1995, 117, 301.

(203) Soler, M.; Wernsdorfer, W.; Abboud, K. A.; Huffman, J. C.; Davidson, E. R.;
Hendrickson, D. N.; Christou, G. J. Am. Chent. Soc. 2003, 125, 3576.

(204) Bian, G. Q.; Kuroda-Sowa, T.; Konaka, H.; Hatano, M.; Maekawa, M.; Munakata, M.;
Miyasaka, H.; Yamashita, M. hzorg. Chent. 2004, 43, 4790.

(205) Zhao, H. H.; Berlinguette, C. P.; Bacsa, J.; Prosvirin, A. V.; Bera, J. K.; Tichy, S. E.;
Schelter, E. J.; Dunbar, K. R. hzorg. Chent. 2004, 43, 1359.

(206) Coronado, E.; Torment-Aliaga, A.; Gaita-Arino, A.; Giminez-Saiz, C.; Romero, F. M.;
Wernsdorfer, W. Angew. Chenz. hIt. Ed. 2004, 43, 6152.

(207) Bagai, R.; Christou, G. hnroly Chent. 2007, 46, 10810.

(208) Tasiopoulos, A. J.; Wernsdorfer, W.; Abboud, K. A.; Christou, G.1Inorg. Chent. 2005, 44,
6324.

(209) King, P.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. Inorg. Chent. 2005, 44, 8659.

(210) Foguet-Albiol, D.; O'Brien, T. A.; Wernsdorfer, W.; Moulton, B.; Zaworotko, M. J.;
Abboud, K. A.; Christou, G. Angew. Chenz. hIt. Ed. 2005, 44, 897.

(211) Boskovic, C.; Brechin, E. K.; Streib, W. E.; Folting, K.; Hendrickson, D. N.; Christou, G.
Chent. Conanun. 2001, 467.

(212) Boskovic, C.; Brechin, E. K.; Streib, W. E.; Folting, K.; Bollinger, J. C.; Hendrickson, D.
N.; Christou, G. J. Am. Chent. Soc. 2002, 124, 3725.










(213) Rumberger, E. M.; Shah, S. J.; Beedle, C. C.; Zakharov, L. N.; Rheingold, A. L.;
Hendrickson, D. N. hzorg. Chent. 2005, 44, 2742.

(214) Yao, H. C.; Li, Y. Z.; Song, Y.; Ma, Y. S.; Zheng, L. M.; Xin, X. Q. hzorg. Chent. 2006,
45, 59.

(215) Dendrinou-Samara, C.; Zaleski, C. M.; Evagorou, A.; Kampf, J. W.; Pecoraro, V. L.;
Kessissoglou, D. P. Chent. Conanun. 2003, 2668.

(216) Cavaluzzo, M.; Chen, Q.; Zubieta, J. Chent. Conanun. 1993, 131.

(217) Koo, B. K.; Lee, U. Bull. Korean Chent. Soc. 2001, 22, 103.

(218) Brechin, E. K.; Clegg, W.; Murrie, M.; Parsons, S.; Teat, S. J.; Winpenny, R. E. P. J. Am.
Chent. Soc. 1998, 120, 7365.

(219) Abu-Nawwas, A. A. H.; Cano, J.; Christian, P.; Mallah, T.; Rajaraman, G.; Teat, S. J.;
Winpenny, R. E. P.; Yukawa, Y. Chent. Conanun. 2004, 314.

(220) Sellmann, D.; Geipel, F.; Heinemann, F. W. Chens. Eur. J. 2002, 8, 958.

(221) Caneschi, A.; Cornia, A.; Fabretti, A. C.; Gatteschi, D. Angew. Chens. Int. Ed. 1999, 38,
1295.

(222) Boskovic, C.; Gudel, H. U.; Labat, G.; Neels, A.; Wernsdorfer, W.; Moubaraki, B.;
Murray, K. S. hnroly Chent. 2005, 44, 3181.

(223) Soler, M.; Wemnsdorfer, W.; Folting, K.; Pink, M.; Christou, G. J. Am. Chent. Soc. 2004,
126, 2156.

(224) Brechin, E. K.; Sanudo, E. C.; Wemnsdorfer, W.; Boskovic, C.; Yoo, J.; Hendrickson, D.
N.; Yamaguchi, A.; Ishimoto, H.; Concolino, T. E.; Rheingold, A. L.; Christou, G. hnroly Chent.
2005, 44, 502.

(225) Weihe, H.; Gudel, H. U. J. Am. Chent. Soc. 1997, 119, 6539.

(226) Wemner, R.; Ostrovsky, S.; Griesar, K.; Haase, W. hzorg. China. Acta 2001, 326, 78.

(227) Bogani, L.; Wernsdorfer, W. Nature Mat. 2008, 7, 179.

(228) Oshio, H.; Nihei, M.; Yoshida, A.; Nojiri, H.; Nakano, M.; Yamaguchi, A.; Karaki, Y.;
Ishimoto, H. Chens. Eur. J. 2005, 11, 843.

(229) Li, Y. G.; Wernsdorfer, W.; Clerac, R.; Hewitt, I. J.; Anson, C. E.; Powell, A. K. hzorg.
Chent. 2006, 45, 2376-2378.










(230) Maheswaran, S.; Chastanet, G.; Teat, S. J.; Mallah, T.; Sessoli, R.; Wernsdorfer, W.;
Winpenny, R. E. P. Angew. Chem. Int. Ed. 2005, 44, 5044-5048.

(231) Beltran, L. M. C.; Long, J. R. Acc. Chem. Res. 2005, 38, 325.

(232) Oshio, H.; Nakano, M. Chem. Eur. J. 2005, 11, 5178.

(233) Hamamatsu, T.; Yabe, K.; Towatari, M.; Osa, S.; Matsumoto, N.; Re, N.; Pochaba, A.;
Mrozinski, J.; Gallani, J. L.; Barla, A.; Imperia, P.; Paulsen, C.; Kappler, J. P. Inorg. Chem.
2007, 46, 4458-4468.

(234) Zaleski, C. M.; Kampf, J. W.; Mallah, T.; Kirk, M. L.; Pecoraro, V. L. Inorg. Chem.
2007, 46, 1954-1956.

(235) Li, D. F.; Clerac, R.; Wang, G. B.; Yee, G. T.; Holmes, S. M. Eur. J. Inorg. Chem. 2007,
1341.

(236) Pointillart, F.; Bernot, K.; Sessoli, R.; Gatteschi, D. Chem. Eur. J. 2007, 13, 1602.

(237) Mori, F.; Nyui, T.; Ishida, T.; Nogami, T.; Choi, K. Y.; Nojiri, H. J. Am. Chem. Soc.
2006, 128, 1440-1441.

(238) Costes, J. P.; Auchel, M.; Dahan, F.; Peyrou, V.; Shova, S.; Wernsdorfer, W.1Inorg.
Chem. 2006, 45, 1924-1934.

(239) Mishra, A.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. J. Am. Chem. Soc. 2004, 126,
15648.

(240) Mishra, A.; Wernsdorfer, W.; Parsons, S.; Christou, G.; Brechin, E. K. Chem. Commun.
2005, 2086.

(241) Zaleski, C. M.; Depperman, E. C.; Kampf, J. W.; Kirk, M. L.; Pecoraro, V. L. Angew.
Chem. Int. Ed. 2004, 43, 3912.

(242) Ma, B. Q.; Gao, S.; Su, G.; Xu, G. X. Angew. Chem. Int. Ed. 2001, 40, 434.

(243) Ferbinteanu, M.; Kajiwara, T.; Choi, K. Y.; Nojiri, H.; Nakamoto, A.; Kojima, N.;
Cimpoesu, F.; Fujimura, Y.; Takaishi, S.; Yamashita, M. J. Am. Chem. Soc. 2006, 128, 9008.

(244) Zaleski, C. M.; Depperman, E. C.; Kampf, J. W.; Kirk, M. L.; Pecoraro, V. L. Inorg.
Chem. 2006, 45, 10022.

(245) Murugesu, M.; Mishra, A.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. Polyhedron
2006, 25, 613.










(246) Osa, S.; Kido, T.; Matsumoto, N.; Re, N.; Pochaba, A.; Mrozinski, J. J. Am. Chem. Soc.
2004, 126, 420-421.

(247) Mereacre, V. M.; Ako, A. M.; Clerac, R.; Wernsdorfer, W.; Filoti, G.; Bartolome, J.;
Anson, C. E.; Powell, A. K. J. Am. Chem. Soc. 2007, 129, 9248.

(248) Kahn, M. L.; Lecante, P.; Verelst, M.; Mathoniere, C.; Kahn, O. Chem. Mater. 2000, 12,
3073.

(249) Kahn, M. L.; Mathoniere, C.; Kahn, O.1Inorg. Chem. 1999, 38, 3692.

(250) Costes, J. P.; Dahan, F.; Dupuis, A.; Laurent, J. P. Chem. Eur. J. 1998, 4, 1616.

(251) Figuerola, A.; Diaz, C.; Ribas, J.; Tangoulis, V.; Granell, J.; Lloret, F.; Mahia, J.;
Maestro, M. Inorg. Chem. 2003, 42, 641.

(252) Zhao, H. H.; Lopez, N.; Prosvirin, A.; Chifotides, H. T.; Dunbar, K. R. Dalton Trans.
2007, 878.

(253) Chen, W. T.; Guo, G. C.; Wang, M. S.; Xu, G.; Cai, L. Z.; Akitsu, T.; Akita-Tanaka, M.;
Matsushita, A.; Huang, J. S. Inorg. Chem. 2007, 46, 2105.

(254) Estrader, M.; Ribas, J.; Tangoulis, V.; Solans, X.; Font-Bardia, M.; Maestro, M.; Diaz, C.
Inorg. Chem. 2006, 45, 8239.

(255) Figuerola, A.; Ribas, J.; Llunell, M.; Casanova, D.; Maestro, M.; Alvarez, S.; Diaz, C.
Inorg. Chem. 2005, 44, 6939.

(256) Figuerola, A.; Ribas, J.; Casanova, D.; Maestro, M.; Alvarez, S.; Diaz, C. Inorg. Chem.
2005, 44, 6949.

(257) Figuerola, A.; Diaz, C.; Ribas, J.; Tangoulis, V.; Sangregorio, C.; Gatteschi, D.; Maestro,
M.; Mahia, J. Inorg. Chem. 2003, 42, 5274.

(258) Sheldrick, G. M. SHELXL-97; University of Gottingen, Germany 1997.

(259) Sanudo, E. C.; Brechin, E. K.; Boskovic, C.; Wernsdorfer, W.; Yoo, J.; Yamaguchi, A.;
Concolino, T. R.; Abboud, K. A.; Rheingold, A. L.; Ishimoto, H.; Hendrickson, D. N.; Christou,
G. Polyhedron 2003, 22, 2267.

(260) Wittick, L. M.; Murray, K. S.; Moubaraki, B.; Batten, S. R.; Spiccia, L.; Berry, K. J.
Dalton Trans. 2004, 1003.

(261) Ako, A. M.; Mereacre, V.; Hewitt, I. J.; Clerac, R.; Lecren, L.; Anson, C. E.; Powell, A.
K. J. Mater. Chem. 2006, 16, 2579.










(262) Shanmugam, M.; Chastanet, G.; Mallah, T.; Sessoli, R.; Teat, S. J.; Timco, G. A.;
Winpenny, R. E. P. Chem. Eur. J. 2006, 12, 8777.

(263) Mishra, A.; Tasiopoulos, A. J.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. Inorg.
Chem. 2007, 46, 3105-3115.

(264) Tasiopoulos, A. J.; Mishra, A.; Christou, G. Polyhedron 2007, 26, 2183-2188.

(265) Tasiopoulos, A. J.; O'Brien, T. A.; Abboud, K. A.; Christou, G. Angew. Chem. Int. Ed.
2004, 43, 345.

(266) Eppley, H. J.; Aubin, S. M. J.; Streib, W. E.; Bollinger, J. C.; Hendrickson, D. N.;
Christou, G. Inorg. Chem. 1997, 36, 109.

(267) Bagai, R.; Abboud, K. A.; Christou, G.1Inorg. Chem. 2008, 47, 621-631.

(268) Bencini, A.; Benelli, C.; Caneschi, A.; Carlin, R. L.; Dei, A.; Gatteschi, D. J. Am. Chem.
Soc. 1985, 107, 8128.

(269) Bencini, A.; Benelli, C.; Caneschi, A.; Dei, A.; Gatteschi, D. Inorg. Chem. 1986, 25, 572.

(270) Kido, T.; Ikuta, Y.; Sunatsuki, Y.; Ogawa, Y.; Matsumoto, N. Inorg. Chem. 2003, 42,
398-408.

(271) Hamamatsu, T.; Yabe, K.; Towatari, M.; Osa, S.; Matsumoto, N.; Re, N.; Pochaba, A.;
Mrozinski, J.; Gallani, J. L.; Barla, A.; Imperia, P.; Paulsen, C.; Kappler, J. P. Inorg. Chem.
2007, 46, 4458.

(272) Chakov, N. E.; Soler, M.; Wernsdorfer, W.; Abboud, K. A.; Christou, G.1Inorg. Chem.
2005, 44, 5304.

(273) Sessoli, R.; Tsai, H. L.; Schake, A. R.; Wang, S. Y.; Vincent, J. B.; Folting, K.;
Gatteschi, D.; Christou, G.; Hendrickson, D. N. J. Am. Chem. Soc. 1993, 115, 1804.

(274) Caneschi, A.; Gatteschi, D.; Sessoli, R.; Barra, A. L.; Brunel, L. C.; Guillot, M. J. Am.
Chem. Soc. 1991, 113, 5873.

(275) Rebilly, J. N.; Mallah, T. Struct. Bonding 2006, 122, 103.

(276) Berlinguette, C. P.; Vaughn, D.; Canada-Vilalta, C.; Galan-Mascaros, J. R.; Dunbar, K.
R. Angew. Chem. Int. Ed. 2003, 42, 1523.

(277) Benelli, C.; Cano, J.; Journaux, Y.; Sessoli, R.; Solan, G. A.; Winpenny, R. E. P.1Inorg.
Chem. 2001, 40, 188.

(278) Price, D. J.; Batten, S. R.; Moubaraki, B.; Murray, K. S. Chem. Commun. 2002, 762.










(279) Koizumi, S.; Nihei, M.; Nakano, M.; Oshio, H. hzorg. Chent. 2005, 44, 1208.

(280) Godbole, M. D.; Roubeau, O.; Mills, A. M.; Kooijman, H.; Spek, A. L.; Bouwman, E.
hzorg. Chent. 2006, 45, 6713.

(281) Ishikawa, N.; Sugita, M.; Wernsdorfer, W. Angew. Chenz. hIt. Ed. 2005, 44, 2931.

(282) Costes, J. P.; Auchel, M.; Dahan, F.; Peyrou, V.; Shova, S.; Wernsdorfer, W. hzorg.
Chent. 2006, 45, 1924.

(283) Stamatatos, T. C.; Foguet-Albiol, D.; Stoumpos, C. C.; Raptopoulou, C. P.; Terzis, A.;
Wernsdorfer, W.; Perlepes, S. P.; Christou, G. J. Am. Chent. Soc. 2005, 127, 15380.

(284) Friedman, J. R.; Sarachik, M. P.; Tejada, J.; Ziolo, R. Phys. Rev. Lett. 1996, 76, 3830.

(285) Hill, S.; Anderson, N.; Wilson, A.; Takahashi, S.; Chakov, N. E.; Murugesu, M.; North,
J. M.; Dalal, N. S.; Christou, G. J. Appl. Phys. 2005, 97, 10M510.

(286) Boskovic, C.; Pink, M.; Huffman, J. C.; Hendrickson, D. N.; Christou, G. J. Am. Chent.
Soc. 2001, 123, 9914.

(287) Artus, P.; Boskovic, C.; Yoo, J.; Streib, W. E.; Brunel, L. C.; Hendrickson, D. N.;
Christou, G. hnroly Chent. 2001, 40, 4199.

(288) Soler, M.; Artus, P.; Folting, K.; Huffman, J. C.; Hendrickson, D. N.; Christou, G. hnroly
Chent. 2001, 40, 4902.

(289) Kuroda-Sowa, T.; Lam, M.; Rheingold, A. L.; Frommen, C.; Reiff, W. M.; Nakano, M.;
Yoo, J.; Maniero, A. L.; Brunel, L. C.; Christou, G.; Hendrickson, D. N. hzorg. Chent. 2001, 40,
6469.

(290) Kuroda-Sowa, T.; Nogami, T.; Konaka, H.; Maekawa, M.; Munakata, M.; Miyasaka, H.;
Yamashita, M. Polyhedron 2003, 22, 1795.

(291) Aubin, S. M. J.; Sun, Z. M.; Pardi, L.; Krzystek, J.; Folting, K.; Brunel, L. C.; Rheingold,
A. L.; Christou, G.; Hendrickson, D. N. hzorg. Chent. 1999, 38, 5329.

(292) Chakov, N. E.; Zakharov, L. N.; Rheingold, A. L.; Abboud, K. A.; Christou, G. hnroly
Chent. 2005, 44, 4555.

(293) Connelly, N. G.; Geiger, W. E. Chens. Rev. 1996, 96, 877.

(294) Sun, Z. M.; Ruiz, D.; Dilley, N. R.; Soler, M.; Ribas, J.; Folting, K.; Maple, M. B.;
Christou, G.; Hendrickson, D. N. Chent. Conanun. 1999, 1973.










(295) Aubin, S. M. J.; Sun, Z. M.; Eppley, H. J.; Rumberger, E. M.; Guzei, I. A.; Folting, K.;
Gantzel, P. K.; Rheingold, A. L.; Christou, G.; Hendrickson, D. N. Inorg. Chem. 2001, 40, 2127.

(296) Takahashi, S.; Hill, S. Rev. Sci. Instrum. 2005, 76, 023114.









BIOGRAPHICAL SKETCH

Rashmi Bagai was born in Delhi, India in 1979. She entered Hindu College, Delhi

University in 1995 and received Bachelor of Science degree in 1998. She then successfully

qualified in the j oint entrance exam of the Indian Institute of Technology, Delhi to pursue her

masters in chemistry. During her masters, she worked in the research group of Prof. Jai Deo

Singh on the preparation of tripodal ligands incorporating three aldehydic moieties

interconnected by a rigid triazine spacer. After the completion of her Master of Science degree in

2000, she was awarded the prestigious Council of Scientific and Industrial Research Junior

Research Fellowship (CSIR-JRF) on the basis of a written competitive test conducted j ointly by

CSIR and UGC (University Grants Commission) at the national level. Thereafter, she j oined the

research group of Prof. Debkumar Bandyopadhyay at IIT, Delhi in July, 2000 as a CSIR-JRF

research assistant. The research involved extensive kinetic studies and preparation of Iron(III)

porphyrin based catalytic systems for the oxidation of various organic and organometallic

compounds. She got married to Ranj an Mitra, her friend for nine years, in May 2003 in Delhi,

India and moved to USA. In August 2003, she j oined the research group of Prof. George

Christou in the Department of Chemistry at the University of Florida. Her doctoral research

primarily involves exploration of new ligands for the preparation and magnetic characterization

of polynuclear oxo-bridged Mn and Fe clusters, some of which behave as single-molecule

magnets .





PAGE 1

1 DIVERSITY OF STRUCTURAL TYPES AND MOLECULAR NANOMAGNETISM IN IRON AND MANGANESE CLUSTERS By RASHMI BAGAI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Rashmi Bagai

PAGE 3

3 To mummi, papa, and Shammi, for their infinite unconditional love

PAGE 4

4 ACKNOWLEDGMENTS Undertaking and com pleting this dissertation ha s been an extraordinary journey for me. It is my pleasure to thank the many people who ha ve encouraged and supported me and made the completion of this work possible. First and foremost, I would like to thank my mentor and research a dvisor, Prof. George Christou, for his guidance and help over the last 5 years. His cons tant encouragement, insightful advice and confidence in my abilities as a chemist ha ve been invaluable in attaining my goal. His impact on my scientific writing skills is enormous I deeply appreciate the freedom that I enjoyed while working under him, nicely balanced by much-needed guidance and support. Most importantly I would like to thank him for always having the time to sit and talk about research. He was always open to my ideas, yet quick to point out weaknesses and incompleteness in my arguments. Without his support, this dissertation would have been impossible. I walk away from his lab infinitely more knowledgeable about crystal growth, magnetochemistry, and a lifelong experience. I wear the title of C hristou group Alumnus with pride. I would also like to thank the other members of my committee, Prof. Stephen Hill, Prof. Lisa McElwee-White, Prof. Michael J. Scott and Prof. Daniel R. Talham for their insightful discussions and comments. I woul d like to express my gratitude to Dr. Khalil A. Abboud and his staff at the UFCXC for not only so lving my crystal structures but most importantly for giving me the rough structures asap. I would also like to acknowledge Dr. Wolfgang Wernsdorfer for providing essential single crystal measurements on Fe9 and Mn10Dy2 compounds below 1.8 K using his micro-SQUID apparatus and answering a ll my questions to help bring these results together. Many thanks to Prof. Stephen Hill, and Saiti Datta at the UF Physics Department for the HFEPR measurements. Their patience as they taught me the fundamentals, from

PAGE 5

5 understanding a HFEPR experiment to interpreting the terms in a spin Hamiltonian of a SMM, was exceptional. I would like to thank the enti re Christou group, past and present, for their friendship, laughs and all the positive and negative interactions which have helped me in some way or the other. I appreciate the friendship and help of all my seniors Jon, Nicole, Abhu, Dolos and Alina who helped me get started in th e lab. Special thanks go to Nicole for her sincere help and for always cheering me up. My juniors Chris, Antonio, Taketo, Arpita, Shreya Jennifer also deserve special mention for the all fun talking. I would also like to acknowledge our postdoc Tasos (Dr. A. J. Tasiopoulos) for helping me in my firs t project and Muralee (D r. M. Murugesu) for introducing me to madras curry paste. Watching Harris (Dr. T. C. Stamatatos) getting new clusters everyday has been very inspiring. During the last five years I had the pleasure of working with two great undergraduates Sarah Nam and Matthew Daniels. Sincere thanks go for the efficient secretaries of Prof. Christou, So ndra and Melinda. Without them, the little things that make everything run smoothly would never get done. A special thanks goes out to Soma, Rumi, Parul, Arpita, Shreya and Ozge for their warm friendship. Finally, I must thank my family, which has been an integral part of my existence. I am forever indebted to my parents and my brother Sh ammi for their constant pride, encouragement, never-ending love, and unwavering confidence in me and for always being there for me. Papa and mummis philosophy of work is worship and time is money have helped me become who I am today and where I am today. Thanks sh ammi for being my greatest critique and best brother in the world. I would like to thank my in-laws for accepting me like a daughter and for all their warmth, love and affection. Thanks just seems too insignificant of a word to express

PAGE 6

6 my gratefulness. I feel that my gratitude can not be expressed in words alone. I am grateful for having such a wonderful, loving fa mily and cannot wait to go back. Last but not least, I would like to thank one very special person, my husband Ranjan. The understanding ear with which he listens to me day and night is commendable. His patience and perfectionism in chemistry and cooking has helped me grow as a person. His love and support carried me through the roughest times. Without him, life would be meaningless. Thanks Ranjan for helping me finish this jour ney so that we may begin anot her. I thank you for the daily experience of helping, sharing, complaining, joking and all the things that make life what it is. The chain of my gratitude would be incomplete if I forget to thank the first cause of this chain, Thou Formless One. My deepest and sin cere gratitude for inspiring and guiding this humble being.

PAGE 7

7 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ........11 LIST OF FIGURES.......................................................................................................................13 LIST OF ABBREVIATIONS........................................................................................................ 18 ABSTRACT...................................................................................................................................19 CHAP TER 1 GENERAL INTRODUCTION.............................................................................................. 22 2 LIGAND-INDUCED DISTORTION OF A T ETRANUCLEAR MANGANESE BUTTERFLY COMPLEX.....................................................................................................34 2.1 Introduction............................................................................................................... ....34 2.2 Experimental Section.................................................................................................... 35 2.2.1 Syntheses........................................................................................................... 35 2.2.2 X-ray Crystallography....................................................................................... 37 2.3 Results and Discussion.................................................................................................. 39 2.3.1 Syntheses........................................................................................................... 39 2.3.2 Description of Structures.................................................................................. 40 2.3.2.1 Structure of [Mn4O2(O2CR)5(salpro)] (R = Me ( 2-1 ), But ( 2-3 ))....... 40 2.3.2.2 Structure of NMe4[Mn(O2CPh)2(salproH)] ( 2-4 )...............................42 2.3.3 Magnetochemistry............................................................................................. 43 2.4 Conclusions................................................................................................................ ...47 3 ROLE OF MIXED-LIGAND AND MIXEDSOLVENT SYSTEM: ROUTE TO Mn4 AND Mn25..............................................................................................................................54 3.1 Introduction............................................................................................................... ....54 3.2 Experimental Section.................................................................................................... 55 3.2.1 Syntheses........................................................................................................... 55 3.2.2 X-ray Crystallography....................................................................................... 56 3.3 Results and Discussion.................................................................................................. 57 3.3.1 Syntheses........................................................................................................... 57 3.3.2 Description of Structures.................................................................................. 59 3.3.2.1 Structure of [Mn4(hmp)4(pdmH)2(MeCN)4](ClO4)4 ( 3-1 )..................59 3.3.2.2 Structure of [Mn25O18(OH)2(hmp)6(pdm)8(pdmH)2(L)2](ClO4)6 ( 3-2 )....................................................................................................60 3.3.3 Magnetochemistry............................................................................................. 61 3.3.3.1 Dc Studies........................................................................................... 61

PAGE 8

8 3.3.3.2 Ac Studies...........................................................................................63 3.4 Conclusions................................................................................................................ ...64 4 DIVERSITY OF STRUCTURAL TYPES IN POLYNUCLEAR IRON CHEMISTRY WITH A (N,N,O)-TRIDENTATE LIGAND .........................................................................70 4.1 Introduction............................................................................................................... ....70 4.2 Experimental Section.................................................................................................... 71 4.2.1 Syntheses........................................................................................................... 71 4.2.2 X-ray Crystallography....................................................................................... 74 4.3 Results and Discussion.................................................................................................. 75 4.3.1 Syntheses........................................................................................................... 75 4.3.2 Description of Structures.................................................................................. 77 4.3.2.1 Structure of [Fe7O4(O2CPh)11(dmem)2] ( 4-1 ).....................................77 4.3.2.2 Structure of [Fe7O4(O2CMe)11(dmem)2] ( 4-2 )....................................78 4.3.2.3 Structure of [Fe6O2(OH)4(O2CBut)8(dmem)2] ( 4-3 )...........................79 4.3.2.4 Structure of [Fe3O(O2CBut)2(N3)3(dmem)2] ( 4-4 )..............................79 4.3.3 Magnetochemistry of Complexes 4-1 to 4-4 .....................................................80 4.3.3.1 Dc Studies........................................................................................... 80 4.3.3.2 Rationalization of the Ground State Spin of 4-1 and 4-3 ...................82 4.3.3.3 Determination of the Exchange Interactions in 4-4 ............................83 4.3.4 High-Frequency EPR Spectroscopy..................................................................85 4.4 Conclusions................................................................................................................ ...89 5 A NEW N, N, O CHELATE FOR TRANSIT ION METAL CLUSTER CHEMISTRY: Fe5 AND Fe6 CLUSTERS FROM THE USE OF 6-HYDROXYMETHYL-2, 2 BIPYRIDINE..........................................................................................................................99 5.1 Introduction............................................................................................................... ....99 5.2 Experimental Section.................................................................................................. 100 5.2.1 Syntheses......................................................................................................... 100 5.2.2 X-ray Crystallography..................................................................................... 102 5.3 Results and Discussion................................................................................................ 104 5.3.1 Syntheses......................................................................................................... 104 5.3.2 Description of Structures................................................................................ 105 5.3.2.1 Structure of [Fe5O2(OH)(O2CMe)5(hmbp)3](ClO4)2 ( 5-1 )...............105 5.3.2.2 Structure of [Fe6O2(OH)2(O2CPh)6(hmbp)4](NO3)2 ( 5-2 )................106 5.3.3 agnetochemistry of Complexes 5-1 to 5-4 ......................................................107 5.3.3.1 Dc Studies......................................................................................... 107 5.3.3.2 Rationalization of the Ground State Spin......................................... 109 5.4 Conclusions................................................................................................................ .111 6 NEW STRUCTURAL TYPES IN POLYNUCLEAR IRON CLUSTERS INCORP ORATING O,N,N,O LIGAND: A SNAKE LIKE CHAIN AND A SUPRAMOLECULAR DIMER OF SMMs......................................................................... 117 6.1 Introduction............................................................................................................... ..117

PAGE 9

9 6.2 Experimental Section.................................................................................................. 119 6.2.1 Syntheses......................................................................................................... 119 6.2.2. X-ray Crystallography..................................................................................... 120 6.3 Results and Discussion................................................................................................ 123 6.3.1 Syntheses......................................................................................................... 123 6.3.2 Description of Structures................................................................................ 124 6.3.2.1 Structure of [Fe18O8(OH)2(O2CBut)28(heen)4] ( 6-1 ).........................124 6.3.2.2 Structure of [Fe9O4(OH)4(O2CPh)13(heenH)2] ( 6-2 )........................125 6.3.2.3 Structure of [Fe7O3(OMe)3(MeOH)1(heen)3Cl4.5(H2O)1.5]Cl1.25[FeCl4] ( 6-3 )....126 6.3.2.4 Structure of [Fe6O2(O2CPh)5(heen)3(heenH)](ClO4)2 ( 6-4 )..............127 6.3.3 Magnetochemistry of complexes 6-1 to 6-4 ....................................................128 6.3.3.1 Dc Studies......................................................................................... 128 6.3.3.2 Ac Studies......................................................................................... 131 6.3.3.3 Single-Crystal Hysteresis Studies..................................................... 131 6.4 Conclusions................................................................................................................ .132 7 UNUSUAL STRUCTURAL TYPES IN Mn AND Fe CHEMISTRY FROM THE USE OF N,N, N ,N TETRAKIS (2-HYDROXYETHYL)ETHYLENEDIAMINE......................142 7.1 Introduction............................................................................................................... ..142 7.2 Experimental Section.................................................................................................. 143 7.2.1 Syntheses......................................................................................................... 143 7.2.2 X-ray Crystallography..................................................................................... 146 7.3 Results and Discussion................................................................................................ 149 7.3.1 Syntheses......................................................................................................... 149 7.3.2 Description of Structures................................................................................ 153 7.3.2.1 Structure of [Mn8O3(OH)(OMe)(O2CPh)7(edte)(edteH2)](O2CPh) ( 7-1 )..................................................................................................153 7.3.2.2 Structure of [Mn12O4(OH)2(edte)4Cl6(H2O)2] ( 7-2 )..........................155 7.3.2.3 Structure of [Mn20O8(OH)4(O2CMe)6(edte)6](ClO4)2 ( 7-3 )..............157 7.3.2.4 Structure of [Fe5O2(O2CPh)7(edte)(H2O)] ( 7-4 )...............................158 7.3.2.5 Structure of [Fe6O2(O2CBut)8(edteH)2] ( 7-5 )....................................159 7.3.2.6 Structure of [Fe12O4(OH)2(O2CMe)6(edte)4(H2O)2](ClO4)4 ( 7-6 )....160 7.3.3 Magnetochemistry........................................................................................... 161 7.3.3.1 Dc Studies on 7-1 to 7-3 ...................................................................161 7.3.3.2 Dc Studies on 7-4 to 7-8 ...................................................................163 7.3.3.3 Ac Studies on 7-1 to 7-5 ...................................................................165 7.3.3.4 Rationalization of the Ground State Spin of 7-4 and 7-5 .................167 7.4 Conclusions................................................................................................................ .168 8 SINGLE-MOLECULE MAGNETISM AND MAGNETOSTRUCTURAL CORRELATION W ITHIN A FAMILY OF [MnIII 10LnIII 2] COMPLEXES........................ 184 8.1 Introduction............................................................................................................... ..184 8.2 Experimental Section.................................................................................................. 186 8.2.1 Syntheses......................................................................................................... 186

PAGE 10

10 8.2.2 X-ray Crystallography..................................................................................... 189 8.3 Results and Discussion................................................................................................ 190 8.3.1 Syntheses......................................................................................................... 190 8.3.2 Description of Structures................................................................................ 191 8.3.3 Magnetochemistry........................................................................................... 193 8.3.3.1 Complexes 8-9 (Mn10Y2) and 8-4 (Mn10Gd2)...................................193 8.3.3.2 Comparison of 8-9 (Mn10Y2) with 8-4 (Mn10Gd2), 8-5 (Mn10Tb2), 8-6 (Mn10Dy2), 8-7 (Mn10Ho2), and 8-8 (Mn10Er2)..........................195 8.3.3.3 Comparison of 8-9 (Mn10Y2) with 8-1 (Mn10Pr2), 8-2 (Mn10Nd2), 8-3 (Mn10Sm2)..................................................................................198 8.3.3.4 Out-of-Phase ac Susceptibil ity Signals and Magnetization Hysteresis L oops...............................................................................199 8.4 Conclusions................................................................................................................ .202 9 A FOURTH ISOLATED OXIDATION LEVEL OF THE [Mn12O12(O2CR)16(H2O)4] FAMILY OF SINGLE MOLECULE MAGNETS............................................................... 212 9.1 Introduction............................................................................................................... ..212 9.2 Experimental Section.................................................................................................. 214 9.2.1 Syntheses......................................................................................................... 214 9.3 Results and Discussion................................................................................................ 215 9.3.1 Syntheses......................................................................................................... 215 9.3.2 Magnetochemistry........................................................................................... 218 9.3.2.1 Dc Studies......................................................................................... 218 9.3.2.2 Comparison of the Magnetic Properties of the [Mn12]z(z = 0 3) Family...............................................................................................219 9.3.2.3 Ac Studies......................................................................................... 222 9.4 Conclusions................................................................................................................ .224 APPENDIX A BOND DISTANCES AND ANGLES..................................................................................233 B LIST OF COMPOUNDS......................................................................................................256 C PHYSICAL MEASUREMENTS.........................................................................................258 D VAN VLECK EQUATIONS............................................................................................... 259 LIST OF REFERENCES.............................................................................................................266 BIOGRAPHICAL SKETCH.......................................................................................................284

PAGE 11

11 LIST OF TABLES Table page 2-1 Crystallographic Data for 2-1 MeC N, 2-3 MeOHCH2Cl2C7H16 and 2-4 CH2Cl2.........47 2-2 Bond-valence sums for the Mn atoms of complexes 2-1 2-3 and 2-4 ..............................48 2-3 Comparison of core parameters of selected [Mn4O2]8+ complexes (, ).............................. 48 2-4 Comparison of exchange parameters in [Mn4O2]8+ complexes......................................... 48 2-5 Spin states of 2-1 C H2Cl2 in the |ST, SA, SB> format.........................................................49 3-1 Crystallographic Data for 3-1 and 3-2 MeCN MeOH...................................................65 3-2 Bond-valence sums for the Mn atoms of complex 3-2 ......................................................65 3-3 Bond-valence sums for the O atoms of complex 3-2 .........................................................65 4-1 Crystallographic data for 4-1 MeCN, 4-2MeCN, 4-3MeCN and 4-4 CH2Cl2.............91 5-1 Crystallographic Data for 5-1 MeC N and 5-2MeCNH2O......................................... 112 6-1 Crystallographic data for 6-1 C5H12CH2Cl2, 6-2 MeCN, 6-3 MeOHH2O and 6-4 EtOHH2O..............................................................................................................134 7-1 Crystallographic Data for 7-12CH2Cl2MeOH, 7-2 MeCNH2O and 7-3MeOH..169 7-2 Crystallographic Data for 7-4 C H2Cl2, 7-5 2CHCl3, 7-6 MeCN and 7-7 ........................170 7-3 Bond-valence sums for the Mn atoms of complex 7-1 and 7-3 .......................................170 7-4 Bond-valence sums for the O atoms of complex 7-1 .......................................................171 7-5 Bond-valence sums for the Mn and O atom s of complex 7-2 .........................................171 7-6 Bond-valence sums for the O atoms of complex 7-5 and 7-6..........................................172 7-7 Selected Fe-O distances and Fe-O-Fe angles for 75 .......................................................172 8-1 Crystallographic data for 8-4 MeCNMeOH, 8-6 MeCNMeOH and 8-9 MeCN..... 204 8-2 Bond-valence sums for the Mn atoms of complex 8-4 8-6 an d 8-9 ................................204 9-1 Magnetism Data for [Mn12]z(z = 0 3) Complexes 9-1 to 9-5 .......................................226 A-1 Selected interatomic distances () and angles () for 2-1 MeCN .................................... 233

PAGE 12

12 A-2 Selected interatomic distances () and angles () for 2-3 MeOH2CH2Cl2C7H16..........234 A-3 Selected interatomic distances () and angles () for 2-4 C H2Cl2...................................235 A-4 Selected interatomic distances () and angles () for 3-1 ................................................236 A-5 Selected interatomic distances () and angles () for 3-2 MeCNMeOH .................... 237 A-6 Selected interatomic distances () and angles () for 4-1 MeCN .................................. 238 A-7 Selected interatomic distances () and angles () for 4-2 MeCN .................................... 239 A-8 Selected interatomic distances () and angles () for 4-3 MeCN .................................. 240 A-9 Selected interatomic distances () and angles () for 4-4 C H2Cl2.................................241 A-10 Selected interatomic distances () and angles () for 5-1 MeCN .................................. 242 A-11 Selected interatomic distances () and angles () for 5-2 M eCNH2O..........................243 A-12 Selected interatomic distances () and angles () for 6-1 C5H12CH2Cl2....................244 A-13 Selected interatomic distances () and angles () for 6-2 MeCN .................................. 245 A-14 Selected interatomic distances () and angles () for 6-3 6MeOHH2O.........................246 A-15 Selected interatomic distances () and angles () for 6-4 EtOH.5H2O......................247 A-16 Selected interatomic distances () and angles () for 7-1 C H2Cl2MeOH.....................248 A-17 Selected interatomic distances () and angles () for 7-2 M eCNH2O.......................249 A-18 Selected interatomic distances () and angles () for 7-3 MeOH................................ 250 A-19 Selected interatomic distances () and angles () for 7-5 C HCl3..................................251 A-20 Selected interatomic distances () and angles () for 7-6 MeCN .................................. 252 A-21 Selected interatomic distances () and angles () for 8-4 MeCNMeOH ...................... 253 A-22 Selected interatomic distances () and angles () for 8-6 MeCNMeOH ...................... 254 A-23 Selected interatomic distances () and angles () for 8-9 MeCN .................................. 255

PAGE 13

13 LIST OF FIGURES Figure page 1-1 Representations of magn etic dipole arrangem ents in paramagnetic, ferromagnetic, antiferromagnetic, and ferrimagnetic materials................................................................. 31 1-2 Schematic diagram of a hysteresis curve for a typical ferromagnet. ................................. 31 1-3 Schematic representation of a multidomain ferromagnetic particle in the unm agnetized state............................................................................................................. 31 1-4 Representation of the [Mn12O12]16+ core and [Mn12O12(O2CMe)16(H2O)4] complex with peripheral ligation...................................................................................................... 32 1-5 Representative plots of the potenti al energy versus or ien tation of the ms vector along the z axis and the ms sublevels for a Mn12 complex.......................................................... 32 1-6 Magnetization hysteresis loops for a typical [Mn12O12(O2CR)16(H2O)4] complex........... 33 1-7 Representation of the change in energy of the ms sublevels as a function of the applied magnetic field........................................................................................................ 33 2-1 Structure of SalproH3.........................................................................................................50 2-2 Labeled representation of the structure of 2-1 and 2-3 ......................................................50 2-3 Comparison of the cores of 2-1 and 2-3 with that of the norm al butterfly complexes...... 51 2-4 Labeled representation of the structure of 2-4 ...................................................................51 2-5 Plots of MT vs T for complexes 2-1 CH2Cl2, 2-2 CH2Cl2 and 2-3 CH2Cl2....................52 2-6 The core of 2-1 defining the pairwise exchange in teractions and rationalization of ground state spin of 2-1 ......................................................................................................53 2-7 MT vs. T and reduced magnetization ( M/N B) vs H/T plots for 2-4 CH2Cl2..................53 2-8 Two-dimensional contour plot of the fitting error surface vs D and g for 2-4 CH2Cl2.............................................................................................................................53 3-1 Structure of ligands: 2-hydroxymethyl pyridine (hm pH), 2,6-pyridine dimethanol (pdmH2), 6-hydroxymethyl 2-pyrid ine carboxylic acid (L).............................................. 66 3-2 Labeled representation of the structure of 3-1 ..................................................................66 3-3 Structure of the cation of 3-2 ............................................................................................66 3-4 Centrosymmetric core of 3-2 and its three types of constituent layers.............................. 67

PAGE 14

14 3-5 Plots of MT vs T for complex 3-1 .....................................................................................67 3-6 The core of 3-1 defining the pairwise exchange interactions. ........................................... 68 3-7 Plot of reduced magnetization (M / N B) vs H / T and two-dimensional contour plot of the fitting error surface vs D and g for complex 3-1 ..........................................................68 3-8 Plot of in-phase ( M'T ) and out-of-phase ( M") ac susceptibility data for complex 3-1 .....69 3-9 MT vs T and reduced magnetization ( M/N B) vs H/T plots for complex 3-2 H2O.......... 69 3-10 Plot of M'T vs T (in-phase) ac susceptibility data for 3-2 H2O....................................... 69 4-1 Structure of ligands: mdaH2 and dmemH.......................................................................... 91 4-2 Labeled representation of the structure of 4-1 and 4-2 ......................................................92 4-3 Comparison of cores of 4-1 4-2 and 4-3 ...........................................................................92 4-4 Labeled representation of the structures of 4-3 and 4-4. ...................................................93 4-5 Plots of MT vs T for complexes 4-1 4-2 4-3 and 4-4 ......................................................93 4-6 Plot of reduced magnetization (M / N B) vs H / T for complex 4-1 MeCN........................ 94 4-7 Two-dimensional contour plot of the fitting e rror surface vs D and g for 4-1 MeCN..............................................................................................................................94 4-8 Plot of reduced magnetization (M / N B) vs H / T and two-dimensional contour plot of the fitting error surface vs D and g for 4-2 MeCN.......................................................... 95 4-9 Plot of reduced magnetization (M / N B) vs H / T and two-dimensional contour plot of the fitting error surface vs D and g for 4-3 ........................................................................95 4-10 Rationalization of the ground state spin of 4-1 and 4-3 ....................................................96 4-11 Plot of reduced magnetization (M / N B) vs H / T and two-dimensional contour plot of the fitting error surface vs D and g for 4-4 CH2Cl2........................................................96 4-12 Core of 4-4 defining the pairwise exchange intera ctions and the rationalization of its ground state spin ................................................................................................................97 4-13 HFEPR peak positions for 4-3 Me CN from angle-dependent studies and frequency dependence for 4-3 MeCN............................................................................................... 97 4-14 Simulated Zeem an diagram for a spin S = 5 and 5/2 system with D < 0 and D > 0.......... 98 4-15 Temperature dependent spectra a nd easy-plane peak positions for 4-1 M eCN plotted versus frequency....................................................................................................98

PAGE 15

15 5-1 Structure of ligands: dmemH and hmbpH.......................................................................112 5-2 Synthetic scheme for hmbpH........................................................................................... 112 5-3 Labeled representation of the structure of 5-1 .................................................................113 5-4 Labeled representation of the structure of 5-2 .................................................................113 5-5 Plots of MT vs T for complexes 5-1 5-2 H2O, 5-3 H2O and 5-4 H2O............................114 5-6 Plot of reduced magnetization (M / N B) vs H / T and two-dimensional contour plot of the fitting error surface vs D and g for 5-1 ......................................................................114 5-7 Plot of reduced magnetization (M / N B) vs H / T and two-dimensional contour plot of the fitting error surface vs D and g for 5-2H2O, 5-3H2O and 5-4H2O.......................... 115 5-8 Rationalization of spin ground state of complex 5-1 and 5-2 .........................................116 6-1 Structure of chelates: dmemH, heenH2............................................................................134 6-2 Labeled representation of the structure of 6-1 .................................................................135 6-3 Labeled representation of the structure of 6-2 .................................................................135 6-4 [Fe9]2 dimer showing intermolecular and in tramolecular hydrogen-bonding, and the resultant ON and OFF state with respect to the coupling of the two molecules.............. 136 6-5 Labeled representation of the structure of 6-3 ................................................................137 6-6 Labeled representation of the cation of 6-4 .....................................................................137 6-7 Plot of MT vs T for complexes 6-1 to 6-4.......................................................................138 6-8 Spin alignments at the Fe atoms of 6-1 rationalizing its ground state spin ..................... 138 6-9 Plot of reduced magnetization (M/ N B) vs H/T and two-dimensional contour plot of the r.m.s. error vs D and g for the fit for 6-2 ....................................................................139 6-10 Plot of reduced magnetization (M/ NB) vs H/T for 6-3 and 6-4 ......................................139 6-11 Two-dimensional contour plot of the r.m.s. error vs D and g for the f it for complexes 6-3 and 6-4 .......................................................................................................................140 6-12 Plot of the in-phase (M T ) and out-of-phase ( M" ) ac susceptibility data for 6-2 ............140 6-13 Plot of in-phase ac susceptibility data for 6-3 and 6-4 ..................................................... 141 6-14 Single-crystal magnetization ( M ) vs dc field ( H ) hysteresis loops for 6-2 .....................141

PAGE 16

16 7-1 Structure of ligands....................................................................................................... ...172 7-2 Labeled representation of the cation of 7-1 .....................................................................173 7-3 Crystallographically establis hed coordination m odes of edte4and edteH2 2found in complexes 7-1 to 7-3 ........................................................................................................173 7-4 Labeled representation of the structure of 7-2 .................................................................174 7-5 The structure of the cation of 7-3 .....................................................................................175 7-6 Labeled representation of the core of 7-3 .......................................................................176 7-7 Labeled representation of the structure of 7-4 .................................................................177 7-8 Crystallographically establis hed coordination m odes of edte4and edteH3found in complexes 7-4 to 7-6 ........................................................................................................177 7-9 Labeled representation of the structure of 7-5 .................................................................177 7-10 Labeled representation of the cation of 7-6 .....................................................................178 7-11 Labeled representation of the cation of 7-7 .....................................................................179 7-12 Plots of MT vs T for complexes 7-1 H2O, 7-2 and 7-3 H2O.......................................179 7-13 Plots of reduced magnetization (M / N B) vs H / T for 7-1 H2O and 7-2 ..........................180 7-14 Two-dimensional contour plot of the r.m.s. error surface vs D and g for t he magnetization fit for 7-1 H2O and 7-2 ...........................................................................180 7-15 Plots of reduced magnetization (M / N B) vs H / T and two-dimensional contour plot of the r.m.s. error surface vs D and g for the magnetization fit for 7-3 H2O..................... 181 7-16 Plots of MT vs T for complexes 7-4 to 7-8.....................................................................181 7-17 Plot of reduced magnetization (M / N B) vs H / T for 7-4 and 7-5 2CHCl3H2O..............182 7-18 Two-dimensional contour plot of the fitting e rror surface vs D and g for complexes 7-4 and 7-5 CHCl3H2O...............................................................................................182 7-19 Plot of M T vs. T for complexes 7-1 7-5......................................................................183 7-20 Spin alignments at the Fe atoms of 7-5 rationalizing its S = 5 ground state .................... 183 8-1 Labeled structures of 8-4 8-6 and 8-9 ...........................................................................205 8-2 Centrosymmetric core of 8-6 e mphasizing the ABCBA layer structure......................... 206

PAGE 17

17 8-3 Plot of MT vs T and M'T vs T for complexes 8-4 and 8-9 ..............................................206 8-4 Plot of reduced magnetization (M/ NB) vs H/T for 8-9 and 8-4 ......................................207 8-5 Two-dimensional contour plot of the error surface for the D vs g fit for 8-9 and 8-4 .....207 8-6 Plots of dc MT vs T for Gd(NO3)3 and Dy(NO3)3...........................................................208 8-7 Plots of MT vs T for 8-9 8-5 and 8-6 ..............................................................................208 8-8 Plots of MT vs T for 8-9 8-7 and 8-8 ..............................................................................208 8-9 Plots of MT vs T for 8-1 8-2, 8-3 and 8-9 ......................................................................209 8-10 Plots of out-of-phase M" vs T ac susceptibility data for 8-9 and 8-4 ..............................209 8-11 Plots of out-of-phase M" vs T ac susceptibility data for 8-1 and 8-2 ..............................209 8-12 Plots of in-phase, M'T vs T and out-of-phase M" vs T ac susceptibility data for 8-5 8-6 and 8-7 .......................................................................................................................210 8-13 Plot of relaxation rate vs recip rocal temperature for 8-5 8-7 ........................................210 8-14 Magnetization vs. time decay plots in zero dc field and vs 1/ T plot for 86 .................211 8-15 Single-crystal magnetization vs dc field hysteresis loops for 86 MeCNMeOH.......... 211 9-1 Cyclic voltammogram and diffe rential pulse voltammogra m for 9-1 .............................226 9-2 Proposed structural core of 9-2 9-4 ...............................................................................227 9-3 Plot of MT vs T for 9-2 9-5 ..........................................................................................228 9-4 Plot of reduced magnetization (M / N B) vs H / T for 9-4 and 9-5 ......................................228 9-5 Plot of reduced magnetization (M / N B) vs H / T for 9-2 and 9-3 ......................................229 9-6 Two-dimensional contour plot of the error surface for the D vs g fit and D vs E fit for complex 9-4 .....................................................................................................................229 9-7 Plot of the in-phase ( M T ) and out-of-phase ( M ) ac susceptibility data for 9-4 ............230 9-8 Plot of the in-phase ( M T ) and out-of-phase ( M ) ac susceptibility data for 9-5 ............230 9-9 M" vs T plots for vacuum-dried complexes [Mn12]z(z = 0-3)........................................ 231 9-10 Comparison of the M" vs T plots for vacuum-dried complexes [Mn12]z(z =0-3).......... 232

PAGE 18

18 LIST OF ABBREVIATIONS But tertiary butyl BVS bond valence sun CV cyclic voltammogram dmemH: 2-{[2-(dimethylamino)ethyl]-methylamino}ethanol DPV differential pulse voltammogram edteH4 N,N,N',N' -tetrakis(2-hydroxyethyl)ethylenediamine heenH2 N ,N '-bis(2-hydroxyethyl)ethylenediamine HFEPR high Frequency electron paramagnetic resonance hmbpH 6-hydroxymethyl-2,2 -bipyridine hmpH: 2-hydroxymethyl pyridine L 6-hydroxymethyl-2-pyridine carboxylic acid pdmH2: 2,6-pyridine dimethanol PS II photosystem II Py pyridine salproH3 1,3-bis(salicylideneamino)-2-propanol TIP temperature independent paramagnetism ZFS zero-field splitting

PAGE 19

19 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DIVERSITY OF STRUCTURAL TYPES AND MOLECULAR NANOMAGNETISM IN IRON AND MANGANESE CLUSTERS By Rashmi Bagai May 2008 Chair: George Christou Major: Chemistry The primary reason for the current interest in high nuclearity manganese and iron-oxo clusters is because of their relevance in mo lecular magnetism and bio-inorganic chemistry. A particularly appealing area in molecular magnetism is that of molecules which show slow relaxation of magnetization at low temperatures behaving as tiny magnets, and thus known as single molecule magnets (SMMs). One of the first SMM to be synthesized was [Mn12O12(O2CMe)16(H2O)4] (Mn12), which now serves as th e drosophila of molecular magnetism. The various modifications of the Mn12 family of SMMs have permitted advances in our knowledge and understanding of Mn12 complexes and the SMM phenomenon in general. For the first time, Mn12 family of SMMs has been extended to a fourth isolated member by the successful isolation and characterization of [Mn12]3complexes with S = 17/2 ground-state spin and D = 0.24(1) cm-1. When studied by ac susceptib ility techniques, the [Mn12]3complexes exhibit frequency-dependent out-of-phase signals indicating them to be SMMs, albeit with smaller barriers than the other Mn12 oxidation levels. The [Mn12]3complexes represent a fourth isolated oxidation level of the Mn12 family of SMMs, by far the la rgest range of oxidation levels yet encountered within si ngle-molecule magnetism.

PAGE 20

20 Towards the synthesis of polynuclear molecula r clusters, various al cohol-based ligands have been explored. Among them is a family of ligands incorporat ing one, two and four hydroxyethyl arms on ethylenediamine backbone. Use of dmemH (2-{[2-(dimethylamino)ethyl]methylamino}ethanol) has led to two new Fe7 clusters and one Fe6 cluster, depending on the identity of the carboxylate employed. Unlik e dmemH, use of hmbpH (6-hydroxymethyl-2,2 bipyridine), one that amalgamates the ch elating property of bipyridine and hmpH (hydroxymethyl pyridine), resulted in Fe6 cluster irrespective of the carboxylate employed. These contrasting results from flexible dm emH and rigid hmbpH underline the exquisite sensitivity of the reaction product on a variety of reaction conditions and reagents used. The magnetochemical characterization of these clus ters emphasize how ground state spin values of significant magnitude can result from spin-frust ration effects even though all the pair wise exchange interactions are antiferromagnetic. The use of heenH2 ( N ,N '-bis(2-hydroxyethyl)ethylenediamine) has provided an entry into new cluster types, including a discrete Fe18 molecular chain with an unusual double-headed serpentine structure and a Fe9 SMM, both having unprecedented st ructures in Fe chemistry. Fe18 represents the highest nu clearity, chain-like meta l-containing molecule to be yet discovered, and Fe9 SMM contains a mixture of ON and OFF dimers with respect to the quantum-mechanical coupling through the hydrogen-bond. The initial use of edteH4 ( N,N,N',N' -tetrakis(2-hydroxyethyl)et hylenediamine) in Mn and Fe chemistry has resulted in novel complexes of high nuclearity and archite ctural beauty ranging from Mn8 to Mn20 and Fe5 to Fe12. The complexes all possess rare or novel core topologies. The combined results demonstrate the ligating flexib ility of alkoxide containing chelates and their usefulness in the synthesis of a variety of Fex and Mnx molecular clusters.

PAGE 21

21 A family of tetranuclear Mn clusters has been synthesi zed using Schiff-base ligand, salproH3 (1,3-bis(salicylideneamino)-2-propanol). The structure of these is much more closed than the previously reported bu tterfly-like complexes as a result of the alkoxide oxygen of salpro bridging the two wingtip Mn atoms. Fitting of the dc magnetic sus ceptibility data revealed that the various exchange parameters are all antifer romagnetic, and the core thus experiences spin frustration effects. Use of hmpH and pdmH2 (2,6-pyridine dimethanol) has resulted in an aesthetically pleasing Mn25 with a ground-state spin of 65/2, which is the second highest in Mn chemistry. Achieving high spin ground state is one of the elus ive goals in the search for obtaining superior SMMs. A family of isostructural heterometallic Mn-Ln clusters with a [MnIII 10LnIII 2] core (Ln = Pr, Nd, Sm, Gd, Tb, Dy, Ho and Er) has been synthesized as well as the [MnIII 10YIII 2] analog with diamagnetic YIII to assist the magnetic studies of the nature of MnLn exchange interactions. Complexes containing Tb, Dy and Ho exhibi t strong frequency-dependent out-of-phase ac susceptibility signals characteri stic of SMMs which was confirmed for the Dy complex by the observation of magnetization hysteresis.

PAGE 22

22 CHAPTER 1 GENERAL INTRODUCTION Magnetism has been known to humans for m illennia. What is magnetism? This question has fascinated people ever since Thales of Miletus (about 634-546 BC) first described the phenomenon as the attraction of iron by lodestone.1 Over the last 2500 years, magnetism has played an important role in the development of civilization, we have not only extensively used the phenomenon for navigation, power production and h igh tech applications but we have also come a long way in exploring its origin. Today we more specifically associate lodestone with the spinel magnetite, Fe3O4, which is magnetically aligned in nature, most likely by the earths magnetic field during the coo ling process of hot lava. One of the most fundamental ideas in magne tism is the concept of a magnetic field. A magnetic field is produced whenever there is an el ectrical charge in motion, specifically the spin and orbital angular momenta of electrons within atoms of a material. Thus the essential component of any magnetic material is the presence of an unpaired electron and how they interact with each other determines the magnetic behavior of all magnets.2 The different types of magnetic materials are usually classified on the basis of their susceptibility ( ). A magnetic susceptibility is merely the quantitative measure of the response of a material to an applied magnetic field. Substances were first classifi ed as diamagnetic or paramagnetic by Michael Faraday in 1845, but it was not until many year s later that these phenomena came to be understood in terms of electronic structure.3 Diamagnetism is an underlying property of all matter and arises from the interaction of paired electrons with the magnetic field while paramagnetism is a property exhibited by substa nces containing unpaired electrons. Diamagnets are slightly repelled by a magnetic field and for these is small negative (-10-5) and paramagnets are attracted into an applied field and for these is small positive (10-3 to 10-5). Diamagnetic

PAGE 23

23 susceptibilities are independent of field and temperature while paramagnetic susceptibility varies inversely with temperature, = C/T, where C is Curie constant.4 Upon bringing the spins closer together, spin-coupling enables a tendency toward parallel ( ) or antiparallel ( ) alignment. This behavior can be modeled as a function of temperature by the Curie-Weiss expression, = C/(T), where is proportional to the strength of c oupling between adjacent spins. Pairwise ferromagnetic coupling ( ) can lead to long-range ferromagnetic order, whereas antiferromagnetic order may arise from pairwise antiferromagentic coupling ( ) as shown in Figure 1-1. The ferromagnetic solids are the most widely recognized magnetic materials with ranging from 50 to 10,000. Examples of these materi als are iron, cobalt, nick el and several rare earth metals and their alloys.5 Ferrimagnets (e.g. magnetite, Fe3O4) arise from antiferromagnetic coupling, which does not lead to complete cancellation, and thus they have a net magnetic moment (Figure 1-1). Ferro-, an tiferroand ferrimagnetic ordering occurs below a critical temperature, Tc. Below Tc, the magnetic moments for ferroand ferrimagnets align in small domains. In the absence an appl ied magnetic field, despite the natu re of interactions, a net zero magnetization is thermodynamically favored, as diffe rent domains have their net magnetizations randomly oriented. A useful property of these materi als is that when an external field is applied to them, the magnetic fields of the individual domains tend to line up in the direction of the external field. At some particular value of the field, no domains are presen t as all the spins are aligned and net magnetization is at its saturation point. The system will remain in that situation unless enough energy is given to overcome th e energy barrier for domain formation. This property can be monitored in a plot of magne tization vs applied fiel d called hysteresis loop (Figure 1-2).2 Because an additional field is required to reverse the direction of magnetization, magnetic storage of information is possible in ferromagnetic and ferrimagnetic materials.

PAGE 24

24 In addition to the ferriand ferromagnetic behavior, other magnetic-ordering phenomena, such as metamagnetism, canted ferromagne tism, and spin-glass behavior may occur.6 The transformation from an antiferromagnetic state to a high moment state is called metamagnetism. A canted antiferromagnet (or weak ferromagnet) results from the relative canting of antiferromagnetically coupled spins that lead to a net moment. A spin glass occurs when local spatial correlations with neighboring spins exist, but long-ra nge order does not. The spin alignment for a spin glass is that of parama gnet; however, unlike paramagnets, for which spin directions vary with time, the spin orientations of a spin glass remain fixed or vary only very slowly with time.7 As throughout history, todays magnetism research remains closely tied to applications. It is therefore no surprise that some of the forefr ont research areas in magnetism today are driven by the smaller and faster mantra of advanced technology. Future of the magnetic data storage and memory technology is concerne d with cramming information into smaller and smaller bits and manipulating these bits faster and faster. T hus, the need to devel op magnetic particles of nanoscale dimensions is unavoidable. The synthesis of such nanomagnets can be accomplished by fragmentation of bulk ferromagnets or ferrimagnets, for example, crystals of magnetite can be broken down such that each fragment is smaller in size than a single domain (20-200 nm); these subdomain nanoscale magnetic particles with varyi ng sizes are called superparamagnets (Figure 1-3).8 The magnetic moments within one superpar amagnetic particle are ferromagnetically aligned due to short range order. These clusters are thermally unstable, i.e., their magnetic moments (represented by moment v ectors) experience thermal fluctua tions with great ease, as is the case with paramagnetic species due to lack of long-range ordering. In other words, even though the temperature is below the Curie or Neel temperature and the thermal energy is not

PAGE 25

25 sufficient to overcome the coupling forces betw een neighboring atoms, the thermal energy is sufficient to change the direction of magnetization of the entire crystallite. The resulting fluctuations in the direction of magnetization cause the magnetic moment to average to zero. When cooled below a critical blocking temperature ( TB), superparamagnetic systems experience a very slow relaxation time, thei r net magnetic moments align parallel to the applied field and appear to behave as if they had an apparent bulk-like ferromagnetic behavior.8 This aspect will result in hysteresis of apparent ferromagnetic be havior. But this approach unfortunately gives a distribution of particle sizes, and this complicates detailed study of these systems, making difficult, for example, an accurate assessment of va riation of properties as a function of particle size.9 One approach being investigated for new magnets is based on molecules as building blocks also called bottom-up approach. Molecule-based ma gnets present several at tributes unavailable in conventional metal and me tal-oxide magnets. These prope rties include low density, mechanical flexibility, low temperature pr ocessability, high magnetic susceptibility, biocompatibility and several other desirable characteristics.10 This strategy has the advantage that a single particle size can be en sured, that organic ligands on th e periphery can be chosen or systematically altered to ensure cr ystallinity and solubility in a va riety of solvents, and that X-ray crystallographic and various spectroscopic and p hysicochemical studies can be readily performed in solution and/or solid state.6,11 Worldwide interest in molecule-based magnets has arisen for both fundamental scientific and technological reasons. Mol ecular magnetic materials have been added to the library of magnetism only at the end of twentieth century. The first molecule-based magnet, reported in 1967 was that of [FeIII(S2CNEt2)2Cl] molecules, which orders at 2.46 K.4 Subsequently, there

PAGE 26

26 was very little publishe d activity in this area until 1987 when Miller et al. characterized a ferromagnetic transition in the organomettalic donor acceptor salt, [Fe(C5Me5)2][TCNE]. The spins associated with both donor and acceptor units are strongly coupled along the chains in a ferromagnetic fashion resulting in bulk fe rromagnetic properties with a spontaneous magnetization below Tc = 4.8 K.7 Since then, plethora of molecu le-based magnets exhibiting a wide variety of bonding and structural motifs ha ve been synthesized. These include molecules with extended bonding within chains (1D), within layers (2D), and within 3D network structures. Molecule-based magnets include materials with sp ins only in organic moietie s (in p orbitals), and those with spins both on metal ions and organic moieties and those materials with spins on metal ions with exchange pathways provided by organic moieties that do not contain spin.10 A big breakthrough in molecule-based systems was the discovery of zero-dimensional (isolated molecules) nanoscale magnets now called Single Molecule Magmets (SMMs). One of the first SMM to be synthesized was [Mn12O12(O2CMe)16(H2O)4] (Mn12),12 which by now serves as the drosophila of molecular magnetism (Figure 1-4).13 The ease of preparation, stability, high ground state spin, high magnetic anisotropy coupled with its hi ghly crystalline nature and high symmetry space group, which simplifies the spin Hamiltonian by precluding second-order transverse (rhombic) terms, has made Mn12 the favorite for detailed study within the chemistry and physics communities by a myriad of techniques.14 Structurally, the family of Mn12 molecules contain an external crown of eight MnIII ions ( S = 2), which are ferromagnetically coupled and an inner core of four MnIV ions ( S = 3/2), also ferromagnetically coupled. The crown and core are antiferromagnetically coupled to produce a total spin of S = 10.15 Remarkably, a negative axial zero-field splitting (ZFS), D leads to a loss in the dege neracy of the associated ms levels, such that ms = +10 and ms = -10 are lowest in energy (Figure 1-5) Thus there exists an energy barrier

PAGE 27

27 for the conversion of spin up to spin down. Th e slow relaxation of magnetization, which is at the origin of the interesting behavior, is due to the presence of an energy barrier to be overcome in the reversal of the magnetic moment. The energy barrier is proportional to S2| D | for integer spin and (S2-1/4)| D | for half-integer spin system.13 A unique feature of Mn12 is that slow relaxation of magne tization gives rise to hysteresis cycle, similar to that observed in bulk magnets, but of mol ecular origin (Figure 1-6). The magnet-like behavior of Mn12 has sparked the idea that informa tion might one day be stored as the direction of magnetizati on in individual molecules.16 The most information that can be stored on hard drives and other devices currently is 3 billion bits, or 3 gigabits, in 1 cm2 area of a cobalt-based magnetic material. The much sm aller size of SMMs means that one can get 30 trillion of them into 1 cm2, and thus a storage density of 30 tril lion bits, or 30 terabits, is feasible. This is 10,000 times greater than the cu rrent best by computer manufacturers.17 One of the research challenges now is to find better SMMs that function at higher temperatures. The second appealing feature of Mn12 is that relaxation of its magneti zation shows clear quantum effects, which is manifested in the step -like hysteresis lo op (Figure 1-6).18 The observed steps correspond to an increase in the re laxation rate of magnetization that occurs when there is an energy coincidence of ms sublevels on the opposite sides of the potential energy barrier. For these critical field values, H = nD/gB, at which steps occur, quantum tunneling of magnetization (QTM)19 is allowed, resulting in an increase in the relaxation rate of the molecule. Thus the relaxation of the magnetization of an SMM occurs not just by thermal activation over the energy barrier, but also by QTM through the energy barrier (Figure 1-7). A transverse component contained in the Hamiltonian of the molecule must be present to promote tunneling through the energy barrier; such transverse components can be provided in three ways 1) by low symmetry

PAGE 28

28 components of the crystal field 2) by a magne tic field provided by magnetic nuclei 3) by a magnetic field provided by neighboring molecules.19 Although tunneling provides a route for rapid reversal of magnetization and, hence, a less attractive memory storage device but, these features can be used to devel op new classes of quantum comput ers in which quantum coherence is used to store information. The SMM phenomenon is not unique to Mn12, the family of SMM compounds has been extended to various other meta ls including Fe, V, Co, Ni and combinations of 3d with 4d, 5d, 4f paramagne tic ions, and homometallic LnIII species.9,13,20-23 In order to be considered for real applications, the challenge th at falls within the domain of inorganic chemistry is to synthesize molecules with higher spin-rever sal barriers, capable of storing information at more practical temperatures. In order to observe SMM behavior from a mo lecular compound, it is necessary to have a high spin ground state, S and a large negative ZFS of the ground state and negligib le interactions between molecules. The combination of these properti es can lead to energy barrier so that, at low temperatures, the system can be trapped in one of the high-spin energy wells In principle, large spin can be achieved using a small number of ions if the individual components have a large spin and ZFS of clusters is determin ed by single ion anisotropy and sp in-spin interactions which can be both magnetic dipolar and exchange in nature.20 Two ions with S = 5/2 ferromagnetically coupled can have a ground state with S = 5, while it is necessary to assemble a cluster of ten S = ions in order to achieve the same result. This is the reason why ions w ith large spin such as high spin FeIII and MnIII have been largely used for the preparation of SMMs.13,22 In addition to their relevance in magnetoch emistry, Mn and Fe are two of the most important elements in biochemistry.24-27 The ability to exist in diffe rent oxidations states makes manganese well suited as the active site for redo x reactions in a number of metalloproteins and

PAGE 29

29 enzymes.28 The most important of which is the wateroxidizing complex of PS II. This reaction is responsible for the generation of almost all the oxygen for this planet.29 The relevance of Fe in geology and biochemistry is even larger than that of Mn. Iron is the fourth most abundant element in the earths crust. In addition to its av ailability, iron also po ssesses chemical properties such as Lewis acidity and redox capability that allows it to perform diverse set of metabolic functions. Iron storage protein ferritin can store upto 4500 iron atoms.30 In addition, ferritin is also considered as a nanosize magnetic particle and has been investigated for quantum tunneling effects of magnetization.31 The work presented in this th esis focuses on the synthesis and characterization of new complexes of Mn and Fe, stimulated by the search for inorganic models of important metalloproteins and by the remark able magnetic properties that these complexes display. In order to design clusters, it is necessary to have available both connecting blocks, which provide efficient bridges and determine the growth of the cluster, and terminal blocks, which stop the growth of the cluster at a finite size. The bridging blocks must not only provide the right connection between metal ions but also provide efficient exchange pathways thus assuring strong magnetic coupling. In the design and synthesis of polynuclear molecular cl usters, the choice of appropriate ligand is the most im portant step. The ligand (from the latin word ligare, to bind) is any molecular moiety that has at least one donor atom i.e. an atom with a non-binding electron pair. A ligand is called mono-, bi-, trietc, dentate if it possesses one, two, three etc donor atoms. Commonly employed are the polydentate ligands th at possess more than one donor atom. If the different donor atoms coordinate the same metal ion it is called a chelating ligand. An important class of polydentate ligands is constituted by polyalcohols since this functi onality is an excellent bridging group and fosters higher nuclear ity product formation on deprotonation.32 In this work

PAGE 30

30 we decided to explore the possibi lity of creating new structural types using a variety of alcohol based ligands: 1) a pentadentate Schiff-base ligand, 2) biand tridentate ligands containing one and two hydroxymethyl arms on pyridine backbone, 3) tri-, tetraand hexadentat e ligands incorporating one, tw o and four hydroxyethyl arms on ethylenediamine backbone. The layout of the thesis is as follows. Chapters 2 to 7 are organized according to the types of ligands used. In Chapter 2, a Schiff-based ligand (salproH3) has been used to synthesize tetranuclear Mn clusters and magnetic interactio ns between different spin centers have been determined. Chapter 3 is devoted to the mix ed-ligand approach where synthesis of a high nuclearity, high spin complex has b een achieved using hmpH and pdmH2. In Chapter 4 and 5, a flexible and a rigid N,N,O based ligand, dmemH and hmbpH respectively, is employed for the synthesis of Fe5, Fe6 and Fe7 clusters. Chapter 6 explores the potential of O,N,N,O based ligand, heenH2, which has led to the largest molecular chain complex, Fe18, and also provides insight into the micro SQUID measurements of a supramolecular Fe9 dimer. Denticity of ligand is further extended in Chapter 7, which em ploys O,O,N,N,O,O based ligand, edteH4, for the preparation of novel Mnx and Fex clusters. Synthesis and magneto structural correlation of a family of heterometallic Mn-Ln SMMs is the th eme of Chapter 8. The chemical reduction of [Mn12O12(O2CCHCl2)16(H2O)4] forming one-, twoand threeelectron reduced complexes with identical peripheral ligation is detailed in Chapter 9.

PAGE 31

31 ParamagneticFerromagneticAntiferromagnetic Ferrimagnetic Figure 1-1. Representations of magnetic dipole arrangement s in 1) paramagnetic, 2) ferromagnetic, 3) antiferromagnetic, and 4) ferrimagnetic materials. M H Ms Ms Figure 1-2. Schematic diagram of a hysteresis curve for a typical ferromagnet showing magnetization (M) as a function of the applied magnetic field (H). Saturation magnetization is indicated by Ms. Figure 1-3. Schematic representation of a multidomain ferromagnetic particle in the unmagnetized state. Each of the three do mains with net moments remain randomized in this state. Circles of varying sizes represent the subdomain superparamagnetic clusters.

PAGE 32

32 Figure 1-4. (left) Representation of the [Mn12O12]16+ core. (right) [Mn12O12(O2CMe)16(H2O)4] complex with peripheral ligation. Color Code: MnIV, green; MnIII, blue; O, red; C, grey. ms= +10 +9 +8 +7 +6 +5 +4 +3 +2 +1 ms= 0 -1 -2 -3 -4 -5 -6 -7 -8 -9 ms= -10 ms Energy 100|D| ms= 0 ms= +10 +9 +8 +7 +6 +5 +4 +3 +2 ms= -10 -9 -8 -7 -6 -5 -4 -3 -2 +1 -1Energy Orientation of msvector ( ) (a) (b) ms= +10 +9 +8 +7 +6 +5 +4 +3 +2 +1 ms= 0 -1 -2 -3 -4 -5 -6 -7 -8 -9 ms= -10 ms Energy 100|D| ms= 0 ms= +10 +9 +8 +7 +6 +5 +4 +3 +2 ms= -10 -9 -8 -7 -6 -5 -4 -3 -2 +1 -1Energy Orientation of msvector ( ) (a) (b) Figure 1-5. Representative plots of the potential energy versus a) the orientation of the ms vector ( ) along the z axis and b) the ms sublevels for a Mn12 complex with an S = 10 ground state, experiencing zero-field splitting.

PAGE 33

33 -1 -0.5 0 0.5 1 -1-0.500.51 M/Ms 0 H ( T ) 0.004 T/s 1.3 K 1.8 K 2 K 2.2 K 2.4 K 2.6 K 2.8 K 3 K 3.6 K M/Ms -1 -0.5 0 0.5 1 H0(T) 4 mT/s 1.3 K 1.8 K 2 K 2.2 K 2.4 K 2.6 K 2.8 K 3 K 3.6 K -1 0 0.5 1 -0.5 Figure 1-6. Magnetization hystere sis loops for a typical [Mn12O12(O2CR)16(H2O)4] complex in the 1.3-3.6 K temperature range at 4 mT/s fi eld sweep rate, M is normalized to its saturation value, Ms. H = 0 H = 0 H = 0 Figure 1-7. Representation of the change in energy of the ms sublevels as the magnetic field is swept from zero to a non-zero value. Res onant magnetization tunneling occurs when the ms sublevels are aligned between th e two halves of the diagram.

PAGE 34

34 CHAPTER 2 LIGAND-INDUCED DISTORTION OF A T ETRANUC LEAR MANGANESE BUTTERFLY COMPLEX 2.1 Introduction Manganese cluster chem istry has been receivi ng a great deal of a ttention for two main reasons: (i) the occurrence of this metal in a va riety of manganese-cont aining biomolecules, the most important of which is the water oxidizing complex (WOC) in the ph otosynthetic apparatus of green plants and cyanobacteria. This contains an oxide-bridged Mn4 unit and is responsible for essentially all the oxyge n gas on this planet.33 This has stimulated the se arch for tetranuclear Mn complexes with oxide bridges that can serve as models for the WOC.29 (ii) High nuclearity Mn clusters often display large ground state spin ( S) states as a result of ferromagnetic exchange interactions and/or sp in frustration effects.34,35 If such molecules with large S values also possess significant magnetoanisotropy of the Ising (easy-axis) type, then th ey have the potential to be single-molecule magnets (SMMs).13 These are individual molecules that possess a significant barrier (vs kT ) to magnetization relaxation and thus exhi bit the ability to function as magnets below their blocking temperature ( TB). Our group has had a strong interest over ma ny years in the development of synthesis methodologies to oxide-bridged Mn clusters, primarily with carboxylate ligands. One synthetic strategy that has proven particularly useful has been the use of the preformed clusters of general formula [Mn3O(O2CR)6(py)3]0,+ as starting materials in reactions with a variety of co-reagents.3638 A wide range of the latter have been employed, almost always bidentate or higher denticity chelates, and often ones that also contain pote ntially bridging alkoxide groups. Such reactions have often caused higher-nuclearity products to form, both homoand mixed-valent.36-38 The present work represents an extension of this a pproach. As part of our continuing search for new preparative routes to high nuclearity Mn cluste rs, we have investigated the reactivity of the

PAGE 35

35 pentadentate Schiff-base ligand 1,3-bis (salicylideneamino)-2 -propanol (salproH3, Figure 2-1). This group has been used previously by others in Mn chemistry and had afforded mononuclear, dinuclear and polymeric complexes.39-42 With this precedent, we believed that salproH3 might prove a route to more new Mn compounds under a ppropriate reaction conditions, and decided to investigate its reactions with the [Mn3O(O2CR)6(py)3] complexes. It was obvious that pentadentate salproH3 cannot bind to these Mn3 species without resulting in a serious structural perturbation, and a possible nuclearity change. Indeed, as will be described below, these reactions have yielded new types of Mn4 complexes with a core structure that is distinctly different from those seen before. Additionall y, a mononuclear complex has also been obtained. The syntheses, structures and magnetochemical pr operties of these comple xes are the subject of this chapter.43 2.2 Experimental Section 2.2.1 Syntheses All preparations were perfor med under aerobic conditions using reagents and solvents as received. The compound salproH3 was synthesized using the reported procedure.44 [Mn3O(O2CMe)6(py)3]py, [Mn12O12(O2CMe)16(H2O)4], [Mn3O(O2CEt)6(py)3]py, [Mn3O(O2CBut)6(py)3] and [Mn3O(O2CPh)6(py)2(H2O)] were synthesized as reported elsewhere.12,45 [Mn4O2(O2CMe)5(salpro)] (2-1). Method A. To a stirred solution of salproH3 (0.05 g, 0.17 mmol) in CH2Cl2/MeOH (3/2 mL) was added triethylamine (0.08 mL, 0.55 mmol) followed by the addition of a solution of complex [Mn3O(O2CMe)6(py)3]py (0.22 g, 0.25 mmol) in CH2Cl2 (10 mL). This solution was left under magne tic stirring for 30 minut es and then filtered through a medium frit. The brown filtrate was left undisturbed to evaporate slowly, giving X-ray quality crystals that grew slowly over five days These were collected by filtration, washed with

PAGE 36

36 CH2Cl2 and dried in vacuo. Yield 56%. Anal. Calcd (Found) for 2-1 CH2Cl2 (C28H32Mn4N2O15Cl2): C, 36.27 (36.82); H, 3.48 (3.64); N, 3.02 (2.97). IR (KBr, cm-1): 3446br, 1625s, 1568s, 1445s, 1394s, 1297m, 1153m, 109 2w, 1026m, 759m, 677s, 595s, 468m. Method B. To a stirred solution of salproH3 (0.20 g, 0.67 mmol) in MeCN/MeOH (3/2 mL) was added triethylamine (0.08 mL, 0.55 mmol) followed by the addition of a solution of complex [Mn12O12(O2CMe)16(H2O)4] (0.32 g, 0.17 mmol) in MeCN (10 mL). This solution was left under magnetic stirring for one hour and then filtered through a medium frit. The homogeneous brown solution was left undisturbed for slow eva poration, giving Xray quality crystals that grew slowly over the course of one week. These we re collected by filtration, washed with acetonitrile, and dried in vacuo Yield 16%. The product was identified as 2-1 by IR spectral comparison with ma terial from method A. Method C. To a stirred solution of salproH3 (0.05 g, 0.17 mmol) in MeCN/MeOH (3/2 mL) was added triethylamine (0.08 mL, 0.55 mmol) followed by the addition of a solution of Mn(O2CMe)3H2O (0.09 g, 0.34 mmol) in MeCN (10 mL).This solution was left under magnetic stirring for 30 minutes and then work ed up as for Method B. Yield 28%. The product was identified as 2-1 by IR spectral comparison with material from method A. [Mn4O2(O2CEt)5(salpro)] (2-2). To a stirred so lution of salproH3 (0.05 g, 0.17 mmol) in CH2Cl2/MeOH (3/2 mL) was added triethylam ine (0.08 mL, 0.55 mmol) followed by the addition of a solution of complex [Mn3O(O2CEt)6(py)3]py (0.31 g, 0.34 mmol) in CH2Cl2 (10 mL). This solution was left under magnetic stir ring for 30 minutes and then filtered through a medium frit. X-ray quality crystals were obtai ned over the course of three days by vapourdiffusing diethyl ether into the f iltrate. The resulting crystals we re collected by filtration, washed with ether, and dried in vacuo Yield 22%. Anal. Calcd (Found) for 2-2 CH2Cl2

PAGE 37

37 (C33H42Mn4N2O15Cl2): C, 39.74 (39.66); H, 4.24 (4.21); N, 2.80 (2.68). IR (KBr, cm-1): 3441br, 2879m, 1626s, 1572s, 1446m, 1404 m, 1299m, 1150w, 1080w, 1031w 750w, 676m, 598s, 467m. [Mn4O2(O2CBut)5(salpro)] (2-3). To a stirred solution of salproH3 (0.05 g, 0.17 mmol) in CH2Cl2/MeOH (3/2 mL) was added triethylam ine (0.08 mL, 0.55 mmol) followed by the addition of solution of complex [Mn3O(O2CBut)6(py)3] (0.29 g, 0.28 mmol) in CH2Cl2 (10 mL). This solution was left under magnetic stirring for 30 minutes and then filtered through a medium frit. X-ray quality crystals were obtained during the course of five days by layering the filtrate with heptane and allowing the solvents to slow ly mix. The resulting crys tals were collected by filtration, washed with heptane and dried in vacuo Yield 25%. Anal. Calcd (Found) for 2-3 CH2Cl2 (C42.5H61Mn4N2O15Cl): C, 47.58 (47.44); H, 5.83 (5.86); N, 2.49 (2.74). IR (KBr, cm-1): 3442br, 2959m, 1627m, 1560s, 1482m 1447w, 1408m, 1358m, 1301 m, 1221m, 1150w, 1029w, 895w, 757w, 678m, 599s, 439m. NMe4[Mn(O2CPh)2(salproH)] (2-4). To a stirred solution of salproH3 (0.05 g, 0.17 mmol) in MeCN was added 25 wt % solution of NMe4OH in MeOH (0.02 mL, 0.49 mmol) followed by the addition of a solution of complex [Mn3O(O2CPh)6(py)2(H2O)] (0.08 g, 0.07 mmol) in CH2Cl2 (10 mL). This solution was left under ma gnetic stirring for one hour and then filtered through a medium frit. X-ray quality crys tals were obtained during the course of five days by layering the filtrate with diethyl ether:hexane (1:1 v/v) Yield 25%. Anal. Calcd (Found) for 2-4 CH2Cl2 (C35.5H39MnN3O7Cl): C, 60.04 (60.34); H, 5.54 (5.90); N, 5.92 (6.02). IR (KBr, cm-1): 1623s, 1599s, 1541m, 1530m, 1467m, 1446s, 1393w, 1399m, 1304m, 1193m, 1149m, 1089w, 1070w, 1008w, 905m, 861w, 758s 614s, 599m, 577m, 459m. 2.2.2 X-ray Crystallography Data were collected by Dr. Khalil A Abboud using a Siemens SMART PLATFORM equipped with a CCD area detector and a graphite monochromator utilizing Mo-K radiation (

PAGE 38

38 = 0.71073 ). Suitable crystals of 2-1 MeCN, 2-3 MeOH2CH2Cl2C7H16 and 2-4 CH2Cl2 were attached to glass fibers using silicone grease and transferred to a goniostat where they were cooled to 173 K for data collection. Cell parame ters were refined using up to 8192 reflections. A full sphere of data (1850 fram es) was collected using the -scan method (0.3 frame width). The first 50 frames were remeasured at the end of data collection to monitor the instrument and crystal stability (maximum correction on I was < 1 %). Absorption corrections by integration were applied based on measured indexed crystal faces. The structures were solved by direct methods in SHELXTL6 ,46 and refined using full-matrix least squares. The non-H atoms were treated an isotropically, whereas th e hydrogen atoms were placed in ideal, calculated positions and refined as riding on their respective carbon atoms. Refinement was done using F2. In 2-1 MeCN, the asymmetric unit consists of the Mn4 cluster and a disordered MeCN molecule. A total of 465 pa rameters were refined in the final cycle of refinement using 5581 reflections with I > 2 ( I ) to yield R1 and wR2 of 4.13 and 8.28%, respectively. In 2-3 MeOHCH2Cl2C7H16, the asymmetric unit consists of the cluster and one heptane, one methanol, and two dichloromethane molecules. The three methyl groups on C17 are disordered, and each dichloromethane molecule has one chlorine atom disordered. In each case, two disorder sites were included, and their site occupation factors independently refined. A total of 723 parameters were included in the final cy cle of refinement using 9336 reflections with I > 2 ( I ) to yield R1 and wR2 of 4.73% and 12.20%, respectively. In 2-4 CH2Cl2, A total of 920 parameters were refined in the final cycl e of refinement using 7287 reflections with I > 2 ( I ) to yield R1 and wR2 of 4.58% and 8.44%, respectively. The crystallographic data and structure refinement details for 2-1 MeCN, 2-3 MeOHCH2Cl2C7H16 and 2-4 CH2Cl2 are listed in Table 2-1.

PAGE 39

39 2.3 Results and Discussion 2.3.1 Syntheses Trinuclear [ Mn3O(O2CR)6(py)3]0,+ clusters have proven to be ve ry useful starting materials for the synthesis of higher nuclearity products, affording complexes of nuclearity 4 18. For example, reaction of [Mn3O(O2CMe)6(py)3]+ with 2,2' bipyridine (bpy)36 or picolinate (pic-)37 gave [Mn4O2(O2CMe)7(bpy)2]+ and [Mn4O2(O2CMe)7(pic)2]salts, respectively. In addition, the reaction of a mixture of [Mn3O(O2CMe)6(py)3]+ and [Mn3O(O2CMe)6(py)3]py with 2(hydroxyethyl)pyridine (hepH) gave [Mn18O14(O2CMe)18(hep)4(hepH)2(H2O)2]2+.38 However, a chelating reagent is not always necessary: Treatment of [Mn3O(O2CPh)6(py)2(H2O)] with phenol gives [Mn6O2(O2CPh)10(py)2(MeCN)2], the phenol merely acting as a reducing agent and trigerring dimerization.47 Thus, the choice of chelating or other co-reagent and the reaction conditions have significant eff ect not only on the nuclearity of the product but also on its metal topology. Along the same lines, the reaction of [Mn3O(O2CMe)6(py)3]py with salproH3 and NEt3 in a roughly 3:2:6 molar ratio in CH2Cl2:MeOH gave the novel tetranuclear complex [Mn4O2(O2CMe)5(salpro)]( 2-1 ). Its formation can be summari zed in eq. 2-1.The reaction is sensitive to the Mn3:salproH3 ratio, and complex 2-1 is obtained only when 0.5 0.7 equivalents of salproH3 per Mn3 is employed. We also found that the mixed CH2Cl2:MeOH solvent system is very important; no reaction was observed when the reaction was performed in CH2Cl2 alone, and only starting material was rec overed. Presumably, the more polar MeOH facilitates the necessary proton transfer steps. However, the yield of th e product decreases as the concentration of MeOH increases beyond that described in the Experimental Section, presumab ly due to the solubility of the product. Thus, a controlled am ount of MeOH is essential for a high yield reaction. However, the same product was obtained using a CH2Cl2:EtOH solvent system. Further investigation

PAGE 40

40 showed that the same complex 2-1 was also obtained from MeCN:MeOH. Also the same product is obtained in the absence of base but in lower yield. 2 [Mn3O(O2CMe)6(py)3] + salproH3 [Mn4O2(O2CMe)5(salpro)] + 7MeCO2 + 2Mn2+ + 6py + 3H+ (2-1) Since complex 2-1 contains 4 MnIII ions, we wondered if a hi gher oxidation state product might form if we employed a higher oxidation state reagent, and thus explored the reaction of salproH3 with [Mn12O12(O2CMe)16(H2O)4], which contains 8MnIII,4MnIV. Thus, an MeCN:MeOH solution of [Mn12O12(O2CMe)16(H2O)4] was treated with 4 equivalents of salproH3, but the same product, complex 2-1 was again obtained, but only in poor yield (16%). We also obtained complex 2-1 when the [Mn3O(O2CMe)6(py)3]py starting material was replaced with Mn(O2CMe)3H2O. This MnIII acetate is really a polymer of Mn3 trinuclear units similar to those in [Mn3O(O2CMe)6(py)3]py, and so it was perhaps not surprising that its reaction with salproH3 gave the same product. Access to other carboxyl ate derivatives of 2-1 is possible using the described procedure of Method A. Thus, the reaction usi ng the R = Et (propionate) or But (pivalate) derivatives of [Mn3O(O2CR)6(py)3] gave the corresponding [Mn4O2(O2CR)5(salpro)] complexes 2-2 and 2-3 respectively. This provides access to more soluble versions of this new structural type of Mn4 complex. However, all our attempts to prepare benzoate derivative of 2-1 resulted in the formation of the monomeric complex NMe4[Mn(O2CPh)2(salproH)] ( 2-4 ). 2.3.2 Description of Structures 2.3.2.1 Structure of [Mn4O2(O2CR)5(salpro)] (R = Me (2-1), But (2-3)) The structures of 2-1 and 2-3 are shown in Figure 2-2. Select ed interatomic distances and angles for 2-1 are listed in Table A-1. Complex 2-1 crystallizes in monoclinic space group P 21/n. It contains a [Mn4O2]8+ core with peripheral ligation pr ovided by five doublybridging acetate

PAGE 41

41 groups and one pentadentate salpro3ligand (Figure 2-2). The core can be described as derived from two triangular, oxide-bridged [Mn3O] units sharing an edge and thus giving a [Mn4O2] butterfly-like core as found in several other Mn4 complexes (as well as with other transition metals), for example, [Mn4O2(O2CMe)7(bpy)2]+ (Mn4-bpy).36 Atoms Mn1 and Mn3 occupy the body positions of the butterfly and are five-c oordinate (square pyramid al), and Mn2 and Mn4 occupy the wingtip positions and are six-co ordinate (octahedral). However, the [Mn4O2] in 2-1 is much more closed up, i.e. a more acute V-sh ape, than normally found in such butterfly units (Figure 2-3) This is reflected in the dihedral angle between the two Mn3 planes, which is 79.2 in 2-1 compared with 134.3 in Mn4-bpy,36 which is typical of previous butterfly complexes. This can clearly be assigned to the fact that the wingtip Mn atoms of the butterfly topology, Mn2 and Mn4, are mono-atomically bridged by the salpro3oxygen atom O1 (Figure 2-2, top). In fact, this drastic closing up of the bu tterfly makes the core of 2-1 intermediate between a butterfly and a cubane (i.e. tetrahedral) Mn4 topology. There is a carboxylate group bridging each body-wingtip Mn pair, and a fifth carboxylate bridges the body-body Mn pair. The pentadentate salpro3ligand completes the peripheral ligation, chelating each wingtip Mn atom and bridging them via O1. All the Mn atoms are in the +3 oxidation state. This was established by qualitative consideration of the bond distances at each M n, and confirmed quantitatively by bond valence sum (BVS) calculations (Table 2-2).48,49 This also agreed with charge considerations and the overall neutrality of the molecule as well as the clear presence of a Jahn-Teller (JT) distortion at near-octahedral Mn2 and Mn4, the JT axes being along the O9, O7 and O15,O13 vectors, respectively. Complex 2-3 crystallizes in monoclinic space group P 21/n. Selected interatomic distances and angles are listed in Table A-2. Complex 2-3 is isostructural with complex 2-1 except for the

PAGE 42

42 difference in the carboxylate R groups. In par ticular, the dihedral angle between the two Mn3 planes is 79.8, and the MnIII JT axes have the same relative orientation. The bulky But groups thus have only a minimal effect on the structur e, as expected from the lack of any steric interactions. As discussed above, the structures of 2-1 and 2-3 can be considered cl osed-up versions of the familiar butterfly structures observed on severa l previous occasions in MnIII chemistry. It is thus of interest to structurally compare the two types, and this is done in Table 2-3. The metric parameters are fairly similar, as expected given that they are all MnIII species, but some overall conclusions can nevertheless be drawn. The closing up of the core of 2-1 and 2-3 which is effectively a pivoting of the wingtip Mn atoms (Mnw) about the 3-O2ions, has the effect of greatly decreasing the Mnw-O-Mnb as expected (by ~10-15), but also slightly decreasing the Mnb-O-Mnb angles (by ~1-2) as the central [Mn2O2] rhombus buckles into a non-planar conformation. These angle changes are also refl ected in the MnMn separations, which all decrease by ~ 0.1-0.2 except for the MnwMnw separation which is much shorter in 2-1 and 2-3 2.3.2.2 Structure of NMe4[Mn(O2CPh)2(salproH)] (2-4) Complex 2-4 crystallizes in monoclinic space group P 21/c. Selected interatomic distances and angles are listed in Table A-3. The asymmetr ic unit contains two monomers, the structures of which are essentially superimposable, therefore structure of one of these will be discussed here. It contains a MnIII ion in octahedral geometry as shown in Figure 2-4. The four equatorial sites are occupied by two N and two alkoxide oxygen atoms of salproH molecule and two 1 benzoate groups disposed trans to each other along the axial dir ection forming the JT axis. The protonation level of the central alcohol arm (O7) of salproH wa s established by a BVS of 0.9.

PAGE 43

43 O7 also forms an intramolecular hydrogen bond w ith the O3 of the benzoate anion; O7O3 = 2.65 2.3.3 Magnetochemistry Variable temperature dc m agnetic susceptibi lity data were collected in the 5.0 300 K range in a 0.1 T magnetic field on powd ered microcrystalline samples of 2-1 CH2Cl2, 2-2 CH2Cl2, 2-3 CH2Cl2 and 2-4 CH2Cl2 restrained in eicosane to prevent torquing. The MT vs T for 2-1 to 2-3 are shown in Figure 2-5. For 2-1 CH2Cl2, MT smoothly decreases from 9.1 cm3Kmol-1 at 300 K to 0.4 cm3Kmol-1 at 5.0 K. The 300 K value is much less than the spin-only value of 12.0 cm3Kmol-1 ( g = 2.0) expected for four MnIII ions with non-interacting metal centers, indicating the presence of appreciable intramolecular antiferromagnetic interactions between the Mn ions, with the low temperature data suggesting a spin S = 0 ground state. Similar data were obtained for 2-2 CH2Cl2 and 2-3 CH2Cl2, consistent with their isostructural nature and a minimal influence of the different li gands in the three complexes (Figure 2-5). The isotropic Heisenberg-Dirac-VanVleck (HDVV) spin Hamiltonian describing the exchange interactions within these Mn4 complexes with virtual C2V symmetry is given by eq. 22, where b = body, w = wingtip, i (i = 1 4) is the spin operator for metal atom Mni, and J is the exchange parameter. The exchange and atom labeling are summarized in the Figure 2-6. H = 2Jbb1 3 2Jbw( 1 2 + 1 4 + 2 3 + 3 4) 2Jww 24 (2-2) The eigenvalues of the spin Hamiltonian of eq. 2-2 can be determined analytically using the Kambe vector coupling method,50 as described elsewhere for the more common butterfly complexes such as Mn4-bpy, which also have C2V symmetry.36 Thus, use of the coupling scheme A = 1 + 3, B = 2 + 4, and T = A + B allows the spin Hamiltonian to be transformed into the equivalent form given by eq. 2-3, where ST is the total spin of the molecule. The eigenvalues of eq. 2-3 can be determined using the relationship i 2 = Si(Si+1) and are given in eq. 2-4,

PAGE 44

44 where E ST, SA, SB> is the energy of state ST, SA, SB>, and constant terms contributing equally to all states have been omitted. The overall multip licity of the spin system is 625, made up of 85 individual spin states ranging from ST = 0 8 H = Jbb( A 2 1 2 3 2) Jbw( T 2 A 2 B 2) Jww( B 2 2 2 4 2) (2-3) E ST, SA, SB> = Jbb[SA(SA+1)] Jbw[ST(ST+1) SA(SA+1) SB(SB+1)] Jww[SB(SB+1)] (2-4) An expression for the molar paramagnetic susceptibility, M, was derived using the above and the Van Vleck equation,51 and assuming an isotropic g tens or (Appendix D-1). This equation was then used to fit the experimental MT vs T data in Figure 2-5 as a function of the three exchange parameters Jbb, Jbw and Jww and the g factor. A contribution from temperature independent paramagnetism (TIP) was held constant at 400 x 10-6 cm3mol-1. The obtained fits are shown as the solid lines in Figure 2-5. The fitting parameters were: For 2-1 Jbb = -6.37 cm-1, Jbw = -5.72 cm-1, Jww = -1.78 cm-1 and g = 2.00; for 2-2 Jbb = -7.64 cm-1, Jbw = -6.73 cm-1, Jww = 2.49 cm-1 and g = 2.00; and for 2-3 Jbb = -7.37 cm-1, Jbw = -6.57 cm-1, Jww = -1.79 cm-1 and g = 1.99. The obtained values indicate that the ground state of the molecules is ST, SA, SB> = 0, 4, 4>, as anticipated from the low temperature data in Figure 2-5. The exchange interactions within the Mn4 cores of 2-1 to 2-3 are thus all antiferromagnetic and weak. The weakest is the Jww between the wingtip MnIII ions, but note that this is nevertheless a significant interaction relative to the others, unlike the more common types of butterfly species where this Jww interaction is not a major contri butor since the wingtip Mn atoms are not directly (monoatomically) bridged. Sinc e the wingtip Mn atoms are bridged by an alkoxide O atom whereas the other Mn pairs are a ll bridged by either one or two oxide O atoms, it is qualitatively reasonable for Jww to be the weakest interaction in the molecule, although the

PAGE 45

45 precise values of all the J pa rameters are the net sum of contributions from ferroand antiferromagnetic pathways and thus it is difficult to rationalize their differences precisely. It is, however, of interest to compare the exchange parameters for 2-1 to 2-3 with those for the the more common type of MnIII butterfly complexes and see if any observed differences can be correlated with the structural differences me ntioned earlier. In Table 2-4 are compared the exchange parameters for 2-1 to 2-3 with those for [Mn4O2(O2CMe)7(bpy)2]+ (Mn4-bpy),36 [Mn4O2(O2CMe)7(pic)2](Mn4-pic)37 and [Mn4O2(O2CEt)7(bpya)2]+ (Mn4-bpya).52 Although the Jbw interaction in 2-1 to 2-3 is within the rang e found for the previous complexes, the Jbb interaction in the former is distinctly weaker th an in the latter. This is consistent with the significantly more acute angles at the 3-O2ions, since these will presumably weaken the antiferromagnetic cont ributions to the Jbb interaction by weakening the Mn(d )-O(p )-Mn(d ) overlap that would be stronger wh en mediated by an essentially trigonal planar O atom as in Mn4-bpy, Mn4-pic and Mn4-bpya. The buckling of the central [Mn2O2] no doubt also contributes to the change in Jbb by affecting the orbital overlap. The fact that Jbw Jbb in 2-1 to 2-3 is expected to have a cl ear impact on the ground state because the butterfly [Mn4O2]8+ core in Mn4-bpy, Mn4-pic and Mn4-bpya has been well established from previous work to experience spin frustration effects as a result of the presence of triangular Mn3 within its structure.36,37,52 Since the interactions are all antiferromagnetic, they are competing and the precise ground state spin alignment is thus very sensitive to the Jbw:Jbb ratio, with Jww not being a factor in Mn4-bpy, Mn4-pic and Mn4-bpya because of its weakness. For example, the typical butterfly complexes Mn4-bpy and Mn4-pic have an ST = 3 ground state spin, the ST, SA, SB> = 3, 1, 4> state, which results from the dominating Jbb interaction aligning the Mnb spins almost perfectly antipa rallel, but not quite (i.e. SA = 1 not 0). The Jbw

PAGE 46

46 interactions are indivi dually weaker than Jbb, but there are four of them, and as a result prevent SA being 0. An intermediate resultant spin is thus obtained in the ground state. In Mn4-bpya, Jbb >> Jwb, and the Mnb spins are now aligned antiparallel, i.e. SA = 0, with the weak Jww serving to couple the Mnw spins antiparallel and giving an overall ST = 0 ground state, the ST, SA, SB> = 0, 0, 0> state. The ground state of 2-1 to 2-3 can now be satisfactorily rationalized within this description. In fact, it represents the situation at the other extreme compared to Mn4-bpya, i.e. the Jbb is now weakened relative to Jbw and the two interactions are comparable in magnitude. However, there are four of the latter, and thus Jbw now dominates the spin alignments, aligning the spins antiparallel to their neighbors along the outer edges of the [Mn4O2]8+ butterfly. The Jbb interaction is antiferromagne tic but nevertheless complete ly frustrated, as is Jww, with the two Mnb spins and the two Mnw spins both being aligned parallel by the Jbw interactions. Thus, the ground state is again ST = 0, as in Mn4-bpya, but now it is the ST, SA, SB> = 0, 4, 4> state as depicted in Figure 2-6. Table 2-5 calculates the spin states of 2-1 CH2Cl2in the |ST, SA, SB> format arranged as a function of their ener gy calculated using the calculated exchange parameters, Jbb, Jbw and Jww, and the Van Vleck equation The MT vs T for 2-4 is shown in Figure 2-7. For 2-4 CH2Cl2, the value of MT decreases very smoothly from 3.19 cm3Kmol-1 at 300 K to 2.57 cm3Kmol-1 at 15.0 K and then drops sharply to 2.21 cm3Kmol-1 at 5.0 K. Complexes 2-4 exhibits behavior expected for high-spin MnIII ( S = 2) center exhibiting zero-field splitting (ZFS). To characterize the ZFS parameter D further, magnetization vs field data were collect ed in the 2.0 10.0 K range in fields of 0.1 7 T (Figure 2-7). Fitting of th e data using the program MAGNET,53 described elsewhere,36,54-56 that involves diagnolization of the sp in Hamiltonian matrix, assuming that only the ground state spin is populated at these temperatures includes axial ZFS and Zeeman interactions and carries out a

PAGE 47

47 full powder average, gave S = 2, g = 1.86, and D = -4.09 cm-1. To ensure that the global minimum had been located, we calculated the ro ot-mean-square error surface for the fit as a function of D and g (Figure 2-8). The plot cl early shows only the above mentioned minima. 2.4 Conclusions SalproH3 has proved an effective route to a novel type of tetranuclear Mn complex whose core can be described as a more closed up versio n of the butterfly-like core that is relatively common. Three isostructural complexes of this new family have been synthesized and characterized. These complexes also complement and extend the currently rich area of MnIII schiff-base species. Complexes 2-1 to 2-3 extend the type of spin frustration effects observed within the Mn4 butterfly core, giving an ST = 0 ground state due to domination of the spin alignments by Jbw. Table 2-1. Crystallographic Data for 2-1MeCN, 2-3MeOH2CH2Cl2C7H16 and 2-4 CH2Cl2 2-1 2-3 2-4 Formulaa C29H33Mn4N3O15 C52H84Cl4Mn4N2O16 C36H40Cl2MnN3 O7 Fw, g/mola 883.34 1354.77 752.55 Space group P 21/ n P 21/ n P 21/ c a, 9.3368(6) 17.7518(13) 24.2193(12) b, 22.5058(15) 17.3654(12) 17.4454(9) c 16.5079(11) 21.4029(15) 18.3088(9) 90 90 90 90.945(1) 100.856(1) 110.569(1) 90 90 90 V 3 3468.4(4) 6479.7(8) 7242.6(6) Z 4 4 8 T K 173(2) 173(2) 173(2) b 0.71073 0.71073 0.71073 calc, g/cm3 1.692 1.389 1.380 mm-1 1.097 0.988 0.563 R1 c,d 0.0413 0.0473 0.0458 wR2 e 0.0828 0.1220 0.0844 a Including solvate molecules. b Graphite monochromator. c I > 2 ( I ). d R 1 = (|| Fo| | Fc||) / | Fo|. e wR 2 = [ [ w ( Fo 2 Fc 2)2] / [ w ( Fo 2)2]]1/2, w = 1/[ 2( Fo 2) + [( ap)2 + bp ], where p = [max ( Fo 2, O) + 2 Fc 2]/3.

PAGE 48

48 Table 2-2. Bond-valence sums for the Mn atoms of complexes 2-1 2-3 and 2-4 a 2-1 MeCN 2-3MeOH2CH2Cl2C7H16 2-4 CH2Cl2 Atom MnII MnIII MnIV MnII MnIII MnIV MnII MnIII MnIV Mn1 3.07 2.81 2.95 2.98 2.73 2.87 3.29 2.84 3.12 Mn2 3.28 3.04 3.13 3.30 3.06 3.14 3.30 2.88 3.14 Mn3 3.04 2.78 2.92 3.33 3.26 3.18 Mn4 3.23 3.16 3.08 3.05 2.79 2.93 a The underlined value is th e one closest to the charge for which it was cal culated. The oxidation state of a particular atom can be taken as the nearest w hole number to the underlined value Table 2-3. Comparison of core parameters of selected [Mn4O2]8+ complexes (, ) ComplexMnbMnb MnbMnwMnwMnwMnb-O Mnw-O Mnb-OMnb Mnb-OMnw Mnw-OMnw Re f 2-1 2.770 3.1713.196 3.651 1.8861.906 1.909, 1.914a1.977, 1.991 94.30, 93.87 112.7114.5 133.8843 2-3 2.794 3.1363.193 3.640 1.8871.910 1.907, 1.910a1.970, 1.984 94.97, 94.74 110.6114.6 134.1 43 Mn4-bpy 2.848 3.2993.385 5.593 1.8891.930 1.804, 1.84495.7, 96.8 123.3131.3 36 Mn4-pic 2.842 3.3083.406 1.8881.910 1.840, 1.84796.9 123.2129.7 37 Mn4bpya 2.871 3.3073.344 1.8731.957 1.833, 1.838 97.07, 97.25 125.3131.4 52 a Top and bottom entries refer to distances to oxide and alkoxide O atoms, respectively. Table 2-4. Comparison of exchange parameters in [Mn4O2]8+ complexes Complex Jbb a Jbw a Jww a g Ref 2-1 -6.37 -5.72 -1.78 2.00 43 2-2 -7.64 -6.73 -2.49 2.00 43 2-3 -7.37 -6.57 -1.79 1.99 43 Mn4-bpy -23.5 -7.8 2.0 36 Mn4-pic -24.6 -5.3 1.96 37 Mn4-bpya -25.7 -3.3 -0.77 1.99 52 a cm-1

PAGE 49

49 Table 2-5. Spin states of 2-1 CH2Cl2 in the |ST, SA, SB> format arranged as a function of their energy calculated using the calculated exchange parameters, Jbb, Jbw and Jww, and the Van Vleck equation |ST,SA,SB> E (cm-1) |ST,SA,SB> E (cm-1) |ST,SA,SB> E (cm-1) |0,4,4> |1,4,4> |1.3.4> |0,3,3> |2,4,4> |2,3,4> |2,2,4> |1,4,3> |1,3,3> |1,2,3> |0,2,2> |2,4,3> |1,3,2> |2,3,3> |1,2,2> |1,2,2> |2,2,3> |2,1,3> |3,4,4> |0,1,1> |3,3,4> |3,2,4> |3,1,4> |0,0,0> |1,2,1> |1,1,1> |1,0,1> |2,4,2> |2,3,2> 0.00 12.40 14.00 33.20 37.20 38.80 40.00 44.00 45.60 46.80 58.10 68.80 69.30 70.40 70.50 71.30 71.60 72.40 74.40 74.70 76.00 77.20 78.00 83.00 86.30 87.10 87.50 92.50 94.10 |1,1,0> |2,2,2> |2,1,2> |2,0,2> |3,4,3> |3,3,3> |3,2,3> |3,1,3> |2,3,1> |3,0,3> |2,2,1> |2,1,1> |2,2,0> |4,4,4> |4,3,4> |4,2,4> |4,1,4> |4,0,4> |3,4,2> |3,3,2> |3,2,2> |3,1,2> |3,4,1> |3,3,1> |3,2,1> |3,3,0> |4,4,3> |4,3,3> |4,2,3> 95.00 95.30 96.10 96.50 106.00 107.60 108.80 109.60 109.90 110.00 111.10 111.90 119.00 124.00 125.60 126.80 127.60 128.00 129.70 131.30 132.50 133.30 145.50 147.10 148.30 155.00 155.60 157.20 158.40 |4,1,3> |4,4,2> |4,3,2> |4,2,2> |5,4,4> |5,3,4> |5,2,4> |5,1,4> |4,4,1> |4,3,1> |4,4,0> |5,4,3> |5,3,3> |5,2,3> |5,4,2> |5,3,2> |5,4,1> |6,4,4> |6,3,4> |6,2,4> |6,4,3> |6,3,3> |6,4,2> |7,4,4> |7,3,4> |7,4,3> |8,4,4> 159.20 179.30 180.90 182.10 186.00 187.60 188.80 189.60 195.10 196.70 203.00 217.60 219.20 220.40 241.30 242.90 257.10 260.40 262.00 263.20 292.00 293.60 315.70 347.20 348.80 378.80 446.40

PAGE 50

50 Figure 2-1. Structure of SalproH3 Figure 2-2. Labeled representa tion of the structure of 2-1 (top) and 2-3 (bottom). Hydrogen atoms have been omitted for clarity. JT axis on MnIII are shown in black. Color code: MnIII, green; O, red; N, blue; C, grey.

PAGE 51

51 Figure 2-3. Comparison of the cores of 2-1 and 2-3 (left) with that of the normal butterfly complexes (right). Color code: MnIII, green; O, red. Figure 2-4. Labeled representa tion of the structure of 2-4 Hydrogen atoms have been omitted for clarity. JT axis on MnIII are shown in black. Color code: MnIII, green; O, red; N, blue; C, grey.

PAGE 52

52 Figure 2-5. Plots of MT vs T for complexes 2-1 CH2Cl2 (top), 2-2 CH2Cl2 (middle) and 23CH2Cl2 (bottom). The solid line is the fit of the data; see the text for the fit parameters.

PAGE 53

53 Figure 2-6. (left) The core of 2-1 defining the pairwise exch ange interac tions. (right) Rationalization of ground state spin of 2-1 Figure 2-7. (left) Plots of MT vs. T for complex 2-4 CH2Cl2. (right) Plot of reduced magnetization (M / N B) vs H / T for complex 2-4 CH2Cl2. The solid lines are the fit of the data; see the text for the fit parameters. Figure 2-8. Two-dimensional contour plot of th e fitting-error surface vs D and g for complex 24CH2Cl2.

PAGE 54

54 CHAPTER 3 ROLE OF MIXED-LIGAND AND MIXEDSOLVENT SYSTEM: ROUTE TO Mn4 AND Mn25 3.1 Introduction For a num ber of years, we have been engage d in developing manganese cluster chemistry with oxide bridges. The motivation for this is the relevance of manganese in bio-inorganic chemistry and molecular magnetism, not least their aesthetic qualities.27,35,57 In the field of molecular magnetism, single-molecule magnets (SMMs) hold great current interest. SMMs are molecular species that can function as nanoscale magnets as a result of their intrinsic properties rather than as a result of inte r-unit interactions and long-rang e ordering as would be found in traditional magnetic materials (metals, metal oxides, etc).13 Thus each SMM is a single-domain magnetic particle, and this arises from th e combination of a large ground state spin ( S ) and an Ising type (easy-axis) magnetoanisotropy (-ve zero field splitting parameter, D ).13 As part of the search of new SMMs, it is important to build high spin and high anisotropy molecules. It is not possible to achieve a rational synthesis of a molecule with a high D but there are certain approaches to high spin molecules.22 One of them being replacement of hydroxides by end-on azides and second is the use of lig ands which generally promote ferromagnetic interactions.58-60 More recently, our group ha s demonstrated spin tweak ing, the conversion of an already high-spin Mn25 SMM with S = 51/2 into a structurally similar one with S = 61/2 by modification of the peripheral ligan ds and conversion of a low-spin Mn3O triangular cluster to a high spin Mn3 cluster by the use of appropriate chelate.34,61 We and others have extensively investigat ed chelating agent 2hydroxymethyl pyridine (hmpH) and 2,6-pyridinedimethanol (pdmH2) in manganese cluster chemistry and these have proven to be very useful in affording high nuc learity and high spin complexes in Mn cluster chemistry.34,62,63 Examples of these include Mn10 ( S = 22)63 and Mn25 ( S = 51/2, 61/2).34,62 We

PAGE 55

55 have been continuing our investigation into the usefulness of the mixed-chelate system and therefore we decided to us e mixture of hmpH and pdmH2, which have already been proved to be very promising in high nuclearity Mn cluster chemistry, with [Mn3O(O2CMe)6(py)3]+ complexes. As mentioned in Chapter 2, Mn3O trinuclear complexes have prove n to be extremely useful in producing polynuclear clusters with high gr ound state spin and possible SMM behavior.64-67 Indeed, as will be described below, reacti on of mixed chelate sy stem (hmpH and pdmH2) with [Mn3O(O2CMe)6(py)3]+ has led to a ferromagnetically coupled Mn4 cluster and a Mn25 cluster with a ground state spin of 65/2. 3.2 Experimental Section 3.2.1 Syntheses All m anipulations were performe d under aerobic conditions and all chemicals were used as received unless otherwise noted. [Mn3O(O2CMe)6(py)3]ClO4 was synthesized as reported elsewhere.45 [Mn4(hmp)4(pdmH)2(MeCN)4](ClO4)4 (3-1). Method A. To a stirred solution of hmpH (0.05 mL, 0.50 mmol) and pdmH2(0.033 g, 0.25 mmol) in MeCN (15 mL) was added MeCO2Na (0.08 g, 1.0 mmol) and Mn(ClO4)2 ( 0.40 g, 1.1 mmol) followed by the addition of NEt3 (0.07 mL, 0.5 mmol). The resulting reddish-brown solution was stirred for two hours and then filtered to remove any undissolved solid and then left fo r slow evaporation. X-ra y quality crystals of 3-1 grew over a period of two weeks in 20% yield. Th ese were collected by filteration, washed with MeCN and dried in vacuo Anal. Calcd (Found) for 3-1 (C46H52N10Mn4O24Cl4): C, 37.07 (36.85); H, 3.52 (3.46); N 9.40 (9.07)%. IR ( cm-1): 1607m, 1578w, 1440m, 1369w, 1281w, 1145s, 1088s, 768w, 720w, 673w, 626m, 575w. Method B. To a stirred solution of hm pH (0.02 mL, 0.21 mmol) and pdmH2 (0.06 g, 0.45 mmol) in MeCN (15 mL) was added [Mn3O(O2CMe)6(py)3]ClO4 (0.22 g, 0.25 mmol). The

PAGE 56

56 resulting solution was stirred for 30 minutes and then filtered and left for slow evaporaton. Reddish brown crystals of 3-1 were obtained after 15 days in 5% yield. The product was identifield as 3-1 by IR spectral comparison wi th material from method A. [Mn25O18(OH)2(hmp)6(pdm)8(pdmH)2(L)2](ClO4)6 ( 3-2 ). To a stirred solution of [Mn3O(O2CMe)6Py3]ClO4 (0.20 g, 0.23 mmol) in MeOH/MeCN (5/15 mL) was added hmpH (0.02 mL, 0.23 mmol) and pdmH2 (0.06 g, 0.46 mmol). This was followed by the addition of NEt3 (0.03 mL, 0.23 mmol). The resulting reddish brown solution was left under magnetic stirring for 30 minutes and then filtered through a medium frit. Th e resulting brown filterate was left undisturbed to evaporate slowly at room temperature. The X-ray quality crystals of 3-2 MeCNMeOH were obtained over a period of two weeks in 10% yield. These were collected by filteration, washed w ith MeCN and MeOH and dried in vacuo. Anal. Calcd (Found) for 3-2 H2O (C120H128N18Mn25O79Cl6): C, 30.85 (30.70); H, 2.76 (2.79); N 5.39 (5.05)%. IR ( cm-1): 1600s, 1577s, 1440s, 1388m, 1276w, 1102s, 1074s, 774w, 676m, 635s, 559m, 503w, 438w. 3.2.2 X-ray Crystallography Data were collected by Dr. Khalil A A bboud on a Siemens SMART PLATFORM equipped with a CCD area detector and a grap hite monochromator utilizing MoK radiation ( = 0.71073 ). Suitable crystals of 3-1 and 3-2 MeCNMeOH were attached to glass fibres using silicone grease and transferred to a goniostat where they were cooled to 173 K for data collection. Cell parameters were refined using up to 8192 reflections. A full sphe re of data (1850 frames) was collected using the -scan method (0.3 frame width). The first 50 frames were re-measured at the end of data collection to monitor instrument and crystal stability (maximum correction on I was < 1 %). Absorption corrections by integrat ion were applied based on measured indexed crystal faces. The structure was so lved by the Direct Methods in SHELXTL6 ,46 and refined using

PAGE 57

57 full-matrix least squares. The non-H atoms were treated anisotropicall y, whereas the hydrogen atoms were calculated in ideal positions and were riding on their respective carbon atoms. Refinement was done using F2. In 3-1 the asymmetric unit consists of a half Mn4 cluster and two perc hlorate anions. A total of 403 parameters were refined in the fina l cycle of refinement using 10860 reflections with I > 2 ( I ) to yield R1 and wR2 of 4.97% and 11.82%, respectively. In 3-2 MeCNMeOH, the asymmetric unit consists of a half Mn25 cluster, three perchlorate anions, four acetonitrile molecules and two methanol molecules. All solvent molecules and the anions were disordered and could not be modeled properly, thus program SQUEEZE,68 a part of the PLATON package of crysta llographic software, was used to calculate the solvent disorder area and remove its contributi on to the overall intensity data. In particular, the perchlorate anions were extensively disorder ed and each was refined in three parts before their contributions were removed from the intens ity data by SQUEEZE. The cluster also has a couple of disorders. Two of the pdmH2 ligands seem to have been oxidized during synthesis to produce, in each case, a carboxylat e instead of one of the hydroxyl group. Both parts of each disorder were fixed at 50% occupancy after the re finement produced value cl ose to 50%. A total of 979 parameters were refined in the final cycle of refinement using 55761 reflections with I > 2 ( I ) to yield R1 and wR2 of 6.70% and 15.77%, respectively. The crystallographic data and structure refinement details for 3-1 and 3-2 MeCNMeOH are listed in Table 3-1. 3.3 Results and Discussion 3.3.1 Syntheses The reaction of prefor med trinuclear complexes [Mn3O(O2CMe)6(py)3]0,+ with chelating ligands has been used extensively in the past to trigger structural rearrangements leading to higher nuclearity products. The choice of chelate and the reaction conditions have significant

PAGE 58

58 effect not only on the nuclearity of the product but also on its metal topology. Along the same lines, the reaction of [Mn3O(O2CMe)6Py3]ClO4 with hmpH and pdmH2 in 1:1:1 molar ratio in MeCN gave [Mn4(hmp)4(pdmH)2(MeCN)4]4+ ( 3-1 ). Its formation is summarized in eq. 3-1. The yield of the reaction is extremel y low and the attempts to increa se the yield gave a mixture of products and were not further pu rsued. Instead, we sought its synthesis from a more convenient procedure that doesnt employ preformed triangul ar complex but only simple starting materials. This was successfully accomplished from the reaction of Mn(ClO4)2, hmpH and pdmH2 in the presence of MeCO2Na. Use of sodium acetate is found to be essential in this reaction, it does not appear in the final product but presumably acts as a base. 2[Mn3O(O2CMe)6Py3]+ + 4hmpH + 2pdmH2 + 4MeCN + 2e[Mn4(hmp)4(pdmH)2(MeCN)4]4+ + 2Mn3+ + 12MeCO2 + 6Py +2H2O + 2H+ (3-1) On the other hand, 3-2 MeCNMeOH is obtaine d upon the reaction of [Mn3O(O2CMe)6Py3]ClO4 with hmpH and pdmH2 in MeCN/MeOH system. This reaction is very sensitive to the ratio of Mn3:hmpH:pdmH2 and the clean product is obtained only when ratio mentioned in the experimental section is use d. Using higher concentrat ion of hmpH or pdmH2 beyond that described in the expe rimental section gives a mixtur e or orange and red crystals. And the presence of MeOH is absolutely essential for the formation of product. Presumably, the more polar MeOH facilitates the necessary pr oton transfer steps. In addition, 3:1 v/v MeCN:MeOH is found to be the best for clean reaction. However, the yield of the product decreases as the concentration of MeOH increases most likely due to the solubility of the product, while reducing the amount of MeOH gi ves white crystals along with the desired product. Also, the same product is obtained in th e absence of base but in extremely low yield.

PAGE 59

59 One remarkable feature of this reaction is the in situ formation of the oxidized pdmH2 ligand. In 3-2 MeCNMeOH, oxidation of the chelate provides the reducing equivalents to lower the metal oxidatio n state of the product. The reduction of the MnIII centers to MnII is associated with oxidation of some of the pdmH2, possibly to the corr esponding aldehyde, 6(hydroxymethyl)-2-pyridine carboxaldehyde fo llowed by the further oxidation to 6(hydroxymethyl)-2-pyridine carboxylic acid (L, Fi gure 3-1). This has been seen before with hmpH69 and pdmH2 70 ligands in high-oxidation state Mn chemistry. The redox chemistry is also accompanied by the fragmentation and structural rearrangement of the Mn3 cluster to yield Mn25. Also filterate after the isolation of the product is still dark colored, thus there is a possibility that disproportionation of MnIII is occurring during the reaction and filterate contains both MnII and MnIV ions. For both compounds, the final yields are lo w, nevertheless the preparations are reproducible. As is usually the case for such react ions, several species in equilibrium are likely present in solution, and the low solubility of one of them is undoubtedly the reason a pure product can be obtained. 3.3.2 Description of Structures 3.3.2.1 Structure of [Mn4(hmp)4(pdmH)2(MeCN)4](ClO4)4 (3-1) The labeled structure of complex 3-1 is shown in Figure 3-2 and selected interatomic distances and angles are lis ted in Table A-4. Complex 3-1 crystallizes in monoclinic space group P 21/c with the cation lying on an inversion center. The centrosymmetric structure of 3-1 can be described as a planar mixed-valence Mn4 rhombus consisting of 2MnII and 2MnIII ions. Mn1 and Mn2 are assigned as MnIII and MnII respectively on the basis of structural parameters and Bond Valence Sum (BVS)48,49 calculations that gave a value of 2.99 and 1.88 for Mn1 and Mn2 respectively and the presence at Mn1 of a Jahn Teller (JT) el ongation axis (O4-Mn1-N1), as

PAGE 60

60 expected for a high spin MnIII. The rhombus is composed of two Mn3 triangular faces each bridged by a 3oxygen, O4 from a bidentate chelating pdmHwhose other arm O3 is protonated as evident from the BVS of 1.09. Hmpgroups are also bidentate with O1 bridging Mn1 and Mn2 and O2 bridging Mn1 and Mn2 Peripheral ligation is provide d by four MeCN molecules at Mn2 and Mn2 ; which are seven coordinated, whereas Mn1 is distorted octahedral. Structure of complex 3-1 is similar to the Mn4 complexes seen with hmpand pdmHalone. 3.3.2.2 Structure of [Mn25O18(OH)2(hmp)6(pdm)8(pdmH)2(L)2](ClO4)6 (3-2) The structure of complex 3-2 is shown in Figure 3-3 and selected interatomic distances and angles are listed in Table A-5. Complex 3-2 crystallizes in triclinic space group P with the cation lying on an inversion center. It can be described as having a ba rrel-like cage structure consisting of 6MnII, 18MnIII and Mn1V ions. The 12 4-O2, 63-O2and 23-OHions hold the core together as well as chelating bridging hmpand pdmHand L groups, where L = 6hydroxymethyl-2-pyridine carboxylic acid (Figure 3-1). The manga nese oxidation states and protonation levels of OH, hmp-, pdm2-, pdmHand L oxygen atoms were established by Mn and O BVS calculations (Table 3-2 and 3-3), inspecti on of metric parameters and detection of MnIII JT elongation axes. The core can be dissected into five parallel layers of three types with an ABCBA arrangement (Figure 3-4). Layer A is a MnII 3 triangular unit (Mn1, Mn11 and Mn12) with a capping 3-OHions; layer B is a MnIII 6 triangle (Mn2, Mn3, Mn5, Mn7, Mn10 and Mn13) comprising three corner-sharing MnIII 3 triangles; and layer C is a MnIII 6 hexagon (Mn4, Mn6, Mn8 and their symmetry equivalents) with a central MnIV ion (Mn9). Layer C has the Anderson type structure seen before in some Mn complexes. Each layer is held together and linked to its neighboring layers by a combination of oxide, alkoxide bridges. The outer coordination shell is occupied by hmp, pdm2-, pdmHand L ligands. There are two types of

PAGE 61

61 MnIII ions; those in layer B are nearly octahedral while those in layer C are pentagonal bipyramidal. There are two Mn25 complexes reported in literature, [Mn25O18(OH)2(N3)12(pdm)6(pdmH)6](Cl)2 ( Mn25 +2) and [Mn25O18(OH)(OMe)(hmp)6(pdm)6(pdmH)6](N3)2(ClO4)6 ( Mn25 +8). All these Mn25 complexes have the same layered arrang ement of Mn ions, however the precise means by which layers are connected to each other are different. The main difference between Mn25 +2 with S = 51/2 and Mn25 +8 with S = 61/2 is that 12 bound azides in Mn25 +2 are replaced by six 1: 1: 2 hmpligands in Mn25 +8. As a result, all intraand inter layer bridges as well as all M2 pairwise exchange interactions are now through oxide anions. Complex 3-2 is very much similar to Mn25 +8 except for the arrangement of layer A, where one alcohol arm of the pdmHgroup has been oxidized and the three MnII ions are bridged by an OHinstead of OMein Mn25 +8 3.3.3 Magnetochemistry 3.3.3.1 Dc Studies Variable-temperatue dc m agnetic susceptibility measurements were performed on dried micro crystalline sample of 3-1 and 3-2 H2O at 0.1 T and 5.0-300 K range. The data for 3-1 is shown as MT vs T plot in Figure 3-5. For 3-1 the value of MT increases from 16.44 cm3Kmol-1 at 300 K to 35.3 cm3Kmol-1 at 5 K. The MT value at 300 K is higher th an the spin only value of 14.75 cm3Kmol-1 expected for two MnIII and two MnII ions suggesting predominant ferromagnetic interactions within the molecule. The value of MT at 5 K suggests S = 9 ground state spin. To determine the individual pair-wise exchange interactions Jij between MniMnj pairs within the molecule, the MT vs T data for complex 3-1 was fit to a simulation curve deduced from the Heisenberg-Dirac-VanVleck spin Hamiltonian given in eq. 3-2 where A = 1 + 3, B = 2 + 4, and T = A + B. The exchange and the atom labeling scheme are summarized in Figure 3-6.

PAGE 62

62 The energies of the resultant total spin states ST, which are eigen functions of the Hamiltonian in this coupling scheme, are given by eq. 3-3. The overall multiplicity of the spin system is 900, made up of 110 individual sp ins states ranging from ST = 1 to 9. An expression for the molar paramagnetic susceptibility was derived for this complex using the Van Vleck equation51 (see Appendix D-2, Refer Chapter 2 for details of vector coupling) H = Jwb( T 2 A 2 B 2) Jbb( A 2 1 2 3 2) (3-2) E|ST,SA,SB> = -Jwb[ST(ST+1)-SA(SA+1)-SB(SB+1)]-Jbb[SA(SA+1)] (3-3) Excellent fits were achieved when data below 10 K was omitted and TIP fixed to 6 X 10-4 cm3mol-1. The obtained fit is shown as solid lines in Figure 3-5 and was obtained with Jbb = 6.67 cm-1, Jbw = 0.36 cm-1 and g = 2.06. These values identify the ground state as |ST,SA,SB> = |9,4,5>. The values of Jbb, Jbw and g are similar to those reported in literature.54,65 To confirm the above ground state spin estima tes and to determine the magnitude of zero field splitting parameter ( D ), variable field (H ) and temperature-magnetization ( M ) data were collected in the 1.8-10 K and 2-6 T ranges. The resulting data for 3-1 are plotted in Figure 3-7 as reduced magnetization ( M / NB) vs H / T where N is Avogadros number and B is the Bohr magneton. The data were fit, using the program MAGNET,53 by diagonalization of the spin Hamiltonian matrix assuming that only the gr ound state is populate d, incorporating axial anisotropy ( D z 2) and Zeeman terms, and employing a full powder average.36,54-56 The corresponding spin Hamiltonian is given by eq 3-4, where z is the easy-axis spin operator, g is the Lande g factor, 0 is the vacuum permeability, and H is the applied field. The last term in eq 3-3 is the Zeeman energy associated with an applied magnetic field. The best fit for 3-1 is shown as the solid lines in Figur e 3-7 and was obtained with S = 9, g = 2.00 and D = -0.27 cm-1. H = D z 2 + gB0H (3-4)

PAGE 63

63 To ensure that the true global minimum had b een located and to ensure the hardness of the fit, root-mean-square error surface for the fit was generated using the program GRID ,71 which calculates the relative difference between the experimental M/NB data and those calculated for various combinations of D and g. This is shown as two-dimensi onal contour plot in Figure 3-7, the plot shows that fit minimum is a soft one, consistent with significant uncertainty in the precision of the obtained g and D fit values, which we estimate as 0.04 on g and 10 % on D. For 3-2 H2O, the value of MT steadily increases from 117.5 cm3Kmol-1 at 300 K to a maximum of 540.9 cm3Kmol-1 at 15 K before dropping to 489.4 cm3Kmol-1 at 5 K (Figure 3-9). The value of MT at 15 K suggests a very large ground-state spin ( S) value, with the sharp decrease at the lowest temperat ure, assigned to Zeeman effects, zero-field splitting, and/or weak intermolecular interactio ns. To determine the ground state, magnetization ( M ) data were collected in 0.1-0.8 T and 1.8-10.0 K ranges and are plotted as (M/N B) vs H/T in Figure 3-9. We used only the low-field data ( 0.8 T) to avoid problems with the low-lying excited states. The best fit (solid-lines in figure 3-9) gave S = 65/2, g = 1.99 and D =0.0082 cm-1. The fits for 63/2, 67/2 were also good, with best fit parameters of g =2.08/ D = -0.01 and g = 1.93/ D = -0.0077 cm-1 respectively. We conclude that 3-2H2O has a ground state of 65/2 1. 3.3.3.2 Ac Studies Ac susceptibility m easurements were performed on polycrystalline sample of 3-1 and 3-2 H2O under 3.5 G oscillating ac field and zero dc field as a function of temperature and frequency. The obtained data for 3-1 is shown as M'T vs T and M" vs T plots in Figure 3-8. Extrapolation of the M'T plot to 0 K from temperatures above ~ 3.5 K gives value of ~ 43 cm3Kmol-1, corresponding to S = 9 with g = 1.95 in very satisf ying agreement with the conclusions from the fits of dc magnetization data. Complex 3-1 shows frequency-dependent in-

PAGE 64

64 phase ( M' ) and out-of-phase ( M") signals, which are a signatu re of slow relaxation of magnetization i.e. SMM behavior by analogy to related [Mn4] compounds. The in-phase ac susceptibility ( M ) signal for 3-2 H2O is shown as M T vs T in Figure 310, and extrapolation of the M T signal to 0 K from above about 8 K (to avoid the effect of intermolecular interac tions at lower temperatures) gave 550 cm3Kmol-1, consistent with the dc data. The value of 550 cm3Kmol-1 is consistent with 1) S = 65/2 / g = 2.01, 2) S = 63/2 / g = 2.07 and 3) S = 67/2 / g = 1.95. The ac data thus confirms a ground state of 65/2 1. No out-of-phase signal was observed down to 1.8 K. The very small D value seen for 3-2 H2O is consistent with the nearly perpendicular disposition of the MnIII anisotropy axes. We have not done with microSQUID hysteresis measurements on single crystals of 3-2 but on the basis of previously reported Mn25 complexes ( Mn25 +2 and Mn25 +8), which show hysteresis loops and are SMMs, we believe that 3-2H2O is also an SMM, albeit at very low temperature. 3.4 Conclusions The com bination of mixed-ch elate system (hmpH and pdmH2) in Mn cluster chemistry has yielded a ferromagnetic coupled Mn4 rhombus and an unusually high spin Mn25 complex depending on the identity of the solvent empl oyed. The latter point emphasizes the exquisite sensitivity of the reaction product on reaction cond itions. Dc and ac studies have established that Mn25 possesses an S = 65/2 ground state spin. Such a high sp in value is extremely rare; in fact, it is the second highest in Mn chemistry. Achieving high spin ground state is one of the elusive goals in the search for obtaining superior SMMs. Once again manganese cluster chemistry continues to surprise and astound with the rema rkable variety and aesthetic beauty of its molecular offspring.

PAGE 65

65 Table 3-1. Crystallographic Data for 3-1 and 3-2 MeCNMeOH 3-1 3-2 Formulaa C46H52Cl4Mn4N10O24C140H162Cl6Mn25N26O80 Fw, g/mola 1490.54 5075.16 Space group P 21/ n C 2/ c a, 11.2401(9) 30.710(2) b, 21.2408(15) 30.818(2) c 12.6654(2) 18.0521(13) 90 90 101.312(2) 91.415(2) 90 90 V 3 2965.1(4) 17080(2) Z 2 4 T K 173(2) 173(2) b 0.71073 0.71073 calc, g/cm3 1.669 1.974 mm-1 1.101 1.976 R1 c,d 0.0497 0.0670 wR2 e 0.1182 0.1577 aIncluding solvate molecules. b Graphite monochromator. c I > 2 (I). d R1 = (|| Fo| | Fc||) / | Fo|. e wR 2 = [ [ w( Fo 2 Fc 2)2] / [ w(Fo 2)2]]1/2, w = 1/[ 2(Fo 2) + [( ap)2 + bp ], where p = [max ( Fo 2, O) + 2 Fc 2]/3. Table 3-2. Bond-valence sums for the Mn atoms of complex 3-2a Atom MnII MnIII MnIV Atom MnII MnIII MnIV Mn1 1.97 1.82 1.88 Mn8 3.21 2.97 3.07 Mn2 3.23 2.99 3.08 Mn9 4.29 3.92 4.12 Mn3 3.25 2.98 3.12 Mn103.19 2.96 3.05 Mn4 3.17 2.92 3.03 Mn111.92 1.77 1.83 Mn5 3.26 2.98 3.13 Mn121.96 1.81 1.87 Mn6 3.22 2.97 3.07 Mn133.19 2.95 3.04 Mn7 3.24 2.97 3.11 aThe underlined value is the one closest to the charge for which it was calculated. The oxidation state of a particular atom can be taken as the near est whole number to the underlined value Table 3-3. Bond-valence sums for the O atoms of complex 3-2a Atom BVS AssignmentAtomBVSAssignment O2 1.94 O2O13 1.85 O2O3 2.06 O2O15 1.16 OHO4 1.85 O2o17 1.85 O2O11 1.81 O2O26 1.05 OHaThe BVS values for O atoms of O2-, OHand H2O groups are typically 1.8-2.0, 1.0-1.2 and 0.2-0.4, respectively, but can be affected somewhat by hydrogen-bonding.

PAGE 66

66 N OH OH O N OH N OH OH hmpH pdmH2L Figure 3-1. Structure of ligands: 2-hydroxymet hyl pyridine (hmpH), 2,6-pyridine dimethanol (pdmH2), 6-hydroxymethyl 2-pyridin e carboxylic acid (L). Figure 3-2. Labeled representa tion of the structure of 3-1 JT axis on MnIII are shown in black. Color code: MnIII, green; MnII, purple; O, red; N, blue; C, grey. Figure 3-3. Structure of the cation of 3-2 Color code: MnIV, yellow; MnIII, green; MnII, purple; O, red; N, blue; C, grey.

PAGE 67

67 Figure 3-4. Centrosymmetric core of 3-2 (top) and its three types of constituent layers (bottom). JT axis on MnIII are shown in bold. Color code: MnIV, yellow; MnIII, green; MnII, purple; O, red; N, blue; C, grey. Figure 3-5. Plots of MT vs T for complex 3-1 The solid line is the fit of the data; see the text for the fit parameters.

PAGE 68

68 Figure 3-6. The core of 3-1 defining the pairwise exchange interactions. Figure 3-7. (left) Plot of reduced magnetization ( M / N B) vs H / T for complex 3-1 The solid lines are the fit of the data; see the text for the fit parameters. (ri ght) Two-dimensional contour plot of the fitting error surface vs D and g for complex 3-1

PAGE 69

69 Figure 3-8. Plot of in-phase ( M'T ) and out-of-phase ( M") ac susceptibility data for complex 3-1 Figure 3-9. (left) Plot of MT vs T for 3-2 H2O at 0.1 T. (right) Plot of reduced magnetization ( M / N B) vs H / T for complex 3-2 H2O. The solid lines are the fit of the data; see the text for the fit parameters. Figure 3-10. Plot of M'T vs T (in-phase) ac susceptibility data for 3-2 H2O.

PAGE 70

70 CHAPTER 4 DIVERSITY OF STRUCTURAL TYPES IN POLYNUCLEAR IRON CHEMISTRY WITH A (N, N, O)TRIDENTATE LIGAND 4.1 Introduction Polynuclear iron com pounds with oxygen-based ligation are releva nt to a variety of fields such as bioinorganic chemistry and magnetic ma terials. Iron-oxo centers are found in several non-heme metalloproteins and metalloenzymes; fo r example, in mammals iron is stored as ferritin, a protein that sequesters FeIII as a polymeric oxo-hydroxo complex.26 A number of polynuclear iron complexes have thus been synt hesized and studied as possible models for ferritin in order to gain insights into the biomin eralization process involved in the formation of its metal core.31,72,73 On the other hand, the paramagnetic na ture of Fe in its common oxidation states can often lead to interesting magnetic prop erties for polynuclear Fe clusters, such as high ground state spin values and ev en single-molecule magnetism.13 Although the exchange interactions between FeIII centers are almost always antiferromagnetic, certain Fex topologies can nevertheless posse ss large ground state spin values as a result of spin frustration. Th e latter is here defined in its more general sense of competing exchange interactions of comparable magn itude, preventing (frustrating) the preferred antiparallel alignment of a ll spins, and thus giving larger gro und state spin values than might be predicted.74-78 In favorable cases, where these larg e ground state spins are coupled to a significant magnetic anisotropy, the compounds can behave as single-molecule magnets (SMMs).20 This is the case for SMMs such as [Fe8O2(OH)12(tacn)6]8+ and [Fe4(OMe)6(dpm)6] etc.79-82 The above considerations and others conti nue to stimulate groups around the world to develop new synthetic methods that can yi eld new polynuclear Fe/O clusters. A common approach has been to use chelat es in order to encourage aggr egation while ensuring discrete

PAGE 71

71 products. Examples include 2,2-bipyridine (bpy),77 1,4,7-triazacyclononane (tacn),81 and the anion of dibenzoylmethane (dbm-).83 When the chelate also contai ns potentially bridging groups such as alkoxides, new high-nuclearity products can be obtained. Examples of this include the deprotonated, tridentate (N,O,O) form of N-methyldiethanolamine (mdaH2) and (O,O,O) form of tris(hydroxymethyl)ethane (thmeH3), and others.84-89 We decided to extend this approach to the potentially tridentate (N,N,O) chelate 2-{[ 2-(dimethylamino)ethyl]-methylamino}ethanol (dmemH, Figure 4-1). This has some similarity to mdaH2, but it only has one alcohol group and it was thus anticipated to give new structural types of produc ts. We were unable to locate previous examples in the literature of tran sition metal complexes with this chelate. Our first investigations with dmemH have been in Fe chemistry using the triangular [Fe3O(O2CR)6(H2O)3]+ complexes as reagents, a common strategy in both FeIII and MnIII chemistry.37,38,90-92 We have found from these reactions th at dmemH is indeed a good route to a variety of interesting new cluster types. These results are described in this chapter, which reports the syntheses, structures, and magnetochemical characterization of f our new Fe clusters containing dmem-.93 4.2 Experimental Section 4.2.1 Syntheses All preparations were perfor med under aerobic conditions using reagents and solvents as received. [Fe3O(O2CPh)6(H2O)3](NO3), [Fe3O(O2CBut)6(H2O)3](NO3), [Fe3O(O2CMe)6(H2O)3](NO3) and (NEt4)2(Fe2OCl6) were synthesized as reported elsewhere.94-97 [Fe7O4(O2CPh)11(dmem)2] ( 4-1 ) Method A. An orange red solution of [Fe3O(O2CPh)6(H2O)3](NO3) (0.20 g, 0.19 mmol) in MeCN (20 mL) was treated with dmemH (0.06 mL, 0.38 mmol) and the solution stirred overnight at room temperature. It was then filtered to remove undissolved starting material, and th e filtrate was allowed to stand undisturbed at

PAGE 72

72 room temperature. X-ray qua lity orange crystals of 4-1 MeCN slowly formed over 5 days in 45% yield. These were collected by filtration, washed with MeCN, and dried under vacuum. Anal. Calcd (Found) for 4-1 MeCN (C92H90.5N4.5Fe7O28): C, 52.66 (52.55); H, 4.35 (4.38); N, 3.00 (3.05). Selected IR data (cm-1): 1598(m), 1567(m), 1539(m), 1413(vs), 1175(w), 1069(w), 1025(w), 717(m), 675(w), 644(m), 461(m). Method B. A solution of FeCl3H2O (0.20 g, 0.74 mmol) and NaO2CPh (0.21 g, 1.48 mmol) in MeCN (15 mL) was treated with dmemH (0.06 mL, 0.37 mmol) and stirred for 3 hours. The resultant red brown solution was filtere d to remove NaCl, and the filtrate was left undisturbed at room temperature for slow evaporation. Orange crystals slowly formed over 5 days in 30% yield; the product was identified by IR spectral comparison with material from Method A. Method C. A solution of (NEt4)2(Fe2OCl6) (0.20 g, 0.33 mmol) and NaO2CPh (0.14 g, 0.99 mmol) in MeCN (15 mL) was treated with dmemH (0.11 mL, 0.66 mmol) and stirred for few hours. The resultant red-brown solution was filtered and kept undisturbed at room temperature for slow evaporation. Orange crystals slowly formed over 3 days in 40% yield; the product was identified by IR spectral co mparison with material from Method A. [Fe7O4(O2CMe)11(dmem)2] (4-2). Method A. A solution of FeCl3H2O (0.20 g, 0.74 mmol) and NaO2CMe3H2O (0.25 g, 1.85 mmol) in MeCN (15 mL) was treated with dmemH (0.06 mL, 0.37 mmol) and stirred for 3 hours. The resultant red-brown solution was filtered to remove NaCl, and the filtrate was left undisturbe d at room temperature for slow evaporation. Xray quality, dark orange crystals appeared over 20 days in 15 % yield. These were collected by filtration, washed with MeCN, and drie d under vacuum. Anal. Calcd (Found) for 4-2 MeCN (C40H73N6Fe7O28): C, 32.53 (32.66); H, 4.98 (5.26); N, 5.69 (5.46). Selected IR data (cm-1):

PAGE 73

73 3431(br), 2985(w), 2875(w), 1565(vs), 1426(vs ), 1088(w), 1052(w), 1033( w), 886(w), 709(w), 668(m), 637(m), 615(m), 539(m), 487(m). Method B. An orange-red solution of [Fe3O(O2CMe)6(H2O)3](NO3) (0.20 g, 0.03 mmol) in MeCN (15 mL) was treated with dmemH (0.10 mL, 0.06 mmol) and the solution stirred overnight at room temperature. It was then filt ered and the filtrate allo wed to stand undisturbed at room temperature. Orange crystals of the product formed over 25 days in 10% yield; the product was identified by IR spectral co mparison with material from Method A. [Fe6O2(OH)4(O2CCBut)8(dmem)2] ( 4-3 ). A solution of dmemH (0.03 ml, 0.19 mmol) in MeCN (5 mL) was treated with pyridine (15 L, 0.19 mmol), followed by the addition of a solution of [Fe3O(O2CBut)6(H2O)3](NO3) (0.18 g, 0.19 mmol) in MeCN (12 mL). The resultant solution was filtered and the filtrate left undistur bed at room temperature. X-ray quality orange needles of 4-3 MeCN grew over 10 days in 20 % yiel d. These were collected by filtration, washed with MeCN, and dried under vacuum. Dried solid analyzed as solv ent-free. Anal. calcd (Found) for 4-3 (C54H110N4Fe6O24): C, 42.27 (42.53); H, 7.23 (7.40); N, 3.65 (3.68). Selected IR data (cm-1): 2960(m), 2925(w), 2866(w), 1558(vs), 1484(s), 1427(vs), 1376(w), 1332(w), 1228(m), 1073(w), 903(w), 787(w), 662(m), 608(m), 530(m), 427(m). [Fe3O(O2CBut)2(N3)3(dmem)2] ( 4-4 ). A solution of [Fe3O(O2CBut)6(H2O)3](NO3) (0.10 g, 0.11 mmol) in EtOH (15 mL) wa s treated with dmemH (34 L, 0.20 mmol) and solid sodium azide (0.03 g, 0.46 mmol), and then stirred overnight at room temp erature to give an orange precipitate. The solid was collected by filtration, wa shed with a little EtOH. It was recrystallized from a CH2Cl2/hexanes layering to give X-ra y quality orange crystals of 4-4 CH2Cl2 over 3 days in 25 % yield. Anal calcd (Found) for 4-4 CH2Cl2 (C24.5H53N13Fe3O7Cl): C, 34.83 (34.77); H, 6.32 (6.30); N, 21.55 (21.16). Selected IR data (cm-1): 3390(br), 2959(w), 2870(w), 2066(vs),

PAGE 74

74 1543(m), 1480(w), 1418(m), 1342( w), 1225(w), 1087(m), 986(w) 720(m), 633(w), 606(w), 429(w). 4.2.2 X-ray Crystallography Data were collected by Dr. Khalil A Abboud on a Siemens SMART PLATFORM equipped with a CCD area detector and a graphite monochromator utilizing Mo-K radiation ( = 0.71073 ). Suitable crystals of 4-1 MeCN, 4-2 MeCN, 4-3MeCN, and 4-4CH2Cl2 were attached to glass fibers using silicone grease and transferred to a goniostat where they were cooled to 173 K for data collection. Cell parame ters were refined using up to 8192 reflections. A full sphere of data (1850 fram es) was collected using the -scan method (0.3 frame width). The first 50 frames were remeasured at the end of da ta collection to monitor instrument and crystal stability (maximum correction on I was < 1 %). Absorption corrections by integration were applied based on measured indexed crystal f aces. The structure was solved by the Direct Methods in SHELXTL6,46 and refined using full-matrix leas t squares. The non-H atoms were treated anisotropically, whereas the hydrogen atoms were calculated in ideal positions and were riding on their respective carbon atom s. Refinement was done using F2. In 4-1 MeCN, the asymmetric unit consists of half the Fe7 cluster and two MeCN molecules of crystallization. A total of 644 parameters were re fined in the final cycle of refinement using 32986 reflections with I > 2 ( I ) to yield R1 and wR2 of 4.57 and 8.99 %, respectively. In 4-2 MeCN, a total of 721 parameters we re refined in the final cycle of refinement using 18637 reflections with I > 2 ( I ) to yield R1 and wR2 of 3.53 and 8.64 %, respectively. In 4-3 MeCN, a total of 924 parameters were refined in the final cycle of refinement using 17808 reflections with I > 2 ( I ) to yield R1 and wR2 of 4.15 and 10.0 %, respectively. In 4-4 CH2Cl2, the azide ligand at N11 was diso rdered and it was refined in two positions with the site occupation factors dependently refined. A total of 472 parameters were

PAGE 75

75 refined in the final cycle of refi nement using 9205 reflections with I > 2 ( I ) to yield R1 and wR2 of 4.97 and 10.23 %, respectively. Unit cell data and details of the structure refinements for the four complexes are listed in Table 4-1. 4.3 Results and Discussion 4.3.1 Syntheses Many synthetic procedures to polynuclea r iron clusters rely on the reaction of [Fe3O(O2CR)6(H2O)3]+ species with a potentially chelating ligand, and this was one of the procedures chosen in the present work. In such reactions, the [Fe3O]7+ core of the trinuclear iron complex serves as a building block for higher nu clearity species, but the exact nuclearity and structure of the product depends on several factors; in the present work, we have found that the identity of the carboxylate group is one of these. Reaction of [Fe3O(O2CPh)6(H2O)3](NO3) with 1 3 equiv of dmemH in MeCN gave the heptanuclear complex [Fe7O4(O2CPh)11(dmem)2] ( 4-1 ) with a core topolo gy not previously encountered (eq. 4-1) The same product was obtaine d from an EtOH reaction solvent, and also from the treatment of an MeCN solution of FeCl3H2O with sodium benzoate and dmemH in a 2:4:1 ratio. Increasing the amount of sodium benzoate or dmemH still gave complex 4-1 but the reaction was not so clean. Reactions in wh ich the MeCN was replaced by EtOH, and the FeCl3H2O by Fe(ClO4)3xH2O or (NEt4)2(Fe2OCl6), also gave the same product, for Fe:dmemH ratios of both 1:1 a nd 1:2. Clearly, complex 4-1 is a preferred product of these components and particular carboxylate group. 7[Fe3O(O2CPh)6(H2O)3]+ + 2dmemH + 4e3[Fe7O4(O2CPh)11(dmem)2] + 9PhCO2 + 16H2O + 12H+ (4-1) If the carboxylate employed was acetate instead of benzoate, then the product from the FeCl3/MeCO2Na/dmemH (2:5:1) reaction system in Me CN (Method A of Experimental Section)

PAGE 76

76 was another heptanuclear complex, [Fe7O4(O2CMe)11(dmem)2] ( 4-2 ) (eq. 4-2). Its formula is the same as that of 4-1 except for the carboxylate identity, but structurally the two complexes are very different (vide infra). The same product 4-2 was obtained using [Fe3O(O2CMe)6(H2O)3]+ as the starting material in a reacti on with two equiv. of dmemH in MeCN (Method B). The yields of 4-2 were much lower than 4-1 although they could be improved somewhat by addition of some NEt3 base to the reaction. 7FeCl3 + 11MeCO2 + 2dmemH + 4H2O [Fe7O4(O2CMe)11(dmem)2] + 21Cl+ 10H+ (4-2) In contrast to the heptanuclear products from the use of benzoate and acetate reagents, the use of pivalate ones led to a hexa nuclear product. Treatment of [Fe3O(O2CBut)6(H2O)3](NO3) with dmemH in MeCN led to subsequent isolation of [Fe6O2(OH)4(O2CBut)8(dmem)2] ( 4-3 ) (eq. 4-3). The addition of one equivalent of NEt3 or pyridine as base improves the yield from 10 to 20 %. The same product is obtained on increasing the amount of dmemH to three equivalents. 2[Fe3O(O2CBut)6(H2O)3]+ + 2dmemH [Fe6O2(OH)4(O2CBut)8(dmem)2] + 4ButCO2 + 6H+ + 2H2O (4-3) It is clear that the r eactions that lead to 4-1 to 4-3 are very complicated, and the reaction solutions likely contain a complicated mixture of several species in equilibrium. In such cases, factors such as relative solubil ity, lattice energies, crystallizati on kinetics, and others determine the identity of the isolated products, and one (o r more) of these factor s is undoubtedly the reason that the reaction product is so dependent on the exact carboxylate employed. Since complex 4-3 contains bridging hydroxide groups, a similar reaction was explored in the presence of sodium azide. Perlepes and cowo rkers have demonstrated that replacement of bridging hydroxide groups (which almost always mediate antiferromagnetic exchange interactions) with end-on br idging azide groups (which medi ate ferromagnetic exchange) in

PAGE 77

77 cobalt, nickel and iron clusters leads to products with much higher ground state spin values.58-60 Thus, we explored a variety of reaction conditions differing in the azide amount and/or solvent, and it was found that reaction of [Fe3O(O2CBut)6(H2O)3](NO3), dmemH and azide in a 1:2:4 ratio gave the new trinuclear complex [Fe3O(O2CBut)2(N3)3(dmem)2] ( 4-4 ) (eq. 4-4). The complex has its azide groups all in terminal site s, but it nevertheless has an interesting core structure. Complex 4-4 was also obtained in lower yield from the reaction of preformed complex 4-3 with four equivalents of sodium azide in EtOH. [Fe3O(O2CBut)6(H2O)3]+ + 2dmemH + 3N3 [Fe3O(O2CBut)2(N3)3(dmem)2] + 3H2O + 4ButCO2 + 2H+ (4-4) 4.3.2 Description of Structures 4.3.2.1 Structure of [Fe7O4(O2CPh)11(dmem)2] (4-1) A labeled representation of complex 4-1 is shown in Figure 4-2(a). Selected interatomic distances and angles are summarized in Table A-6. Complex 4-1 MeCN crystallizes in the monoclinic space group C 2/c with the Fe7 molecule lying on a crystallographic C2 axis passing through the central Fe4 atom. The co re can be described as two [Fe4( 3-O)2] planar-butterfly units fused at body atom Fe4, one butterfly unit being atoms Fe1', Fe2, Fe3, Fe4, O9, O10'. Further, each butterfly unit can be considered as two edge-sharing Fe3O triangular units, with the 3-O2bridging atoms O9 and O10 slightly above and below their Fe3 planes. These O atoms bridge somewhat asymmetrically; th e bonds to the wingtip Fe atoms (Fe1 O10, 1.828 and Fe3 O9, 1.844 ) are shorter than th e bonds to the body Fe atoms (Fe2 O10', 1.941 and Fe2 O9, 1.923 ). The two dmemgroups bind as tridentate chel ates to Fe1 and its symmetry partner Fe1', with their alkoxide O atoms bridging wingtip atom Fe1 in one Fe4 unit with body atom Fe2 in the other. The remain ing peripheral ligation about the [Fe7O4] core is provided by

PAGE 78

78 eleven benzoate groups, nine in their common 1: 1: bridging mode and the other two in the rare 2 chelating mode on Fe3 and Fe3'. 4.3.2.2 Structure of [Fe7O4(O2CMe)11(dmem)2] (4-2) A labeled representation of complex 4-2 is provided in Figure 42(b). Selected interatomic distances and angles are gi ven in Table A-7. Complex 4-2 MeCN crystallizes in the triclinic space group P The molecule contains a remarkable [Fe7( 3-O)4] core. It can be described as consisting of a central [Fe3O3] ring containing Fe2, Fe3 and Fe5, with each of the doublybridging O2ions of this hexagon becoming 3 by also bridging to a thir d, external Fe atom (Fe1, Fe4, Fe6). The fourth O2ion bridges ring atom Fe5, Fe6 and a seventh Fe atom (Fe7) on the periphery of the molecule. The two dmemgroups bind one each to the external atoms Fe1 and Fe4 in a tridentate chelating manner with their alkoxide O atoms also bridging to ring atoms Fe2 and Fe3, respectively. Peripheral ligation is completed by eleven acetate groups, ten in 1: 1: bridging modes and one 2 chelating to Fe7. The molecular structures of 4-1 and 4-2 can be said to represent two different ways of linking a number of Fe3O triangular units, as is clear in Figure 4-3 where the cores of 4-1 to 4-3 are compared. The core topologies of complexes 4-1 and 4-2 are unprecedented within FeIII chemistry. Indeed, there are only a few Fe7 complexes in the literature, and they are all mixedvalent except for the one reported by Winpe nny and coworkers containing phenylphosphonate ligand and Zheng and coworkers cont aining cyclohexenephosphonate ligand.98,99 In addition, very recently few disklike and domelike heptanuclear FeIII clusters were also reported,100-102 but they are significantly different from the ones obtained with dmemH. Complexes 4-1 and 4-2 are thus the novel heptanuclear FeIII carboxylate complexes.

PAGE 79

79 4.3.2.3 Structure of [Fe6O2(OH)4(O2CBut)8(dmem)2] (4-3) A labeled representation of complex 4-3 is shown in Figure 4-4(a). Selected interatomic distances and angles are gi ven in Table A-8. Complex 4-3 MeCN crystallizes in the triclinic space group P with the asymmetric unit co ntaining two independent Fe6 clusters, both lying on inversion centers; since the two molecules are e ssentially superimposable, we show and discuss the structure of only one of them here. The core consists of an [Fe4( 3-O)2] unit (Fe1, Fe1', Fe2 and Fe2'] on either side of which is attached a [Fe( -OH)2( -OR)] unit containing Fe3; the OHions are O9 and O10 on one side, and thei r symmetry partners on the other. One OHbridge (O10) connects Fe3 to central Fe1 whereas the other (O9) connects to Fe2. The OHnature of O9 and O10 was confirmed by BVS calculations,103 which gave values of 1.14 for O9 and 1.05 for O10. Peripheral ligation is provided by two dmemand eight pivalate groups. There are three types of pivalate binding m odes: four are in the common 1: 1: bridging mode, two are in the rare 2 chelating mode, and the remaining two are in an 1 terminal mode. A number of other Fe6 complexes have been previously reported possessin g a variety of metal topologies, such as planar, twisted boat, chair, parallel tr iangles, octahedral, ladder-like, cyclic, etc.104 However, the only previous comp ounds structurally similar to 4-3 are [Fe6O2(OMe)12(tren)2]2and [Fe6O2(OR)8(O2CPh)6].105,106 Both of the latter complexes contain a central [Fe4( 3-O)2]8+ core with an additional Fe atom on each side, as in 4-3 but the precise means of attachment are different. 4.3.2.4 Structure of [Fe3O(O2CBut)2(N3)3(dmem)2] (4-4) A labeled representation of 4-4 is provided in Figure 4-4(b). Selected interatomic distances and angles are given in Table A-9. Complex 4-4 CH2Cl2 crystallizes in the monoclinic space group P 21/n. The structure consists of an Fe3 isosceles triangle bridged by a 3-O2ion (O1) with a rare T-shaped geometry, rather than the comm on trigonal planar geometry usually seen in

PAGE 80

80 triangular metal carboxylates.107 The Fe1Fe2 and Fe2Fe3 edges are each additionally bridged by an alkoxide O atom of the dmemligand and a 1: 1: pivalate group. As a result, Fe2Fe1 (2.997(1) ) and Fe2Fe3 (2.980(1) ) are much shorter than Fe1Fe3 (3.694(2) ). Similarly, Fe2-O1 (2.070(19) ) is noticeably longer than Fe1-O1 (1.872(19) ) and Fe3-O1 (1.865(19) ). Fe1, Fe2, Fe3 and O1 are co-planar, and O2 and O3 are slightly above and below this plane. A chelating dmemand a terminal azide on each Fe atom complete the ligation at the metal atoms, which are all near-octahed ral. The overall asymmetric Fe3O is with little precedent in iron chemistry, the only previous discrete example being [Fe3O(TIEO)2(O2CPh)2Cl3], where TIEO is 1,1,2-tris(1-methylimidazol-2-yl)ethoxide.73,108 4.3.3 Magnetochemistry of Complexes 4-1 to 4-4 4.3.3.1 Dc Studies Solid-s tate, variable-temperature dc magnetic susceptibility data in a 0.1 T field and in the 5.0-300 K range were collected on po wdered crystalline samples of 4-1 to 4-4 restrained in eicosane. The obtained data are plotted as MT vs T in Figure 4-5. For 4-1 MeCN, MT steadily decreases from 6.95 cm3Kmol-1 at 300 K to 4.07 cm3Kmol-1 at 5.0 K. The 300 K value is much less than the spin-only (g = 2.0) value of 30.62 cm3Kmol-1 for seven non-interacting FeIII ions, indicating the presence of strong antiferromagnetic interactions. The 5.0 K value suggests an S = 5/2 ground state. For 4-2 MeCN, MT steadily decreases from 8.19 cm3Kmol-1 at 300 K to 4.14 cm3Kmol-1 at 34 K, stays essentially constant until 10 K, and then decreas es slightly to 3.85 cm3Kmol-1 at 5.0 K. As for 4-1 MeCN, this behavior is in dicative of antiferromagnetic interactions and an S = 5/2 ground state. For 4-3 MT increases from 9.73 cm3Kmol-1 at 300 K to a maximum of 14.10 cm3Kmol-1 at 20K and then drops to 12.92 at 5.0 K. The 300 K value is again much less than the spin-only value of 26.25 cm3Kmol-1 expected for six non-interacting FeIII ions, indicating predominantly antiferro magnetic interactions. The increase in MT as the

PAGE 81

81 temperature then decreases suggests the lowest lying states are of high spin values, and the maximum at 20 K of 14.10 cm3Kmol-1 is very close to the spin-only value of 15.00 cm3Kmol-1 for an S = 5 ground state. The decrease in MT at the lowest temperatures is very likely due to zero-field splitting (ZFS) within the S = 5 ground state and perhaps some weak intermolecular interactions. For 4-4 CH2Cl2, MT steadily decreases from 5.74 cm3Kmol-1 at 300 K to 4.07 cm3Kmol-1 at 50 K, and then stays approximately c onstant until 15 K, below which it decreases slightly to 3.75 cm3Kmol-1 at 5.0 K. The latter value suggests an S = 5/2 ground state. To confirm the above ground state sp in estimates, variable-field ( H ) and -temperature magnetization (M ) data were collected in the 0.1 to 7.0 T and 1.8-10 K ranges. The resulting data for 4-1 MeCN are plotted in Figure 4-6 as reduced magnetization ( M/N B) vs. H / T where N is Avogadro's number and B is the Bohr magneton. The saturati on value at the highest fields and lowest temperatures is ~4.8, as expected for an S = 5/2 and g slightly less than 2; the saturation value should be gS. The data were fit, using the program MAGNET53 described elsewhere.54-56 The best-fit for 4-1 MeCN is shown as the solid lines in Figure 4-6, and was obtained with S = 5/2 and either of two sets of parameters, g = 1.94 and D = -0.56 cm-1, or g = 1.95 and D = 0.77 cm-1. Alternative fits with S = 3/2 or 7/2 were rejected because they gave unreasonable values of g and D It should be noted that it is common to obta in two acceptable fits of magnetization data for a given S value, one with D > 0 and the other with D < 0. This was indeed the case for the magnetization fits for all of the complexes 4-1 to 4-4 in this work. In order to assess which is the superior fit in these cases, and also to ensure that the true global minimum had been located for each compound, we calculated the root-mean-square error surface for the f its as a function of D and g, and have plotted them as two-dimensional contour plots in Figure 4-7. For 4-1 MeCN,

PAGE 82

82 the plot clearly shows only the above-mentioned minima with positive and negative D values, with both fits being of comparable quality as shown in Figure 4-7. For 4-2 MeCN, the reduced magnetization plot saturates at ~4.5, ag ain suggesting an S = 5/2 ground state. The fit, shown as th e solid lines in Figure 4-8(a), gave S = 5/2 with either g = 1.91 and D = -0.76 cm-1, or g = 1.91 and D = 0.98 cm-1. The error surface contour plot is shown in Figure 4-8(b) and shows the above minima, with the one with negative D clearly the superior fit since it has a lower (deeper) minimum. Figure 48(b) also clearly shows that the fit minimum is a soft one, consistent with a significan t uncertainty in the pr ecision of the obtained g and D fit values, which we estimate as 0.01 on g and 5 10 % on D For 4-3 the reduced magnetization plot saturates at ~ 9.5, suggesting an S = 5 state with g < 2 (Figure 4-9(a)). A satisfactory fit could only be obtained if data collected at fields above 5 T were excluded, suggesting that some low-lying excited states with S > 5 are being stabilized by the applied field to the point that they are significantly populated at these temperatures. To avoid this, the data at 6 and 7 T were excluded, and now a good fit was obtained (solid lines in Figure 4-9(a)) with S = 5 and either g = 1.95 and D = -0.28 cm-1, or g = 1.92 and D = 0.33 cm-1. The error surface for the fit shows ag ain that the fit with negative D is far superior (Figure 4-9(b)), suggesting this is the true sign of D 4.3.3.2 Rationalization of the Ground State Spin of 4-1 and 4-3 It is of interest to tr y to rationalize the observed ground state spin values of 4-1 to 4-3 It is assum ed that all Fe2 pairwise exchange interactions are antiferromagnetic, as is essentially always the case for high-spin FeIII, and there will thus be compe ting antiferromagnetic exchange interactions and spin frustration effects within the many Fe3 triangular units in these complexes. The ground state of 4-1 is the easiest to rationalize: the discrete Fe4 butterfly (planar or bent rhombus) topology is known to usually give an S = 0 ground state as a result of the

PAGE 83

83 antiferromagnetic interactions al ong the four outer edges overcoming, and thus frustrating, the diagonal interaction.77,109-111 In 4-1 two such Fe4 units are fused at body (central) Fe4 of the two butterfly units, and assuming the same spin alignments as in the discrete Fe4 molecules, then the ground state spin alignments are predicted to be those shown in Figure 4-10(a), giving an S = 5/2 ground state, as observed experime ntally. It is worth mentioning here that disklike and domelike heptanuclear amino alkoxo clusters reported in literature have also been shown to possess S = 5/2 ground state spin.101,102 The ground state for 4-2 is not so easy to rationalize convincingly because of its high content of triangular units. For 4-3 the recognizable Fe4 unit as in 4-1 suggests that the spin of this sub-unit is zero, and then the two Fe atoms Fe3 and Fe3 above and below would have their spins parallel to each other by both being antiparallel to the spins of Fe1 and Fe1 as shown in Figure 4-10(b). This would thus rationalize an overall S = 5 ground state for 4-3 For 4-4 CH2Cl2, the reduced magnetization saturates at ~4.7, suggesting an S = 5/2 ground state and g < 2 (Figure 4-11(a)). The fit of the da ta (solid lines in Figure 4-11(a)) gave S = 5/2 with either g = 1.92 and D = -0.69 cm-1, or g = 1.92 and D = 0.82 cm-1. The D vs. g error surface (Figure 4-11(b)) shows that the fit with negative D is again superior suggesting this may be the true sign of D Since complex 4-4 is only trinuclear, we determined its pairwise Fe2 exchange interactions by fitting the variable temp erature susceptibility data to the appropriate theoretical expression. 4.3.3.3 Determination of the Exchange Interactions in 4-4 The Heisenb erg spin Hamiltonian describing the isotropic exchange interactions within an isosceles Fe3 triangle of C2v symmetry (Figure 4-12(a)) is given by eq. 4-5, where Ja refers to the interactions between Fe 2Fe3 and Fe1Fe2, and Jb refers to the Fe3Fe1 interaction; Si refers

PAGE 84

84 to the spin of atom Fei. The energies of the resultant total spin states ST, which are eigenfunctions of the Hamiltonian in this coupli ng scheme, are given by eq. 4-6, where A = 1 + 3. The overall multiplicity of the spin system is 216 made up of 27 individual spin states ranging from ST = 1/2 to 15/2. H = -2Ja( 23 + 2 1) 2 Jb( 3 1) (4-5) E| ST, SA > = Ja[ ST( ST+1) SA( SA+1)] Jb[ SA( SA+1)] (4-6) An expression for the molar paramagnetic susceptibility was derived for this complex using the Van Vleck equation.51 This was then used to fit the experimental MT vs T data, with fit parameters Ja, Jb and an isotropic g value (see Appendix D-3). The fit is shown as the solid line in Figure 4-5, which gave Ja = 3.6 cm-1, Jb = 45.9 cm-1 and g = 1.93. These values identify the ground state as the | ST, SA > = | 5/2, 0 > state shown in Figur e 4-12(b), which is in agreement with the reduced magnetization fit. The marked inequality in the exchange constants, Jb >> Ja is as expected on the basis of the iron-oxo bond lengths, where Fe3-O1 = Fe 1-O1 < Fe2-O1. A similar situation was also observed in the previous Fe3 complex with a similar core, [Fe3O(TIEO)2(O2CPh)2Cl3], for which Ja = -8(4) cm-1 and Jb = -55(6) cm-1. It has been established that the magnitude of the exchange coupling constant J for an oxo-bridged FeIII 2 unit can be approximately correlated with a single structural parameter P by the equation in eq. 47, if the Fe-O-Fe does not alter too much. In this relationship, A = 8.763 x 1011 B = -12.663 and P is the shor test superexchange pathway.112 Applying this relationship to complex 4-4 gives Ja = -12.9 cm-1 and Jb = 46.4 cm-1, which are in reasonable overall agreement with the experimental values obtained from fitting the susceptibility data, and given that an angular dependence is of lesser importance than the radial one, and is ignored by eq. 4-7. In particular, the acute values of angles of Fe2-O2-Fe3 and Fe2-

PAGE 85

85 O3-Fe1 (96.07(8) and 96.77(8) resp ectively), which lead to the weak Ja coupling, are significantly smaller than those found in dinuclear FeIII 2 complexes on which the relationship of eq. 4-7 was based and probably reflect a gr eater angular dependence. The value of Jb is stronger than the magnitude of the an tiferromagnetic coupling consta nt found for the triangular FeIII carboxylate complexes with an a pproximately equilateral [Fe3O]7+ core (~30cm-1),113 but weaker than the 80-130 cm-1 values observed for the [Fe2O]4+ and [Fe2O(O2CR)2]2+ dinuclear cores.114116 J = A eBP (4-7) None of the compounds 4-1 to 4-4 exhibited an out-of-phase AC magnetic susceptibility signal down to 1.8 K in an AC field of 3.5 Oe oscillating with frequencies up to 997 Hz. This indicates that they do not e xhibit a large enough barrier (vs kT ) to exhibit the characteristic signature of slow magnetization relaxation charac teristic of single-molecule magnets (SMMs), at least down to 1.8 K. As discussed above, fits of variable-temperatu re and variable-field magnetization data are not the most reliable way to obtain the most precise and accurate values of D or its sign. The magnetization fits suggested D to be negative for 4-2 and 4-3 but they could not suggest the sign of D for 4-1 Since the sign and magnitude of D are crucial to the potential ability of a complex to function as a SMM, we desired to better characterize D for these new and relatively rare examples of Fex clusters with significant ground state spin values. The perfect technique for this is high-frequency electron para magnetic resonance spectroscopy. 4.3.4 High-Frequency EPR Spectroscopy A detailed single-crystal study of representative com plexes 4-1 MeCN and 4-3 MeCN has been carried out by HFEPR spectroscopy. Th e main overall objective was to measure the ZFS parameters in the spin Hamiltonian of eq. 48, which is the same as that in eq. 3-4 except

PAGE 86

86 that it also now includes the rhombic ZFS term, E ( x 2 y 2), where E is the rhombic ZFS parameter, and x and y are the x and y components of the total spin operator EPR is a high resolution spectroscopic technique that can be us ed to investigate the more complete spin Hamiltonian of eq. 4-8, whereas fits of bulk ma gnetization data are essentially insensitive to inclusion of the rhombic E term. H = D z 2 + E ( x 2 y 2) + gB0 H (4-8) Single-axis angle dependence studies were first performed to roughly determine the orientation of each crystal in the magnetic field. Both complexes 4-1 MeCN and 4-3 MeCN possess low symmetry structures. Thus, determining the precise symmetry directions represents a highly complex task requiring de tailed two-axis rotation studie s. However, one can readily obtain basic information from single-axis studies; in particular, the sign of D which is the crucial factor in whether a part icular complex is a SMM.100,117 Figure 4-13(a) displays the a ngle-dependence of the field pos itions of the strongest EPR transitions determined from field-swept spec tra recorded at 116 GHz and 1.4 K for complex 4-3 MeCN; given the low temperatur e, these data points must co rrespond to transitions from the lowest-lying mS levels. Two series of resonances are obs erved (black and red data points) which shift significantly upon rotating the field, thus pr oviding the clearest evidence for a significant magnetoanisotropy. Both series exhibit 180o periodicity, with virtually identical amplitudes. The source of the two series has a natural explanation for complex 4-3 MeCN for which there are two differently oriented molecules in the unit cell. Thus, one natu rally expects two distinct EPR signatures, one from each species. The solid curv es represent phenomenological fits to the two sets of data, and are intended to capture the qualitative nature of the angle dependence. The phase shift between the two data sets is 75o.

PAGE 87

87 In order to determine the sign of D frequencyand temperature-dependent data were collected on complex 4-3 MeCN with the field oriented along one of the minima in Figure 413(a) (191o). Figure 4-13(b) displays the frequency dependence of the angle-dependent peak from Figure 4-13(a), and the inset displays repres entative spectra taken at higher temperatures. A remarkable feature of the frequency-dependent data is that all peaks lie on a straight line which extrapolates to a finite frequenc y on the vertical axis, i.e. there is no evidence for curvature in the data. Assuming | DS | 1.5 cm-1 (from reduced magnetization measurements), one realizes that at least a 3 T magnetic field would be required to overcome the axial term in eq. 4-8. This suggests that the Zeeman interaction commutes with the do minant axial term in eq. 4-8 across the entire range of fields for which data were collected (0.6 to 2 T). In other words, the minima in Figure 413(a) and the data in Figure 4-13( b) correspond to field or ientations parallel t o, or very close to the z axes of the two species. This is quite coincidental, as the sample orientation was not previously known. Figure 4-14(a) displays a simulation of the Zeeman diagram for a SMM with S = 5, i.e. with D < 0. As can clearly be seen, the transition from the lowest-lying mS level occurs at the lowest field; the excited state transitions all o ccur at higher field. This agrees qualitatively with the data in Figure 4-13(b). Ther efore, we can conclude that D is negative, and that 4-3 is a SMM. The intercept on the frequency axis in Figure 4-13(b) (66.4 GHz) then corresponds to the ZFS between the ground and first excite d state. If one assumes that S = 5, then D = -0.25(1) cm-1, which is in reasonable agreement with the value from the magnetization fits ( D = -0.28(3) cm-1). Because of the uncertainty in the precise orientat ion of the field relative to the easy-axis, we cannot quote a precise value for g; the main purpose of the HFEPR measurements was to unambiguously determine the sign of D which was successfully achieved.

PAGE 88

88 Single-axis rotation experiments for complex 4-1 MeCN were not able to locate the axial direction (presumably the rotation plane was inclined significantly with respect to the magnetic z -axis of the molecule). Neverthele ss, we were able to locate the plane perpendicular to the axial direction ( xy plane) from measurements similar to thos e shown in Figure 4-13(a). Thus, all of the temperatureand frequencydependent studies were carried out with the field aligned within the magnetic xy plane of the Fe7 molecule. Only a single molecular species was anticipated for complex 4-1 MeCN, making interpretation of the data more straightforward. Furthermore, this complex exhibits sharper EPR peaks, as eviden t from Figure 4-15(a), which shows the high-field xy-plane spectra obtained at different temperat ures and a frequency of 197 GHz. Comparison of the data in Figure 4-14(a) with the simulated Zeeman diagram in Figur e 4-13(b) reveals that complex 4-1 cannot be a SMM because its D value is positiv e. As can be seen from Figure 414(b), upon reducing the temperature, the stronger EPR peaks should be observed at the lowest fields for an easy-plane magnet ( D > 0) when the field is appl ied parallel to the easy ( xy ) plane; this is exactly what is seen in the data. If 4-1 MeCN were a SMM, the intensities of the five transitions (labeled a to e in the Figure 4-15(a)) would be reversed. Figure 4-15(b) displays the results of a multi-frequency study for complex 4-1 MeCN, with the field applied within the easy plane; the temperature was 20 K. Fits (solid curves) to the positions of the EPR peaks were performed via exact diagonalization of eq. 4-8. It is very clear from Figure 4-15(b) that the data lie on a series of lines that are not evenly spaced, and exhibit significant curvature at low fr equencies and fields. These trends are a characteristic of xy-plane spectra obtained for a system with a significant uniaxial anisotropy (both positive and negative D ), due to the competition between the orthogonal Zeeman and ZFS ( DSz 2) interactions. In other words, the data displayed in Figure 4-15(b) provid e further confirmation that the field is in the

PAGE 89

89 xy-plane and, when combined with the temperat ure dependence in Figure 4-15(a), also confirm the positive sign of D The fit assumes an S = 5/2 ground state, and yields g = 2.0 and D = +0.62 cm-1. This value again agrees reasonably well with that from the reduced magnetization studies ( D = +0.77(7) cm-1). The low value of g obtained from the reduced magnetization fits can be explained as the limitation of the fitting pr ogram Magnet which assumes axial anisotropy; therefore HFEPR data is more reliable. The best fit to the data required the inclusion of a rhombic ZFS anisotropy, | E | 0.067 cm-1. This is not unexpected, given the low symmetry of the molecule. Our estimate of E represents a lower bound, as the orie ntation of the field within the easy plane was not known. Low-temperat ure EPR measurements on domelike Fe7 cluster yield a D value of 0.28 cm-1,101 which is considerably lower than calculated for 4-1 MeCN. This can be attributed to their different structural arrang ements leading to the di fferences in single-ion anisotropy and spin-spin anisotropy. 4.4 Conclusions The tridentate N, N, O ligand dmemhas proven to be a very fru itful new route to a variety of new FeIII clusters comprising two Fe7 and one Fe6 species, depending on the identity of the carboxylate employed. The latter point emphasizes the exquisite sensitivity of the reaction product on a variety of reaction conditions and reagents employed. For example, even though complexes 4-1 and 4-2 have the same formula except for the identity of the carboxylate, the structures of the two complexes are very different It was interesting that the azide ligands in 4-4 were only terminal rather than bridging, but yet ne vertheless fostered formation of a product very different from that of the non-azide product 4-3 Fitting of the reduced magnetization vs H/T data established that 4-1 4-2 and 4-4 each possess an S = 5/2 ground state spin, whereas 4-3 has an S = 5 ground state. The complexes all serve to clearly emphasize again how ground state sp in values of significan t magnitude can result

PAGE 90

90 from spin frustration effects even though all the pairwise exch ange interaction constants are antiferromagnetic. The ma gnetization fits of 4-1 to 4-4 serve to emphasize, however, the difficulty of determining the sign of D for FeIII clusters from such measurements, making it thus difficult to predict whether a given cluster might be a new example of a SMM. Representative complexes 4-1 and 4-3 were therefore studied by HFEPR spectroscopy, a tremendously powerful and sensitive technique, not least for obtaining accurate and precise values for spin Hamiltonian parameters such as D including an unequivocal determin ation of its sign. From these measurements, we concluded that complex 4-3 has D < 0 and thus is a potential SMM, whereas complex 4-1 has D > 0 and is not. In fact, none of the compounds 4-1 to 4-4 exhibited an out-ofphase ac magnetic susceptibility signal down to 1.8 K in ac frequencies up to 997Hz. Even for 43, which was confirmed by HFEPR spectroscopy to have a negative D value, its S = 5 and D = 0.25 cm-1 gives a barrier ( U ) to magnetization relaxation with an upper value of U = S2| D | = 6.3 cm-1 (= 9.0 K). Remembering that th e true or effective barrier ( Ueff) is less that U due to quantum tunneling of the magnetization (QTM) through the ba rrier, it is not surp rising that no sign of slow relaxation is seen at temperatures above 1.8 K. Studies significantly below 1 K will be required in order to better investigate the potential SMM behavior. Neve rtheless, the present work does establish intere sting new examples of Fex clusters with si gnificant ground state S values and negative D values. Finally, the preparation of complexes 4-1 to 4-4 again serves to em phasize the utility of alkoxide-containing chelates in pol ynuclear metal cluster chemistry.

PAGE 91

91 Table 4-1. Crystallographic data for 4-1 MeCN, 4-2MeCN, 4-3MeCN and 4-4 CH2Cl2 4-1 4-2 4-3 4-4 Formulaa C99H101Fe7N8O28 C38H70Fe7N5O28 C58H116Fe6N6O24 C25H54Cl2Fe3N13O7 Fw, g/mola 2241.83 1435.94 1616.67 887.26 Space group C 2/c P P P 21/n a, 18.6028(14) 12.4586(8) 12.9769(10) 12.3260(8) b, 26.8523(14) 13.5495(9) 14.4142(11) 25.3961(17) c 20.8083(13) 18.690(12) 23.9082(18) 13.1400(9) 90 70.636(2) 87.6240(10) 90 103.879(2) 79.731(2) 88.5620(10) 99.1490(10) 90 73.099(2) 66.0920(10) 90 V 3 10090.9(11) 2836.2(3) 4084.7(5) 4060.9(5) Z 4 2 2 4 T K 173(2) 173(2) 173(2) 173(2) Radiation, b 0.71073 0.71073 0.71073 0.71073 calc, g/cm3 1.476 1.681 1.311 1.451 mm-1 1.058 1.828 1.105 1.244 R1 c,d 0.0457 0.0463 0.0415 0.0497 wR2 e 0.0899 0.0927 0.1009 0.1023 a Including solvate molecules. b Graphite monochromator. c I > 2 (I). d R1 = (|| Fo| |Fc||) / | Fo|. e wR 2 = [ [ w(Fo 2 Fc 2)2] / [ w(Fo 2)2]]1/2, w = 1/[ 2(Fo 2) + [( ap)2 + bp], where p = [max ( Fo 2, O) + 2 Fc 2]/3. N N HO dmemHN OH HO mdaH 2 Figure 4-1. Structure of ligands: mdaH2 and dmemH.

PAGE 92

92 Figure 4-2. (a) Labeled represen tation of the structure of 4-1 Hydrogen atoms and phenyl rings (except for the ipso carbon atoms) have been omitted for clarity. The C2 symmetry axis is approximately vertical. Color code: FeIII, green; O, red; N, blue; C, grey. (b) Labeled representation of the structure of 4-2 Hydrogen atoms have been omitted for clarity. Color code: FeIII, green; O, red; C, grey; N, blue Figure 4-3. Comparison of cores of 4-1 (a), 4-2 (b), and 4-3 (c). Color code: FeIII, green; O, red

PAGE 93

93 Figure 4-4. (a) Labeled repr esentation of the centrosymmetric structure of 4-3 Hydrogen atoms and methyl groups on pivalate groups have been omitted for clarity. Color code: FeIII, green; O, red; C, grey; N, blue. (b) La beled representation of the structure of 4-4 Hydrogen atoms have been omitted for clarity. Color code: FeIII, green; O, red; C, grey; N, blue. Figure 4-5. Plots of MT vs T for complexes 4-1 ( ), 4-2 ( ), 4-3 ( ) and 4-4 ( ). The solid line is the fit of the data for 4-4 ; see the text for the fit parameters.

PAGE 94

94 H/T(kG/K) 01 02 03 04 05 0 M/N B 0 1 2 3 4 5 6 0.1T 0.5T 1T 2T 3T 4T 5T 6T 7T Fitting Figure 4-6. Plot of reduced magnetization ( M / N B) vs H / T for complex 4-1 MeCN. The solid lines are the fit of the data; see the text for the fit parameters. Figure 4-7. Two-dimensional contour pl ot of the fitting error surface vs D and g for 4-1MeCN.

PAGE 95

95 Figure 4-8. (a) Plot of reduced magnetization ( M / N B) vs H / T for complex 4-2 MeCN. The solid lines are the fit of the data; see the text for the fit parameters. (b) Twodimensional contour plot of the fitting error surface vs D and g for 4-2 MeCN. Figure 4-9. (a) Plot of reduced magnetization ( M / N B) vs H / T for complex 4-3 The solid lines are the fit of the data; see the text for th e fit parameters. (b) Tw o-dimensional contour plot of the fitting error surface vs D and g for 4-3

PAGE 96

96 (a) Fe O Fe Fe O Fe O O Fe Fe Fe Fe Fe O O Fe Fe O Fe O O O Fe O O Fe (b) Figure 4-10. Rationalization of the ground state spin of (a) 4-1 and (b) 4-3 Figure 4-11. (a) Plot of reduced magnetization ( M / N B) vs H / T for 4-4 CH2Cl2. The solid lines are the fit of the data; see the text for th e fit parameters (b) Two-dimensional contour plot of the fitting error surface vs D and g for 4-4 CH2Cl2.

PAGE 97

97 Fe3 O1 Fe1 Fe2 O3 O2 Figure 4-12. (left) The core of 4-4 defining the pairwise exch ange interactions. (right) Rationalization of the ground state spin of 4-4 Figure 4-13. (a) Plot of the HFEPR peak positions for 4-3 MeCN obtained from angledependent studies at 116 GHz and 1.4 K. (b) Frequency dependence for 4-3 MeCN with the field oriented along one of the minima in Figure 4-13(a) (191o); the inset displays temperature-dependent spectra obtained at 106 GHz.

PAGE 98

98 Figure 4-14. (a) Simulated Zeeman diagram for a spin S = 5 system with D < 0 with the magnetic field parallel to the z axis. The red lines (labeled a to d) correspond to the transitions shown in the inset of Figure 4-13(b). (b) Simulated Zeeman diagram for a spin S = 5/2 system with D > 0 with the magnetic field parallel to the xy plane. The red lines (labeled a to c ) correspond to the transitions shown in Figure 4-15(a). Figure 4-15. (a) Temperature dependent spectra for 4-1 MeCN at 197 GHz with the DC magnetic field applied within the easy ( xy) plane. (b) Easy-plane peak positions for 4-1 MeCN plotted versus frequency at 20 K. The solid lines are simulations using the ZFS parameters given in the main text

PAGE 99

99 CHAPTER 5 A NEW N, N, O CHELATE FOR TRANSIT ION METAL CLUSTER CHEMISTRY: Fe5 AND Fe6 CLUSTERS FROM THE USE OF 6-HYDROXYMETHYL-2, 2 BIPYRIDINE 5.1 Introduction There continues to be a great interest by m any groups around the world in the synthesis and study of 3d transition metal cluster compoun ds, not least for the st ructural aesthetics possessed by such species. Other reasons for this interest are varied. For FeIII chemistry, for example, there are bioinorganic areas of relevance such as the great desire to understand and model the assembly of the polynuclear iron core of the iron stor age protein ferritin.26,30,31 There is also a materials interest in that high nuclearity iron compoun ds can sometimes exhibit unusual and occasionally novel magnetic properties, with so me of them even being examples of singlemolecule magnets (SMMs);13,20,82 the latter are molecules with a combination of a relatively large ground state spin and a significant magne toanisotropy of the easy-axis (Ising) type.20,118 That FeIII is one area where high nuclearity species are often encountered is as expected from the high charge-to-size ratio of this oxidation stat e and the resulting propensity to favor oxidebridged multinuclear products. Indeed, the forma tion of the Fe/O/OH core of ferritin that was mentioned as a bioinorganic area of interest is merely an extr eme example of such polynuclear chemistry. As a result, many large FeIII clusters have been reported to date with nuclearities up to 22.72,75,87,119 Although the exchange interactions between FeIII centers are almost always antiferromagnetic, certain Fex topologies can nevertheless posse ss large ground state spin values as a result of spin frustration. Th e latter is here defined in its more general sense of competing exchange interactions of comparable magn itude, preventing (frustrating) the preferred antiparallel alignment of a ll spins, and thus giving larger gro und state spin values than might be expected.74-78 In some cases, as mentioned above, the compounds can behave as SMMs. This is

PAGE 100

100 the case for clusters such as [Fe8O2(OH)12(tacn)6]8+,81,82,120 and [Fe4(OMe)6(dpm)6]79 for example. For the above reasons and more, there is a continuing search for new synthetic methods that can yield new polynuclear Fe/O clusters. One approach that has proven successful in this regard is the use of alcoholcontaining chelate groups that on deprotonation, can provide alkoxide groups that are excelle nt bridging units and thus foster formation of high nuclearity products.74,86,88 In chapter 4, we reported the use of deprotonated 2-{[2-(dimethylamino)ethyl]methylamino}ethanol (dmemH; Fi gure 5-1) as a new and flex ible N,N,O chelate for the synthesis of Fe3, Fe6 and two Fe7 complexes, some of which possess novel Fex topologies. As part of these continuing e fforts to synthesize new Fex clusters, we have now turned to another potential chelating group that has also never before been employed, to our knowledge, in transition metal chemistry. This is 6-hydroxymethyl-2,2 -bipyridine (hmbpH; Figure 5-1), whose deprotonated form, like dmem-, would be a potential N,N,O-chelate, but a more rigid one than dmem-. In fact, hmbpH was selected as a fusion of two ch elates that have each proven a rich source of Mx, and particularly Fex, species in the past, 2,2 -bipyridine and the anion of 2(hydroxymethyl)pyridine (hmpH). 77,109,121-126 We thus considered it a potentially viable route to new clusters, and probably of a different type than previously encountered with bpy or hmpseparately. Our first investigatio ns with hmbpH have been in FeIII chemistry and we have indeed found it to lead to new structural types of pr oducts. We herein report the syntheses, crystal structures and magnetochemical characterization of new Fe5 and Fe6 molecular species.127 5.2 Experimental Section 5.2.1 Syntheses All preparations were perfor med under aerobic conditions using reagents and solvents as received. [Fe3O(O2CPh)6(H2O)3](NO3), [Fe3O(O2CBut)6(H2O)3](NO3) and

PAGE 101

101 [Fe3O(O2CMe)6(H2O)3](NO3) were synthesized as reported elsewhere.96 The known organic molecule hmbpH was synthesized, according to Figure 5-2, using previously reported procedures.128-130 [Fe5O2(OH)(O2CMe)5(hmbp)3](ClO4)2 (5-1). To a stirred solution of Fe(ClO4)3xH2O (0.19 g, 0.54 mmol) and sodium acetate (0.18 g, 2.0 mmol) in EtOH (15 mL) was added hmbpH (0.10 g, 0.54 mmol). The resulting orange solution was stirred for 3 hours at room temperature, during which time precipitated an orange soli d. The precipitate was collected by filtration, washed with EtOH, and dried. It was then dissolved in MeCN (15 mL), filtered, and the filtrate layered with Et2O. X-ray quality crystals of 5-1 MeCN slowly grew over five days in 18% yield. These were collected by filtration, washed with MeCN and dried in vacuo ; dried solid analyzed as solvent-free. Anal. Calcd (Found) for 5-1 (C43H43N6Cl2Fe5O24): C, 37.48 (37.05); H, 3.14 (3.12); N, 6.09 (5.89). Selected IR data (cm-1): 1599(s), 15429s), 1490(m), 1402(s), 1175(m), 1068(m), 1025(m), 937(w), 820(w), 777( m), 719(s), 663(s), 600(w), 547(w), 463(m). [Fe6O2(OH)2(O2CPh)6(hmbp)4](NO3)2 (5-2) An orange-red solution of [Fe3O(O2CPh)6(H2O)3](NO3) (0.14 g, 0.14 mmol) in MeCN (20 mL) was treated with hmbpH (0.05 g, 0.27 mmol). The solution was stirred for 2 hours, filtered, and the ora nge-red filtrate left undisturbed to concentrate slowly by evaporation. X-ray quality, or ange-red crystals of 5-2 MeCNH2O formed over five days in 25% yield. These were collected by filtration, washed with MeCN and dried in vacuo Anal. Calcd (Found) for 5-2 H2O (C86H70N10Fe6O27): C, 51.37 (51.34); H, 3.51 (3.37); N, 6.97 (6.82) Selected IR data (cm-1): 3426(br), 1602(m), 1539(s), 1490(m), 1433(s), 1350(m), 1299(w), 1254(w), 1226(w), 1088(s), 1044(s), 779(m), 690(m), 662(m), 623(m), 556(m), 429(w).

PAGE 102

102 [Fe6O2(OH)2(O2CMe)6(hmbp)4](NO3)2 (5-3). An orange-red solution of [Fe3O(O2CMe)6(H2O)3](NO3) (0.11 g, 0.17 mmol) in MeCN (20 mL) was treated with hmbpH (0.065 g, 0.35 mmol). The solution was stirred for 2 hours, filtered, and the orange-red filtrate left undisturbed to concentrate slowly by evapora tion. Orange-red crystals formed over five days in 15% yield. These were collected by filtration, washed with MeCN and dried in vacuo Anal. Calcd (Found) for 5-3 H2O (C56H58N10Fe6O27): C, 41.06 (41.13); H, 3.57 (3.61); N, 8.55 (8.33). Selected IR data (cm-1): 3399(br), 1601(m), 1546(s), 1491(m), 1438(s), 1384(s), 1256(w), 1225(w), 1166(w), 1090(w), 1037(s), 905(w), 833(w), 781(m), 664(s), 645(m), 556(m), 434(w), 413(w). [Fe6O2(OH)2(O2CBut)6(hmbp)4](NO3)2 (5-4). An orange-red solution of [Fe3O(O2CBut)6(H2O)3](NO3) (0.13 g, 0.14 mmol) in MeCN (20 ml) was treated with hmbpH (0.05 g, 0.27 mmol). The solution was stirred for 2 hours, filtered, and the orange-red filtrate layered with Et2O. Orange crystals slowly grew over four days in 20% yield. Anal. Calcd (Found) for 5-4 H2O (C74H94N10Fe6O27): C, 47.01 (46.64); H, 5.01 (4.76); N, 7.41 (7.50). Selected IR data (cm-1): 3408(br), 3077(w), 2962(m), 1601( w), 1539(s), 1484(m), 1459(m), 1425(s), 1383(s), 1361(s), 1300(w), 1227(m), 1163(w), 1091(w), 1036(m), 900(w), 832(w), 785(m), 663(s), 600(m), 553(m), 434(m). 5.2.2 X-ray Crystallography Data were collected by Dr. Khalil A Abboud on a Siemens SMART PLATFORM equipped with a CCD area detector and a graphite monochromator utilizing Mo-K radiation ( = 0.71073 ). Suitable crystals of 5-1 MeCN and 5-23MeCNH2O were attached to glass fibers using silicone grease and transferred to a goniosta t where they were cooled to 173 K for data collection. Cell parameters were refined using up to 8192 reflections. A full sphere of data (1850 frames) was collected using the -scan method (0.3 frame width). The first 50 frames were re-

PAGE 103

103 measured at the end of the data collection to m onitor instrument and crystal stability (maximum correction on I was <1 %). Absorption correct ions by integration were applied based on measured indexed crystal faces. The struct ure was solved by the direct methods in SHELXTL6,46 and refined on F2 using full-matrix least-squares. The non-H atoms were treated anisotropically, whereas the hydrogen atoms were calculated in ideal positions and refined as riding on their respective carbon atoms. In 5-1 MeCN, the asymmetric unit consists of the Fe5 cluster, two ClO4 anions, and five MeCN solvent molecules. The latter molecule s were disordered and could not be modeled properly, thus program SQUEEZE,68 a part of the PLATON p ackage of crystallographic software, was used to calculate the solvent disorder area and remove its contribution to the overall intensity data. One of the ClO4 anions was disordered and was refined in two positions with their site occupation factors refined dependently. Both disorder components are H-bonded to the hydroxyl proton on O4. A total of 720 para meters were refined in the final cycle of refinement using 42942 reflections with I > 2 ( I ) to yield R1 and wR2 of 6.11 and 17.05%, respectively. In 5-2 MeCNH2O, the asymmetric unit consists of a half Fe6 cluster, one and a half MeCN molecules, one NO3 anion disordered over three posi tions, and a half water molecule which exists counter to one of the half NO3 anions. One nitrate exists with 50% occupancy, while the other has all but one O atom disorder ed. The latter was refi ned in two parts with occupation factors fixed at 20 and 30%. A total of 633 parameters we re refined in the final cycle of refinement using 15661 reflections with I > 2 ( I ) to yield R1 and wR2 of 4.85 and 12.76%, respectively. Unit cell data and structure refinement details for the two compounds are listed in Table 5-1.

PAGE 104

104 5.3 Results and Discussion 5.3.1 Syntheses The reaction of Fe(ClO4)3 with hmbpH and sodium acetate in a 1:1:4 ratio in EtOH gave an orange precipitate that afte r recrystallization from MeCN/Et2O gave crystals of the novel pentanuclear cluster [Fe5O2(OH)(O2CMe)5(hmbp)3](ClO4)2 ( 5-1 ). The acetate acts as the proton acceptor in this reaction, as well as providing ligand groups. The same product was also obtained using MeCN as the reaction solvent, but the precipitate was found to be contaminated with some other solid products. The formation of 5-1 is summarized in eq. 51. Decreasing the amount of acetate from 4 to 2 equiv reduces the reaction yi eld, as expected from eq. 5-1. Other reactions with small variations in the Fe3+:hmbpH:MeCO2 ratio also gave compound 5-1 5Fe3+ + 3hmbpH + 13MeCO2 + 3H2O [Fe5O2(OH)(O2CMe)5(hmbp)3]2+ + 8MeCO2H (5-1) Many synthetic procedures to polynuclea r iron clusters rely on the reaction of [Fe3O(O2CR)6(H2O)3]+ species with a potentially chelating ligand,84,91,92,98,119,125,126 and we thus also explored this starting material for reactions with hmbpH. In such reactions, the [Fe3O]7+ core of the trinuclear iron complex serves as a useful building bl ock for higher nuclearity species, but we have occasionally found that the exact nuclearity and structure of the product is sensitive to the identity of the carboxylate empl oyed. An example of th is is the reaction of [Fe3O(O2CR)6-(H2O)3]+ species with dmemH.93 Thus, we have also studied the product of reactions with hmbpH as a function of the carboxylate, but in this case ha ve found that we obtain the same structural type in each case. Thus, the reaction of [Fe3O(O2CR)6(H2O)3]+ (R = Ph, Me, But) with 1 3 equiv of hmbpH in MeCN led to the isolation of the corresponding hexanuclear cluster [Fe6O2(OH)2(O2CR)6(hmbp)4]2+ (R = Ph ( 5-2 ), Me ( 5-3 ), But ( 5-4 )). The formation of this family is summarized in eq. 5-2.

PAGE 105

105 2[Fe3O(O2CR)6(H2O)3]+ + 4hmbpH [Fe6O2(OH)2(O2CR)6(hmbp)4]2++ 6RCO2H + 4H2O (5-2) 5.3.2 Description of Structures 5.3.2.1 Structure of [Fe5O2(OH)(O2CMe)5(hmbp)3](ClO4)2 (5-1) A labeled representation of complex 5-1 is shown in Figure 5-3. Selected interatomic distances and angles are lis ted in Table A-10. Complex 5-1 crystallizes in monoclinic space group P 21/c. The core can be described as consisting of a [Fe4( 3-O)2]8+ butterfly-like subunit (Fe1, Fe3, Fe4 and Fe5), on the top of which is attached a [Fe( -OH)( -OR)3] unit containing Fe2. There is an O atom monoatomically bridging Fe2 to each of the four Fe atoms of the butterfly. Three of these O atoms (O3, O5, O6 ) are the alkoxide arms of the three hmbpgroups, and the fourth is the OHion (O4). The protonated OHnature of O4 was confirmed by bond valence sum (BVS) calculations,103,131 which gave a value of 1.09. Th e bipyridyl portions of two hmbpgroups chelate one each to the two wingt ip Fe atoms, Fe1 and Fe3, while the third chelates Fe2. Peripheral ligation about the core is then completed by five acetate groups in the common 1 : 1 : bridging mode. It is interesting to note that bpy itself will react with FeIII in the presence of carboxylate groups to give the Fe4 butterfly complexes of formula [Fe4O2(O2CR)7(bpy)2]+ with the two bpy groups attached at the wingtip Fe atoms.77 Thus, the bpy fragments of the hmbpchelates are giving the analogous Fe4 butterfly unit, but the alkoxide arms also then foster attachment of the fifth Fe atom. The core of complex 5-1 is unprecedented in pentanuclear FeIII chemistry. Indeed, there are re latively few pentanuclear FeIII complexes in the literature, and these have Fe5 topologies such as a square pyramid, a centered tetrahedron, and a partial cubane extende d at one face by a partial admantane unit.132-141

PAGE 106

106 5.3.2.2 Structure of [Fe6O2(OH)2(O2CPh)6(hmbp)4](NO3)2 (5-2) A labeled representation of complex 5-2 is shown in Figure 5-4, and selected interatomic distances and angles are lis ted in Table A-11. Complex 5-2 crystallizes in triclinic space group P The core can be described as a modification of the structure of 5-1 and consists of a central [Fe4( 3-O)2]8+ flattened-butterfly unit (Fe2, Fe2 Fe3 and Fe3 ) on either side of which is attached a [Fe( -OH)( -OR)2] unit containing Fe1 and Fe1'. There are now four hmbpgroups, two again on the wingtip positions of the butterfly unit, and one each on Fe1 and Fe1'. Unlike 28, there are now only three monoatomically-bridging O atoms linking Fe1 to the butterfly unit, two hmbpalkoxide arms (O3', O9) and the OHgroup (O7). The OHnature of O7 was again confirmed by BVS calculations, which gave a value of 1.16. The periphe ral ligation about the [Fe6O2(OH)2(hmbp)4]8+ core is completed by six be nzoate groups, four in the 1 : 1 : bridging mode and two in an 1 terminal mode. The main over all difference between the core structures of 5-1 and 5-2 is that one of the wingtip hmbpgroups of 5-1 has rotated by 180, bringing its alkoxide arm to the opp osite side of the molecule from the fifth Fe atom and thus allowing attachment of a sixth Fe atom. A number of other Fe6 complexes have been reported in the literature, and these possess a variety of metal topologies such as planar, twisted boat, chair, parallel triangles, octahedral, ladder-like, cyclic, etc.104 However, the only previous compounds somewhat structurally similar to 5-2 are [Fe6O2(OMe)12(tren)2]2-,105 [Fe6O2(OR)8(O2CPh)6]106 and [Fe6O2(OH)2(O2CBut)8(dmem)2].93 In these compounds there is again a central [Fe4( 3-O)2]8+ core with an additional Fe atom on each side (as in 5-2 ), but the precise means by which the latter are connected to the Fe4 unit are different from the situation in 5-2

PAGE 107

107 5.3.3 Magnetochemistry of Complexes 5-1 to 5-4 5.3.3.1 Dc Studies Solid-s tate, variable-temperature dc magnetic susceptibility data in a 0.1 T field and in the 5.0-300 K range were collected on po wdered crystalline samples of 5-1 to 5-4 restrained in eicosane. The obtained data are plotted as MT vs T in Figure 5-5. For 5-1 MT steadily decreases from 6.69 cm3mol-1K at 300 K to 4.01 cm3mol-1K at 5.0 K. The 300 K value is much less than the spin-only ( g = 2) value of 21.87 cm3mol-1K for five non-interacting FeIII ions, indicating the presence of strong antiferro magnetic interactions, as expected for oxo-bridged FeIII systems. The 5.0 K value of 4.01 cm3mol-1K suggests a spin S = 5/2 ground state. The MT vs T plots for the three complexes 5-2 to 5-4 in Figure 5-5 are very similar, indicating a minimal influence of the different carboxylate groups and supporting the conclusion that not just their formulations are identical but also their structures. MT for 5-2 H2O, 5-3 H2O and 5-4 H2O increases from 11.88, 11.11 and 11.51 cm3mol-1K at 300 K to a maximum of 14.95, 14.52 and 15.07 cm3mol-1K at 30 K, and then decreases very slightly to 14.65, 14.35 and 14.91 cm3mol-1K respectively at 5.0 K. The MT at 300 K is again much less than the spin-only value of 26.25 cm3mol-1K expected for six non interacting FeIII ions indicating the presence of strong antiferromagnetic interactions. However, the increase in MT with decreasing temperature suggests that the lowest lying spin states are of high spin values and the near-plateau value of 14.6 14.9 cm3mol-1K at low temperatures is very clos e to the spinon ly value of 15.0 cm3mol1K for an S = 5 ground state. The small decrease in MT at the lowest temperatures is very likely due to the zero-field sp litting (ZFS) within the S = 5 ground state and perhaps some weak intermolecular interactions. The differences in MT vs T for the three complexes are almost certainly just reflecting small differences in g values, intramolecular exchange coupling constants ( J ), zero-field splitting parameters ( D ), and intermolecular antiferromagnetic

PAGE 108

108 interactions, but the overall almost identical MT vs T plots indicates these factors are nevertheless almost identical for 5-2 to 5-4 To confirm the above ground state sp in estimates, variable-field ( H ) and -temperature magnetization (M ) data were collected in the 0.1 to 7.0 T and 1.8 to 10 K ranges. The resulting data for 5-1 are plotted in Figure 5-6(a) as reduced magnetization ( M/N B) vs. H / T where N is Avogadro's number and B is the Bohr magneton. The saturati on value at the highest fields and lowest temperatures is ~4.86, as expected for an S = 5/2 ground state and g slightly less than 2; the saturation value should be gS in the absence of complications from low-lying excited states. The data were fit, using the program MAGNET,53 by diagonalization of the spin Hamiltonian matrix assuming only the ground state is popu lated, incorporating axial anisotropy ( D z 2) and Zeeman terms, and employing a full powder average. The corresponding spin Hamiltonian is given by eq. 5-3, where z is the easy-axis spin operator, g is the electronic g factor, 0 is the vacuum permeability, and H is the applied field. The last term in eq. 5-3 is the Zeeman energy associated with an applied magnetic field. H = D z 2 + gB0 H (5-3) The best-fit for 5-1 is shown as the solid lines in Fi gure 5-6(a), and was obtained with S = 5/2 and either of the tw o sets of parameters: g = 1.96 and D = 0.75 cm-1, or g = 1.95 and D = 0.59 cm-1. Alternative fits with S = 3/2 or 7/2 were rejected because they gave unreasonable values of g and D It is common to obtain two acceptable fits of magnetization data for a given S value, one with D > 0 and the other with D < 0, since magnetization fits are not very sensitive to the sign of D This was indeed the case for the magnetization fits for all complexes 5-1 to 5-4 In order to assess which is the superior fit in all th ese cases, and also to ensure that the true global minimum had been located for each compound, we calculated the root-mean-square error surface

PAGE 109

109 for the fits as a function of D and g. For 5-1 the error surface (Figure 5-6(b)) clearly shows only two minima with positive and negative D values, with both fits being of comparable quality. The reduced magnetization pl ots saturate at 9.78 for 5-2 H2O, 9.64 for 5-3 H2O and 9.88 for 5-4 H2O, suggesting an S = 5 ground state and g < 2. The best fit for 5-2 H2O is shown as the solid lines in Figure 5-7(a, top), and was obtained with S = 5 and either g = 2.00 and D = 0.36 cm-1, or g = 1.97 and D = -0.20 cm-1. In this case, the fit error surface (Figure 5-7b, top) clearly shows that the fit with positive D is far superior, s uggesting that this is the true sign of D The best fit for 5-3 H2O was obtained with S = 5 and either g = 1.98 and D = 0.46 cm-1, or g = 1.94 and D = -0.21 cm-1. For 5-4 H2O, the best fit was with S = 5 and either g = 2.02 and D = 0.36 cm-1, or g = 1.99 and D = -0.19 cm-1. The corresponding figures and the two-dimensional D vs g error plots for 5-3 H2O and 5-4 H2O are provided in Figure 5-7 (middle and bottom respectively). 5.3.3.2 Rationalization of the Ground State Spin It is interesting to try to rationalize the observed ground state spin values of 5-1 and 5-2 It is assum ed that all Fe2 pairwise exchange inte ractions are antiferromagnetic, as is essentially always the case for high-spin FeIII, and there will thus be compe ting antiferromagnetic exchange interactions and spin frustration effects within the many Fe3 triangular units in these complexes. The ground state of 5-1 is the easiest to rationalize: the discrete Fe4 butterfly (rhombus) topology is known to usually give an S = 0 ground state as a result of the antiferromagnetic interactions along the four outer (wingtipbody) edges overcoming, and thus frustrating, the diagonal (bodybody) interaction.77,109-111,142,143 The structure of 5-2 comprises such an Fe4 unit with an additional Fe above it, and assuming the same spin alignments as in the discrete Fe4 molecules, then the ground state spin alignments are predicte d to be those shown in Figure 5-8(a), giving the S = 5/2 ground state observed experi mentally. Note that whether the spin of the fifth Fe atom is

PAGE 110

110 aligned parallel to the wingtip sp ins (as shown) or parallel to the body spins, an S= 5/2 ground state will still result as long as the interactions within the butterfly are stronger than those between it and the fifth Fe atom. This seems re asonable given that the in teractions within the butterfly involve monoatomically bridging oxide ions. For 5-2 we can again rationalize the ground state, using a simple extension of the argument for 5-1 on the basis of a central S = 0 planar-butterfly unit coupling with the fifth and sixth Fe atoms as shown in Figure 5-8(b). This will give an overall S = 5 ground state for 5-2 as observed experimentally. It should be noted that we have sought to rationalize the ground states of 5-1 and 5-2 on the basis of previous observations for the Fe4 butterfly units, and with as straig htforward a description as possible. Thus, we have not invoked intermed iate spin alignments of individua l spins. In reality, the spin alignments leading to the observed S = 5/2 and 5 ground states could be more complicated than shown in Figure 5-8. None of the compounds exhibited an out-of-pha se ac magnetic susceptibility signal down to 1.8 K in an ac field of 3.5 Oe oscillating with frequencies up to 997 Hz, indicating that they do not exhibit a barrier large enough vs kT, down to 1.8 K at least, to exhibit slow relaxation of their magnetization vectors, i.e. they ar e not single-molecule magnets. This is not surprising that the D values for 5-2 to 5-4 were concluded to be positive, whereas negative D values are required to yield the easy-axis (Ising) anis otropy necessary for SMMs. For 5-1 we could not conclude the sign of D : assuming it is negative, the combination of S = 5/2 and D = -0.59 cm-1 would yield an upper limit to the magnetiza tion relaxation barrier ( U ) of U = ( S2-1/4)| D | = 3.5 cm-1 = 5.1 K. Remembering that the actual or effective barrier ( Ueff) is significantly less than U it is not surprising that even with a negative D complex 5-1 does not display slow relaxation down to 1.8

PAGE 111

111 K. Studies at much lower temperatures would be required to search for what would at best be a tiny barrier. 5.4 Conclusions We have reported the initial use of a new N, N,O based tr identate chelate in coordination chemistry, one that amalgamates the chelating property of 2,2 -bipyridine (bpy) with the chelating/bridging properties of the anion of 2-(hydroxymethyl)pyridine (hmpH). The resulting hmbphas been employed in FeIII chemistry, and it has provid ed clean access to four new polynuclear FeIII clusters 5-1 to 5-4 The structures of 5-2 to 5-4 are concluded to be the same, given their identical formulation and almost s uperimposable magnetic properties. Note that identical formulation by itself does not mean iden tical structure: we reported two compounds in chapter 4, [Fe7O4(O2CPh)11(dmem)2] ( 4-1 ) and [Fe7O4(O2CMe)11(dmem)2] ( 4-2 ) that have the same formula (except for the carboxyla te) but very different structures. The structures of 5-1 and 5-2 show the manifestation of the hybrid bpy/hmpnature of hmbpin that the bpy portion gives an Fe4 butterfly subunit, as does bpy itself, while the alkoxide arm acts as an additional bridging group and raises the nuclearity to five or six. As a result, the complexes have significant ground state spin values of S = 5/2 and 5, respectively.

PAGE 112

112 Table 5-1. Crystallographic Data for 5-1 MeCN and 5-2 MeCNH2O. 5-1 5-2 Formulaa C53H58Cl2Fe5N11O24 C92H79Fe6N13O27 Fw, g/mola 1583.25 2133.77 Space group P 21/c P a 21.6352(2) 13.8233(6) b 13.4154(6) 14.0671(6) c 23.1971(11) 14.2856(6) 90 65.175(2) 102.456(2) 70.147(2) 90 89.561(2) V 3 6574.4(5) 2341.55(17) Z 4 1 T K 173(2) 173(2) b 0.71073 0.71073 calc, g/cm3 1.600 1.512 mm-1 1.244 0.990 R1 c,d 0.0611 0.0487 wR2 e 0.1705 0.1276 a Including solvate molecules. b Graphite monochromator. c I > 2 ( I ). d R 1 = (|| Fo| |Fc||) / | Fo|. e wR 2 = [ [ w ( Fo 2 Fc 2)2] / [ w ( Fo 2)2]]1/2, w = 1/[ 2( Fo 2) + [( ap )2 + bp], where p = [max (Fo 2, O) + 2 Fc 2]/3. N N HO N N HO dmemH hmbpH Figure 5-1. Structure of ligands. Figure 5-2. Synthetic sc heme for hmbpH (i) H2O2, CF3CO2H (Ref. 117) (ii) Me3SiCN, PhCOCl, CH2Cl2 (Ref. 118) (iii) NaOMe, Me OH (Ref. 118) (iv) (a) NaBH4, EtOH (b) H2SO4, H2O (Ref. 119)

PAGE 113

113 Figure 5-3. Labeled representa tion of the structure of 5-1 Hydrogen atoms have been omitted for clarity. Color code: FeIII, green; O, red; N, blue; C, grey. Figure 5-4. Labeled representa tion of the structure of 5-2 Hydrogen atoms have been omitted for clarity. Color code: FeIII, green; O, red; C, grey; N, blue.

PAGE 114

114 Figure 5-5. Plots of MT vs T for complexes 5-1 ( ), 5-2 H2O ( ), 5-3 H2O ( ) and 5-4 H2O ( ). Figure 5-6 (a) Plot of reduced magnetization ( M / N B) vs H / T for complex 5-1 The solid lines are the fit of the data; see the text for the fit parameters. (b) Two-dimensional contour plot of the fitting error surface vs D and g for 5-1

PAGE 115

115 Figure 5-7 (a) Plot of reduced magnetization ( M / N B) vs H / T for complexes 5-2H2O (top), 53H2O (middle), 5-4H2O (bottom). The solid lines are the fit of the data (b) Twodimensional contour plot of the fitting error surface vs D and g for complexes 5-2H2O (top), 5-3H2O (middle), 5-4H2O (bottom).

PAGE 116

116 Fe Fe O O Fe Fe O O Fe O O Fe Fe O O Fe Fe O Fe O O O Fe O O Fe (a) (b) Figure 5-8. Rationalization of sp in ground state of complex (a) 5-1 and (b) 5-2

PAGE 117

117 CHAPTER 6 NEW STRUCTURAL TYPES IN POLYNUCLEAR IRON CLUSTERS INCORPORATING O,N,N, O LIGAND: A SNAKE LIKE CHAI N AND A SUPRAMOLECULAR DIMER OF SMMs 6.1 Introduction Molecu lar clusters of magnetic transition meta l ions have been gene rating great interest since the discovery that these molecules can be have as nanoscale magnets and show magnetic bistability of pure molecular origin whic h can be used for information storage.13 The first molecule to show this behaviour was [Mn12O12(O2CMe)16(H2O)4,15 which led to the field of single molecule magnetism. Since then much effort has been put into fi nding new systems with interesting magnetic properties. Although a variety of Mn containing SMMs have been reported in recent years but to date, SMMs based on FeIII are still rare.22 In addition, oxo-bridged FeIII clusters of various nuclearitie s have been studied as models of Fe sites in proteins and enzymes, as well as models of intermediate stages of the growth of the giant Fe/O core of the Fe storage protein ferritin.144 In fact, the biologi cal and magnetic areas essentially involve the same Fe/O chemistry as emphasized by the fact that Fe/O core of ferritin can be considered a nanoscale magnetic particle and has been investigated for quantum tunneling effects of magnetization.31 SMMs derive their properties from the combin ation of a large ground-state spin quantum number ( S) and a magnetoanisotropy of the easy-axis (Ising)-type (negative zero-field splitting parameter, D ), rather than from intermolecular inte ractions and long-range ordering as in traditional magnets.13 This combination leads to a significant barrier ( U ) to relaxation (reorientation) of the magnetization vector, whose maximum value is given by S2| D | or ( S2 1/4) | D | for integer and half-int eger spin, respectively. The use of high spin FeIII can theoretically lead to large spin values even for quite small number of paramagnetic centers.20 Although the

PAGE 118

118 interactions between FeIII are generally antiferromagnetic, some topological arrangements can result in large ground state spin due to the pheno menon of spin frustration. In general, molecules with high spin ground state have many spin states which are thermally populated at room temperature; therefore the SMM properties manifest itself at low temperature when excited states are depopulated. FeIII clusters do not provide SMMs by design because the conditions for large Ising type anisotropy depend very unpredictably on minor feat ures of the coordination environment of the metal ion.20 The above considerations a nd others continue to stim ulate groups around the world to develop new synthetic methods that can yield new polynuclear Fe/O clusters. However there is no obvious and guaranteed route to such species. Different approaches to the synthesis of ironoxo clusters are controlled hydrol ysis or alcoholysis reactions of simple iron salts or more complicated starting materials.30,122 Also, trinuclear iron oxo clusters have also proved to be very useful building blocks for the preparation of higher nuclearity clusters in the presence of appropriately chosen chelate.145 One successful strategy in produc ing polynuclear clusters is the use of alkoxide-based ligands,32 since this functionality is an excellent bridging group that fosters higher nuclearity product formation. In chapter 4, we reported our resu lts in polynuclear iron clusters with N,N,O based chelate 2-{[2-(d imethylamino)ethyl] methylamino}ethanol (dmemH).93 In the present work, we have been inve stigating the use of a new O,N,N,O-based chelate N ,N '-bis(2-hydroxyethyl)ethylenediamine (heenH2; Figure 6-1) for transition metal cluster chemistry. It has been used before in the literature to make mononuclear Pt and Cu molecules in its fully protonated form but ther e are no previous use of it in polynuclear metal (i.e. cluster) chemistry, protonated or otherw ise. Our first investigations with heenH2 have been in FeIII chemistry, and we have indeed found it to l ead to new structural types of products. We

PAGE 119

119 herein report the syntheses, crystal structures and magentochemical charac terization of four new iron complexes Fe6, Fe7, Fe9 and Fe18.145,146 6.2 Experimental Section 6.2.1 Syntheses All preparations were perfor med under aerobic conditions using reagents and solvents as received. [Fe3O(O2CBut)6(H2O)3]NO3 and [Fe3O(O2CPh)6(H2O)3]NO3, were synthesized as reported elsewhere.96 [Fe18O8(OH)2(O2CBut)28(heen)4] (6-1) To a stirred dichloromethane solution (20 mL) of [Fe3O(O2CBut)6(H2O)3]NO3 (0.20 g, 0.21 mmol) was added heenH2 (0.05 g, 0.31 mmol) and the reaction mixture was stirred for 3 hours and th en filtered to remove undissolved solid and layered with pentane. Orange crystals of 6-1 C5H12CH2Cl2 suitable for X-ray crystallography formed over a week in 30% yield. Anal. calcd (Found) for 6-1 (C164H308N8Fe18O74): C, 42.98 (43.25); H, 6.82 (6.99); N, 2.44 (2.43) Selected IR data (cm-1): 2961(s), 2928(m), 2871(w), 1580(s), 1551(s), 1484(vs), 1421(vs), 1376(m), 1228(m), 1090(w), 906( w), 787(w), 663(w), 602(m), 508(w), 438(m). [Fe9O4(OH)4(O2CPh)13(heenH)2] (6-2) Ligand heenH2 (0.11 g, 0.74 mmol) was added to a stirred MeCN solution (15 mL) of FeCl3H2O (0.20 g, 0.74 mmol) and NaO2CPh (0.21 g, 0.15 mmol). The reaction mixture was stirred for 2 hours and then filtered. Slow evaporation of the filtrate gave X-ray quality crystals of 6-2 MeCN in three weeks days in 15% yield. Anal. calcd (Found) for 6-2 (C103H99N4Fe9O38): C, 49.41 (48.99); H, 3.98 (3.90); N, 2.24 (2.57). Selected IR data (cm-1): 3421(br), 1599(m), 1558(s), 1597(s) 1176(w), 1025(w), 827(w), 715(m), 676(m), 592(w), 469(m). [Fe7O3(OMe)3(MeOH)1(heen)3Cl4.5(H2O)1.5]Cl1.25[FeCl4] (6-3) To a stirre d solution of FeCl2H2O (0.20 g, 1.0 mmol) in methanol (10 mL) was added heenH2 (0.15 g, 1.9 mmol) and

PAGE 120

120 the resulting solution was refluxed for 2 hours. It was filtered hot and kept for slow evaporation. Orange needle like crystals of 6-3 2MeOHH2O appeared in seven days in 10 % yield. Anal. Calcd (Found) for 6-3 (C22H58N6Fe7.25O14.5Cl6.75): C, 20.60 (21.09); H, 4.56 (4.93); N, 6.56 (6.08). Selected IR data (cm-1): 3224(s,br), 2952(m), 2868(m), 1634(m), 1455(m), 1344(w), 1236(w), 1090(s), 1066(s), 963(s), 886(w), 810(w) 670(s), 633(m), 558(m), 505(m), 407(m). [Fe6O2(O2CPh)5(heen)3(heenH)](ClO4)2 (6-4) Method A. A solution of Fe(ClO4)3 (0.20 g, 0.56 mmol) and NaO2CPh (0.16 g, 0.11 mmol) in EtOH (15 mL) was treated with heenH2 (0.08 g, 0.57 mmol) and stirred for 3 hours. The re sultant red brown solution was filtered to remove NaCl and the filterate was left undisturbe d for slow evaporation. X-ray quality crystals of 6-4 EtOH.5H2O appeared in 5 days in 25% yield. Anal. calcd. (Found) for 6-4 H2O (C59H84N8Fe6O29Cl2): C, 39.92 (39.84); H, 4.77 (4.62); N, 6.31 (6.19). Selected IR data (cm-1): 2968(m), 1596(m), 1557(m), 1400(s), 1341(w) 1302(w), 1176(w), 1190(s), 1024(m), 981(m), 829(w), 724(s), 624(m), 533(w), 468(m). Method B. An orange red solution of [Fe3O(O2CPh)6(H2O)3]NO3 (0.20 g, 0.19 mmol) in EtOH (15 mL) was treated with heenH2 (0.06 g, 0.39 mmol) and NaClO4 (0.05 g, 0.41 mmol) and the resulting solution stirred for 2 hours at room temperature. Next, it was filtered to remove undissolved starting material and the filterate was allowed to stand undisturbed at room temperature. Orange crystals of the product formed over 5 days in 15 % yield. 6.2.2. X-ray Crystallography Data were collected by Dr. Khalil A Abboud on a Siemens SMART PLATFORM equipped with a CCD area detector and a graphite monochromator utilizing Mo-K radiation ( = 0.71073 ). Suitable crystals of 6-1 C5H12CH2Cl2, 6-2 MeCN, 6-3 MeOHH2O and 64EtOH.5H2O were attached to glass fibers using s ilicone grease and transferred to a goniostat

PAGE 121

121 where they were cooled to 173 K for data colle ction. Cell parameters were refined using up to 8192 reflections. A full sphere of data (1850 frames) was collected using the -scan method (0.3 frame width). The first 50 frames were remeasur ed at the end of data collection to monitor instrument and crystal stability (maximum corre ction on I was < 1 %). Ab sorption corrections by integration were applied based on measured inde xed crystal faces. The structure was solved by the Direct Methods in SHELXTL6,46 and refined using full-matrix least squares. The non-H atoms were treated anisotropically, whereas the hydrogen atoms were calculated in ideal positions and were riding on their respective carbon atoms. Refinement was done using F2 In 6-1 the asymmetric unit consists of a half Fe18 cluster, two pentane molecules and two dichloromethane molecules. The latter molecu les were disordered and could not be modeled properly, thus program SQUEEZE68 was used to calculate the solv ent disorder area and remove its contribution to the overall intensity data. A total of 1183 parameters were included in the final cycle of refinement 42915 reflections with I > 2 ( I ) to yield R1 and wR2 of 4.67 % and 12.52 %, respectively. In 6-2 the asymmetric unit consists of the cluste r and 9 acetonitrile molecules. The latter were disordered and could not be m odeled properly, thus program SQUEEZE68 was used to calculate the solvent disorder area and remove its contribution to the overall intensity data. There are three phenyl groups that are disordered (a t C81, C91 and C121) and each was refined as a rigid group and in two parts with their site occu pation factors dependently refined. The last two of them cause a disorder in an uncoordinate d EtOH arm of the N4 diamine. Both hydroxy group protons of the uncoordinated diamines were calcul ated in idealized positions and refined riding on their O atoms. A total of 1375 parameters were included in the final cycle of refinement using 42636 reflections with I > 2 ( I ) to yield R1 and wR2 of 7.41 % and 18.03 %, respectively.

PAGE 122

122 In 6-3 the asymmetric unit consists of one Fe7 cluster and and FeCl4 located on a 222 symmetry site (thus is in the asymmetric unit). The structure has se veral disorders one of which is in the N1 ligand where two CH2 units are disordered and were refined in two parts with their site occupation factors de pendently refined. The O4 pos ition is a disorder between a coordinated water molecule and a methanol solv ent (O18) in one parts against a coordinated methanol ligand and a water so lvent (O19) molecule. Anothe r disorder has a coordinated methanol and a solvent methanol in one part (O11 and O15) against c oordinated water and a solvent methanol molecule (O11 and O16). The coordinated Cl4 ligand is disordered with a water (O4 ) ligand while a lattice Cl7 anion is di sordered against a lattice water (O7 ) molecule. The last two disorders lead to ba lanced charges in the asymmetric unit. A total of 566 parameters were refined in the final cycle of refinement using 11605 reflections with I > 2 ( I ) to yield R1 and wR2 of 4.64 % and 10.44 %, respectively. In 6-4 the asymmetric unit consists of a cluste r, a perchlorate anion, disordered ethanol molecules and three partial water molecules. The cluster has one arm of the diamine ligand disordered against a benzoate anion. Due to sy mmetry, the two parts of the disorder were 0.5 occupation factors. The perchlorat e anion is disordered and was refined in two parts with Cl and Cl'. Further, the Cl' part was also disordered al ong a three fold rotation axis. The ethanol solvent molecule was also disordered and refined in two parts with their site occupation factors dependently refined. A total of 586 parameters we re included in the final cycle of refinement using 8642 reflections with I > 2 ( I ) to yield R1 and wR2 of 6.22 % and 15.51 %, respectively. Unit cell data and details of the structure refinements for 6-1 to 6-4 are listed in Table 6-1.

PAGE 123

123 6.3 Results and Discussion 6.3.1 Syntheses Reaction of [Fe3O(O2CBut)6(H2O)3]+ and heenH2 in a 2 : 3 molar ratio in CH2Cl2, followed by layering with pentanes, gave orange crystals of an octadecanuclear product [Fe18O8(OH)2(O2CBut)28(heen)4] ( 6-1 ) with the core topology not previously encountered. Its formation is summarized in eq. 6-1. Use of me thylene chloride as the reaction medium was found to be very essential for obtaining a cl ean product. Using acetonitrile as the reaction medium gives a mixture of white and orange crys tals, which were difficult to separate, and hence could not be characterized. Further we decided to use polar alcoholic solv ents seeking formation of higher nuclerity products via methanolysis or ethanolysis, but alwa ys insoluble yellow precipitates were obtained. 6[Fe3O(O2CBut)6(H2O)3]+ + 4heenH2 [Fe18O8(OH)2(O2CBut)28(heen)4] + 14H2O + 8ButCO2H + 6H+ (6-1) A similar reaction involving FeCl3, NaO2CPh and heenH2 in 1:2:1 molar ratio in MeCN, gave [Fe9O4(OH)4(O2CPh)13(heenH)2] ( 6-2 ) (eq. 6-2). On the other hand, reaction of Fe(ClO4)3, O2CPhand heenH2 in 1:2:1 ratio in ethanol gave [Fe6O2(O2CPh)5(heen)3(heenH)](ClO4)2 ( 6-4 ) (eq. 6-3). The reaction procedure employed contai ns an excess of ligand over that required by stoichiometric reaction and this mi ght be beneficial in providing H+ acceptors. 9FeCl3 + 13PhCO2 + 2heenH2 + 8H2O [Fe9O8(O2CPh)9(HO2CPh)4(heenH)2] + 14H+ + 27Cl(6-2) 6Fe(ClO4)3 + 5PhCO2 + 4heenH2 +2H2O [Fe6O2(O2CPh)5(heen)3(heenH)]2+ + 18ClO4 + 11H+ (6-3) In both these reactions (eq. 6-2 and 6-3) choice of solvent and ratio of starting materials is very crucial as other ratios do not give clean reaction. In th e former reaction, using lower

PAGE 124

124 concentration of sodium benzoate and heenH2 gives the crystals of trinuclear iron cluster and using higher concentration of sodium benzoate and ligand gives messy reactions. Additionally, nothing crashes out of reaction mixture if ethanol is used as the reaction medium. In the latter reaction also, 1:2:1 ratio is very important in getting single crystals. In ot her ratios, different products are obtained, whose IR is different from 6-4 but these could not be characterized as single crystals could no t be obtained. Further, the reacti on mixture gives oily droplets if acetonitrile is used as the solvent, making ethanol absolutely essential for the formation of 6-4 Refluxing FeCl2 and heenH2 in 1:1 ratio in MeOH gave 6-3 while nothing comes out of reaction mixture if EtOH is used instead of MeOH. Using lowe r or higher concentration of ligand also gives the same product but ve ry bad quality crys tals are obtained. It is clear that the r eactions that lead to 6-1 to 6-4 are very complicated, and the reaction solutions likely contain a complicated mixture of several species in equilibrium. In such cases, factors such as relative solubil ity, lattice energies, crystallizati on kinetics, and others determine the identity of the isolated products, and one (o r more) of these factor s is undoubtedly the reason that the reaction product change s from an octadecanuclear comp lex to a nonanuclear complex on changing the carboxylate employed from pivalate to benzoate. 6.3.2 Description of Structures 6.3.2.1 Structure of [Fe18O8(OH)2(O2CBut)28(heen)4] (6-1) A labeled representation of 6-1 is shown in Figure 6-2. Select ed interatomic distances and angles are given in Table A-12. Complex 6-1 crystallizes in triclinic P space group. The structure of centrosymmetric 6-1 comprises a remarkable Fe18 cluster that can be described as seven [Fe2O2] rhombs linked into a chain, and attached to four end Fe atoms (Fe8, Fe8', Fe9, Fe9'). Alternatively, it can be bett er described as the linkage by heen2alkoxide arms O5 and O5' of two central [Fe4( 3-O)2]8+ butterfly units (Fe1, Fe2, Fe3 and Fe4, and its symmetry partner),

PAGE 125

125 and then connected to additional [Fe4( 3-O)2]8+ butterfly units at each end via intermediate Fe atoms Fe5 and Fe5', the bridges being heen2alkoxide arms O16 and O17 on one side of Fe5, and heen2alkoxide arm O18 and a hydroxide ion O19 on the other side. Peripheral ligation is provided by a total of four chelating heen2and twenty-eight pivalate groups, twenty-two of the latter in their common 1: 1: -mode, four in a 2 chelating mode on Fe8, Fe8',Fe9 and Fe9', and two in a 1 terminal mode on Fe4 and Fe4'. Two of the four heen2groups are in a 2: 1: 1: 2: 3mode (N3 and N4 are bound to Fe5 while O17 is bridging Fe4 and Fe5, and O18 is bridging Fe5 and Fe6) and the other two are in 2: 1: 1: 2: 3mode (N1, N2 are chelating Fe3 and O5 is bridging Fe1 and Fe1', and O16 is bridging Fe4 and Fe5). The overall topology of 6-1 is chain-like and resembles a double-headed serpent or alligator with both sets of jaws wide open. Such a molecule is not only unprecedented in Fe chemistry, it represents the highe st-nuclearity, chain-like metal-containing molecule to be yet discovered, and can reasonably be called a molecular chain. The next longest such molecular chain is a Cr12Ni species.147 There are only two previous Fe18 clusters known, and they are both wheel complexes, i.e. closed molecular chains.89,148 6.3.2.2 Structure of [Fe9O4(OH)4(O2CPh)13(heenH)2] (6-2) A labeled representation of 6-2 is shown in Figure 6-3 with selected geometric parameters listed in Table A-13. Complex 6-2 crystallizes in the triclinic P space group. The structure consists of pairs of Fe9 clusters arranged as head-to-head dimers. Each Fe9 contains two [Fe4( 3O)( 3-OH)]8+ butterfly-like sub-units (Fe1, Fe5, Fe 6, Fe7, O5, O27 and Fe4, Fe7, Fe8, Fe9, O21, O31) fused at body atom Fe7 (Fi gure 6-3), and attached to two additional Fe atoms Fe2 and Fe3 by two 3-OH(O6 and O11) and two 3-O2ions (O7 and O14). There are thirteen PhCO2 groups, nine 1: 1: -bridging and four 1 terminal on Fe1, Fe4, Fe6 and Fe8, with their noncoordinated O atom intramolecularly H-bonded to OHions (O6, O11, O5 and O21,

PAGE 126

126 respectively). Each heenHgroup is a tridentate chelate to an Fe atom (Fe5 and Fe9) and bridges through its deprotonated arm to a neighbor ing Fe atom. Each remaining heenHprotonated arm (O30 and O38) is unbound and involved in hydrogen-bonding. The one on Fe9 forms a H-bond to benzoate O atom O32 (O38...O32 = 2.788 ) w ith no disorder. The one on Fe5, however, is disordered, forming intramolecular H-bonds to benzoate O atom O28 (O30...O28 = 2.882 ) or to the same heenHarm of the corresponding chelate on the neighboring Fe9 molecule (O30...O39 = 2.753 ) (Figure 6-4). Refinement of th e disorder components gave an essentially statistical 36:64% mixture of the intramolecular and intermolecu lar H-bonding situations, since there are two equivalent forms of the latter, as shown in Figure 64. Thus, 2/3 of the molecules in the crystal are linked within H-bonded [Fe9]2 dimers, whereas the other third of the molecules are within non-H-bonded [Fe9]2 dimers. The core of 6-2 is unprecedented in nonanuclear Fe (III) chemistry. Indeed, there are relatively few Fe9 clusters known in literature and they are described as ferric Triple-Decker, nonanuclear ring, rhomb-like array, two distorted Fe4 tetrahedra linked to another Fe atom via oxide and alkoxides, central Fe atom surrounded by four dinuclear Fe2 units etc.58,80,149-152 6.3.2.3 Structure of [Fe7O3(OMe)3(MeOH)1(heen)3Cl4.5(H2O)1.5]Cl1.25[FeCl4] (6-3) A labeled representation of 6-3 is shown in Figure 6-5. Select ed interatomic distances and angles are given in Table A-14. Complex 6-3 crystallizes in the hi gh symmetry orthorhombic I222 space group and it consists of a planar arrangement of six FeIII ions with a seventh central FeIII ion 1.437 below the Fe6 plane. The central Fe atom, Fe1, is tetracoordinated. It establishes bonds with three 3-O2ions and with a terminal Cl ligand. Every peripheral iron is connected to one of its adjacent Fe atoms through one -OMeand one alkoxide group from heen2and to the neighboring iron atom through th e second alkoxide arm of heen2and a 3-O2bridge, which in turn are bridging the alternate peripheral iron atoms to the central tetrahedral iron atom. N atoms

PAGE 127

127 of the heen2group complete the octahedral coordination of alternati ng iron atoms, Fe3, Fe5 and Fe7 while terminal Cl ligand and H2O/MeOH molecules complete the octahedral coordination of alternating iron atoms Fe2, Fe4 and Fe6. There is one molecule of FeCl4 anion per four formula units of Fe7 for charge balance. Additionally there are lattice Clions for charge balance and mol ecules of water and methanol as solvents of crystallization. Ther e are several features which ma kes this structure especially remarkable. First, it is one of very few polynuc lear iron-oxo complexes with an odd number of metal ions. Second striking feature of this compound is rather unusual coordination of central iron atom. Fe1 has an almost regular tetrahedral geometry with O-Fe-O angles ranging between 108.1-111.4. The core topology of 6-3 is new within FeIII chemistry. There are only a few Fe7 complexes in the literature and they are de scribed as cagelike, disklike and domelike.98,101,102 Also, we reported two new heptanuclear clusters in chapter 4, [Fe7O4(O2CPh)11(dmem)2] and [Fe7O4(O2CMe)11(dmem)2]. These complexes had the same formula except for the identity of the carboxylate but the structures were very different. The former had two [Fe4( 3-O)2]8+ butterfly units sharing a common body iron atom wh ile in the other a number of Fe3O triangular units were linked in an unusual way.93 6.3.2.4 Structure of [Fe6O2(O2CPh)5(heen)3(heenH)](ClO4)2 (6-4) A labeled representation of 6-4 is shown in Figure 6-6. Select ed interatomic distances and angles are given in Table A-15. Complex 6-4 crystallizes in triclinic P space group. The asymmetric unit consists of half of the cluster, a perchlorate ion and an ethanol molecule and one and a half water molecule as the solvents of cr ystallization. The structure of the centrosymmetric cation of 6-4 can be described as a central [Fe4( 3-O)2]8+ butterfly-like unit (Fe1, Fe1', Fe2 and Fe2') connected through both its body (Fe1, Fe 1') and wingtip atoms (Fe2, Fe2') to two

PAGE 128

128 additional Fe atoms Fe3 and Fe3' by heen2(O8, O9, O10) alkoxide arms. At one end of the molecule (Fe3), the remaining heen2alkoxide arm (O11) binds terminally; at the other end (Fe3'), this arm (O20) is protonated and unbound (i.e. a heenHgroup) and there is instead a PhCO2 bound terminally (Figure 6-6; only benzoate ipso C atoms shown). These two situations at the two ends are, of cour se, disordered by the centre of symmetry which makes Fe3 and Fe3 equivalent; only one of the diso rder components is shown in Fi gure 6-6. Ligation is completed by four benzoate groups in the common 1: 1: -mode bridging the body and wingtip iron atoms of the central butterfly unit, only the ipso C atoms of benzoate rings have been shown for clarity in Figure 6-6. A number of Fe6 complexes have been previously reported possessing a variety of metal topologies such as planar, twisted boat, chair, parallel triangles, linked triangles, fused butterflies, octahedral, ladder-like, cyclic etc.104 6.3.3 Magnetochemistry of complexes 6-1 to 6-4 6.3.3.1 Dc Studies Solid-state, variable temperature dc m agnetic susceptibility data in a 0.1 T and 5.0-300 K range were collected on powde red crystalline samples of 6-1 to 6-4 restrained in eicosane. The obtained data are plotted as MT vs T in Figure 6-7. For 6-1 MT steadily decreases from 22.83 cm3Kmol-1 at 300 K to 1.16 cm3Kmol-1 at 5.0 K indicating predominant antiferromagnetic interactions between FeIII magnetic centers. Value of MT at 300 K is much less than spin-only (g = 2.0) value of 78.75 cm3Kmol-1 for eighteen non-interacting FeIII ions, the behavior with decreasing temperature and the low value of MT at 5 K are indicative of S = 0 ground state spin. This is not unexpected given that this is the most common ground state for FeIII x clusters where x is even. There are common exceptions to this rule however, when the topology is such so as to introduce competing antiferromagnetic exchange inte ractions and spin frus tration effects that result in often significant ground state S values. For complex 6-1 the S = 0 ground state can be

PAGE 129

129 rationalized as shown in Figure 6-8: the central butterfly uni ts are known to exhibit spin frustration effects with in their triangular s ubunits and to possess an S = 0 ground state as a result of the four wingtip-body interactions of the four edges ove rcome (frustrating) the body-body interaction.77 The antiferromagnetic interactions betw een separate butterfly units and between them and Fe5/Fe5 then lead to an expected S = 0 ground state, as found experimentally. Of course, since 6-1 has an S = 0 ground state, the depicted spin alignments in Figure 6-8 represent only one of the component wave functions of the ground state eigenstate of the molecule. For 6-2 the value of the MT gradually decreases from 13.35 cm3Kmol-1 at 300 K to 11.65 cm3Kmol-1 at 150 K, stays essentially constant until 60 K and then decreases to 8.73 cm3Kmol-1 at 5.0 K. The 300 K value is again less than the spin-only value of 39.37 cm3Kmol-1 expected for nine non-interacting FeIII ions, indicating predominantly an tiferromagnetic interactions. The 5.0 K value suggests S = 7/2 ground state. For 6-3 the value of MT gradually increases from 29.92 cm3Kmol-1 at 300 K to 48.16 cm3Kmol-1 at 50 K before dropping slowly to 32.83 cm3Kmol-1 at 5.0 K. The decrease of MT at low temperature may be a consequence of zeeman effects from the DC field and/or the presence of antiferromagnetic interactions between clusters. Subtracting out the MT contribution for the [FeCl4]anion indicates an S = 15/2 ground state. For 6-4 H2O, the value of MT steadily decreases from a value of 10.05 cm3Kmol-1 at 300 K to 8.13 cm3Kmol-1 at 70 K and then rises to a value of 10.32 cm3Kmol-1 at 5.0 K. The 300 K value is again much less than the spin-only value of 26.25 cm3Kmol-1 expected for six non-interacting FeIII ions, indicating predominant antiferromagnetic interactions. The increase in MT as the temperature then decreases suggests the lowest lying states are of high spin values, and the value at 5.0 K suggests S = 4 ground state.

PAGE 130

130 To confirm the above ground state spin estimates, variable field ( H ) and temperature magnetization (M ) data were collected in the 0.1 T to 7 T and 1.8 K-10 K ranges. The resulting data for 6-2 to 6-4 are plotted as reduced magnetization ( M / N B) vs H / T in Figures 6-9 and 6-10. These data were fit using the program MAGNET,53 described elsewhere.56 The best-fit for 6-2 is shown as the solid lines in Figur e 6-9. A satisfactory fit could only be obtained if data collected at fields above 4 T were ex cluded, suggesting that some lo w-lying excited states with S > 7/2 are being stabilized by the applied field to the point that they are signifi cantly populated at these temperatures. The best fit was obtai ned using only the low-field data ( 4 T) (solid lines in Figure 6-9) gave S = 7/2, D = -0.85.01 cm-1 and g = 2.06.01. Alternative fits with S = 5/2 or 9/2 were rejected because they gave unreasonable values of g and D In order to ensure that the true global minimum had been located for each compound, we calculated the root-mean-square error surface for the f its as a function of D and g, and have plotted them as two-dimensional contour plots. For 6-2 the plot clearly shows only the a bove-mentioned minima with negative D value (Figure 6-9). For 6-3 the best fit (obtaine d using low-field data; 2T) is shown as so lid lines in Figure 6-10 (magnetization data was corrected for th e paramagnetic anion) and was obtained with S = 15/2 and either of the two sets of parameters, g = 2.12/ D = -0.13 cm-1 or g = 2.07/ D = 0.16 cm-1. The error surface contour plot is shown in Fi gure 6-11 and shows the above two minima, with the one with negative D clearly the superior fit since it has a lower (deeper) minimum. For 6-4 H2O, a good fit was obtained using 0.1-4 T data (solid lines in Figure 6-10) with S = 4 and either of the two sets of parameters, g = 2.06/ D = -0.29 cm-1 or g = 2.05/ D = 0.35 cm-1. A rms error analysis shows that the one with D < 0 is superior, suggesting this to be the true sign of D (Figure 6-11).

PAGE 131

131 6.3.3.2 Ac Studies Ac m agnetic susceptibility studies were pe rformed on vacuum-dried microcrystalline samples of 6-2 to 6-4 in the temperature range 1.8-10 K with a zero dc field and a 3.5 G ac field oscillating at frequencies in th e 5 1000 Hz range. The in-phase ( M ) component of the ac susceptibility for 6-2 plotted as M T vs T, is shown in Figure 6-12. The in-phase M T decreases with decreasing T before exhibiting a frequency dependent drop below ~3 K. Extrapolation from above 3 to 0 K gives a value of ~7.9 cm3Kmol-1, confirming an S = 7/2 ground state. The drop below ~3 K and the concomitant frequency dependent out-of-phase ( M ) signal suggest 6-2 to possibly be a SMM. The M T vs T plot for 6-3 and 6-4 H2O is shown in Figure 6-13. Extrapolation from above ~4 to 0 K gives a value of ~30 and ~10 cm3Kmol-1 for 6-3 and 64H2O, confirming an S = 15/2 and 4 ground state respectivel y. No out-of-phase signals were seen for both complexes. 6.3.3.3 Single-Crystal Hysteresis Studies To probe the possible S MM behavior further, single-crystal hystere sis loop and relaxation measurements were performed using a micro-SQUID153 setup. It was found that complex 6-2 indeed behaves as an SMM as shown by the presence of magnetic hysteresis loops whose coercivity is strongly temper ature and sweep rate dependent increasing with decreasing temperature and increasing sweep rate as expected for the superparamagnetic like behavior of a SMM, as shown in Figure 6-14. QTM steps were observed, the first appear ing before zero-field as expected for an exchange-bias effect from the neighbor within the [Fe9]2 dimer, and as seen for [Mn4]2.154 However, a QTM step at zero field was also seen, and this is not expected for an exchange biased dimer. However, this and the ~2:1 ratio of the steps at 0.11 and 0.0 T can be explained with reference to Figure 6-4 as due to an antiferromagnetic exchange interaction between the two Fe9 units mediated by the intermolecular H-bond. Thus, 64% of the [Fe9]2

PAGE 132

132 dimers show an exchange-bias (ON), whereas the remaining 36% do not (OFF). The spin Hamiltonian ( H) for each Fe9 SMM with Ising-like anisotropy is given by H = H 1 + H 2 -2 J 12 From the Hex = 0.11 T ( Hex = exchange-bias field) and the relationship J = gBHex/(2 S) (H = -2 J ij convention)155 can be calculated that the intermolecular interaction is J = -0.04 K, that is, very weakly antiferromagnetic. Thus, 64% of the dimers in the crystal are in an ON state and 36% are OFF (Figure 6-4). Since the intermolecular interaction J will serve to quantum mechanically entangle the two molecules and generate superposition states,154,156,157 the ON and OFF states with respect to the interaction thus correspond to potential ON and OFFstates of a coupled two-qubit system for quantum computation. Being able to have the interaction ON or OFF in some simple way is important, and the present work shows that a super exchangepathway via a single H-bond will suffice. Note that the [Fe9]2 head to-head dimer structure does not depend on the intermolecular H-bond, unlike [Mn4]2 where a total of six equivalent C-HCl intermolecular H-bonds clearly control the crystallization of the dimer. 6.4 Conclusions The use of dialcohol based ligand heenH2 in Fe(III) chemistry has led to fascinating new polynuclear clusters, Fe6, Fe7, Fe9 and Fe18. Complex Fe18 is a new structur al type with a serpentine topology. The diamagnetic ground state of Fe18 is not unexpected give n that this is the most common ground state for FeIII x clusters where x is even. Some important exceptions however include [Fe8O2(OH)12(tacn)6]8+ ( S = 10) and Fe6 presented in this paper, where the nonzero ground state is caused by sp in frustration effects induced by the architecture of these complexes. A new Fe9 single-molecule magnet has been synthesized and found to crystallize as head-to-head dimers. Two-thirds of th ese are exchange-coup led through a hydrogen-bond

PAGE 133

133 whereas the other third of the dimers are non-interacting, as mon itored by magnetization hysteresis measurements. The crystal thus cont ains a mixture of ON and OFF dimers with respect to the quantum mechanical coupling through the hydrogen-bond. Identification of these new clusters suggests th at introduction of alc ohol based ligands can lead to new Fex topologies/high spin molecules/SMMs not seen or accessible with oxide and carboxylate ligands alone using a simple bottom up approach. These results continue to emphasize the utility of small nuclearity prefor med clusters as stepping-stones to higher nuclearity products, in this case through the use of alcohol based ch elate. Overall we feel that this system is fascinating both in terms of the aes thetic appeal of the structural concepts and its potential to provide nanoscale magnets.

PAGE 134

134 Table 6-1. Crystallographic data for 6-1 C5H12CH2Cl2, 6-2MeCN, 6-3 MeOHH2O and 6-4 EtOHH2O 6-1 6-2 6-3 6-4 Formulaa C188H364Cl8Fe18N8 O74 C103H99Fe9N4O38 C24H67N6Fe7.25O17 Cl6.75 C63H97Cl2Fe6N8 O31.5 fw, g/mola 5209.77 2503.546 1356.04 1876.49 Space group P P I222 P 21/n a, 15.7923(12) 17.0412(15) 20.8859(14) 11.7319(8) b, 17.9984(13) 19.8622(16) 21.5851(15) 13.5034(9) c 24.8805(19) 19.9519(17) 22.6380(16) 13.8033(10) 106.938(2) 89.227(2) 90 113.332(1) 104.711(2) 77.146(2) 90 92.028(1) 90.867(2) 77.433(2) 90 99.130(1) V 3 6513.0(8) 6421.9(9) 10205.8(12) 1970.8(2) Z 1 2 8 1 T K 173(2) 173(2) 173(2) 173(2) Radiation,b 0.71073 0.71073 0.71073 0.71073 calc, g/cm3 1.328 1.474 1.765 1.581 mm-1 1.124 1.277 2.420 1.231 R1 c,d 0.0467 0.0741 0.0464 0.0622 wR2 e 0.1252 0.1803 0.1044 0.1551 a Including solvate molecules. b Graphite monochromator. c I > 2 ( I ). d R 1 = (|| Fo| | Fc||) / | Fo|. e wR 2 = [ [ w ( Fo 2 Fc 2)2] / [ w ( Fo 2)2]]1/2, w = 1/[ 2( Fo 2) + [( ap )2 + bp], where p = [max ( Fo 2, O) + 2Fc 2]/3. Figure 6-1. Structure of chelates: dmemH, heenH2

PAGE 135

135 Figure 6-2. Labeled representa tion of the structure of 6-1 with core Fe-O bonds shown as thicker black lines and pivalate Me groups omitted. Color code: Fe, green; O, red; N, blue; C, grey. Figure 6-3. Labeled representa tion of the structure of 6-2 with only the ipso C atoms of benzoate rings shown for clarity. The core is outlined in bold. Color code: Fe, green; O, red; N, blue; C, grey.

PAGE 136

136 Figure 6-4. The [Fe9]2 dimer showing intermolecular (t op, bottom) or only intramolecular (middle) O-HO hydrogen-bonding, and the resultant ON or OFF state with respect to the coupling of the two molecules. Color code: Fe, green; O, red; N, blue; C, grey; H, sky blue.

PAGE 137

137 Figure 6-5. Labeled representa tion of the structure of 6-3 with core Fe-O bonds shown as thicker black lines. Color code: Fe, green; Cl, cyan; O, red; N, blue; C, grey. Figure 6-6. Labeled represen tation of the cation of 6-4 with only the ipso C atoms of benzoate rings shown for clarity. Core Fe-O bonds are shown as thicker lines. Colour code: Fe, green; O, red; N, blue; C, grey.

PAGE 138

138 T/K 050100150200250300 MT/cm3mol-1K 0 10 20 30 40 50 6-1 6-2 6-3 6-4 Figure 6-7. Plot of MT vs T for complexes 6-1 to 6-4. Fe Fe O O O O Fe O O Fe Fe Fe Fe Fe O O Fe O O Fe Fe Fe O O O OF eO O Fe Fe Fe Fe Fe O O Figure 6-8. Spin alignments at the eighteen S = 5/2 Fe(III) atoms of 6-1 rationalizing its S = 0 ground state, based on the argum ents given in the text

PAGE 139

139 Figure 6-9. (left) Plot of reduced magnetization ( M/N B) vs H/T for 6-2 (right) Two-dimensional contour plot of the r.m.s. error vs D and g for the fit for 6-2 The asterisk marks the best-fit position (error minimum). Figure 6-10. Plot of reduced magnetization ( M/NB) vs H/T for 6-3 (left) and 6-4 (right).

PAGE 140

140 Figure 6-11. Two-dimensional contou r plot of the r.m.s. error vs D and g for the fit for 6-3 (left) and 6-4 (right). The asterisk marks the best-fit position (error minimum). Figure 6-12. Plot of the in-phase ( M T ) and out-of-phase ( M" ) ac susceptibility data for 6-2

PAGE 141

141 Figure 6-13. Plot of in-phase ac susceptibility data for 6-3 and 6-4 Figure 6-14. Single-crystal magnetization ( M ) vs dc field ( H ) hysteresis loops for 6-2 at different scan rates (left) and at various temperatures (right).

PAGE 142

142 CHAPTER 7 UNUSUAL STRUCTURAL TYPES IN Mn AND Fe CHEMISTRY FROM THE USE OF N,N,N ,N T ETRAKIS (2-HYDROXYETHYL)ETHYLENEDIAMINE 7.1 Introduction Interest in the preparation of polynuclear Mn and Fe com p lexes has developed worldwide for both fundamental scientific a nd technological reasons since the discovery that some of these molecules can behave as zero-dimensional nanoscale magnets now called single-molecule magnets (SMMs).13,22 Since then, many polynuclear clusters containing 3d transition metals have been reported to be SMMs,20,158-160 the vast majority of bring Mn complexes.21,161-164 In addition, polynuclear Mn and Fe compounds with O and N base d ligation are of interest because of their relevance in bioinorganic chemistry.26 For the above reasons and more, we continue to seek new synthetic methods to new Mnx and Fex complexes. In the design of a potentially ne w synthetic route to a polynuclear cluster, the choice of the ligands and bridging groups is vital. As part of our continuing search for such new methods, we have begun exploring the use of chelating/bridging groups based on the ethylenediamine backbone. In chapter 4 and 6, we reported the use of dmemH and heenH2 (Figure 7-1) as new and flexible N,N,O and O,N,N,O chelates, re spectively, for the synthesis of Fe3, Fe6, Fe7, Fe9 and Fe18 complexes, some of which possess novel Fex topologies.93,145,146 The hydroxyethyl arms, on deprotonation, usually act as bridging groups a nd thus foster formation of a high nuclearity product. In the present work, we have extended this study by exploring the use in Mn and Fe cluster chemistry of the re lated, potentially hexadentate ligand N,N,N',N'tetrakis(2-hydroxyethyl)ethylenediamine (edteH4; Figure 7-1). The edteH4 molecule now provides four hydroxyethyl arms on an ethylen ediamine backbone, and was considered an attractive potential new route to high nuc learity products. Previous use of edteH4 in the literature with other metals has been limited to the prep aration only of mononuclear Ca and dinuclear Ba,

PAGE 143

143 Cu and V complexes.165-167 We herein report that the use of edteH4 in a variety of reactions with Mn reagents has yielded novel Mn8, Mn12 and Mn20 complexes with core structures that are distinctly different from any seen previously.168 The use of edteH4 in Fe chemistry also leads to interesting new structural types of products, Fe5, Fe6 and Fe12 complexes. The syntheses, structures and magnetochemical properties of these complexes will be described.169 7.2 Experimental Section 7.2.1 Syntheses All preparations were perfor med under aerobic conditions using reagents and solvents as received. [Mn3O(O2CPh)6(py)2H2O],45 [Fe3O(O2CPh)6(H2O)3]NO3 and [Fe3O(O2CBut)6(H2O)3]OH was synthesized as reported elsewhere.94,95,170 [Mn8O3(OH)(OMe)(O2CPh)7(edte)(edteH2)](O2CPh) (7-1). Method A. To a stirred solution of edteH4 (0.05 g, 0.21 mmol) in CH2Cl2/MeOH (16/4 mL) was added [Mn3O(O2CPh)6(py)2H2O] (0.23 g, 0.21 mmol). The mixture wa s stirred for 30 minutes, filtered, and the filtrate layered with Et2O. X-ray quality dark orange-brown crystals of 7-1 CH2Cl2 MeOH slowly formed over a week. They we re collected by filtration, washed with Et2O and dried in vacuo The yield was 40%. Dried solid appeared to be hygroscopic, analyzing as the dihydrate. Anal. Calcd (Found) for 7-1 H2O (C77H90N4Mn8O31): C, 46.08 (45.97); H, 4.52 (4.36); N, 2.79 (2.81). Selected IR data (cm-1): 2868(w), 1649(w), 1595(m), 1547(m), 1447(w), 1383(s), 1315(m), 1174(w), 1122(w), 1066(m), 9 11(w), 719(m), 676(w), 603(m), 536(m). Method B. To a stirred solution of edteH4 (0.10 g, 0.42 mmol) in MeCN/MeOH (10/5 mL) was added NEt3 (0.18 mL, 1.28 mmol) followed by Mn(O2CPh)2 (0.42 g, 1.26 mmol). The resulting mixture was stirred for one hour, filtered, and the filtrate layered with Et2O. Dark orange-brown crystals of 7-1 slowly formed over 5 days and were then isolated as in Method A. The yield was 20%. The product was identified by IR spectral comparison with material from

PAGE 144

144 Method A and elemental analysis. Anal. Calcd (Found) for 7-1 H2O (C77H90N4Mn8O31): C, 46.08 (45.79); H, 4.52 (4.37); N, 2.79 (2.76). [Mn12O4(OH)2(edte)4Cl6(H2O)2] (7-2) To a stirred solution of edteH4 (0.15 g, 0.64 mmol) in MeCN/MeOH (10/1 mL) was added NEt3 (0.09 mL, 0.64 mmol) followed by MnCl2H2O (0.25 g, 1.26 mmol). The solution was stirred for 2 hours and then filtered, and the brown filtrate was left undisturbed to evaporate slowly, giving X-ray quality crystals of 7-2 MeCNH2O over five days. These were collected by filtration, washed with MeCN, and dried in vacuo The yield was 25%. Dried solid analyzed as solvent-free. Anal. Calcd (Found) for 7-2 (C40H86N8Mn12O24Cl6): C, 24.83 (24.57); H, 4.48 (4.50); N, 5.79 (6.20); Cl, 10.99 (11.89). Selected IR data (cm-1): 2854(m), 1631(w), 1465(w), 1359(w), 1270(w), 1160(w), 1088(s), 1059(s), 926(m), 899(m), 741(w), 669(m), 619(m), 557(m). [Mn20O8(OH)4(O2CMe)6(edte)6](ClO4)2 (7-3). To a stirred solution of edteH4 (0.10 g, 0.42 mmol) in MeOH (12 mL) was added NEt3 (0.12 mL, 0.85 mmol) followed by Mn(O2CMe)2H2O (0.21 g, 0.86 mmol) and then NaClO4 (0.05 g, 0.41 mmol). The mixture was stirred for one hour, filtered, a nd the filtrate layered with Et2O. X-ray quality dark orange-brown crystals of 7-3 MeOH slowly formed over a week. They were collected by filtration, washed with a little Et2O, and dried in vacuo The yield was 20%. Dried solid appeared to be hygroscopic, analyzing as the penta hydrate. Anal. Calcd (Found) for 7-3 H2O (C72H152N12Mn20O61Cl2): C, 25.96 (25.92); H, 4.60 (4.69); N, 5.04 (4.54). Selected IR data (cm-1): 2929(w), 1560(s), 1418(s), 1145(m), 1112(m), 1088(s) 910(m), 627(s), 563(m). [Fe5O2(O2CPh)7(edte)(H2O)] (7-4). To a stirred solution of edteH4 (0.05 g, 0.21 mmol) in CH2Cl2 (15 mL) was added [Fe3O(O2CPh)6(H2O)3](NO3)(0.38 g, 0.37 mmol). The mixture was stirred for 30 minutes, filtered to remove undissolv ed solid, and the filtrate layered with a 1:1

PAGE 145

145 (v/v) mixture of Et2O and hexanes. X-ray quali ty orange crystals of 7-4 CH2Cl2 slowly formed over a period of five days. These were collected by filtration, washed with Et2O, and dried in vacuo The yield was 20%. Anal. Calcd (Found) for 7-4 (C59H57N2Fe5O21): C, 50.28 (50.63); H, 4.07 (4.27); N, 1.99 (1.85). Se lected IR data (cm-1): 2862(w), 1597(m), 1552(s), 1534(s), 1400(s), 1175(w), 1087(m), 1067(m), 1024(w), 928(w), 892(w), 863(w), 720(s), 653(m), 602(w), 529(w), 465(m). [Fe6O2(O2CBut)8(edteH)2] (7-5). To a stirred solution of edteH4 (0.10 g, 0.42 mmol) in CHCl3 (15 mL) was added [Fe3O(O2CBut)6(H2O)3](OH) (0.18 g, 0.21 mmol). The mixture was stirred for 30 minutes, filtered to remove undi ssolved solid, and the filtrate layered with pentanes. X-ray quality orange crystals of 7-5 CHCl3 slowly formed over a week. These were collected by filtration, washed with pentanes, and dried in vacuo. The yield was 10 %. Dried solid appeared to be very hygroscopic, analyzing as the tetrahydrate. Anal. Calcd (Found) for 7-5 2CHCl3H2O (C62H124N4Fe6Cl6O30): C, 38.12 (37.98); H, 6.40 (6.33); N, 2.87 (3.24). Selected IR data (cm-1): 2960(m), 2869(m), 1562(s), 1483(s), 1421(s), 1375(m), 1360(m), 1227(m), 1098(m), 1042(w), 909(w), 788(w), 694(m), 603(m), 554(m), 480(w), 429(m). [Fe12O4(OH)2(O2CMe)6(edte)4(H2O)2](ClO4)4 (7-6). To a stirred solution of edteH4 (0.10 g, 0.42 mmol) in MeCN (15 mL) was added NaO2CMe3H2O (0.23 g, 1.69 mmol) followed by Fe(ClO4)3H2O (0.39 g, 0.85 mmol). The mixture was stirre d for 30 minutes, filtered to remove undissolved solid, and the filtrate le ft to slowly concentrate by evaporation. X-ray quality orange crystals of 7-6 MeCN slowly formed over a week. These were collected by filtration, washed with MeCN, and dried in vacuo The yield was 40%. Anal. Calcd (Found) for 7-6 (C52H104N8Fe12Cl4O52): C, 25.13 (24.82); H, 4.22 (4.21); N, 4 .51 (4.47). Selected IR data (cm-1):

PAGE 146

146 2884(m), 1559(s), 1455(s), 1336(w) 1271(w), 1086(s), 933(m), 904( m), 744(w), 624(s), 532(m), 466(w), 437(w). [Fe12O4(OH)8(edte)4(H2O)2](ClO4)4 (7-7). To a stirred solution of edteH4 (0.05 g, 0.21 mmol) in EtOH (15 mL) was added Fe(ClO4)3H2O (0.39 g, 0.85 mmol). The mixture was stirred for 30 minutes, filtered to remove undisso lved solid, and the filt rate left to slowly concentrate by evaporation. X-ray quality orange crystals of 7-7 slowly formed over a week. These were collected by filtrati on, washed with EtOH, and dried in vacuo The yield was 20%. Anal. Calcd (Found) for 7-7 (C40H92N8Fe12Cl4O46): C, 21.66 (21.51); H, 3.74 (4.15); N, 5.31 (5.02). Selected IR data (cm-1): 2867(m), 1628(w), 1469(w), 1363( w), 1270(w), 1088(s), 935(m), 910(m), 740(w), 661(m), 627(s), 583(w), 490(m). [Fe12O4(OH)8(edte)4(H2O)2](NO3)4 (7-8). To a stirred so lution of edteH4 (0.10 g, 0.42 mmol) in MeOH (15 mL) was added NEt3 (0.12 mL, 0.85 mmol) followed by Fe(NO3)3H2O (0.34 g, 0.85 mmol). The mixture was stirred for 30 minutes, and filtered to remove undissolved solid. Vapor diffusion of THF into the filtrate gave needle-like orange crystals of 7-8 These were collected by filtration, washed with THF, and dried in vacuo The yield was 10%. Anal. Calcd (Found) for 7-8 (C40H92N12Fe12O42): C, 23.29 (23.06); H, 4.61 (4.45); N, 8.12 (8.07). Selected IR data (cm-1): 2938(w), 2677 (m), 1650(w), 1385(s), 1171(w), 1057(m), 934(m), 909(m), 825(m), 636(m), 613(w), 525(w), 492(m). 7.2.2 X-ray Crystallography Data were collected by Dr. Khalil A Abboud on a Siemens SMART PLATFORM equipped with a CCD area detector and a graphite monochromator utilizing Mo-K radiation ( = 0.71073 ). Suitable crystals of 7-1 CH2Cl2MeOH, 7-2MeCNH2O, 7-310MeOH, 74CH2Cl2, 7-5 2CHCl3, 7-6 MeCN and 7-7 were attached to glass fibe rs using silicone grease and

PAGE 147

147 transferred to a goniostat where they were cool ed to 173 K for data collection. Cell parameters were refined using up to 8192 re flections. A full sphere of da ta (1850 frames) was collected using the -scan method (0.3 frame width). The first 50 frames were remeasured at the end of data collection to monitor instrument and crystal stability (maximum co rrection on I was < 1 %). Absorption corrections by integration were applie d based on measured indexed crystal faces. The structure was solved by direct methods in SHELXTL6,46 and refined on F2 using full-matrix least squares. The non-H atoms were treated anisotropi cally, whereas the hydrogen atoms were placed in ideal, calculated positions and were refined as riding on their respective C atoms. For 7-1 CH2Cl2MeOH, the asymmetric unit consists of the Mn8 cation, a benzoate anion, and one MeOH and two CH2Cl2 molecules. The solvent molecu les were disordered and could not be modeled properly, thus program SQUEEZE,68 a part of the PLATON package of crystallographic software, was used to calcula te the solvent disorder area and remove its contribution to the overall intensity data. A tota l of 1068 parameters were refined in the final cycle of refinement using 10920 reflections with I > 2 ( I ) to yield R1 and wR2 of 3.34 and 8.12%, respectively. For 7-2 MeCNH2O, the asymmetric unit contains of the Mn12 cluster, two MeCN molecules, one MeCN in a general position, a nd a 1/8 water molecule. All solvent molecules were disordered and could not be m odeled properly, thus program SQUEEZE68 was used to calculate the solvent disorder area and remove its contribution to the overall intensity data. A total of 204 parameters were refined in the fina l cycle of refinement using 2799 reflections with I > 2 ( I ) to yield R1 and wR2 of 4.64 and 10.46%, respectively. For 7-3 MeOH, the asymmetric un it consists of half the Mn20 cluster, one ClO4 anion, and five MeOH molecules. The la tter were disordered and could not be modeled properly, thus

PAGE 148

148 program SQUEEZE68 was again used to calculate the so lvent disorder area and remove its contribution to the overall intensity data. The cluste r exhibits two disorders: A large part of the N4 ligand is disordered and was refined in two pa rts; attempts to resolve the two parts of C17'C18' were not successful, and thus this part re mained common to both. C28 is also disordered and was also refined in two parts. The two part s of each disorder were dependently refined. Three O atoms of the ClO4 anion containing Cl2 were also di sordered and were refined in two parts related by rotation along the Cl2-O32 axis. A total of 687 parameters were refined in the final cycle of refinement using 6689 reflections with I > 2 ( I ) to yield R1 and wR2 of 8.12 and 21.74%, respectively. For 7-5CHCl3, the asymmetric unit consists of half an Fe6 cluster and a CHCl3 molecule. Two But groups are disordered and were refined in two parts each. A total of 481 parameters were refined in the final cycle of refinement using 6037 reflections with I > 2 ( I ) to yield R1 and wR2 of 5.75 and 13.64%, respectively. For 7-6 MeCN, the asymmetric unit consists of the Fe12 cluster, three wh ole and two half perchlorate anions, which are all disordered, and four MeCN molecu les, three of which are very disordered. Program SQUEEZE,68 a part of the PLATON packag e of crystallographic software, was used to calculate the solvent disorder area and remove its contribution to the overall intensity data. The N9 MeCN mol ecule was not removed by SQUEEZE68 because it is hydrogenbonded to the O17-H17 hydroxyl group and not diso rdered. The Cl1 perchlorate is hydrogenbonded to the opposite hydroxyl group (O17-H17) through O38. While each disordered perchlorate anion was refined in two pa rts, the second part of the Cl3 (Cl3 ) was not complete, only one O atom being located. The charges are balanced by the fact th at the groups occupying the O5 and O27 positions represent a disorder between a water molecule and a carboxyl group.

PAGE 149

149 The others could not be found due to the heavy disorder. Finally, the hydroxyl protons and the coordinated water protons were obtained from a difference Fourier map and included as riding on their parent O atoms. A total of 1165 parameters were refined in the final cycle of refinement using 10158 reflections with I > 2 ( I ) to yield R1 and wR2 of 7.88 and 22.19%, respectively. Severe disorder problems were encountered for 7-4CH2Cl2 and 7-7 For 7-4CH2Cl2, the asymmetric unit consists of an Fe5 cluster and a dichlorometh ane molecule; the structure exhibited much disorder in th e benzoate phenyl rings and edte4groups, preventing satisfactory refinement of the structure. However, the core was well observed and showed no disorder. For 7-7 the asymmetric unit consists of half a Fe12 cluster and two perchlorate anions; again, the structure exhibited bad disord er among the peripheral ligands. Despite examination of many crystals of both compounds, we could not find one s that diffracted well enough to allow data of sufficient quantity and quality to be obtained for satisfactory structure refinement. Thus, the structures were refined as far as possible so that we could at least identify the overall structure and nuclearity of the complexes for comparison with 7-5 and 7-6 which we were able to successfully do. Knowing the numbe r and arrangement of the Fe atoms in the core was also essential for the interpretation of the magnetic data of 7-4 and 7-7 We include and briefly describe the structures of these two complexes in this chapter only fo r the mentioned purposes; the metric parameters are unreliable and are no t discussed. Unit cell data and details of the structure refinements for complexes 7-1 to 7-3 are listed in Table 7-1 and 7-4 to 7-7 are listed in Table 7-2. 7.3 Results and Discussion 7.3.1 Syntheses In order to m ake clusters containing MnIII ions, it is generally necessary to either oxidize simple MnII salts, or use preformed higher oxidation state Mnx clusters. Both of these strategies

PAGE 150

150 have previously proved to be us eful routes to a variety of hi gher nuclearity complexes with chelating ligands ranging from bidentate to pentadentate.34,35,43,63,171,172 Therefore, we decided to employ them both with the potentially hexadentate ligand edteH4. Thus, a variety of reaction ratios, reagents, and other conditions we re investigated. The reaction of edteH4 with Mn(O2CPh)2 and NEt3 in a 1:3:3 molar ratio in MeOH affo rded a reddish-brown solution from which was subsequently obtai ned the octanuclear complex [Mn8O3(OH)(OMe)(O2CPh)7(edte)(edteH2)](O2CPh) ( 7-1 ) in ~20% yield. Its formation is summarized in eq. 7-1, where atmospheric oxygen gas is assumed to provide the oxidizing equivalents required to form the mixed-valence 6MnIII, 2MnII product. 8[Mn(O2CPh)2] + 2edteH4 + MeOH + H2O + 3/2 O2 8PhCO2H + [Mn8O3(OH)(OMe)(O2CPh)7(edte)(edteH2)](O2CPh) (7-1) Complex 7-1 was also obtained, and in a higher yiel d of ~40%, from the reaction of edteH4 with [Mn3O(O2CPh)6(py)2H2O] in a 1:1 molar ratio in CH2Cl2/MeOH. Such trinuclear [Mn3O] clusters have often proved to be very useful starting materials for the synthesis of higher nuclearity products, some of which have also been new SMMs.38,173 The formation of 7-1 via this route is summarized in eq. 7-2. The mixed so lvent system was needed to ensure adequate solubility of all reagents, and it also led to methoxide incorpor ation; no isolable products were obained when only CH2Cl2 was used. Small variations in the Mn:edteH4:PhCO2 ratio also gave complex 7-1 which clearly is a preferre d product of these components, and with benzoate. When [Mn3O(O2CR)6(py)3]0,+ clusters with other R groups were employed, we were unable to isolate pure, crystalline materials for satisfactory characterization. 3[Mn3O(O2CPh)6(py)2(H2O)] + 2edteH4 + MeOH Mn2+ + 6py + 8H+ + 2H2O + 10PhCO2 + [Mn8O3(OH)(OMe)(O2CPh)7(edte)(edteH2)](O2CPh) (7-2)

PAGE 151

151 The reaction of edteH4 with MnCl2H2O, and NEt3 in 1:2:1 molar ratio in MeCN/MeOH gave a brown solution from which was isolated [Mn12O4(OH)2(edte)4Cl6(H2O)2] ( 7-2 ) in 25% yield. As found for 7-1 complex 7-2 is mixed-valent, containing 8MnIII and 4MnII ions, and its formation is summarized in eq. 7-3, again a ssuming the participation of atmospheric O2. Increasing or decreasing the amount of edteH4 or NEt3 also gave complex 7-2 but the product was not as pure. We were also unable to isol ate any clean products when we employed an MeCN/EtOH solvent system. 12MnCl2 + 4edteH4 + 4H2O + 2O2 [Mn12O4(OH)2(edte)4Cl6(H2O)2] + 18HCl (7-3) Finally, the reaction of edteH4 with Mn(O2CMe)2 and NEt3 in a 1:2:2 ratio in MeOH, followed by addition of NaClO4, gave a dark orange-brown solution from which was subsequently isolated [Mn20O8(OH)4(O2CMe)6(edte)6](ClO4)2 ( 7-3 ) in 20% yield. This product is once again mixed-valence, containing 12MnIII, 8MnII ions, and its formation is summarized in eq. 7-4, with O2 again included as the oxidizing agent. 20Mn(O2CMe)2 + 6edteH4 + 6H2O + 3O2 [Mn20O8(OH)4(O2CMe)6(edte)6]2+ + 32MeCO2H + 2MeCO2 (7-4) A variety of reactions of edteH4 were explored with a number of different Fe(III) starting materials and under different reag ent ratios, solvents and other conditions before the following successful procedures were id entified. The reaction of [Fe3O(O2CPh)6(H2O)3](NO3) with edteH4 in a ~3:2 molar ratio in CH2Cl2 followed by layering with Et2O hexanes (1:1 v/v) gave orange needle-like crystals of [Fe5O2(O2CPh)7(edte)(H2O)] ( 7-4 ). Its formation is summarized in eq. 7-5. The benzoate groups clearly function as proton acceptors facilitating the deprotonation of edteH4 in the absence of added base.

PAGE 152

152 5[Fe3O(O2CPh)6(H2O)3]+ + 3edteH4 3[Fe5O2(O2CPh)7(edte)(H2O)] + 9PhCO2H + 11H2O + 5H+ (7-5) With other chelates such as dmemH,93 we have found that the identity of the Fex product depends on the carboxylate employed,93 and thus we also explored reactions of edteH4 with other [Fe3O(O2CR)6(H2O)3]+ reagents. With pivalate (R = But), a related reaction to that which gave 7-4 but with a [Fe3O(O2CBut)6(H2O)3]+ to edteH4 molar ratio of 1:2 in CHCl3, gave a brown solution and subsequent isolation of [Fe6O2(O2CBut)8(edteH)2] ( 7-5 ) on layering with pentanes. The proton acceptors in this reaction are the carboxylate groups and the OHanions; the formation of 7-5 is summarized in eq. 7-6. 2[Fe3O(O2CBut)6(H2O)3](OH) + 2edteH4 [Fe6O2(O2CBut)8(edteH)2] + 4ButCO2H + 8H2O (7-6) The same product was also obtained using CH2Cl2 as the solvent, but in poor crystallinity and decreased yield. We also explored the use of simple Fe(III) salts as reagents, in the presence of added carboxylate groups as proton acceptors. The reaction of Fe(ClO4)3H2O with edteH4 and NaO2CMe3H2O in a 2:1:4 ratio in MeCN gave a brown solution from which was obtained [Fe12O4(OH)2(edte)4(O2CMe)6(H2O)2](ClO4)4 ( 7-6 ). Its preparation is summarized in eq. 7-7. 12Fe3+ + 32MeCO2 + 4edteH4 + 8H2O [Fe12O4(OH)2(edte)4(O2CMe)6(H2O)2]4+ + 26MeCO2H (7-7) Decreasing the the amount of acetate from 4 to 2 equiv drastically reduced the reaction yield, as expected from eq. 7-7. Complex 7-6 was also obtained from a MeCN:MeOH solvent system. However, when the reaction of Fe(ClO4)3H2O with edteH4 in a 4:1 ratio was carried out in neat EtOH in the absence of NaO2CMe, the product was [Fe12O4(OH)8(edte)4(H2O)2](ClO4)4 ( 7-7), obtained as orange needles on layering the solution with CHCl3. Complex 7-7 is

PAGE 153

153 structurally very similar to 7-6 except that the acetate groups have been replaced by hydroxide ions (vide infra). In a related fashion, the reaction of Fe(NO3)3 with edteH4 and NEt3 in a 2:1:2 ratio in MeOH gave [Fe12O4(OH)8(edte)4(H2O)2](NO3)4 ( 7-8 ) on vapor diffusion with tetrahydrofuran. The product was identified by elemental analysis, and IR and magnetic comparisons with complexes 7-6 and 7-7 (vide infra). It is clear that the described reactions to complexes 7-1 to 7 8 are very complicated and involve acid/base and redox chemistry, as well as structural fragmentations and rearrangements. As a result, the reaction soluti ons likely contain a complicated mixture of several species in equilibrium. For this reason, we were happy to settle for the rela tively low yields of 7-1 to 78, given that the products were reproducibly obtained in a pure, crystalline form from the described procedures. 7.3.2 Description of Structures 7.3.2.1 Structure of [Mn8O3(OH)(OMe)(O2CPh)7(edte)(edteH2)](O2CPh) (7-1) A labeled representati on of the cation of 7-1 is shown in Figure 7-2, and selected interatomic distances and angles ar e summarized in Table A-16. Complex 7-1 crystallizes in the monoclinic space group P 21/n. The core is mixed valence (6MnIII, 2MnII), with Mn3 and Mn6 being the MnII ions, and contains a [Mn7O7] subunit consisting of two distorted [MnIII 3MnII( 5O)( 3-O)(3-OR)2]5+ cubanes sharing the Mn6-O1 edge. This double-cubane unit is additionally bridged by a -OMe(O21) between Mn7/Mn8 and a 3-OH(O6) between Mn2/Mn4. The latter 3-OHadditionally connects the double-cubane to the eighth Mn atom, Mn3. The edteH2 2group is hexadentate-chelating on MnII atom Mn3, with the two deprotonated alkoxide O atoms, O7 and O10, bridging to Mn2 and Mn4, respectively, and thus the edteH2 2group is overall 3bridging; oxygen atoms O8 and O9 are protonated. The edte4group is also hexadentate chelating to a MnII atom, Mn6, with the four deprotonated alkoxide O atoms all adopting 3 bridging

PAGE 154

154 modes within the double-cubane, and thus the edte4group is overall 7-bridging. The chelating/bridging modes of the 7-edte4and 3-edteH2 2groups are shown in Figures 7-3(a) and (b), respectively. The remaining ligation is provided by seven benzoate groups, five of which are 1: 1: -bridging and the remaining two are 1 terminal on Mn1 and Mn5. The oxidation states of the Mn at oms and the protonation levels of O2-, OH-, OMeand ORgroups were determined from a combination of charge balance considerations, inspection of bond lengths, and bond valence sum (BVS) calculations.48,49,103 BVS values for Mn and O atoms are listed in Table 7-3 an d 7-4, respectively. All the MnIII atoms are six-coordinate and display a Jahn-Teller (JT) elongation, as expected for high spin MnIII in octahedral geometry, with the JT axes (shown as thicker black bonds in Figure 7-2) not co-parallel. The MnII atoms, Mn3 and Mn6, are seven coordinate. The anion of complex 1 is a PhCO2 group (not shown in Figure 7-2), which forms an intimate ion-pair with the Mn8 cation by hydrogen-bonding with O6 of the 3OHion (O6O29 = 2.580 ) and O8 of an edteH2 2protonated alcohol arm (O8O28 = 2.584 ). This also has the effect of raising the BVS of 3-OHion O6 to 1.43, higher than normally expected for a OHgroup (1.0-1.2). The BVS for O8 (1.21) is much less affected by the hydrogen-bonding (compare with 1.17 for O9), no doubt due to the only monodentate binding of O8, which thus retains greater basicity and a stronger O-H bond than the 3-OHion. A number of other Mn8 complexes have previously been reported. These possess a variety of metal topologies such as rod-like, serpentine rectangular, linked Mn4 butterfly units, linked tetrahedral, etc,67,174-188 but none have possessed the core of complex 7-1 which is unprecedented. In addition to this novel core structure, another unusual feature of 7-1 is the presence of its 5-O2ion, O1. There are only two prev ious structural types with a 5-O2ion in molecular Mn chemistry, certain Mn12 189,190 and Mn13 191-193 complexes.

PAGE 155

155 7.3.2.2 Structure of [Mn12O4(OH)2(edte)4Cl6(H2O)2] (7-2) The labeled structure of 7-2 is shown in Figure 7-4, and sele cted interatomic distances and angles are listed in Table A-17. Complex 7-2 crystallizes in the tetragonal space group P 4/ncc with the Mn12 molecule lying on an S4 symmetry axis and thus onl y of it in the asymmetric unit. The structure consists of a [MnIII 8MnII 4( 4-O)4( -OH)2( -Cl)4( 3-OR)4( -OR)12]2+ core consisting of two near-planar Mn6 layers sandwiched between th ree near-planar layers of O atoms (Figure 7-4, bottom). For the sake of brevit y, reference to specific atoms in the following discussion includes their symmetry-related partne rs. BVS calculations for the Mn atoms (Table 7-5) identified Mn1, Mn2 and Mn3 as MnII, MnIII and MnIII atoms, respectively. Mn1 and Mn2 are six coordinate while Mn3 is seven coordinate. The four 4-O2ions (O1) together serve to connect all twelve Mn atoms. Each edte4group is hexadentate-chelating on a MnIII atom, Mn3, with each of its deprotonated alkoxide arms br idging to either one (O3, O5, O6) or two (O2) additional Mn atoms. Thus, the edte4groups are overall 5-bridging, as shown in Figure 7-3(c). Charge balance considerations require that, with eight MnIII, four MnII, four O2and four edte4groups, there must be eight additional negatively charged ligands to give neutral complex 7-2 The simplest conclusion is that with eight apparent Clions in the complex, the two -O atoms O4 and O4 bridging Mn2 atoms belong to H2O groups. However, we were unhappy with this conclusion, being unaware of a ny precedent in the literature for H2O groups bridging two MnIII atoms; the high Lewis acidity of two MnIII would be expected to make the water molecule in a [MnIII 2( -OH2)] unit very (Bronsted) acidic (very low pKa) and unlikely to be stable. In contrast, a water molecule bridging two MnII atoms is known.194-197 We thus decided to determine the protonation levels of all O atoms in 7-2 by BVS calculations, and the results are listed in Table 7-5. The oxide and edte4O atoms have BVS values of >1.89, confirming them as completely deprotonated, as concluded above from their bridging modes. In contrast, O4 has a

PAGE 156

156 BVS of only 0.99, as expected for an OHgroup. In addition, the Mn2-O4 bond length of 2.0098(15) is typical of MnIII-OHbond lengths in the literature.198-200 This is a more realistic bridging group between two MnIII atoms. Thus, we conclude that O4 and O4 are OHgroups. This now requires six additional anionic ligands fo r a neutral molecule, and we suspected that the S4 symmetry was masking a disorder between the Cland a neutral ligand such as H2O at the terminal positions (Cl1). Crystallographic re finement of these terminal Cl atoms was inconclusive as to whether there was a Cl/H2O disorder, and so we inves tigated the Cl content of the molecule more directly with a chlorine elem ental analysis. This did indeed give a value less than expected for eight Cl atoms, whose formula of [Mn12O4(H2O)2(edte)4Cl8] would have a calculated 14.40% Cl content, much higher than the experime ntal 11.89%. The latter is, however, consistent with the expected six Clions that are required for the observed neutrality of complex 7-2 if O4 is a OHion. Thus, we conclude that the correct formula of 7-2 is [Mn12O4(OH)2(edte)4Cl6(H2O)2]. Note that for the reasons al ready mentioned, we disfavor the H2O groups being disordered with Cl at the -Clpositions (Cl2) bridging MnIIIMnII pairs, but this is not as unlikely as water bridging two MnIII atoms, and thus cannot be completely ruled out. Indeed, maybe the two water molecules ar e disordered amongst the eight bridging and terminal positions, and this is why the crystallographic refinement is fine with eight Cl atoms. Recently, a complex apparently identical to complex 7-2 was reported by Zhou et al., but who instead formulated it as [Mn12O4(H2O)2(edte)4Cl8].201 There are many other structural types of Mn12 complexes already in the literature, the most well-studied being the [Mn12O12(O2CR)16(H2O)4] (Mn12) family, which has been extended over the years to include one, two and three-electron reduced [Mn12]z(z = 0 3) versions.12,202-207 Another Mn12 family of complexes was more recently obtained by reduc tive aggregation of

PAGE 157

157 MnO4 in MeOH-containing media; this family differs from the previous one in having a central MnIV 4 rhombus rather than a MnIV 4 tetrahedron.208,209 The remaining Mn12 complexes cover a variety of other structural t ypes, including loops and more co mplicated face-sharing cuboidal units, amongst others.189,210-215 7.3.2.3 Structure of [Mn20O8(OH)4(O2CMe)6(edte)6](ClO4)2 (7-3) The structure of the cation of 7-3 is shown in Figures 7-5 and 7-6, the latter providing the atom labeling. Selected interatomic distances and angles are listed in Table A-18. Complex 7-3 crystallizes in the monoclinic space group P 21/c with the Mn20 cation lying on a crystallographic inversion center; again, reference to a specific atom will include its symmetry-related partner. The cation can be described as consisting of two sets of three edge-sharing [Mn4O4] cubanes (Figure 7-5, middle), with the upper and lower se ts connected by face-sharing to give a 3 x 2 arrangement of six cubanes. This central [Mn14O16] unit is then attached to three additional Mn ions at each end by additional O atoms (Figure 7-5, bottom). This gives an overall tube-like arrangement of twenty Mn atom s inside of which are four O2ions. Note that the Mn7 edgesharing double-cubane structure of complex 7-1 is a recognizable subfragment of the central [Mn14O16] unit of 7-3 and thus 7-3 can be considered a more extended version of 7-1 The overall core is thus [MnIII 12MnII 8( 6-O)2( 4-O)2( 3-O)4( -OH)4( 3-OR)10( -OR)14]2+ with the two 6-O (O24), two 4-O (O19) and four 3-O (O20, O22) atoms being O2ions. The four -OHgroups, O21 and O23, (The BVS values of O21 and O23 are 1.19 and 1.20 respectively) bridge Mn8/Mn9' and Mn5/Mn9, respectively, and thus provide additional linkages between the cubanes. The ten 3-OR and fourteen -OR oxygen atoms are provided by the alkoxide arms of six edte4groups. As seen in 7-1 and 7-2 each edte4group binds as a hexadentate chelate to one Mn and then bridges through its deprotonated al koxide arms to a number of additional Mn atoms; four edte4groups are overall 5-bridging, and the remaining two are 7-bridging, and

PAGE 158

158 these modes are shown in Figures 7-3(c) and (d ), respectively. The remaining ligation in the molecule is provided by six acetate groups, two of which are 1: 1: -bridging, two are 2 chelating on Mn1, and two are 1 terminal on Mn7. Inspection of metric parameters and BVS calcula tions (Table 7-3) indicate that there are twelve MnIII and eight MnII atoms in the molecule. The BVS values for Mn5 and Mn8 are a little higher than normally expected for MnII, and those for Mn4 and Mn9 are a little lower than normally expected for MnIII, so it is possible there is some MnII/MnIII static disorder within the core. All the Mn ions are six-coordinate ex cept Mn5, which is seven coordinate. The JT elongation axes on six-coordinate MnIII atoms are shown as thicker black bonds in Figure 7-6. There is only one other Mn20 cluster in the literature, a co mplex with benzylphosphonate ligands reported by Winpenny and coworkers,173 which contains twelve Mn atoms in one plane. Complex 7-3 is thus structurally very different from this previous example. In addition to the novel overall structure, there is again, as for 7-1 another unusual feature, namely the presence of 6 O2ions (O24). There are only two previous examples of such a 6 O2in Mn chemistry, [Mn10O2Cl8(thme)6]2and [Mn18O14(OMe)14(O2CBut)8(MeOH)6].216-218 7.3.2.4 Structure of [Fe5O2(O2CPh)7(edte)(H2O)] (7-4) A labeled representation of 7-4 is shown in Figure 7-7. Complex 7-4 crystallizes in monoclinic space group P 21/c. The core can be described as consisting of a [Fe4( 3-O)2]8+ butterfly like subunit (Fe2, Fe3, Fe4 and Fe5) on the top of which is attached a [Fe( -OR)4] unit containing Fe1. There is an O atom monoatomically bridging Fe1 to each of the four Fe atoms of the butterfly. These four O atoms (O4, O6, O 10 and O11) are the alkoxide arms of edte4group. The edte4group is hexadentate with the four de protonated alkoxide O atoms all adopting bridging modes and thus edte4group is overall 5-bridging as shown in Figure 7-8(a). Peripheral ligation about the core is provided by one water molecule on Fe5 and seven benzoates, out of

PAGE 159

159 which five are in 1: 1: bridging mode and one is 1 terminal on Fe5 and one is 2 chelating on Fe2. There are relatively few Fe5 clusters reported in literature, and these have Fe5 topologies such as square pyramid, a centered tetrahedron, and a partial cubane ex tended at one face by a partial admantane unit.132-141 However, the only previous compound structurally similar to 7-4 is [Fe5O2(OH)(O2CMe)5(hmbp)3]2+ ( 5-1 ), where hmbpH = 6-hydroxy methyl-2,2 bipyridine.127 In 5-1 also, there is a [Fe4( 3-O)2]8+ core with an additional iron atom on top (as in 7-4 ) but the precise means by which latter is connected to the Fe4 unit is different from the situation in 7-4 Specifically, one of the bridging alkoxide in 7-4 is replaced by a hydroxide. 7.3.2.5 Structure of [Fe6O2(O2CBut)8(edteH)2] (7-5) A labeled representation of 7-5 is shown in Figure 7-9. Select ed interatomic distances and angles are summarized in Table A-19. Complex 7-5 crystallized in m onoclinic space group C 2/c. The structure comprises roughly planar arrangeme nt of six Fe atoms and the core can be described as consisting of two triangular [Fe3( 3-O)]7+ units joined together via six alkoxide edte O atoms. Specifically Fe2 one tria ngular unit is bridged to Fe3 of next triangular unit by O4. In addition, Fe3 and Fe3 are bridged via O13 and O13 Each triangular unit is essentially isosceles, (Fe1Fe2 = 2.986 Fe2Fe3 = 3.3 13 and Fe1Fe3 = 3.344 ) and essentially planar (the oxide, O12, is only 0.359 from the Fe3 plane). The two Fe3 triangular units are trans to each other. The peripheral liga tion is provided by 8 pivalates out of which 6 are 1: 1: bridging and two are 1 terminal on Fe1 and Fe1 All the Fe atoms are six-c oordinate. Additionally, Fe1 and Fe2 of each triangular unit are bridged by alkoxide arm of edteH3(O3), while fourth arm (O5) is protonated. The BVS for the O atoms of edteH3is provided in Table 7-6. The protonated oxygen atom, O5, is involved in intermolecular H-bonding to O1 (pivalate) of next molecule forming one dimensional chains that run in two directions in lattice. The edteH3group is overall 4bridging as shown in Figure 78(b). The core of complex 7-5 is unprecedented in hexanuclear

PAGE 160

160 Fe(III) chemistry. A number of Fe6 clusters have been reported in the literature and a recent listing of these, together with their st ructural types, is available elsewhere.104 Among these are a family of Fe6 clusters whose core s comprise linked [Fe3( 3-O)]7+ triangular subunits as in 7-5 but the two units are bridged by multiple hydrox o or alkoxo groups and ove rall all these prior complexes possess core structures different from that of the present complex 7-5 7.3.2.6 Structure of [Fe12O4(OH)2(O2CMe)6(edte)4(H2O)2](ClO4)4 (7-6) The labeled structure of cation of 7-6 is shown in Figure 7-10, and selected interatomic distances and angles are lis ted in Table A-20. Complex 7-6 crystallizes in the monoclinic space group C 2/c. The structure consists of a [FeIII 12( 4-O)4( -OH)2( -O2CMe)4( 3-OR)4( -OR)12]2+ core consisting of two near-planar Fe6 layers sandwiched between th ree near-planar layers of O atoms (Figure 7-10, bottom). All the iron atoms ar e six coordinate except Fe1, Fe3, Fe9 and Fe12 which are seven coordinate. The four 4-O2ions (O7, O13, O29 and O37) together serve to connect all twelve Fe atoms. Each edte4group is hexadentat e-chelating on a FeIII atom, Fe1, Fe3, Fe9 and Fe12, with each of its deprotonated alkoxide arms bridging to either one or two additional Fe atoms. Thus, the edte4groups are overall 5-bridging, as shown in Figure 7-8(c). The protonation levels of O2-, OH-, and ORgroups were determined from a combination of charge balance considerations, inspection of bond lengths, and BV S calculations (Table 7-6). The edte4O atoms have BVS values of >1.87, confirming them as completely deprotonated, as concluded above from their bridging modes. In contrast, O17 and O18 have a BVS of 1.24 and 1.20 as expected for an OHgroup. Peripheral ligation is provided by two terminal water molecules and six acetate groups, out of which 4 are 1: 1: bridging and two are 1 terminal on Fe8 and Fe10. Complex 7-6 is only one of a very few dodecanuclear Fe(III) clusters known in the literature, of which the majority have a wheel or loop structure.219-221 Among the remainder, one

PAGE 161

161 is composed of face-sharing defect cuboidal unit s in the central fragment of the core, and the other consists of four edge sharing [Fe3( 3-O)]7+ units.91,222 The structure of complex 7-6 is thus unprecedented in Fe chemistry, but is similar to Mn cluster 7-2 with the formula [Mn12O4(OH)2(edte)4Cl6(H2O)2] and a mixed-valence MnIII 8MnII 4 description.168 The labeled structure of the cation of [Fe12O4(OH)8(edte)4(H2O)2](ClO4)4 ( 7-7 ) is shown in Figure 7-11. The core is essent ially the same as that of 7-6 except that acetat e groups have been replaced by hydroxide ones. Complex 7-8 gave an elemental analysis consistent with it being the NO3 salt of the same cation as 7-7 and is thus formulated as [Fe12O4(OH)8(edte)4(H2O)2](NO3)4. This conclusion is also supported by th e very similar magne tic properties of 7-7 and 7-8 (vide infra), and indeed the very similar magne tic properties of all three complexes 7-6 to 7-8 which is consistent with the conclusion that they all possess the same or very similar Fe12 core structure. 7.3.3 Magnetochemistry 7.3.3.1 Dc Studies on 7-1 to 7-3 Solid-s tate, variable-temperature dc magnetic susceptibility data in a 0.1 T field and in the 5.0-300 K range were collected on powder ed microcrystalline samples of 7-12H2O, 7-2 and 73H2O restrained in eicosane. The ob tained data are plotted as MT vs T in Figure 7-12. The MT at 300 K is 26.8, 37.5 and 50.4 cm3Kmol-1 for 7-1 to 7-3 respectively. The 300 K value is equal to or less than the spin-only ( g = 2) value of 26.75, 41.5, and 71.0 cm3Kmol-1 expected for noninteracting MnIII 6MnII 2, MnIII 8MnII 4, and MnIII 12MnII 8 mixed-valence situations of 7-1 to 7-3 respectively. For 7-12H2O, MT stays essentially constant with decreasing temperature until 25 K and then increases to 32.1 cm3Kmol-1 at 8.0 K before drop ping slightly to 31.4 cm3Kmol-1 at 5.0 K. The MT at the lowest temperatures suggests S = 8 ground state spin with g < 2, as expected for Mn. For 7-2 MT again stays essentially constant with decreasing temperature until 70 K and then decreases smoothly to 26.2 cm3Kmol-1 at 5.0 K, which is suggestive of an S = 7

PAGE 162

162 ground state. For 7-3 H2O, MT decreases smoothly with decreas ing temperature to a minimum of 28.2 cm3mol-1K at 35 K and then increases to 36.0 cm3Kmol-1 at 5.0 K, which again suggests an S = 8 ground state. To confirm the above initial estimates of the ground state spin of the three compounds, variable-field ( H ) and -temperature magnetization ( M ) data were collected in the 0.1-7 T and 1.810 K ranges. The resulting data for 7-12H2O are plotted in Figure 7-13 (left) as reduced magnetization (M/N B) vs. H / T where N is Avogadro's number and B is the Bohr magneton. The data were fit, using the program MAGNET,53 described elsewhere.56 The best-fit for 71H2O is shown as the solid lines in Figure 7-13 and was obtained with S = 8, g = 2.00 and D = -0.30 cm-1. Alternative fits with S = 7 and S = 9 gave unreasonable values of g of 2.28 and 1.78, respectively. In order to ensure that the true global minimum ha d been located and to assess the hardness of the fit, a root-mean-square D vs g error surface for the fit was generated using the program GRID ,71 which calculates the relative difference between the experimental M/N B data and those calculated for various combinations of D and g. This is shown as a 2-D contour plot in Figure 7-14 (left) covering the D = -0.10 to -0.50 cm-1 and g =1.90 to 2.10 ranges. Only one minimum was observed, and this was a relative ly soft minimum; we thus estimate the fit uncertainties as D = -0.30 0.01 cm-1 and g = 2.00 0.02. For 7-2 we could not obtain a sati sfactory fit if data collected at all field values were employed. In our experience, the usual reason for this is the presence of lo w-lying excited states because (i) the excited states are close e nough to the ground state and they have a non-zero Boltzmann population even at the low temperatures used in the magnetization data collection, and/or (ii) even excited states that are mo re separated from the ground state but have an S value greater than that of the ground-state become populated as their larger MS levels rapidly decrease

PAGE 163

163 in energy in the applied dc magne tic field and approach (or even cross) those of the ground state. Either (or both) of these two e ffects will lead to poor fits because the fitting program assumes population of only the ground state. A large density of low-lying excited states is expected for higher nuclearity complexes and/or t hose with a significant content of MnII atoms, which give weak exchange couplings. Thus, it is reasonable that such problems are more likely for 7-2 than for 7-1 given both the higher nuclear ity and the higher relative MnII content of 7-2 vs 7-1 As described elsewhere,69,70,223,224 one way around effect (ii) is to use only data collected at low fields. Indeed, a satisfactory fit (s olid lines in Figure 7-13, right) was now obtained using data in fields up to 0.8 T, with fit parameters S = 7, D = -0.16 cm-1 and g = 1.90. Alternative fits with S = 6 and S = 8 gave g = 2.20 and 1.67, respectively. The corresponding error surface vs D and g (Figure 7-14, right) gives a ha rder minimum than that for 7-1 with estimated fit uncertainties of D = -0.16 0.01 cm-1 and g = 1.90 0.01. For 7-3 H2O, the even higher metal nuclearity and MnII content again necessitated using data collected at lower fields in the fit, and in th is case a satisfactory fit (so lid lines in Figure 7-15) was obtained for data up to 1 T with fit parameters S = 8, g = 1.90 and D = -0.16 cm-1. Alternative fits with S = 7 and S = 9 gave g = 2.16 and 1.70, respect ively. The corresponding error surface vs D and g (Figure 7-15) is similar to that for 7-2 and gives estimated fit uncertainties of D = -0.16 0.01 cm-1 and g = 1.90 0.01. The magnetization fits confirmed the preliminary estimates of the ground state spin S of 7-1 to 7-3 7.3.3.2 Dc Studies on 7-4 to 7-8 Solid-state, variable-temperature dc m agnetic susceptibility da ta were collected in a 0.1 T field and in the 5.0-300 K range on powdered crystalline samples of 7-4 to 7-8 restrained in eicosane. The obtained data are plotted as MT vs T in Figure 7-16. For 7-4 MT steadily decreases from 6.73 cm3Kmol-1 at 300 K to 3.88 cm3Kmol-1 at 40.0 K, then stays approximately

PAGE 164

164 constant until 25.0 K, and incr eases slightly to 4.02 cm3Kmol-1 at 5.0 K. The 300 K value is much less than the spin-only (g = 2) value of 21.87 cm3Kmol-1 for five non-interacting Fe(III) atoms, indicating the presence of strong antifer romagnetic interactions, as expected for oxobridged Fe(III) systems. The 5.0 K value of 4.02 cm3Kmol-1 suggests an S = 5/2 ground state spin. MT for 7-5 2CHCl3H2O is 11.03 cm3Kmol-1 at 300 K, and stays ap proximately constant with decreasing temperature to 100 K and then increases to 13.83 cm3Kmol-1 at 5 K. MT at 300 K is again much less than the spin-only value of 26.25 cm3Kmol-1 expected for six noninteracting Fe(III) ions indicating strong antifer romagnetic interactions. The 5.0 K value of 13.83 cm3Kmol-1 suggests an S = 5 ground state spin. The MT vs T plots for the three complexes 7-6 to 7-8 in Figure 7-16 are very similar, indicating a minimal influence of the periphe ral groups and supporting the conclusions above that they possess simila r core structures. For 7-6 to 7-8 MT steadily decreases from 22.04, 23.37, 20.53 cm3Kmol-1 at 300 K to 0.25, 0.51, 0.50 cm3Kmol-1 at 5.0 K, respectively. The change in MT with decreasing temperature and the lo w value at 5 K are indicative of an S = 0 ground state. The differences in MT vs T for the three complexes are almost certainly just reflecting small differences in intram olecular exchange coupling constants ( J ), and perhaps in ZFS parameters ( D ) and any intermolecular interactions. To confirm the initial ground st ate spin estimates above for 7-4 and 7-5 variable-field ( H ) and -temperature magnetization ( M ) data were collected in the 0.1-7.0 T and 1.8-10 K ranges. The resulting data for 7-4 are plotted in Figure 7-17 as reduced magnetization ( M / NB) vs H / T where N is Avogadros number and B is the Bohr magneton. The sa turation value at the highest fields and lowest temperatures is ~4.90, as expected for an S = 5/2 ground state and g slightly less than 2; the saturation value should be gS in the absence of complications from low-lying

PAGE 165

165 excited states. The data were fit, using the program MAGNET ,53 described elsewhere.56 The best fit for 7-4 is shown as the solid lines in Figur e 7-17 (left) and was obtained with S = 5/2 and either of the two sets of parameters: g = 1.96 and D = 0.58 cm-1, and g = 1.96 and D = -0.50 cm-1. Alternative fits with S = 3/2 or 7/2 were rejected because they gave unreasonable values of g and D It is common to obtain two acceptable f its of magnetization data for a given S value, one with D > 0 and the other with D < 0, since magnetization fits are no t very sensitive to the sign of D This was indeed the case for the magnetization fits for both the complexes 7-4 and 7-5 In order to assess which is the superior fit for these co mplexes and also to ensu re that the true global minimum had been located in each case, we calcu lated the root-mean-square error surface for the fits as a function of D and g using the program GRID .71 For 7-4 the error surface (Figure 7-18, left) clearly shows the two mini ma with positive and negative D values, with the fit with negative D being clearly superior and suggesti ng that this is the true sign of D However, it would require a more sensitive technique such as EPR spectroscopy to confirm this. The obtained magnetization data for 7-5 are plotted in Figure 7-17 (right) as M / NB vs H / T and it can be seen to saturate at ~9.29, suggesting an S = 5 ground state and g < 2. The resulting best fit of the data is shown as the so lid lines in Figure 7-17 (right), and was obtained with S = 5 and either g = 1.90, D = 0.45 cm-1 or g = 1.89, D = -0.28 cm-1. In this case also, the fit error surface (Figure 7-18, right) clea rly shows that the fit with negative D is far superior, suggesting this to be the true sign of D 7.3.3.3 Ac Studies on 7-1 to 7-5 To independently confirm the ground state S va lues, ac susc eptibility data was collected on microcrystalline samples of 7-1 to 7-5 in a 3.5 G ac field. The in-phase ( M' ) ac susceptibility signal is invaluable for assessing S without any complications from a dc field,62,69,224 and these signals for complexes 7-1 to 7-3 at 997 Hz are plotted as M'T vs T in Figure 7-19. Extrapolation

PAGE 166

166 of the plots to 0 K, from temperatures above ~5 K to avoid the effect of weak intermolecular interactions (dipolar and superexchang e), gives values of ~ 33, ~27 and ~37 cm3Kmol-1 for 7-1 to 7-3, respectively, corresponding to S = 8, 7 and 8, respectively, with g ~ 1.91, 1.96 and 2.02, in very satisfying agreement with the conclusion s from the fits of the dc magnetization data. None of the complexes displayed out-of-phase ( M") ac susceptibility peaks above 1.8 K. There were some very weak signs of the beginn ing of signals whose p eaks would lie well below 1.8 K, and these may correspond to the very small dips in the M'T plots of Figure 7-19 (left) at T < 2 K. However, it is clear that if any of the complexes 7-1 to 7-3 were single-molecule magnets (SMMs), they would at best have very small barriers to magnetizati on relaxation. In fact, complex 7-1 with its combination of S = 8 and D = -0.30 cm-1, would be predicted to have the largest barrier of the three complexes, with an upper limit ( U ) of U = S2| D | = 19.2 cm-1. However, the true or effective barrier ( Ueff) is expected to be significantly less than U especially given the low symmetry of the molecule, and it is perhaps not surprising that even if 7-1 were a SMM, it would be one only at very low temper atures <1 K and thus not a significant new addition to the family of SMMs. In-phase ac susceptibility signals for complexes 7-4 and 7-5 at 997 Hz are plotted as M'T vs T in Figure 7-19 (right). The M'T is essentially temperature independent below 15 K until ~ 4 5 K where there is a small decrease that can be assigned to low temperature effects such as ZFS of the ground state and/or very weak intermolecula r interactions. The essen tially constant values at > 5 K of ~4 and ~14 cm3Kmol-1 for 7-4 and 7-5 respectively, confirm S = 5/2 and 5 ground states with g < 2, whose spin-only ( g = 2.0) values are 4.38 and 15.0 cm3Kmol-1, respectively. Neither complex displayed out-of-phase ( M") ac susceptibility peaks above 1.8 K.

PAGE 167

167 7.3.3.4 Rationalization of the Ground State Spin of 7-4 and 7-5 It is of interest to attem pt to rationa lize the observed ground st ate spin values of 7-4 and 75. It is assumed that all Fe2 pairwise exchange interactions are antiferromagnetic, as is essentially always the case for high-spin Fe(III), and there will thus be competing antiferromagnetic exchange interactions and spin fr ustration effects within the many Fe3 triangular units in these complexes. In fact, for complex 7-4 its S = 5/2 ground state can be rationalized in an identical fashion based on spin frustration as we previously described for complex 5-1 in chapter 5, which has a similar [Fe5O6] core topology as 7-4 as stated earlier, and an identical S = 5/2 ground state.127 For complex 7-5 the spin alignments giving rise to the S = 5 ground state are again not obvious owing to spin frustration with in the triangular units of the Fe6 core. There are five inequivalent types of exchange interactions, J12, J13, J23, J23 and J33, the subscripts referring to the atom labels of Figure 7-9. In Table 7-7, we list the average Fe-O distances and the Fe-O-Fe angles for bridged Fe2 pairs within the molecule. It is we ll known that short Fe-O bond distances and large Fe-O-Fe angles lead to the larger J values.76,112,225,226 In complex 7-5 the Fe1/Fe3 and Fe2/Fe3 pairs, with only a single monoatomic brid ge, have both the shortest Fe-O superexchange pathways and the largest Fe-O-Fe angles in the molecule and are thus expected on the basis of magnetostructural correlations76 to have the strongest J values, in the order of ~38 cm-1. The Fe2/Fe3 pair, also with a single monoatomic bridge, has a slightly longer Fe-O pathway but still a large Fe-O-Fe angle; thus, it would also be expected to have a relatively strong J value, in the ~15 cm-1 region. In contrast, the Fe1/Fe2 and Fe3/Fe3 pairs, which are now bis-monoatomically bridged, have Fe-O distances similar to that for Fe2/Fe3 but by far the smallest Fe-O-Fe angles in the molecule, and would thus be expected to have the weakest J values, in the ~7 cm-1 region. The estimates given for all the pairs are based on the J values of Fe2 pairs with similar metric

PAGE 168

168 parameters in other Fe(III) clusters. Thus, we conclude that the Fe1/Fe2 and Fe3/Fe3 exchange will be frustrated by the other, stronger inte ractions, and as a result, the ground state spin alignments in the molecule will be as s hown in Figure 7-20 (top). The spins within Fe2 pairs monoatomically bridged by a single O atom are a ligned antiparallel, whereas those within the three bis-monoatomically bridged Fe2 pairs are spin frustrated by the other stronger interactions and forced to align parallel even though th eir exchange interactio ns are intrinsically antiferromagnetic. This situation predicts an S = 5 ground state for 7-5 as experimentally obtained. Note that it is not easy to formul ate reasonable alternativ e ways of getting an S = 5 ground state, and thus we feel confident on the above proposal. For example, if the Fe2/Fe3' interaction were considered weak enough to be frustrated by the Fe 3/Fe3' interaction, then this situation would give the spin alignments of Figure 7-20 (bottom) and an S = 0 ground state. It is difficult to rationalize the S = 0 ground state spin for 7-6 due to the high content of triangular units and large number of non-zero exchange interactions. 7.4 Conclusions The initial u se of edteH4 has proven to be a useful ne w route to a variety of novel Mnx and Fex clusters. It is a hexadentate chelate whose four alcohol groups o ffer, on deprotonation, the possibility of each bridging to one or more a dditional Mn and Fe atoms and thus fostering formation of high nuclearity products. In the pres ent work, we have described the synthesis and characterization of new Mn8, Mn12 and Mn20 products, all with unpreced ented structural features and all with significant gr ound state spin values of S = 7 or 8. Although the core of complex 7-4 is overall similar to that of a previous Fe5 complex with hmbp-, those of Fe6 complex 7-5 and particularly that of Fe12 complex 7-6 are unprecedented in Fe(III) ch emistry. The structures of the cations 7-7 and 7-8 are concluded to be the same give n their identical formulations and almost superimposable magnetic properties. We have also successfully rationalized the S = 5

PAGE 169

169 ground state of 7-6 The combined results emphasize the usefulness of the poly-alcohol-based chelate edteH4 as a route to new high-nuclearity products. Table 7-1. Crystallographic Data for 7-12CH2Cl2MeOH, 7-2 MeCNH2O and 7-3 10MeOH 7-1 7-2 7-3 Formulaa C80H94Cl4Mn8N4O30 C52H99Cl8Mn12N14O22.5 C82H182Cl2Mn20N12O66 Fw, g/mola 2172.91 2223.33 3562.05 Space group P 21/ n P 4/ ncc P 21/ c a, 16.0450(16) 19.7797(12) 17.0570(16) b, 17.6428(17) 19.7797(12) 25.409(2) c 31.896(3) 20.851(2) 15.9322(15) 90 90 90 95.425(2) 90 100.463(2) 90 90 90 V 3 8988.6(15) 8157.7(12) 6790.3(11) Z 4 4 2 T K 173(2) 173(2) 173(2) b 0.71073 0.71073 0.71073 calc, g/cm3 1.577 1.810 1.573 mm-1 1.340 2.125 1.886 R1 c,d 0.0334 0.0464 0.0812 wR2 e 0.0812 0.1046 0.2174 a Including solvate molecules. b Graphite monochromator. c I > 2 (I ). d R1 = (|| Fo| |Fc||) / | Fo|. e wR 2 = [ [ w( Fo 2 Fc 2)2] / [ w(Fo 2)2]]1/2, w = 1/[ 2(Fo 2) + [( ap)2 + bp ], where p = [max ( Fo 2, O) + 2 Fc 2]/3.

PAGE 170

170 Table 7-2. Crystallographic Data for 7-4 CH2Cl2, 7-5 2CHCl3, 7-6 MeCN and 7-7 7-4 7-5 7-6 7-7 Formulaa C60H59Cl2Fe5N2O21 C62H116Cl6Fe6N4O26 C60H116Cl4Fe12N12O52 C40H92Cl4Fe12N8O46 Fw, g/mola 1494.27 1881.39 2649.61 2233.17 Space group P 21/c C 2/c C 2/c C 2/c a, 21.3735(10) 14.211(2) 29.590(4) 30.502(3) b, 18.6612(9) 24.297(2) 29.641(4) 11.9702(11) c 17.7842(8) 25.676(3) 23.174(3) 30.517(3) 90 90 90 90 113.280(1) 94.783(3) 104.088(2) 111.404(1) 90 90 90 90 V 3 6515.81 8988.6(15) 19714(5) 10373.7 Z 4 4 8 4 T K 173(2) 173(2) 173(2) 173(2) b 0.71073 0.71073 0.71073 0.71073 calc, g/cm3 1.414 1.764 mm-1 1.210 1.916 R1 c,d 0.0575 0.0788 wR2 e 0.1364 0.2219 a Including solvate molecules. b Graphite monochromator. c I > 2 (I ). d R1 = (|| Fo| |Fc||) / | Fo|. e wR 2 = [ [ w( Fo 2 Fc 2)2] / [ w(Fo 2)2]]1/2, w = 1/[ 2(Fo 2) + [( ap)2 + bp ], where p = [max ( Fo 2, O) + 2 Fc 2]/3. Table 7-3. Bond-valence sums for the Mn atoms of complex 7-1 and 7-3a 7-1 7-3 Atom MnII MnIII MnIV MnII MnIII MnIV Mn1 3.21 2.93 3.08 1.85 1.70 1.78 Mn2 3.18 2.91 3.05 3.12 2.90 2.98 Mn3 2.01 1.87 1.92 3.21 2.93 3.08 Mn4 3.13 2.87 3.01 2.99 2.77 2.85 Mn5 3.14 2.87 3.01 2.48 2.26 2.38 Mn6 1.96 1.83 1.86 3.17 2.90 3.04 Mn7 3.17 2.90 3.04 3.22 2.94 3.09 Mn8 3.18 2.91 3.05 2.45 2.24 2.35 Mn9 3.05 2.79 2.93 Mn10 1.88 1.75 1.79 a The underlined value is the one closest to the charge for which it was calculated. The oxidation state of a particular atom can be taken as the neares t whole number to the underlined value.

PAGE 171

171 Table 7-4. Bond-valence sums for the O atoms of complex 7-1a Atom BVS Assignment group O21 1.99 OROMeO6 1.43 OHOHO15 1.95 ORedte4O16 1.93 ORedte4O17 1.94 ORedte4O18 1.96 ORedte4O7 2.02 ORedte4O8 1.21 ROH edteH2 2O9 1.17 ROH edteH2 2O10 1.98 ORedte4aThe BVS values for O atoms of O2-, OHand H2O groups are typically 1.8-2.0, 1.0-1.2 and 0.2-0.4, respectively, but can be affected somewhat by hydrogen-bonding. Table 7-5. Bond-valence sumsa for the Mn and O atoms of complex 7-2 Manganese BVS Atom MnII MnIII MnIV Mn1 2.23 2.12 2.17 Mn2 3.34 3.09 3.22 Mn3 3.05 2.82 2.91 Oxygen BVS Atom BVS assignmentgroup O1 1.98 O2O2O4 0.99 OHOHO2 1.89 ORedte4O3 1.97 ORedte4O5 1.97 ORedte4O6 2.03 ORedte4a The underlined value is the one closest to the charge for which it was calculated. The oxidation state of a particular atom can be taken as the neares t whole number to the underlined value. aThe BVS values for O atoms of O2-, OHand H2O groups are typically 1.8-2.0, 1.0-1.2 and 0.2-0.4, respectively.

PAGE 172

172 Table 7-6. Bond-valence sums for the O atoms of complex 7-5 and 7-6a 7-5 Atom BVS assignmentgroup O3 1.82 ORedte4O4 1.94 ORedte4O5 0.90 ROH edteH O13 1.88 ORedte47-6 Atom BVS assignmentgroup O1 1.95 ORedte4O2 1.87 ORedte4O3 1.97 ORedte4O4 1.98 ORedte4O5 0.50 H2O H2O O16 0.41 H2O H2O O17 1.24 OHOHO18 1.20 OHOHaThe BVS values for O atoms of O2-, OHand H2O groups are typically 1.8-2.0, 1.0-1.2 and 0.2-0.4, respectively. Table 7-7. Selected Fe-O distances and Fe-O-Fe angles for 7-5 Fe2 pair Avg. Fe-O () Angle (deg) Fe1/Fe3 1.895 123.8 Fe2/Fe3 1.883 123.2 Fe2/Fe3' 2.000 118.7 Fe3/Fe3' 2.026 102.3 Fe1/Fe2 1.980 98.1 (avg.) N N HO dmemHNH HN HO OH N N HO OH OH HO heenH2edteH4 Figure 7-1. Structure of ligands.

PAGE 173

173 Figure 7-2. (left) Labeled representation of the cation of 7-1 Hydrogen atoms and phenyl rings (except for the ipso carbon atoms) have been omitted for clarity. JT axes are shown as thicker black bonds. (right) The core of 7-1 Color code: MnIII, green; MnII, purple; O, red; N, blue; C, grey. Figure 7-3. Crystallographically esta blished coordination modes of edte4and edteH2 2found in complexes 7-1 to 7-3. Color code: MnIII, green; MnII, purple; O, red; N, blue; C, grey.

PAGE 174

174 Figure 7-4. (top) Labeled repres entation of the structure of 7-2 (bottom) The core of 7-2 viewed along (left) the c axis, and (righ t) the b axis. Color code: MnIII, green; MnII, purple; Cl, cyan; O, red; N, blue; C, grey.

PAGE 175

175 Figure 7-5. (top) The stru cture of the cation of 7-3 (middle and bottom) The core of 7-3 from different viewpoints emphasizing the 3 x 2 cubane arrangement. Color code: MnIII, green; MnII, purple; O, red; N, blue; C, grey.

PAGE 176

176 Figure 7-6. Labeled representation of the core of 7-3 JT axes are shown as thicker black bonds. Color code: MnIII, green; MnII, purple; O, red; N, blue; C, grey.

PAGE 177

177 Figure 7-7. Labeled representa tion of the structure of 7-4 with core Fe-O bonds as thick black lines; only the ipso benzoate carbon atoms are shown. Color code: Fe, green; O, red; N, blue; C, grey. Figure 7-8. Crystallographically esta blished coordination modes of edte4and edteH3found in complexes 7-4 to 7-6. Color code: Fe, green; O, red; N, blue; C, grey. Figure 7-9. Labeled representa tion of the structure of 7-5 with core Fe-O bonds as thick black lines; methyl groups on pivalate groups have been omitted for clarity. Color code: Fe, green; O, red; N, blue; C, grey.

PAGE 178

178 Figure 7-10. (top) Labeled repr esentation of the cation of 7-6 with core Fe-O bonds as thick black lines. (bottom) Core of 7-6 Color code: Fe, green; O, red; N, blue; C, grey.

PAGE 179

179 Figure 7-11. Labeled repres entation of the cation of 7-7 Color code: Fe, green; O, red; N, blue; C, grey. Figure 7-12. Plots of MT vs T for complexes 7-1 H2O ( ), 7-2 ( ) and 7-3 H2O ( )

PAGE 180

180 Figure 7-13. Plots of reduced magnetization ( M / N B) vs H / T for complexes 7-1 H2O (left) and 7-2 (right). The solid lines are the fits of th e data; see the text for the fit parameters. Figure. 7-14.Two-dimensional contour plot of the r.m.s. error surface vs D and g for the magnetization fit for 7-1 H2O (left) and 7-2 (right).

PAGE 181

181 Figure 7-15. (left) Plots of reduced magnetization ( M / N B) vs H / T for complex 7-3 H2O. (right) Two-dimensional contour plot of the r.m.s. error surface vs D and g for the magnetization fit for 7-3 H2O. Figure 7-16. Plots of MT vs T for complexes 7-4 to 7-8.

PAGE 182

182 Figure 7-17. Plot of reduced magnetization ( M / N B) vs H / T for complexes 7-4 (left) and 7-5 2CHCl3H2O (right). The solid lines are the fit of the data; see the text for the fit parameters. Figure 7-18. Two-dimensional contour plot of the fitting error surface vs D and g for 7-4 (left) and 7-5 2CHCl3H2O (right).

PAGE 183

183 Figure 7-19. (left) Plot of M T vs. T for complexes 7-1 H2O ( ), 7-2 ( ) and 7-3 H2O () at 997 Hz. (right) Plot of M T vs. T for complexes 7-4 () and 7-5 2CHCl3H2O ( ) at 997 Hz. O Fe3' Fe2 O Fe1 O O Fe3 Fe2' O Fe1' O O O O Fe3' Fe2 O Fe1 O O Fe3 Fe2' O Fe1' O O O S=5 S=0 Figure 7-20. (top) Spin alignments at the six S = 5/2 Fe(III) atoms of 7-5 rationalizing its S = 5 ground state, based on the arguments given in the text. (bottom) Spin alignments if the strengths of the Fe2/Fe3 and Fe3/Fe3 couplings were reversed, showing that the wrong ground state would be obtained.

PAGE 184

184 CHAPTER 8 SINGLE-MOLECULE MAGNETISM AND M AGNE TOSTRUCTURAL CORRELATION WITHIN A FAMILY OF [MnIII 10LnIII 2] COMPLEXES 8.1 Introduction Single-m olecule magnets (SMMs) are individual molecules that can function as nanoscale magnetic particles below th eir blocking temperature ( TB). They thus represent a molecular (or bottom-up) approach to nanoscale magnetic materials, and one that retains all the advantages of molecular chemistry, particularly monodispersity, solubility, crystallinity, and a periphery of organic ligands.13,16,154,227 The SMM property results from a combination of a high-spin ground state (S) and an easy-axis type magnetoanisotropy (negative zero-field splitting parameter, D ) and can be determined experimentally by the observation of frequencydependent out-of-phase ( ") signals in ac magnetic susceptibility measur ements, and hysteresis loops in magnetization vs. dc field scans.13,15,19,228 Since the initial discovery of the Mn12 family of SMMs, a number of different structural types have been synthesized, the majority of which have been Mn clusters composed completely or partially of MnIII atoms, since the Jahn-Tell er distortion of high-spin MnIII in near-octahedral geometry is the main source of the molecular anisotropy.21,22,171,213,228-232 In order to develop new synthetic routes fo r the synthesis of new SMMs, we and others have recently been focusing on polynuclear 3d-4f complexes.233-245 The hope has been that the large anisotropy of most of the LnIII ions, as well as their often large number of unpaired electrons, would enhance the SMM property by ra ising the barrier to magnetization relaxation. The initial success in 2004 in the synthesis of a mixed-meta l, ferromagnetically-coupled Tb2Cu2 SMM complex with a slow ma gnetization relaxation rate,246 provided a proof-of-feasibility that amalgamation of transition metals with anisotropic LnIII ions can have a major impact on the resulting relaxation barrier. Since then, this area has steadily built up momentum and there are now several types of 3d-4f SMMs, most of which are in Mn-Ln chemistry: [Mn11Dy4]239,

PAGE 185

185 [Mn2Dy2]240 [Mn6Dy6]241, [Mn6Dy4]234 and [Mn11Gd2].247 All these complexes display frequency-dependent out-of-phase ac susceptibility signals at low temperatures, but it was the [Mn11Dy4]239 complex that first confirmed that such Mn-Ln SMMs can exhibit clear magnetization hysteresis and quantum tunneling of magnetization (QTM) th rough the anisotropy barrier. The first 3d-4f SMMs in Fe chem istry have also now been reported, [Fe2Ho2(OH)2(teaH)2(O2CPh)4(NO3)2] and [Fe2Dy2(OH)2(teaH)2(O2CPh)6] (teaH3 = triethanolamine).245 The total number of examples of 3d-4f SMMs is thus still limited, and new synthetic methodologies to additional examples ar e of continuing interest, especially with the hope that some may have increased magnetizat ion relaxation barriers, i.e. higher blocking temperatures, TB. These points represent a major stimulus of this work. As part of our continuing inte rest in the synthesis of new Mn-Ln SMMs, we have explored the use of 2-hydroxymethylpyridine (hmpH) in mi xed-metal reactions. This alcohol containing group on deprotonation has been a comm on route to various homometallic Mnx clusters in our previous work,34,63,211 and we expected that it might also prove a useful route to new Mn-Ln species. Indeed, we report in this paper our development of a synthetic procedure employing hmpH that has successfully led to a family of MnIII 10LnIII 2 isostructural complexes spanning most LnIII ions: Ln = Pr ( 8-1 ), Nd ( 8-2 ), Sm ( 8-3 ), Gd ( 8-4 ), Tb ( 8-5 ), Dy ( 8-6 ), Ho ( 8-7 ) and Er (8-8 ). In addition to showing that some of these are new 3d-4f SMMs, we also take advantage of the availability of an isostructu ral series to map out the 3d4f exchange coupling (and SMM properties) as a function of the LnIII ion employed. It is rare to ha ve such an opportunity in high nuclearity Ln-containing mixed-metal complexes.23 In general, the large orbital angular momentum in LnIII ions, except for GdIII, prevents the convenient use of an isotropic (Heisenberg) spin Hamiltonian fo r interpretation of the magnetic properties of polynuclear 3d-4f

PAGE 186

186 clusters. Costes et al. and Kahn et al. introduced an empirical appr oach of comparing a 3d-4f system with an isostructural one in which the LnIII is replaced by a diamagnetic MIII ion, thus allowing the effect of the LnIII to be factored out.248-250 In 2003, Figuerola et al. extended this to an FeIII-LnIII species by comparison with the isostructural CoIII-LnIII and FeIII-LaIII complexes, thus factoring out both the individual FeIII and LnIII properties.251 Since then, a variety of such studies have been carried out on dinuclear, trin uclear, tetranuclear, 1-D and 2-D cyanide-bridged 3d-4f assemblies.233,236,238,252-257 However, this strategy has not been employed as yet on polynuclear Mn-Ln oxide clusters relevant to the SMM field, and to the search for increased barriers to magnetization relaxation. In order to do so in the present work, we have also prepared the isostructural MnIII 10YIII 2 with diamagnetic YIII for comparison with the Mn-Ln complexes, which has allowed characterization of the magnetic properties of the MnIII 10 portion of the structure. We herein describe the combined results of this comparative investigation of the properties of this family of complexes. 8.2 Experimental Section 8.2.1 Syntheses All preparations were perfor med under aerobic conditions using reagents and solvents as received. (NBun 4)[Mn4O2(O2CPh)9(H2O)] was prepared as previously described.185 [Mn10Dy2O8(O2CPh)10(hmp)6(NO3)4] ( 8-6 ). To a stirred solution of (NBun 4)[Mn4O2(O2CPh)9(H2O)] (0.36 g, 0.23 mmol) in MeOH/MeCN (1/19 mL) was added Dy(NO3)3H2O (0.10 g, 0.23 mmol) followed by hmpH (0.02 mL, 0.23 mmol). The mixture was stirred for an hour, filtered to remove undissolv ed solid, and the filtrate layered with Et2O. X-ray quality red-brown crystals of 8-6 MeCNMeOH were obtained over a period of one week. These were collected by f iltration, washed with Et2O, and dried in vacuo ; the yield was 15%. The dried solid analyzed as solvent free. Anal. Calcd. (Found) for 8-6 (C106H86Dy2Mn10N10O46):

PAGE 187

187 C, 40.93 (40.50); H, 2.79 (2.90); N, 4.50 (4.24)%. Selected IR data (cm-1): 3434(br), 3063(w), 1707(w), 1605(m), 1566(s), 1487(m), 1385(s), 1308(w), 1228(w), 1175( w), 1157(w), 1069(m), 1051(w), 1027(w), 817(w), 765(w), 718(m), 668 (m), 614(w), 548(m), 465(w), 429(w). [Mn10Pr2O8(O2CPh)10(hmp)6(NO3)4] (8-1). Complex 8-1 was prepared following the same procedure as for 8-6 but with Pr(NO3)3H2O (0.095 g, 0.23 mmol). The yield was 25%. Anal. Calcd. (Found) for 8-1 (C106H86N10Mn10O46Pr2): C, 41.51 (41.55); H, 2.83 (2.84); N, 4.57 (4.38)%. Selected IR data (cm-1): 3434(br), 3063(w), 1707(w), 1606(m), 1566(m), 1403(s), 1290(w), 1176(w) 1069(m), 1051(w), 1027(w), 820( w), 763(w), 718(m), 661(m), 549(m), 429(w). [Mn10Nd2O8(O2CPh)10(hmp)6(NO3)4] (8-2). Complex 8-2 was prepared following the same procedure as for 8-6 but with Nd(NO3)3H2O (0.10 g, 0.23 mmol). The yield was 20%. Anal. Calcd. (Found) for 8-2 (C106H86N10Mn10O46Nd2): C, 41.42 (41.77); H, 2.82 (2.88); N, 4.56 (4.76)%. Selected IR data (cm-1): 3446(br), 3063(w), 1698(w), 1607(m), 1566(s), 1473(m), 1401(s), 1315(w), 1291(w), 1175(w), 1157(w), 1069(m), 1050(w), 762(w), 718(m), 660(m), 612(w), 549(m), 460(w), 429(w). [Mn10Sm2O8(O2CPh)10(hmp)6(NO3)4] (8-3). Complex 8-3 was prepared following the same procedure as for 8-6 but with Sm(NO3)3H2O (0.10 g, 0.23 mmol). The yield was 12%. Anal. Calcd. (Found) for 8-3 (C106H86N10Mn10O46Sm2): C, 41.26 (41.36); H, 2.81 (2.73); N, 4.54 (4.21)%. Selected IR data (cm-1): 3436(br), 3063(w), 1707(w), 1606(m), 1566(s), 1485(m), 1402(s), 1291(w), 1230(w), 1176(w), 1070(w), 1051( m), 1027(w), 819(w), 765(w), 718(m), 663(m), 614(w), 551(m), 468(w). [Mn10Gd2O8(O2CPh)10(hmp)6(NO3)4] ( 8-4 ). Complex 8-4 was prepared following the same procedure as for 8-6 but with Gd(NO3)3H2O (0.10 g, 0.23 mmol). X-ray crystallography

PAGE 188

188 characterized the obtained crystals as 8-4 MeCNMeOH. The yield was 10%. Anal. Calcd. (Found) for 8-4 (C106H86Gd2Mn10N10O46): C, 41.07 (41.35); H, 2.80 (2.81); N, 4.52 (4.22)%. [ Mn10Tb2O8(O2CPh)10(hmp)6(NO3)4] (8-5). Complex 8-5 was prepared following the same procedure as for 8-6 but with Tb(NO3)3H2O (0.10 g, 0.23 mmol). The yield was 20%. Anal. Calcd. (Found) for 8-5 (C106H86N10Mn10O46Tb2): C, 41.03 (41.03); H, 2.79 (2.80); N, 4.51 (4.31)%. Selected IR data (cm-1): 3432(br), 3062(w), 1710(w), 1605(m), 1566(s), 1487(m), 1401(s), 1291(w), 1175(w), 1157(w), 1070(m), 1051(w), 1026(w), 818(w), 765(w), 718(m), 665(m), 616(w), 550(m), 465(w). [Mn10Ho2O8(O2CPh)10(hmp)6(NO3)4] (8-7). Complex 8-7 was prepared following the same procedure as for 8-6 but with Ho(NO3)3H2O (0.10 g, 0.23 mmol). The yield was 15%. Anal. Calcd. (Found) for 8-7 H2O (C106H90N10Mn10O48Ho2): C, 40.40 (40.11); H, 2.88 (2.79); N, 4.44 (4.23)%. Selected IR data (cm-1): 3432(br), 3063(w), 1602(m), 1565(s), 1488(m), 1385(s), 1309(w), 1175(w), 1157(w), 1069(m), 1051(w), 1025(w), 765(w), 718(m), 670(m), 547(m), 466(w). [Mn10Er2O8(O2CPh)10(hmp)6(NO3)4] (8-8). Complex 8-8 was prepared following the same procedure as for 8-6 but with Er(NO3)3H2O (0.10 g, 0.23 mmol). The yield was 10%. Anal. Calcd. (Found) for 8-8 H2O (C106H92N10Mn10O48Er2): C, 40.11 (39.85); H, 2.92 (2.84); N, 4.41 (4.23)%. Selected IR data (cm-1): 3430(br), 3063(w), 1709(w), 1603(m), 1565(s), 1488(m), 1400(s), 1306(w), 1175(w), 1157(w), 1070(m), 1051(w), 1027(w), 817(w), 765(w), 717(m), 673(m), 612(w), 549(m). [Mn10Y2O8(O2CPh)10(hmp)6(NO3)4] ( 8-9 ). Complex 8-9 was prepared following the same procedure as for 8-6 but with Y(NO3)3H2O (0.087 g, 0.23 mmol). X-ray crystallography characterized the obtained crystals as 8-9 MeCN. The yield was 15%. Anal. Calcd. (Found) for

PAGE 189

189 8-9 (C106H86N10Mn10O46Y2): C, 42.97 (42.69); H, 2.92 (2.83); N, 4.73 (5.17)%. Selected IR data (cm-1): 3421(br), 3063(w), 1698(w), 1602(m), 1564(s), 1506(m), 1385(s), 1311(w), 1176(w), 1069(m), 1051(w), 765(w), 718(m) 667(m), 547(w), 467(w). 8.2.2 X-ray Crystallography Data were collected by Dr. Khalil A Abboud on a Siemens SMART PLATFORM equipped with a CCD area detector and a graphite monochromator utilizing Mo-K radiation ( = 0.71073 ). Suitable crystals of 8-4 MeCNMeOH, 8-6 MeCNMeOH and 8-9 MeCN were attached to glass fibers using silicone grease and transferred to a goniostat where they were cooled to 173 K for data collection. Cell parame ters were refined using up to 8192 reflections. A full sphere of data (1850 fram es) was collected using the -scan method (0.3 frame width). The first 50 frames were re-measured at the end of the data collection to monitor instrument and crystal stability (maximum correction on I was <1 %). Absorption co rrections by integration were applied based on measured indexed crystal faces. The structure was solved by the direct methods in SHELXTL6,258 and refined on F2 using full-matrix least-squares. The non-H atoms were treated anisotropically, whereas the H atom s were placed in calculated, ideal positions and refined as riding on their respective C atoms. For 8-4 MeCNMeOH and 8-6 MeCNMeOH, the asymmetric unit consists of a half Mn10Gd2 or Mn10Dy2 cluster, a disordered MeCN in a general position, and a half MeCN disordered against a half MeOH molecule a bout an inversion cente r. A total of 825 ( 8-4 ) or 817 ( 8-6 ) parameters were refined in the final cycle of refinement using 20616 ( 8-4 ) or 10817 (8-6 ) reflections with I > 2 (I) to give R1( wR2) of 6.43(17.13) and 5.08(13.30)% for 8-4 and 8-6 respectively. For 8-9 MeCN, the asymmetric unit consists of a half Mn10Y2 cluster and two MeCN molecules. The latter are disordered and could not be modeled properly, thus program

PAGE 190

190 SQUEEZE,68 a part of the PLATON package of crysta llographic software, was used to calculate the solvent disorder area and remove its contribut ion to the overall intensity data. A total of 778 parameters were refined in the final cycl e of refinement using 6197 reflections with I > 2 (I) to yield R1( wR2) of 7.10(16.14)%, respectively. Unit cell data and structure refinement details for 8-4 3MeCNMeOH, 8-6 MeCNMeOH and 8-9 MeCN are listed in Table 8-1. 8.3 Results and Discussion 8.3.1 Syntheses Many synthetic m ethods to high-nuclearity MnIII-containing clusters and SMMs have involved the reaction of a ch elate with preformed [Mn3O(O2CR)6L3]0,+ oxide-centered triangular complexes.32,259-262 The chelate has the dual function of encouraging molecular products rather than polymers, and fostering high-nuclearity pr oducts if good binding gro ups such as alkoxides are present. In the present wor k, the preformed tetranuclear MnIII 4 complex (NBun 4)[Mn4O2(O2CPh)9(H2O)] was employed.185 This has also proven in the past to be a good stepping-stone to high nuclearity products.179,263,264 Reaction of (NBun 4)[Mn4O2(O2CPh)9(H2O)] with 1 equiv each of hmpH and Ln(NO3)3 (Ln = Pr ( 8-1 ), Nd ( 8-2 ), Sm ( 8-3 ), Gd ( 8-4 ), Tb ( 8-5 ), Dy ( 8-6 ), Ho ( 8-7 ) and Er ( 8-8 )) in MeCN:MeOH led to subse quent isolation of red-brown crystals of [Mn10Ln2O8(O2CPh)10(hmp)6(NO3)4] in fair yields of 1030%. The yields are not optimized and we were happy to settle for lower yi elds of well-formed, pure material rather than add more Et2O co-solvent, which would contaminate th e products with by-products such as NBun 4NO3 and others. The reaction is summarized in eq. 8-1. 5[Mn4O2(O2CPh)9(H2O)]+ 4Ln3+ + 12hmpH + H2O 2[Mn10Ln2O8(O2CPh)10(hmp)6]4+ + 25PhCO2 + 24H+ (8-1)

PAGE 191

191 We had hoped to make the complete series fo r every Ln (except Pm). However, the later lanthanides Tm, Yb and Lu gave products that were clearly not isostructural with 8-1 8-8 and we assume this is related to their decreased si ze compared with the othe rs. We were also unable to get the Eu analogue. The Ce reaction suffere d from reactions that we assume involve redox reactions; we have seen elsewhere on multiple occasions that Ce favors the CeIV oxidation state in mixed-metal Mn-Ce chemistry involving high oxidation state MnIII/MnIV.264,265 To benefit the magnetism studies, as stated in the Introduction, we also attempted and succeeded in preparing the isostructural complex [Mn10Y2O8(O2CPh)10(hmp)6(NO3)4] ( 8-9 ), confirmed crystallographically (vide infra), by the same route. All these reactions were very sensitive to the Mn4:hmpH:LnIII ratio; other ratios were found to give poor crystallinity and/or mixtures of products. The mixed MeCN:MeOH solven t system is also very important to give clean products 8-1 8-9. The structures of representative LnIII complexes 8-4 and 8-6 and the YIII complex 8-9 were determined by X-ray crystallography; all the complexes gave essentially superimposable IR spectra, and elemental analys es in agreement with the given formulations, and we conclude on these bases that the complexe s are all isostructural. The compounds are air stable, but interstitial solvent molecules are easil y lost during vacuum drying and the solids are slightly hygroscopic. 8.3.2 Description of Structures The labeled structures of 8-4 8-6 and 8-9 are shown in Figure 8-1, and selected interatomic distances and angles for 8-4 8-6 and 8-9 are provided in Tables A-21, A-22 and A23 respectively. The complexes all cr ystallize in triclinic space group P with the Mn10Ln2 molecules lying on inversion centers and their structures are esse ntially identical except for the identity of the LnIII or YIII atoms. Therefore, we will only de scribe the structure of complex 8-6 here. The complex contains a [Mn10Dy2(3-O2-)4(4-O2-)4]20+ core that can be dissected into five

PAGE 192

192 parallel layers of three types with an ABCBA arrangement (Figure 8-2). Layer A is the Dy1 atom, layer B consists of a triangular MnIII 3 unit (Mn1, Mn2, Mn3), and layer C consists of a MnIII 4 rhombus (Mn4, Mn4', Mn5, Mn5'). Each laye r is held together and linked to its neighboring layers by a combination of four 3-O2-, two 4-O2and ten benzoate groups, six of which are in 1: 1: -bridging modes and four are in 2: 1: 3-bridging modes. Peripheral ligation is provided by four chelating NO3 groups, two on each Dy1 atom, and six chelating hmpgroups, one each on the Dy1 atoms, and Mn2, Mn2 Mn3 and Mn3 of layer B ; the hmpalkoxide arms also bridge Dy atoms with Mn atoms of layer B or vice-versa. The Mn and Dy atoms are six and nine coordi nate, respectively, and the MnIII oxidation states were determined using a combination of charge-b alance considerations, inspection of metric parameters and bond valence sum (BVS) calculations (Table 8-2).48 As expected, the MnIII centers exhibit a JahnTeller (JT) distortion, as e xpected for a high-spin d4 ion in near-octahedral geometry, and takes the form of the usual axial elonga tion. The JT elongation axes of the MnIII atoms of layer C are approximately parallel to each other (thicker bl ack bonds in Figure 8-2, bottom right). In layer B the JT axes of Mn1 and Mn3 are approximately pa rallel but that of Mn2 is roughly perpendicular to them (Figure 8-2, bottom left). The overall structure of these Mn10Ln2 complexes is unprecedented in 3d-4f chemistry, but that of the central BCB Mn10 subunit is similar to that in the homometallic complexes [Mn10O8(O2CPh)6L8] (L is the anion of picolin ic acid or dibenzoylmethane),266 which also comprises two Mn3 triangular units above and below a central Mn4 planar unit; however, there are significant differences in the exact dispositio n of the three units and in the resulting metric parameters.

PAGE 193

193 8.3.3 Magnetochemistry Solid-s tate variable-temperature dc magnetic susceptibility data were collected on powdered microcrystalline samples of complexes 8-1 8-9 in the 5.0 300 K range and in a 0.1 T magnetic field. We will first discuss the data for [Mn10Y2] ( 8-9 ) and [Mn10Gd2] ( 8-4 ): the first will allow characterization of the magnetic properties of the Mn10 sub-unit alone, and the second will show the resultant of its ex change coupling with isotropic Gd3+ ions ( S = 7/2, 8S7/2 free-ion term). This will assist the interpretation of the data for the other complexes, which contain strongly anisotropic LnIII ions. 8.3.3.1 Complexes 8-9 (Mn10Y2) and 8-4 (Mn10Gd2) The plots of MT vs T for 8-4 and 8-9 are shown in Figure 8-3. For 8-9 the value of MT smoothly decreases from 28.4 cm3Kmol-1 at 300 K to 10.7 cm3Kmol-1 at 5 K. The 300 K value is slightly less than the spin-only ( g = 2) value of 30 cm3Kmol-1 for ten MnIII ions with noninteracting metal centers and decreases with decr ease in temperature, indicating the presence of dominant intramolecular antiferromagnetic excha nge interactions. The 5.0 K value suggests an S = 4 ground state. For 8-4 the value of MT decreases from 37.9 cm3Kmol-1 at 300 K to 32.4 cm3Kmol-1 at 50 K, stays roughly constant down to 15 K, and then decreases rapidly to 27.4 cm3Kmol-1 at 5 K. The 300 K value is much less than the spin-only value of 45.7 cm3Kmol-1 for ten MnIII and two GdIII non-interacting ions. The value of MT at 5 K suggests an S = 7 ground state. To confirm the above ground state spin estimates, magnetization ( M ) data were collected at various fields up to 7 T and in the 1.8-10 K temper ature range. The resultin g data are plotted in Figure 8-4 as reduced magnetization ( M / N B) vs H / T where N is Avogadros number and B is the Bohr magneton. The data were fit, using the program MAGNET,53 by diagonalization of the spin Hamiltonian matrix assuming only the gr ound state is populate d, incorporating axial

PAGE 194

194 anisotropy ( D z 2) and Zeeman terms, and employing a full powder average. The corresponding spin Hamiltonian is given by eq 8-2, where Sz is the z-axis spin operator, g is the electronic g H = D z 2 + gB0 H (8-2) factor, 0 is the vacuum permeability, and H is the applied field. The last term in eq 8-2 is the Zeeman energy associated with the app lied magnetic field. An acceptable fit for 8-9 was obtained with S = 4, g = 2.01 and D = 0.89 cm-1 using data collected in the 0.1-0.8 T field range; this fit is shown as the solid lines in Figure 8-4 (left). Alternative fits with S = 3 or 5 were rejected because they gave unreasonable values of g of 2.67 and 1.65, respectively. We could not get a good fit when we included data collected at fields higher than 0.8 T, as is often the case for such polynuclear complexes where there are lowlying excited states as a result of weak interactions and/or sp in frustration effects.62,267 To assess the hardness of the fit and the resulting uncertainties in g and D a root-mean square D vs g error surface for the fit was generated for 8-9 using the program GRID,71 which calculates the differe nce between the experimental M/NB data and those calculated for various combinations of D and g. The corresponding D vs g 2-D contour plot for 8-9 is provided in Figure 8-5 and shows a soft minimum from which we estimate the uncertainties in the fit parameters to be S = 4, g = 2.01(1), and D = 0.89(5) cm-1. For Mn10Gd2 complex 8-4 even more difficulty was encount ered in obtaining a good fit of the data, which makes sense given that exchange couplings involving lant hanide ions are very weak as a result of the buried na ture of f orbitals, and thus ther e will be an even greater number of very low-lying excited st ates. The best fit was with S = 7, g = 2.00 and D = -0.11 cm-1, and this is shown as the solid lines in Figure 84 (right). The overestim ation by the fit of the experimental data at the lower fields of the plot suggests that the true ground state may be S = 6 with a very low-lying S = 7 excited state that is stabili zed below the true ground state by the

PAGE 195

195 applied field. Thus, the only safe conclusion to be drawn for the magneti zation fit is that the ground state of 8-4 is probably either S = 6 or S = 7. We shall return to this point below. The corresponding D vs g 2-D contour plot for 8-4 is provided in Figure 8-5 (right) and also shows a soft minimum from which we estimate the uncertainties in the fit parameters to be S = 7, g = 2.00(2), and D = -0.11(1) cm-1. To probe the ground states of 8-4 and 8-9 further, and to assess their magnetization relaxation dynamics, ac susceptibility data were collected in the 1.8-10 K range under a 3.5 G ac field oscillating at frequencies in the 50-1000 Hz range. The obtain ed in-phase ac susceptibility ( M ) signals are plotted as M'T vs T in Figure 8-3 (right) and are invaluable as an additional and independent means to determine the ground state sp in of a molecule without any complications from a dc field.69,224 M'T for 8-9 steadily decreases with decreasing T consistent with depopulation of low-lying excited states, and extrapolates to just under 10 cm3Kmol-1 at 0 K; this indicates an S = 4 ground state and g ~ 2, in agreement with the dc magnetization fit. M'T for 8-4 again decreases steadily down to 8 K and then more rapidly, appearing to be heading for ~19 cm3Kmol-1. Assuming the latter decr ease is due to depopulation of excited states, the extrapolation indicates an S = 6 ground state and a very low-ly ing excited state(s). The ac data thus confirm the conclusions of the dc magnetization fit that 8-4 has a greater ground state S value than 8-9 ; in any case, the precise ground state spin value of 8-4 is not essential for the analyses described below. 8.3.3.2 Comparison of 8-9 (Mn10Y2) with 8-4 (Mn10Gd2), 8-5 (Mn10Tb2), 8-6 (Mn10Dy2), 8-7 (Mn10Ho2), and 8-8 (Mn10Er2) The study of 3d-4f complexes has been of incr eased interest to magnetochemists since the discovery of intrinsic ferromagnetic coupling in the CuII-GdIII pair.268,269 In the present work, we have similarly sought to identify the na ture of the interactions between the Mn10 unit and the

PAGE 196

196 LnIII atoms of 8-1 8-8 The availability of [Mn10Y2] ( 8-9 ) allows an empirical approach analogous to that introduced by Costas et al and Kahn et al .249,250 The magnetic properties of a [Mn10Ln2] complex are governed by (i) the magnetic properties of the central Mn10 unit resulting from its many MnIIIMnIII interactions; (ii) MnIIILnIII interactions between the Mn10 unit and the LnIII atoms; and (iii) the thermal population of the spin-orbit (Stark) components of the LnIII ion. Complex 8-9 has allowed point (i) to be separately el ucidated. Insights into the nature of the MnIIILnIII interactions can now be obtained by comparing the MT vs T data for [Mn10Ln2] and [Mn10Y2] (assuming that the three separate MnIIILnIII interactions at each end of the molecule are insignificantly different, which is reasona ble given that they are all bridged by a 4-O2and an hmpalkoxide arm (Figures 8-1 and 8-2). This is carried out by determining the difference ( MT ) between MT for a [Mn10Ln2] cluster and [Mn10Y2] (eq 8-3). ( MT ) will thus reflect both points (ii) and (iii), except for isotropic GdIII ( 8-4 ) where it directly probes point (ii). ( MT ) = (MT )MnLn ( MT )MnY (8-3) We do not have access to the analogous CoIII 10Ln2 complexes, i.e. containing diamagnetic CoIII in place of the MnIII of 8-1 8-8 and thus cannot therefore separately determine the exact LnIII MT vs T behavior in these complexes, i.e. point (i ii). However, the latter is available in the literature for LnIII ions in a variety of coordination environments similar to 8-1 8-8 and thus eq 8-3 can be employed to obtain importan t insights into the nature of the MnIIILnIII interactions in these [Mn10Ln2] complexes. Application of eq 8-3 to 8-4 and 8-9 gives the ( MT ) shown as a dashed line in Figure 8-3 (left). ( MT ) increases from 9.5 cm3Kmol-1 at 300 K to 18.7 cm3Kmol-1 at 10 K, and then drops to 16.6 cm3Kmol-1 at 5 K. Since MT vs T for GdIII is essentially temperature-independent4 (Figure 8-6), the steady increase in ( MT ) with decreasing T suggests the MnIIILnIII exchange

PAGE 197

197 interactions to be weakly ferromagnetic. This is consistent with the increase in ground state spin between 8-9 ( S = 4) and 8-4 ( S = 6 or 7). Note that it is not expected that an S = 11 ground state should result for 8-4 from weak ferromagnetic MnIIIGdIII interactions; there will be extensive spin frustration within the many triangular units of the Mn10 core, and resulting intermediate spin alignments, so weak ferromagnetic MnIIIGdIII interactions (likely comparable in strength to some MnIIIMnIII interactions), are by no means expect ed to give a simple spin summation. Indeed, it is even possible that the Mn10 portion of 8-4 may not still be effectively S = 4 once the extra MnIIIGdIII interactions are introduced. The MT for 8-5 (Mn10Tb2) increases from 46.7 cm3Kmol-1 at 300 K to 64.3 cm3Kmol-1 at 5 K (Figure 8-7, left). The 300 K value is sli ghtly less than the spin only value of 53.6 cm3Kmol-1 expected for ten MnIII ions and two TbIII (4f8, 7F6, 23.6 cm3Kmol-1) non-interacting ions (using the free-ion approximation for Tb) owing to the antiferromagnetic interactions within the Mn10 unit. The difference ( MT ) increases sharply with decrea sing temperature, from 18.2 cm3Kmol-1 at 300 K to 53.5 cm3Kmol-1 at 5 K. This indicates ferromagnetic MnIIITbIII interactions. The MT for 8-6 (Mn10Dy2) decreases slightly from 52.3 cm3Kmol-1 at 300 K to 48.0 cm3Kmol-1 at 50 K and then increases again reaching 51.9 cm3Kmol-1 at 5 K (Figure 8-7, right). The 300 K value is slightly less than th e spin only value of 58.3 cm3Kmol-1 expected for ten MnIII ions and two DyIII ions (4f9, 6H15/2, 28.3 cm3Kmol-1). The difference ( MT ) again increases sharply as the temperature is lowered, from 23.8 cm3Kmol-1 at 300 K to 41.1 cm3Kmol-1 at 5 K, again indicating ferromagnetic MnIIIDyIII interactions. Note that MT vs T plots for isolated DyIII (and TbIII) complexes show decreases with decreasing T ,236,246,249-251,270 and thus they by themselves (i.e. point (iii) above) cannot be responsible for the increasing ( MT ). The MT for 8-7 decreases from 51.9 cm3Kmol-1 at 300 K to 40.5 cm3Kmol-1 at 5 K (Figure 8-8, left). The 300 K value is

PAGE 198

198 again slightly less than the spin only value of 58.1 cm3Kmol-1 expected for ten MnIII ions and two HoIII ions (4f10, 5I8, 28.1 cm3Kmol-1). The ( MT ) increases from 23.5 cm3Kmol-1 at 300 K to only 30.0 cm3Kmol-1 at 5 K. Thus, the increase in ( MT ) is much smaller than those for 8-5 and 8-6 and this makes it difficult to concl ude with safety the nature of the MnIIIHoIII interactions. The MT for 8-8 decreases continuously from 49.5 cm3Kmol-1 at 300 K to 21.3 cm3Kmol-1 at 5 K (Figure 6d). The 300 K value is less than the spin only value of 52.9 cm3Kmol-1 expected for ten MnIII ions and two ErIII ions (4f11, 4I15/2, 22.9 cm3Kmol-1). The ( MT ) stays approximately constant at 21.1 cm3Kmol-1 from 300 K to 100 K and then drops to 10.5 cm3Kmol-1 suggesting weak antiferromagnetic MnIIIErIII interactions in this complex. 8.3.3.3 Comparison of 8-9 (Mn10Y2) with 8-1 (Mn10Pr2), 8-2 (Mn10Nd2), 8-3 (Mn10Sm2) For these three complexes with early lanthanides, the MT vs T plots were essentially identical (Figure 8-9). The MT at 300 K is 27.4, 28.4, and 26.8 cm3Kmol-1 for 8-1 8-2 and 8-3 respectively. In each case, MT then decreases with decreas ing temperature to 14.6, 15.2, and 16.0 cm3Kmol-1, respectively, at 5 K. The MT vs T behaviors of 8-1 8-3 are thus essentially parallel to that of 8-9 except at the lowest temperatures. Consequently, ( MT ) is small and negative at higher temperatures, and then slightly positive at lower temperatures. Since we do not know the exact MT vs T behaviors of the LnIII ions in the absence of the MnIII, we cannot draw any safe conclusions from these small magnitudes of ( MT ) about the exact sign of MnIIILnIII. In other work, the interactions of these early lanthanide ions w ith transition metals have been found to be typically antiferromagnetic.248-251,254,271 The combined results suggest that complex 8-5 contains ferromagnetic MnIIITbIII interactions, as is likely also the case in 8-4 and 8-6 but the data are not clear enough to come to safe conclusions for the other complexes. It would require access to the corresponding M10Ln2 complexes with M = CoIII or other diamagnetic ion to probe this point further. It should also be

PAGE 199

199 reiterated that the exchange interactions within the Mn10 unit may be sligh tly altered as the LnIII changes (as a result of small changes to bond di stances and angles), and that this and the introduction of MnIIILnIII interactions will complicate further the spin frustration effects. Thus, small changes to ( MT ) are not considered reliable indicators of the sign of the MnIIILnIII interactions. 8.3.3.4 Out-of-Phase ac Susceptibility Signal s a nd Magnetization Hysteresis Loops As stated in the Introduction, one of the objectives of the present work was to obtain new SMMs. We have thus explored all complexes by ac susceptibility studies. The in-phase signals for 8-4 and 8-9 were shown in Figure 8-3 (right); their out-of-phase ( M ) signals are shown in Figure 8-10. Complex 8-9 shows only the merest hint of a frequency-dependent signal down to 1.8 K, the operating minimum of our SQUID. Complex 8-4 shows a stronger but still very weak signal. Since the upper limit (U ) to the magnetization relaxa tion barrier is given by S2|D| and (S2)|D| for integer and half-integer sp in, respectively, the difference between 8-4 and 8-9 must reflect a bigger barrier as a result of the fo rmers increased ground state spin isotropic GdIII is not expected to significantly increase the anisotropy. Similarly very weak signals were also observed for 8-1 8-3 (Figure 8-11). Thus, these complexe s are at best only poor SMMs with very small relaxation barriers. However, much more encouraging results were observed for 8-5 8-7 where the LnIII ions bring both a large spin and large anisotropy to the molecules. In Figure 8-12 are shown the M'T vs T and M" vs T plots for these complexes, and two points should be noted: (a) in each case, the M'T at 5K is essentially identical to that in the corresponding dc MT vs T plot (Figure 8-7) showing th at the latter was not unduly affected by the dc field used; and (b) in each case, below ~3 K ther e is a frequency-dependent drop in M'T and a strong and frequency-dependent M" peak, approximately one order of magnitude larger than those in 8-4

PAGE 200

200 These data suggest that 8-5 8-7 are SMMs with significant relaxa tion barriers, and thus their magnetization vector cannot relax fast enough to stay in phase w ith the oscillating ac field. The ac M vs T plots can be used as a source of relaxation rate vs T kinetic data for determining the true or effective energy barrier ( Ueff) to magnetization relaxation, because at the M peak maximum the relaxation rate (1/ where is the relaxation time) is equal to the angular frequency (2 ) of the ac field. The obtained data for 8-5 8-7 are shown as Arrhenius plots in Figure 8-13, based on the Arrhenius Law of eq 8-4, where k is the Boltzmann constant and 0 is the pre-exponential factor. The fit of the data for 8-6 to eq 8-4 (solid line in Figure 8-13) gave = 0 exp ( Ueff / kT ) (8-4) Ueff ~ 39 K and 0 ~ 1 x 10-11 s. A similar analysis for 8-7 gave Ueff ~ 41 K and 0 ~ 3 x 10-12 s. Because the ac M vs T data were over a small temperature range (~0.4 K), these values must be considered only approximations. The M peaks for 8-5 are at lower temperatures and thus over too small a range for a meaningful plot, but its Ueff was comparable with those for 8-6 and 8-7 Nevertheless, the 0 values appear smaller than those typically seen for transition metal SMMs (10-7-10-9 s) but are comparable with other 3d-4f ones.247 Because we were worried that the above Ueff and 0 values were not very accurate owing to the small T range employed, we carried out a more accura te analysis for representative complex 8-6 by supplementing the ac data with additional relaxation rate vs T data down to 0.04 K obtained from dc magnetization decay vs time measurements on a single crystal of 8-6 MeCN MeOH. The samples magnetization was first satu rated in one direction at ~ 5 K with a large applied dc field, the T was then decreased to a chosen va lue in the 0.04 1.6 K range, the field removed, and the magnetization decay monitored with time. The obtained magnetization vs time plots are characteristic of a di stribution of relaxati on barriers, and are shown in Figure 8-14

PAGE 201

201 (left). The dc decay and ac M data were combined and used to construct an Arrhenius plot over a wider T range (Figure 8-14, right). Fitting of the data in the thermally activated region gave Ueff = 30 K and 0 = 6 x 10-10 s. Below ~0.5K, the vs 1/T plot deviates from linearity as the thermally activated relaxation rate diminishes and the relaxation via quantum tunneling through the anisotropy barrier be comes more dominant. Eventually at T < 0.2 K, it becomes essentially temperature-independent, as expected for th e relaxation now being exclusively by quantum tunneling through the lowest lying MS ( MJ) levels. Such quantum tunneling was first observed for a 3d-4f complex in a [Mn11Dy4] complex.21 The Ueff value of 30 K for 8-6 is the highest yet reported for a Mn-Ln SMM. Previous examples of [Mn11Dy4],239 [Mn2Dy2],240 [Mn6Dy4],234 and [Mn11Gd2]247 SMMs have been reported to have Ueff values of 9.3, 15, 16 and 18.4 K respectively. It should be noted that the Ueff value of 39 K obtained for 8-6 using just the ac M'' data is very different from the more reliable 30 K obt ained from the combined dc and ac data. These results thus represent an important caveat th at inaccuracies of a large magnitude can be introduced by determining the barrier Ueff from too small a temperature range. Inversely, it should be accepted that when there is indeed no choice but to determine Ueff values over a limited T range, then large inaccuracies are likely. To confirm whether these complexes are trul y SMMs, magnetization vs dc field sweeps were carried out on single crysta ls of representative complex 8-6 3MeCNMeOH.153 Hysteresis loops were observed below 1.6 K (Figure 8-15), whose coercivity incr eases with decreasing temperature and increasing scan rate, as expect ed for the superparama gnet-like properties of a SMM. The loops are dominated by a large step at zero field due to quantum tunneling of magnetization (QTM) through the anisotropy barrier. The large step is indicative of fast QTM

PAGE 202

202 rates, as is typical of low symmetry molecules. Steps at other field positions are barely visible, and such smearing out is typi cal of broadening effects from a distribution of molecular environments, as already concluded to be pr esent in the magnetization decay vs time plots mentioned above. Such distributions are due to solvent and/or lig and disorder, and the presence of low-lying excited states. 8.4 Conclusions The reaction of a preformed Mn4 cluster, hmpH and simple LnIII salts has provided entry into a new family of 3d-4f [Mn10Ln2] clusters for most of the Ln ions. Three representative crystal structures have shown the complexes to be isostructural, including the corresponding [Mn10Ln2] analogue with diamagnetic YIII. Comparisons of the combined dc and ac magnetic susceptibility studies have allowed importa nt insights into the effect of the LnIII ions on the magnetic properties. Complex 8-4 containing isotropic GdIII with S = 7/2 has demonstrated an increase in the spin compared with [Mn10Y2] complex 8-9 as a result of net ferromagnetic interactions with the Mn10 core. This is reflected in the appearance of out-of-phase ac signals indicating the slow relaxation of a SMM. The result demonstrates that as long as the anisotropy of a MnIII x unit is sufficient, the large spin of GdIII can boost the spin S if couplings are ferromagnetic and thus improve the SMM properties. The other LnIII ions also bring significant anisotropy to the table, but the couplings for earl y lanthanide ions Pr, Nd and Sm do not improve the SMM properties above those of 8-9 presumably due to their small spins and possibly antiferromagnetic couplings. The later lanthani de ions Tb, Dy and Ho give much more encouraging results, however, with all three compounds exhib iting strong out-of-phase signals above 1.8 K. This can be attributed to both their significant spin and anisotropy, and the observance of ferromagnetic MnLn coupling between them and the central Mn10 unit; the similar ac behavior between 8-7 and 8-5 / 8-6 suggests the coupling really is ferromagnetic in each

PAGE 203

203 case. The calculated Ueff value of 30 K for representative complex 8-6 is the highest yet for a Mn-Ln SMM. Single-crystal studies using a micro-SQUID on representative complex 8-6 confirm the SMM property by the clear observation of hysteresis loops. The present work thus emphasizes that synthesis of MnIII-LnIII complexes incorporating Tb, Dy and Ho is a somewhat promising appro ach to higher-barrier SMMs. We have reported elsewhere preliminary results on a related family of Mn11Ln4 complexes, and showed there that the Dy analogue again is an SMM exhibiting magnetization hysteresis. However, the barrier Ueff in that case was only 9.3 K, ~30% of that in 8-6 This and related reports of Mn-Ln complexes234,239-241 serve to emphasize that merely incorporating one or more anisotropic LnIII ions into a MnIII-containing cluster, SMM or otherwise, wi ll not automatically increase barriers and thus switch on or improve the SMM prope rty. The nature of the Mn-Ln coupling, the presence of low-lying excited states, the QTM rates, the symmetry of the molecule, and other factors, all impact the observed magnetization re laxation barrier. Nevertheless, the present work does emphasize that it is possible to get barriers from Mn-Ln species that are fairly high, akin to those of two-electron reduced versio ns of the prototypical SMM family, [Mn12O12(O2CR)16(H2O)4]2-, which also show out-of-phase ac peaks in the 2-4 K range.203,272.

PAGE 204

204 Table 8-1. Crystallographic data for 8-4 3MeCNMeOH, 8-6 MeCNMeOH and 8-9 MeCN. 8-4 8-6 8-9 Formulaa C113H99Mn10Gd2N13O47C113H99Mn10Dy2N13O47C114H98Mn10Y2N14O46 FW, g/mola 3245.94 3265.46 3127.28 Space group P P P a, 14.7083(7) 14.7358(11) 14.737(3) b, 15.2173(7) 15.2150(12) 15.080(3) c 16.7604(8) 16.6441(13) 16.569(3) 67.414(2) 67.6290(10) 67.175(4) 65.549(2) 65.6580(10) 65.668(4) 87.627(2) 87.6360(10) 87.374(4) V 3 3122.0(3) 3114.3(4) 3063.9(10) Z 1 1 1 T K 173(2) 173(2) 173(2) b 0.71073 0.71073 0.71073 calc, g/cm3 1.726 1.741 1.712 mm-1 2.112 2.252 2.024 R1 c,d 0.0643 0.0508 0.0710 wR2 e 0.1713 0.1330 0.1614 a Including solvate molecules. b Graphite monochromator. c I > 2 (I). d R1 = (|| Fo| |Fc||) / | Fo|. e wR 2 = [ [ w(Fo 2 Fc 2)2] / [ w(Fo 2)2]]1/2, w = 1/[ 2(Fo 2) + [( ap)2 + bp], where p = [max ( Fo 2, O) + 2 Fc 2]/3. Table 8-2. Bond-valence sums for the Mn atoms of complex 8-4 8-6 and 8-9 a 8-4 8-6 8-9 MnII MnIII MnIV MnII MnIII MnIV MnII MnIII MnIV Mn1 3.09 2.83 2.97 3.05 2.79 2.93 3.13 2.87 3.01 Mn2 3.33 3.08 3.18 3.32 3.07 3.17 3.33 3.08 3.18 Mn3 3.31 3.06 3.16 3.31 3.06 3.16 3.34 3.09 3.18 Mn4 3.15 2.88 3.03 3.16 2.89 3.03 3.22 2.95 3.09 Mn5 3.06 2.80 2.94 3.07 2.81 2.94 3.09 2.83 2.97 a The underlined value is the one closest to the charge for which it was calculated. The oxidation state of a particular atom can be taken as the nearest whole number to the underlined value

PAGE 205

205 Figure 8-1. The structures of 8-4 (top), 8-6 (middle), and 8-9 (bottom). Hydrogen atoms and phenyl rings (except for the ipso carbon atoms) have been omitted for clarity. Color code: Gd, cyan, Dy, yellow; Y, pink; M n, green; O, red; N, blue; C, grey.

PAGE 206

206 Figure 8-2. (top) Centro symmetric core of 8-6 emphasizing the ABCBA layer structure. (bottom) The B (left) and C (right) layers showing the Jahn-T eller elongation axes as thicker black bonds. Color code: Dy, yellow; Mn, green; O, red; N, blue; C, grey. Figure 8-3. (left) Plot of MT vs T for complexes 8-4 and 8-9 at 0.1 T. The difference, ( MT ), is shown as the dashed line. (right) Pl ot of in-phase ac susceptibility ( M' ) as M'T vs T for complexes 8-4 and 8-9

PAGE 207

207 Figure 8-4. Plot of reduced magnetization ( M/NB) vs H/T for complexes (left) 8-9 and (right) 8-4 See the text for the fit parameters. Figure 8-5. Two-dimensional contour pl ot of the error surface for the D vs g fit for (left) 8-9 and (right) 8-4 The asterisk indicates the point of minimum error (best fit).

PAGE 208

208 Figure 8-6. Plots of dc MT vs T for (left) Gd(NO3)3.6H2O and (right) Dy(NO3)3.5H2O. Figure 8-7. Plots of MT vs T for 8-9 and (left) 8-5 and (right) 8-6 In each case, the difference, ( MT ), is shown as a dashed line. Figure 8-8. Plots of MT vs T for 8-9 and (left) 8-7 and (right) 8-8 In each case, the difference, ( MT ), is shown as a dashed line.

PAGE 209

209 Figure 8-9. Plots of MT vs T for 8-1 8-2, 8-3 and 8-9 The dashed line is the difference, ( MT ), between 8-3 and 8-9 ; the others are similar. Figure 8-10. Plots of out-of-phase M" vs T ac susceptibility data for (left) 8-9 (Mn10Y2), and (right) 8-4 (Mn10Gd2). Figure 8-11. Plots of out-of-phase M" vs T ac susceptibility data for (left) 8-1 (Mn10Pr2), and (right) 8-2 (Mn10Nd2).

PAGE 210

210 Figure 8-12. Plots of in-phase, M'T vs T and out-of-phase M" vs T ac susceptibility data for (top) 8-5 (Mn10Tb2), (middle) 8-6 (Mn10Dy2), and (bottom) 8-7 (Mn10Ho2). Figure 8-13. Plot of relaxation rate vs reciprocal temperature for 8-5 8-7

PAGE 211

211 Figure 8-14. (left) Magnetization (M ) vs. time decay plots in zero dc field for 8-6 MeCN MeOH. The magnetization is normali zed to its saturation value, Ms (right) Plot of relaxation time vs 1/ T for 8-6 using combined ac M" and magnetization decay data. Figure 8-15. Single-crystal magnetization ( M ) vs dc field ( H ) hysteresis loops fo r a single crystal of 8-6 MeCNMeOH at different scan rates (left) and temperatures (right).

PAGE 212

212 CHAPTER 9 A FOURTH ISOLATED OXIDATION LEVEL OF THE [Mn12O12(O2CR)16(H2O)4] FAMILY OF SINGLE MOLECULE MAGNETS 9.1 Introduction Single-m olecule magnets (SMMs) are molecule s that possess a significant barrier (vs kT ) to reorientation of their magnetization (magnetic moment) vector as a result of the combination of a large ground state spin ( S) and Ising (easy-axis) magnetoan isotropy (negative axial zerofield splitting parameter ( D ).13 As such, they represent a mol ecular (bottom-up) approach to nanomagnetism. The first SMM was [Mn12O12(O2CMe)16(H2O)4]HO2CMe4H2O15,273,274 (Mn12-Ac; 4MnIV,8MnIII) and many more have since been synthesized.21,22,231,275 Although complexes displaying SMM behavior are known for a variety of 3d, 4d, 4f and mixed-metal complexes,9,20,58,79,86,118,162,163,173,190,271,276-282 manganese carboxylate clusters have proven to be the most fruitful source of SMMs.21,22,32 Using only a limited palette of ligands and starting materials, a wide range of Mn SMMs has been obta ined with their nuclearities ranging from 2 to 84.164,172,283 Amongst the known Mn SMMs, the Mn12 family continues to be attractive for study as a result of its ease of preparation, stability, ready modification in a variety of ways, high ground state spin ( S = 10) and anisotropy, and the access to derivatives that crystallize in high symmetry (tetragonal) space groups.14,284,285 The various modifications of the Mn12 family of SMMs that have been accomplished to date have proven extremely useful for a myriad of reasons and studies, an d have permitted great advances in our knowledg e and understanding of Mn12 complexes and the SMM phenomenon in general. In this regard, carboxylate substitution273,286-288 represented a big step forward because it provided an extremely useful and convenient means of accessing other carboxylate analogues, which provided benefits such as isotopic labe ling, tunability of redox properties and increased solubility in a variety of organi c solvents. One of the most informative impacts of the latter two

PAGE 213

213 points was the observation of multiple, reversible redox processes and the subsequent generation and isolation of one-electron reduced complexes i.e. salts of the [Mn12O12(O2CR)16(H2O)4]anion, abbreviated [Mn12]-. The crystal structures of such sa lts revealed minimal change to the structure on reduction, with the added elec tron localized on an outer, formerly MnIII atom giving a trapped-valence MnIV 4MnIII 7MnII situation.202 The [Mn12]salts allowed an assessment of the structural, magnetic and spectrosc opic consequences of changing the electron count, as well as allowing the study of the differences in quantum properties due to the integer vs half-integer S value, since [Mn12]salts have an S = 9 ground state.202,289-291 The subsequent introduction of carboxylates with more electron-with drawing substituents into the Mn12 complexes made twoelectron reduction easier and led to the succe ssful generation and isol ation of two-electron reduced [Mn12O12(O2CR)16(H2O)4]2complexes, [Mn12]2-, such as salts of [Mn12O12(O2CCHCl2)16(H2O)4].203 The [Mn12]2anion was again found to be trapped-valence, with a MnIV 4MnIII 6MnII 2 oxidation state description, and the spin was found to be S = 10, the same as the Mn12 parent compound. The above efforts had thus provided the Mn12 family of SMMs in three oxidation states, providing a wealth of comparativ e chemical and physical data. So much so that it was clearly desirable to extend this family to a fourth oxida tion level if at all possi ble. The three-electron reduction of Mn12 complexes is in fact observa ble in the cyclic voltammetry,203 and so we decided to pursue the generation an d isolation of this oxi dation state. Indeed, this effort has been successful, and we herein report the synthesis and characterization of the [Mn12]3salts (NPrn 4)3[Mn12O12(O2CCHCl2)16(H2O)4] and (NMe4)3[Mn12O12(O2CCHCl2)16(H2O)4].207

PAGE 214

214 9.2 Experimental Section 9.2.1 Syntheses All m anipulations were performed under aerobic conditions using ma terials as received, except if otherwise noted. [Mn12O12(O2CCHCl2)16(H2O] ( 9-1 ) was prepared as described elsewhere.203 (NPrn 4)[Mn12O12(O2CCHCl2)16(H2O)4] (9-2) Solid NPrn 4I (0.03 g, 0.1 mmol) was added to a stirred dark brown solution of complex 9-1 (0.30 g, 0.10 mmol) in MeCN (15 mL). The resulting solution was stirred for 4 hours with no noticeabl e color change. After 4 hours, hexanes (20 mL) were added causing the formation of two phases, a nd the mixture shaken to facilitate extraction of I2 into the hexanes phase. The hexanes layer was then removed, and the extraction process repeated a few more times until the hexanes layer was colorless. The two layers were then separated and the MeCN solution evaporated to dryness. The resi due was dissolved in MeCN (10 mL), and Et2O/hexanes (1:1 v/v, 20 mL) added. The re sulting microcrystalline product was isolated and dried in vacuo. Yiel d, 70 %. Anal. Calcd (Found) for 9-2 MeCN (C46H55N2Mn12O48Cl32): C, 17.28 (17.60); H, 1.73 (1.62); N, 0.87 (0.53) %. (NPrn 4)2[Mn12O12(O2CCHCl2)16(H2O)4] (9-3). Complex 9-3 was synthesized following the same procedure as for complex 9-2 except that two equivalents of NPrn 4I (0.06 g, 0.2 mmol) were employed and the reaction mixture was s tirred for 10 hours. Yield, 65 %. Anal. Calcd (Found) for 9-3 MeCN (C58H83N3Mn12O48Cl32): C, 20.59(20.82); H, 2.47(2.26); N, 1.24(0.88) %. (NPrn 4)3[Mn12O12(O2CCHCl2)16(H2O)4] (9-4) Complex 9-4 was synthesized following the same procedure as for complex 9-2 except that three equivalents of NPrn 4I (0.09 g, 0.3 mmol) were employed. The reaction mixture was stirred for 40 hours and the product was not recrystallized. Yield, 85 %. Anal. Calcd (Found) for 9-4 (C68H108N3Mn12O48Cl32): C, 23.14 (22.85); H, 3.08 (2.78); N, 1.19 (1.16) %.

PAGE 215

215 (NMe4)3[Mn12O12(O2CCHCl2)16(H2O)4] (9-5) Complex 9-5 was synthesized following the same procedure as for complex 9-4 except that three equivalents of NMe4I (0.06 g, 0.3 mmol) were employed and the reaction mixture was s tirred for 48 hours. Yield, 80 %. Anal. Calcd (Found) for 9-5 MeCN (C46H63N4Mn12O48Cl32): C, 17.08(16.92); H, 1.96(1.90); N, 1.73(1.78) %. 9.3 Results and Discussion 9.3.1 Syntheses Electrochemical studies on various [ Mn12O12(O2CR)16(H2O] complexes have revealed a rich redox chemistry involving several quasi-reversible oxidation and reduction processes.202,203,292 In addition, the redox pot entials are, as expecte d, very sensitive to the electron-withdrawing and donating ability of the carboxylate ligand. For example, E1/2 (vs ferrrocene) for the first reduction varies by almost a volt from 0.91 V for the R = CHCl2 complex to 0.00 V for the R = p-C6H4OMe complex. The particularly high electron-withdrawing capability of the R = CHCl2 group, as reflected in the very low pKa of 1.48 for CHCl2CO2H, brought the second reduction potential to 0.61 V (Figure 9-1), well within the reducing capability of our preferred reducing agent, iodide (0.14 V vs ferrocene in MeCN),293 and this led to the subsequent successful genera tion and isolation of (PPh4)2[Mn12O12(O2CCHCl2)16(H2O)4] reported elsewhere.203 Similarly for the R = C6F5 substituent, which has also been used for the synthesis of the two-elec tron reduced complex (NMe4)2[Mn12O12(O2CC6F5)16(H2O)4]272 from the reaction of [Mn12O12(O2CC6F5)16(H2O)4] with two equivalents of I-. For the present work, we chose to employ the R = CHCl2 carboxylate complex because it has a particularly well resolved third one-elect ron reduction in the cyclic voltammogram (CV) and differential pulse voltammogram (DPV) (Figure 91), and one that is s till within the reducing capability of I-. A fourth, clearly irreversib le reduction at ~0.1 V re presented a potential problem, so we avoided the use of an excess of reducing agent beyond the stoichiometric three

PAGE 216

216 equivalents. Thus, complex 9-1 was treated with three equivalents of NMe4I, NPrn 4I, NBun 4I and PPh4I in MeCN for different lengths of time; the formation of I2 was confirmed by its extraction into a hexane phase. It was found that longer reaction time of 40 h were required to give complete conversion of [Mn12] to [Mn12]3, as established by subseque nt characterization of the product; these are much longer times than routinely employed for the [Mn12]and [Mn12]2complexes.203,272 Samples of [Mn12]3 salts that were analytically pure (and subsequently shown by magnetism studies to be pure [Mn12]3-) were obtained with the NMe4 + and NPrn 4 + cations, but we were not satisfied with the purity of the NBun 4 + and PPh4 + salts or their prolonged stability in the solid state (vide infra). Thus, we used the NMe4 + and NPrn 4 + salts for the detailed studies below. In addition, for better comparisons of [Mn12]z(z = 0 3) complexes with the same cation, we also prepared the NPrn 4 + salts of the [Mn12]and [Mn12]2complexes. The transformations of 9-1 into 9-2 to 9-4 are summarized by general eq. 9-1, where z = 1, 2 or 3. [Mn12O12(O2CR)16(H2O)4] + z I[Mn12O12(O2CR)16(H2O)4]z+ z/2 I2 (9-1) It soon became apparent that the [Mn12]3anion is far less stable in solution than [Mn12]and [Mn12]2-. Numerous attempts to grow crystals of a [Mn12]3salt with a variety of cations and under various crystallization conditions were al l unsuccessful, giving amorphous powders and/or crystals that turned out to be the [Mn12]2salt on analysis and magnetic examination. However, in reality a crystal structure would not have told us anything that we did not feel we already knew about the [Mn12]3anion from previous observations of what happens to the structure of a Mn12 complex on oneand two-electron reduction. The mo st important structural question in these other complexes had been where does(do) the ad ded electron(s) go, and the answer was on the outer ring of MnIII atoms. This is summarized in Figure 9-2, which shows the distribution of Mn oxidation states within the [Mn12O12] cores of these compounds. The neutral Mn12 (Figure 9-2,

PAGE 217

217 top) has four central MnIV atoms within a non-planar ring of eight outer MnIII atoms. The latter divide by symmetry into two cla sses, and addition of one or two extra electrons leads to localization of these electrons onto MnIII atoms of only one class lead ing to their conversion to MnII, giving MnIV 4MnIII 7MnII and MnIV 4MnIII 6MnII 2 oxidation state descri ptions, respectively. This was established from the crystal structures of multiple [Mn12]and [Mn12]2complexes,202,203,272,291 and is shown in the two central figures of Figure 9-2. This counterintuitive preferential reduction of a MnIII rather than a MnIV was rationalized on the basis that reduction of a central MnIV would convert it into a MnIII atom that would show a characteristic Jahn-Teller distortion, as e xpected for a high-spin d4 configuration (and exhibited by the outer MnIII atoms). This would introduce strain into the relatively rigid central Mn4O4 cubane, and so reduction of an outer MnIII becomes thermodynamically preferre d since it causes no significant structural perturbation. We ar e thus certain that the thir d added electron in the [Mn12]3complexes also has added to a formerly MnIII atom of the same symmetry class, giving the MnIV 4MnIII 5MnII 3 situation depicted in Figure 9-2, bottom. On the basis of the above arguments, the decreased stability of the [Mn12]3anion in solution compared with the [Mn12]z(z = 0 2) is perhaps not surprising given the now high content of MnII in a complex that still contains four MnIV atoms. It is reasonable that such a species would be unstable to structural degrada tion initiated perhaps by the lability of the MnII centers and/or intramolecular redox transitions. This would also rationalize that observation in Figure 9-1 that the four-ele ctron reduced species [Mn12]4-, which would be expected to be MnIV 4MnIII 4MnII 4, rapidly degrades even on the electroch emical timescale and thus does not show a well-formed peak in the DPV or even a re versible CV wave on the faster CV timescale.

PAGE 218

218 9.3.2 Magnetochemistry 9.3.2.1 Dc Studies Solid-s tate, variable-temperature dc magnetic susceptibility data in a 0.1 T field and in the 5.0-300 K range were collected on powdered crystalline samples of complexes 9-2 to 9-5 restrained in eicosane to prevent torquing. The obtain ed data are plotted as MT vs T in Figure 93. The MT values for 9-2 9-3 9-4 and 9-5 slowly increase from 22.7, 21.8, 21.5 and 22.3 cm3 K mol-1 at 300 K to a maximum of 46.2, 50.1, 38.8, and 38.1 cm3 K mol-1 at 10 K, respectively, and then decrease at lower temperatures due to Z eeman effects from the applied field, any weak intermolecular interactions, etc. The MT vs T profiles of 9-2 and 9-3 are essentially identical to those of previously reported [Mn12]and [Mn12]2complexes.203,272,291 Their maxima of 46.2 and 50.1 cm3 K mol-1 at 10 K are in agreement with S = 19/2 and S = 10 ground states, respectively, and g < 2 as expected for Mn. This is in agreemen t with the ground states found in previous work for [Mn12]and [Mn12]2complexes.203,272,291 The calculated, spin-only (g = 2) values are 49.9 and 55.0 cm3 K mol-1 for S = 19/2 and S = 10, respectively. The MT vs T profiles of 9-4 and 9-5 are essentially superimposable with each other throughout the whole temperatur e range, and with those of 9-2 and 9-3 in the 100-300 K range. Below 100 K, they diverge from those of the la tter, and reach maxima significantly below those of 9-2 and 9-3 This shows that 9-2 and 9-5 are indeed at a different oxidation level from either 9-2 or 9-3 and also that they ha ve a smaller ground state S value than them. Remembering that a MnIV 4MnIII 5MnII 3 complex must have a half-i nteger ground state, then the MT maxima at 10 K of 38.8 and 38.1 cm3 K mol-1 suggest that 9-4 and 9-5 have an S = 17/2 ground state with g < 2; the spin-only ( g = 2) value is 40.4 cm3 K mol-1. Confirmation of the above preliminary conclu sion was sought from fits of magnetization ( M ) data collected on complexes 9-4 and 9-5 in the 0.1 4 T and 1.8 10 K ranges. The obtained

PAGE 219

219 data are shown as reduced magnetization ( M/NB) vs H/T plots in Figure 9-4, where N is Avogadros number and B is the Bohr magneton. The data were fit using the program MAGNET,53 described elsewhere.56 The best fits for 9-4 and 9-5 are shown as the solid lines in Figure 9-4, and the fit parameters were S = 17/2, D = -0.25 cm-1, g = 1.91 for 9-4 and S = 17/2, D = -0.23 cm-1, g = 1.90 for 9-5 Fits of the data with S = 15/2 or 19/2 gave unreasonable g values of 2.21 and 1.72, respectively, and were th erefore discounted. For a comparison of data for complexes with different degrees of reduction but with the same cation, we also collected variable-temperature and -field magnetization data for complexes 9-2 and 9-3 ; the corresponding ( M/NB) vs H/T plots and fits are provided in Figure 9-5. The fit parameters were S = 19/2, D = 0.35 cm-1, g = 1.95 for 9-2 and S = 10, D = -0.28 cm-1, g = 1.98 for 9-3 The obtained ground state S values of 9-2 and 9-3 are the same as those previously found for several other [Mn12]and [Mn12]2complexes.203,272,291 To confirm that the obtained fit minima were the true global minima and to assess the hardness of the fit, a root-mean square D vs g error surface for the fit was generated for representative complex 9-4 using the program GRID ,71 which calculates the relative difference between the experimental ( M/NB) data and those calculated fo r various combinations of D and g. This is shown as a 2-D contour plot in Figure 9-6 covering the D = -0.10 to -0.50 cm-1 and g = 1.86 to 1.98 ranges. Only one minimum was observed, and this was a relatively soft minimum; we thus estimate the fitting uncertainties as D = -0.25 0.01 cm-1 and g = 1.91 0.01. 9.3.2.2 Comparison of the Magnetic Properties of the [Mn12]z(z = 0 3) Family The combined results for complexes 9-2 to 9-5 as well as those for neutral complex 9-1 ,203 are collected in Table 9-1. Co nsidering first the S values, it is well known that the spin ground state changes very little on oneand two-electron reduction, from S = 10 to S = 19/2 and then back to S = 10 along the series 9-1 (z = 0), 9-2 (z = 1) and 9-3 (z = 2), respectively. Thus,

PAGE 220

220 the Mn12 core acts almost as a spin buffer, picking up electrons with little change to the ground state S value. However, on three-electron reduction to complexes 9-4 and 9-5 there is a more significant change to S = 17/2. This is no doubt due to the increased MnII content and the general weakening of many of the exchange intera ctions in the core. However, the [Mn12O12] core is a complicated one with many symmetry inequivalent exchange interactions, many of them competing and it is thus not easy to provide a rationalization of the S = 17/2 ground state, as indeed it has not been possible in the past to rationalize those of the [Mn12]and [Mn12]2complexes either. The g values given in Table 9-1 are provided only for completeness and should not be taken as particularly accurate. It is well known that fits of bul k magnetization data are not a good way to obtain accurate g values. While we prefer to quote the actual values obtained by having the g value as a free parameter, rather than fixing it at a more realistic value at or near 2.0, we do not attempt to draw any conclusions from resulting differences in g. It would require studies with a more sensitive technique such as EPR spectroscopy to provide more accurate g values. In contrast to the S value, the axial zero-field splitting parameter D does exhibit a monotonic change with the extent of reduc tion; there is a cl ear decrease in | D | with progressive one-electron reduction. This is exactly as expected because the molecular anisotropy, as gauged by the magnitude of | D |, is the projection of the single-i on Mn anisotropies onto the molecular anisotropy axis. MnIV and MnII are relatively isotropic ions, an d the primary contributions to the molecular D are thus the MnIII ions, which are significan tly Jahn-Teller distorted. Since reduction involves addition of electrons onto formerly MnIII centers converting them to MnII, the greater the extent of reduction, the fewer will be the remaining MnIII ions, and the smaller will thus be the molecular anisotropy | D |. This assumes other factors remain the same, such as the

PAGE 221

221 overall structure of the Mn12 complex, and the relative orientation of the MnIII JT axes essentially parallel to the molecular z axis. It should be added that D values in Table 9-1 have been obtained by fitting the magnetization data with the rhombic (transverse) zero-field splitting parameter ( E ) fixed at E = 0. In fact, these complexes do not have axial symmetry, and E is unlikely to be exactly zero. In our experience, however, magnetization fi ts are usually not very sensitive to E and the D values in Table 9-1 are therefore expected to be reasonable, especially for assessment of relative magnitudes within a series, as here. Nevertheless, for information purposes, we provide in Figure 9-6, the fit of the magnetization data of 9-4 as a function of D and E with g held constant at 2.0; the fit is shown as a contour plot of the error surface. The best-fit parameters are D = -0.24 cm-1 and |E | = 0.065 cm-1. The D value changes very slightly (from 0.25 cm-1 obtained with E = 0), while the non-zero E is consistent with the low symmetry of a threeelectron reduced [Mn12]3complex. The final entries in Table 9-1 for each compound are the values of the U the anisotropy barrier to magnetization relaxa tion, whose upper limit is given by S2| D |. In practice, the true or effective barrier ( Ueff) is smaller than this upper limit because the magnetization vector need not go over the top of the barrie r but can tunnel th rough its upper regions via higher-lying MS levels. This quantum tunneling of magnetization (QTM) is a characteristic of all SMMs. Since | D | monotonically decreases w ith reduction, whereas the S stays roughly the same or decreases, then it would be expected that U would decrease with reduction, and th is is what is indeed seen. The U for [Mn12]3complexes 9-4 and 9-5 coming from their S = 17/2 spin and a | D | value that has been decreased but is still reasona ble, is still relatively large, and even decreased by QTM might still be sufficient for them to function as SMMs. In order to expl ore whether these [Mn12]3-

PAGE 222

222 complexes might indeed exhibit slow relaxation; we investigated thei r magnetization dynamics using ac susceptibility. 9.3.2.3 Ac Studies In ac stud ies, a weak field (1 5 G) oscillat ing at a particular fr equency, typically up to 1500 Hz, is applied to a sample to probe the dynamics of the magnetizat ion relaxation. If the magnetization vector can relax fast enough to keep up with the oscill ating field, then there is no imaginary (out-of-phase) susceptibility signal ( M), and the real (in-phase) susceptibility ( M ) is equal to the dc susceptibility. However, if th e barrier to magnetization relaxation is significant compared to thermal energy (kT ), then there is a non-zero M signal and the in-phase signal decreases. In addition, the M signal will be frequency-dependent. The ac susceptibilities of [Mn12]z(z = 0 2) complexes 9-1 to 9-3 have been previously reported, but they were remeasured here for better comparison with those of 9-4 and 9-5 under identical conditions. The ac susceptibilities for complexes 9-1 to 9-5 were collected on microcrystalline samples in a 3.5 G ac field, and the obtained data for complexes 9-4 and 9-5 at representative frequencies of 50, 250 and 1000 Hz are shown in Fi gures 9-7 and 9-8, respectively, as M'T vs T and M" vs T plots. The in-phase (M'T ) ac signal is invaluable as an additional and independent means to determine the ground state spin of a molecule, without any complications from a dc field.62,69,224 Inspection of Figures 9-7 and 9-8 shows that the M'T values are essentially temperature independent down to ~ 5 K, below which they show decreases due to slow relaxation (vide infra). The temperature independent M'T shows that only the spin gr ound state of the molecule is populated at these temperatures, a nd can be used to calculate its S value. The M'T values of 40.5 cm3 K mol-1 and 39.5 cm3 K mol-1 for 9-4 and 9-5 respectively, correspond to S = 17/2, g = 2.00, and S = 17/2, g = 1.98, in very satisfying agreement with the conclusions from the fits of the dc magnetization data discussed above. Note that for S = 15/2 or 19/2 states, a M'T value of ~40

PAGE 223

223 cm3 K mol-1 would require g values of 2.24 and 1.79, which are unreasonable for Mn. We conclude that [Mn12]3complexes 9-4 and 9-5 are confirmed to possess S = 17/2 ground states. Below ~5K, the in-phase M'T signals for 9-4 and 9-5 in Figures 9-7 and 9-8 exhibit a frequency-dependent decrease c oncomitant with the appearance of frequency-dependent out-ofphase ( M") signals. This is indicative of th e onset of slow magnetiza tion relaxation relative to the ac field, i.e. the magnetization vector can no longer relax fast enough to stay in-phase with the oscillating field. This is the characteristic superparamagne t-like behavior of a SMM, and parallels that previously observed for the other oxidation levels of the [Mn12]z(z = 0 2) family. On the basis of the comparative data pres ented in Table 9-1, the appearance of the M" signals at very low temperatures of ~2.5 K and below ar e as expected for a barrier to magnetization relaxation in [Mn12]3complexes 9-4 and 9-5 being smaller than those in [Mn12]z(z = 0 2) complexes. This is emphasized by the comparativ e ac data presented in Figure 9-9, which shows the M" signals for complexes 9-1 to 9-4 at equivalent frequencies of 50, 250 and 1000 Hz. In each case, the M" signals are frequency-dependent a nd exhibit a monotoni c shift to lower temperatures with increasing reduction: 6 8 K for 9-1 [Mn12]; 4 6 K for 9-2 [Mn12]-; 2 4 K for 9-3 [Mn12]2-; and 2.5 K for 9-4 [Mn12]3-. For better comparisons at identical ac frequencies, the M" signals for 9-1 to 9-4 at 50 and 1000 Hz are plotted t ogether in Figure 9-10, bottom and top, respectively. A clear shift to lower temper ature is seen with each reduction step. The combined data in Figures 9-9 and 9-10 are thus perfectly consistent with the conclusions from the data in Table 9-1 and the discussion above, since the barrier to magnetization relaxation scales with S2| D |, and either one or both of these quantities decrease with each one-electron increase in the extent of reduction.

PAGE 224

224 Note that it is not expected that that there should be a linear decrease in barrier with reduction, since there are so many factors that de termine the actual magnitude of the true or effective barrier, Ueff, including S, D the rhombic ZFS parameter ( E ), fourth order spin Hamiltonian parameters, precise QTM ra te and tunneling channel (i.e., which Ms levels are involved), spin-phonon coupling stre ngths, and others. Thus, there are too many parameters that contribute to the observed Ueff to permit a more quantitativ e comparison between different oxidation levels. Since we have not been able to obtain singlecrystals, micro-SQUID hysteresis measurements could not be perf ormed. Note also that the M" signals for complex 9-1 in Figures 9-9 and 9-10 also exhibit a weaker signal at lower temperatures, wh ich is due to a faster-relaxing form arising from a different Jahn-Teller is omer, i.e. a form in which one of the MnIII Jahn-Teller isomers is abnormally oriented towards a bridging oxide ion in the molecule.205,294,295 These isomeric forms are known to possess smaller barr iers to magnetization relaxation and thus to exhibit their M" signals at lower temperatures. 9.4 Conclusions The Mn12 family of SMMs has been successfully exte nded to four isolated oxidation states by the three-electron reduction of [Mn12O12(O2CCHCl2)16(H2O)4] to (NR4)3[Mn12O12(O2CCHCl2)16(H2O)4] (R = Me, Prn) with NR4I. The [Mn12]3complexes are unstable in solution, which has prevented us from obtaining crystals suitable for X-ray crystallography, but this is not unduly disappoint ing, because it is clear on the basis of the structural characterization of the three other Mn12 oxidation states that the third electron will have added to an outer, formerly MnIII ion giving a MnIV 4MnIII 5MnII 3 trapped-valence situation. We do not believe it will be possible to extend the Mn12 family of SMMs to five members by four-electron reduction, given the instabi lity demonstrated by the putative [Mn12]4species in the electrochemical studies.

PAGE 225

225 The [Mn12]3complexes 9-4 and 9-5 both possess a half-integer S = 17/2 ground state, and a | D | value smaller than that for the [Mn12]3complex 9-4 which supports the above assertion that the third added electron is localized on a formerly MnIII ion, since the Jahn-Teller distorted MnIII ions are the primary source of the molecular anisotropy. As a result of the decreased S and D relative to the other Mn12 oxidation states, th e barrier to magnetization relaxation U is also smaller than for the other oxidation states, but is still sufficient to yield out-of-phase ( M") ac susceptibility signals indicative of slow magnetization relaxation. Thus, we conclude that the [Mn12]3complexes 9-4 and 9-5 are SMMs. Note that the observation of M" signals is indicative of a SMM but not normally sufficient proof of on e. In this case, howev er, the well-established fact that the M" ac signals for the other Mn12 oxidation states are correctly identifying SMMs, as proven by single-crystal hysteresis studies, leaves little doubt that these same signals for the [Mn12]3complexes 9-4 and 9-5 are also due to SMMs. Thus, although we do not have single crystals with which to carry out micro-SQUI D studies down to 0.04 K in order to observe magnetization hysteresis loops for 9-4 and 9-5 there seems little doubt that the available data are indicating that the Mn12 family of SMMs now spans four oxidation levels.

PAGE 226

226 Table 9-1. Magnetism Data for [Mn12]z(z = 0 3) Complexes 9-1 to 9-5 z = 0 ( 9-1 ) z = 1( 9-2 ) z = 2( 9-3 ) z = 3( 9-4 ) z = 3( 9-5 ) S 10 19/2 10 17/2 17/2 g 1.86 1.95 1.98 1.91 1.90 D /cm-1 -0.45 -0.35 -0.28 -0.25 -0.23 D /K -0.65 -0.50 -0.40 -0.35 -0.33 U /Ka 65 45 40 25 24 a Calculated as S2| D | for 9-1 and 9-3 and as ( S2-)| D | for 9-2 9-4 and 9-5 Potential (V) -0.4 0.0 0.4 0.8 1.2 1.6 50 A 0.86 0.56 0.24 0.03 0.95 0.64 0.34 0.91 0.61 0.29 10 A Current Potential (V) -0.4 0.0 0.4 0.8 1.2 1.6 50 A 0.86 0.56 0.24 0.03 0.95 0.64 0.34 0.91 0.61 0.29 10 A Current Figure 9-1. Cyclic voltammogram at 100 mV/s (top) and differential pulse voltammogram at 20 mV/s (bottom) for complex 9-1 in MeCN containing 0.1 M NBun 4PF6 as supporting electrolyte. The indicated potentials are vs ferrocene.

PAGE 227

227 Figure 9-2. Proposed structural core of 9-1 9-2 9-3 and 9-4 Color code: MnIV, green; MnIII, blue; MnII, yellow, O, red.

PAGE 228

228 Figure 9-3. Plot of MT vs T for 9-2 to 9-5 Figure 9-4. Plot of reduced magnetization ( M / N B) vs H / T for 9-4 (left) and 9-5 (right). The solid lines are the fit of the data; see the text for the fit parameters.

PAGE 229

229 Figure 9-5. Plot of reduced magnetization ( M / N B) vs H / T for 9-2 (left) and 9-3 (right). Figure 9-6. (left)Two-dimensional contour plot of the error surface for the D vs g fit for complex 9-4 (right) Two-dimensional contour plot of the root-mean-square error surface for the D vs E fit for complex 9-4

PAGE 230

230 Figure 9-7. Plot of the in-phase ( M T ) and out-of-phase ( M ) ac susceptibility data for 9-4 Figure 9-8. Plot of the in-phase ( M T ) and out-of-phase ( M ) ac susceptibility data for 9-5

PAGE 231

231 Figure 9-9. M" vs T plots for vacuum-dried complexes [Mn12]z(z = 0, 1, 2, 3) at the indicated frequencies.

PAGE 232

232 Figure 9-10. Comparison of the M" vs T plots for vacuum-dried complexes [Mn12]z(z = 0, 1, 2, 3) at 1000 Hz (top) and 50 Hz (bottom).

PAGE 233

233 APPENDIX A BOND DISTANCES AND ANGLES Table A-1. Selected interatom ic dist ances () and angles () for [Mn4O2(O2CMe)5(salpro)] MeCN ( 2-1 MeCN ) Mn1-O3 1.8859(19) Mn1-O2 1.8906(18) Mn1-O4 1.946(2) Mn1-O6 1.956(2) Mn1-O5 2.0768(19) Mn1-Mn3 2.7704(6) Mn1-Mn2 3.1713(6) Mn2-O8 1.864(2) Mn2-O2 1.9090(18) Mn2-N1 1.972(2) Mn2-O1 1.9905(18) Mn2-O7 2.206(2) Mn2-O9 2.238(2) O3-Mn1-O2 81.26(8) O3-Mn1-O4 168.79(9) O2-Mn1-O4 93.60(8) O2-Mn3-O3 80.79(8) O2-Mn3-O12 170.65(8) O3-Mn3-O12 91.79(8) O2-Mn3-O10 94.91(8) O3-Mn3-O10 165.26(8) O12-Mn3-O10 90.87(8) O14-Mn4-O3 92.81(8) O14-Mn4-N2 90.08(9) O3-Mn4-N2 174.03 (9) Mn2-Mn3 3.1763(6) Mn3-O2 1.8884(19) Mn3-O3 1.9062(18) Mn3-O12 1.9452(19) Mn3-O10 1.950(2) Mn3-O11 2.085(2) Mn3-Mn4 3.1803(6) Mn4-O14 1.8768(19) Mn4-O3 1.9145(18) Mn4-N2 1.972(2) Mn4-O1 1.9776(18) Mn4-O15 2.222(2) Mn4-O13 2.271(2) O3-Mn4-O1 94.64(8) N2-Mn4-O1 82.55(9) O3-Mn4-O15 86.21(8) N2-Mn4-O15 98.94(9) O1-Mn4-O15 88.57(8) Mn4-O1-Mn2 133.88(9) Mn3-O2-Mn1 94.30(8) Mn3-O2-Mn2 113.53(9) Mn1-O2-Mn2 113.16(9) Mn1-O3-Mn3 93.87(8) Mn1-O3-Mn4 114.47(9) Mn3-O3-Mn4 112.69(9)

PAGE 234

234 Table A-2. Selected interatomic dist ances () and angles () for [Mn4O2(O2CBut)5(salpro)] MeOH2CH2Cl2C7H16 ( 2-3 MeOH2CH2Cl2C7H16) Mn1-O2 1.887(2) Mn1-O5 1.910(2) Mn1-O1 1.955(2) Mn1-O3 1.956(2) Mn1-O4 2.110(2) Mn1-Mn4 2.7942(6) Mn1-Mn3 3.1688(6) Mn2-O8 1.884(2) Mn2-O5 1.907(2) Mn2-N2 1.978(3) Mn2-O9 1.984(2) Mn2-O7 2.183(2) Mn2-O6 2.196(2) Mn1-O2-Mn4 94.97(9) Mn1-O2-Mn3 113.17(10) Mn4-O2-Mn3 110.64(9) Mn4-O5-Mn2 114.58(10) Mn2-Mn4 3.1932(7) Mn3-O11 1.877(2) Mn3-O2 1.9096(19) Mn3-N1 1.966(3) Mn3-O9 1.970(2) Mn3-O12 2.181(2) Mn3-O10 2.224(2) Mn3-Mn4 3.1361(7) Mn4-O5 1.888(2) Mn4-O2 1.904(2) Mn4-O13 1.946(2) Mn4-O14 1.949(2) Mn4-O15 2.081(2) Mn4-O5-Mn1 94.74(9) Mn2-O5-Mn1 111.16(10) Mn3-O9-Mn2 134.10(10)

PAGE 235

235 Table A-3. Selected interatomic di stances () and angles () for NBu4[Mn(O2CPh)2(salproH)]CH2Cl2 ( 2-4 CH2Cl2) Mn1-O5 1.8927(18) Mn1-O6 1.9031(19) Mn1-N1 2.019(2) Mn1-N2 2.043(2) Mn1-O2 2.1392(19) Mn1-O4 2.2005(19) O5-Mn1-O6 91.53(8) O5-Mn1-N1 88.77(9) O6-Mn1-N1 176.05(9) O5-Mn1-N2 174.10(9) O6-Mn1-N2 88.20(9) N1-Mn1-N2 91.10(9) O5-Mn1-O2 98.88(8) O6-Mn1-O2 92.15(8) N1-Mn1-O2 91.69(9) N2-Mn1-O2 87.02(8) O5-Mn1-O4 87.08(7) O6-Mn1-O4 88.89(8) N1-Mn1-O4 87.19(8) N2-Mn1-O4 87.02(8) O2-Mn1-O4 173.91(7) Mn2-O13 1.8955(19) Mn2-O12 1.9062(19) Mn2-N3 2.025(2) Mn2-N4 2.040(2) Mn2-O9 2.118(2) Mn2-O11 2.1862(19) O13-Mn2-O12 91.10(8) O13-Mn2-N3 88.66(9) O12-Mn2-N3 177.79(9) O13-Mn2-N4 175.42(9) O12-Mn2-N4 88.51(9) N3-Mn2-N4 91.56(9) O13-Mn2-O9 98.21(8) O12-Mn2-O9 90.70(8) N3-Mn2-O9 91.51(9) N4-Mn2-O9 86.36(9) O13-Mn2-O11 88.57(8) O12-Mn2-O11 90.04(8) N3-Mn2-O11 87.76(8) N4-Mn2-O11 86.87(8) O9-Mn2-O11 173.16(8)

PAGE 236

236 Table A-4. Selected interatomic di stances () and angles () for [Mn4(hmp)2(pdmH)2(MeCN)4](ClO4)4 ( 3-1 ) Mn1-O2 1.876(3) Mn1-O1 1.898(3) Mn1-O4' 1.968(2) Mn1-N2 2.064(3) Mn1-N1 2.205(3) Mn1-O4 2.254(2) Mn1Mn1' 3.2092(10) Mn1-O1-Mn2 112.43(11) Mn1-O2-Mn2' 109.12(11) Mn1'-O4-Mn1 98.73(10) Mn2-O1 2.157(2) Mn2-O2' 2.215(2) Mn2-N3 2.267(3) Mn2-N4 2.272(3) Mn2-O3 2.303(3) Mn2-N5 2.310(4) Mn2-O4 2.363(2) Mn1'-O4-Mn2 100.45(10) Mn1-O4-Mn2 93.85(9)

PAGE 237

237 Table A-5. Selected interatomic di stances () and angles () for [Mn25O18(OH)2(hmp)6(pdm)8(pdmH)2(L)2](ClO4)6MeCNMeOH ( 3-2 MeCNMeOH) Mn(1)-O(26)' 2.130(5) Mn(1)-O(18) 2.180(5) Mn(1)-O(1) 2.215(7) Mn(2)-O(4) 1.874(5) Mn(2)-O(19) 1.904(5) Mn(2)-O(18) 1.918(4) Mn(3)-O(18) 1.887(5) Mn(3)-O(21) 1.901(5) Mn(3)-O(3) 1.928(5) Mn(4)-O(20)' 1.887(4) Mn(4)-O(23) 1.889(4) Mn(4)-O(27) 2.109(4) Mn(4)-O(21) 2.112(5) Mn(4)-N(3) 2.140(6) Mn(4)-O(6) 2.241(5) Mn(4)-O(5) 2.331(5) Mn(5)-O(24) 1.866(5) Mn(5)-O(27) 1.890(5) Mn(5)-O(16) 1.936(5) Mn(6)-O(19) 1.875(5) Mn(6)-O(23) 1.877(5) Mn(6)-O(21) 2.109(4) Mn(12)-O(17) 2.182(5) Mn(12)-O(24) 2.207(5) Mn(12)-N(8) 2.258(8) Mn(3)-O(3)-Mn(11)' 99.8(2) Mn(2)-O(4)-Mn(1) 102.2(2) Mn(3)-O(5)-Mn(10)' 96.68(17) Mn(4)-O(6)-Mn(13) 88.63(19) Mn(2)-O(7)-Mn(3) 86.92(16) Mn(7)-O(8)-Mn(13) 95.5(2) Mn(5)'-O(9)-Mn(8) 97.1(2) Mn(8)-O(10)-Mn(10) 88.12(16) Mn(6)-O(22) 2.133(4) Mn(6)-O(7) 2.259(5) Mn(6)-O(8) 2.297(5) Mn(7)-O(25) 1.884(4) Mn(7)-O(22) 1.889(4) Mn(7)-O(14) 1.939(5) Mn(8)-O(20) 1.874(4) Mn(8)-O(19) 1.885(4) Mn(8)-O(22) 2.105(5) Mn(9)-O(27) 1.879(5) Mn(9)-O(27)' 1.879(5) Mn(9)-O(22)' 1.892(4) Mn(10)-O(11) 1.883(5) Mn(10)-O(20) 1.908(5) Mn(10)-O(25) 1.909(5) Mn(11)-O(26) 2.116(5) Mn(11)-O(25) 2.205(4) Mn(11)-O(11) 2.223(6) Mn(11)-O(14) 2.493(5) Mn(12)-O(26) 2.112(5) Mn(13)-O(17) 1.887(5) Mn(13)-O(23) 1.903(5) Mn(13)-O(24) 1.919(4) Mn(13)-N(9) 2.027(6) Mn(13)-O(6) 2.258(5) Mn(13)-O(17)-Mn(12) 103.7(2) Mn(3)-O(18)-Mn(2) 108.5(2) Mn(6)-O(19)-Mn(8) 110.7(2) Mn(4)'-O(20)-Mn(10) 115.8(2) Mn(9)-O(21)-Mn(3) 136.2(2) Mn(8)-O(22)-Mn(6) 93.69(19) Mn(6)-O(23)-Mn(4) 109.6(2) Mn(7)-O(24)-Mn(12) 100.21(18)

PAGE 238

238 Table A-6. Selected interatomic di stances () and angles () for [Fe7O4(O2CPh)11(dmem)2]MeCN ( 4-1 MeCN) Fe1-O10 1.8276(18) Fe1-O2 1.9966(18) Fe1-O4 2.0424(18) Fe1-O1 2.0519(19) Fe1-N2 2.248(2) Fe1-N1 2.269(3) Fe2-O9 1.9234(17) Fe2-O10 1.941(2) Fe2-O5 2.051(2) Fe2-O2 2.0534(17) Fe2-O3 2.0537(18) Fe2-O7 2.1053(18) O10 -Fe1-O2 98.25(8) O10 -Fe1-O4 95.05(8) O2-Fe1-O4 166.09(8) O10 -Fe1-O1 103.89(8) O2-Fe1-O1 90.28(8) O9-Fe2-O10 83.84(8) O9-Fe2-O5 94.27(8) O10-Fe2-O5 176.11(8) O9-Fe2-O2 94.30(7) O10-Fe2-O2 97.13(7) O9-Fe2-O3 174.58(8) O10-Fe2-O3 94.32(8) O2-Fe2-O3 90.99(7) O9-Fe2-O7 94.97(7) O10-Fe2-O7 91.06(8) O2-Fe2-O7 168.24(8) O3-Fe2-O7 79.95(7) O9-Fe3-O8 102.13(8) Fe2-Fe4 2.9287(5) Fe3-O9 1.8436(18) Fe3-O8 2.0092(19) Fe3-O6 2.027(2) Fe3-O13 2.038(2) Fe3-O11 2.0547(19) Fe3-O12 2.200(2) Fe3-C34 2.470(3) Fe4-O9 1.989(2) Fe4-O10 1.9915(17) Fe4-O14 2.0680(17) Fe4-O14 2.0681(18) O9-Fe3-O6 95.81(9) O9-Fe4-O9 176.62(10) O9-Fe4-O10 80.88(7) O9 -Fe4-O10 101.66(7) O9-Fe4-O10 101.66(7) O9 -Fe4-O10 80.87(7) O10-Fe4-O10 84.82(10) O9-Fe4-O14 87.48(7) O9-Fe4-O14 90.19(8) O10-Fe4-O14 92.39(7) O10 -Fe4-O14 167.19(8) Fe1-O2-Fe2 125.76(10) Fe3-O9-Fe2 120.44(9) Fe3-O9-Fe4 125.57(10) Fe2-O9-Fe4 96.92(8) Fe1 -O10-Fe2 124.59(9) Fe1 -O10-Fe4 134.38(10) Fe2-O10-Fe4 96.25(8)

PAGE 239

239 Table A-7. Selected interatomic di stances () and angles () for [Fe7O4(O2CMe)11(dmem)2]MeCN ( 4-2 MeCN) Fe1-O1 1.8783(17) Fe1-O6 1.9992(19) Fe1-O8 2.0106(18) Fe1-O5 2.0147(17) Fe1-N2 2.191(2) Fe1-N1 2.282(2) Fe1-Fe2 2.9585(5) Fe2-O3 1.8672(17) Fe2-O1 1.9833(18) Fe2-O12 2.0291(18) Fe2-O11 2.0413(19) Fe2-O5 2.0541(18) Fe2-O7 2.1647(18) Fe3-O1 1.8621(17) Fe3-O2 1.9900(17) Fe3-O16 2.0380(18) Fe3-O10 2.0658(19) Fe3-O14 2.094(2) Fe3-O9 2.0976(18) Fe3-Fe4 2.9540(5) Fe4-O2 1.8787(17) Fe3-O1-Fe1 133.53(10) Fe3-O1-Fe2 123.22(9) Fe1-O1-Fe2 99.98(8) Fe4-O2-Fe5 127.18(9) Fe4-O2-Fe3 99.52(8) Fe5-O2-Fe3 130.01(9) Fe2-O3-Fe5 128.04(9) Fe4-O16 1.9894(18) Fe4-O17 2.0107(18) Fe4-O15 2.057(2) Fe4-N3 2.246(2) Fe5-O2 1.9402(17) Fe5-O3 1.9568(17) Fe5-O4 2.0141(17) Fe5-O19 2.0554(19) Fe5-O21 2.0590(19) Fe5-O18 2.0922(18) Fe5-Fe6 2.9066(5) Fe6-O4 1.9053(18) Fe6-O3 1.9579(17) Fe6-O13 2.0416(18) Fe6-O23 2.0525(19) Fe6-O22 2.055(2) Fe6-O25 2.0619(19) Fe7-O4 1.8175(18) Fe7-O24 2.011(2) Fe7-O20 2.057(2) Fe7-O27 2.0623(19) Fe2-O3-Fe6 129.56(9) Fe5-O3-Fe6 95.88(7) Fe7-O4-Fe6 121.04(9) Fe7-O4-Fe5 132.30(10) Fe6-O4-Fe5 95.69(7) Fe1-O5-Fe2 93.29(7) Fe4-O16-Fe3 94.35(7)

PAGE 240

240 Table A-8. Selected interatomic di stances () and angles () for [Fe6O2(OH)4(O2CBut)8(dmem)2]MeCN ( 4-3 MeCN) Fe1-O5 1.9366(16) Fe1-O5 1.9382(17) Fe1-O1 2.0251(18) Fe1-O2 2.0457(17) Fe1-O10 2.0471(17) Fe1-O3 2.0504(18) Fe1-Fe1 2.8651(7) Fe2-O5 1.8441(17) Fe2-O9 1.9580(18) Fe2-O4 2.0412(18) O5 -Fe1-O5 84.64(7) O5 -Fe1-O1 94.52(7) O5-Fe1-O1 178.33(8) O5 -Fe1-O2 95.30(7) O5-Fe1-O2 88.98(7) O1-Fe1-O2 89.67(7) O5 -Fe1-O10 87.64(7) O5-Fe1-O10 91.72(7) O1-Fe1-O10 89.68(7) O2-Fe1-O10 177.02(7) O5 -Fe1-O3 171.93(7) O5-Fe1-O3 93.52(7) O2-Fe1-O3 92.51(7) O10 -Fe1-O3 84.56(7) O5-Fe2-O9 103.36(8) O5-Fe2-O4 96.43(7) O9-Fe2-O4 91.06(8) O5-Fe2-O8 96.06(7) Fe2-O8 2.0448(18) Fe2-O6 2.0662(19) Fe2-O7 2.2189(19) Fe2-C11 2.481(3) Fe3-O10 1.9420(17) Fe3-O9 1.9651(18) Fe3-O2 2.0181(18) Fe3-O11 2.0331(19) Fe3-N1 2.231(2) Fe3-N2 2.272(2) O9-Fe2-O8 89.35(8) O5-Fe2-O6 99.44(8) O9-Fe2-O6 157.16(8) O5-Fe2-O7 160.49(7) O9-Fe2-O7 96.06(7) O10-Fe3-O9 99.75(8) O10-Fe3-O2 96.45(7) O9-Fe3-O2 90.25(8) O10-Fe3-O11 93.63(8) O9-Fe3-O11 89.97(8) O2-Fe3-O11 169.73(7) Fe3-O2-Fe1 125.40(8) Fe2-O5-Fe1 128.19(9) Fe2-O5-Fe1 126.51(9) Fe1 -O5-Fe1 95.36(7) Fe2-O9-Fe3 117.76(9) Fe3-O10-Fe1 129.75(9)

PAGE 241

241 Table A-9. Selected interatomic di stances () and angles () for [Fe3O(O2CBut)2(N3)3(dmem)2]CH2Cl2 ( 4-4 CH2Cl2) Fe1-O1 1.8716(19) Fe1-N8 2.007(2) Fe1-O3 2.0291(18) Fe1-O7 2.0608(19) Fe1-N4 2.220(2) Fe1-N3 2.243(2) Fe2-O3 1.9787(19) Fe2-O2 1.9834(19) Fe2-N11 2.007(2) O1-Fe1-O3 80.87(8) O1-Fe1-O7 96.21(8) O3-Fe1-O7 88.65(8) O1-Fe1-N4 97.88(9) O3-Fe1-N4 95.48(8) O7-Fe1-N4 165.78(9) O1-Fe1-N3 157.25(8) N8-Fe1-N3 95.53(10) O3-Fe1-N3 76.70(8) O7-Fe1-N3 87.04(9) N4-Fe1-N3 80.71(9) O3-Fe2-O2 155.14(8) O3-Fe2-O4 92.88(8) O2-Fe2-O4 88.52(8) O3-Fe2-O6 89.19(8) O2-Fe2-O6 91.02(8) O4-Fe2-O6 176.04(8) O3-Fe2-O1 77.41(7) Fe2-O4 2.0469(19) Fe2-O6 2.0605(19) Fe2-O1 2.0700(19) Fe3-O1 1.8647(19) Fe3-N5 2.020(2) Fe3-O2 2.0245(19) Fe3-O5 2.066(2) Fe3-N2 2.211(2) Fe3-N1 2.241(2) O2-Fe2-O1 77.73(7) O4-Fe2-O1 93.04(8) O6-Fe2-O1 90.70(7) O1-Fe3-N5 106.84(10) O1-Fe3-O2 81.60(8) O1-Fe3-O5 95.92(8) O2-Fe3-O5 88.49(8) O1-Fe3-N2 98.10(9) O2-Fe3-N2 95.09(9) O5-Fe3-N2 165.89(9) O1-Fe3-N1 157.93(8) O2-Fe3-N1 76.51(8) O5-Fe3-N1 86.38(9) N2-Fe3-N1 81.21(9) Fe3-O1-Fe1 162.82(11) Fe3-O1-Fe2 98.34(8) Fe1-O1-Fe2 98.85(8) Fe2-O2-Fe3 96.07(8) Fe2-O2-Fe1 96.77(8)

PAGE 242

242 Table A-10. Selected interatomic di stances () and angles () for [Fe5O2(OH)(O2CMe)5(hmbp)3](ClO4)2MeCN ( 5-1 MeCN) Fe1-O16 1.825(3) Fe1-O3 2.021(3) Fe1-O15 2.058(3) Fe1-O1 2.075(3) Fe1-N2 2.104(3) Fe1-N1 2.190(3) Fe2-O4 1.933(3) Fe2-O5 1.980(3) Fe2-O3 2.023(3) Fe2-O6 2.052(3) Fe2-N4 2.116(4) Fe2-N3 2.155(5) Fe3-O11 1.826(3) Fe3-O6 2.008(3) Fe3-O7 2.048(3) O16-Fe1-O3 100.08(12) O16-Fe1-O15 93.21(12) O16-Fe1-O1 94.68(12) O4-Fe2-O5 111.55(12) O4-Fe2-O3 90.87(13) O5-Fe2-O3 89.21(13) O4-Fe2-O6 88.51(13) O5-Fe2-O6 88.59(13) O11-Fe3-O6 104.43(12) O11-Fe3-O9 93.41(13) O16-Fe4-O11 84.34(11) O16-Fe4-O14 92.60(11) O11-Fe4-O14 176.42(12) O16-Fe4-O10 176.14(12) O11-Fe4-O12 88.83(11) O16-Fe4-O5 88.71(12) O11-Fe4-O5 87.71(11) O14-Fe4-O5 90.36(12) Fe3-O9 2.063(3) Fe3-N5 2.094(4) Fe3-N6 2.189(4) Fe4-O16 1.951(3) Fe4-O11 1.956(3) Fe4-O14 2.003(3) Fe4-O10 2.017(3) Fe4-O12 2.039(3) Fe4-O5 2.058(3) Fe5-O11 1.951(3) Fe5-O16 1.951(3) Fe5-O2 2.017(3) Fe5-O8 2.021(3) Fe5-O4 2.031(3) Fe5-O13 2.041(3) O11-Fe5-O16 84.45(11) O11-Fe5-O8 91.98(12) O16-Fe5-O8 176.20(12) O11-Fe5-O4 88.95(12) O16-Fe5-O4 87.62(11) O2-Fe5-O4 91.79(13) O8-Fe5-O4 93.62(12) O11-Fe5-O13 88.85(11) O16-Fe5-O13 89.14(11) O4-Fe5-O13 176.24(12) Fe1-O3-Fe2 114.99(14) Fe2-O4-Fe5 119.95(15) Fe2-O5-Fe4 118.09(14) Fe3-O6-Fe2 112.45(14) Fe3-O11-Fe5 125.89(15) Fe3-O11-Fe4 127.03(14) Fe5-O11-Fe4 94.45 O10-Fe4-O5 91.81(13)

PAGE 243

243 Table A-11. Selected interatomic di stances () and angles () for [Fe6O2(OH)2(O2CPh)6(hmbp)4](NO3)2MeCNH2O ( 5-2 MeCNH2O) Fe1-O7 1.912(2) Fe1-O3 1.974(2) Fe1-O9 2.046(2) Fe1-O1 2.059(2) Fe1-N2 2.097(3) Fe1-N1 2.186(3) Fe2-O8 1.938(2) Fe2-O8 1.978(2) Fe2-O7 2.011(2) Fe2-O5 2.029(2) O7 -Fe1-O3 105.96(9) O7 -Fe1-O9 94.49(9) O3-Fe1-O9 91.67(10) O7 -Fe1-O1 89.85(10) O3-Fe1-O1 96.07(10) O9 -Fe1-O1 169.75(10) O8-Fe2-O8 82.56(9) O8-Fe2-O7 88.09(9) O8 -Fe2-O7 94.18(9) O8-Fe2-O5 97.21(9) O8 -Fe2-O5 176.80(9) O7-Fe2-O5 89.00(9) O8-Fe2-O4 175.77(9) O8 -Fe2-O4 94.14(9) O7-Fe2-O4 89.52(9) O5-Fe2-O4 86.24(9) O8-Fe2-O3 92.59(9) O8 -Fe2-O3 91.68(9) O7-Fe2-O3 174.13(9) O5-Fe2-O3 85.13(9) Fe2-O4 2.033(2) Fe2-O3 2.063(2) Fe2-Fe2 2.9430(8) Fe3-O8 1.832(2) Fe3-O9 2.015(2) Fe3-O6 2.031(2) Fe3-O10 2.049(2) Fe3-N3 2.095(3) Fe3-N4 2.183(3) O4-Fe2-O3 90.14(9) O8-Fe3-O9 99.68(9) O8-Fe3-O6 98.96(9) O9-Fe3-O6 93.29(10) O8-Fe3-O10 93.08(9) O9-Fe3-O10 95.63(10) O6-Fe3-O10 163.60(10) O8-Fe3-N3 173.82(10) O9-Fe3-N3 77.43(10) O6-Fe3-N3 86.72(10) O10-Fe3-N3 81.85(10) O8-Fe3-N4 109.05(10) O9-Fe3-N4 151.19(10) O6-Fe3-N4 80.31(10) O10-Fe3-N4 85.27(11) Fe1 -O7-Fe2 125.85(11) Fe3-O8-Fe2 124.38(11) Fe3-O8-Fe2 129.12(11) Fe2-O8-Fe2 97.44(9) Fe3-O9-Fe1 116.32(10)

PAGE 244

244 Table A-12. Selected interatomic di stances () and angles () for [Fe18O8(OH)2(O2CBut)28(heen)4] 4C5H12CH2Cl2 ( 6-1C5H12CH2Cl2) Fe1-O6 1.848(2) Fe1-O5' 1.981(3) Fe1-O5 2.022(3) Fe1-O1 2.046(3) Fe1-O3 2.055(3) Fe1-O7 2.102(3) Fe2-O6 1.923(2) Fe2-O11 1.993(2) Fe2-O12 1.994(3) Fe2-O4 2.019(3) Fe2-O2 2.028(3) Fe2-O9 2.083(3) Fe3-O6 1.894(2) Fe3-O11 1.937(3) Fe3-O8 2.019(3) Fe3-O10 2.060(2) Fe3-N2 2.163(3) Fe3-N1 2.220(3) Fe4-O11 1.819(3) Fe4-O14 1.872(4) Fe4-O17 1.974(3) Fe4-O13 2.013(3) Fe4-O16 2.029(3) Fe5-O18 1.954(2) Fe5-O16 1.966(3) Fe5-O19 2.000(2) Fe5-O17 2.011(2) Fe1'-O5-Fe1 104.35(11) Fe1-O6-Fe3 132.53(13) Fe1-O6-Fe2 122.96(12) Fe3-O6-Fe2 96.22(11) Fe4-O11-Fe3 138.87(13) Fe4-O11-Fe2 125.23(13) Fe3-O11-Fe2 92.62(11) Fe5-N4 2.175(3) Fe5-N3 2.202(3) Fe6-O23 1.943(2) Fe6-O30 1.978(2) Fe6-O31 2.030(3) Fe6-O21 2.032(3) Fe6-O18 2.034(2) Fe6-O29 2.042(3) Fe6-Fe7 2.9008(8) Fe7-O23 1.941(2) Fe7-O30 1.949(2) Fe7-O24 2.024(3) Fe7-O35 2.042(2) Fe7-O19 2.048(2) Fe7-O33 2.073(3) Fe8-O23 1.849(2) Fe8-O28 1.994(3) Fe8-O25 2.036(3) Fe8-O22 2.037(3) Fe8-O27 2.108(3) Fe8-O26 2.140(3) Fe9-O30 1.843(2) Fe9-O34 1.980(3) Fe9-O32 2.018(3) Fe9-O36 2.084(3) Fe9-O37 2.114(3) Fe9-O38 2.123(3) Fe5-O18-Fe6 127.19(13) Fe5-O19-Fe7 126.89(12) Fe8-O23-Fe7 126.05(13) Fe8-O23-Fe6 120.88(12) Fe7-O23-Fe6 96.64(11) Fe9-O30-Fe7 120.39(13) Fe9-O30-Fe6 126.41(13)

PAGE 245

245 Table A-13. Selected interatomic di stances () and angles () for [Fe9O4(OH)4(O2CPh)13(heenH)2]9MeCN ( 6-2 MeCN) Fe1-O7 1.939(4) Fe1-O1 1.984(4) Fe1-O8 1.985(4) Fe1-O5 2.013(4) Fe1-O3 2.088(4) Fe1-O6 2.145(4) Fe2-O7 1.912(4) Fe2-O12 1.990(4) Fe2-O4 2.017(4) Fe2-O9 2.045(4) Fe2-O11 2.057(4) Fe2-O6 2.076(4) Fe3-O14 1.887(4) Fe3-O15 1.989(4) Fe3-O17 2.010(4) Fe3-O10 2.049(4) Fe3-O6 2.069(4) Fe3-O11 2.093(4) Fe4-O14 1.964(4) Fe4-O22 1.979(4) Fe4-O19 2.000(4) Fe4-O21 2.048(4) Fe4-O18 2.067(4) Fe4-O11 2.115(4) Fe5-O27 1.853(4) Fe5-O22 1.986(4) Fe5-O25 2.038(5) Fe1-O5-Fe6 130.80(18) Fe1-O5-Fe7 121.71(18) Fe6-O5-Fe7 92.87(14) Fe3-O6-Fe2 97.96(15) Fe3-O6-Fe1 133.2(2) Fe2-O6-Fe1 90.39(15) Fe8-O7-Fe2 125.50(19) Fe5-O23 2.098(6) Fe5-N3 2.189(6) Fe5-N4 2.205(6) Fe6-O14 1.905(4) Fe6-O27 1.961(4) Fe6-O24 2.027(4) Fe6-O16 2.035(4) Fe6-O5 2.058(4) Fe6-O28 2.093(4) Fe7-O27 1.960(4) Fe7-O31 1.963(4) Fe7-O36 2.010(4) Fe7-O26 2.012(4) Fe7-O5 2.086(4) Fe7-O21 2.141(4) Fe8-O7 1.889(4) Fe8-O31 1.968(4) Fe8-O34 2.008(4) Fe8-O32 2.044(4) Fe8-O21 2.048(4) Fe8-O13 2.071(4) Fe9-O31 1.850(4) Fe9-O8 2.016(4) Fe9-O37 2.032(4) Fe9-O35 2.071(4) Fe9-N1 2.195(5) Fe9-N2 2.223(5) Fe3-O14-Fe6 125.3(2) Fe3-O14-Fe4 101.98(17) Fe6-O14-Fe4 129.7(2) Fe8-O21-Fe4 127.88(19) Fe8-O21-Fe7 93.22(15) Fe4-O21-Fe7 122.76(19) Fe4-O22-Fe5 120.5(2)

PAGE 246

246 Table A-14. Selected interatomic di stances () and angles () for 4[Fe7O3(OMe)3(MeOH)1.5(heen)3Cl4.5(H2O)]Cl5[FeCl4]6MeOHH2O ( 6-3 MeOHH2O) Fe1-O2 1.844(4) Fe1-O3 1.848(3) Fe1-O1 1.856(4) Fe1-Cl1 2.2564(15) Fe2-O1 1.955(4) Fe2-O6 1.973(4) Fe2-O14 1.993(4) Fe2-O5 2.028(4) Fe2-O4 2.128(4) Fe2-Cl2 2.3670(15) Fe3-O2 1.947(4) Fe3-O5 1.970(4) Fe3-O7 1.996(4) Fe3-O6 2.015(4) Fe3-N1 2.175(6) Fe3-N2 2.180(6) Fe4-O9 1.964(4) Fe4-O7 1.987(4) Fe4-O2 1.987(4) Fe4-O8 2.037(4) Fe4-O4" 2.043(13) Cl8'-Fe8-Cl8 109.79(8) Cl8"-Fe8-Cl8 110.36(8) Cl8-Fe8-Cl8'" 108.27(9) Fe1-O1-Fe7 123.50(18) Fe1-O1-Fe2 128.89(19) Fe7-O1-Fe2 102.33(17) Fe1-O2-Fe4 128.2(2) Fe3-O2-Fe4 102.21(17) Fe1-O3-Fe5 124.31(19) Fe1-O3-Fe6 128.29(19) Fe4-Cl3 2.3995(15) Fe4-Cl4 2.458(4) Fe5-O3 1.944(3) Fe5-O8 1.960(4) Fe5-O10 2.000(4) Fe5-O9 2.022(4) Fe5-N3 2.157(5) Fe5-N4 2.195(4) Fe6-O3 1.949(4) Fe6-O13 1.972(4) Fe6-O10 1.981(4) Fe6-O12 2.025(4) Fe6-O11 2.131(4) Fe6-Cl5 2.3727(15) Fe7-O1 1.941(4) Fe7-O12 1.960(4) Fe7-O14 1.998(4) Fe7-O13 2.018(4) Fe7-N6 2.169(5) Fe7-N5 2.183(5) Fe8-Cl8 2.2078(15) Fe5-O3-Fe6 102.64(15) Fe3-O5-Fe2 102.14(18) Fe2-O6-Fe3 102.49(17) Fe4-O7-Fe3 100.48(17) Fe5-O8-Fe4 102.71(17) Fe4-O9-Fe5 103.09(17) Fe6-O10-Fe5 99.53(15) Fe7-O12-Fe6 102.13(16) Fe6-O13-Fe7 101.98(16) Fe2-O14-Fe7 99.01(16)

PAGE 247

247 Table A-15. Selected interatomic di stances () and angles () for [Fe6O2(O2CPh)5(heen)3(heenH)](ClO4)2EtOH.5H2O ( 6-4 EtOH.5H2O) Fe1-O7' 1.918(3) Fe1-O10 1.957(3) Fe1-O7 1.980(3) Fe1-O1 2.027(3) Fe1-O8 2.044(3) Fe1-O4' 2.116(3) Fe1-Fe1' 2.9113(11) Fe2-O7 1.846(3) Fe2-O9 2.022(3) Fe2-O2' 2.048(3) Fe2-O7-Fe1' 120.52(15) Fe2-O7-Fe1 130.32(16) Fe1'-O7-Fe1 96.63(12) Fe2-O3 2.053(3) Fe2-N1 2.171(4) Fe2-N2 2.200(4) Fe3-O11 1.853(10) Fe3-O8 1.984(4) Fe3-O9 1.995(3) Fe3-O10 1.997(3) Fe3-O5 2.025(6) Fe3-N3 2.187(6) Fe3-N4 2.212(5) Fe3-O8-Fe1 101.62(16) Fe3-O9-Fe2 121.73(16) Fe1-O10-Fe3 104.30(16)

PAGE 248

248 Table A-16. Selected interatomic di stances () and angles () for [Mn8O3(OH)(OMe)(O2CPh)7(edte)(edteH2)](O2CPh)2CH2Cl2MeOH ( 7-1 CH2Cl2MeOH) Mn1-O3 1.8836(19) Mn1-O25 1.8844(17) Mn1-O4 1.9492(18) Mn1-O17 1.9591(17) Mn1-O27 2.176(2) Mn1-O18 2.3097(18) Mn2-O7 1.8554(18) Mn2-O25 1.9105(18) Mn2-O18 1.9424(17) Mn2-O6 1.9566(17) Mn2-O5 2.1596(18) Mn2-O1 2.4368(17) Mn3-O10 2.1431(17) Mn3-O7 2.1543(18) Mn3-O8 2.180(2) Mn3-O9 2.239(2) Mn3-N2 2.330(2) Mn3-N1 2.359(2) Mn3-O6 2.4054(17) Mn4-O10 1.8661(17) Mn4-O22 1.9136(18) Mn4-O15 1.9297(17) Mn4-O6 1.9652(17) Mn4-O11 2.1725(19) Mn4-O1 2.4695(18) Mn7-O1-Mn8 97.12(8) Mn8-O1-Mn2 88.95(6) Mn6-O1-Mn2 92.52(6) Mn7-O1-Mn4 89.31(6) Mn6-O1-Mn4 92.09(6) Mn2-O1-Mn4 80.05(5) Mn2-O6-Mn4 107.14(8) Mn2-O6-Mn3 97.31(7) Mn5-O14 1.8923(18) Mn5-O22 1.8941(17) Mn5-O16 1.9467(18) Mn5-O12 1.9517(19) Mn5-O19 2.195(2) Mn5-O15 2.3662(18) Mn6-O1 2.2242(17) Mn6-O16 2.2495(18) Mn6-O18 2.2606(17) Mn6-O15 2.2617(18) Mn6-O17 2.2647(18) Mn6-N4 2.279(2) Mn6-N3 2.285(2) Mn7-O22 1.8957(18) Mn7-O1 1.8997(17) Mn7-O20 1.9505(19) Mn7-O21 1.9565(18) Mn7-O23 2.1235(19) Mn7-O16 2.3692(18) Mn8-O25 1.8937(17) Mn8-O1 1.9079(17) Mn8-O21 1.9517(18) Mn8-O26 1.9571(19) Mn8-O24 2.1360(19) Mn8-O17 2.3122(18) Mn2-O7-Mn3 109.95(9) Mn4-O10-Mn3 110.16(8) Mn6-O15-Mn5 97.48(7) Mn5-O16-Mn6 111.83(8) Mn8-O21-Mn7 93.83(8) Mn5-O22-Mn7 105.86(8) Mn1-O25-Mn8 104.91(8) Mn4-O6-Mn3 97.19(7)

PAGE 249

249 Table A-17. Selected interatomic di stances () and angles () for [Mn12O4(OH)2(edte)4Cl6(H2O)2]6MeCNH2O ( 7-2 MeCNH2O) Mn1-O6' 2.065(3) Mn1-O3 2.088(3) Mn1-O1 2.138(3) Mn1-O2' 2.306(3) Mn1-Cl1 2.4050(13) Mn1-Cl2 2.5813(14) Mn2-O1 1.841(2) Mn2-O5' 1.910(2) Mn2-O4 2.0098(15) Mn2-O2 2.015(3) Mn2-O4-Mn2' 133.9(2) O6'-Mn1-O3 168.16(11) O2'-Mn1-Cl1 108.95(7) O1-Mn1-Cl2 82.17(7) O2'-Mn1-Cl2 147.38(7) O1-Mn2-O5' 172.08(12) O1-Mn2-O4 92.17(10) Mn2-O1' 2.033(3) Mn2-Cl2 2.5255(14) Mn3-O6 1.882(3) Mn3-O3 1.905(3) Mn3-O5 2.107(2) Mn3-O1 2.163(3) Mn3-O2 2.209(3) Mn3-N2 2.275(3) Mn3-N1 2.320(3) O2-Mn2-Cl2 92.62(8) O2-Mn2-Cl2 92.62(8) O1'-Mn2-Cl2 171.93(8) O6-Mn3-O5 87.60(11) O5-Mn3-O1 72.73(9) O3-Mn3-O2 105.80(11) O5-Mn3-N2 75.17(10) O6-Mn3-N1 100.89(12)

PAGE 250

250 Table A-18. Selected interatomic di stances () and angles () for [Mn20O8(OH)4(O2CMe)6(edte)6](ClO4)2MeOH ( 7-3 10MeOH) Mn1-O2 2.111(7) Mn1-O7 2.116(7) Mn1-O14 2.148(11) Mn1-O13 2.258(9) Mn1-O6 2.293(7) Mn1-O4 2.333(6) Mn2-O2 1.874(6) Mn2-O3 1.903(5) Mn2-O1 1.926(7) Mn2-N2 2.165(7) Mn2-O4 2.175(6) Mn2-N1 2.308(8) Mn3-O20 1.882(6) Mn3-O22 1.895(6) Mn3-O6 1.973(6) Mn3-O4 1.988(6) Mn3-O19 2.227(6) Mn4-O5 1.892(7) Mn4-O7 1.910(7) Mn4-O8' 1.934(8) Mn4-N3 2.159(9) Mn4-O6 2.228(6) Mn4-N4' 2.35(2) Mn4-N4 2.356(16) Mn5-O22 1.986(6) Mn5-O23 1.994(5) Mn5-O9 2.040(6) Mn5-O19 2.085(7) Mn5-O3 2.145(6) Mn2-O1-Mn7 128.2(4) Mn2-O2-Mn1 110.8(3) Mn2-O3-Mn5 110.8(3) Mn6-O22 1.890(6) Mn6-O20 1.900(5) Mn6-O10' 1.927(6) Mn6-O12' 1.944(5) Mn6-O17 2.132(6) Mn6-O24 2.523(5) Mn7-O19 1.856(6) Mn7-O16 1.919(6) Mn7-O1 1.979(7) Mn7-O8' 2.002(8) Mn7-O11 2.142(8) Mn7-O9 2.177(7) Mn8-O21 1.990(6) Mn8-O20 1.999(6) Mn8-O11 2.039(7) Mn8-O5 2.121(6) Mn8-O24 2.498(5) Mn9-O23 1.916(6) Mn9-O21' 1.929(6) Mn9-O24' 1.939(5) Mn9-O24 1.946(5) Mn9-O10 2.270(5) Mn9-O12' 2.270(5) Mn10-O9 2.229(6) Mn10-O12 2.246(6) Mn10-O11 2.260(6) Mn10-O10 2.264(6) Mn10-N6 2.303(8) Mn10-N5 2.304(8) Mn10-O10-Mn9 93.0(2) Mn8-O11-Mn7 95.2(3) Mn6'-O12-Mn9' 99.1(2)

PAGE 251

251 Table A-19. Selected interatomic di stances () and angles () for [Fe6O2(O2CBut)8(edteH)2]CHCl3 ( 7-5 2CHCl3) Fe1-O12 1.931(2) Fe1-O6 1.957(3) Fe1-O10 2.025(3) Fe1-O8 2.026(3) Fe1-O1 2.037(3) Fe1-O3 2.051(2) Fe2-O12 1.907(2) Fe2-O4 1.965(3) Fe2-O3 2.030(3) Fe3-O12-Fe2 123.25(13) Fe3-O12-Fe1 123.83(13) Fe2-O12-Fe1 102.16(11) Fe2-O2 2.050(3) Fe2-N2 2.251(3) Fe2-N1 2.289(3) Fe3-O12 1.858(2) Fe3-O13' 2.019(3) Fe3-O13 2.033(2) Fe3-O4' 2.034(2) Fe3-O11 2.039(3) Fe3-O9 2.040(3) Fe3'-O13-Fe3 102.33(11) Fe2-O3-Fe1 94.06(10) Fe2-O4-Fe3' 118.77(12)

PAGE 252

252 Table A-20. Selected interatomic di stances () and angles () for [Fe12O4(OH)2(O2CMe)6(edte)4(H2O)2](ClO4)4MeCN ( 7-6 MeCN) Fe1-O2 1.977(6) Fe1-O3 1.985(6) Fe1-O1 2.040(6) Fe1-O7 2.187(6) Fe1-O4 2.223(6) Fe2-O2 1.964(6) Fe2-O11 1.984(7) Fe2-O5 2.010(7) Fe2-O13 2.053(6) Fe2-O4 2.104(6) Fe3-O10 1.972(7) Fe3-O11 1.988(7) Fe3-O9 2.038(6) Fe3-O13 2.160(6) Fe3-O12 2.239(6) Fe4-O3 1.975(6) Fe4-O21 1.981(7) Fe4-O7 2.004(6) Fe4-O16 2.080(6) Fe4-O23 2.098(6) Fe5-O17 1.944(6) Fe5-O1 1.968(7) Fe5-O37 1.971(6) Fe5-O7 2.032(6) Fe5-O23 2.104(6) Fe6-O18 1.939(6) Fe6-O9 1.967(6) Fe6-O7 1.980(6) Fe6-O13 2.013(6) Fe12-O29 2.156(6) Fe7-O17-Fe5 135.1(3) Fe6-O4 2.074(6) Fe7-O17 1.926(6) Fe7-O33 1.969(6) Fe7-O13 1.982(6) Fe7-O29 2.037(6) Fe7-O12 2.092(6) Fe8-O19 1.972(7) Fe8-O10 1.976(7) Fe8-O31 2.001(7) Fe8-O29 2.068(6) Fe8-O12 2.098(7) Fe9-O21 1.959(6) Fe9-O24 1.979(6) Fe9-O22 2.031(7) Fe9-O37 2.186(6) Fe9-O23 2.237(7) Fe10-O32 1.954(6) Fe10-O24 1.974(6) Fe10-O26 1.995(8) Fe10-O37 2.031(6) Fe10-O27 2.032(7) Fe10-O30 2.103(7) Fe11-O18 1.952(6) Fe11-O29 1.963(6) Fe11-O22 1.980(6) Fe11-O37 2.026(6) Fe11-O30 2.093(6) Fe12-O32 1.962(6) Fe12-O31 1.976(7) Fe12-O30 2.236(7) Fe6-O18-Fe11 134.1(3)

PAGE 253

253 Table A-21. Selected interatomic di stances () and angles () for [Mn10Gd2O8(O2CPh)10(hmp)6(NO3)4]3MeCNMeOH ( 8-4 MeCNMeOH) Gd1-O13 2.312(5) Gd1-O11 2.325(5) Gd1-O12 2.360(5) Gd1-O8 2.469(7) Gd1-N3 2.504(7) Gd1-O1 2.509(4) Gd1-O6 2.512(9) Gd1-O5 2.535(8) Gd1-O9 2.567(7) Gd1-N2 2.949(8) Gd1-N1 2.951(10) Gd1-Mn2 3.4185(10) Mn1-O11 1.916(5) Mn1-O3 1.931(4) Mn1-O1 1.942(4) Mn1-O20 1.973(5) Mn1-O18 2.137(5) Mn1-O22 2.308(5) Mn1-Mn2 3.1104(14) Mn2-O4 1.841(4) Mn2-O13 1.908(5) Mn3-O1-Mn1 133.2(2) Mn3-O1-Mn2 108.5(2) Mn1-O1-Mn2 106.31(19) Mn3-O1-Gd1 101.54(16) Mn1-O1-Gd1 102.50(18) Mn2-O1-Gd1 99.50(17) Mn3-O2-Mn5 125.8(2) Mn3-O2-Mn4 125.3(2) Mn5-O2-Mn4 98.54(18) Mn1-O3-Mn5' 124.1(2) Mn1-O3-Mn4' 118.5(2) Mn5'-O3-Mn4' 96.31(18) Mn2-N5 2.030(6) Mn2-O22 2.190(5) Mn2-O14 2.223(5) Mn2-Mn3 3.1327(13) Mn3-O2 1.882(4) Mn3-O12 1.905(4) Mn3-O1 1.915(4) Mn3-N4 2.026(5) Mn3-O16 2.086(5) Mn3-O14 2.380(6) Mn4-O2 1.912(4) Mn4-O3' 1.947(4) Mn4-O19' 1.954(5) Mn4-O15 1.994(6) Mn4-O4 2.125(4) Mn4-O21' 2.151(6) Mn5-O4' 1.885(4) Mn5-O2 1.907(4) Mn5-O3' 1.938(4) Mn5-O17 1.962(4) Mn5-O23' 2.211(5) Mn5-O3 2.476(4) Mn1-O3-Mn5 124.90(19) Mn5'-O3-Mn5 93.99(17) Mn4'-O3-Mn5 90.99(16) Mn2-O4-Mn5' 124.6(2) Mn2-O4-Mn4 118.7(2) Mn5'-O4-Mn4 104.62(19) Mn1-O11-Gd1 110.4(2) Mn3-O12-Gd1 107.37(19) Mn2-O13-Gd1 107.8(2) Mn2-O14-Mn3 85.71(18) Mn2-O22-Mn1 87.45(16)

PAGE 254

254 Table A-22. Selected interatomic di stances () and angles () for [Mn10Dy2O8(O2CPh)10(hmp)6(NO3)4]3MeCNMeOH ( 8-6 MeCNMeOH) Dy1-O13 2.301(4) Dy1-O11 2.315(4) Dy1-O12 2.341(4) Dy1-O6 2.454(6) Dy1-O8 2.477(4) Dy1-O1 2.478(3) Dy1-N3 2.495(5) Dy1-O5 2.500(5) Dy1-O9 2.515(5) Dy1-N1 2.901(6) Dy1-N2 2.921(5) Mn1-O11 1.915(4) Mn1-O3 1.940(3) Mn1-O1 1.952(3) Mn1-O20 1.977(4) Mn1-O18 2.139(4) Mn1-O22 2.301(4) Mn1-Mn2 3.1220(11) Mn2-O4 1.842(3) Mn2-O13 1.902(4) Mn3-O1-Mn1 132.96(18) Mn3-O1-Mn2 108.24(17) Mn1-O1-Mn2 106.15(15) Mn3-O1-Dy1 101.87(13) Mn1-O1-Dy1 102.70(15) Mn2-O1-Dy1 100.00(13) Mn3-O2-Mn5 126.65(18) Mn3-O2-Mn4 124.95(18) Mn5-O2-Mn4 98.51(14) Mn5'-O3-Mn1 124.39(17) Mn5'-O3-Mn4' 96.34(14) Mn1-O3-Mn4' 118.47(17) Mn2-N5 2.039(5) Mn2-O22 2.186(4) Mn2-O14 2.217(5) Mn3-O2 1.878(3) Mn3-O12 1.904(3) Mn3-O1 1.911(3) Mn3-N4 2.027(5) Mn3-O16 2.092(4) Mn3-O14 2.384(5) Mn4-O2 1.919(3) Mn4-O19' 1.949(4) Mn4-O3' 1.949(3) Mn4-O15 1.995(5) Mn4-O4 2.117(3) Mn4-O21' 2.143(5) Mn5-O4' 1.887(3) Mn5-O2 1.905(3) Mn5-O3' 1.939(3) Mn5-O17 1.961(3) Mn5-O23' 2.212(4) Mn5-O3 2.466(3) Mn5'-O3-Mn5 94.11(13) Mn1-O3-Mn5 124.33(15) Mn4'-O3-Mn5 91.20(13) Mn2-O4-Mn5' 124.77(19) Mn2-O4-Mn4 118.35(16) Mn5'-O4-Mn4 104.75(15) Mn1-O11-Dy1 110.13(18) Mn3-O12-Dy1 107.18(14) Mn2-O13-Dy1 108.13(17) Mn2-O14-Mn3 85.69(16) Mn2-O22-Mn1 88.14(13)

PAGE 255

255 Table A-23. Selected interatomic di stances () and angles () for [Mn10Y2O8(O2CPh)10(hmp)6(NO3)4]MeCN ( 8-9 MeCN) Y1-O11 2.290(5) Y1-O13 2.296(6) Y1-O12 2.334(4) Y1-O6 2.432(7) Y1-O8 2.450(5) Y1-O1 2.450(4) Y1-N3 2.465(7) Y1-O5 2.475(7) Y1-O9 2.497(6) Y1-N1 2.885(8) Y1-N2 2.910(7) Mn1-O11 1.908(5) Mn1-O3 1.921(4) Mn1-O1 1.947(4) Mn1-O20 1.967(5) Mn1-O18 2.137(5) Mn1-O22 2.291(5) Mn2-O4 1.843(4) Mn2-O13 1.895(5) Mn2-O1 1.947(5) Mn2-N4 2.014(6) Mn2-O22 2.181(5) Mn3-O1-Mn2 107.8(2) Mn3-O1-Mn1 132.9(2) Mn2-O1-Mn1 105.72(18) Mn3-O1-Y1 102.26(16) Mn2-O1-Y1 100.61(17) Mn1-O1-Y1 103.05(19) Mn3-O2-Mn5 126.5(2) Mn3-O2-Mn4 124.8(2) Mn5-O2-Mn4 98.53(17) Mn1-O3-Mn5' 124.1(2) Mn1-O3-Mn4' 118.8(2) Mn2-O14 2.285(7) Mn3-O2 1.869(4) Mn3-O12 1.892(4) Mn3-O1 1.911(4) Mn3-N5 2.015(5) Mn3-O16 2.091(5) Mn3-O14 2.470(8) Mn4-O2 1.917(4) Mn4-O19' 1.948(5) Mn4-O3' 1.962(4) Mn4-O15 1.998(7) Mn4-O4 2.081(4) Mn4-O21' 2.100(6) Mn5-O4' 1.895(4) Mn5-O2 1.910(4) Mn5-O3' 1.933(4) Mn5-O17 1.945(5) Mn5-O23' 2.213(5) Mn5-O3 2.439(4) Mn5-Mn4' 3.1527(14) Mn5-Mn5' 3.2167(19) O3-Mn5' 1.933(4) Mn1-O3-Mn5 124.78(18) Mn5'-O3-Mn5 94.02(17) Mn4'-O3-Mn5 90.85(16) Mn2-O4-Mn5' 124.5(2) Mn2-O4-Mn4 118.21(19) Mn5'-O4-Mn4 104.82(19) Mn1-O11-Y1 110.5(2) Mn3-O12-Y1 107.22(17) Mn2-O13-Y1 108.0(2) Mn2-O14-Mn3 81.8(3) Mn2-O22-Mn1 87.87(16)

PAGE 256

256 APPENDIX B LIST OF COMPOUNDS [Mn4O2(O2CMe)5(salpro)] ( 2-1 ) [Mn4O2(O2CEt)5(salpro)] ( 2-2 ) [Mn4O2(O2CBut)5(salpro)] ( 2-3 ) NMe4[Mn(O2CPh)2(salproH)] ( 2-4 ) [Mn4(hmp)4(pdmH)2(MeCN)4](ClO4)4 ( 3-1 ) [Mn25O18(OH)2(hmp)6(pdm)8(pdmH)2(L)2](ClO4)6 ( 3-2 ) [Fe7O4(O2CPh)11(dmem)2] ( 4-1 ) [Fe7O4(O2CMe)11(dmem)2] ( 4-2 ) [Fe6O2(OH)4(O2CCBut)8(dmem)2] ( 4-3 ) [Fe3O(O2CBut)2(N3)3(dmem)2] ( 4-4 ) [Fe5O2(OH)(O2CMe)5(hmbp)3](ClO4)2 ( 5-1 ) [Fe6O2(OH)2(O2CPh)6(hmbp)4](NO3)2 ( 5-2 ) [Fe6O2(OH)2(O2CMe)6(hmbp)4](NO3)2 ( 5-3 ) [Fe6O2(OH)2(O2CBut)6(hmbp)4](NO3)2 ( 5-4 ) [Fe18O8(OH)2(O2CBut)28(heen)4] ( 6-1 ) [Fe9O4(OH)4(O2CPh)13(heenH)2] ( 6-2 ) 4[Fe7O3(OMe)3(MeOH)1.5(heen)3Cl4.5(H2O)]Cl5[FeCl4] ( 6-3 ) [Fe6O2(O2CPh)5(heen)3(heenH)](ClO4)2 ( 6-4 ) [Mn8O3(OH)(OMe)(O2CPh)7(edte)(edteH2)](O2CPh) ( 7-1 ) [Mn12O4(OH)2(edte)4Cl6(H2O)2] ( 7-2 ) [Mn20O8(OH)4(O2CMe)6(edte)6](ClO4)2 ( 7-3 ) [Fe5O2(O2CPh)7(edte)(H2O)] ( 7-4 )

PAGE 257

257 [Fe6O2(O2CBut)8(edteH)2] ( 7-5 ) [Fe12O4(OH)2(O2CMe)6(edte)4(H2O)2](ClO4)4 ( 7-6 ) [Fe12O4(OH)8(edte)4(H2O)2](ClO4)4 ( 7-7 ) [Fe12O4(OH)8(edte)4(H2O)2](NO3)4 ( 7-8 ) [Mn10Ln2O8(O2CPh)10(hmp)6(NO3)4] (Ln = Pr ( 8-1 ), Nd ( 8-2 ), Sm (8-3 ), Gd ( 8-4 ), Tb ( 8-5 ), Dy ( 8-6 ), Ho ( 8-7 ), Er ( 8-8 ) Y ( 8-9 )) [Mn12O12(O2CCHCl2)16(H2O)4] ( 9-1 ) (NPrn 4)[Mn12O12(O2CCHCl2)16(H2O)4] ( 9-2 ) (NPrn 4)2[Mn12O12(O2CCHCl2)16(H2O)4] ( 9-3 ) (NPrn 4)3[Mn12O12(O2CCHCl2)16(H2O)4] ( 9-4 ) (NMe4)3[Mn12O12(O2CCHCl2)16(H2O)4] ( 9-5 )

PAGE 258

258 APPENDIX C PHYSICAL MEASUREMENTS Infrared spectra were recorded in th e solid state (KBr pellets) on a Nicolet Nexus 670 FTIR spectrometer in the 400 4000 cm-1 range. Elemental analyses (C, H and N) were performed by the in-house facil ities of the University of Florida, Chemistry Department. Cl analysis was performed by Complete Analysis Laboratories, Inc. in Parsippany, New Jersey. Variable-temperature dc and ac magnetic susceptibil ity data were collected at the University of Florida using a Quantum Design MPMS-XL SQUI D susceptometer equipped with a 7 T magnet and operating in the 1.8 300 K range. Samples were embedded in solid eicosane to prevent torquing. Magnetization vs. field and temper ature data was fit using the program MAGNET.53 Pascal's constants were used to estimate the diamagnetic correction, which was subtracted from the experimental susceptibility to give the molar paramagnetic susceptibility (M). Double-axis angledependent high-frequency electron pa ramagnetic resonance (HFEPR) studies were performed on single crysta ls using a rotating cavity296 and a 7 T transverse magnetic field, which can be rotated about an axis perpendicular to the axis of the rotating cavity. In addition, a 17 T axial magnet was employed for some single-axis measurements. The experiments were carried out over a wide range of frequencies (50-200 GHz) and with the sample at temperatures in the 1.8-20 K range.

PAGE 259

259 APPENDIX D VAN VLECK EQUATIONS p = param agnetic impurity c = N B 2/3k N = Avogadro's number g = Lande's factor k = Boltzmann constant T = Temperature TIP = Temperature independent paramagnetism D-1[ Mn4O2(O2CMe)5(salpro)] ( 2-1 ) M = (c g2)/T (Num/Den) + TIP l=Jbb/k/T m=Jbw/k/T n=Jww/k/T Num=+ 300.0000 *exp( 0.0000 *l+ 0.0000 *m+ 0.0000 *n) + 6.0000 *exp( 2.0000 *l+ 0.0000 *m+ 0.0000 *n) + 0.0000 *exp( 2.0000 *l+ -4.0000 *m+ 2.0000 *n) + 300.0000 *exp( 2.0000 *l+ -2.0000 *m+ 2.0000 *n) + 30.0000 *exp( 2.0000 *l+ 2.0000 *m+ 2.0000 *n) + 6.0000 *exp( 2.0000 *l+ -6.0000 *m+ 6.0000 *n) + 84.0000 *exp( 2.0000 *l+ 4.0000 *m+ 6.0000 *n) + 30.0000 *exp( 2.0000 *l+ -8.0000 *m+ 12.0000 *n) + 180.0000 *exp( 2.0000 *l+ 6.0000 *m+ 12.0000 *n) + 84.0000 *exp( 2.0000 *l+ -10.0000 *m+ 20.0000 *n) + 330.0000 *exp( 2.0000 *l+ 8.0000 *m+ 20.0000 *n) + 114.0000 *exp( 6.0000 *l+ 0.0000 *m+ 0.0000 *n) + 300.0000 *exp( 6.0000 *l+ -6.0000 *m+ 2.0000 *n) + 30.0000 *exp( 6.0000 *l+ -2.0000 *m+ 2.0000 *n) + 414.0000 *exp( 6.0000 *l+ 4.0000 *m+ 2.0000 *n) + 30.0000 *exp( 6.0000 *l+ -12.0000 *m+ 6.0000 *n) + 6.0000 *exp( 6.0000 *l+ -10.0000 *m+ 6.0000 *n) + 180.0000 *exp( 6.0000 *l+ 8.0000 *m+ 6.0000 *n) + 6.0000 *exp( 6.0000 *l+ -16.0000 *m+ 12.0000 *n) + 180.0000 *exp( 6.0000 *l+ 2.0000 *m+ 12.0000 *n) + 330.0000 *exp( 6.0000 *l+ 12.0000 *m+ 12.0000 *n)

PAGE 260

260 + 30.0000 *exp( 6.0000 *l+ -20.0000 *m+ 20.0000 *n) + 84.0000 *exp( 6.0000 *l+ -14.0000 *m+ 20.0000 *n) + 546.0000 *exp( 6.0000 *l+ 16.0000 *m+ 20.0000 *n) + 84.0000 *exp( 12.0000 *l+ 0.0000 *m+ 0.0000 *n) + 30.0000 *exp( 12.0000 *l+ -8.0000 *m+ 2.0000 *n) + 414.0000 *exp( 12.0000 *l+ -2.0000 *m+ 2.0000 *n) + 510.0000 *exp( 12.0000 *l+ 6.0000 *m+ 2.0000 *n) + 6.0000 *exp( 12.0000 *l+ -16.0000 *m+ 6.0000 *n) + 294.0000 *exp( 12.0000 *l+ -12.0000 *m+ 6.0000 *n) + 84.0000 *exp( 12.0000 *l+ -6.0000 *m+ 6.0000 *n) + 180.0000 *exp( 12.0000 *l+ 2.0000 *m+ 6.0000 *n) + 330.0000 *exp( 12.0000 *l+ 12.0000 *m+ 6.0000 *n) + 0.0000 *exp( 12.0000 *l+ -24.0000 *m+ 12.0000 *n) + 6.0000 *exp( 12.0000 *l+ -22.0000 *m+ 12.0000 *n) + 30.0000 *exp( 12.0000 *l+ -18.0000 *m+ 12.0000 *n) + 180.0000 *exp( 12.0000 *l+ -4.0000 *m+ 12.0000 *n) + 546.0000 *exp( 12.0000 *l+ 18.0000 *m+ 12.0000 *n) + 6.0000 *exp( 12.0000 *l+ -30.0000 *m+ 20.0000 *n) + 30.0000 *exp( 12.0000 *l+ -26.0000 *m+ 20.0000 *n) + 84.0000 *exp( 12.0000 *l+ -20.0000 *m+ 20.0000 *n) + 546.0000 *exp( 12.0000 *l+ 10.0000 *m+ 20.0000 *n) + 840.0000 *exp( 12.0000 *l+ 24.0000 *m+ 20.0000 *n) + 180.0000 *exp( 20.0000 *l+ 0.0000 *m+ 0.0000 *n) + 414.0000 *exp( 20.0000 *l+ -10.0000 *m+ 2.0000 *n) + 510.0000 *exp( 20.0000 *l+ -2.0000 *m+ 2.0000 *n) + 330.0000 *exp( 20.0000 *l+ 8.0000 *m+ 2.0000 *n) + 294.0000 *exp( 20.0000 *l+ -20.0000 *m+ 6.0000 *n) + 84.0000 *exp( 20.0000 *l+ -14.0000 *m+ 6.0000 *n) + 180.0000 *exp( 20.0000 *l+ -6.0000 *m+ 6.0000 *n) + 330.0000 *exp( 20.0000 *l+ 4.0000 *m+ 6.0000 *n) + 1386.0000 *exp( 20.0000 *l+ 16.0000 *m+ 6.0000 *n) + 6.0000 *exp( 20.0000 *l+ -30.0000 *m+ 12.0000 *n) + 30.0000 *exp( 20.0000 *l+ -26.0000 *m+ 12.0000 *n) + 180.0000 *exp( 20.0000 *l+ -12.0000 *m+ 12.0000 *n) + 546.0000 *exp( 20.0000 *l+ 10.0000 *m+ 12.0000 *n) + 840.0000 *exp( 20.0000 *l+ 24.0000 *m+ 12.0000 *n) + 0.0000 *exp( 20.0000 *l+ -40.0000 *m+ 20.0000 *n) + 6.0000 *exp( 20.0000 *l+ -38.0000 *m+ 20.0000 *n) + 30.0000 *exp( 20.0000 *l+ -34.0000 *m+ 20.0000 *n) + 84.0000 *exp( 20.0000 *l+ -28.0000 *m+ 20.0000 *n) + 546.0000 *exp( 20.0000 *l+ 2.0000 *m+ 20.0000 *n) + 1224.0000 *exp( 20.0000 *l+ 32.0000 *m+ 20.0000 *n) Den=+ 25.0000 *exp( 0.0000 *l+ 0.0000 *m+ 0.0000 *n) + 3.0000 *exp( 2.0000 *l+ 0.0000 *m+ 0.0000 *n) + 1.0000 *exp( 2.0000 *l+ -4.0000 *m+ 2.0000 *n)

PAGE 261

261 + 24.0000 *exp( 2.0000 *l+ -2.0000 *m+ 2.0000 *n) + 5.0000 *exp( 2.0000 *l+ 2.0000 *m+ 2.0000 *n) + 3.0000 *exp( 2.0000 *l+ -6.0000 *m+ 6.0000 *n) + 7.0000 *exp( 2.0000 *l+ 4.0000 *m+ 6.0000 *n) + 5.0000 *exp( 2.0000 *l+ -8.0000 *m+ 12.0000 *n) + 9.0000 *exp( 2.0000 *l+ 6.0000 *m+ 12.0000 *n) + 7.0000 *exp( 2.0000 *l+ -10.0000 *m+ 20.0000 *n) + 11.0000 *exp( 2.0000 *l+ 8.0000 *m+ 20.0000 *n) + 12.0000 *exp( 6.0000 *l+ 0.0000 *m+ 0.0000 *n) + 24.0000 *exp( 6.0000 *l+ -6.0000 *m+ 2.0000 *n) + 5.0000 *exp( 6.0000 *l+ -2.0000 *m+ 2.0000 *n) + 18.0000 *exp( 6.0000 *l+ 4.0000 *m+ 2.0000 *n) + 6.0000 *exp( 6.0000 *l+ -12.0000 *m+ 6.0000 *n) + 3.0000 *exp( 6.0000 *l+ -10.0000 *m+ 6.0000 *n) + 9.0000 *exp( 6.0000 *l+ 8.0000 *m+ 6.0000 *n) + 3.0000 *exp( 6.0000 *l+ -16.0000 *m+ 12.0000 *n) + 9.0000 *exp( 6.0000 *l+ 2.0000 *m+ 12.0000 *n) + 11.0000 *exp( 6.0000 *l+ 12.0000 *m+ 12.0000 *n) + 5.0000 *exp( 6.0000 *l+ -20.0000 *m+ 20.0000 *n) + 7.0000 *exp( 6.0000 *l+ -14.0000 *m+ 20.0000 *n) + 13.0000 *exp( 6.0000 *l+ 16.0000 *m+ 20.0000 *n) + 7.0000 *exp( 12.0000 *l+ 0.0000 *m+ 0.0000 *n) + 5.0000 *exp( 12.0000 *l+ -8.0000 *m+ 2.0000 *n) + 18.0000 *exp( 12.0000 *l+ -2.0000 *m+ 2.0000 *n) + 20.0000 *exp( 12.0000 *l+ 6.0000 *m+ 2.0000 *n) + 3.0000 *exp( 12.0000 *l+ -16.0000 *m+ 6.0000 *n) + 21.0000 *exp( 12.0000 *l+ -12.0000 *m+ 6.0000 *n) + 7.0000 *exp( 12.0000 *l+ -6.0000 *m+ 6.0000 *n) + 9.0000 *exp( 12.0000 *l+ 2.0000 *m+ 6.0000 *n) + 11.0000 *exp( 12.0000 *l+ 12.0000 *m+ 6.0000 *n) + 1.0000 *exp( 12.0000 *l+ -24.0000 *m+ 12.0000 *n) + 3.0000 *exp( 12.0000 *l+ -22.0000 *m+ 12.0000 *n) + 5.0000 *exp( 12.0000 *l+ -18.0000 *m+ 12.0000 *n) + 9.0000 *exp( 12.0000 *l+ -4.0000 *m+ 12.0000 *n) + 13.0000 *exp( 12.0000 *l+ 18.0000 *m+ 12.0000 *n) + 3.0000 *exp( 12.0000 *l+ -30.0000 *m+ 20.0000 *n) + 5.0000 *exp( 12.0000 *l+ -26.0000 *m+ 20.0000 *n) + 7.0000 *exp( 12.0000 *l+ -20.0000 *m+ 20.0000 *n) + 13.0000 *exp( 12.0000 *l+ 10.0000 *m+ 20.0000 *n) + 15.0000 *exp( 12.0000 *l+ 24.0000 *m+ 20.0000 *n) + 9.0000 *exp( 20.0000 *l+ 0.0000 *m+ 0.0000 *n) + 18.0000 *exp( 20.0000 *l+ -10.0000 *m+ 2.0000 *n) + 20.0000 *exp( 20.0000 *l+ -2.0000 *m+ 2.0000 *n) + 11.0000 *exp( 20.0000 *l+ 8.0000 *m+ 2.0000 *n) + 21.0000 *exp( 20.0000 *l+ -20.0000 *m+ 6.0000 *n) + 7.0000 *exp( 20.0000 *l+ -14.0000 *m+ 6.0000 *n)

PAGE 262

262 + 9.0000 *exp( 20.0000 *l+ -6.0000 *m+ 6.0000 *n) + 11.0000 *exp( 20.0000 *l+ 4.0000 *m+ 6.0000 *n) + 28.0000 *exp( 20.0000 *l+ 16.0000 *m+ 6.0000 *n) + 3.0000 *exp( 20.0000 *l+ -30.0000 *m+ 12.0000 *n) + 5.0000 *exp( 20.0000 *l+ -26.0000 *m+ 12.0000 *n) + 9.0000 *exp( 20.0000 *l+ -12.0000 *m+ 12.0000 *n) + 13.0000 *exp( 20.0000 *l+ 10.0000 *m+ 12.0000 *n) + 15.0000 *exp( 20.0000 *l+ 24.0000 *m+ 12.0000 *n) + 1.0000 *exp( 20.0000 *l+ -40.0000 *m+ 20.0000 *n) + 3.0000 *exp( 20.0000 *l+ -38.0000 *m+ 20.0000 *n) + 5.0000 *exp( 20.0000 *l+ -34.0000 *m+ 20.0000 *n) + 7.0000 *exp( 20.0000 *l+ -28.0000 *m+ 20.0000 *n) + 13.0000 *exp( 20.0000 *l+ 2.0000 *m+ 20.0000 *n) + 17.0000 *exp( 20.0000 *l+ 32.0000 *m+ 20.0000 *n) D-2 [Mn4(hmp)4(pdmH)2(MeCN)4](ClO4)4 ( 3-1 ) M = (c g2)/T (Num/Den) + TIP l=Jbb/k/T m=Jbw/k/T Num=+ 630.0000 *exp( 0.0000 *l+ 0.0000 *m)+ 6.0000 *exp( 2.0000 *l+ 0.0000 *m) + 0.0000 *exp( 2.0000 *l+ -4.0000 *m)+ 630.0000 *exp( 2.0000 *l+ -2.0000 *m) + 30.0000 *exp( 2.0000 *l+ 2.0000 *m)+ 6.0000 *exp( 2.0000 *l+ -6.0000 *m) + 84.0000 *exp( 2.0000 *l+ 4.0000 *m)+ 30.000 0 *exp( 2.0000 *l+ -8.0000 *m) + 180.0000 *exp( 2.0000 *l+ 6.0000 *m)+ 84.0000 *exp( 2.0000 *l+ -10.0000 *m) + 330.0000 *exp( 2.0000 *l+ 8.0000 *m)+ 180.0000 *exp( 2.0000 *l+ -12.0000 *m) + 546.0000 *exp( 2.0000 *l+ 10.0000 *m)+ 114.0000 *exp( 6.0000 *l+ 0.0000 *m) + 630.0000 *exp( 6.0000 *l+ -6.0000 *m)+ 30.0000 *exp( 6.0000 *l+ -2.0000 *m) + 414.0000 *exp( 6.0000 *l+ 4.0000 *m)+ 30.0000 *exp( 6.0000 *l+ -12.0000 *m) + 6.0000 *exp( 6.0000 *l+ -10.0000 *m)+ 180.0000 *exp( 6.0000 *l+ 8.0000 *m) + 186.0000 *exp( 6.0000 *l+ -16.0000 *m)+ 180.0000 *exp( 6.0000 *l+ 2.0000 *m) + 330.0000 *exp( 6.0000 *l+ 12.0000 *m)+ 30.0000 *exp( 6.0000 *l+ -20.0000 *m) + 84.0000 *exp( 6.0000 *l+ -14.0000 *m)+ 546.0000 *exp( 6.0000 *l+ 16.0000 *m) + 84.0000 *exp( 6.0000 *l+ -24.0000 *m)+ 546.0000 *exp( 6.0000 *l+ 6.0000 *m)

PAGE 263

263 + 840.0000 *exp( 6.0000 *l+ 20.0000 *m)+ 630.0000 *exp( 12.0000 *l+ 0.0000 *m) + 30.0000 *exp( 12.0000 *l+ -8.0000 *m )+ 414.0000 *exp( 12.0000 *l+ -2.0000 *m) + 510.0000 *exp( 12.0000 *l+ 6.0000 *m)+ 6.0000 *exp( 12.0000 *l+ -16.0000 *m) + 624.0000 *exp( 12.0000 *l+ -12.0000 *m)+ 84.0000 *exp( 12.0000 *l+ -6.0000 *m) + 180.0000 *exp( 12.0000 *l+ 2.0000 *m)+ 330.0000 *exp( 12.0000 *l+ 12.0000 *m) + 0.0000 *exp( 12.0000 *l+ -24.0000 *m)+ 186.0000 *exp( 12.0000 *l+ -22.0000 *m) + 30.0000 *exp( 12.0000 *l+ -18.0000 *m)+ 180.0000 *exp( 12.0000 *l+ -4.0000 *m) + 546.0000 *exp( 12.0000 *l+ 18.0000 *m)+ 90.0000 *exp( 12.0000 *l+ -30.0000 *m) + 30.0000 *exp( 12.0000 *l+ -26.0000 *m )+ 84.0000 *exp( 12.0000 *l+ -20.0000 *m) + 546.0000 *exp( 12.0000 *l+ 10.0000 *m)+ 840.0000 *exp( 12.0000 *l+ 24.0000 *m) + 30.0000 *exp( 12.0000 *l+ -36.0000 *m)+ 840.0000 *exp( 12.0000 *l+ 14.0000 *m) + 1224.0000 *exp( 12.0000 *l+ 30.0000 *m)+ 180.0000 *exp( 20.0000 *l+ 0.0000 *m) + 414.0000 *exp( 20.0000 *l+ -10.0000 *m)+ 510.0000 *exp( 20.0000 *l+ -2.0000 *m) + 330.0000 *exp( 20.0000 *l+ 8.0000 *m)+ 624.0000 *exp( 20.0000 *l+ -20.0000 *m) + 84.0000 *exp( 20.0000 *l+ -14.0000 *m)+ 180.0000 *exp( 20.0000 *l+ -6.0000 *m) + 330.0000 *exp( 20.0000 *l+ 4.0000 *m)+ 1386.0000 *exp( 20.0000 *l+ 16.0000 *m) + 186.0000 *exp( 20.0000 *l+ -30.0000 *m)+ 30.0000 *exp( 20.0000 *l+ -26.0000 *m) + 180.0000 *exp( 20.0000 *l+ -12.0000 *m)+ 546.0000 *exp( 20.0000 *l+ 10.0000 *m) + 840.0000 *exp( 20.0000 *l+ 24.0000 *m)+ 0.0000 *exp( 20.0000 *l+ -40.0000 *m) + 90.0000 *exp( 20.0000 *l+ -38.0000 *m )+ 30.0000 *exp( 20.0000 *l+ -34.0000 *m) + 84.0000 *exp( 20.0000 *l+ -28.0000 *m)+ 546.0000 *exp( 20.0000 *l+ 2.0000 *m) + 1224.0000 *exp( 20.0000 *l+ 32.0000 *m)+ 6.0000 *exp( 20.0000 *l+ -48.0000 *m) + 30.0000 *exp( 20.0000 *l+ -44.0000 *m)+ 546.0000 *exp( 20.0000 *l+ -8.0000 *m) + 840.0000 *exp( 20.0000 *l+ 6.0000 *m)+ 1224.0000 *exp( 20.0000 *l+ 22.0000 *m) + 1710.0000 *exp( 20.0000 *l+ 40.0000 *m) Den=+ 36.0000 *exp( 0.0000 *l+ 0.0000 *m )+ 3.0000 *exp( 2.0000 *l+ 0.0000 *m) + 1.0000 *exp( 2.0000 *l+ -4.0000 *m)+ 35.0000 *exp( 2.0000 *l+ -2.0000 *m) + 5.0000 *exp( 2.0000 *l+ 2.0000 *m)+ 3.0000 *exp( 2.0000 *l+ -6.0000 *m) + 7.0000 *exp( 2.0000 *l+ 4.0000 *m)+ 5.0000 *exp( 2.0000 *l+ -8.0000 *m) + 9.0000 *exp( 2.0000 *l+ 6.0000 *m)+ 7.0000 *exp( 2.0000 *l+ -10.0000 *m) + 11.0000 *exp( 2.0000 *l+ 8.0000 *m)+ 9.0000 *exp( 2.0000 *l+ -12.0000 *m) + 13.0000 *exp( 2.0000 *l+ 10.0000 *m)+ 12.0000 *exp( 6.0000 *l+ 0.0000 *m) + 35.0000 *exp( 6.0000 *l+ -6.0000 *m)+ 5.0000 *exp( 6.0000 *l+ -2.0000 *m) + 18.0000 *exp( 6.0000 *l+ 4.0000 *m)+ 6.0000 *exp( 6.0000 *l+ -12.0000 *m) + 3.0000 *exp( 6.0000 *l+ -10.0000 *m)+ 9.0000 *exp( 6.0000 *l+ 8.0000 *m) + 12.0000 *exp( 6.0000 *l+ -16.0000 *m)+ 9.0000 *exp( 6.0000 *l+ 2.0000 *m) + 11.0000 *exp( 6.0000 *l+ 12.0000 *m)+ 5.0000 *exp( 6.0000 *l+ -20.0000 *m) + 7.0000 *exp( 6.0000 *l+ -14.0000 *m)+ 13.0000 *exp( 6.0000 *l+ 16.0000 *m) + 7.0000 *exp( 6.0000 *l+ -24.0000 *m)+ 13.0000 *exp( 6.0000 *l+ 6.0000 *m) + 15.0000 *exp( 6.0000 *l+ 20.0000 *m)+ 20.0000 *exp( 12.0000 *l+ 0.0000 *m) + 5.0000 *exp( 12.0000 *l+ -8.0000 *m)+ 18.0000 *exp( 12.0000 *l+ -2.0000 *m) + 20.0000 *exp( 12.0000 *l+ 6.0000 *m)+ 3.0000 *exp( 12.0000 *l+ -16.0000 *m) + 32.0000 *exp( 12.0000 *l+ -12.0000 *m)+ 7.0000 *exp( 12.0000 *l+ -6.0000 *m) + 9.0000 *exp( 12.0000 *l+ 2.0000 *m)+ 11.0000 *exp( 12.0000 *l+ 12.0000 *m) + 1.0000 *exp( 12.0000 *l+ -24.0000 *m)+ 12.0000 *exp( 12.0000 *l+ -22.0000 *m)

PAGE 264

264 + 5.0000 *exp( 12.0000 *l+ -18.0000 *m)+ 9.0000 *exp( 12.0000 *l+ -4.0000 *m) + 13.0000 *exp( 12.0000 *l+ 18.0000 *m)+ 10.0000 *exp( 12.0000 *l+ -30.0000 *m) + 5.0000 *exp( 12.0000 *l+ -26.0000 *m)+ 7.0000 *exp( 12.0000 *l+ -20.0000 *m) + 13.0000 *exp( 12.0000 *l+ 10.0000 *m)+ 15.0000 *exp( 12.0000 *l+ 24.0000 *m) + 5.0000 *exp( 12.0000 *l+ -36.0000 *m)+ 15.0000 *exp( 12.0000 *l+ 14.0000 *m) + 17.0000 *exp( 12.0000 *l+ 30.0000 *m)+ 9.0000 *exp( 20.0000 *l+ 0.0000 *m) + 18.0000 *exp( 20.0000 *l+ -10.0000 *m )+ 20.0000 *exp( 20.0000 *l+ -2.0000 *m) + 11.0000 *exp( 20.0000 *l+ 8.0000 *m)+ 32.0000 *exp( 20.0000 *l+ -20.0000 *m) + 7.0000 *exp( 20.0000 *l+ -14.0000 *m)+ 9.0000 *exp( 20.0000 *l+ -6.0000 *m) + 11.0000 *exp( 20.0000 *l+ 4.0000 *m)+ 28.0000 *exp( 20.0000 *l+ 16.0000 *m) + 12.0000 *exp( 20.0000 *l+ -30.0000 *m)+ 5.0000 *exp( 20.0000 *l+ -26.0000 *m) + 9.0000 *exp( 20.0000 *l+ -12.0000 *m)+ 13.0000 *exp( 20.0000 *l+ 10.0000 *m) + 15.0000 *exp( 20.0000 *l+ 24.0000 *m)+ 1.0000 *exp( 20.0000 *l+ -40.0000 *m) + 10.0000 *exp( 20.0000 *l+ -38.0000 *m)+ 5.0000 *exp( 20.0000 *l+ -34.0000 *m) + 7.0000 *exp( 20.0000 *l+ -28.0000 *m)+ 13.0000 *exp( 20.0000 *l+ 2.0000 *m) + 17.0000 *exp( 20.0000 *l+ 32.0000 *m)+ 3.0000 *exp( 20.0000 *l+ -48.0000 *m) + 5.0000 *exp( 20.0000 *l+ -44.0000 *m)+ 13.0000 *exp( 20.0000 *l+ -8.0000 *m) + 15.0000 *exp( 20.0000 *l+ 6.0000 *m)+ 17.0000 *exp( 20.0000 *l+ 22.0000 *m) + 19.0000 *exp( 20.0000 *l+ 40.0000 *m) D.3 [Fe3O(O2CBut)2(N3)3(dmem)2 ( 4-4 ) M = (c g2)/T (Num/Den) m=Ja/k/T n=Jb/k/T Num = + 52.5000 exp( 8.7500 m+ 0.0000 n) + 15.0000 exp( 1.7500 m+ 2.0000 n) + 52.5000 exp( 6.7500 m+ 2.0000 n) + 126.0000 exp( 13.7500 m+ 2.0000 n) + 1.5000 exp( -5.2500 m+ 6.0000 n) + 15.0000 exp( -2.2500 m+ 6.0000 n) + 52.5000 exp( 2.7500 m+ 6.0000 n) + 126.0000 exp( 9.7500 m+ 6.0000 n) + 247.5000 exp( 18.7500 m+ 6.0000 n) + 1.5000 exp( -11.2500 m+ 12.0000 n) + 15.0000 exp( -8.2500 m+ 12.0000 n) + 52.500 0 exp( -3.2500 m+ 12.0000 n) + 126.0000 exp( 3.7500 m+ 12.0000 n)+ 247.5000 exp( 12.7500 m+ 12.0000 n) + 429.0000 exp( 23.7500 m+ 12.0000 n)+ 15.00 00 exp( -16.2500 m+ 20.0000 n) + 52.5000 exp( -11.2500 m+ 20.0000 n)+ 126.0000 exp( -4.2500 m+ 20.0000 n)

PAGE 265

265 + 247.5000 exp( 4.7500 m+ 20.0000 n)+ 429.0000 exp( 15.7500 m+ 20.0000 n) + 682.5000 exp( 28.7500 m+ 20.0000 n)+ 52.50 00 exp( -21.2500 m+ 30.0000 n) + 126.0000 exp( -14.2500 m+ 30.0000 n)+ 247.5000 exp( -5.2500 m+ 30.0000 n) + 429.0000 exp( 5.7500 m+ 30.0000 n)+ 682.5000 exp( 18.7500 m+ 30.0000 n) + 1020.0000 exp( 33.7500 m+ 30.0000 n) Den = + 6.0000 exp( 8.7500 m+ 0.0000 n) + 4.0000 exp( 1.7500 m+ 2.0000 n) + 6.0000 exp( 6.7500 m+ 2.0000 n)+ 8.0000 exp( 13.7500 m+ 2.0000 n) + 2.0000 exp( -5.2500 m+ 6.0000 n)+ 4.0000 exp( -2.2500 m+ 6.0000 n) + 6.0000 exp( 2.7500 m+ 6.0000 n)+ 8.0000 exp( 9.7500 m+ 6.0000 n) + 10.0000 exp( 18.7500 m+ 6.0000 n)+ 2.0000 exp( -11.2500 m+ 12.0000 n) + 4.0000 exp( -8.2500 m+ 12.0000 n)+ 6.0000 exp( -3.2500 m+ 12.0000 n) + 8.0000 exp( 3.7500 m+ 12.0000 n)+ 10.0000 exp( 12.7500 m+ 12.0000 n) + 12.0000 exp( 23.7500 m+ 12.0000 n)+ 4.0000 exp( -16.2500 m+ 20.0000 n) + 6.0000 exp( -11.2500 m+ 20.0000 n)+ 8.0000 exp( -4.2500 m+ 20.0000 n) + 10.0000 exp( 4.7500 m+ 20.0000 n)+ 12.0000 exp( 15.7500 m+ 20.0000 n) + 14.0000 exp( 28.7500 m+ 20.0000 n)+ 6.0000 exp( -21.2500 m+ 30.0000 n) + 8.0000 exp( -14.2500 m+ 30.0000 n)+ 10.0000 exp( -5.2500 m+ 30.0000 n) + 12.0000 exp( 5.7500 m+ 30.0000 n)+ 14.0000 exp( 18.7500 m+ 30.0000 n) + 16.0000 exp( 33.7500 m+ 30.0000 n)

PAGE 266

266 LIST OF REFERENCES (1) Stohr, J.; Siegmann, H. C. Magnetism: From Fundamentals to Nanoscale Dynamics; Springer 2006; Vol. 152. (2) Jiles, D. Introduction to Magnetism and Magnetic Materials ; Chapman & Hall: NewYork, 1991. (3) Huheey, J. E.; Keiter, E. A.; Keiter, R. L. Inorganic Chemistry: Principles of Structure and Reactivity; 4th ed.; HarperCollin College: New York, 1993. (4) Kahn, O. Molecular Magnetism ; VCH: Weinheim, Germany, 1993. (5) Smart, L.; Moore, E. Solid State Chemistry ; Chapman & Hall: New York, 1993. (6) Miller, J. S.; Epstein, A. J. Angew. Chem. Int. Ed. 1994, 33, 385. (7) Epstein, A. J. MRS Bulletin 2000, 25, 33. (8) Drago, R. S. Physical Methods for Chemists ; 2nd ed.; Surfside Scient ific Publishers, 1992. (9) Aromi, G.; Aubin, S. M. J.; Bolcar, M. A.; Christou, G.; Eppley, H. J.; Folting, K.; Hendrickson, D. N.; Huffman, J. C.; Squire, R. C.; Tsai, H. L.; Wang, S.; Wemple, M. W. Polyhedron 1998, 17, 3005. (10) Miller, J. S.; Epstein, A. J. MRS Bulletin 2000, 25 21. (11) Magnetism: A Supramolecular Function ; Kahn, O., Ed.; Kluwer: Dordrecht, 1996. (12) Lis, T. Acta Crystallogr., Sect. B: Struct. Sci. 1980 36, 2042. (13) Christou, G.; Gatteschi, D.; Hendrickson, D. N.; Sessoli, R. MRS Bulletin 2000, 25, 66. (14) Chakov, N. E.; Lee, S. C.; Harter, A. G.; Kuhns, P. L.; Reyes, A. P.; Hill, S. O.; Dalal, N. S.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. J. Am. Chem. Soc. 2006, 128, 6975. (15) Sessoli, R.; Gatteschi, D.; Caneschi, A.; Novak, M. A. Nature 1993, 365, 141. (16) Leuenberger, M. N.; Loss, D. Nature 2001, 410, 789. (17) Ritter, S. K. In Chem. Eng. News 2004; Vol. 82, p 29. (18) Chudnovsky, E. M. Science 1996, 274, 938. (19) Gatteschi, D.; Sessoli, R. Angew. Chem. Int. Ed. 2003, 42, 268. (20) Gatteschi, D.; Sessoli, R.; Cornia, A. Chem. Commun. 2000, 725.

PAGE 267

267 (21) Christou, G. Polyhedron 2005, 24, 2065. (22) Aromi, G.; Brechin, E. K. Struct. Bonding 2006, 122, 1. (23) Benelli, C.; Gatteschi, D. Chem. Rev. 2002, 102, 2369. (24) Christou, G. Acc. Chem. Res. 1989, 22, 328. (25) Kurtz, D. M. Chem. Rev. 1990, 90, 585. (26) Bertini, I.; Gray, H. B.; Lippard, S. J.; Valentine, J. S. Bioinorganic Chemistry ; University Science Books: Mill Valley, CA, 1994. (27) Manganese Redox Enzymes ; Pecoraro, V. L., Ed.; VCH Publishers, Inc., 1992. (28) Wu, A. J.; Penner-Hahn, J. E.; Pecoraro, V. L. Chem. Rev. 2004, 104, 903. (29) Mukhopadhyay, S.; Mandal, S. K.; Bhaduri, S.; Armstrong, W. H. Chem. Rev. 2004, 104, 3981. (30) Gorun, S. M.; Papaefthymiou, G. C.; Frankel, R. B.; Lippard, S. J. J. Am. Chem. Soc. 1987, 109, 3337. (31) Taft, K. L.; Papaefthymiou, G. C.; Lippard, S. J. Science 1993 259, 1302. (32) Brechin, E. K. Chem. Commun. 2005 5141. (33) Yachandra, V. K.; Sauer, K.; Klein, M. P. Chem. Rev. 1996, 96, 2927. (34) Stamatatos, T. C.; Abboud, K. A.; Wernsdorfer, W.; Christou, G. Angew. Chem. Int. Ed. 2007, 46, 884. (35) Ako, A. M.; Hewitt, I. J.; Mereacre, V.; Clerac, R.; Wernsdorfer, W.; Anson, C. E.; Powell, A. K. Angew. Chem. Int. Ed. 2006, 45, 4926. (36) Vincent, J. B.; Christmas, C.; Chang, H. R.; Li, Q. Y.; Boyd, P. D. W.; Huffman, J. C.; Hendrickson, D. N.; Christou, G. J. Am. Chem. Soc. 1989, 111, 2086. (37) Libby, E.; McCusker, J. K.; Schmitt, E. A. ; Folting, K.; Hendrickson, D. N.; Christou, G. Inorg. Chem. 1991, 30, 3486. (38) Brechin, E. K.; Boskovic, C.; Wernsdor fer, W.; Yoo, J.; Yamaguchi, A.; Sanudo, E. C.; Concolino, T. R.; Rheingold, A. L.; Ishimo to, H.; Hendrickson, D. N.; Christou, G. J. Am. Chem. Soc. 2002, 124, 9710.

PAGE 268

268 (39) Bertoncello, K.; Fallon, G. D.; Murray, K. S.; Tiekink, E. R. T. Inorg. Chem. 1991, 30, 3562. (40) Bonadies, J. A.; Kirk, M. L.; Lah, M. S. ; Kessissoglou, D. P.; Hatfield, W. E.; Pecoraro, V. L. Inorg. Chem. 1989, 28, 2037. (41) Gelasco, A.; Pecoraro, V. L. J. Am. Chem. Soc. 1993, 115, 7928. (42) Mikuriya, M.; Yamato, Y.; Tokii, T. Bull. Chem. Soc. Jpn. 1992, 65, 1466. (43) Bagai, R.; Abboud, K. A.; Christou, G. Dalton Trans. 2006, 3306. (44) Nishida, Y.; Kida, S. J. Chem. Soc., Dalton Trans. 1986, 2633. (45) Vincent, J. B.; Chang, H. R.; Folting, K.; Huffman, J. C.; Christou, G.; Hendrickson, D. N. J. Am. Chem. Soc. 1987, 109, 5703. (46) SHELXTL6; Bruker-AXS:Madison, WI 2000. (47) Schake, A. R.; Vincent, J. B.; Li, Q. Y.; Boyd, P. D. W.; Folting, K.; Huffman, J. C.; Hendrickson, D. N.; Christou, G. Inorg. Chem. 1989, 28, 1915. (48) Liu, W. T.; Thorp, H. H. Inorg. Chem. 1993, 32, 4102. (49) Palenik, G. J. Inorg. Chem. 1997, 36, 4888. (50) Kambe, K. J. Phys. Soc. Jpn. 1950, 5, 48. (51) VanVleck, J. H. The Theory of Electrical and Magnetic Susceptibilities ; Oxford Press, London, 1932. (52) Aromi, G.; Bhaduri, S.; Artus, P.; Huffman, J. C.; Hendrickson, D. N.; Christou, G. Polyhedron 2002, 21, 1779. (53) Davidson, E. R. MAGNET; Indiana Universi ty: Bloomington, IN. 1999. (54) Yoo, J.; Yamaguchi, A.; Nakano, M.; Krzystek, J.; Streib, W. E.; Brunel, L. C.; Ishimoto, H.; Christou, G.; Hendrickson, D. N. Inorg. Chem. 2001, 40, 4604. (55) Hendrickson, D. N.; Christou, G.; Schmitt, E. A.; Libby, E.; Bashkin, J. S.; Wang, S. Y.; Tsai, H. L.; Vincent, J. B.; B oyd, P. D. W.; Huffman, J. C.; Folti ng, K.; Li, Q. Y.; Streib, W. E. J. Am. Chem. Soc. 1992, 114, 2455. (56) Aliaga-Alcalde, N.; Edwards, R. S.; Hill, S. O.; Wernsdorfer, W.; Folting, K.; Christou, G. J. Am. Chem. Soc. 2004, 126, 12503.

PAGE 269

269 (57) Brockman, J. T.; Huffman, J. C.; Christou, G. Angew. Chem. Int. Ed. 2002, 41, 2506. (58) Boudalis, A. K.; Donnadieu, B.; Nastopoul os, V.; Clemente-Juan, J. M.; Mari, A.; Sanakis, Y.; Tuchagues, J. P.; Perlepes, S. P. Angew. Chem. Int. Ed. 2004, 43, 2266. (59) Papaefstathiou, G. S.; Perlepes, S. P.; Es cuer, A.; Vicente, R.; F ont-Bardia, M.; Solans, X. Angew. Chem. Int. Ed. 2001, 40, 884. (60) Papaefstathiou, G. S.; Escuer, A.; Vicente, R.; Font-Bardia, M.; Solans, X.; Perlepes, S. P. Chem. Commun. 2001 2414. (61) Stamatatos, T. C.; Foguet-Albiol, D.; L ee, S. C.; Stoumpos, C. C.; Raptopoulou, C. P.; Terzis, A.; Wernsdorfer, W.; Hill, S. O.; Perlepes, S. P.; Christou, G. J. Am. Chem. Soc. 2007, 129, 9484. (62) Murugesu, M.; Habrych, M.; Werns dorfer, W.; Abboud, K. A.; Christou, G. J. Am. Chem. Soc. 2004, 126, 4766. (63) Stamatatos, T. C.; Abboud, K. A.; Wernsdorfer, W.; Christou, G. Angew. Chem. Int. Ed. 2006, 45, 4134. (64) Brechin, E. K.; Yoo, J.; Nakano, M.; Huffm an, J. C.; Hendrickson, D. N.; Christou, G. Chem. Commun. 1999, 783. (65) Yoo, J.; Brechin, E. K.; Yamaguchi, A.; Nakano, M.; Huffman, J. C.; Maniero, A. L.; Brunel, L. C.; Awaga, K.; Ishimoto, H.; Christou, G.; Hendrickson, D. N. Inorg. Chem. 2000, 39, 3615. (66) Brechin, E. K.; Soler, M.; Davidson, J. ; Hendrickson, D. N.; Parsons, S.; Christou, G. Chem. Commun. 2002, 2252. (67) Brechin, E. K.; Soler, M.; Christou, G.; Helliwell, M.; Teat, S. J.; Wernsdorfer, W. Chem. Commun. 2003, 1276. (68) Vandersluis, P.; Spek, A. L. Acta Crystallogr.; Sect. A 1990, 46 194. (69) Sanudo, E. C.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. Inorg. Chem. 2004, 43, 4137. (70) Boskovic, C.; Wernsdorfer, W.; Folting, K.; Huffman, J. C.; Hendrickson, D. N.; Christou, G. Inorg. Chem. 2002, 41, 5107. (71) Davidson, E. R. GRID; Indiana University: Bloomington, IN. 1999. (72) Goodwin, J. C.; Sessoli, R.; Gatteschi, D.; Wernsdorfer, W.; Powell, A. K.; Heath, S. L. J. Chem. Soc., Dalton Trans. 2000, 1835.

PAGE 270

270 (73) Gorun, S. M.; Papaefthymiou, G. C.; Frankel, R. B.; Lippard, S. J. J. Am. Chem. Soc. 1987, 109, 4244. (74) Jones, L. F.; Brechin, E. K.; Collison, D.; Helliwell, M.; Mallah, T.; Piligkos, S.; Rajaraman, G.; Wernsdorfer, W. Inorg. Chem. 2003, 42, 6601. (75) Powell, A. K.; Heath, S. L.; Gatteschi, D.; Pardi, L.; Sessoli, R.; Spina, G.; Del Giallo, F.; Pieralli, F. J. Am. Chem. Soc. 1995, 117, 2491. (76) Canada-Vilalta, C.; O'Brien, T. A.; Brechi n, E. K.; Pink, M.; Davidson, E. R.; Christou, G. Inorg. Chem. 2004, 43, 5505. (77) McCusker, J. K.; Vincent, J. B.; Schmitt, E. A.; Mino, M. L.; Shin, K.; Coggin, D. K.; Hagen, P. M.; Huffman, J. C.; Christou, G.; Hendrickson, D. N. J. Am. Chem. Soc. 1991, 113, 3012. (78) Kahn, O. Chem. Phys. Lett. 1997, 265 109. (79) Barra, A. L.; Caneschi, A.; Cornia, A.; de Biani, F. F.; Gatteschi, D.; Sangregorio, C.; Sessoli, R.; Sorace, L. J. Am. Chem. Soc. 1999, 121, 5302. (80) Powell, G. W.; Lancashire, H. N.; Brechi n, E. K.; Collison, D.; H ealth, S. L.; Malluh, T.; Wernsdorfer, W. Angew. Chem. Int. Ed. 2004, 43, 5772. (81) Delfs, C.; Gatteschi, D.; Pardi, L.; Sessoli, R.; Wieghardt, K.; Hanke, D. Inorg. Chem. 1993, 32, 3099. (82) Sangregorio, C.; Ohm, T.; Paul sen, C.; Sessoli, R.; Gatteschi, D. Phys. Rev. Lett. 1997 78, 4645. (83) Caneschi, A.; Cornia, A.; Lippard, S. J. Angew. Chem. Int. Ed. 1995, 34, 467. (84) Murugesu, M.; Abboud, K. A.; Christou, G. Dalton Trans. 2003, 4552. (85) Jones, L. F.; Batsanov, A.; Brechin, E. K.; Collison, D.; Helliwell, M.; Mallah, T.; McInnes, E. J. L.; Piligkos, S. Angew. Chem. Int. Ed. 2002, 41, 4318. (86) Moragues-Canovas, M.; Riviere, P.; Ricard, L.; Paulsen, C.; Wernsdorfer, W.; Rajaraman, G.; Brechin, E. K.; Mallah, T. Adv. Mater. 2004 16, 1101. (87) Foguet-Albiol, D.; Abboud, K. A.; Christou, G. Chem. Commun. 2005, 4282. (88) Saalfrank, R. W.; Bernt, I.; Uller, E.; Hampel, F. Angew. Chem. Int. Ed. 1997, 36, 2482. (89) King, P.; Stamatatos, T. C.; Abboud, K. A.; Christou, G. Angew. Chem. Int. Ed. 2006, 45, 7379.

PAGE 271

271 (90) Ammala, P.; Cashion, J. D.; Kepert, C. M.; Moubaraki, B.; Murray, K. S.; Spiccia, L.; West, B. O. Angew. Chem. Int. Ed. 2000, 39, 1688. (91) Murugesu, M.; Abboud, K. A.; Christou, G. Polyhedron 2004 23, 2779. (92) Taft, K. L.; Lippard, S. J. J. Am. Chem. Soc. 1990, 112, 9629. (93) Bagai, R.; Datta, S.; Betancur-Rodriguez, A.; Abboud, K. A.; Hill, S.; Christou, G. Inorg. Chem. 2007, 46, 4535. (94) Duncan, J. F.; Kanekar, C. R.; Mok, K. F. J. Chem. Soc. A 1969, 480. (95) Earnshaw, A.; Figgis, B. N.; Lewis, J. J. Chem. Soc. A 1966, 1656. (96) Bond, A. M.; Clark, R. J. H.; Humphrey, D. G.; Panayiotopoulos, P.; Skelton, B. W.; White, A. H. J. Chem. Soc., Dalton Trans. 1998, 1845 and references there in. (97) Armstrong, W. H.; Lippard, S. J. Inorg. Chem. 1985, 24, 981. (98) Tolis, E. I.; Helliwell, M.; Langley, S.; Raftery, J.; Winpenny, R. E. P. Angew. Chem. Int. Ed. 2003, 42, 3804. (99) Yao, H. C.; Li, Y. Z.; Zheng, L. M.; Xin, X. Q. Inorg. Chim. Acta 2005, 358 2523. (100) Datta, S.; Betancur-RodrigueZ, A.; Lee, S. C.; Hill, S.; Foguet-Albiol, D.; Bagai, R.; Christou, G. Polyhedron 2007, 26, 2243. (101) Jones, L. F.; Jensen, P.; Moubaraki, B.; Berry, K. J.; Boas, J. F.; Pilbrow, J. R.; Murray, K. S. J. Mater. Chem. 2006, 16, 2690. (102) Ako, A. M.; Waldmann, O.; Mereacre, V.; Klower, F.; Hewitt, I. J.; Anson, C. E.; Gudel, H. U.; Powell, A. K. Inorg. Chem. 2007, 46, 756. (103) Brown, I. D.; Wu, K. K. Acta Crystallogr., Sect. B: Struct. Sci. 1976, 32, 1957. (104) Taguchi, T.; Stamatatos, T. C.; Abboud, K. A.; Jones, C. M.; O' Brien, T. A.; Christou, G. Inorg. Chem. 2007, In Press and references there in. (105) Nair, V. S.; Hagen, K. S. Inorg. Chem. 1992, 31, 4048. (106) Ammala, P. S.; Batten, S. R.; Cashion, J. D.; Kepert, C. M.; Moubaraki, B.; Murray, K. S.; Spiccia, L.; West, B. O. Inorg. Chim. Acta 2002 331, 90. (107) Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry ; Wiley: New York, 1980. (108) Gorun, S. M.; Lippard, S. J. J. Am. Chem. Soc. 1985, 107, 4568.

PAGE 272

272 (109) Wemple, M. W.; Coggin, D. K.; Vincent, J. B.; McCusker, J. K.; Streib, W. E.; Huffman, J. C.; Hendrickson, D. N.; Christou, G. J. Chem. Soc., Dalton Trans. 1998, 719. (110) Boudalis, A. K.; Tangoulis, V.; Raptopoul ou, C. P.; Terzis, A.; Tuchagues, J. P.; Perlepes, S. P. Inorg. Chim. Acta 2004 357, 1345. (111) Boudalis, A. K.; Lalioti, N.; Spyroulia s, G. A.; Raptopoulou, C. P.; Terzis, A.; Bousseksou, A.; Tangoulis, V.; Tuchagues, J. P.; Perlepes, S. P. Inorg. Chem. 2002, 41, 6474. (112) Gorun, S. M.; Lippard, S. J. Inorg. Chem. 1991, 30, 1625. (113) Dziobkowski, C. T.; Wrobleski, J. T.; Brown, D. B. Inorg. Chem. 1981, 20, 671. (114) Armstrong, W. H.; Lippard, S. J. J. Am. Chem. Soc. 1983, 105, 4837. (115) Armstrong, W. H.; Spool, A.; Papaefthymiou, G. C.; Frankel, R. B.; Lippard, S. J. J. Am. Chem. Soc. 1984, 106, 3653. (116) Armstrong, W. H.; Lippard, S. J. J. Am. Chem. Soc. 1985, 107, 3730. (117) Lee, S. C.; Stamatatos, T. C.; Hill, S.; Perlepes, S. P.; Christou, G. Polyhedron 2007, 26, 2225. (118) Oshio, H.; Hoshino, N.; Ito, T.; Nakano, M. J. Am. Chem. Soc. 2004, 126, 8805. (119) Benelli, C.; Parsons, S.; Solan, G. A.; Winpenny, R. E. P. Angew. Chem. Int. Ed. 1996 35, 1825. (120) Barra, A. L.; Gatteschi, D.; Sessoli, R. Chem. Eur. J. 2000, 6, 1608. (121) Bacsa, J.; Zhao, H.; Dunbar, K. R. Acta Crystallogr. Sect. C 2003, C59 m561. (122) Canada-Vilalta, C.; O'Brien Ted, A.; Pink, M.; Davidson Ernest, R.; Christou, G. Inorg. Chem. 2003, 42, 7819. (123) Yan, B.; Chen, Z. D. Inorg. Chem. Commun. 2001 4, 138. (124) Seddon, E. J.; Huffman, J. C.; Christou, G. Dalton 2000, 4446. (125) Brechin, E. K.; Knapp, M. J.; Huffma n, J. C.; Hendrickson, D. N.; Christou, G. Inorg. Chim. Acta 2000, 297, 389. (126) Christmas, C. A.; Tsai, H. L.; Pardi, L.; Kesselman, J. M.; Gantzel, P. K.; Chadha, R. K.; Gatteschi, D.; Harvey, D. F.; Hendrickson, D. N. J. Am. Chem. Soc. 1993, 115, 12483. (127) Bagai, R.; Abboud, K. A.; Christou, G. Inorg. Chem. 2007, 46, 5567.

PAGE 273

273 (128) Demnitz, F. W. J.; D'Heni, M. B. Org. Prep. Proced. Int. 1998 30, 467. (129) Heirtzler, F. R. Synlett 1999, 1203. (130) Norrby, T.; Borje, A.; Zhang, L.; Akermark, B. Acta Chem. Scand. 1998, 52, 77. (131) Brown, I. D.; Altermatt, D. Acta Crystallogr., Sect. B: Struct. Sci. 1985, B41 244. (132) Tabernor, J.; Jones, L. F.; Heath, S. L.; Muryn, C.; Aromi, G.; Ribas, J.; Brechin, E. K.; Collison, D. Dalton Trans. 2004, 975. (133) Boskovic, C.; Sieber, A.; Chaboussant, G.; Guedel, H. U.; Ensling, J.; Wernsdorfer, W.; Neels, A.; Labat, G.; Stoeckli-Evans, H.; Janssen, S. Inorg. Chem. 2004, 43, 5053. (134) Boskovic, C.; Labat, G.; Neels, A.; Gudel, H. U. Dalton Trans. 2003, 3671. (135) Lachicotte, R. J.; Hagen, K. S. Inorg. Chim. Acta 1997, 263, 407. (136) Reynolds, R. A.; Coucouvanis, D. Inorg. Chem. 1998, 37, 170. (137) O'Keefe, B. J.; Monnier, S. M. ; Hillmyer, M. A.; Tolman, W. B. J. Am. Chem. Soc. 2001, 123, 339. (138) Mikuriya, M.; Nakadera, K. Chem. Lett. 1995, 213. (139) Mikuriya, M.; Hashimoto, Y.; Nakashima, S. Chem. Commun. 1996, 295. (140) Herold, S.; Lippard, S. J. Inorg. Chem. 1997, 36, 50. (141) Krishnamurthy, D.; Sarjeant, A. N.; Goldbe rg, D. P.; Caneschi, A.; Totti, F.; Zakharov, L. N.; Rheingold, A. L. Chem. Eur. J. 2005, 11, 7328. (142) Overgaard, J.; Hibbs, D. E.; Rentschler, E.; Timco, G. A.; Larsen, F. K. Inorg. Chem. 2003, 42, 7593. (143) Chaudhuri, P.; Rentschler, E.; Birkelbac h, F.; Krebs, C.; Bill, E.; Weyhermuller, T.; Florke, U. Eur. J. Inorg. Chem. 2003, 541. (144) Taft, K. L.; Papaefthymiou, G. C.; Lippard, S. J. Inorg. Chem. 1994, 33, 1510. (145) Bagai, R.; Abboud, K. A.; Christou, G. Chem. Commun. 2007 3359. (146) Bagai, R.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. J. Am. Chem. Soc. 2007, 129, 12918.

PAGE 274

274 (147) Affronte, M.; Carretta, S.; Timco, G. A.; Winpenny, R. E. P. Chem. Commun. 2007, 1789. (148) Watton, S. P.; Fuhrmann, P.; Pence, L. E.; Caneschi, A.; Cornia, A.; Abbati, G. L.; Lippard, S. J. Angew. Chem. Int. Ed. 1997, 36, 2774. (149) Schmitt, W.; Anson, C. E.; Wernsdorfer, W.; Powell, A. K. Chem. Commun. 2005, 2098. (150) Jones, L. F.; Low, D. M.; Helliwell, M. ; Raftery, J.; Collison, D.; Aromi, G.; Cano, J.; Mallah, T.; Wernsdorfer, W.; Brechin, E. K.; McInnes, E. J. L. Polyhedron 2006, 25, 325. (151) Yao, H. C.; Wang, J. J.; Ma, Y. S.; Wa ldmann, O.; Du, W. X.; Song, Y.; Li, Y. Z.; Zheng, L. M.; Decurtins, S.; Xin, X. Q. Chem. Commun. 2006 1745. (152) Bino, A.; Shweky, I.; Cohen, S.; Bauminger, E. R.; Lippard, S. J. Inorg. Chem. 1998, 37, 5168. (153) Wernsdorfer, W. Adv. Chem. Phys. 2001, 118, 99. (154) Wernsdorfer, W.; Aliaga-Alcalde N.; Hendrickson, D. N.; Christou, G. Nature 2002, 416, 406. (155) Tiron, R.; Wernsdorfer, W.; Aliaga-Alcalde, N.; Christou, G. Phys. Rev. B 2003, 68, 140407. (156) Hill, S.; Edwards, R. S.; Aliaga-Alcalde, N.; Christou, G. Science 2003, 302, 1015. (157) Tiron, R.; Wernsdorfer, W.; Foguet-Al biol, D.; Aliaga-Alcalde, N.; Christou, G. Phys. Rev. Lett. 2003, 91, 227203. (158) Sun, Z. M.; Grant, C. M.; Castr o, S. L.; Hendrickson, D. N.; Christou, G. Chem. Commun. 1998, 721. (159) Yang, E. C.; Hendrickson, D. N.; Wern sdorfer, W.; Nakano, M.; Zakharov, L. N.; Sommer, R. D.; Rheingold, A. L.; Ledezma-Gairaud, M.; Christou, G. J. Appl. Phys. 2002, 91, 7382. (160) Andres, H.; Basler, R.; Blake, A. J.; Ca diou, C.; Chaboussant, G.; Grant, C. M.; Gudel, H. U.; Murrie, M.; Parsons, S.; Paulsen, C.; Semadini, F.; Villar, V.; Wernsdorfer, W.; Winpenny, R. E. P. Chem. Eur. J. 2002, 8, 4867. (161) Aubin, S. M. J.; Wemple, M. W.; Adams, D. M.; Tsai, H. L.; Christou, G.; Hendrickson, D. N. J. Am. Chem. Soc. 1996, 118, 7746. (162) Dendrinou-Samara, C.; Alexiou, M.; Zalesk i, C. M.; Kampf, J. W.; Kirk, M. L.; Kessissoglou, D. P.; Pecoraro, V. L. Angew. Chem. Int. Ed. 2003, 42, 3763.

PAGE 275

275 (163) Milios, C. J.; Raptopoulou, C. P.; Terzis, A. ; Lloret, F.; Vicente, R.; Perlepes, S. P.; Escuer, A. Angew. Chem. Int. Ed. 2004, 43, 210. (164) Miyasaka, H.; Clerac, R.; Wernsdorfer, W.; Lecren, L.; Bonhomme, C.; Sugiura, K.; Yamashita, M. Angew. Chem. Int. Ed. 2004, 43, 2801. (165) Hundal, G.; Hundal, M. S.; Obrai, S.; Poonia, N. S.; Kumar, S. Inorg. Chem. 2002, 41, 2077. (166) Plass, W. Eur. J. Inorg. Chem. 1998, 799. (167) de Sousa, A. S.; Fernandes, M. A. Polyhedron 2002, 21, 1883. (168) Bagai, R.; Abboud, K. A.; Christou, G. Inorg. Chem. 2007, In Press. (169) Bagai, R.; Daniels, M. R.; Abboud, K. A.; Christou, G. Inorg. Chem., 2008, In Press. (170) Wu, R. W.; Poyraz, M.; Sowrey, F. E. ; Anson, C. E.; Wocadlo, S.; Powell, A. K.; Jayasooriya, U. A.; Cannon, R. D.; Nakamoto, T.; Katada, M.; Sano, H. Inorg. Chem. 1998, 37, 1913. (171) Milios, C. J.; Vinslava, A.; Wernsdorfer, W.; Prescimone, A.; Wood, P. A.; Parsons, S.; Perlepes, S. P.; Christou, G.; Brechin, E. K. J. Am. Chem. Soc. 2007, 129, 6547. (172) Tasiopoulos, A. J.; Vinslava, A.; Wern sdorfer, W.; Abboud, K. A.; Christou, G. Angew. Chem. Int. Ed. 2004, 43, 2117. (173) Maheswaran, S.; Chastanet, G.; Teat, S. J.; Mallah, T.; Sessoli, R.; Wernsdorfer, W.; Winpenny, R. E. P. Angew. Chem. Int. Ed. 2005, 44, 5044. (174) Godbole, M. D.; Roubeau, O.; Clerac, R.; Kooijman, H.; Spek, A. L.; Bouwman, E. Chem. Commun. 2005, 3715. (175) Rajaraman, G.; Murugesu, M.; Sanudo, E. C.; Soler, M.; Wernsdorfer, W.; Helliwell, M.; Muryn, C.; Raftery, J.; Teat, S. J.; Christou, G.; Brechin, E. K. J. Am. Chem. Soc. 2004, 126, 15445. (176) Murugesu, M.; Wernsdorfer, F.; Abboud, K. A.; Christou, G. Angew. Chem. Int. Ed. 2005, 44, 892. (177) Brechin, E. K.; Christou, G.; Soler, M.; Helliwell, M.; Teat, S. J. Dalton Trans. 2003, 513. (178) Jones, L. F.; Brechin, E. K.; Collison, D.; Raftery, J.; Teat, S. J. Inorg. Chem. 2003, 42, 6971.

PAGE 276

276 (179) Tsai, H. L.; Wang, S. Y.; Folting, K.; St reib, W. E.; Hendrickson, D. N.; Christou, G. J. Am. Chem. Soc. 1995, 117, 2503. (180) Tanase, S.; Aromi, G.; Bouwman, E.; Kooijman, H.; Spek, A. L.; Reedijk, J. Chem. Commun. 2005, 3147. (181) Milios, C. J.; Kefalloniti, E.; Raptopoul ou, C. P.; Terzis, A.; Vicente, R.; Lalioti, N.; Escuer, A.; Perlepes, S. P. Chem. Commun. 2003 819. (182) Alvareza, C. S.; Bond, A. D.; Cave, D.; Mosquera, M. E. G.; Harron, E. A.; Layfield, R. A.; McPartlin, M.; Rawson, J. M.; Wood, P. T.; Wright, D. S. Chem. Commun. 2002 2980. (183) Boskovic, C.; Huffman, J. C.; Christou, G. Chem. Commun. 2002, 2502. (184) Saalfrank, R. W.; Low, N.; Deml eitner, B.; Stalke, D.; Teichert, M. Chem. Eur. J. 1998, 4, 1305. (185) Wemple, M. W.; Tsai, H. L.; Wang, S. Y.; Claude, J. P.; Streib, W. E.; Huffman, J. C.; Hendrickson, D. N.; Christou, G. Inorg. Chem. 1996, 35, 6437. (186) Tasiopoulos, A. J.; Abboud, K. A.; Christou, G. Chem. Commun. 2003, 580. (187) Milios, C. J.; Fabbiani, F. P. A.; Parsons, S.; Murugesu, M.; Christou, G.; Brechin, E. K. Dalton Trans. 2006, 351. (188) Saalfrank, R. W.; Low, N.; Trummer, S. ; Sheldrick, G. M.; Teichert, M.; Stalke, D. Eur. J. Inorg. Chem. 1998, 559. (189) Murugesu, M.; Wernsdorfer, W.; A bboud, K. A.; Brechin, E. K.; Christou, G. Dalton Trans. 2006, 2285. (190) Li, Y. G.; Wernsdorfer, W.; Clerac, R.; Hewitt, I. J.; Anson, C. E.; Powell, A. K. Inorg. Chem. 2006, 45, 2376. (191) Masello, A.; Murugesu, M.; Abboud, K. A.; Christou, G. Polyhedron 2007 26, 2276. (192) Lampropoulos, C.; Murugesu, M.; Abboud, K. A.; Christou, G. Polyhedron 2007, 26, 2129. (193) Sun, Z. M.; Gantzel, P. K.; Hendrickson, D. N. Inorg. Chem. 1996, 35, 6640. (194) Caneschi, A.; Ferraro, F.; Gatteschi, D.; Melandri, M. C.; Rey, P.; Sessoli, R. Angew. Chem. Int. Ed. 1989, 28, 1365. (195) Yu, S. B.; Lippard, S. J.; Shweky, I.; Bino, A. Inorg. Chem. 1992, 31, 3502.

PAGE 277

277 (196) Ye, B. H.; Mak, T.; Williams, I. D.; Li, X. Y. Chem. Commun. 1997, 1813. (197) Coucouvanis, D.; Reynolds, R. A.; Dunham, W. R. J. Am. Chem. Soc. 1995, 117, 7570. (198) Cheng, B. S.; Fries, P. H.; Marchon, J. C.; Scheidt, W. R. Inorg. Chem. 1996, 35, 1024. (199) Komatsuzaki, H.; Ichikawa, S.; Hikichi, S.; Akita, M.; Moro-oka, Y. Inorg. Chem. 1998, 37, 3652. (200) Biswas, S.; Mitra, K.; Adhikary, B.; Lucas, C. R. Transition Met. Chem. 2005, 30, 586. (201) Zhou, A. J.; Qin, L. J.; Beedle, C. C.; Ding, S.; Nakano, M.; Leng, J. D.; Tong, M. L.; Hendrickson, D. N. Inorg. Chem. 2007, 46, 8111. (202) Eppley, H. J.; Tsai, H. L.; Devries, N. ; Folting, K.; Christou, G.; Hendrickson, D. N. J. Am. Chem. Soc. 1995, 117, 301. (203) Soler, M.; Wernsdorfer, W.; Abboud, K. A.; Huffman, J. C.; Davidson, E. R.; Hendrickson, D. N.; Christou, G. J. Am. Chem. Soc. 2003, 125, 3576. (204) Bian, G. Q.; Kuroda-Sowa, T.; Konaka, H.; Hatano, M.; Maekaw a, M.; Munakata, M.; Miyasaka, H.; Yamashita, M. Inorg. Chem. 2004, 43, 4790. (205) Zhao, H. H.; Berlinguette, C. P.; Bacsa, J.; Prosvirin, A. V.; Bera, J. K.; Tichy, S. E.; Schelter, E. J.; Dunbar, K. R. Inorg. Chem. 2004, 43, 1359. (206) Coronado, E.; Torment-Aliaga, A.; Gaita-A rino, A.; Giminez-Saiz, C.; Romero, F. M.; Wernsdorfer, W. Angew. Chem. Int. Ed. 2004, 43, 6152. (207) Bagai, R.; Christou, G. Inorg. Chem. 2007, 46, 10810. (208) Tasiopoulos, A. J.; Wernsdorfe r, W.; Abboud, K. A.; Christou, G. Inorg. Chem. 2005, 44, 6324. (209) King, P.; Wernsdorfer, W. ; Abboud, K. A.; Christou, G. Inorg. Chem. 2005, 44, 8659. (210) Foguet-Albiol, D.; O'Brien, T. A.; Wern sdorfer, W.; Moulton, B.; Zaworotko, M. J.; Abboud, K. A.; Christou, G. Angew. Chem. Int. Ed. 2005, 44, 897. (211) Boskovic, C.; Brechin, E. K.; Streib, W. E.; Folting, K.; Hendrickson, D. N.; Christou, G. Chem. Commun. 2001, 467. (212) Boskovic, C.; Brechin, E. K.; Streib, W. E.; Folting, K.; Bollinger, J. C.; Hendrickson, D. N.; Christou, G. J. Am. Chem. Soc. 2002, 124, 3725.

PAGE 278

278 (213) Rumberger, E. M.; Shah, S. J.; Beedle, C. C.; Zakharov, L. N.; Rheingold, A. L.; Hendrickson, D. N. Inorg. Chem. 2005, 44, 2742. (214) Yao, H. C.; Li, Y. Z.; Song, Y.; Ma, Y. S.; Zheng, L. M.; Xin, X. Q. Inorg. Chem. 2006 45, 59. (215) Dendrinou-Samara, C.; Zaleski, C. M.; Ev agorou, A.; Kampf, J. W.; Pecoraro, V. L.; Kessissoglou, D. P. Chem. Commun. 2003, 2668. (216) Cavaluzzo, M.; Chen, Q.; Zubieta, J. Chem. Commun. 1993, 131. (217) Koo, B. K.; Lee, U. Bull. Korean Chem. Soc. 2001, 22, 103. (218) Brechin, E. K.; Clegg, W.; Murrie, M.; Parsons, S.; Teat, S. J.; Winpenny, R. E. P. J. Am. Chem. Soc. 1998, 120, 7365. (219) Abu-Nawwas, A. A. H.; Cano, J.; Christia n, P.; Mallah, T.; Rajaraman, G.; Teat, S. J.; Winpenny, R. E. P.; Yukawa, Y. Chem. Commun. 2004, 314. (220) Sellmann, D.; Geipel, F.; Heinemann, F. W. Chem. Eur. J. 2002, 8, 958. (221) Caneschi, A.; Cornia, A.; Fabretti, A. C.; Gatteschi, D. Angew. Chem. Int. Ed. 1999, 38, 1295. (222) Boskovic, C.; Gudel, H. U.; Labat, G. ; Neels, A.; Wernsdorfer, W.; Moubaraki, B.; Murray, K. S. Inorg. Chem. 2005, 44, 3181. (223) Soler, M.; Wernsdorfer, W.; Folting, K.; Pink, M.; Christou, G. J. Am. Chem. Soc. 2004, 126, 2156. (224) Brechin, E. K.; Sanudo, E. C.; Wernsdorfe r, W.; Boskovic, C.; Yoo, J.; Hendrickson, D. N.; Yamaguchi, A.; Ishimoto, H.; Conco lino, T. E.; Rheingold, A. L.; Christou, G. Inorg. Chem. 2005, 44, 502. (225) Weihe, H.; Gudel, H. U. J. Am. Chem. Soc. 1997, 119, 6539. (226) Werner, R.; Ostrovsky, S.; Griesar, K.; Haase, W. Inorg. Chim. Acta 2001, 326, 78. (227) Bogani, L.; Wernsdorfer, W. Nature Mat. 2008, 7 179. (228) Oshio, H.; Nihei, M.; Yoshida, A.; Nojiri, H.; Nakano, M.; Yamaguchi, A.; Karaki, Y.; Ishimoto, H. Chem. Eur. J. 2005, 11, 843. (229) Li, Y. G.; Wernsdorfer, W.; Clerac, R.; Hewitt, I. J.; Anson, C. E.; Powell, A. K. Inorg. Chem. 2006, 45, 2376-2378.

PAGE 279

279 (230) Maheswaran, S.; Chastanet, G.; Teat, S. J.; Mallah, T.; Sessoli, R.; Wernsdorfer, W.; Winpenny, R. E. P. Angew. Chem. Int. Ed. 2005, 44, 5044-5048. (231) Beltran, L. M. C.; Long, J. R. Acc. Chem. Res. 2005, 38, 325. (232) Oshio, H.; Nakano, M. Chem. Eur. J. 2005, 11, 5178. (233) Hamamatsu, T.; Yabe, K.; Towatari, M.; Os a, S.; Matsumoto, N.; Re, N.; Pochaba, A.; Mrozinski, J.; Gallani, J. L.; Barla, A.; Imperia, P.; Paulsen, C.; Kappler, J. P. Inorg. Chem. 2007, 46, 4458-4468. (234) Zaleski, C. M.; Kampf, J. W.; Ma llah, T.; Kirk, M. L.; Pecoraro, V. L. Inorg. Chem. 2007, 46, 1954-1956. (235) Li, D. F.; Clerac, R.; Wang, G. B.; Yee, G. T.; Holmes, S. M. Eur. J. Inorg. Chem. 2007, 1341. (236) Pointillart, F.; Bernot, K.; Sessoli, R.; Gatteschi, D. Chem. Eur. J. 2007, 13, 1602. (237) Mori, F.; Nyui, T.; Ishida, T.; Nogami, T.; Choi, K. Y.; Nojiri, H. J. Am. Chem. Soc. 2006, 128, 1440-1441. (238) Costes, J. P.; Auchel, M.; Dahan, F. ; Peyrou, V.; Shova, S.; Wernsdorfer, W. Inorg. Chem. 2006, 45, 1924-1934. (239) Mishra, A.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. J. Am. Chem. Soc. 2004, 126, 15648. (240) Mishra, A.; Wernsdorfer, W.; Parsons, S.; Christou, G.; Brechin, E. K. Chem. Commun. 2005, 2086. (241) Zaleski, C. M.; Depperman, E. C.; Kampf, J. W.; Kirk, M. L.; Pecoraro, V. L. Angew. Chem. Int. Ed. 2004, 43, 3912. (242) Ma, B. Q.; Gao, S.; Su, G.; Xu, G. X. Angew. Chem. Int. Ed. 2001, 40, 434. (243) Ferbinteanu, M.; Kajiwara, T.; Choi, K. Y.; Nojiri, H.; Nakamoto, A.; Kojima, N.; Cimpoesu, F.; Fujimura, Y.; Takaishi, S.; Yamashita, M. J. Am. Chem. Soc. 2006, 128, 9008. (244) Zaleski, C. M.; Depperman, E. C.; Kampf, J. W.; Kirk, M. L.; Pecoraro, V. L. Inorg. Chem. 2006, 45, 10022. (245) Murugesu, M.; Mishra, A.; Werns dorfer, W.; Abboud, K. A.; Christou, G. Polyhedron 2006, 25, 613.

PAGE 280

280 (246) Osa, S.; Kido, T.; Matsumoto, N.; Re, N.; Pochaba, A.; Mrozinski, J. J. Am. Chem. Soc. 2004, 126, 420-421. (247) Mereacre, V. M.; Ako, A. M.; Clerac, R.; Wernsdorfer, W.; Filoti, G.; Bartolome, J.; Anson, C. E.; Powell, A. K. J. Am. Chem. Soc. 2007, 129, 9248. (248) Kahn, M. L.; Lecante, P.; Verelst, M.; Mathoniere, C.; Kahn, O. Chem. Mater. 2000, 12 3073. (249) Kahn, M. L.; Mathoniere, C.; Kahn, O. Inorg. Chem. 1999, 38, 3692. (250) Costes, J. P.; Dahan, F.; Dupuis, A.; Laurent, J. P. Chem. Eur. J. 1998, 4, 1616. (251) Figuerola, A.; Diaz, C.; Ribas, J.; Tangou lis, V.; Granell, J.; Lloret, F.; Mahia, J.; Maestro, M. Inorg. Chem. 2003, 42, 641. (252) Zhao, H. H.; Lopez, N.; Prosvirin, A.; Chifotides, H. T.; Dunbar, K. R. Dalton Trans. 2007, 878. (253) Chen, W. T.; Guo, G. C.; Wang, M. S.; Xu, G.; Cai, L. Z.; Akitsu, T.; Akita-Tanaka, M.; Matsushita, A.; Huang, J. S. Inorg. Chem. 2007, 46, 2105. (254) Estrader, M.; Ribas, J.; Tangoulis, V.; Sola ns, X.; Font-Bardia, M.; Maestro, M.; Diaz, C. Inorg. Chem. 2006, 45, 8239. (255) Figuerola, A.; Ribas, J.; Llunell, M.; Ca sanova, D.; Maestro, M.; Alvarez, S.; Diaz, C. Inorg. Chem. 2005, 44, 6939. (256) Figuerola, A.; Ribas, J.; Casanova, D.; Maestro, M.; Alvarez, S.; Diaz, C. Inorg. Chem. 2005, 44, 6949. (257) Figuerola, A.; Diaz, C.; Ribas, J.; Tangou lis, V.; Sangregorio, C.; Gatteschi, D.; Maestro, M.; Mahia, J. Inorg. Chem. 2003, 42, 5274. (258) Sheldrick, G. M. SHELXL-97; University of Gttingen, Germany 1997. (259) Sanudo, E. C.; Brechin, E. K.; Boskovic, C.; Wernsdorfer, W.; Yoo, J.; Yamaguchi, A.; Concolino, T. R.; Abboud, K. A.; Rheingold, A. L. ; Ishimoto, H.; Hendrickson, D. N.; Christou, G. Polyhedron 2003, 22, 2267. (260) Wittick, L. M.; Murray, K. S.; Moubaraki, B.; Batten, S. R.; Spiccia, L.; Berry, K. J. Dalton Trans. 2004, 1003. (261) Ako, A. M.; Mereacre, V.; Hewitt, I. J.; Cl erac, R.; Lecren, L.; Anson, C. E.; Powell, A. K. J. Mater. Chem. 2006, 16, 2579.

PAGE 281

281 (262) Shanmugam, M.; Chastanet, G.; Mallah, T.; Sessoli, R.; Teat, S. J.; Timco, G. A.; Winpenny, R. E. P. Chem. Eur. J. 2006, 12, 8777. (263) Mishra, A.; Tasiopoulos, A. J.; Wern sdorfer, W.; Abboud, K. A.; Christou, G. Inorg. Chem. 2007, 46, 3105-3115. (264) Tasiopoulos, A. J.; Mishra, A.; Christou, G. Polyhedron 2007, 26, 2183-2188. (265) Tasiopoulos, A. J.; O'Brien, T. A.; Abboud, K. A.; Christou, G. Angew. Chem. Int. Ed. 2004, 43, 345. (266) Eppley, H. J.; Aubin, S. M. J.; Streib, W. E.; Bollinger, J. C.; Hendrickson, D. N.; Christou, G. Inorg. Chem. 1997, 36, 109. (267) Bagai, R.; Abboud, K. A.; Christou, G. Inorg. Chem. 2008, 47, 621-631. (268) Bencini, A.; Benelli, C.; Caneschi, A. ; Carlin, R. L.; Dei, A.; Gatteschi, D. J. Am. Chem. Soc. 1985, 107, 8128. (269) Bencini, A.; Benelli, C.; Cane schi, A.; Dei, A.; Gatteschi, D. Inorg. Chem. 1986, 25, 572. (270) Kido, T.; Ikuta, Y.; Sunatsuki, Y.; Ogawa, Y.; Matsumoto, N. Inorg. Chem. 2003, 42, 398-408. (271) Hamamatsu, T.; Yabe, K.; Towatari, M.; Os a, S.; Matsumoto, N.; Re, N.; Pochaba, A.; Mrozinski, J.; Gallani, J. L.; Barla, A.; Imperia, P.; Paulsen, C.; Kappler, J. P. Inorg. Chem. 2007, 46, 4458. (272) Chakov, N. E.; Soler, M.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. Inorg. Chem. 2005, 44, 5304. (273) Sessoli, R.; Tsai, H. L.; Schake, A. R. ; Wang, S. Y.; Vincent, J. B.; Folting, K.; Gatteschi, D.; Christou, G.; Hendrickson, D. N. J. Am. Chem. Soc. 1993, 115, 1804. (274) Caneschi, A.; Gatteschi, D.; Sessoli, R. ; Barra, A. L.; Brunel, L. C.; Guillot, M. J. Am. Chem. Soc. 1991, 113, 5873. (275) Rebilly, J. N.; Mallah, T. Struct. Bonding 2006, 122, 103. (276) Berlinguette, C. P.; Vaughn, D.; Canada-Vilal ta, C.; Galan-Mascaros, J. R.; Dunbar, K. R. Angew. Chem. Int. Ed. 2003, 42, 1523. (277) Benelli, C.; Cano, J.; Journaux, Y.; Sesso li, R.; Solan, G. A.; Winpenny, R. E. P. Inorg. Chem. 2001, 40, 188. (278) Price, D. J.; Batten, S. R.; Moubaraki, B.; Murray, K. S. Chem. Commun. 2002, 762.

PAGE 282

282 (279) Koizumi, S.; Nihei, M.; Nakano, M.; Oshio, H. Inorg. Chem. 2005, 44, 1208. (280) Godbole, M. D.; Roubeau, O.; Mills, A. M.; Kooijman, H.; Spek, A. L.; Bouwman, E. Inorg. Chem. 2006, 45, 6713. (281) Ishikawa, N.; Sugita, M.; Wernsdorfe