Citation
Exploration of Environmental Influences on the Properties of Mn12 Single-Molecule Magnets

Material Information

Title:
Exploration of Environmental Influences on the Properties of Mn12 Single-Molecule Magnets
Creator:
Fournet, Adeline D
Publisher:
University of Florida
Publication Date:
Language:
English

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Chemistry
Committee Chair:
CHRISTOU,GEORGE
Committee Co-Chair:
BOWERS,CLIFFORD RUSSELL
Committee Members:
ANGERHOFER,ALEXANDER
BRUCAT,PHILIP J
SIKIVIE,PIERRE

Subjects

Subjects / Keywords:
cluster
magnetism
manganese
molecular

Notes

General Note:
Single-molecule magnets (SMMs) are individual molecules that function as nanoscale magnetic particles. Below a certain blocking temperature, they behave as small permanent magnets, and their properties can be exploited. They have generated great interest in fundamental research on magnetic properties at the nanoscale, and have been suggested for potential applications such as spintronic devices and quantum computing. [Mn12O12(O2CCH3)16(H2O)4] (Mn12-acetate) was the first discovered and, still to date, one of the most well-studied SMMs. Through a well-developed ligand substitution method, it has engendered a large family of Mn12 derivatives with the formula [Mn12O12(O2CR)16(H2O)4], which has been the primary source of current knowledge of the SMM phenomenon. This work focuses on exploring new properties in new or previously known derivatives of the Mn12 family, through various new characterization methods and types of study. In Chapter 2, a new method of ligand substitution to replace the water ligands on the Mn12 core with alcohols is presented. The substitution relies on the exposition of the compound to excess alcohol at elevated temperature. The substitution has minor effects on the magnetism of individual molecules, but the modifications in the crystal packing induce some important modifications in the results of experiments which rely on the alignment of the crystal to a magnetic field. In Chapter 3, the structure of Mn12 derivatives in solution is explored using variable-temperature nuclear magnetic resonance (VT-NMR) spectroscopy on various nuclei (1H, 2D, 19F). A fluxional process between some of the carboxylate and the water ligands is observed, causing the symmetry of the molecule to appear higher in the room-temperature NMR spectrum than in the solid state structure. Using fluorine-containing ligands, the spectra are clear enough that the process can be characterized, and activation parameters obtained. Finally, in Chapter 4, the magnetic properties of a new Mn12 derivative, [Mn12O12(p-FPhCO2)16(H2O)4], are investigated. At low temperature, its magnetic susceptibility is higher than expected for Mn12 derivatives suggesting that weak ferromagnetic intermolecular interactions are present throughout the crystal. This effect relies on the presence of solvent molecules in the lattice, and can be modified upon drying and exposure to air of the crystal.

Record Information

Source Institution:
UFRGP
Rights Management:
All applicable rights reserved by the source institution and holding location.
Embargo Date:
5/31/2018

Downloads

This item has the following downloads:


Full Text

PAGE 1

EXPLORATION OF ENVIRONMENTAL INFLUENCES ON THE PROPERTIES OF M n 12 SINGLE MOLECULE MAGNETS By ADELINE D. FOURNET 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 2016

PAGE 2

2016 Adeline D. Fournet

PAGE 3

To my family and friends for their love and support

PAGE 4

4 ACKNOWLEDGMENTS These six years spent at the University of Florida have been an incredible journey. From the challenge of learning a new language, to the excitement of discovering research, my graduate school years have been filled with eye opening experiences, and unexpe cted events. It feels like yesterday that I stepped off the plane, ready to begin a new adventure. Little did I know how far it would take me, or how much it would change me. It was originally meant to last a year, and guess curiosity was too strong. I am looking forward to the next chapter, but I will certainly miss many aspects of this one, and most importantly many people. The completion of this work would not have been made possible without many of them, and I owe th em all a big thank you. First and foremost, I would like to thank my advisor Pr of George Christou for his guidance through this adventure I appreciate his patience during long discussions, his positive attitude, and his ability to see the best in me and my results and to give me the willpower to carry on during the toughest times. His passion and enthusiasm for research and teaching was and will continue to be an endless source of motivation for me. He will always be a true role m odel for me in my profess ional career and I will always have for him, and for his work ethic, the most profound respect I would also like to thank my committee members, Prof. Alexander Angerhofer, Prof. Russ Bowers, Prof. Philip Brucat, and Prof. Pierre Sikivie for their guidanc e and helpful suggestions regarding this document and my research work Additionally, I would like to extend my gratitude to Dr. Khalil Abboud and his team for solving the crystal structures included in this work, and for helpful discussions about them, as well as Dr. Ion Ghiviriga and Robert Harker for their help with NMR spectroscopy data collection. I would also like to thank my collaborators, Dr. Stephen Hill and his student Thiago Szymanski for the collection and interpretation of EPR

PAGE 5

5 spectroscopy data Dr. Rodolphe Clrac for his help with the magnetism data fitting, and Dr. Wolfgang Wernsdorfer and his student Yanhua Lan for the hysteresis measurements. I would also like to thank all of the Christou group members for six years of friendship and for he lping at various times. I truly believe that our group dynamic was a true contributor to the completion of this work as it made even the hardest of days more enjoyable and helped me carry on. I would like to thank more particularly Annaliese Thuijs, And rew Mowson, and Kylie Mitchell. They are true friends to me, a big part of my moral support team, and their encouragements during all those years have been essential. They also inherited the tedious task of proofreading this document and offering helpful sugg estions for improvement. For the long days in the lab, the hard and the fun ones, for the gatherings around drinks, meals, games or movies, and to the many more years of friendship to come: Cheers guys! Nothing is over for us! Speaking of friends, there ar e a few more to whom I owe a big thank you. I will start with my roommate, Corey, who made my last year so much more enjoyable I will remember the nights at Boca enjoy ing a nice margarita, the episodes of Chopped and Departures, which let us dream of futu re journeys, and Kacey Musgraves in the background for weeks at a time. I would also like to thank the members of my synchronized swimming team, and in particular Shannen, Greg, and Melanie, for the never ending support and stress relief opportunities they provided without realizing it. I want to address a very special thank you to my two best friends, Ccile and Amlie. The kilometers never mattered for us, nor did the time between phone calls. The true friends are the ones with whom you can stay for month s without speaking, and coming back to as if only a day had gone by. I want to thank my parents, for being so supportive of my choice to go study abroad for so long, for the unconditional love they have provided me, for working so hard so that the path I

PAGE 6

6 w anted to take would always be accessible, and for helping me become the person I am today. Nobody ever says thank you enough to their parents, and everybody takes many things from them for granted and I hope to have rectified this, at least partially. Fin ally, I want to address the biggest thank you to my boyfriend, Russ Winkel, for putting up with me, every day and always. Especially for these past few months when, in spite of being so far away, he still was in the front line to deal with my stressed and emotional self. I am looking forward to opening this new page of our lives. "Y'a que les routes qui sont belles, et peu importe o elles nous mnent." J.J. Goldman

PAGE 7

7 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 9 LIST OF FIGURES ................................ ................................ ................................ ....................... 10 LIST OF ABBREVIATIONS ................................ ................................ ................................ ........ 19 ABSTRACT ................................ ................................ ................................ ................................ ... 20 CHAPTER 1 GENERAL INTRODUCTION ................................ ................................ .............................. 22 1.1 Molecules, Clusters, and Nanoparticles ................................ ................................ ............ 22 1.2 Magnetism, Molecular Magnetism, and Single Molecule Magnets ................................ 24 1.3 The Mn 12 Family ................................ ................................ ................................ ............... 32 2 NEW LIGAND SUBSTITUTION METHOD TO REPLACE THE WATER LIGANDS WITH ALCOHOL LIGANDS ON Mn 12 SINGLE MOLECULE MAGNETS ..................... 47 2.1 Background ................................ ................................ ................................ ....................... 47 2.2 Experimental Section ................................ ................................ ................................ ........ 48 2.2.1 Syntheses ................................ ................................ ................................ ................ 48 2.2.2 X Ray Crystallography ................................ ................................ ........................... 51 2.2. 3 DC and AC Magnetometry ................................ ................................ ..................... 51 2.2.4 NMR Spectroscopy ................................ ................................ ................................ 52 2.2.5 Other Studies ................................ ................................ ................................ .......... 52 2.3 Results and Discussion ................................ ................................ ................................ ..... 52 2.3.1 Discussion of the Synthesis ................................ ................................ .................... 52 2.3.2 Description of the Structure ................................ ................................ .................... 53 2.3.3 Magnetometry Studies ................................ ................................ ............................ 55 2.3.3.1 DC studies ................................ ................................ ................................ .... 55 2.3.3.2 AC stu dies ................................ ................................ ................................ .... 56 2.3.4 NMR Spectroscopy ................................ ................................ ................................ 57 2.4 Conclusion ................................ ................................ ................................ ........................ 57 3 CHARACTERIZATION OF A FLUXIONAL PROCESS IN Mn 12 SINGLE MOLECULE MAGNETS THROUGH VT NMR SPECTROSCOPY ................................ 72 3.1 Background ................................ ................................ ................................ ....................... 72 3.2 Experimental Section ................................ ................................ ................................ ........ 74 3.2.1 Synthesis ................................ ................................ ................................ ................. 74

PAGE 8

8 3.2.2 NMR Spectroscopy ................................ ................................ ................................ 75 3.2.3 Structural Analyses ................................ ................................ ................................ 76 3.3 Results and Discussion ................................ ................................ ................................ ..... 76 3.3.1 Synthesis ................................ ................................ ................................ ................. 76 3.3.2 Descriptio n of the Structures ................................ ................................ .................. 76 3.3.3 Room Temperature NMR Spectroscopy ................................ ................................ 76 3.3.4 Hypothesis of a Fluxional Process ................................ ................................ ......... 78 3.3.5 Variable Temperature NMR Spectroscopy ................................ ............................ 78 3.4 Conclusion ................................ ................................ ................................ ........................ 83 4 SUBTLE ENVIRONMENTAL INFLUENCES ON THE MAGNETIC PROPERTIES OF Mn 12 SINGLE MOLECULE MAGNETS ................................ ................................ ..... 103 4.1 Background ................................ ................................ ................................ ..................... 103 4.2 Experimental Section ................................ ................................ ................................ ...... 104 4.2.1 Synthesis of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] (4 2) ................................ ........... 104 4.2.2 X Ray Crystallography ................................ ................................ ......................... 105 4. 2.3 DC and AC Magnetometry ................................ ................................ ................... 107 4.2.4 Other Studies ................................ ................................ ................................ ........ 107 4.3 Results and Discussion ................................ ................................ ................................ ... 108 4.3.1 Discussion of the Synthesis ................................ ................................ .................. 108 4.3.2 Description of the Structure of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ....................... 108 4.3.3 DC and AC Magnetometry ................................ ................................ ................... 110 4. 3.3.1 Magnetic susceptibility of the wet sample ................................ ................. 110 4.3.3.2 Drying and exposure to air: time dependence study ................................ .. 112 4.3.3.3 Evaluation of the strength of the net ferromagnetic interaction ................. 115 4.3.3.4 Hysteresis studies ................................ ................................ ....................... 116 4.4 Conclusion ................................ ................................ ................................ ...................... 116 APPENDIX A ADDITIONAL NMR SPECTRA ................................ ................................ ......................... 138 B OTHER NEW Mn 12 DERIVATIVES ................................ ................................ .................. 144 C TEST OF THE VALIDITY OF A RECENTLY DEVELOPPED METHOD TO ESTIMATE KINETIC PARAMETERS IN WEAK SINGLE MOLECULE MAGNETS 171 D BOND DISTANCES AND ANGLES ................................ ................................ .................. 181 E BOND VALENCE SUM CALCULATIONS ................................ ................................ ...... 185 F PERMISSION TO REPRODUCE COPYRIGHTED MATERIAL ................................ ..... 186 LIST OF REFERENCES ................................ ................................ ................................ ............. 188 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 195

PAGE 9

9 LIST OF TABLES Table page 2 1 Unit cell parameters of crystals of complex 2 4 2 5 and 2 6 ................................ ........... 58 2 2 Crystal data and structure refinement parameters for complex 2 3 ................................ ... 58 2 3 Reduced magnetization fitting parameters ................................ ................................ ......... 59 2 4 Arrhenius parameters ................................ ................................ ................................ ......... 59 3 1 Peak positions, and fitting data for 3 4 ................................ ................................ ............. 85 3 2 Peak positions, and fitting data for 3 6 ................................ ................................ ............. 85 3 3 Peak positions, and fitting data for 3 8 ................................ ................................ ............. 86 3 4 Peak positions, and fitting data for 3 11 ................................ ................................ ........... 86 3 5 Peak positions, and fitting data for 3 7 ................................ ................................ ............. 87 4 1 Crystal data and structure refinement parameters for complex 4 2 8MeCN ................... 118 4 2 Unit cell parameters of crystals of complex 4 2 8MeCN, 4 2 and 4 2 3H 2 O ................ 118 4 3 List of root mean square deviations for analogous atoms between 4 2 and Mn 12 acetate ................................ ................................ ................................ .............................. 119 4 4 List of root mean square deviations for analogous atoms between 4 2 and Mn 12 benzoate ................................ ................................ ................................ ........................... 119 4 5 Distances () between the fluorine atoms and the hydrogen atoms with which they form hydrogen bonds, and corresponding carbon fluorine bond distances. .................... 120 C 1 Effective barriers (K) calculated with the Kramers Kronig method at 1000, 250, and 50 Hz, and through the normal Arrhenius method. ................................ ......................... 173 D 1 Selected interatomic distances () and angles () for [Mn 12 O 12 (O 2 CCH 3 ) 16 (CH 3 OH) 4 2 3 ) ................................ ................................ 181 D 2 Selected interatomic distances () and angles () for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 4 2 ) ................................ ................................ ................. 183 E 1 Bond valence sum calculations and assignment for the Mn ions for [Mn 12 O 12 (O 2 CCH 3 ) 16 (CH 3 OH) 4 2 3 ) a ................................ ............................... 185 E 2 Bond valence sum calculations and assignment for the Mn ions for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 4 2 ) a ................................ ................................ ................ 185

PAGE 10

10 LIST OF FIGURES Figure page 1 1 Magnetic field lines in different situations. A) Magnetic field line profile in vacuum, B) Magnetic field line profile around a diamagnetic substance, C) Magnetic field line profile around a paramagnetic substance. ................................ ................................ .......... 39 1 2 Spin alignments in different types of materials: A) paramagnet, B) ferromagnet, C) antiferromagnet, D) ferrimagnet. ................................ ................................ ....................... 39 1 3 A qualitative plot of m T vs. T for a paramagnet, a ferromagnet, and an antiferromagnet. ................................ ................................ ................................ ................. 39 1 4 Schematic representation of a hysteresis loop for a typical permanent magnet. H is the applied field, M is the magnetization, M s is the magnetization at saturation, M r is the remanent magne tization, and H c is the coercive field of the magnet. .......................... 40 1 5 Relative energies of the m s states for a system with an S = 10 ground state with negative D value in zero field. ................................ ................................ ........................... 40 1 6 Effect of negative zero field splitting on the m s state energies fo r an S = 10 system ........ 41 1 7 Schematic representation of the change in relative energy of the m s states with increasing applied field, showing allowed and forbidden quantum tunneling transitions. ................................ ................................ ................................ .......................... 41 1 8 Structure of [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] (Mn 12 acetate), A) full molecule, B) core of the molecule with Mn and 3 O 2 ions. ................................ ................................ ......... 42 1 9 DC susceptibility vs. tem perature plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] (Mn 12 acetate) in a 0.1 T applied field. ................................ ................................ ......................... 42 1 10 Spin alignment for [Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ] in the ground state that results in a S = 10 system. Mn III are shown in blue, Mn IV in green, and O in red. ................................ ..... 43 1 11 Reduced magnetization plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] (Mn 12 acetate). Fit obtained for g = 1.95 0.0044 and D = 0.43 0.0050 cm 1 ................................ ........... 43 1 12 AC susceptibility plots vs. temperature for [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] (Mn 12 acetate). AC in phase (top) and out of phase (bottom). ................................ .................... 44 1 13 Arrhenius plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] (Mn 12 acetate). The fitting parameters give a n effective barrier U eff /k B = 61 K. ................................ .......................... 44 1 14 Portion of the [Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ] structure showing two Jahn Teller axes in the normal position (Mn III on each side), and one in an abnormal position (central Mn III ) pointing towards an oxide ion. ................................ ................................ ................ 45

PAGE 11

11 1 15 Cyclic voltammogram (top) and differential pulse voltammogram (bottom) for [Mn 12 O 12 (O 2 CCHCl 2 ) 16 (H 2 O) 4 ]. The indicated potentials are vs. ferrocene. .................... 45 1 16 Structure of the [Mn 12 O 12 (O 2 CCHCl 2 ) 16 (H 2 O) 4 ] z core (z = 0 3). Mn II are shown in yellow, Mn III in blue, Mn IV in green, and O in red. ................................ ........................... 46 2 1 Structure of [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ). Mn III are shown in blue, Mn IV in green, O in red, C in grey, O and C of MeOH ligands are shown in pink and black respectively. Hydrogen atoms are omitted for clarity. ................................ ....................... 59 2 2 Packing diagram of [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ). A) Top view of the molecules along the z axis. B) Side view. ................................ ................................ ......... 60 2 3 DC magnetic susceptibility plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ) in an applied 0.1 T field. ................................ ................................ ................................ ............. 61 2 4 DC magnetic susceptibil ity plot for [Mn 12 O 12 (O 2 CC 6 H 5 ) 16 (MeOH) 4 ] ( 2 4 ) in an applied 0.1 T field. ................................ ................................ ................................ ............. 61 2 5 DC magnetic susceptibility plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( t BuOH) 4 ] ( 2 5 ) in an applied 0.1 T field. ................................ ................................ ................................ ............. 62 2 6 DC magnetic susceptibility plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ((CH 3 ) 2 CHOH) 4 ] ( 2 6 ) in an applied 0.1 T field. ................................ ................................ ................................ ........ 62 2 7 Reduced magnetization plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ). Fit obtained for S = 10, g = 1.92 0.0059, and D = 0.43 0.0068 cm 1 ................................ ............. 63 2 8 Contour plot of the fitting error for the reduced magnetization data for [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ). ................................ ................................ ............... 63 2 9 Reduced magnetization plot for [Mn 12 O 12 (O 2 CC 6 H 5 ) 16 (MeOH) 4 ] ( 2 4 ). Fit obtained for S = 10, g = 2.09 0.0048, and D = 0.38 0.0047 cm 1 ................................ ............. 64 2 10 Contour plot of the fitting error for the reduced magnetization data for [Mn 12 O 12 (O 2 CC 6 H 5 ) 16 (MeOH) 4 ] ( 2 4 ). ................................ ................................ .............. 64 2 11 Reduced magnetization plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( t BuOH) x (H 2 O) 4 x ] ( 2 5 ). Fit obtained for S = 10, g = 1.95 0.0080, and D = 0.39 0.0084 cm 1 .............................. 65 2 12 Contour plot of the fitting error for the reduced magnetization data for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( t BuOH) x (H 2 O) 4 x ] ( 2 5 ).. ................................ ................................ 65 2 13 Reduced magnetization plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ((CH 3 ) 2 CHOH) 4 ] ( 2 6 ). Fit obtained for S = 10, g = 1. 98 0.0051, and D = 0.43 0.0058 cm 1 .............................. 66 2 14 Contour plot of the fitting error for the reduced magnetization data for [Mn 12 O 12 (O 2 CCH 3 ) 16 ((CH 3 ) 2 CHOH) 4 ] ( 2 6 ).. ................................ ................................ ... 66

PAGE 12

12 2 15 AC magnetic susceptibility plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ). AC in phase (top) and out of phase (bottom). ................................ ................................ .............. 67 2 16 Arrhenius plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ). The fitting parameters give an effective barrier U eff /k B = 71.9 K and a pre exponential factor 0 = 1 .18 10 8 s. .......... 67 2 17 AC magnetic susceptibility plot for [Mn 12 O 12 (O 2 CC 6 H 5 ) 16 (MeOH) 4 ] ( 2 4 ). AC in phase (top) and out of phase (bottom). ................................ ................................ .............. 68 2 18 Arrhenius plot for [Mn 12 O 12 (O 2 CC 6 H 5 ) 16 (MeOH) 4 ] ( 2 4 ). The fitting parameters give an effective barrier U eff /k B = 71.7 K and a pre exponential factor 0 = 4.46 10 9 s. .......... 68 2 19 AC magnetic susceptibility plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( t BuOH) x (H 2 O) 4 x ] ( 2 5 ). AC in phase (top) and out of phase (bottom). ................................ ................................ ... 69 2 20 Arrhenius plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( t BuOH) x (H 2 O) 4 x ] ( 2 5 ). The fitting parameters give an effective barrier U eff /k B = 75.6 K and a pre exponential factor 0 = 5.83 10 9 s. ................................ ................................ ................................ ........................ 69 2 21 AC magnetic susceptibility plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ((CH 3 ) 2 CHOH) 4 ] ( 2 6 ). AC in phase (top) and out of phase (bottom). ................................ ................................ ......... 70 2 22 Arrhenius plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ((CH 3 ) 2 CHOH) 4 ] ( 2 6 ). T he fitting parameters give an effective barrier U eff /k B = 75.3 K and a pre exponential factor 0 = 5.70 10 9 s. ................................ ................................ ................................ ........................ 70 2 23 1 H NMR spectrum of [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ) in CD 3 CN at room temperature (*: methylene chloride, acetonitrile, and diethyl ether peaks). ...................... 71 3 1 1 H NMR spectrum of Mn 12 acetate ( 3 1 ) in CD 3 CN at room temperature (*: acetic acid and acetonitrile peaks). ................................ ................................ ............................... 88 3 2 1 H NMR spectrum of Mn 12 dichloroacetate ( 3 2 ) in CD 2 Cl 2 at room temperature (*: dichloroacetic acid, acetonitrile, hexanes, methylene chlor ide, and toluene peaks) .......... 88 3 3 1 H NMR spectrum of Mn 12 propionate ( 3 3 ) in CD 2 Cl 2 at room temperature (*: propionic acid, hexanes, methylene chloride, and toluene peaks) ................................ ..... 89 3 4 Highlight of the different types of carboxylates by symmetry in Mn 12 acetate. Water ligands are shown in teal, A type carboxylate ligands are shown in blue, B type in orange, and C type in red and green. ................................ ................................ ................. 89 3 5 Highlight of the different point groups found among Mn 12 derivatives in the solid state: A) Mn 12 acetate with S 4 symmetry, B) Mn 12 benzoate with D 2 symmetry, Mn 12 paramethyl benzoate with C 2 and C 1 symmetry (C and D respectively). ................. 90

PAGE 13

13 3 6 Schematic representation of the seven isomers spanning four point groups available for Mn 12 derivatives with four water ligands. Green dots represent Mn III ions that can have water ligands. ................................ ................................ ................................ ............. 91 3 7 Schematic representatio n of the fluxional process between carboxylate type A and water ligands. ................................ ................................ ................................ ..................... 91 3 8 Mirror planes virtually created by a rapid fluxional process exchanging A type carboxylates and water ligands. ................................ ................................ ......................... 92 3 9 2 H VT NMR spectra of [Mn 12 O 12 (O 2 CCD 2 CH 3 ) 16 (H 2 O) 4 ] ( 3 4 ) in D 10 Et 2 O from 175 to 50 ppm. The chemical shift is in ppm relative to deuterated TMS and the temperature in degrees Celsius. (* = Et 2 O). ................................ ................................ ....... 93 3 10 Low temperature peak positions, fit, and extrapolation for the 2 H VT NMR spectra of [Mn 12 O 12 (O 2 CCD 2 CH 3 ) 16 (H 2 O) 4 ] ( 3 4 ) in D 10 Et 2 O from 175 to 50 ppm. .................. 94 3 11 1 H VT NMR (300 MHz) spectra of [Mn 12 O 12 (NO 3 ) 4 (O 2 CCH 2 C(CH 3 ) 3 ) 12 (H 2 O) 4 ] ( 3 6 ) in CD 2 Cl 2 from 10 to 60 ppm, showing the methylene proton region. ......................... 95 3 12 Low temperature peak positions, fit, and extrapolation for the 1 H VT NMR (300 MHz) spectra of [Mn 12 O 12 (NO 3 ) 4 (O 2 CCH 2 C(CH 3 ) 3 ) 12 (H 2 O) 4 ] ( 3 6 ) in CD 2 Cl 2 from 5 to 10 ppm, showing the tert butyl proton region. ................................ ........................... 96 3 13 19 F VT NMR (470 MHz) spectra of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 3 8 ) in CD 2 Cl 2 from 130 to 30 ppm. ................................ ................................ ........................... 97 3 14 Low temperature peak positions, fit, and extrapolation for the 19 F VT NMR spectra of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 3 8 ) in CD 2 Cl 2 ................................ ....................... 98 3 15 19 F VT NMR spectra of [Mn 12 O 12 (3,5 F 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( 3 11 ) in CD 2 Cl 2 from 170 to 20 ppm. ................................ ................................ ................................ .................. 99 3 16 Low temperature peak positions, fit, and extrapolation for the 19 F VT NMR spectra of [Mn 12 O 12 (3,5 F 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( 3 11 ) in CD 2 Cl 2 ................................ .............. 100 3 17 1 H VT NMR spectra of [Mn 8 Fe 4 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] ( 3 7 ) in MeCN from 50 to 8 ppm. The chemical shift is in ppm relative to TMS and the temperature in degrees Celsius. (* = MeCN and Et 2 O). ................................ ................................ ....................... 101 3 18 Low temperature peak positions, fit, and extrapolation f or the 1 H VT NMR spectra of [Mn 8 Fe 4 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] ( 3 7 ) in MeCN. ................................ ......................... 102 4 1 Structure of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ). A) Full molecule, B) structure of the core with atom labels, C) full molecule with disorder of the ligand s. ................... 121 4 2 Two side views of the structure of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ). The bonds in black show the Jahn Teller axes of Mn III atoms. ................................ .............. 122

PAGE 14

14 4 3 Overlay of the core [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) (pink) with Mn 12 acetate (green) (A), and with Mn 12 benzoate (green) (B).. ................................ .......................... 122 4 4 Visualization of the crystals of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ). A) Picture of a batch of crystals B) Schematic representation of the crystal shape. ............................. 123 4 5 Packing diagram of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 4 2 CN). A) Side view of the crystal packing. The dotted lines represent hydrogen bonding. B) and C) side and top view, respectively, of the crystal packing.. ................................ ...... 124 4 6 Visualization of the hydrogen bonds involving the fluorine atoms on equatorial ligands (F1). ................................ ................................ ................................ ..................... 125 4 7 Visualization of the hydrogen bonds involving the fluorine atoms on axial ligands bridging a Mn III and a Mn IV ion (F2). ................................ ................................ .............. 126 4 8 Visualization of the hydrogen bonds involving the fluorine atoms on axial ligands bridging two Mn III ions (F3). ................................ ................................ ........................... 127 4 9 Schematic representation of the changes in unit cell parameters of 4 2 induced by drying under vacuum and subsequent exposure to air ................................ ..................... 127 4 10 Comparison of DC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ( 4 2 12 acetate in a 0.1 T applied field. ................................ ................. 128 4 11 Comparison of AC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ( 4 2 12 acetate. AC in phase (top) and out of phase (bottom). ............ 128 4 12 Scheme of the interactions leading to the spin alignment between two Mn 12 units linked by a hydrogen bond.. ................................ ................................ ............................ 129 4 13 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ( 4 2 phase (top) and out of phase (bottom). ................................ ......... 130 4 14 Arrhenius plot for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 4 2 fitting parameters give an effective barrier U eff /k B = 59.3 K. ................................ ........... 130 4 15 Reduced magnetization plot for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 4 2 ................................ ................................ ................................ ....................... 131 4 16 Comparison of DC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) at different stages of drying. ................................ ................................ ................................ 131 4 17 Comparison of AC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) at different stages of drying. AC in phase (top) and out of phase (bottom). ...................... 132 4 18 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ). AC in phase (top) and out of phase (bottom). ................................ ................................ 133

PAGE 15

15 4 19 Arrhenius plot for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ). The fitting parameters give an effective barrier U eff /k B = 60.6 K. ................................ ................................ ........ 133 4 20 Comparison of DC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) at different times of exposure to air. ................................ ................................ .................... 134 4 21 Comparison of AC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) at different times of exposure to air. AC in phase (top) and out of phase (bottom). .......... 134 4 22 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 2 O ( 4 2 2 O). AC in phase (top) and out of phase (bottom). ................................ ............. 135 4 23 Arrhenius plot for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 2 O ( 4 2 2 O). The fitting parameters give an effective barrier U eff /k B = 59.6 K. ................................ ..................... 1 35 4 24 Comparison of DC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) exposed to air for 32 hours, and 4 2 re exposed to MeCN for 3 days. ............................ 136 4 25 Comparison of AC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) exposed to air for 32 hours, and 4 2 re exposed to MeCN for 3 days. AC in phase (top) and out of phase (bottom). ................................ ................................ ...................... 136 4 26 Simulation of the DC M T data of 4 2 M T data of Mn 12 benzoate and estimation of the coupling constant. ................................ .......................... 137 4 27 Hysteresis measurement data for [Mn 1 2 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 4 2 visualizing the position of the step. ................................ ................................ .................. 137 A 1 1 H VT NMR spectra of [Mn 12 O 12 (O 2 CCH 2 C(CH 3 ) 3 ) 16 (H 2 O) 4 ] ( 3 5 ) in CD 2 Cl 2 from 10 to 15 ppm, showing the t ert butyl proton region. ................................ ....................... 138 A 2 1 H VT NMR spectra of [Mn 12 O 12 (O 2 CCH 2 C(CH 3 ) 3 ) 16 (H 2 O) 4 ] ( 3 5 ) in CD 2 Cl 2 from 22 to 120 ppm, showing the methylene proton region. ................................ .................... 139 A 3 1 H VT NMR spectra of [Mn 12 O 12 (NO 3 ) 4 (O 2 CCH 2 C(CH 3 ) 3 ) 12 (H 2 O) 4 ] ( 3 6 ) in CD 2 Cl 2 from 5 to 10 ppm, showing the tert butyl proton region. ................................ ............... 140 A 4 19 F VT NMR (470 MHz) spectra of [Mn 12 O 12 ( p CF 3 C 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 3 9 ) in CD 2 Cl 2 from 100 to 0 ppm. ................................ ................................ ............................ 141 A 5 19 F VT NMR (470 MHz) spectra of [Mn 12 O 12 ( o FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 3 10 ) in CD 2 Cl 2 from 140 to 20 ppm. ................................ ................................ .......................... 142 A 6 19 F VT NMR (470 MHz) spectra of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 3 8 ) in CD 2 Cl 2 from 130 to 30 ppm around the coalescence temperature. ............................... 143

PAGE 16

16 B 1 DC magnetic susceptibility plot for [Mn 12 O 12 (3,5 (CH 3 O) 2 C 6 H 3 CO 2 ) 1 6 (H 2 O) 4 ] ( B 2 ) in an applied 0.1 T field. ................................ ................................ ................................ .. 152 B 2 AC susceptibility plots vs. temperature for [Mn 12 O 12 (3,5 (CH 3 O) 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 2 ). AC in phase (top) and out of phase (bottom). ........ 152 B 3 Arrhenius plot for [Mn 12 O 12 (3,5 (CH 3 O) 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 2 ). Fitti ng parameters: U eff /k B = 63.5 K; 0 = 4.80 10 9 s. ................................ ............................... 153 B 4 Reduced magnetization plot for [Mn 12 O 12 (3,5 (CH 3 O) 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 2 ). Fitting parameters: g = 1.9450 0.0014; D = 0.4183 0.0016 cm 1 ........................... 153 B 5 DC magnetic susceptibility plot for [Mn 12 O 12 ( p t BuC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 3 ) in an applied 0.1 T field. ................................ ................................ ................................ ........... 154 B 6 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( p t BuC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 3 ). AC in phase (top) and out of phase (bott om). ................................ ................................ 154 B 7 Arrhenius plot for [Mn 12 O 12 ( p t BuC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 3 ). Fitting parameters: U eff /k B = 67.0 K; 0 = 5.05 10 9 s. ................................ ................................ ................... 155 B 8 Reduced magnetization plot for [Mn 12 O 12 ( p t BuC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 3 ). Fitting parameters: g = 1.9044 0.0017; D = 0.3718 0.0018 cm 1 ................................ ...... 155 B 9 DC magnetic susceptibility plot for [Mn 12 O 12 (2,4,6 (CH 3 ) 3 C 6 H 2 CO 2 ) 16 (H 2 O) 4 ] ( B 4 ) in an applied 0.1 T field. ................................ ................................ ................................ .. 156 B 10 AC susceptibility plots vs. temperature for [Mn 12 O 12 (2,4,6 (CH 3 ) 3 C 6 H 2 CO 2 ) 16 (H 2 O) 4 ] ( B 4 ). AC in phase (top) and out of phase (bottom). ........... 156 B 11 Arrhenius plot for [Mn 12 O 12 (2,4,6 (CH 3 ) 3 C 6 H 2 CO 2 ) 16 (H 2 O) 4 ] ( B 4 ). Fitting parameters: U eff /k B = 64.9 K; 0 = 4.74 10 9 s. ................................ ............................... 157 B 12 Reduced magnetization plot for [Mn 12 O 12 (2,4,6 (CH 3 ) 3 C 6 H 2 CO 2 ) 16 (H 2 O) 4 ] ( B 4 ). Fitting parameters: g = 1.9587 0.0116; D = 0.3797 0.0123 cm 1 ........................... 157 B 13 DC magnetic susceptibility plot for [Mn 12 O 12 ( p ClC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 5 ) in an applied 0.1 T field. ................................ ................................ ................................ ........... 158 B 14 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( p ClC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 5 ). AC in phase (top) and out of phase (bottom). ................................ ................................ 158 B 15 Arrhenius plot for [Mn 12 O 12 ( p ClC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 5 ). Fitting parameters: U eff /k B = 68.0 K; 0 = 1.61 10 9 s. ................................ ................................ ................... 159 B 16 Reduced magnetization plot for [Mn 12 O 12 ( p ClC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 5 ). Fitting parameters: g = 1.6715 0.0042; D = 0.4161 0.0055 cm 1 ................................ ...... 159

PAGE 17

17 B 17 DC magnetic susceptibility plot for [Mn 12 O 12 (3,5 F 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 7 ) in an applied 0.1 T field. ................................ ................................ ................................ ........... 160 B 18 AC susceptibility plots vs. temperature for [Mn 12 O 12 (3,5 F 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 7 ). AC in phase (top) and out of p hase (bottom). ................................ ........................... 160 B 19 Arrhenius plot for [Mn 12 O 12 (3,5 F 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 7 ). Fitting parameters: U eff /k B = 57.9 K; 0 = 4.83 10 9 s. ................................ ................................ ................... 161 B 20 Reduced magnetization plot for [Mn 12 O 12 (3,5 F 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 7 ). Fitting parameters: g = 1.7168 0.0053; D = 0.3108 0.0052 cm 1 ................................ ...... 161 B 21 DC magnetic susceptibility plot for [Mn 12 O 12 ( o FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 8 ) in an applied 0.1 T field. ................................ ................................ ................................ ........... 162 B 22 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( o FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 8 ). AC in phase (top) and out of phase (bottom). ................................ ................................ 162 B 23 Arrhenius plot for [Mn 12 O 12 ( o FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 8 ). Fitting parameters: U eff /k B = 62.9 K; 0 = 3.63 10 9 s. ................................ ................................ .............................. 163 B 24 Reduced magnetization plot for [Mn 12 O 12 ( o FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 8 ). Fitting parameters: g = 1.7206 0.0016; D = 0.3306 0.0016 cm 1 ................................ ...... 163 B 25 DC magnetic susceptibility plot for [Mn 12 O 12 ( p CF 3 C 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 9 ) in an applied 0.1 T field. ................................ ................................ ................................ ........... 164 B 26 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( p CF 3 C 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 9 ). AC in phase (top) and out of phase (bottom). ................................ ........................... 164 B 27 Arrhenius plot for [Mn 12 O 12 ( p CF 3 C 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 9 ). Fitting parameters: U eff /k B = 61.9 K; 0 = 2.02 10 9 s. ................................ ................................ ................... 165 B 28 Reduced magnetization plot for [Mn 12 O 12 ( p CF 3 C 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 9 ). Fitting parameters: g = 1.9450 0.0053; D = 0.4201 0.0061 cm 1 ................................ ...... 165 B 29 DC magnetic susceptibility plot for [Mn 12 O 12 (tma) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 10 ) in an applied 0.1 T field. ................................ ................................ ................................ ...... 166 B 30 AC susceptibility plots vs. temperature for [Mn 12 O 12 (tma) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 10 ). AC in phase (top) and out of phase (bottom). ................................ ......................... 166 B 31 Arrhenius plot for [Mn 12 O 12 (tma) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 10 ). Fitting parameters: U eff /k B = 52.11 K; 0 = 3.50 10 9 s. ................................ ................................ ................. 167 B 32 Reduced magnetization plot for [Mn 12 O 12 (tma) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 10 ). .............. 167 B 33 DC magnetic susceptibility plot for [Mn 12 O 12 (add) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 11 ) in an applied 0.1 T field. ................................ ................................ ................................ ........... 168

PAGE 18

18 B 34 AC susceptibility plots vs. temperature for [Mn 12 O 12 (add) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 11 ). AC in phase (top) and out of phase (bottom). ................................ ......................... 168 B 35 Arrhenius plot for [Mn 12 O 12 (add) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 11 ). Fitting parameters: U eff /k B = 63.31 K; 0 = 5.07 10 9 s. ................................ ................................ ................. 169 B 36 DC magnetic susceptibility plot for [Mn 12 O 12 (iso) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 12 ) in an applied 0.1 T field. ................................ ................................ ................................ ........... 169 B 37 AC susceptibility plots vs. temperature for [Mn 12 O 12 (iso) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 12 ). AC in phase (top) and out of phase ( bottom). ................................ ......................... 170 B 38 Arrhenius plot for [Mn 12 O 12 (iso) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 12 ). Fitting parameters: U eff /k B = 50.52 K; 0 = 1.14 10 8 s. ................................ ................................ ................. 170 C 1 ln( / ) versus 1/T plots for [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ]. ................................ ............ 175 C 2 ) versus 1/T plots for 2 3 ................................ ................................ ..................... 175 C 3 ) versus 1/T plots for 2 4 ................................ ................................ ..................... 176 C 4 ) versus 1/T plots for 2 5 ................................ ................................ ..................... 176 C 5 ) versus 1/T plots for 2 6 ................................ ................................ ..................... 177 C 6 ) versus 1/T plots for 4 2 ................................ ................................ ....... 177 C 7 ) versus 1/T plots for 4 2 ................................ ................................ ..................... 178 C 8 ) versus 1/T plots for 4 2 2 O ................................ ................................ ........... 178 C 9 ) versus 1/T plots for B 2 ................................ ................................ .................... 179 C 10 ) versus 1/T plots for B 3 ................................ ................................ .................... 179 C 11 ln( / ) versus 1/T plots for B 11 ................................ ................................ .................. 180

PAGE 19

19 LIST OF ABBREVIATIONS AC Alternating current BVS Bond valence sum DC Direct current FTIR Fourier transform infrared spectroscopy HFEPR High frequency electron paramagnetic resonance spectroscopy MeCN Acetonitrile MeOH Methanol NMR Nuclear magnetic resonance PS II Photosystem II QTM Quantum tunneling of magnetization RM Reduced magnetization RMS Root mean square SMM S ingle molecule magnet SQUID Superconducting quantum interference device ZFS Z ero field splitting

PAGE 20

20 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 EXPLORATION OF ENVIRONMENTAL INFLUENCES ON THE PROPERTIES OF Mn 12 SINGLE MOLECULE MAGNETS By Adeline D. Fournet M a y 2016 Chair: George Christou Major: Chemistry Single molecule magnets (SMMs) are individual molecules that function as nanoscale magnetic particles. Below a certain blocking temperature they behave as small permanent magnets, and their properties can be exploited. T hey have generated great interest in fundamental research on magnetic properties at the nanoscale, and have been suggested for potential a pplications such as spintronic devices and quantum computing. [ Mn 12 O 12 ( O 2 CCH 3 ) 16 (H 2 O) 4 ] (Mn 12 acetate) was the first discovered and, still to date, one of the most well studied SMM s Through a well developed ligand substitution method, it has engendered a large family of Mn 12 derivatives with the formula [ Mn 12 O 12 ( O 2 CR ) 16 (H 2 O) 4 ] which has been the primary source of current knowledge of the SMM phenomenon. This work focuses on exploring new propert ies in new or previously known derivatives of the Mn 12 family, through various new characterization methods and types of study In Chapter 2, a new method of ligand substitution to replace the water ligands on the Mn 12 core with alcohols is presented. The substitution relies on the exposition of the compound to excess alcohol at elevated temperature. The substitution has minor effects on the magnetism of individual molecules, but the modifications in the crystal packing induce some important

PAGE 21

21 modifications in the results of experiments which rely on the alignme nt of the crystal to a magnetic field. In Chapter 3 t he structure of Mn 12 derivatives in solution is explored using variable temperature nuclear magnetic resonance (VT NMR) spectroscopy on various nuclei ( 1 H, 2 D, 19 F) A fluxional process between some of the carboxylate and the water ligands is observed, causing the symmetry of the molecule to appear higher in the room temperature NMR spectrum than in the solid state structure. Using fluorine containing ligands, the spectra are clear enough that the proces s can be characterized, and activation parameters obtained. Finally, i n Chapter 4 the magnetic properties of a new Mn 12 derivative, [ Mn 12 O 12 ( p FPhC O 2 ) 16 (H 2 O) 4 ] are investigated. At low temperature, its magnetic susceptibility is higher than expected for Mn 12 derivatives suggesting that weak ferromagnetic intermolecular interactions are present throughout the crystal. This effect relies on the presence of solvent molecules in the lattice, and can be modified upon drying and exposure to air of the crystal.

PAGE 22

22 CHAPTER 1 GENERAL INTRODUCTION 1.1 Molecules, Clusters, and Nanoparticles 19 60 s in reference to compounds containing several metal centers and substantial metal me tal bonding 1 Since then, its use has broadened, and it is now commonly employed to describe multinuclear compounds containing metal centers interacting either directly throug h metal metal bonds, or indirectly through ligands. 2 3 While a molecule containing two metal ions can strictly a high number of metal ions, or a particle small en ough that the number of atoms inside is used as a measure of the size rather than macroscopic properties such as volume. The field of clusters and nanoparticles is thus considered to be covering the gap between small molecules and bulk materials of small s izes. 4 Although the definition of the term remains broad in the literature, it is defined particle (in shape, size, chemical formula etc. d characteristics (size distribution) although the properties c an be just as precisely known. 4 In inorganic chemistry, two types of clusters are commonly defined: metal and metal rich clusters. Metal clusters are compounds containing mainly metal atoms directly bound to each other through metal metal bonds. Me tal clusters are the direct low size analogues of bulk metals. In metal rich clusters, me tal ions are bridged by non metallic elements such as oxygen, nitrogen, or phosphorus. Metal oxide clusters in particular are of increasing interest due to the relevance of their bulk analogues to many fields (catalysis, dev ices, sensors ). 5 Increasing efforts have been made over the last twenty years to expa nd our knowledge of the synthetic methods towards and properties of these clusters Several important questions are raised with the emergence of this

PAGE 23

23 new field: to what extent will the physical properties be retained upon reduction of a compound's size from bulk to molecula r level? Will the inner structur e be modified? Will the cluster show new properties such as quantum effects due to its small size? 6 These questions became particularly interesting when the size of molecular clusters began to reach the size of the smallest nanoparticles known. 7 8 Many theoretical studies have been conducted in order to predict physical propert ies, in particular for these big molecular clusters which straddle the gap between small molecules and nanoparticles 9 The search for new transition metal oxide clusters is motivated by a desire to bridge the worlds of small molecules and nanoparticles, but also to develop a way of better understanding Nature. Indeed, clusters containing metals such as iron, cobalt, or manganese are commonly found in natural systems and are essential to life processes. 10 The ability of transition metals to reach several oxidation states, and to adopt diverse topologies and a wide range of coordination number s has favored their incorporation into many biomolecules, in particular proteins and enzymes Functions such as charge tra nsport, oxygen transport, or catalysis are often carried out at a metal site of a protein. Among these important metalloproteins, several contain m etal oxo clusters in particular with metals like manganese, iron, or copper. One of the most important example s can be found in photosystem II which is the protein complex responsible for water splitting and the generation of oxygen in plants, alg ae and cyanobacteria. The reaction is photocatalysed by a [Mn 4 CaO 5 ] cluster known as the oxygen evolving complex (OEC). Although a precise structure of the cluster was obtained only recently 11 12 the use of spectroscopic methods had long ago given strong evidence supporting the hypothesis that it comprise s a manganese calcium oxo cube, and many attempts have been made to synthesiz e an artificial model. 13 It is always a synthetic challenge to design new molecular clusters with the

PAGE 24

24 desired properties or topologies, b ut also an opportunity to discover new structures with beautiful topologies, and new behaviors. 1.2 Magnetism, Molecular Magnetism, and Single Molecule Magnets Magnetic materials have been known for over 2500 years since the discovery of magnetite (Fe 3 O 4 ) as rocks capable of attracting iron or each other 14 In the early days, these materials were used in simple devices for purposes such as surgery, navigation, or astronomy. However, it wasn't until the 19 th century that the scientific community started understand ing more complex magnetic behaviors such as the relationship between electricity and magnetism. The work of scientis ts lik e Faraday and Maxwell expa nded our knowledge of electromagnetism, and set the foundations for the research of the 20 th and 21 st centuries The discovery of the Zeeman effect and the emergence of quantum mechanics at the beginning of the 20 th century led to the idea of exchange interactions, and to a better understanding of the origin of magnetism. Magneti sm arises f rom the motion of electrons around the nucleus, and around their own axes. The angular momentum created by the motion causes them to individually behave as tiny magnets. One can distinguish two main types of materials: diamag netic and paramagnetic Diamagnetism is a property of all matter, it comes from the presence of paired electrons and gives materials the ability to weakly repel magnetic fields. Paramagnetism, on the other hand, is a much stronger effect which comes from the presence of unpaired electrons. Paramagnetic materials have a tendency to attract magnetic fields (Figure 1 1). Am ong paramagnets, some materials have the ability to orient their electron s' intrinsic magnetic moments parallel to each other, and to conserve this orientation after removal of the applied magnetic field. This effect results from internal interactions whic h cause the overall energy of the material to be lower when the electrons' magnetic moments are parallel Two types of such interactions between electron spins are possible: ferromagnetic, where the magnetic

PAGE 25

25 moments align parallel and in the same direction and antiferromagnetic, where the magnetic moments align parallel but in opposite directions. This results in three main types of magnetic materials: ferromagnetic materials, in which the alignment of spins results in a net magnetic moment antiferromagne tic mate rials, in which the anti parallel alignment of spins of the same magnitude results in a complete cancellation of the magnetic moment and ferrimagnetic materials in which antiferromagnetic interactions between spins of different magnitude s result i n a non zero net magnetic moment (Figure 1 2). To distinguish these behaviors, one can measure the response of the material when expose d to an external magnetic field: the magnetization. At a given temperature, the magnetization of a sample depends on the sum of the magnetic moments of each state of the sample weighted with its population. For example, for a simple paramagnet with S = 1/2 and no interaction between spins, the magnetization ( M ) is given by eq 1 1 known as the Curie law, where N A is Avogadro' s number, g is the electron g factor, B is the Bohr magneton, H 0 is the applied external magnetic field, k B is the Boltzmann constant, and T is the temperature. (1 1) The volume magnetic susceptibility v is introduced as a measure of the magnitude of the magnetization in an applied field, and is given by eq 1 2. (1 2) It is, however, more convenient to measure the molar magnetic susceptibility M which is obtained with eq 1 3 where M w is the sample's molecular weight, and d its density (1 3)

PAGE 26

26 The differences in type s of interactions within a material can be observed in a plot of M T vs. temperature ( T ) as seen in Figure 1 3 Ferromagnetic interactions cause the M T signal to increase as T is decreasing, whereas antiferromagnetic interactions cause the M T signal to decrease as T is decreasing. Paramagnets exhibit no significant change in the M T value throughout the temperature range. Above a certain temperature, al l materials behave as paramagnets as t he interactions between spins are overcome by thermal energy. As the temperature gets reduced, the interactions become dominant, and the spins align according to them. The temperature below which the spin alignments a re observable depends on the strength of the interactions, or coupling between spins. It is commonly understood that magnets can be divided into three main categories relative to their size, composition, and the orig in of their magnetic properties: bulk ma gnets, molecule based magnets, and single molecule magnets. Bulk magnets were the first discovered, and to this date still the most widely used type of magnets. A bulk magnet is typically composed of a large 3 dimensional array of atoms (metal metal alloy s metal oxides etc ) with interactions throughout the material creating long range ordering. To minimize the internal energy of the material, magnetic domains form and are present in the unmagnetized state. In e ach domain, all the magnetic moments are aligned according to the i nteraction between spins. However, t he orientation of the magnetization between domains is random, which creates an overall zero net magnetization. When an external field is applied to the material, the magnetization of all the do mains tend s to align parallel to the field and results in a macroscopic magnetization for the material. Below a certain temperature, this orientation can be retained when the external field is turn ed off, which gives a magnetized material. It is then requi red to apply a field in the opposite direction to bring the magnetization back to zero. The intensity of the field required to

PAGE 27

27 demagnetize the m aterial corresponds to its coercivity, the value of which depend s on the temperature and the nature of the mater ial. The coercivity of a materi al ( H c ) along with its saturation magnetization ( M s ) and its remanent magnetization ( M r ) are important properties that characterize a magnet. They can be observed in a magnetization vs field plot known as a hysteresis loop (Figure 1 4). Although bulk magnets still count as a major part of the magnetic material industry, the y present a few disadvantages. T hey often require very h arsh synthetic conditions, use chemical elements for which the supply is limited, and can be brit tle and chemically reactive. These limitations, along with the increasing demand for devices requiring magnetic materials, led the research community to increase its interest in molecul e based magnets. 15 The se differ from bulk magnets by their structural building blocks which are molecular in nature, and their low temperature solution based synthetic conditions which enables chemical tailoring of the molecular structure. The precursors used for the synthesis of molecule based magnets are often coordination compounds, or more rarel y purely organic molecules. 15 To form the materia l, the precursors react with each other and covalent bonds are formed between them to obtain a network of molecular building blocks. At the macroscopic level, the magnetic properties of molecule based magnets resemble those of the b ulk magnets, but they arise from different mechanisms: in bulk magnets, the overlap of the magnetic orbitals (d or f orbitals) of the metals either directly (metals or metal alloys) or through bridging oxides (metal oxide materials) allows for direct excha nge or superexchange respectively. In most molecule based magnets, however, the interaction between spins occurs via the connecting ligands or through space (dipolar coupling). Examples of molecule based magnets include the famous Prussian Blue and its analogues.

PAGE 28

28 The focus of this dissertation is the third type of magnets: single molecule magnets (SMMs) These materials are individual independent molecules that can function as nanoscale magnet s below their blocking tempe rature Their magnetic properties are intrinsic to the molecule instead of relying on extended interactions throughout a network like in the other two types. They are often molecules containing 3d, 4f, or 3d/4f metal oxide cores surrounded by a shell of di amagnetic organic ligands. The magnetic bistability observed in the ir hysteresis loops arises from a combination of a large ground state magnetic moment and negative magnetoaniso tropy as described in Figure 1 5 which displays the barrier to reversal of ma gnetization for an S = 10 system The size and dimension decrease from 3 D materials to 0 D single molecule magnets allows for the replacement of energy bands by discrete energy levels and the appearance of quantum effects such as quantum tunneling of mag netizatio n 4 They are thus often called quantum magnets. The origin of the slow m agnetization relaxation displayed by SMMs depends on the nature of the metal included in their structure. For the purpose of this work, we will discuss SMMs which contain only 3d tran sition metals. As stated previous ly the bistability of SMMs comes from t he energy barrier to the reversal of magnetization created by a combination of the ground state spin S and a negative magnetoanisotropy. In 3d transition metals, magnetoanisotropy can be gauged by the axial zero field splitting parameter D and it arises from an anisotropic distribution of the valence electrons around the metal ions created by a lowering in the symmetry of the ligand arrangement around the metal. The effect of zero field splitting is to remove the degeneracy of m s states with different ab solute m s values. The energy given to a particular m s state from zero field splitting is described in eq 1 4. (1 4)

PAGE 29

29 Negative zero field splitting brings the states of higher m s to lower energy, and the states of lowest m s to the highest energy level as shown on Figure 1 6 The lowest m s is 0 or 1/2 for integer or half integer spin systems respectively. E q 1 4 then allows the height of the potential energy barrier to the reversal of magnetization from to to be calculated This barrier is defined as for integer spin systems, and for half integer spin systems. It should be noted that this is a theoretical thermal barrier which doesn't take into account any relaxation processes other t han thermal excitation from the to the lowest m s state to allow the magnetization to relax to the state. At very low temperature, quantum tunneling of magnetization becomes a non negligible relaxation process that will reduce the effective barrier below the maximum value given by the above equations Experimental evidence for the slow magnetization re laxation of SMMs at low temperature can be observed in AC magnetic susceptibility measurements where a weak oscillating AC field is applied to the sample in the absence of a DC static field. It allows dynamic information on the magnetization of a sample t o be obtained The AC magnetic susceptibility measurement yields two quantities: the magnitude of the susceptibility, and the phase shift, From these two quantities, one can obtain the in phase component and the out of phase component (eq 1 5 and 1 6). (1 5) (1 6) In the high temperature limit, the real component of the AC susceptibility ( ) follows the DC susceptibility, and the imaginary component ( '' ) is equal to zero. As the temperature decreases, less thermal energy is available to participate in the magnetization relaxation process Below a certain temperature, the magnetic moment can no longer stay in phase with the

PAGE 30

30 oscillating magnetic field, which is seen as a drop in the in phase M and a peak in the out of phase '' M This phenomenon is also frequency dependent. The observation of a frequency dependent drop in the in phase and a concomitant rise in the out of phase signals is the first indication that a sample may have SMM properties, which can t hen be confirmed by a magnetization vs. field measurement to look for hysteresis The position of the maxim um of the out of phase peak indicates the temperature at which the rate of the magnetic moment reversal is equal to the oscillating frequency of the AC field. The magnetization relaxation of a single molecule magnet obeys the Arrhenius relationship characteristic of a thermally activated Orbach process This relationship is described by eq 1 7 where is the relaxation time, and U eff is the effective e nergy barrier to the reversal of magnetization. (1 7) For a single molecule magnet, magnetization vs. field measurements yield hysteresis loops below its blocking temperature, similar to ones that can be observed for other types of magnets. S ingle molecule magnet s, however, can be distinguished from other typ es of magnets by the presence of steps in their hysteresis loops. These steps arise from the phenomenon of quantum tunneling which can occur when two m s states are degenerate, as shown on Figure 1 7 (i.e. the magnetization reverses by tunneling through the barrier rather than going over the top) This phenomenon was first discovered in the [Mn 12 O 12 ( O 2 CCH 3 ) 16 (H 2 O) 4 ] molecule, and its properties will be detailed later. 16 Magnet ic materials have been important elements of advancing technology for many centuries. After the industrial revolution, and even more so during the 20 th century, their use greatly incr eased, and they became essential pieces of many devices. In our modern society, applications using magnets are ubiquitous and can be found i n a variety of industries, from high

PAGE 31

31 technology domains such as information techno logy or transport to simple consu mer products. The global market for magnetic materials is estimated to be worth over $30 billion, and will keep increasing over the next few decades. 17 The miniaturization of electronic devices has driven the field of magnetic material s research towards the nanoscale. With increasing demand for smaller and smaller magnets, the scientific communit y started looking for new methods to decrease the size of materials while retaining their physical properties. The two methods that can be employed to generate micro to nano scale materials are commonly known as "top down" and "bottom up". The "top down" a pproach consists of using bulk materials, and decreasing their sizes to give smaller units This method is commonly used in the synthesis of nanoparticles. It unfortunately often results in products with a wide or at least significant distribution of shape s and sizes which can be a particularly major problem at the nanoscale where physical properties can be highly dependent on these parameters. 18 It also often requires somewhat harsh conditions in order to decrease the size of the material. The "bottom up" approach, on the other hand uses metal salts and ligands as starting materials to build up a targeted structure. This method can be used to synthesize both nanoparticles or molecular materials such as single molecule magnets. It has the advantage of yielding narrow size distribution products (in the case of nanoparticles), or even truly monodisperse collections of particles (in the case of molecular materials). For single molecule magnets, the syntheses are usually carried out at room temperature or with moderate heating, in solutions open to the atmosphere. The crystallization step acts as purification and yields a crystalline material in which every unit is identical and (usually identically oriented ) and therefore has homogeneous properties. Single molecule magnets are composed of a magnetic core (typically metal and oxide ions ), surrounded by a shell of organic ligands that prevent any

PAGE 32

32 major interaction between cores. This organic shell also incre ases the solubility of SMMs in organic solvents, and can be altered at will to modify properties or optimize the SMMs for particular studies Combined with the quantum properties, these advantages make single molecule magnets ideal candidates for new applications such as quantum computing and molecular spintronic devices and they have been proposed as such 19 1.3 The Mn 1 2 Family The first discovered SMM, and still the most well studied to this date, is [Mn 12 O 12 ( O 2 C CH 3 ) 16 (H 2 O) 4 ], commonly referred as Mn 12 acetate. It was first synthesized in 1980 20 but was not character ized magnetically until 1991. 21 Its synthesis involves a comproportionation reaction between Mn II from Mn(OAc) 2 and Mn VII from KMnO 4 in a 60% aqueous acetic acid solution. The resulting product is a mixed valent complex containing eight Mn III and four Mn IV for an average oxidation state of +3.33 (eq 1 8 ) 44Mn 2+ + 16Mn 7+ 5Mn 3+ 8 Mn 4+ 4 (1 8 ) The structure of Mn 12 acetate can be found in Figure 1 8. The core of the molecule is made of a central [Mn IV 4 O 4 ] cube surrounded by a star shaped ring of eight Mn III bridged to the cube by eight 3 O 2 ions The ligand sphere is comprised of sixteen acetate ligands, eight of them in a xial positions, and eight in equatorial position s along with four water ligands. The latter are positioned on the axial site of every other Mn III alternating sides relative to the disc defined by the ring of Mn III ions Overall, the molecule has S 4 symmetry. The sixteen acetate ligands can be sorted in three types according to their position s relative to the Mn III and Mn IV and whether or not the y lie on Jahn Teller axes Four of the axial acetate ligands have both their oxygens on Jahn Teller ax es of Mn III four have one of their oxygens on a Mn III Jahn Teller axis, and the other on a Mn IV and the eight equatorial ligands have both their oxygens on Mn III non Jahn Teller sites The four water ligands also lie on Mn III Jahn Teller axes The positi on of the

PAGE 33

33 ligands relative to the Jahn Teller axis also defines their lability and therefore the order in which they will exchange in substitution reaction s ( vide infra ). It was discovered early on that analogues of the Mn 12 acetate with the same core but different carboxylate ligands could be synthesized. Some Mn 12 derivatives were synthesized directly from Mn II and Mn VII starting materials 22 but a more efficient method relying on ligand substitution was rapidly developed providing a convenient route to Mn 12 derivatives with a variety of carboxylate ligands. 23 This method is described for a generic carboxylate in eq 1 9 [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] + 16 RCOOH [ Mn 12 O 12 ( O 2 CR ) 16 (H 2 O) 4 ] + 16 CH 3 COOH (1 9 ) Mn 12 acetate is placed in an excess of the carboxylic acid RCOOH, and allowed to react for a few hours. The reduced boiling point of the toluene/acetic acid azeotrope is then used to remove the acetic acid that forms, and therefor e push the equilibrium towards the products. The rate of the forward reaction, and the equilibrium constant depend on the relative acid strength of acetic acid and the new carboxylic acid : for acids significantly stronger than acetic acid, the equilibrium constant is substantial and full substitution can be achieved using only one addition of excess RCOOH followed by removal of the acetic acid by product. To achieve full substituti on with weaker carboxylic acids however, multiple additions of the new acid along with multiple cycles of acetic acid removal are often necessary. Substitution of the carboxylate ligands whose oxygen atoms lie on Jahn Teller axes is facilitated by the longer (and therefore weaker) Mn O bond s Using this property, site selective s ubstitution can be achieved when using a smaller amount of incoming carboxylic acid. 24 The substitution happens with priority for the axial ligands with both their oxygen atoms on Jahn Teller axes, followed by the axial ligands with any one of their oxygen atoms on a Jahn Teller axis, and the equatorial ligands exchange

PAGE 34

34 last. Additionally, it has been shown that, in cases where more than one type of carboxylic acid was present, the most basic carboxylate preferentially binds to the Mn 12 core in the eq uatorial position s where it can form a shorter and stronger bond with the manganese. These substitution methods provide an exceptional level of control over the ligand shell of Mn 12 derivatives, and allowed for the creation of a very large Mn 12 family. It is, to this date, the most well studied families of SMMs, and has provided a large amount of the current knowledge on the SMM phenomenon Mn 12 derivatives usually have very similar magnetic properties In this I ntroduction chapter, the magnetic properties of Mn 12 acetate will be described, and it will be assumed that the properties of the other derivatives are identical, unless otherwise stated. Mn 12 acetate possesses a ground state spin of S = 10 which arises from intramolecular exchange interactions betw een the manganese ions. The DC susceptibility vs. temperature plot is shown in Figure 1 9. A close observation and analysis of that plot allows one to rationalize the relative alignment of each Mn ion spin which results in the S = 10 ground state. The Mn 12 core possesses three types of interactions: the Mn IV Mn IV interaction which is ferromagnetic, and the Mn III Mn IV and Mn III Mn III interactions which are both antiferromagnetic. A complete calculation of the ener gy levels defined by the Hamiltonian was carried out and yielded the values of the coupling constants for all three interactions. 25 The Mn III Mn IV interaction is the strongest, and it is responsible for the decrease in M T with decreasing temperature between 30 0 and 150 K. It forces every Mn III spin to align antiparallel with respect to any Mn IV spin with which they interact. The ferromagnetic Mn IV Mn IV interaction is the second strongest, and it forces the spins of the Mn IV ions of the central cube to align par allel, and therefore forces the spins of all of the Mn III ions in the outer ring to align parallel with one another. This interaction is responsible for the steep rise in M T

PAGE 35

35 between 150 and 15 K. The antiferromagnetic Mn III Mn III interaction is much weaker and cannot compete with the Mn III Mn IV interaction to force an antiparallel spin alignment of the Mn III spin s. Such a phenomenon is called magnetic frustration and is quite common in antiferromagn e tic systems. 26 28 Overall, the parallel alignment of four S = 3/2 Mn IV spins in the opposite direction to eight S = 2 Mn III spins explains the overall S = 10 ground state of the molecule (Figure 1 10). While the DC susceptibility data above 15 K is very similar across Mn 12 derivatives, the low temperature data can be very dependent on the type of carboxylate ligand, and more generally on the close environment of the molecule (crys tal packing, lattice solvents ). It usually shows a small drop in M T which is commonly attribut ed to a combination of Zeeman and zero field splitting effects. The maximum in M T before the low temperature drop usually has a value between 50 and 55 cm 3 .K.mol 1 which is the spin only value expected for an S = 10 system with g slightly lower than 2 .0 (eq 1 10) (1 10) The ground state can be confirmed by a fit of reduced magnetization data ( M/N A B ) plotted vs. field over temperature ( H/T ). This fit also gave values for the zero field splitting parameter D and the electron g factor of the molecule, respectively 0.5 cm 1 and 1.9 (Figure 1 11). 21 The AC susceptibility plots for Mn 12 acetate are shown in Figure 1 1 2. An extrapolation of the in phase data before the frequency dependent drop to 0 K gives a M 'T value around 53.5 cm 3 .K.mol 1 which is again in agreement with an S = 10 system with g slightl y lower than 2.0. The frequency dependent drop in the in phase data along wi th the frequency dependent peak in the out of phase data were the first proof of a frequency dependent freezing of the magnetization, and the first evidence that the material is a permanent magnet below a certain

PAGE 36

36 blocking temperature. 21 The relaxation times obtained from the AC data were found to follow an Arrhenius law for a thermally activated process and the effective energy barr ier to the reversal of magnetization ( U eff /k B ) was calculated to be equal to 61 K (Figure 1 13) 29 It was later discovered that for certain Mn 12 derivatives, a second isomer relative to the orientation of the Jahn Teller axes with a lower energy barrier was able to be formed, either by synthesis, or by r earrangement of the crystal lattice upon loss of solvent. 30 31 U sually, the wet material shows the higher temperature peak only which corresponds to the higher relaxation barrier because all of the Mn III Jahn Teller axes are nearly parallel and along the z axis of the molecule. Upon drying, the solvent molecules leaving the crystal lattice often cause a small rearrangement, and some of the Jahn Teller axes rotate 90 (Figure 1 14) therefore lowering the total anisotropy of the molecule and speeding up quantum tunneling rates, both decreasing the effective energy barrier. The AC data then shows a second peak at lower temperature. In most cases, the dry material contains both isomers and therefore both peaks are present in the AC data with var ying ratios. In some rare cases the low temperature peak alone can be observed in w et materials which crystallize with the "abnormal" orientation of some of the Jahn Teller axes 31 Mn 12 acetate was the first single molecule magnet on which was measured magnetization vs. field data to observe magnetic hysteresis. It provided the first macroscopic observat ion of quantum tunneling of magnetization in the form of steps in the hysteresis loop. 16 Th e se steps occur at regular intervals of applied field when two m s states on opposite sides of the energy barrier are degenerate. The separation between the steps therefore gives a measure of the difference in energy between m s states, which is directly related to the zero field splitting parameter D, as described earlier (eq 1 4).

PAGE 37

37 In addition to its fascinating magnetic properties, the Mn 12 family displays very rich redox chemistry involving several oxidation and reduction proc esses. Many Mn 12 derivatives show a reversible oxidation, and up to two or three reversible reductions. The redox properties of the Mn 12 cluster are very dependent on the carboxylate ligand and its ability to interact electronically with the metals within the core. For example, Mn 12 acetate displays two reversible reduction processes, and a third irreversible one, whereas the dichloroacetate derivative shows three reversible reductions followed by a fourth irreversible one (Figure 1 15 ) 32 33 This phenomenon is attributed to the electron withdrawing ability of the dichlo roacetate ligand lower ing the electron density around the Mn III ions of the core, in turn lowering the energy of the reduced Mn 12 core. Using Mn 12 dichloroacetate, it was possi ble to chemically reduce the Mn 12 core by up to three electrons and to isolate and crystallize the reduced derivatives [Mn 12 ] z with z = 1 3. 34 In each case, the reduction was found to be localized to the outer ring Mn III ions, leaving the central Mn IV ions unaffected (Figure 1 16 ) The isolation of these compounds provided a unique opportunity to study the magnetic properties of a manganese single molecule magnet at different oxidation levels It should be noted that the SMM properties were retained upon reduction of the Mn 12 core, even though the reduction of Mn III to Mn II caused a decrease in axial anisotropy due to the disappearance of Jahn Teller elongations, thus decreasing the energy barrier to the reversal of magnetization. The Mn 12 family is the oldest, and to date the most studied family of single molecule magnets. Yet, t he synthesis of new derivatives and the use of new techniques constantly result in the discovery of new behaviors and properties, mak ing this family of SMM s still one of the most prominent in the field. In Chapter 2, a new ligand substitution method allowing the replacement of the water ligands around the Mn12 with alcohol ligands will be presented. In Chapter 3 a

PAGE 38

38 detailed analysis of a very unique fluxional process occurring in solution between carboxylate and water ligands will be described In Cha pter 4 the magnetic properties of new Mn 12 derivatives will be investigated, along with interesting effects caused by solvent of crystallization and weak intermolecular interaction s

PAGE 39

39 A B C Figure 1 1. Magnetic field lines in different situations A) M agneti c field line profile in vacuum, B ) M agnetic field line profile around a diamagnetic substance, C ) M agnetic field line profile around a paramagnetic substance A B C D Figure 1 2. Spin alignment s in dif ferent types of materials: A) p aramagnet B) f erromagnet C) antiferromagnet, D) f errimagnet. Figure 1 3 A qualitative plot of m T vs. T for a paramagnet, a ferromagnet, and an antiferromagnet.

PAGE 40

40 Figure 1 4 Schematic representation of a hysteresis loop for a typical permanent magnet H is the applied field, M is the magnetization, M s is the magnetization at saturation, M r is the remanent magnetization, and H c is the coercive field of the magnet. Figure 1 5. Relative energies of the m s states for a system with an S = 10 ground state with negative D value in zero field. Figure adapted with permission from reference 35. Copyright 2009 Royal Society of Chemistry. 35

PAGE 41

41 Figur e 1 6. Effect of negative zero field splitting on the m s state energies for an S = 10 system Figure 1 7. Schematic representation of the change in relative energy of the m s states with incr easing applied field showing allowed and forbidden quantum tunneling transitions.

PAGE 42

42 A B Figure 1 8. Structure of [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] (Mn 12 acetate) A) full molecule, B) core of the molecule with Mn and 3 O 2 ions Mn III are show n in blue, Mn IV in green, O in red, water ligands in teal, and C in grey. Hydrogen atoms are omitted for clarity. Fi gure 1 9. DC susceptibility vs. temperature plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] (Mn 12 acetate) in a 0.1 T applied field.

PAGE 43

43 Figure 1 10. Spin alignment for [Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ] in the ground state that results in a S = 10 system Mn III are shown in blue, Mn IV in green, and O in red. Figure 1 11. Reduced magnetization plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] (Mn 12 acetate). Fit obtained for g = 1.9 5 0.0 044 and D = 0.4 3 0.0050 cm 1

PAGE 44

44 Figure 1 12. AC susceptibility plots vs. temperature for [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] (Mn 12 acetate). AC in phase (top) and out of phase (bottom). Figure 1 13. Arrhenius plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] (Mn 12 acetate). The fitting parameters give an effective barrier U eff /k B = 61 K.

PAGE 45

45 Figure 1 14. Portion of the [Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ] structure showing two Jahn Teller axes in the normal position (Mn III on each side), and one in an abnormal position (central Mn III ) pointing towards an oxide ion Mn III are show n in blue, Mn IV in green, O in red, and C in grey. Figure adapted with permission from reference 35. Copyright 2009 Royal Society of Chemistry. 35 Figure 1 15. Cyclic voltammogram (top) and differential pulse voltammogram (bottom) for [Mn 12 O 12 (O 2 CCHCl 2 ) 16 (H 2 O) 4 ] The indicated potentials are vs. ferrocene. Figure adapted with permission from reference 34. Copyright 2007 American Chemical Society. 34

PAGE 46

46 Figure 1 16. Structure of the [Mn 12 O 12 (O 2 CCHCl 2 ) 16 (H 2 O) 4 ] z core (z = 0 3). M n II are shown in yellow, Mn III in blue, Mn IV in green, and O in red. Figure adapted with permission from reference 34. Copyright 2007 American Chemical Society. 34

PAGE 47

47 CHAPTER 2 NEW LIGAND SUBSTITUTION METHOD TO REPLACE THE WATER LIGANDS WITH ALCOHOL LIGANDS ON Mn 12 SINGLE MOLECULE MAGNETS 2.1 Background As the most studied and best understood group of single molecule magnets, the Mn 12 family is of great importance to the field. This family has been expanding over the last twenty years by using a carboxylate substitution method relying on the removal of acetic acid from solution using the low boiling point toluene/acetic acid azeotrope (eq 2 1). 23 [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] + 16 RCOOH [ Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ] + 16 CH 3 COOH (2 1) This method supported the extensive study of the carboxylate ligand's influence on both magnetic and electrochemical properties of the Mn 12 core. It allowed, for example, the isolation of four oxidation levels using the electron withdrawing ligand dichloroactetate, 34 or the identific ation of the Jahn Teller isomerism phenomenon. 31 36 Other types of non carboxylate ligand substitution have been inve stigated within the family The replacement of carboxylate ligands with nitrate ligands was achieved by treating a solution of [ Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ] with a solution of nitric acid. This method led to the isolation of a series of compounds with formula [ Mn 12 O 12 (NO 3 ) x (O 2 CR) 16 x (H 2 O) 4 ] with very little perturbation to the magnetic properties. 37 Similarly, some carboxylate ligands can be replaced with diphenylphosphin ate ligands by treatment with diphenylphosphinic acid. 38 In both cases, the incoming ligand replaces a carboxylate ligand leaving the four water ligands untouched. The replacement of the water ligands with alcohol ligands has been observed in some cases, but a concerted effort to do so, and in particular to rep lace only the water ligands has

PAGE 48

48 never been demonstrated. Herein, a method to specifically replace the water ligands with alcohol ligands is investigated. 2.2 Experimental Section 2.2.1 Syntheses All manipulations were performed under aerobic conditions us ing materials as received, unless otherwise stated. [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] 2CH 3 COOH 4H 2 O ( 2 1 ) and [Mn 12 O 12 (O 2 CC 6 H 5 ) 16 (H 2 O) 4 ] ( 2 2 ) were prepared as described elsewhere. 20 32 [Mn 12 O 12 ( O 2 CCH 3 ) 16 ( CH 3 OH ) 4 ] ( 2 3 ). A dark brown slurry of complex 2 1 (1.5 g, 0.73 mmol) in a mixture of MeOH (35 mL) and glacial MeCO 2 H (2 mL) was refluxed for one h our The solution was filtered and the residue was redissolved in CH 2 Cl 2 (50 mL). The latter was left at room temperature in a closed flask overnight and filtered again the next day. The solution was then carefully layered with Et 2 O (60 mL), and allowed to stand undisturbed at room temperature for 7 10 days. The resulting black crystals of 2 3 MeOH were collected by filtration, washed with cold MeOH (2 x 5 mL) and Et 2 O (2 x 5 mL), and dried in air. The yield was 35% based on Mn Suitable single crystals were maintained in the mother liquor for all crystallographic studies to prevent the loss of interstitial solvent. Dried solid a nalyzed as solvent free 2 3 Anal. Calcd (found) for 2 3 (C 36 H 64 Mn 12 O 48 ): C, 22.47 (22.74); H, 3.35 (3.42); N, 0 (0%). Selected IR data ( KBr pellet, cm 1 ): 1595 (vs), 1571 (vs), 1550 (s), 1529 (s), 1420 (vs), 1400 (s), 1365 (s), 1201 (m), 1120 (w), 962 (m), 900 (w), 736 (m), 682 (s), 650 (s), 599 (s), 551 (s), 520 (m). Upon exposure to air for prolonged time, the compound undergoes a substitution of the coordinated MeOH groups ( vide infra ) by the corresponding number of H 2 O molecules; which was confirmed by elemental analyses studies. Anal. Calcd (found) for [Mn 12 O 12 (O 2 CMe) 16 (H 2 O) 4 ] (C 32 H 56 Mn 12 O 48 ): C, 20.58 (20.43); H, 3.02 (2.87); N, 0 N, 0 (0%). All bulk DC and AC magnetic susceptibility studies were performed on freshly prepared samples of 2 3

PAGE 49

49 [Mn 12 O 12 ( O 2 CC 6 H 5 ) 16 ( CH 3 OH ) 4 ] ( 2 4 ). A dark brown slurry of complex 2 2 ( 2 4 g, 0.73 mmol) in a mixture of MeOH (35 mL) and benzoic acid (4 .0 g, 33 mmol) was refluxed for one h our The solution was filtered, the residue was redissolved in CH 2 Cl 2 (50 mL) and filtered again The solution was then carefully layered with Et 2 O (60 mL), and allowed to stand undisturbed at room temperature for three days. The resulting black crystals of 2 4 were collected by filtration, washed with cold MeOH (2 x 5 mL) and Et 2 O (2 x 5 mL), and dried in air. The yield was 55 % based on Mn Suitable single crystals were maintained in the mother liquor for all crystallographic studies to prevent the loss of interstitial solvent. Unit cell parameters were collected (Table 2 1) and compare d to a partial structure previously collected which confirmed the presence of four MeOH ligands. In house elemental analysis was no t available so the sample had to be exposed to air for a prolonged time while shipping and it underwent substitution of the M eOH ligands with water ligands Anal Calcd (found) for [Mn 12 O 12 (O 2 CC 6 H 5 ) 16 (H 2 O) 3 ] (C 112 H 86 Mn 12 O 47 ): C, 47.31 (47.23); H, 3.04 (2.89); N, 0 N, 0 (0%). Selected IR data (KBr pellet, cm 1 ): 3139 (vb), 1599 (s), 1562 (s), 1521 (s), 1493 (s), 1448 (m), 1416 (vs), 1353 (s), 1308 (m), 1178 (m), 1140 (m), 1070 (m), 1026 (m), 1002 (w), 936 (w), 839 (w), 717 (s), 677 (s), 656 (w), 615 (m), 550 (m), 515 (m) All bulk DC and AC magnetic susceptibilit y studies were performed on freshly prepared samples of 2 4 [Mn 12 O 12 ( O 2 CCH 3 ) 16 ( t BuOH ) x (H 2 O) 4 x ] ( 2 5 ). A dark brown slurry of complex 2 1 ( 0 .5 0 g, 0. 24 mmol) in a mixture of t BuOH ( 20 mL) and glacial MeCO 2 H ( 1 mL) was refluxed for thirty minutes The solution was filtered and the residue was redissolved in MeCN ( 2 0 mL) The solution was then carefully layered with Et 2 O ( 2 0 mL), and allowed to stand undisturbed at room temperature for three days. The resulting black crystals of 2 5 were collected by filtration, washed with cold t BuOH (2 x 5 mL) and Et 2 O (2 x 5 mL), and dried in air. The yield was 70 %

PAGE 50

50 based on Mn Suitable single crystals were maintained in the mother liquor for all crystallographic studies to prevent the loss of interstitial solven t. Unit cell parameters and a partial structure were collected ; the unit cell parameters are shown in Table 2 1. I n house elemental analysis was no t available so the sample had to be exposed to air for a prolonged time while shipping and it underwent substitution of some of the t BuOH ligands with water ligands. Anal. Calcd (found) for [Mn 12 O 12 (O 2 CMe) 16 ( t BuOH) 1.5 (H 2 O) 2.5 ] ( C 3 8 H 6 8 Mn 12 O 48 ): C, 23.38 ( 23 .58 ); H, 3. 51 (3. 56 ); N, 0 (0%). Selected IR data (KBr pellet, cm 1 ): 3958 (m), 3167 (vb), 2981 (m), 2931 (w), 1592 (s), 1562 (s), 1528 (m), 1428 (vs), 1326 (s), 1115 (w), 1089 (w), 1047 (m), 959 (m), 934 (w), 885 (w), 712 (m), 675 (s), 639 (s), 608 (s), 5 62 (m), 548 (w), 516 (m), 409 (m) All bulk DC and AC magnetic susceptibility studies were performed on freshly prepared samples of 2 5 [Mn 12 O 12 ( O 2 CCH 3 ) 16 ( ( CH 3 ) 2 CH OH ) 4 ] ( 2 6 ). A dark brown slurry of complex 2 1 ( 0 .5 0 g, 0. 24 mmol) in a mixture of isopropanol ( 20 mL) and glacial MeCO 2 H ( 1 mL) was refluxed for thirty minutes The solution was filtered and the residue was redissolved in MeCN ( 2 0 mL). The solution was then carefully layered with Et 2 O ( 2 0 mL), and allowed to stand undisturbed at room temperature for three days. The resulting black crystals of 2 6 were collected by filtration, washed with cold isopropanol (2 x 5 mL) and Et 2 O (2 x 5 mL), and dried in air. The yield was 72 % based on Mn Suit able single crystals were maintained in the mother liquor for all crystallographic studies to prevent the loss of interstitial solvent. Unit cell parameters were collected and are shown in Table 2 1 In house elemental analy sis was no t available so the sam ple had to be exposed to air for a prolonged time while shipping and it underwent substitution of some of the isopropanol ligands with water ligands. Anal. Calcd (found) for [Mn 12 O 12 (O 2 CMe) 16 ( (CH 3 ) 2 CHOH ) 2 ( H 2 O) 2 ] ( C 3 8 H 6 8 Mn 12 O 48 ): C, 23.38 ( 23.4 0 ); H, 3. 51 (3. 51 );

PAGE 51

51 N, 0 (0%). Selected IR data (KBr pellet, cm 1 ): 3958 (m), 3163 (vb), 2979 (w), 2930 (w), 1592 (s), 1562 (s), 1528 (m), 1404 (vs), 1326 (s), 1115 (w), 1047 (m), 959 (m), 934 (w), 885 (w), 712 (m), 675 (s), 639 (s), 612 (s), 562 (m), 548 (w), 516 (m ), 409 (m) All bulk DC and AC magnetic susceptibility studies were performed on freshly prepared samples of 2 6 2.2.2 X Ray Crystallography Data for complex 2 3 were collected by Dr. Khalil A. Abboud at 173 K on a Siemens SMART PLATFORM equipped with A C CD area detector and a graphite monochromator utilizing MoK radiation ( 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 measured at the end of data collection to monitor instrument and crystal stability (maxim um correction on I was < 1 %). Absorption corrections by integration were applied based on measured indexed crystal faces. The structure w as solved by the Direct Methods in SHELXTL6 39 and refined us ing 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. The asymmetric unit consists of a Mn 12 cluster and methanol molec ule disordered over two parts. Due to disorder the methanol protons were no t located nor included in the final refinement model. A total of 230 parameters were refined in the final cycle of refinement using 3920 reflections with I > 2 (I) to yield R 1 and wR 2 of 2.75% and 8.49%, respectively. Refinement was done using F 2 The cry stallographic data and structure refinement details are collected in Table 2 2 2.2.3 DC and AC Magnetometry Variable temperature magnetic susceptibility data down to 1.8 K were collected on a Quantum Design MPMS XL SQUID magnetometer equipped with a 7 T D C magnet at the

PAGE 52

52 University of Florida. Pascal's constants were used to estimate the diamagnetic contributions to the susceptibility, which were subtracted from the experimental susceptibility to give the molar magnetic susceptibility ( M ). Samples were embedded in solid eicosane to prevent torquing and solvent loss, unless otherwise stated. DC magnetic susceptibility data were collected using a constant 0.1 T applied field between 5 and 300 K. AC magnetic susceptibility data were collected between 1.8 an d 15 K, in absence of applied DC field, with a 3.5 G field oscillating at frequencies up to 1500 Hz. Magnetization vs field data were collected over a range of fields and temperatures, and fitted using the program MAGNET. 40 Contour plots were obtained using the program GRID. 41 Both programs were written at Indiana University by E. R. Davidson. 2.2.4 NMR Spectroscopy Room temperature 1 H NMR spectra were collected on a VARIAN INOVA 500 spectrometer operating at 499.799 MHz for 1 H. The standard 1D proton pul se sequence was used with a short acquisition time to account for the paramagnetic nature of the samples. 2.2.5 Other Studies Infrared spectra were recorded in the solid state (KBr pellets) on a Nicolet Nexus 670 FTIR spectrophotometer in the 400 4000 cm 1 range. Elemental analyses were performed at the in house facilities of the University of Florida Chemistry Department ( 2 3 ), or at Complete Analysis Laboratories, Inc ( 2 4 2 5 and 2 6 ) 2.3 Results and Discussion 2.3.1 Discussion of the Synthesis The pr eparation s of compounds 2 3 MeOH 2 4 2 5 and 2 6 are described in eq 2 2 eq 2 3 eq 2 4 and eq 2 5 respectively

PAGE 53

53 [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] + 4 MeOH [ Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] + 4 H 2 O (2 2 ) [Mn 12 O 12 (O 2 CC 6 H 5 ) 16 (H 2 O) 4 ] + 4 MeOH [ Mn 12 O 12 (O 2 CC 6 H 5 ) 16 (MeOH) 4 ] + 4 H 2 O (2 3 ) [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] + 4 t BuOH [ Mn 12 O 12 (O 2 CCH 3 ) 16 ( t BuOH) 4 ] + 4 H 2 O (2 4 ) [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] + 4 (CH 3 ) 2 CHOH [ Mn 12 O 12 (O 2 CCH 3 ) 16 ( (CH 3 ) 2 CHOH ) 4 ] + 4 H 2 O (2 5 ) The substitution of the water ligands in 2 1 or 2 2 by alcohol ligands was obtained by exposing the compound to a large excess of alcohol and heating the solution. To ensure the stability of the cluster upon exposure to a reducing solvent an d heat, an excess of acetic or benzoic acid was added t o the solution. It must also be noted that the residue needed to be kept wet with alcohol when redissolved in order to obtain the desired product. The crystals are stable for a long time when kept in the mother liquor, but the alcohol ligands are slowly re placed by water ligands when the crystals are exposed to air. For this reason, all magnetic measurements were performed on crystalline powders obtained from crystals just removed from the mother liquor and lightly dried with absorbent paper to remove the e xcess solvent ( vide infra ). 2.3.2 Description of the Structure The structure of the individual molecule of 2 3 is shown in Figure 2 1 It is almost identical to that of the original Mn 12 acetate, the main difference arising from the replacement of the water ligands with methanol ligands. The difference in the structures and consequently the magnetic properties arises from the difference in the crystal packing of the compounds.

PAGE 54

54 The crystal packing of Mn 12 acetate is known to contain two acetic acid and four water solvent molecules per Mn 12 unit. 42 Each acetic acid molecule lies betw een two Mn 12 molecules, and forms a hydrogen bond with either. This creates six Mn 12 isomers in regard to how many hydrogen bonds the Mn 12 unit forms with neighboring acetic acid molecules (zero to four). Each isomer has a slightly different environment, a nd three different orientations of the easy axis can be observed. This phenomenon is known to cause issues in magnetic measurements which require alignment of the crystals with respect to the magnetic field. 43 In contrast, the crystal packing of c ompound 2 3 contains only one methanol molecule which lies on a symmetry element of the lattice. This causes the r etention of the molecular S 4 symmetry throughout the entire lattice (including the lattice solvents), and the orientation of all Mn 12 units to be strictly identical (Figure 2 2). This effect has been previously observed in other Mn 12 derivatives such as [M n 12 O 12 (O 2 CCH 2 Br) 16 (H 2 O) 4 ] (Mn 12 bromoacetate) or [Mn 12 O 12 (O 2 CCH 2 t Bu) 16 (CH 3 OH) 4 ] (Mn 12 t Bu acetate). 44 45 A partial structure of 2 5 was collected and it was observed that some of the water ligands had been replaced. The nature of the replacing ligand was unclear and further studies would be needed in order to confirm the formula. Because of the ability of the Mn 12 core to accommodate sixteen tert butylacetate ligands, 45 46 it is unlikely that the bulkiness of t BuOH prevents full substitution of the water ligands. However, the recrystallization step for the synthesis of 2 5 was performed in acetonitrile, which has a higher water content than methylene chloride, which was used in the recrystallization step of 2 3 This could cause the t BuOH ligands to be replaced with water ligands during the recrystallization step. It should also be noted that the unit cells of 2 3 2 5 and 2 6 are very similar and similar to the unit cell of Mn 12 acetate crystallized in the same space group. This is expected as the

PAGE 55

55 volume of the molecule is determined mainly by the carboxylate ligands, which are identical in all four compounds. 2.3.3 Magnetometr y Studies Variable temperature magnetic susceptibility studies were performed on microcrystalline powder samples obtained from clean wet crystals of each compound. In order to avoid substitution of the alcohol ligands by water molecules, the powders were q uickly weighed and covered in eicosane. 2.3.3.1 DC studies The results of the DC experiment are plotted as M T vs. T (Figure 2 3 2 4, 2 5, and 2 6). The DC susceptibility profiles for 2 3 2 4 2 5 and 2 6 are very similar to the ones for Mn 12 acetate and Mn 12 benzoate, and the small differences can be attributed to inexactitude of the molecular weights used to process the data due to the solvent content of the samples. No major differences between the susceptibility data of Mn 12 acetate, Mn 12 benzoate, 2 3 2 4 2 5 and 2 6 were expected as the DC susceptibility measured on microcrystalline powder samples mainly depends on properties that are intrinsic to the individual molecule and in particular its core which are very similar in all compounds. To con firm the S = 10 ground state of the molecules and obtain g and D parameters for each compound, magnetization vs field data were collected over a range of fields and temperatures, and fitted using the program MAGNET 40 (Figure 2 7, 2 9, 2 11, and 2 13). Contour plots of the fitting errors were obtained using the program GRID 41 (Figure 2 8, 2 10, 2 12, and 2 14). The fit of the magnetization vs. field data confirms the S = 10 ground state for all four compounds. The obtained fitting parameters are gathered i n Table 2 3 It should be noted that g is expected to be slightly below 2 .0 for Mn III and Mn IV ions For 2 4 g is found to be slightly above 2.0 which indicates that the fit may not reflect reality perfectly. Due to the

PAGE 56

56 inaccuracy in molecular weight, an appreciable uncertainty on the obtained g and D values is expected. High frequency electron paramagnetic resonance (HFEPR) experiments would provide more accurate values of g and D 2.3.3.2 AC studies The results of the A C experiment are plotted as M T vs. T for the in phase data, and as M vs. T for the out of phase data (Figure 2 15, 2 17, 2 19, and 2 21 ). The general behavior of the four compounds under oscillating field is identical to that of the original Mn 12 acet ate ( 2 1 ). B efore the frequency dependent drop caused by blocking of the magnetization, the AC in phase data agrees with the DC data, confirming the S = 10 ground state spin. For 2 4 the AC out of phase data display s two peaks which suggests that two isomers with slightly different anisotropy are present in the sample. This phenomenon is known as Jahn Teller isomerism, and has been seen in many Mn 12 derivatives. 30 31 47 It is usually caused by minor rearrangements within the crystal lattice upon loss of lattice solvents when the sample is dri ed, but it also has been observed in samples prepared directly out of the mother liquor when the crystallization produces the two isomers. 46 The AC out of phase data collected at multiple frequencies was used to determine the kinetic parameters for the ma gnetization relaxation: at the '' M peak maximum, the magnetization relaxation rate ( 1/ ) is equal to the oscillation frequency of the AC field. The obtained set s of relaxation rates vs. temperature w ere fitted to the Arrhenius equation (eq. 2 6 ) to calculate the effective energy barrier to the reversal of magnetization ( U eff ) and the pre exponential relaxation time 0 (Figure 2 16, 2 18, 2 20, and 2 22 ). 43 (2 6 )

PAGE 57

57 The fitting parameters for all four compounds are gathered in Table 2 4 The obtained effective energy barriers are high for Mn 12 derivatives (usually in the 60 65 K range). In fact, the effective energy barrier of 75.6 K for 2 5 is the highest observed for a member of the Mn 12 family. 2.3.4 NMR Spectroscopy The room temperature 1 H NMR spectrum for 2 3 (Figure 2 23 ) is identical to that of the original Mn 12 acetate, which is to be expected as the methanol ligands were most likely replaced by water ligands when the complex is dissolved in the NMR solvent. The two peaks of integration 1 correspond to the axial acetate ligands, and the peak of integration 2 cor responds to the equatorial acetate ligands. The broad peak centered around 8.3 ppm corresponds to water molecules (free in solution and ligands). 2.4 Conclusion A new synthetic method for substitution of the water ligands on the Mn 12 core with alcohol liga nds was investigated It relies on the exposition of the Mn 12 compound to a large excess of alcohol at elevated temperature. The method has proven effective for replacement of the water ligands with methanol ligands, but further investigation will be neede d in order to demonstrate its efficiency with other alcohols The magnetic propertie s of the individual molecule with substituted alcohol ligands are essentially identical to the properties of the original molecule The differences in the crystal packing i nduced by the substitution however, can be of major importance for magnetic measurements relying on single crystals such as HFEPR spectroscopy measurements or investigation of properties related to the quantum tunneling of magnetization as it is the case for 2 3 43

PAGE 58

58 Table 2 1. Unit cell parameters of crystals of complex 2 4 2 5 and 2 6 2 4 2 5 2 6 a () 17.6 17.4 17.4 b () 17.6 17.4 17.4 c () 24.6 12.3 12.3 80.9 90 90 77.0 90 90 89.7 90 90 V ( 3 ) 7326 3707 3707 Crystal system Triclinic P Tetragonal I Tetragonal I Table 2 2 Crystal data and structure refinement parameters for complex 2 3 2 3 MeOH formula a C 37 H 68 Mn 12 O 49 fw, g mol 1 1956.15 space group I 4 a, 17.3444(10) b, 17.3444(10) c, 11.9877(14) 90 deg 90 90 V, 3 3606.2(5) Z 2 T, K 173(2) radiation, b 0.71073 calc g cm 3 1.798 1 2.124 R1 c, d 0.0275 wR2 e 0.0849 a Including solvent molecules. b Graphite monochromator. c I I ). d R 1 = 100 (|| F o | | F c ||)/ | F o |. e wR 2 = 100[ [ w ( F o 2 F c 2 ) 2 ]/ [ w ( F o 2 ) 2 ]] 1/2 w = 1/[ 2 ( F o 2 ) + [( ap ) 2 + bp ], where p = [max ( F o 2 O) + 2 F c 2 ]/3.

PAGE 59

59 Table 2 3 Reduced magnetization fitting parameters 2 3 2 4 2 5 2 6 S 10 10 10 10 g 1.92 0.0059 2. 0 9 0.0048 1. 9 5 0.0080 1. 9 8 0.0051 D (cm 1 ) 0.43 0.0068 0.3 8 0.0047 0.39 0.0084 0. 4 3 0.0058 D (K) 0. 6 2 0. 54 0. 5 7 0. 61 Table 2 4 Arrhenius parameters 2 3 2 4 2 5 2 6 U eff (K) 71.9 71.7 75.6 75.3 0 (s) 1.18 10 8 4.46 10 9 5.83 10 9 5.70 10 9 Figure 2 1. Structure of [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ). Mn III are shown in blue, Mn IV in green, O in red, C in grey, O and C of MeOH ligands are shown in pink and black respectively. Hydrogen atoms are omitted for clarity.

PAGE 60

60 A B Figure 2 2. Packing diagram of [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ). A) Top view of the molecules along the z axis. B) Side view. Mn III are shown in blue, Mn IV in green, O in red, C in grey, O and C of MeOH are shown in pink and black respectively. Hydrogen atoms are omitted for clarity.

PAGE 61

61 Figure 2 3. DC magnetic susce ptibility plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ) in an applied 0.1 T field. Figure 2 4. DC magnetic susceptibility plot for [Mn 12 O 12 (O 2 CC 6 H 5 ) 16 (MeOH) 4 ] ( 2 4 ) in an applied 0.1 T field.

PAGE 62

62 Figure 2 5. DC magnetic susceptibility plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( t BuOH) 4 ] ( 2 5 ) in an applied 0.1 T field. Figure 2 6. DC magnetic susceptibility plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( (CH 3 ) 2 CHOH ) 4 ] ( 2 6 ) in an applied 0.1 T field.

PAGE 63

63 Figure 2 7. Reduced magnetization plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ). Fit obtained for S = 10, g = 1.92 0.0059, and D = 0.43 0.0068 cm 1 A B Figure 2 8. Contour plot of the fitting error for the reduced magnetization data for [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ). A) T he entire range of g and D for which the fittin g error is calculated. B) Zoom ed in view of the error minimum (the red star indicates the fitting parameters obtained from the reduced magnetization fit).

PAGE 64

64 Figure 2 9. Reduced magnetization plot for [Mn 12 O 12 (O 2 CC 6 H 5 ) 16 (MeOH) 4 ] ( 2 4 ). Fit obtained for S = 10, g = 2. 0 9 0.0048, and D = 0.3 8 0.0047 cm 1 A B Figure 2 10. Contour plot of the fitting error for the reduced magnetization data for [Mn 12 O 12 (O 2 CC 6 H 5 ) 16 (MeOH) 4 ] ( 2 4 ). A) The entire range of g and D for which the fitting error is calculated. B) Zoomed in view of the error minimum (the red star indicates the fitting parameters obtained from the reduced magnetization fit).

PAGE 65

65 Figure 2 11. Reduced magnetization plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( t BuOH) x ( H 2 O) 4 x ] ( 2 5 ). Fit obtained for S = 10, g = 1. 9 5 0.0080, and D = 0.39 0.0084 cm 1 A B Figure 2 12. Contour plot of the fitting error for the reduced magnetization data for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( t BuOH) x ( H 2 O) 4 x ] ( 2 5 ). A) The entire range of g and D for which the fitting error is calculated. B) Zoomed in view of the error minimum (the red star indicates the fitting parameters obtained from the reduced magnetization fit).

PAGE 66

66 Figure 2 13. Reduced magnetization plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( (CH 3 ) 2 CHOH ) 4 ] ( 2 6 ). Fit obtained for S = 10, g = 1. 9 8 0.0051, and D = 0. 4 3 0.0058 cm 1 A B Figure 2 14. Contour plot of the fitting error for the reduced magnetization data for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( (CH 3 ) 2 CHOH ) 4 ] ( 2 6 ). A) The entire range of g and D for which the fitting error is calculated. B) Zoomed in view of the error minimum (the red star indicates the fitting parameters obtained from the reduced magnetization fit).

PAGE 67

67 Figure 2 15 AC magnetic susceptibility plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ). AC in phase (top) and out of phase (bottom). Figure 2 16 Arrhenius plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ). The fitting parameters give an effective barrier U eff /k B = 71.9 K and a pre exponential factor 0 = 1.18 10 8 s.

PAGE 68

68 Figure 2 17 AC magnetic susceptibility plot for [Mn 12 O 12 (O 2 CC 6 H 5 ) 16 (MeOH) 4 ] ( 2 4 ). AC in phase (top) and out of phase (bottom). Figure 2 1 8 Arrhenius plot for [Mn 12 O 12 (O 2 CC 6 H 5 ) 16 (MeOH) 4 ] ( 2 4 ). The fitting parameters give an effective barrier U eff /k B = 71.7 K and a pre exponential factor 0 = 4.46 10 9 s.

PAGE 69

69 Figure 2 1 9 AC magnetic susceptibility plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( t BuOH) x ( H 2 O) 4 x ] ( 2 5 ). AC in phase (top) and out of phase (bottom). Figure 2 20 Arrhenius plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( t BuOH) x ( H 2 O) 4 x ] ( 2 5 ). The fitting parameters give an effective barrier U eff /k B = 7 5.6 K and a pre exponential factor 0 = 5.83 10 9 s.

PAGE 70

70 Figure 2 21 AC magnetic susceptibility plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( (CH 3 ) 2 CHOH ) 4 ] ( 2 6 ). AC in phase (top) and out of phase (bottom). Figure 2 22 Arrhenius plot for [Mn 12 O 12 (O 2 CCH 3 ) 16 ( (CH 3 ) 2 CHOH ) 4 ] ( 2 6 ). The fitting parameters give an effective barrier U eff /k B = 75.3 K and a pre exponential factor 0 = 5.70 10 9 s.

PAGE 71

71 Figure 2 2 3 1 H NMR spectrum of [Mn 12 O 12 (O 2 CCH 3 ) 16 (MeOH) 4 ] ( 2 3 ) in CD 3 CN at room temperature (*: methylene chloride, acetonitrile, and diethyl ether peaks).

PAGE 72

72 CHAPTER 3 CHARACTERIZATION OF A FLUXIONAL PROCESS IN M n 12 SINGLE MOLECULE MAGNETS THROUGH VT NMR SPECTROSCOPY 3 .1 Background Single molecule magnets (SMMs) are individual molecules that can function as independent, truly monodisperse magnetic nanoparticles. 48 50 With the miniaturization of technologies and the need for smaller and smaller devices, SMMs have been suggested as potential candidates for many applications. 19 51 52 They bring all the advantages of molecular chemistry to the field of magnetism, in particular the ability to be truly dissolved in solvents. This, of course, is an advantage only if the structure of the molecule does not change upon dissolution. One of the first discovered, and still today the most well studied SMM is [Mn 12 O 12 (CH 3 COO) 16 (H 2 O) 4 ] ( 3 1 ) (Mn 12 acetate) 20 Thanks to its relatively easy and high yield synthesis, it has been, and still is widely used in experiments in the search for unique properties and applicatio ns of single molecule magnets. 16 53 54 It is also the first member of a very large family of derivatives, most of t hem obtained via carboxylate substitution on Mn 12 acetate using the toluene/acetic acid azeotrope method developed almost 20 years ago 23 Mn 12 acetate and its derivatives are compounds known to form high quality single crystals suitable for analysis via X ray diffraction. Over the years, this allowed for the creation of a vast library of Mn 12 crystal structures from which we can distinguish a number of common features: all Mn 12 compounds possess a central Mn IV 4 O 4 cubane surrounded by a star shaped ring of 8 Mn III ions bridged to the Mn IV s by 8 3 oxides, and 16 carboxylate ligands These ligands are divided in 3 types : 4 carboxylates of type A which have both their oxygen atoms on Jahn Teller (JT) axes of Mn III s, 4 carboxylates of type B which have one oxygen atom on a Mn III JT axis, and 8 carboxylates of type C which have none of their oxygen atoms on Mn III JT axes

PAGE 73

73 bec ause t hey are in equatorial positions. 35 In most Mn 12 derivative structures, the four remaining positions in the ligand sphere consist of wa ter ligands although some possess less than four, leaving the remaining Mn III ion to be pentacoordinate. 23 24 55 56 The position s of these water ligand s are of major importance as they determine the point group of the molecule. As has been shown the structure of Mn 12 acetate and its derivatives in the solid state ha ve been well studied. 35 However, when considering the potential application for a single molecule magnet such as Mn 12 ace tate we find that it is equally important to gain knowledge over the behavior of such molecules in solution, if for no other reason than to ensure that their structures and physical properties are retained upon dissolution. NMR spectroscopy is the obvious best choice to carry out structural analysis of compounds in solution, although it is not nearly as popular to study paramagnetic molecules as it is in all other areas of chemistry. Indeed, the unpaired electrons of paramagnetic metals influence the magne tism of the studied nuclei, shifting the NMR signal such that the peak positions can hardly be used for identification, and broadening the peaks decreasing the resolution of the technique sometimes to the extent that it becomes unusable. In Mn 12 compounds however, the fast electronic relaxation of Mn III ions allows for relatively good resolution of NMR spectra, and the technique has been shown to be very useful in monitoring the purity of Mn 12 derivatives after ligand substitution. Many 1 H NMR spectra of Mn 12 derivatives are available in the literature. 23 31 35 37 44 The 1 H NMR sp ectrum of Mn 12 acetate at room temperature presents 3 carboxylate peaks in a 1:2:1 ratio and a broad peak for the water ligands ( F igure 3 1). This spectrum is not in agreement with the S 4 symmetry of the compound which predicts four carboxylate peaks in a 1:1:1:1 ratio, each peak corresponding to four symmetr y equivalent carboxylate ligands ( F igure 3 4 ). Although accidental degeneracy could explain this spectrum, a similar pattern has been

PAGE 74

74 observed in all other Mn 12 derivative spectra (Figure 3 2 and 3 3 ) s uggesting that a deeper analysis is needed to explain this behavior. Furthermore, it can be observed that four different solid state symmetry point groups (S 4 D 2 C 2 C 1 ) can be obtained upon crystallization of Mn 12 derivatives ( F igure 3 5 ). 30 32 No particular criterion seem s to dictate the point group choice, and two different point groups have even been observed for the same carboxylate with different crystallization conditions. 30 Changes to the NMR spectra upon cooling wer e previously noticed by a former student of our group. In this work, an explanation for these two observations was sought via variable temperature NMR spectroscopy. 3 .2 Experimental Section 3 .2.1 Synthesis All manipulations were performed under aerobic con ditions using chemicals as rec eived, unless otherwise stated. [Mn 12 O 12 (O 2 CMe) 16 (H 2 O) 4 ] ( 3 1 ), [Mn 12 O 12 (O 2 C HCl 2 ) 16 (H 2 O) 4 ] ( 3 2 ) [Mn 12 O 12 (O 2 CCH 2 C H 3 ) 16 (H 2 O) 4 ] ( 3 3 ), [Mn 12 O 12 (O 2 CCD 2 CH 3 ) 16 (H 2 O) 4 ] ( 3 4 ), [Mn 12 O 12 (O 2 CCH 2 C(CH 3 ) 3 ) 16 (H 2 O) 4 ] ( 3 5 ), [Mn 12 O 12 (NO 3 ) 4 (O 2 CCH 2 C(CH 3 ) 3 ) 1 2 (H 2 O) 4 ] ( 3 6 ), [Mn 8 Fe 4 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] ( 3 7 ), and [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 3 8 ), were prepared as described elsewhere. 20 23 24 37 57 [Mn 12 O 12 ( p CF 3 C 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 3 9 ), [Mn 12 O 12 ( o FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 3 10 ), and [Mn 12 O 12 (3,5 F 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( 3 1 1 ) were synthesized from ( 3 1 ) using a previously reported method of ligand substitution, and recrystallized from methylene chloride/hexanes. 23 The identities of 3 9 3 10 and 3 11 were confirmed by elemental analysis, and IR spectroscopy. Vacuum dried crystals of 3 9 were analyzed as solvent free: Anal. Calcd (Found): C, 38.93 (38.84); H, 1.84 (1.51); N, 0 (0) %. Vacuum dried crystals of 3 10 were analyzed as solvent free : Anal. Cal cd (Found): C, 42.71 (42.68); H, 2.30 (2.10); N, 0 (0) %.

PAGE 75

75 Vacuum dried crystals of 3 11 were analyzed as 3 11 (3,5) F 2 C 6 H 3 CO 2 H : Anal. Calcd (Found): C, 39.75 (39.60); H, 1.68 (1.5); N, 0 (0) %. Selected IR data for 3 9 (KBr pellet, cm 1 ) : 3139(m), 1707(w), 1599(m), 1562(s), 1514(s), 1431(vs), 1326(vs), 1170(m), 1131(s), 1103(m), 1066(s ) 1019(m), 866(m), 784(m ) 769(w), 711(m), 656(b), 617(b), 547(m). Selected IR data for 3 10 (KBr disk, cm 1 ) : 3148(b), 1714(m), 1613(vs), 1596(vs), 1540(s), 1520(s), 1488(s), 1454(s), 1412(vs), 1267(w), 1228(m), 1164(m), 1098(m), 1033(w), 956(vw), 862(m), 843(w), 800(m), 756(s), 696(m), 659(s), 619(m), 554(m), 523(w). Selected IR data for 3 11 (KBr disk, cm 1 ) : 3102(b), 1621(m), 1584(s), 1524(m), 1473(s), 1442(s), 140 2(vs), 1356(s), 1288(w), 1212(w), 1123(s), 989(s), 962(m), 890(m), 858(m), 781(m), 763(m), 713(w), 662(m), 616(w), 552(w), 509(w), 436(w). 3 .2.2 NMR Spectroscopy Room temperature 1 H and variable temperature 19 F NMR spectra were collected on a VARIAN INOVA 500 spectrometer operating at 499.799 MHz for 1 H. Spectra were recorded at the indicated temperature using a sw2 VARIAN 5mm conventional probe calibrated with methanol. Variable temperature 1 H, and 2 H NMR spectra were collected on a VARIAN Gem300 spectrome ter, and a VARIAN i400 spectrometer respectively. A nitrogen gas flux cooled by liquid nitrogen was used to decrease the temperature. The standard 1D proton pulse sequence was used; however, data was collected over a very short acquisition time (typically 0.05 to 0.5 s) because of the fast relaxation of the nuclei due to the paramagnetic nature of the samples. Approximate values for the rate of the fluxional process ( 1 ) at the coalescence temperature ( T c ) and the free energy of activation ( ) were calculated from a visual estimation of the coalescence temperature and the frequency difference between the coalescing signals in absence of exchange ( 0 ) using the Eyring relationship (eq 3 1 and 3 2 ). 58 60

PAGE 76

76 ( 3 1) ( 3 2) The temperature dependence of chemical shifts of the coalescing signals in the low temperature region (no exchange) was fitted to eq 3 3 and the chemical shifts in absence of fluxional process were extrapolated to the coalescence temperature using this fi t. ( 3 3) 3 .2.3 Structural A nalyses Infrared spectra were recorded in the solid state (KBr pellets) on a Nicolet Nexus 670 FTIR spectrophotometer in the 400 4000 cm 1 range. Elemental analyses (C, H, and N) were performed by the in house facilities of the University of Florida Chemistry Department 3 .3 Results and D iscussion 3 .3.1 Synthesis T he Mn 12 derivatives were synthesized using a well known method of ligand substitution from Mn 12 acetate and recrystallized from methylene chloride and hexanes with the exception of ( 3 7 ) which was recrystallized from methylene chloride and acetonitrile. 3 .3.2 Description of the S tructures The cores of all the compounds mentioned in this chapter are identical, and analog ou s to the Mn 12 acetate core. The central [Mn IV 4 O 4 ] cubane and the eight Mn III star shaped ring around it are conserved with very minor structural modifications related to the ligands. 3 .3.3 Room T emperature NMR S pectroscopy Although NMR spectroscopy is most often used to study diamagnetic systems, it can reveal itself useful to investigate the structure of some paramagnetic complexes. 61 64 Systems

PAGE 77

77 containing Mn III ions, in particular, are known to give relatively good NMR spectra due to the very small electron relaxat ion times arising from their substantial zero field splitting. 65 68 Room temperature NMR spectra of many Mn 12 derivatives have been previously reported. 23 31 35 37 44 Several of them have been measured again for the purpose of this work; a few examples are shown in Figure s 3 1 3 2, and 3 3 As mentioned before, the spectra are consistent with higher symme try than provided by the point groups found in solid state, with the equatorial ligands appearing as only one peak of double integration magnitude The apparent symmetry preserves two types of axial ligands, with those bridging two Mn III (type A) giving rise to the furthest downfield peaks of single integration. For Mn 12 acetate each ligand type appears as one peak with the following peak assignments: 47.5 ppm: type A (axial CH 3 ), 40.8 ppm: type C ( equatorial CH 3 ) 13.7 ppm: type B (axia l CH 3 ). The broad peak centered at 11.6 ppm corresponds to the protons of water either as ligands or free molecules in fast exchange For Mn 12 dichloroacetate, the peak assignment is very similar with the peaks at 40.5, 24.3, and 11.6 ppm corresponding to type A axial ligands, type B axial ligands, and equatorial ligands respectively. For Mn 12 propionate, it should be noted that each ligand is expected to appear as three peaks: one of integration 3 for the terminal methyl group, and two, each of integration 1, for the methylene protons which are diastereotopic. Accounting for the fact that both types of equatorial ligands appear in the same peak with double integration, we would expect to obtain nine peaks for the ligands of ratio 1:1:1:1:2:2:3:3:6. The obse rved spectrum, however, shows only seven peaks of ratio approximately 2:2:2:2:3:3:6 which suggests that the diastereotopic nature of the methylene protons is conserved only in the equatorial ligands. From this analysis, we can assign the peaks at 52.3, 12. 7, 10.4, 3.8, and 5.1 ppm to the type A axial methylene protons, type B axial methylene protons, type A axial methyl protons, type B axial methyl

PAGE 78

78 protons, and equatorial methyl protons respectively, and the two peaks at 47.1 and 45.1 ppm corresponding to the diastereotopic methylene protons of the equatorial ligands. 3 .3.4 Hypothesis of a F luxional P rocess When observing the arrangement of the axial ligands, w e can notice th at simple rotations of the carboxylates shown in blue in Figure 3 4 along with disp lacement of the water ligands allow for the formation of the point groups observed in the solid state (Figure 3 5 and 3 6 ) These carboxylates are the ligands of type A which have both of their oxygen atoms on JT axes of Mn III s. Therefore, it can safely be assumed that the O Mn bond is weak enough that it can be broken and reformed fairly easily in solution and the hypothesis of a fluxional process exchanging these carboxylates and water ligands happening rapidly in solution at room temperature is reasonab le ( F igure 3 7 ). If this fluxional process happens with all four water ligands and all four A type carboxylates, and is faster than the NMR timescale at room temperature, two virtual mirror planes are created making all equatorial carboxylate ligands equivalent by symmetry ( F igure 3 8 ). This process would therefore increase the symmetry of the molecule in solution, explaining the room temperature NMR spectrum which is consistent with D 2d symmetry. In order to test this hypothesis, variable temperature NMR spectroscopy was used in an attempt to observe the fluxional process by decreasing the temperature thereby slowing it down and looking for decoalescence of the appropriate peaks 3 .3.5 Variable T emperature NMR S pectroscopy Variable temperature NMR sp ectroscopy has been widely used to investigate exchange processes. 69 74 In our study, v ariable temperature NMR spectra were collected for a range of compounds ( 3 1 to 3 11 ) on different nuclei ( 1 H, 2 D, and 19 F). The absence of diamagnetic

PAGE 79

79 solvent peaks in the 19 F spectra allowed for better resolution of the paramagnetic peaks particularly at low temperature where these signals become weaker. It can be observed that all spectra become more complex at low temperature, confirming our hypothesis of a fluxional process, and indicating that it can be slowed down enough to be observable. In most cases, the low temperature spectra showed many peaks, indicating the presence of many isomers in solution, and locating the coalescing peaks was not always possible. In some cases however, the low temperatur e spectra were clear enough to find the coalescing peaks and follow their chemical shifts as the temperature increases, allowing for calculation of approximate parameters for the fluxional process. This was possible for compounds 3 4 3 6 3 8 and 3 1 1 (F igures 3 9 to 3 16 ), and the calculated free energy of activation were 9.13, 10.2, 10.8, and 11.0 kcal/mol, respectively. The free energy of activation is expected to be similar for all Mn 12 derivatives as most of the energy barrier to the fluxional proces s is likely of enthalpic nature and comes from the energy required to break the Mn O bond between the manganese and the rotating carboxylate. The small differences in free energy of activation between the Mn 12 derivatives account for the entropic component which likely comes from the nature of the ligand and its bulkiness. The 2 H VT NMR spectra obtained for 3 4 (Figures 3 9 and 3 10 ) confirm the claims made from the room temperature spectrum of 3 3 : out of the three types of ligands, only one possesses dias tereotopic methylene protons. The room temperature spectrum shows four peaks of similar integration, with the peaks at 53.7 and 11.2 ppm corresponding to the type A and type B axial ligands respectively, and the two peaks at 47.6 and 46.7 ppm corresponding to the two diastereotopic methylene deuterium of the equatorial ligands. It can be observed that all of the ligand peaks decoalesce as the temperature is lowered, to finally result in eight peaks at low

PAGE 80

80 temperature: peak 1 and 4, and peak 7 and 8 arising from the type A, and type B axial ligands respectively, and the couples of peaks 2/3 and 5/6 arising from the two equatorial ligand peaks. The coalescence of peaks 1 and 4, and 7 and 8 can be used to calculate the free energy of activation of the fluxional process, with peaks 1 and 4 coalescing around 50C (223.15 K), and peaks 7 and 8 coalescing around 40C (233.15 K). Even though the two pairs of peaks coalesce at different temperatures, the two free energy of activation values obtained are expected to be the same (within error) as the two coalescence phenomena arise from the same fluxional process. The difference in coalescence temperature simply comes from the peak separation at low temperature being different. The two values of free energy of activati on obtained are 8.95 and 9.12 kcal/mol for the peak 1/4 and 7/8 pairs respectively. The peak position, fitting parameters, and the extrapolation of the peak position in absence of exchange to the coalescence temperature are reported in Table 3 1. In an att empt to clarify the 1 H VT NMR spectra of 3 5 ( Figures A 1 and A 2 ), the 1 H VT NMR spectra of 3 6 were r ecorded between 90 and 30 C (F igure 3 11 and 3 12 ). In this compound, one type of axial carboxylate ligand (type A) was replaced with nitrate ligands which do not appear on 1 H NMR spectra. At room temperature, the spectrum exhibits two peaks in a 1:2 ratio for the tert butyl protons ( Figure A 3 ), and thr ee peaks in a 1:1:1 ratio for the methylene protons (Figure 3 11 ). Although much less clear than for 3 4 the 1 H VT NMR spectra of 3 6 still show significant decoalescing of the peaks under 30C (243.15 K), and the peaks in the methylene region allow for calculation of kinetic parameters: from the coalescence of peak 5 and 6, we were able to calculate a free energy of activation value of 10.2 kcal/mol. The peak position, fitting parameters, and the extrapolation of the peak positions in absence of exchange to the coalescence temperature are reported in Table 3 2.

PAGE 81

81 In order to enhance the clarity of the spectra, especially at low temperature, the focus of this work was shifted towards fluorinated compounds to avoid obtaining a high number of intense solvent p eaks in the diamagnetic region. In this perspective, the 19 F VT NMR spectra of 3 8 3 9 3 10 and 3 11 in CD 2 Cl 2 were recorded over the widest range of temperatures allowed by the solvent. For compound 3 9 it seems that we were not able to decrease the temperature enough to observe splitting of the three peaks, and only broadening due to the increase in paramagnetism as the temperature decreases can be seen ( Figure A 4 ). For 3 10 it can be observed that the three peaks start to decoalesce around 15C, but the resolution of the spectra is rapidly lost as the temperature decreases, preventing any use of the data for calculation of kinetic parameters ( Figure A 5 ). The VT NMR spectra of compounds 3 8 and 3 11 however, show clear splitting of one of the roo m temperature peaks into two peaks at low temperature (peak 1 and 2, and 6 and 7 respectively). The room temperature spectrum of 3 8 shows 4 peaks in a 2:1:1 ratio as expected. As the temperature is lowered, the downfield peak splits into two peaks to give a fairly well resolved spectrum containing four main peaks at low temperature. While it can be noticed that a number of low intensity of peaks are visible at low temperature, suggesting that many isomers may be present in solution, the presence of four pe aks of higher intensity suggests that one isomer is favored. The coalescence of peaks 1 and 2 around 16C allows for the calculation of the free energy of activation of the fluxional process for this compound, with a value of 10.8 kcal/mol. The 19 F VT NMR spectra of 3 8 around the coalescence temperature are shown in Figure A 6 The peak position, fitting parameters, and the extrapolation of the peak position in absence of exchange to the coalescence temperature are reported in Table 3 3. The 19 F VT NMR sp ectra of compound 3 11 show very similar behavior to 3 8 with three peaks in a 1:1:2 ratio at room temperature corresponding to the type A axial ligands, type B

PAGE 82

82 axial ligands, and equatorial ligands respectively (Figure 3 15 ). The equatorial ligand peak s tarts splitting into two peaks (peaks 6 and 7) around 10C, with a peak splitting, in absence of exchange and extrapolated to the coalescence temperature, of 3.63 ppm corresponding to a free energy of activation of 11.0 kcal/mol. Like in the previous case it can be observed that the low temperature spectra display four main intense peaks and multiple low intensity peaks suggesting that one isomer may be favored, but many isomers are present in solution. The peak position, fitting parameters, and the extra polation of the peak position in absence of exchange to the coalescence temperature are reported in Table 3 4. Lastly, the 1 H VT NMR spectra of [Mn 8 Fe 4 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] ( 3 7 ) were collected between 20 and 70C. This interesting compound can be considered as part of the Mn 12 family since it is isostructural to the Mn 12 acetate with four of the Mn III ions being replaced by four Fe III which gives rise to interesting properties such as the modification of the spin ground state from 10 to 2. 57 The position of the Fe III ions in the molecule make this compound particularly interesting for our study since they are placed at the position where the water and the carboxylate ligands are exc hanging. Unlike the Mn III ions, the electronic configuration of Fe III which has an isotropic distribution of electrons in its d orbitals, does not lead to a Jahn Teller distortion. For this reason, the Fe O bond is expected to be shorter, and therefore st ronger than the equivalent Mn O bond Consequently, the water and carboxylate ligands bound to the Fe III ions are expected to be much less labile, and the fluxional process should be slowed considerably. The 1 H VT NMR spectra of 3 7 are shown in Figure 3 1 7 It is observed that four peaks are visible at room temperature, and the temperature needs to be increased in order to obtain the three peak pattern in a 1:2:1 ratio normally seen for the Mn 12 compounds at room temperature. This confirms the expectation that the fluxional process is slower, as it needs more thermal energy to proceed faster

PAGE 83

83 than the NMR time scale. Although the resolution of the spectra is not the highest, especially between the two coalescing peaks, it is possible to estimate the free ene rgy of activation for the fluxional process in this compound using the peak splitting extrapolated to the coalescence value of 16.1 kcal/mol is found for this compound, which is much larger than the free energy of activation for the flux ional processes in any of the other Mn 12 derivatives, as expected. The peak position, fitting parameters, and the extrapolation of the peak position in absence of exchange to the coalescence temperature are reported in Table 3 5. 3 .4 Conclusion The examination of the room temperature spectra of Mn 12 derivatives, and of the position of the water ligands on the Mn 12 core in crystal structures led to the hypothesis of a rapid fluxional process exchanging water and carboxylate ligands in solution. Decrea sing the temperature of the samples allowed the fluxional process to slow down enough that it became observable via NMR. Kinetic parameters such as the rate of the fluxional process, and its free energy of activation were calculated using variable temperat ure NMR data collected on multiple nuclei. The free energy of activation values were found to be around 10 kcal.mol 1 for all Mn 12 samples. Variable temperature 1 H NMR data collected on [Mn 8 Fe 4 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] ( 3 7 ) revealed that the fluxional process can be observed at room temperature when the bonds between the exchanging ligands and the metallic core are stronger. It can be observed that simultaneous occurrence of the fluxional process on several water/carboxylate ligand couples leads to the formatio n of the isomers observed in crystals structures. Therefore, it can be safely assumed that the fluxional process happens in the mother liquor, and that the crystallization "freezes" the process to form one of the isomers. No trend could be found to explain the formation of a particular isomer. The difference in energy between

PAGE 84

84 the isomers is most likely extremely small and the formation of one particular isomer is probably only slightly favored by the crystallization environment.

PAGE 85

85 Table 3 1. Peak positions, and fitting data for 3 4 Temperature (C) Peak 1 (ppm) Peak 4 (ppm) Peak 7 (ppm) Peak 8 (ppm) 115 151.9 114.3 14.7 25.2 110 144.4 109.7 15.7 23.0 105 134.8 104.4 16.7 19.6 100 129.0 101.0 17.1 17.7 95 122.6 97.4 17.4 15.7 90 115.8 93.2 17.6 13.5 85 110.2 89.7 17.7 11.4 80 104.8 86.4 17.7 9.6 75 100.1 83.5 17.6 7.9 70 95.7 80.8 17.5 6.2 65 91.4 78.4 17.0 4.6 50 (T c1 calc.) 80.70 71.73 40 (T c 2 calc.) 16.97 1.76 0 + a exp( b T) 0 39.63 41.36 16.97 20.64 a 1.30e+3 6.32e+2 1.68e+6 3.03e+2 b 1.55e 2 1.36e 2 48.5 1.19e 2 Table 3 2. Peak positions, and fitting data for 3 6 Temperature (C) Peak 5 (ppm) Peak 6 (ppm) 90 24.9 14.6 80 24.8 16.2 70 24.7 17.2 60 24.7 18.1 30 (T c calc.) 24.65 19.39 0 + a exp( b T) 0 24.65 20.23 a 1.93e+4 1.95e+3 b 0.061 0.032

PAGE 86

86 Table 3 3. Peak positions, and fitting data for 3 8 Temperature (C) Peak 1 (ppm) Peak 2 (ppm) 85 54.46 47.26 74 59.02 52.7 65 62.43 56.73 55 65.73 60.52 45 68.26 63.58 36 70.4 66.11 16 (T c calc.) 74.1 70.5 0 + a exp( b T) 0 (ppm) 84.4 82.5 a (ppm) 559.1 668.5 b (K 1 ) 0.0155 0.0156 Table 3 4. Peak positions, and fitting data for 3 11 Temperature (C) Peak 6 (ppm) Peak 7 (ppm) 85 134.18 139.74 75 132.6 137.34 65 130.28 134.66 55 128.5 132.6 45 127.05 130.82 35 125.83 129.51 10 (T c calc.) 122.8 126.43 0 + a exp( b T) 0 (ppm) 113.1 120.1 a (ppm) 150.2 335.9 b (K 1 ) 0.0104 0.0151

PAGE 87

87 Table 3 5. Peak positions, and fitting data for 3 7 Temperature (C) Peak 1 (ppm) Peak 2 (ppm) 20 28.93 28.37 30 28.84 28.0 40 28.23 27.91 55 (T c calc.) 27.83 27.63 0 + a exp( b T) 0 (ppm) 17.87 27.06 a (ppm) 28.66 1.01e+3 b (K 1 ) 3.22e 3 0.0228

PAGE 88

88 Figure 3 1. 1 H NMR spectrum of Mn 12 acetate ( 3 1 ) in CD 3 CN at room temperature (*: acetic acid and acetonitrile peaks). Figure 3 2. 1 H NMR spectrum of Mn 12 dichloroacetate ( 3 2 ) in CD 2 Cl 2 at room temperature (*: dichloroacetic acid, acetonitrile, hexanes, methylene chloride, and toluene peaks)

PAGE 89

89 Figure 3 3. 1 H NMR spectrum of Mn 12 propionate ( 3 3 ) in CD 2 Cl 2 at room temperature (*: propionic acid, hexanes, methylene chloride, and toluene peaks) Figure 3 4. Highlight of the different types of carboxylates by symmetry in Mn 12 acetate Water ligands are shown in teal A type carboxylate ligands are shown i n blue, B type in orange, and C type in red and green.

PAGE 90

90 A B C D Figure 3 5. Highlight of the different point groups found among Mn 12 derivatives in the solid state: A) Mn 12 acetate with S 4 symmetry B) Mn 12 benzoate with D 2 symmetry Mn 12 paramethyl benzoate with C 2 and C 1 symmetry (C and D respectively).

PAGE 91

91 Figure 3 6. Schematic representation of the seven isomers spanning four point groups available for Mn 12 derivatives with four water ligands. Green dots represent Mn III ions that can hav e water ligands. Figure 3 7. Schematic representation of the fluxional process between carboxylate type A and water ligands.

PAGE 92

92 Figure 3 8. Mirror planes virtually created by a rapid fluxional process exchanging A type carboxylates and water ligands.

PAGE 93

93 Figure 3 9. 2 H VT NMR spectra of [Mn 12 O 12 (O 2 CCD 2 CH 3 ) 16 (H 2 O) 4 ] ( 3 4 ) in D 10 Et 2 O from 175 to 50 ppm. The chemical shift is in ppm relative to deuterated TMS and the temperature in degrees Celsius. (* = Et 2 O).

PAGE 94

94 Figure 3 10. Low temperature peak positions, fit, and extrapolation for the 2 H VT NMR spectra of [Mn 12 O 12 (O 2 CCD 2 CH 3 ) 16 (H 2 O) 4 ] ( 3 4 ) in D 10 Et 2 O from 175 to 50 ppm.

PAGE 95

95 Figure 3 11. 1 H VT NMR (300 MHz) spectra of [Mn 12 O 12 (NO 3 ) 4 (O 2 CCH 2 C(CH 3 ) 3 ) 12 (H 2 O) 4 ] ( 3 6 ) in CD 2 Cl 2 from 10 to 60 ppm, showing the methylene proton region. The chemical shift is in ppm relative to TMS and the temperature in degrees Celsius.

PAGE 96

96 Figure 3 12. Low temperature peak positions, fit, and extrapolation for the 1 H VT NMR (300 MHz) spectra of [Mn 12 O 12 (NO 3 ) 4 (O 2 CCH 2 C(CH 3 ) 3 ) 12 (H 2 O) 4 ] ( 3 6 ) in CD 2 Cl 2 from 5 to 10 ppm, showing the tert butyl proton region.

PAGE 97

97 Figure 3 13. 19 F VT NMR (470 MHz) spectra of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 1 6 (H 2 O) 4 ] ( 3 8 ) in CD 2 Cl 2 from 130 to 30 ppm.

PAGE 98

98 Figure 3 14. Low temperature peak positions, fit, and extrapolation for the 1 9 F VT NMR spectra of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 1 6 (H 2 O) 4 ] ( 3 8 ) in CD 2 Cl 2

PAGE 99

99 Figure 3 15. 1 9 F VT NMR spectra of [Mn 12 O 12 ( 3,5 F 2 C 6 H 3 CO 2 ) 1 6 (H 2 O) 4 ] ( 3 11 ) in CD 2 Cl 2 from 170 to 20 ppm.

PAGE 100

100 Figure 3 16. Low temperature peak positions, fit, and extrapolation for the 1 9 F VT NMR spectra of [Mn 12 O 12 ( 3,5 F 2 C 6 H 3 CO 2 ) 1 6 (H 2 O) 4 ] ( 3 11 ) in CD 2 Cl 2

PAGE 101

101 Figure 3 17. 1 H VT NMR spectra of [Mn 8 Fe 4 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] ( 3 7 ) in MeCN from 50 to 8 ppm. The chemical shift is in ppm relative to TMS and the temperature in degrees Celsius. (* = MeCN and Et 2 O).

PAGE 102

102 Figure 3 18. Low temperature peak positions, fit, and extrapolation for the 1 H VT NMR spectra of [Mn 8 Fe 4 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] ( 3 7 ) in MeCN

PAGE 103

103 CHAPTER 4 SUBTLE ENVIRONMENTAL INFLUENCES ON THE MAGNETIC PROPERTIES OF M n 12 SINGLE MOLECULE MAGNETS 4 .1 Background The magnetic properties of the Mn 12 family of single molecule magnets (SMM) have been known since the early 1990s starting with Mn 12 acetate and Mn 12 benzoate 21 22 At that time these one of a kind molecular compounds drew the attention of both the chemistry and physics communities. Relatively f ew high nuclea rity magnetic molecular clusters were then known, and the formation of such a n aesthetically pleasing and complex structure from a relatively simple reaction was certainly a novelty. Additionally, the ability of this metal oxide compound to display quantum properties visible in a macroscopic measurement was a major advance in the pursuit of a better understanding of the quantum physics of nanomagnetism and in bridging the quantum and classical worlds. 16 53 75 By definition, the magnetic properties of single molecule magnets are intrinsic to the individual molecule rather than depending on long range interactions as for other types of magnets. Therefore, any intermolecular interactions that may be present within the lattice must be very weak in order to preserve the intrinsic single molecule properties of each SMM. F or this reason, crystals of t hese compounds are often described as collections of isolated magnetic molecules. Although that definition is most often sufficient since the intermolecular interactions are usually negligible, in some cases the close environm ent of the molecule must be considered to fully understand the observed magnetic properties of the compound. 76 82 Some interactions between SMM units have been found to cause the desirable effect known as exchange bia s ed quantum tunneling of mag netization (QTM) by which a magnetized unit influences slightly the magnetization of a neighbor unit by shifting its QTM resonance positions relative to the applied field ; this enables, in some cases the suppression of the

PAGE 104

104 tunneling step at zero field. 76 Many examples of this phenomenon have been found in compounds with intermolecular interactions mediated by hydrogen bonding. 77 79 83 85 A few attempts have been made to control intermolecular interactions by linking SMMs covalently with bridging ligands into 1 2 or 3 D networks 86 90 In most cases, the covalent linkage introduces too strong of an interaction between SM M units and the SMM behavior is lost but in some rare cases, the ex change remains weak enough to allow SMM behavior to be retained. 91 93 More recently, supramolecular aggregates of SMMs have been obtained through careful selection of the SMM building unit and design of the bridging ligand, which provides greater control over the structural arrangement of SM Ms and the resulting exchange biased QTM. 94 97 Herein is presented the magnetic study of a network of Mn 12 single molecule magnets formed with C F H hydrogen bonds mediating very weak intermolecular interactions, which strongly depen d on the solvation of the crystal lattice. 4 .2 Experimental Section 4 .2.1 Synthes i s of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) All manipulations were performed under aerobic conditions using materials as received, unless otherwise stated. [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] ( 4 1 ) was prepared as described elsewhere. 20 To a stirred solution of 4 1 (2.00 g, 0.97 mmol) in methylene chloride (100 mL) was added para fluorobenzoic acid (2.80 g, 20.0 mmol). The resulti ng solution was stirred overnight and the solvent was removed in vacuum. Toluene (25 mL) was added to the residue, and the solution was evaporated again to dryness. The addition and removal of toluene were repeated two more times. The residue was then redi ssolved in methylene chloride (50 mL), filtered, layered with acetonitrile, and left undisturbed for 10 days, during which time large black crystals of 4 2 8CH 3 CN slowly grew. The yield was 75% based on Mn. The crystals were maintained in

PAGE 105

105 mother liquor for the X ray crystallographic analysis, and collected by filtration, washed with acetonitrile, and dri ed under vacuum for other solid state studies. The identity of the product was confirmed by elemental analysis, IR spectral comparison, and X ray crystallog raphic analysis. Anal. Calcd (Found) for 4 2 3H 2 O (vacuum dried sample exposed to air) : C, 41.99 (41.71); H, 2.45 (2.49); N, 0 (0); F, 9.49 (9.63) %. Anal Calcd (Found) for solvent free 4 2 (vacuum dried sample not exposed to air): C, 42.72 (42.98); H, 2.30 (2.14); N, 0 .00 (0 .00 ) %. Se lected IR data (KBr pellet, cm 1 ): 3424(b), 3125(b), 1698(w), 1604(s), 1570(m ), 15 07(s), 1416(vs), 1352( s), 1300(w), 1233(s), 1153(s ), 1093(m), 1015(m), 857(m), 798(w), 778(s), 712(w), 692(m), 632(sb), 552(m). 4 .2.2 X R ay C rystallography X Ray i ntensity data for complex 4 2 8MeCN were collected by Dr. Khalil A. Abboud at 100 K on a Bruker DUO diffractometer using MoK radiation ( = 0.71073 ) and an APEXII CCD area detector. Raw data f rames were read by program SAINT 39 and integrated using 3D profiling algorithms. The resulting data were reduced to produce hkl reflections and their intensities and estima ted standard deviations. The data were corrected for Lorentz and polarization effects and numerical absorption corrections were applied based on indexed and measured faces. The structure was s olved and refined on F 2 in SHELXTL2013 39 using full ma trix least squares cycles of refinement. The non H atoms were refined with anisotropic thermal parameters and all of the H atoms were placed in idealized positions and refin ed riding on their parent atoms. The asymmetric unit consists of 1/8 of a Mn 12 cluster and a molecule of acetonitrile solvent disordered over two positions. The Mn 12 cluster is located on a 4 2 m position, thus only 1/8 of it exists in the asymmetric unit. The 4 2 m symmetry along with the disorder pro duces a complicated structure. There are th ree types of benzoate ligands. The C1

PAGE 106

106 ligand and its oxygen atoms are completely disordered and all of their symmetry equivalent s lie in the equatorial plane of the mol ecul e. The C1 ligand was refined in two parts with the planes of the parts being parallel to each other. T he C11 and C21 benzoates lie above and below the molecular equatorial plane. The C11 ligand has only the C 6 H 4 F part disordered and again refined in tw o parts that are related to ea ch other by simple reflection. The third ligand type, C21, is disordered around a mirror plane which lies on the bisector of the angle formed by three axial oxygen atoms positioned in a row. T he ligand occup ies a pair of them and the third is a water ligand T he other parts is where the ligand and water positions are reversed. The disorder of the ligands can be visualized in Figure 4 1. The acetonitrile solvent molecules were disordered and could not be modeled properly, thus p rogram SQUEEZE 98 a part of the PLATON 99 package of crystallographic software, was used to calculate the solvent disorder area and remove its contribution to the overall intensity data. In the final cycle of refinement, 4361 reflections (of which 3436 are observed with I > 2 (I)) were used to refine 147 parameters and the resulting R 1 wR 2 and S (goodness of fit) were 5.79 %, 15.76 % and 1.076 respectively. The refinement was carried out by minimizing the wR 2 function using F 2 rather than F values. R 1 is calculated to provide a referenc e to the conventional R value but its function is not minimized. The crystallographic data and structure refinement details are collected in Table 4 1 The vacuum dried crystals of complex 4 2 still diffract X rays well enough to obtain unit cell parameters. They were therefore obtained for a vacuum dried crystal not subsequently exposed to air, and a vacuum dried crystal with subsequent exposure to air. The unit cell parameters of all three samples are gathered in Table 4 2 The wet sample is referred to as 4 2 8MeCN from the formula obtained from th e crystal structure, the vacuum dried sample with

PAGE 107

107 no exposur e to air is referred to as 4 2 and the vacuum dried sample with exposure to air is referred to as 4 2 3H 2 O from the formula obtained from the elemental analysis. 4 .2.3 DC and AC Magnetometry Variable temperature magnetic susceptibility data down to 1.8 K were collected on a Quantum Design MPMS XL SQUID magnetometer equipped with a 7 T DC magnet at the University of Florida. Pascal's constants were used to estimate the diamagnetic contributions to the susceptibility, which were subtracted from the experimen tal susceptibility to give the molar magnetic susceptibility ( M ). Samples were embedded in solid eicosane to prevent torquing, unless otherwise stated. DC magnetic susceptibility data were collected between 5 and 300 K using a constant 0.1 T applied field AC magnetic susceptibility data were collected between 1.8 and 15 K, in the absence of applied DC field, with a 3.5 G field with oscillation frequencies up to 1500 Hz. Magnetization vs field and temperature data were collected over range s up to 7 T and 1 0.0 K, respectively and fitted using the program MAGNET written at Indiana University by E. R. Davidson 40 Low temperature (below 1.8 K) hysteresis loop studies were performed by Dr. Wolfgang Wernsdorfer at the Institut Nel of the CNRS in Grenoble, France, using an array of micro SQUIDs. 4 .2. 4 Other Studies Infrared spectra were recorded in the solid state (KBr pellets) on a Nicolet Nexus 670 FTIR spectrophotometer in the 400 4000 cm 1 range. Elemental analyses were performed at the in house facilities of the University of Florida Chemistry Department (C, H, N), and at Atlantic Microlab, Inc (C, H, N, F)

PAGE 108

108 4 .3 Results and Discu ssion 4 .3.1 Discussion of the S ynthesis The preparation of 4 2 8CH 3 CN is described in eq 4 1 [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] + 16 p FC 6 H 4 CO 2 H [ Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] + 16 CH 3 COOH ( 4 1 ) The synthesis of 4 2 wa s performed using the ligand substitution method for Mn 12 derivatives. 23 Complex 4 1 is initially insoluble in methylene chloride, and it is the solubility of the product ( 4 2 ), in combination with the presence of excess p FC 6 H 4 CO 2 H which drives the reaction. Full substitution is achieved by the removal of acetic acid as its toluene/acet ic acid azeotrope. 4 .3.2 Description of the S tructure of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] The structure of the individual molecule of 4 2 is shown in Figure 4 1 The individual molecule has S 4 symmetry as does the original Mn 12 acetate. The oxidation states of the Mn ions are confirmed through bond valence sum (BVS) calculations (Appendix E). The Jahn Teller axes of the Mn III ions are shown in black in Figure 4 2. A root mean square (RMS) deviation calculation was performed between the co res of 4 2 and Mn 12 acetate and those of Mn 12 benzoate to evaluate the differences induced by the ligand substitution. Lists of deviations between analog ous atoms in th e structures are given in Table 4 3 and 4 4 When compared to Mn 12 acetate, t he weighte d RMS deviation over the entire core of the molecule is only 0.1318 which shows that the core of 4 2 is very similar to the core of Mn 12 acetate When compared to Mn 12 benzoate, the weighted RMS deviation over the entire core of the molecule is only 0.1016 which shows that the core of 4 2 is even closer to the core of Mn 12 benzoate This can be visually observed with the superposition of the compared cores in Figure 4 3

PAGE 109

109 The compound crystallizes in the tetragonal space group I 4 2m The crystallization yields large black single crystals of 4 2 8 Me CN The crystals can grow as large as 5 mm along one side and are shaped as octahedr a where the top and bottom vertices are truncated in parallel planes (Figure 4 4 ). Inside the crystal, the molecules are arranged in a body centered cubi c like fashion All the molecules are in the same orientation, with collinear z axes and parallel equato rial planes (Figure 4 5) The fluorine atoms in the para positions of the ligands form hydrogen bonds with hydrogen atoms on ortho and meta positions of the ligands of neighboring molecules. The fluorine atoms on equatorial ligands (F 1) form hydrogen bonds with hydrogens H14 (meta hydrogens of the axial ligands bridging Mn III and Mn IV ions) and H23 (ortho hydrogens of the axial ligands bridging two Mn III ions) as seen in Figure 4 6 Each fluorine atom on the axial ligands bridging Mn III and Mn IV ions (F2) f orms hydrogen bonds with two equatorial meta hydrogens (H6) on different neighbor ing molecules (Figure 4 7 ). Each fluorine atom on the axial ligands bridging two Mn III ions (F3) forms hydrogen bonds with two equatorial ortho hydrogens (H7) on different lig ands of the same neighbor ing molecules (Figure 4 8 ). The hydrogen bond distances are gathered in Table 4 5 Overall, each Mn 12 unit interacts with twelve of its neig hbors t hrough hydrogen bond ing (48 total hydrogen bonds: 4 for each neighbor) A close observation of the unit cell parameters for samples 4 2 8 Me CN (wet sample), 4 2 (vacuum dried sample), and 4 2 3H 2 O (vacuum dried sample exposed to air) reveals that the drying process leads to only small changes for the a and b axes, whereas the c axis undergoes a much larger change from 23.9 to 20.7 The length of the c axis corresponds to the distance between two parallel Mn 12 units stacked directly on top of each other (Figure 4 9 ). It can be observed that the removal of eight acetonitrile mol ecules from the lattice during the drying process induces a collapse of the lattice in the c direction. The subsequent exposure of the

PAGE 110

110 crystals to air causes water molecules to be absorbed in the lattice, which induces a small lattice expansion in the c di rection from 20.7 to 21.6 4 .3.3 DC and AC Magnetometry Variable temperature DC magnetic susceptibility studies were performed on microcrystalli ne powder samples obtained from wet crystals of 4 2 8 Me CN rinsed with acetonitrile and dried with tissue paper, and vacuum dried crystals The samples were restrained in eicosane to prevent torquing, and the data were collected in a 0.1 T magnetic field in the 5.0 300 K range. After some preliminary measuremen ts, however, it became obvious that the magnetic response of the sample was highly dependent on the time of vacuum drying and subsequent exposure to air For this reason, s ystematic measurements were performed with a precise measure of the time of drying a nd subsequent air exposure 4 .3.3.1 Magnetic susceptibility of the wet sample Crystals taken out of the mother liquor and lightly dried with absorbent paper to remove the excess solvent were used to prepare a sample of wet crystalline powder of 4 2 8 Me CN The obtained DC susceptibilit y data obtained are shown in Figure 4 10 as a plot of M T vs. T and compared to the DC susceptibility data obtained for Mn 12 acetate B etween 300 and 50 K the data are identical for both complexes. The low tempe rature data below 50 K however, show significant differences. At its low temperature maximum, M T is typically around 50 55 cm 3 .K.mol 1 for Mn 12 derivatives, which corresponds to the spin only value for a well isolated S = 10 ground state with g slightly lower t han 2.0. F or 4 2 8 Me CN however M T is much larger with a value of 67.5 cm 3 .K.mol 1 at its maximum The same effect can be observed in the AC in phase susceptibility data before the drop due to slow magnetization relaxation (Figure 4 11) indicating that the large DC M T is not a result of the applied DC field This high value of the magnetic susceptibility at low temperature could indicate either a ground state spin higher than S

PAGE 111

111 = 10, or th e presence of weak intermolecular ferromagnetic interactions in the sample. The BVS calculations show that the oxidation states of the Mn ions are the same as in Mn 12 acetate, and the RMS deviation calculation s reveal that the geometry of the core of 4 2 is hardly modified by the ligand substitution. For these reasons, it is very unlikely that the ground state spin of the molecule is different from S = 10. The presen ce of a high number of hydrogen bonds throughout the crystal lattice, however, suggests th at magnetic exchange through interm olecular interactions is likely to affect the magnetism of 4 2 at low temperature and could be the cause of the unusually high M T It can be rationalized that the hydrogen bonds involving hydrogen atoms in the ortho pos ition lead to parallel alignment of the spins of the two Mn 12 units it links, whereas hydrogen bonds involving hydrogen atoms in the meta position lead to antiparallel alignment of the spins of the two Mn 12 units it links. As can be seen i n Figure 4 12 A a contact shift by delocalization from the Mn 12 unit onto the system of the ligand leads to spin up density on the ortho carbon, which causes spin down density on the ortho hydrogen by spin polarization. The spin down density of the hydrogen atom induce s a spin up density in the p orbital of the fluorine atom through direct overlap mediated by the hydrogen bond, which is then transferred to the Mn 12 unit through delocalization, resulting in a parallel alignment of the spins in the two Mn 12 units. The same mechanism is involved in the magnetic exchange that occurs through the meta hydrogen hydrogen bonds, but the meta carbon is at a node of the system, therefore, there is no direct delocalization of spin density onto it. The delocalizati on of the Mn 12 spin density onto the ligand leads to spin up density on both the ortho and para carbon atoms which causes spin down density on the meta carbon, and spin up density on the meta hydrogen by spin polarization. The contact shift by delocali zation mechanism is known to result in a greater spin density on the carbon and hydrogen atoms at the ortho and para positions than at the meta

PAGE 112

112 positions, which explains why the ferromagnetic intermolecular interactions through the ortho hydrogen bonds dom inate. 100 The coupling constant characterizing the exchange mediated by each hydrogen bond is most likely very small, but the high number of path s providing the same exchange renders the overall effect visible. It should be noted that the in phase data of the AC susceptibility shows a slightly positive slope between 15 and 7 K confirming the presence of net ferromagnetic interactions at low temperature. Multifrequency AC measurements were performed on the same wet sample in order to determine the kinetic parameters characterizing the magnetic relaxation process at low temperature (Figure 4 1 3 ). The obtained relaxation rate ( ) vs. temp erature data can be fitted to an Arrhenius equation (Figure 4 1 4 ) and gives the effective barrier to the reversal of magnetization ( U eff ) and the pre exponential factor ( 0 ) characteristic of the relaxation time (eq 4 2). For 4 2 8 Me CN U eff = 59.3 K, an d 0 = 3.01 10 9 s. ( 4 2) The magnetization vs. field data were also collected on 4 2 8 Me CN (Figure 4 1 5 ). No acceptable fit could be obtained most likely due to the presence of multiple Jahn Teller isomers with different anisotropy parameters throughout the sample ( vide infra ) 4 .3.3.2 Drying and exposure to air: time depend ence study In order to elucidate the mechanism responsible for the change in magnetism induced by drying and exposure to air of the sample, a series of measurements were performed on 4 2 at different stages of the sample collection process. One large batch of 4 2 was prepared and used for all samples. The crystals were crushed while still in the mother liquor to prevent solvent loss, and an aliquot of the powder was used to prepare the wet sample ( 4 2 8 Me CN ). The powder was then dried under vacuum, and two more al iquots were taken after an hour and two hours

PAGE 113

113 respectively. The dried powder was then exposed to air and samples were prepared after one, four, sixteen, and thirty two hours of exposure to air. For each sample, the powder was weighed and quickly embedded i n eicosane to stop any solvent gain or loss. The DC magnetic susceptibility data for the wet sample, and the two vacuum dried samples not exposed to air are gathered in Figure 4 1 6 U pon drying, the maximum M T value at low temperature decreases to 49.9 an d 47.5 cm 3 .K.mol 1 for the one and two hours vacuum dried samples respectively. These values are even lower than the value of 50 55 cm 3 .K.mol 1 usually observed for Mn 12 derivatives suggesting that drying the crystalline powder causes the net ferromagnetic intermolecular interaction to become net antiferromagnetic interaction This trend is also observed in the AC magnetic susceptibility d ata for the same set of samples (Figure 4 1 7 ) and it can be observed that the slightly positive slope in M T between 15 and 7 K, seen for the wet sample, decreases and becomes very slightly negative in the vacuum dried samples, once again suggesting the transformation of the net ferromagnetic interaction into net antiferromagnetic interaction This can be cor related to the lattice collapsing upon drying that is observed in the unit cell parameters. This change in spacing between Mn 12 planes most likely causes the original hydrogen bonds to be modified, therefore modifying the intermolecular interactions they c arry. A multifrequency AC study was performed on the two hour vacuum dried sample in order to determine the kinetic parameters characterizing the magnetic relaxation process at low temperature (Figure 4 1 8 ). The obtained relaxation rate ( ) vs. temperature data can be fitted to an Arrhenius equation (Figure 4 1 9 ), and gives the effective barrier to the reversal of magnetization ( U eff ) and the pre exponential factor ( 0 ) characteristic of the relaxation time (eq 4 2). For the solvent free 4 2 U eff = 60.6 K, and 0 = 3.07 10 9 s.

PAGE 114

114 The DC magnetic susceptibility data for samples with different times of exposure to air are shown i n Figure 4 20 The reverse trend from the previous set of samples can be observed: as the sample is exposed to air for an increasin g amount of time, the observed maximum M T value at low temperature increases. The M T maxima values are 47.5, 50.9, 53.9, 59.0, and 61.2 cm 3 .K.mol 1 for the sample not exposed to air (two hour vacuum dried sample), and the samples exposed to air for one, four, sixteen, and thirty two hours respectively. Additionally, the elemental analysis of a sample of 4 2 exposed to air for over a week shows that water is absorbed in the lattice to reach the equilibrium formula of 4 2 3H 2 O. This suggests that the absor bed water molecules help re expand the lattice, and that some of the ferromagnetic intermolecular interactions are able to be re formed. The M T maximum for the sample at equilibrium is, however, still lower than the M T maximum for the wet sample, which s uggests that the number of intermolecular ferromagnetic interaction pathways remains smaller in 4 2 3H 2 O than in 4 2 8 Me CN This trend is once again confirmed by the AC in phase susceptibility data which shows an increase in M T with increased exposure to air along with an increase of its slope between 15 and 7 K (Figure 4 2 1 ) A multifrequency AC study was performed on the last sample (thirty two hour exposure to air) in order to determine the kinetic parameters characterizing t he magnetic relaxation process at low temperature (Figure 4 2 2 ). The obtained relaxation rate ( ) vs. temperature data can be fitted to an Arrhenius equation (Figure 4 2 3 ), and gives the effective barrier to the reversal of magnetization ( U eff ) and the pre exponential factor ( 0 ) characteristic of the relaxation time (eq 4 2). For 4 2 3H 2 O U eff = 59.6 K, and 0 = 2.99 10 9 s. Finally, the sample was vacuum dried once again and re exposed to acetonitrile for three days in an attempt to observe whether acetonitrile molecules would re enter the lattice and give

PAGE 115

115 rise to similar magnetism t o the wet sample ( 4 2 8 Me CN ). The DC and AC susceptibility data for that final sample compared to the previous sample (32 hour exposure to air) are shown in Figure 4 2 4 and 4 2 5 No significant change in the magnetic susceptibility data is observed between these two samples either s uggesting that the vacuum did not extract the water molecules from the lattice, and therefore not creating enough space for acetonitrile molecules to enter, or that the lattice was not able to expand back to its original dimensions. The sample no longer co nsisted of single crystals at that point, so unit cell parameters could not be c ollected in order to verify either hypothesis. It is observed that some samples show two out of phase signals of varying relative intensity, which attests to the presence of tw o isomers with different relaxation rates. This effect is known as Jahn Teller isomerism and is commonly observed in Mn 12 derivatives. 31 36 101 It arises from internal pressures within the crystal lattice forcing the Jahn Teller axes of some Mn III ions to change orientation, therefore modifying the anisotropy of the molecule. This effect is clearly related to the modification of the solvent content of the sample, although a direct correlation is not evident. 4.3.3.3 Evaluation of the strength of the net ferromagnetic interaction In collaboration with Dr. Rodolphe Clrac at the Research Center Paul Pascal of the CNRS in Bordeaux, France, the strength of the net ferromagnetic intermolecular interaction was estimated. The DC M T data of 4 2 8 Me CN was compared to the DC M T data of Mn 12 benzoate. Using a mean field approximation, the net strength of the intermolecular interactions can be extracted. This method allowed to simulate the DC M T data of 4 2 8 Me CN with good accuracy (Figure 4 26), and gave an estimate of the coupling constan t of 70 mK. As expected, this value is very small.

PAGE 116

116 4.3.3.4 Hysteresis studies The studies were performed on an aligned single crystal of 4 2 8MeCN using a micro SQUID apparatus, and the resulting hysteresis loops at 3.6 K and with sweep rate s from 0.002 to 0.280 T/s are shown in Figure 4 27 Hysteresis is observed down to at least a sweep rate of 0.002 T/s at 3.6 K Additionally the loops exhibit a step around zero field due to quantum tunneling of magnetization (QTM) To determine with accuracy the field at which the step occurs, a differential plot of the magnetization vs. applied field was built (Figure 4 27). It allows observation that the step occurs at 0.012 T, indicating the presence of a weak net antiferromagne tic intermolecular interaction in the direction of the applied field. At 3.6 K each Mn 12 unit is treated as a spin of m s = 4. An antiferromagnetic coupling strength of J = 0.0007 cm 1 ( 1 0 mK) was calculated using eq. 4 3. (4 3) It should be noted that the magnetic field is applied along the z axis of the molecule, which corresponds to the c axis of the crystal lattice. Therefore, only the intermolecular interactions with a c axis component in their direction will affect the resul ts of this experiment. In particular, the ferromagnetic interactions mediated by the F1 H23 hydrogen bonds, which are in the ab plane, will have no effect on the position of the step. The intermolecular interactions with a component in the c direction re sult in a net antiferromagnetic coupling, which explains the small negative shift of the QTM step. 4 .4 Conclusion The magnetic properties of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) at low temperature are strongly influenced by intermolecular interaction s mediated by numerous hydrogen bonds throughout the crystal lattic e. Particularly hydrogen bonds involving benzoate ortho hydrogens

PAGE 117

117 promote net ferromagnetic intermolecular interactions which cause the magnetic susceptibility value at low temperature to be much higher than expected for Mn 12 derivatives. This effect is related to the distance between Mn 12 units within the lattice, and is greatly affected by the modification of the solvent content of the lattice. The loss of acetonitrile molecules upon drying of the crystal induces a collapse of the lattice and the disappearance of some ferromagnetic interactions. The subsequent insertion of water molecules from atmospheric exposure allows the lattice to slightly re expand, and re gain some ferromagnetic inter actions. The molecular nature of this material allows for the identification and analysis of these effects which may be expandable to other type of molecule based magnetic nanomaterials.

PAGE 118

118 Table 4 1. Crystal data and structure refinement parameters for comp lex 4 2 8MeCN 4 2 8MeCN formula a C 128 H 96 F 16 Mn 12 N 8 O 48 fw, g mol 1 3477.40 space group I 4 2m a, 17.4099(12) b, 17.4099(12) c, 23.8737(17) 90 deg 90 90 V, 3 7236.2(11) Z 2 T, K 100(2) radiation, b 0.71073 calc g cm 3 1.596 1 1.114 R1 c, d 0.0579 wR2 e 0.1576 a Including solvent molecules. b Graphite monochromator. c I I ). d R 1 = 100 (|| F o | | F c ||)/ | F o |. e wR 2 = 100[ [ w ( F o 2 F c 2 ) 2 ]/ [ w ( F o 2 ) 2 ]] 1/2 w = 1/[ 2 ( F o 2 ) + [( ap ) 2 + bp ], where p = [max ( F o 2 O) + 2 F c 2 ]/3. Table 4 2. Unit cell parameters of crystals of complex 4 2 8MeCN, 4 2 and 4 2 3H 2 O 4 2 8MeCN a 4 2 b 4 2 3H 2 O c a () 17.4 17.0 17.4 b () 17.4 17.0 17.4 c () 23.9 20.7 21.6 90 90 90 90 90 90 90 90 90 V ( 3 ) 7236 5982 6504 Crystal system Tetragonal I Tetragonal P Tetragonal P a As isolated from the mother liquor b Vacuum dried for two hours c Vacuum dried and exposed to air for 32 hours

PAGE 119

119 Table 4 3 List of root mean square deviations for analog ous atoms between 4 2 and Mn 12 acetate Atom Label Deviation () Mn 1 0.050 M n 2 0.278 M n 3 0.112 O1 0.044 O2 0.063 O3 0.078 Weighted RMS Deviation = 0.1318 Table 4 4. List of root mean square deviations for analogous atoms between 4 2 and Mn 12 benzoate Atom Label Deviation () Mn1 0.0 25 Mn2 0. 045 Mn3 0. 098 O1 0.0 20 O2 0.0 91 Mn1' 0.0 45 Mn2' 0.177 Mn3' 0.016 O1' 0.025 O2' 0.031 Mn1'' 0.029 Mn2'' 0.313 Mn3'' 0.094 O1'' 0.045 O2'' 0.022 Mn1''' 0.050 Mn2''' 0.211 Mn3''' 0.046 O1''' 0.024 O2''' 0.097 O3 0.066 O3' 0.046 O3'' 0.062 O3''' 0.112 Weighted RMS Deviation = 0.1 016

PAGE 120

120 Table 4 5 Distances () between the fluorine atoms and the hydrogen atom s with which they form hydrogen bonds and corresponding carbon fluorine bond distances. F1 H14 2.06 F1 H23 2.39 F2 H6 2.16 F3 H7 2.39 C5 F1 1.33 C15 F2 1.28 C25 F3 1.28

PAGE 121

121 A B C Figure 4 1. Structure of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) A) F ull molecule, B) structure of the core with atom labels C) full molecule with disorder of the ligands Mn III are shown in blue, Mn IV in green, O in red, C in grey, F in neon green, and water ligands in teal. Hydrogen atoms are omitted for clarity.

PAGE 122

122 A B Figure 4 2. Two side views of the structure of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ). The bonds in black show the Jahn Teller axes of Mn III atoms. A B Figure 4 3 Overlay of the core [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) (pink) with Mn 12 acetate (green) (A), and with Mn 12 benzoate (green) (B) The oxygen atoms and the first two carbon atoms of the carboxylate ligands are shown for clarity but were not included in the RMS comparison.

PAGE 123

123 A B Figure 4 4 Visualization of the crystals of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ). A) Picture of a batch of crystals B) S chematic representation of the crystal shape 1 mm

PAGE 124

124 A B C Figure 4 5. Packing diagram of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] 8 Me CN ( 4 2 8 Me CN ). A) Side view of the crystal packing. The dotted lines represent hydrogen bonding. B) and C) s ide and top view, respectively, of the crystal packing. The carbon, hydrogen, and fluorine atoms are omitted for clarity.

PAGE 125

125 Figure 4 6 Visua lization of the h ydrogen bonds involving the fluorine atoms on equatorial ligands (F1).

PAGE 126

126 Figure 4 7 Visualization of the hydrogen bonds involving the fluorine atoms on axial ligands bridging a Mn III and a Mn IV ion (F2).

PAGE 127

127 Figure 4 8 Visualization of t he hydrogen bonds involving the fluorine atoms on axial ligands bridging two Mn III ions (F3). Figure 4 9 Schematic representation of the changes in unit cell parameters of 4 2 induced by drying under vacuum and subsequent exposure to air

PAGE 128

128 Figure 4 10 Comparison of DC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] 8 Me CN ( 4 2 8 Me CN ) and Mn 12 acetate in a 0.1 T applied field. Figure 4 11 Comparison of AC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] 8 Me CN ( 4 2 8 Me CN ) and Mn 12 acetate. AC in phase (top) and out of phase (bottom).

PAGE 129

129 A B Figure 4 12. Scheme of the interactions leading to the spin alignment between two Mn 12 units linked by a hydrogen bond. A) For a hydrogen bond involving a hydrogen atom in the ortho position. B) For a hydrogen bond involving a hydrogen atom in the meta position.

PAGE 130

130 Figure 4 1 3 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] 8 Me CN ( 4 2 8 Me CN ). AC in phase (top) and out of phase (bottom). Figure 4 1 4 Arrhenius plot for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] 8 Me CN ( 4 2 8 Me CN ). The fitting parameters give an effective barrier U eff /k B = 59.3 K.

PAGE 131

131 Figure 4 1 5 Reduced magnetization plot for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] 8 Me CN ( 4 2 8 Me CN ). Figure 4 1 6 Comparison of DC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) at different stages of drying.

PAGE 132

132 Figure 4 1 7 Comparison of AC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) at different stages of drying. AC in phase (top) and out of phase (bottom).

PAGE 133

133 Figure 4 1 8 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ). AC in phase (top) and out of phase (bottom). Figure 4 1 9 Arrhenius plot for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ). The fitting parameters give an effective barrier U eff /k B = 60.6 K.

PAGE 134

134 Figure 4 20 Comparison of DC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) at different times of exposure to air. Figure 4 2 1 Comparison of AC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) at different times of exposure to air. AC in phase (top) and out of phase (bottom).

PAGE 135

135 Figure 4 2 2 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] 3H 2 O ( 4 2 3H 2 O). AC in phase (top) and out of phase (bottom). Figure 4 2 3 Arrhenius plot for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] 3H 2 O ( 4 2 3H 2 O). The fitting parameters give an effective barrier U eff /k B = 59.6 K.

PAGE 136

136 Figure 4 2 4 Comparison of DC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) exposed to air for 32 hours, and 4 2 re exposed to MeCN for 3 days. Figure 4 2 5 Comparison of AC susceptibility data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( 4 2 ) exposed to air for 32 hours, and 4 2 re exposed to MeCN for 3 days. AC in phase (top) and out of phase (bottom).

PAGE 137

137 Figure 4 26. Simulation of the DC M T data of 4 2 8 Me CN from the DC M T data of Mn 12 benzoate and estimation of the coupling constant. A B Figure 4 27. Hysteresis measurement data for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] 8 Me CN ( 4 2 8 Me CN ). A) Hysteresis loops at 3.6 K and different scan rates. B) Differential plot visualizing the position of the step.

PAGE 138

138 APPENDIX A ADDITIONAL NMR SPECTRA Figure A 1. 1 H VT NMR spectra of [Mn 12 O 12 (O 2 CCH 2 C(CH 3 ) 3 ) 16 (H 2 O) 4 ] ( 3 5 ) in CD 2 Cl 2 from 10 to 15 ppm, showing the t ert butyl proton region. The chemical shift is in ppm relative to TMS and the temperature in degrees Celsius. (* = CH 2 Cl 2 HO 2 CCH 2 C(CH 3 ) 3 and MeNO 2 )

PAGE 139

139 Figure A 2. 1 H VT NMR spectra of [Mn 12 O 12 (O 2 CCH 2 C(CH 3 ) 3 ) 16 (H 2 O) 4 ] ( 3 5 ) in CD 2 Cl 2 from 22 to 120 ppm, showing the methylene proton region. The chemical shift is in ppm relative to TMS and the temperature in degrees Celsius.

PAGE 140

140 Figure A 3. 1 H VT NMR spectra of [Mn 12 O 12 (NO 3 ) 4 (O 2 CCH 2 C(CH 3 ) 3 ) 12 (H 2 O) 4 ] ( 3 6 ) in CD 2 Cl 2 from 5 to 10 ppm, showing the tert butyl proton region. The chemical shift is in ppm relative to TMS and the temperature in degrees Celsius. (* = CH 2 Cl 2 and HO 2 CCH 2 C(CH 3 ) 3 )

PAGE 141

141 Figure A 4. 19 F VT NMR (470 MHz) spectra of [Mn 12 O 12 ( p CF 3 C 6 H 4 CO 2 ) 1 6 (H 2 O) 4 ] ( 3 9 ) in CD 2 Cl 2 from 100 to 0 ppm.

PAGE 142

142 Figure A 5. 19 F VT NMR (470 MHz) spectra of [Mn 12 O 12 ( o FC 6 H 4 CO 2 ) 1 6 (H 2 O) 4 ] ( 3 10 ) in CD 2 Cl 2 from 140 to 20 ppm.

PAGE 143

143 Figure A 6. 19 F VT NMR (470 MHz) spectra of [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 1 6 (H 2 O) 4 ] ( 3 8 ) in CD 2 Cl 2 from 130 to 30 ppm around the coalescence temperature

PAGE 144

144 APPENDIX B OTHER NEW Mn 12 DERIVATIVES The Mn 12 family is known to be one of the largest family of single molecule magnets. All of the Mn 12 derivatives can be obtained from the original [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] 2CH 3 COOH 4H 2 O ( B 1 ), through a ligand substitution method developed twenty years ago (eq B 1). 23 [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] + 16 RCOOH [ Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ] + 16 CH 3 COOH (B 1) Almost any carboxylic acid can be used to replace the acetate ligands on B 1 and the obtained derivative's properties highly depend on the nature of the substituting ligand. Some ligands have helped develop and explore the electrochemical properties of Mn 12 and many have allowed to study diverse aspects of its magnetic properties. Each new derivative is an opportunity to expand our knowledge of single molecule magnetism and to understand magnetic properties potentially applicable to other magnetic nanomat erials. Syntheses All manipulations were performed under aerobic conditions using materials as received, unless otherwise stated. [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] 2CH 3 COOH 4H 2 O ( B 1 ) was prepared as described elsewhere. 20 Infrared spectra were recorded in the solid state (KBr pellets) on a Nicolet Nexus 670 FTIR spectrophotometer in the 400 4000 cm 1 range. Elemental analyses were performed at the in house facilities of the University of Florida Chemistry Depart ment, or within Complete Analysis Laboratories, Inc. [Mn 12 O 12 ( 3,5 (CH 3 O) 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] (B 2). To a stirred solution of complex B 1 ( 0.5 g, 0.24 mmol) in acetonitrile ( 40 mL) was added ( 3,5) dimethoxybenzoic acid ( 1 .8 2 g, 1 0.0 mmol). While t he resulting solution was stirred

PAGE 145

145 overnight a dark brown precipitate formed The solution was filtered, and the residue redissolved in methylene chloride (25 mL) filtered, and layered with diethyl ether. After3 5 days, black crystals of B 2 appeared on the sides and bottom of the vial. The yield was 85 % based on Mn. The crystals were collected by filtration, washed with acetonitrile and ether and dried under vacuum. The identity of the product was confirmed by elemental analysis, and IR spectroscopy Anal. Calcd (Fo und) for B 2 : C, 45.25 ( 45.46 ); H, 4.01 ( 3.93 ); N, 0 ( 0 ) %. Se lected IR data (KBr pellet, cm 1 ): 3440(b), 3000(w), 2939(w), 2838(w), 2076(w), 1578(s), 1515(m), 1449(s), 1405(vs), 1292(w), 1249(w), 1204(s), 1156(s), 1063(s), 991(m), 942(m), 846(m), 782(s), 756(m), 676(m), 613(m), 523(wb) [Mn 12 O 12 ( p t BuC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] (B 3 ). To a stirred solution of complex B 1 ( 0.5 g, 0.24 mmol) in methylene chloride ( 40 mL) was added para tert butyl benzoic acid ( 1 43 g, 8 .0 mmol). The resulting solution was stirred overnight and the solvent was removed in vacuum. Toluene ( 15 mL) was added to the residue, and the solution was evaporated again to dryness. The addition and removal of toluene was repeated two more times. The remaining solid was redissolved in methylene chloride ( 25 mL), filtered, layered with acetonitrile nitrom ethane, ethanol, or methanol and left undisturbed for 5 days, during which time black crystals of B 3 slowly grew. The yield was 82 % based on Mn The crystals were collected by filtration, washed with the layering solvent and dried under vacuum. The iden tity of the product was confirmed by elemental analysis, and IR spectroscopy Anal. Calcd (Found) for B 3 : C, 56.24 ( 56.5 ); H, 5.79 ( 5.98 ); N, 0 ( 0 ) %. Se lected IR data (KBr pellet, cm 1 ): 3129(vb), 2964(m), 2906(w), 2870(w), 1693(vw), 1610(m), 1597(m), 1550(m), 1524(m), 1461(w), 1411(vs), 1268(m), 1193(m), 1145(w), 1106(w), 1017(m), 857(m), 783(s), 712(s), 655(w), 616(m), 545(w), 504(w)

PAGE 146

146 [Mn 12 O 12 ( 2,4,6 (CH 3 ) 3 C 6 H 2 CO 2 ) 16 (H 2 O) 4 ] (B 4 ). To a stirred solution of complex B 1 ( 0.5 g, 0.24 mmol) in acetonitrile ( 40 mL) was added (2,4,6 ) trimethylbenzoic acid ( 1 31 g, 8 .0 mmol). The resulting solution was stirred overnight and the solvent was removed in vacuum. Toluene ( 15 mL) was added to the residue, and the solution was evaporated again to dryness. The addition and removal of toluene was repeated t hree more times. The remaining solid was redissolved in methylene chloride ( 2 0 mL), filtered, layered with ethanol, or methan ol and left undisturbed for 5 days, during which time black crystals of B 4 slowly grew. The yield was 67 % based on Mn The crystals were collected by filtration, washed with the layering solvent and dried under vacuum. The identity of the product was co nfirmed by elemental analysis, and IR spectroscopy Anal. Calcd (Found) for B 4 : C, 54.37 ( 54.62 ); H, 5.25 ( 4.95 ); N, 0 ( 0 ) %. Se lected IR data (KBr pellet, cm 1 ): 3425 (vb), 3130 (vb), 1610 (m), 1582 (m), 1512 (m), 1435 (s), 1399 (vs), 1184 (m), 1117 (m), 850 (m), 823 (w), 798 (w), 714 (w), 661 (m), 613 (m), 560 (m), 535 (w), 479 (w) [Mn 12 O 12 ( p ClC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] (B 5 ). To a stirred solution of complex B 1 ( 0.2 g, 0.1 mmol) in methylene chloride ( 20 mL) was added para chlorobenzoic acid ( 0 31 g, 2 .0 mmol). The resulting solution was stirred overnight and the solvent was removed in vacuum. Toluene ( 10 mL) was added to the residue, and the solution was evaporated again to dryness. The addition and removal of toluene was repeated two more times. The r emaining solid was redissolved in methylene chloride ( 25 mL), filtered, layered with hexanes and left undisturbed for 4 days, during which time black crystals of B 5 slowly grew. The yield was 72 % based on Mn The crystals were collected by filtration, washed with hexanes and dried under vacuum. The identity of the product was confirmed by elemental analysis, and IR spectroscopy Anal. Calcd (Found) for B 5 2.5 p

PAGE 147

147 ClC 6 H 4 CO 2 H : C, 40.89 ( 40.86 ); H, 2.24 ( 2.51 ); N, 0 ( 0 ) %. Se lected IR data (KBr pellet, cm 1 ): 3431 (vb), 3134 (vb), 1696 (w), 1593 (s), 1543 (m), 1405 (vs), 1348 (s), 1281 (w), 1174 (m), 1141 (w), 1090 (s), 1016 (s), 853 (m), 770 (s), 731 (w), 687 (m), 658 (m), 616 (m), 546 (b), 476 (w), 432 (w) [Mn 12 O 12 ( O 2 CCCl 3 ) 16 (H 2 O) 4 ] (B 6 ). To a stirred solution of complex B 1 ( 0.2 g, 0.1 mmol) in methylene chloride ( 20 mL) was added trichloroacetic acid ( 0 33 g, 2 .0 mmol). The resulting solution was stirred overnight and the solvent was removed in vacuum. Toluene ( 10 mL) was added to the res idue, and the solution was evaporated again to dryness. The addition and removal of toluene was repeated two more times. The remaining solid was redissolved in methylene chloride ( 20 mL), filtered, layered with hexanes and left undisturbed in the fridge f or 5 days, during which time black crystals of B 6 slowly grew. The yield was 6 5 % based on Mn The crystals were collected by filtration, washed with hexanes and dried under vacuum. The identity of the product was confirmed by elemental analysis, and IR s pectroscopy Anal. Calcd (Found) for B 6 : C, 10.91 ( 10.82 ); H, 0.23 ( 0.35 ); N, 0 ( 0 ) %. Se lected IR data (KBr pellet, cm 1 ): 3492 (vb), 2960 (w), 1678 (vs), 1603 (vs), 1361 (vs), 967 (m), 858 (vs), 837 (vs), 748 (vs), 690 (vs), 558 (m), 519 (w), 459 (m) [Mn 12 O 12 ( 3,5 F 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] (B 7 ). To a stirred solution of complex B 1 ( 0.2 g, 0.1 mmol) in methylene chloride ( 20 mL) was added (3,5) difluorobenzoic acid ( 0 32 g, 2 .0 mmol). The resulting solution was stirred overnight and the solvent was removed in vacuum. Toluene ( 10 mL) was added to the residue, and the solution was evaporated again to dryness. The addition and removal of toluene was repeated two more times. The remaining solid was redissolved in methylene chloride ( 20 mL), filtered, layered wit h hexanes and left undisturbed for 2 days, during which time black crystals of

PAGE 148

148 B 7 slowly grew. The yield was 85 % based on Mn The crystals were collected by filtration, washed with hexanes and dried under vacuum. The identity of the product was confirme d by elemental analysis, and IR spectroscopy Vacuum dried crystals of B 7 were analyzed as B 7 (3,5) F 2 C 6 H 3 CO 2 H : Anal. Calcd (Found): C, 39.75 (39.60); H, 1.68 (1.5); N, 0 (0) %. Selected IR data for B 7 (KBr disk, cm 1 ) : 3102(b), 1621(m), 1584(s), 1524(m), 1473(s), 1442(s), 1402(vs), 1356(s), 1288(w), 1212(w), 1123(s), 989(s), 962(m), 890(m), 858(m), 781(m), 763(m), 713(w), 662(m), 616(w), 552(w), 509(w), 436(w). [Mn 12 O 12 ( o FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] (B 8 ). To a stirred solution of complex B 1 ( 0.2 g, 0.1 mmol) in methylene chloride ( 20 mL) was added ortho fluorobenzoic acid ( 0 28 g, 2 .0 mmol). The resulting solution was stirred overnight and the solvent was removed in vacuum. Toluene ( 10 mL) was added to the residue, and the solution was evaporated again to dryness. The addition and removal of toluene was repeated two more times. The remaining solid was redissolved in methylene chloride ( 20 mL), filtered, layered with hexanes and left un disturbed in the fridge for 2 days, during which time black crystals of B 8 slowly grew. The yield was 85 % based on Mn The crystals were collected by filtration, washed with hexanes and dried under vacuum. The identity of the product was confirmed by ele mental analysis, and IR spectroscopy Vacuum dried crystals of B 8 were analyzed as solvent free : Anal. Calcd (Found): C, 42.71 (42.68); H, 2.30 (2.10); N, 0 (0) %. Selected IR data for B 8 (KBr disk, cm 1 ) : 3148(b), 1714(m), 1613(vs), 1596(vs), 1540(s), 1 520(s), 1488(s), 1454(s), 1412(vs), 1267(w), 1228(m), 1164(m), 1098(m), 1033(w), 956(vw), 862(m), 843(w), 800(m), 756(s), 696(m), 659(s), 619(m), 554(m), 523(w).

PAGE 149

149 [Mn 12 O 12 ( p CF 3 C 6 H 4 CO 2 ) 16 (H 2 O) 4 ] (B 9 ). To a stirred solution of complex B 1 ( 0.2 g, 0.1 mmol) in methylene chloride ( 20 mL) was added para trifl uoromethylbenzoic acid ( 0 38 g, 2 .0 mmol). The resulting solution was stirred overnight and the solvent was removed in vacuum. Toluene ( 10 mL) was added to the residue, and the solution was evaporate d again to dryness. The addition and removal of toluene was repeated two more times. The remaining solid was redissolved in methylene chloride ( 20 mL), filtered, layered with hexanes and left undisturbed for 10 days, during which time black crystals of B 9 slowly grew. The yield was 68 % based on Mn The crystals were collected by filtration, washed with hexanes and dried under vacuum. The identity of the product was confirmed by elemental analysis, and IR spectroscopy Vacuum dried crystals of B 9 were analyzed as solvent free: Anal. Calcd (Found): C, 38.93 (38.84); H, 1.84 (1.51); N, 0 (0) %. Selected IR data for B 9 (KBr pellet, cm 1 ) : 3139(m), 1707(w), 1599(m), 1562(s), 1514(s), 1431(vs), 1326(vs), 1170(m), 1131(s), 1103(m), 1066(s ) 1019(m), 866(m), 784(m ) 769(w), 711(m), 656(b), 617(b), 547(m). [Mn 12 O 12 ( tma ) x ( O 2 CCH 3 ) 16 x (H 2 O) 4 ] n (B 10 ). To a stirred solution of complex B 1 ( 0.5 g, 0.25 mmol) in acetonitrile ( 20 mL) was added a solution of trimesic acid (tma H 3 ) ( 0 84 g, 4 .0 mmol) in ethanol (10 mL) The resulting solution was stirred for fifteen minutes, and left undisturbed overnight during which time a black gel formed The gel was collected by filtration, washed with ethanol, and dried under vacuum. The identity of the product was confirmed by elem ental analysis, and IR spectroscopy Anal. Calcd (Found) for B 10 : C, n/a ( 30.66 ); H, n/a ( 2.90 ); N, n/a ( 0 ) %. Se lected IR data (KBr pellet, cm 1 ): 3434 (vb), 3165 (vb), 1699 (s), 1624 (vs), 1555 (s), 1442 (s), 1373 (vs), 1271 (s), 1111 (m), 932 (m), 908 (w), 754 (s), 716 (s), 693 (s), 618 (m)

PAGE 150

150 [Mn 12 O 12 ( add ) x ( O 2 CCH 3 ) 16 x (H 2 O) 4 ] n (B 11 ). To a stirred solution of complex B 1 ( 0.1 g, 0.05 mmol) in acetonitrile ( 10 mL) was added a solution of (1,3) adamantanedicarboxylic acid (addH 2 ) ( 0 36 g, 1.6 mmol) in tetrahydrofuran (10 mL) The resulting solution was stirred for fifteen minutes, and left undisturbed overnight during which time a brown precipita te formed The precipitate was collected by filtration, washed with tetrahydrofuran, and dried under vacuum. The identity of the product was confirmed by elemental analysis, and IR spectroscopy Anal. Calcd (Found) for B 11 : C, n/a ( 42.29 ); H, n/a ( 3.61 ); N, n/a ( 0 ) %. Se lected IR data (KBr pellet, cm 1 ): 3421 (vb), 3133 (vb), 2907 (s), 2856 (m), 1699 (m), 1579 (m), 1402 (vs), 1312 (m), 1121 (m), 1047 (w), 975 (w), 956 (w), 876 (w), 674 (m), 614 (m), 518 (m) [Mn 12 O 12 ( iso ) x ( O 2 CCH 3 ) 16 x (H 2 O) 4 ] n (B 12 ). To a stirred solution of complex B 1 ( 0.1 g, 0.05 mmol) in acetonitrile ( 10 mL) was added a solution of isophtalic acid (isoH 2 ) ( 0 27 g, 1.6 mmol) in tetrahydrofuran (10 mL) The resulting solution was stirred for fifteen minutes, and left undisturbed over night during which time a brown precipitate formed The precipitate was collected by filtration, washed with tetrahydrofuran and diethyl ether, and dried under vacuum. The identity of the product was confirmed by elemental analysis, and IR spectroscopy An al. Calcd (Found) for B 12 : C, n/a ( 33.40 ); H, n/a ( 2.69 ); N, n/a ( 0 ) %. Se lected IR data (KBr pellet, cm 1 ): 3137 (vb), 1705 (m), 1609 (s), 1557 (s), 1481 (m), 1399 (vs), 1277 (w), 1162 (m), 1077 (w), 933 (b), 739 (m), 652 (m), 615 (m) DC and AC Magnetometry Variable temperature magnetic susceptibility data down to 1.8 K were collected on a Quantum Design MPMS XL SQUID magnetometer equipped with a 7 T DC magnet at the University of Florida. Pascal's constants were used to estimate the diamagnetic

PAGE 151

151 contributions to the susceptibility, which were subtracted from the experimental susceptibility to give the molar magnetic susceptibility ( M ). Samples were embedded in solid eicosane to prevent torquing, unless otherwise stated. DC magnetic susceptibility data were collected using a constant 0.1 T applied field between 5 and 300 K. AC magnetic susceptibility data were collected between 1.8 and 15 K, in absence of applied DC field, with a 3.5 G field oscillating at frequencies up to 1500 Hz. Magnetization v s field data were collected over a range of fields and temperatures, and fitted using the program MAGNET written at Indiana University by E. R. Davidson 40

PAGE 152

152 Figure B 1. DC magnetic susceptibility plot for [ Mn 12 O 12 (3,5 (CH 3 O) 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 2 ) in an applied 0.1 T field. Figure B 2 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( 3,5 (CH 3 O) 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 2 ). AC in phase (top) and out of phase (bottom).

PAGE 153

153 Figure B 3 Arrhenius plot for [Mn 12 O 12 ( 3,5 (CH 3 O) 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 2 ). F itting parameters : U eff /k B = 63 5 K ; 0 = 4.80 10 9 s Figure B 4. Reduced magnetization plot for [Mn 12 O 12 ( 3,5 (CH 3 O) 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 2 ). Fitting parameters: g = 1.9450 0.0014; D = 0.4183 0.0016 cm 1

PAGE 154

154 Figure B 5. DC magnetic susceptibility plot for [ Mn 12 O 12 ( p t BuC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 3 ) in an applied 0.1 T field. Figure B 6. AC susceptibility plots vs. temperature for [Mn 12 O 12 ( p t BuC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 3 ). AC in phase (top) and out of phase (bottom).

PAGE 155

155 Figure B 7 Arrhenius plot for [Mn 12 O 12 ( p t BuC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 3 ). F itting parameters : U eff /k B = 67 0 K ; 0 = 5.05 10 9 s Figure B 8. Reduced magnetization plot for [Mn 12 O 12 ( p t BuC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 3 ). Fitting parameters: g = 1.9044 0.0017; D = 0.3718 0.0018 cm 1

PAGE 156

156 Figure B 9. DC magnetic susceptibility plot for [ Mn 12 O 12 ( 2,4,6 (CH 3 ) 3 C 6 H 2 CO 2 ) 16 (H 2 O) 4 ] ( B 4 ) in an applied 0.1 T field. Figure B 10. AC susceptibility plots vs. temperature for [Mn 12 O 12 ( 2,4,6 (CH 3 ) 3 C 6 H 2 CO 2 ) 16 (H 2 O) 4 ] ( B 4 ). AC in phase (top) and out of phase (bottom).

PAGE 157

157 Figure B 11 Arrhenius plot for [Mn 12 O 12 ( 2,4,6 (CH 3 ) 3 C 6 H 2 CO 2 ) 16 (H 2 O) 4 ] ( B 4 ). F itting parameters : U eff /k B = 64.9 K ; 0 = 4.74 10 9 s Figure B 12. Reduced magnetization plot for [Mn 12 O 12 ( 2,4,6 (CH 3 ) 3 C 6 H 2 CO 2 ) 16 (H 2 O) 4 ] ( B 4 ). Fitting parameters: g = 1.9587 0.0116; D = 0.3797 0.0123 cm 1

PAGE 158

158 Figure B 13. DC magnetic susceptibility plot for [ Mn 12 O 12 ( p ClC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 5 ) in an applied 0.1 T field. Figure B 14. AC susceptibility plots vs. temperature for [Mn 12 O 12 ( p ClC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 5 ). AC in phase (top) and out of phase (bottom).

PAGE 159

159 Figure B 15 Arrhenius plot for [Mn 12 O 12 ( p ClC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 5 ). F itting parameters : U eff /k B = 68.0 K ; 0 = 1.61 10 9 s Figure B 16. Reduced magnetization plot for [Mn 12 O 12 ( p ClC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 5 ). Fitting parameters: g = 1.6715 0.0042 ; D = 0.4161 0.0055 cm 1

PAGE 160

16 0 Figure B 17 DC magnetic susceptibility plot for [ Mn 12 O 12 ( 3,5 F 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 7 ) in an applied 0.1 T field. Figure B 18 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( 3,5 F 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 7 ). AC in phase (top) and out of phase (bottom).

PAGE 161

161 Figure B 19 Arrhenius plot for [Mn 12 O 12 ( 3,5 F 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 7 ). F itting parameters : U eff /k B = 57.9 K ; 0 = 4.83 10 9 s Figure B 2 0 Reduced magnetization plot for [Mn 12 O 12 ( 3,5 F 2 C 6 H 3 CO 2 ) 16 (H 2 O) 4 ] ( B 7 ). Fitting parameters: g = 1.7168 0.0053 ; D = 0.3108 0.0052 cm 1

PAGE 162

162 Figure B 2 1 DC magnetic susceptibility plot for [ Mn 12 O 12 ( o FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 8 ) in an applied 0.1 T field. Figure B 2 2 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( o FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 8 ). AC in phase (top) and out of phase (bottom).

PAGE 163

163 Figure B 2 3 Arrhenius plot for [Mn 12 O 12 ( o FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 8 ). F itting parameters : U eff /k B = 62.9 K ; 0 = 3.63 10 9 s Figure B 2 4 Reduced magnetization plot for [Mn 12 O 12 ( o FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 8 ). Fitting parameters: g = 1.7206 0.0016 ; D = 0.3306 0.0016 cm 1

PAGE 164

164 Figure B 2 5 DC magnetic susceptibility plot for [ Mn 12 O 12 ( p CF 3 C 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 9 ) in an applied 0.1 T field. Figure B 26 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( p CF 3 C 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 9 ). AC in phase (top) and out of phase (bottom).

PAGE 165

165 Figure B 27 Arrhenius plot for [Mn 12 O 12 ( p CF 3 C 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 9 ). F itting parameters : U eff /k B = 61.9 K ; 0 = 2.02 10 9 s Figure B 28 Reduced magnetization plot for [Mn 12 O 12 ( p CF 3 C 6 H 4 CO 2 ) 16 (H 2 O) 4 ] ( B 9 ). Fitting parameters: g = 1.9450 0.0053 ; D = 0.4201 0.0061 cm 1

PAGE 166

166 Figure B 29 DC magnetic susceptibility plot for [ Mn 12 O 12 (tma) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 10 ) in an applied 0.1 T field. Figure B 3 0 AC susceptibility plots vs. temperature for [Mn 12 O 12 (tma) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 10 ). AC in phase (top) and out of phase (bottom).

PAGE 167

167 Figure B 3 1 Arrhenius plot for [Mn 12 O 12 (tma) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 10 ). F itting parameters : U eff /k B = 52.11 K ; 0 = 3.50 10 9 s Figure B 3 2 Reduced magnetization plot for [Mn 12 O 12 (tma) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 10 ).

PAGE 168

168 Figure B 3 3 DC magnetic susceptibility plot for [ Mn 12 O 12 (add) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 11 ) in an applied 0.1 T field. Figure B 3 4 AC susceptibility plots vs. temperature for [Mn 12 O 12 (add) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 11 ). AC in phase (top) and out of phase (bottom).

PAGE 169

169 Figure B 3 5 Arrhenius plot for [Mn 12 O 12 (add) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 11 ). F itting parameters : U eff /k B = 63.31 K ; 0 = 5.07 10 9 s Figure B 36 DC magnetic susceptibility plot for [ Mn 12 O 12 ( iso ) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 12 ) in an applied 0.1 T field.

PAGE 170

170 Figure B 37 AC susceptibility plots vs. temperature for [Mn 12 O 12 ( iso ) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 12 ). AC in phase (top) and out of phase (bottom). Figure B 38 Arrhenius plot for [Mn 12 O 12 ( iso ) x (O 2 CCH 3 ) 16 x (H 2 O) 4 ] ( B 12 ). F itting parameters : U eff /k B = 50.52 K ; 0 = 1.14 10 8 s

PAGE 171

171 APPENDIX C TEST OF THE VALIDITY OF A RECENTLY DEVELOPPED METHOD TO ESTIMATE KINETIC PARAMETERS IN WEAK SINGLE MOLECULE MAGNETS In the characterization of single molecule magnets, a major step is the calculation of the effective energy barrier to the reversal of magnetization. Under the assumption that the relaxation has only one characteristic time, it is traditionally obtained from the temperatures of the peaks observed in the out of phase susceptibility at different frequencies using an Arrhenius law (eq C 1). 102 (C 1) The relaxation rate at the temperature at which occurs the peak maximum is given by the inverse of the angular frequency of the oscillating AC field. Using data from a multi frequency AC experiment, the relaxation rates at different temperatures ca n be fitted to eq C 1 to obtain the effective barrier to the reversal of magnetization ( U eff ) and the pre exponential factor 0 characteristic of the relaxation time. SMMs with a low blocking temperature, however, may not display full out of phase peaks w ithin the range of temperature s reachable by the instrument (common SQUID magnetometers only go down to 1.8 K). In that case, only the tail of the out of phase peak is visible. In 2009, Bartolom et al. described a method to use a relationship between the real and imaginary parts of the AC susceptibility to estimate the effective barrier of a weak single molecule magnet. 103 It relies on finding a simple relationship between the ratio of the imaginary ( ) to the real part ( ) of the magnetic suscep tibility as a function of the relaxation time (eq C 2 to C 4), and using the Arrhenius law (eq C 1) to obtain eq C 5 to which the AC data can be fitted to calculate the kinetic parameters U eff and 0 In these equations, T is the isothermal susceptibility (i.e. the susceptibility of the material in equilibrium with phonons), S is the

PAGE 172

172 adiabatic susceptibility (i.e. the susceptibility of the isolated material), is the relaxation time, and is the frequency of the applied AC field. (C 2) (C 3) (C 4) (C 5) This method has often be called the Kramers Kronig method in reference to the Kramers Kronig relations connecting the real and imaginary part o f some complex functions. In recent years, many chemistry papers use this method in order to estimate the kinetic parameters of newly synthesized single molecule magnets. 104 107 However, it h as recently been criticized, and many question its reliability. The aim of this work is therefore to attempt to validate it using the magnetic data of a few compounds for which the full out of phase peaks are visible and therefore the kinetic parameters ca n be calculated through the normal Arrhenius method. The many Mn 12 derivatives synthesized and characterized for the purpose of this dissertation seemed to offer a good opportunity to test the validity of this method. For each compound, the AC data at 1000 250, and 50 Hz were used to calculate the effective barrier calculated via the Kramers Kronig method. These three frequencies were chosen because they are the frequencies used in a standard three frequency AC sequence used when an out of phase tail is ob served. Only the data above a certain temperature, chosen so that the peak maximum of the 1000 Hz out of phase data wasn't visible, were used. The effective barriers

PAGE 173

173 calculated using the Kramers Kronig method are gathered in Table C 1 along with the effect ive barriers obtained with the normal Arrhenius method. Table C 1. Effective barriers (K) calculated with the Kramers Kronig method at 1000, 250, and 50 Hz, and through the normal Arrhenius method. U eff 1000 Hz U eff 250 Hz U eff 50 Hz U eff Arrhenius [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ] 109.2 103.6 86.0 70.7 2 3 106.3 109.6 92.8 71.9 2 4 92.4 96.5 77.1 71.7 2 5 103.7 97.4 88.2 75.6 2 6 104.6 97.4 88.1 75.3 4 2 8 Me CN 56.7 66.6 64.2 59.3 4 2 59.0 66.4 66.1 60.6 4 2 3H 2 O 66.5 67.1 67.2 59.6 B 2 74.5 71.5 63.5 B 3 57.1 75.9 66.1 67.0 B 11 61.5 72.0 69.2 63.3 For most of the samples used, the effective barriers calculated using the Kramers Kronig method are higher than the barrier obtained with the normal Arrhenius method. For most of the samples, there is also a disparity in the effective barriers obtained for each frequenc y and the barrier calculated using data at the lowest frequency is often the closest to the barrier calculated with the normal Arrhenius method. In order to investigate the origin of the se issues, we can look at the original equation giving the complex susceptibility vs. frequency (eq. C 6) from which can be extracted the real and imaginary parts of the susceptibility (eq. C 7 and C 8). 48 (C 6) (C 7)

PAGE 174

174 (C 8) Eq. C 2 and C 3 used to develop the Kramers Kronig method are obtained from eq. C 7 and C 8 if the adiabatic susceptibility S is close to zero, which is usually assumed to be the case in molecular nanomagnets. 48 If the S term is retained, and the same method used (i.e. replacement of by the Arrhenius law in the ''( )/ '( ) ratio), the resulti ng equation becomes much more complicated (eq. C 9) and a fitting of the data to that equation would involve too many parameters ( T S 0 and U eff ). (C 9) In conclusion, the Kramers Kronig method offers a simplified path to estimate the effective energy barrier to the reversal of magnetization of a weak single molecule magnet. The obtained values should be assumed to be higher than the true effective energy barrier, and the lower frequency data should be given priority. At any time this method is used, it should be emphasized that the obtained value is only a rough approximation of the effective energy barrier.

PAGE 175

175 Figure C 1. ln( / ) versus 1/T plots for [Mn 12 O 12 (O 2 CCH 3 ) 16 (H 2 O) 4 ]. Figure C 2. ln( / ) versus 1/T plots for 2 3

PAGE 176

176 Figure C 3 ) versus 1/T plots for 2 4 Figure C 4 ) versus 1/T plots for 2 5

PAGE 177

177 Figure C 5 ) versus 1/T plots for 2 6 Figure C 6 ) versus 1/T plots for 4 2 8 Me CN

PAGE 178

178 Figure C 7 ) versus 1/T plots for 4 2 Figure C 8 ) versus 1/T plots for 4 2 3H 2 O

PAGE 179

179 Figure C 9 ) versus 1/T plots for B 2 Figure C 10 ) versus 1/T plots for B 3

PAGE 180

180 Figure C 11. ln( / ) versus 1/T plots for B 11

PAGE 181

181 APPENDIX D BOND DISTANCES AND ANGLES Table D 1. Selected interatomic distances () and angles () for [Mn 12 O 12 (O 2 CCH 3 ) 16 (CH 3 OH) 4 ] ( 2 3 ) Mn1 O2 1.8679(19) Mn2 O8 1.939(2) Mn1 O3' 1.887(2) Mn2 O6 2.188(2) Mn1 O1' 1.899(2) Mn2 O4 2.228(2) Mn1 O5 1.913(2) Mn3 O2 1.890(2) Mn1 O1 1.9225(19) Mn3 O3 1.891(2) Mn1 O1' 1.9260(19) Mn3 O9 1.963(2) Mn1 Mn2 2.7775(6) Mn3 O10 1.980(2) Mn1 Mn1''' 2.8172(8) Mn3 O7''' 2.111(2) Mn1 Mn1' 2.8173(8) Mn3 O12 2.232(3) Mn1 Mn1'' 2.9569(8) O1 Mn1''' 1.899(2) Mn2 O2 1.884(2) O1 Mn1'' 1.9260(19) Mn2 O3' 1.899(2) O3 Mn1''' 1.887(2) Mn2 O11' 1.934(2) O3 Mn2''' 1.899(2) O7 Mn3' 2.111(2) O11 Mn2''' 1.934(2) O2 Mn1 O3' 84.83(9) O2 Mn2 O3' 84.04(8) O2 Mn1 O1' 90.01(9) O2 Mn2 O11' 174.90(10) O3' Mn1 O1' 90.26(9) O3' Mn2 O11' 96.47(9) O2 Mn1 O5 94.63(9) O2 Mn2 O8 94.62(9) O3' Mn1 O5 91.79(9) O3' Mn2 O8 176.75(10) O1' Mn1 O5 175.08(9) O11' Mn2 O8 84.60(10) O2 Mn1 O1 95.93(9) O2 Mn2 O6 92.25(10) O3' Mn1 O1 174.37(9) O3' Mn2 O6 92.56(9) O1' Mn1 O1 84.16(10) O11' Mn2 O6 92.79(11) O5 Mn1 O1 93.70(9) O8 Mn2 O6 90.45(10) O2 Mn1 O1'' 172.89(9) O2 Mn2 O4 85.13(9) O3' Mn1 O1'' 99.09(8) O3' Mn2 O4 84.23(9) O1' Mn1 O1'' 84.07(9) O11' Mn2 O4 89.88(10) O5 Mn1 O1'' 91.19(9) O8 Mn2 O4 92.71(10) O1 Mn1 O1'' 79.59(9) O6 Mn2 O4 176.04(9) O2 Mn1 Mn2 42.47(6) O2 Mn2 Mn1 42.02(6) O3' Mn1 Mn2 42.98(6) O3' Mn2 Mn1 42.64(6) O1' Mn1 Mn2 95.94(6) O11' Mn2 Mn1 137.51(7) O5 Mn1 Mn2 88.60(6) O8 Mn2 Mn1 135.53(7) O1 Mn1 Mn2 138.33(6) O6 Mn2 Mn1 98.92(7) O1'' Mn1 Mn2 142.02(6) O4 Mn2 Mn1 77.15(6) O2 Mn1 Mn1''' 86.98(6) O2 Mn3 O3 92.05(9) O3' Mn1 Mn1''' 132.45(7) O2 Mn3 O9 91.21(9)

PAGE 182

182 Table D 1. Continued O1' Mn1 Mn1''' 42.93(6) O3 Mn3 O9 176.27(10) O5 Mn1 Mn1''' 135.58(7) O2 Mn3 O10 173.04(9) O1 Mn1 Mn1''' 42.19(7) O3 Mn3 O10 93.61(9) O1'' Mn1 Mn1''' 86.00(6) O9 Mn3 O10 83.02(10) Mn2 Mn1 Mn1''' 119.34(2) O2 Mn3 O7''' 93.41(10) O2 Mn1 Mn1' 132.50(7) O3 Mn3 O7''' 95.26(9) O3' Mn1 Mn1' 89.31(6) O9 Mn3 O7''' 86.36(10) O1' Mn1 Mn1' 42.83(6) O10 Mn3 O7''' 90.11(11) O5 Mn1 Mn1' 132.71(7) O2 Mn3 O12 93.99(11) O1 Mn1 Mn1' 86.07(6) O3 Mn3 O12 91.39(11) O1'' Mn1 Mn1' 42.20(7) O9 Mn3 O12 86.57(11) Mn2 Mn1 Mn1' 121.54(2) O10 Mn3 O12 81.83(13) Mn1''' Mn1 Mn1' 63.305(19) O7''' Mn3 O12 169.87(12) O2 Mn1 Mn1'' 135.55(6) Mn1''' O1 Mn1 94.98(9) O3' Mn1 Mn1'' 138.62(6) Mn1''' O1 Mn1'' 94.87(9) O1' Mn1 Mn1'' 82.53(6) Mn1 O1 Mn1'' 100.41(9) O5 Mn1 Mn1'' 92.98(6) Mn1 O2 Mn2 95.50(9) O1 Mn1 Mn1'' 39.84(6) Mn1 O2 Mn3 134.22(11) O1'' Mn1 Mn1'' 39.75(6) Mn2 O2 Mn3 128.65(11) Mn2 Mn1 Mn1'' 177.64(2) Mn1''' O3 Mn3 133.21(11) Mn1''' Mn1 Mn1'' 58.347(10) Mn1''' O3 Mn2''' 94.38(9) Mn1' Mn1 Mn1'' 58.346(9) Mn3 O3 Mn2''' 122.24(11)

PAGE 183

183 Table D 2 Selected interatomic distances () and angles () for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] 8MeCN ( 4 2 ) Mn1 O2' 1.874(3) Mn2 O6 2.124(4) Mn1 O2 1.874(3) Mn3 O2' 1.890(3) Mn1 O1'' 1.891(5) Mn3 O2 1.890(3) Mn1 O8 1.924(4) Mn3 O3' 1.910(5) Mn1 O1 1.927(2) Mn3 O3 1.910(5) Mn1 O1''' 1.927(2) Mn3 O3'' 2.001(6) Mn1 Mn3 2.7920(12) Mn3 O3' 2.001(6) Mn1 Mn1'' 2.8097(16) Mn3 O7 2.183(5) Mn1 Mn1''' 2.9626(15) Mn3 O5 2.202(6) Mn2 O2 1.883(3) O1 Mn1''' 1.927(2) Mn2 O4' 1.916(5) O4 Mn2'' 1.917(5) Mn2 O4'' 2.049(5) O4' Mn2'' 2.049(5) O2' Mn1 O2 84.08(18) O4' Mn2 O6 88.28(18) O2' Mn1 O1'' 90.11(13) O4'' Mn2 O6 76.95(18) O2 Mn1 O1'' 90.11(13) O2' Mn3 O2 83.20(16) O2' Mn1 O8 91.99(15) O2' Mn3 O3' 178.8(2) O2 Mn1 O8 91.99(15) O2 Mn3 O3' 96.75(17) O1'' Mn1 O8 177.17(19) O2' Mn3 O3 96.75(17) O2' Mn1 O1 173.98(17) O2 Mn3 O3 178.8(2) O2 Mn1 O1 97.94(11) O3' Mn3 O3 83.3(3) O1'' Mn1 O1 84.24(17) O2' Mn3 O3'' 167.2(2) O8 Mn1 O1 93.60(17) O2 Mn3 O3'' 94.89(19) O2' Mn1 O1''' 97.94(11) O2' Mn3 O3' 94.89(19) O2 Mn1 O1''' 173.99(17) O2 Mn3 O3' 167.2(2) O1'' Mn1 O1''' 84.24(17) O3'' Mn3 O3' 84.2(4) O8 Mn1 O1''' 93.59(17) O2' Mn3 O7 85.47(15) O1 Mn1 O1''' 79.50(15) O2 Mn3 O7 85.47(15) O2' Mn1 Mn3 42.35(9) O3' Mn3 O7 95.75(18) O2 Mn1 Mn3 42.35(9) O3 Mn3 O7 95.75(18) O1'' Mn1 Mn3 95.82(11) O3'' Mn3 O7 81.72(19) O8 Mn1 Mn3 87.00(14) O3' Mn3 O7 81.72(19) O1 Mn1 Mn3 140.22(8) O2' Mn3 O5 93.28(16) O1''' Mn1 Mn3 140.22(8) O2 Mn3 O5 93.28(16) O2' Mn1 Mn1'' 88.35(10) O3' Mn3 O5 85.5(2) O2 Mn1 Mn1'' 132.63(12) O3 Mn3 O5 85.50(19) O1'' Mn1 Mn1'' 43.11(7) O3'' Mn3 O5 99.5(2) O8 Mn1 Mn1'' 135.05(10) O3' Mn3 O5 99.5(2) O1 Mn1 Mn1'' 86.12(11) O7 Mn3 O5 178.3(2) O1''' Mn1 Mn1'' 42.11(13) O2' Mn3 Mn1 41.91(8) Mn3 Mn1 Mn1'' 120.44(4) O2 Mn3 Mn1 41.91(8) O2' Mn1 Mn1''' 137.43(9) O3' Mn3 Mn1 138.21(15)

PAGE 184

184 Table D 3. Continued O2 Mn1 Mn1''' 137.43(9) O3 Mn3 Mn1 138.21(15) O1'' Mn1 Mn1''' 82.43(10) O3'' Mn3 Mn1 133.16(18) O8 Mn1 Mn1''' 94.75(14) O3' Mn3 Mn1 133.16(18) O1 Mn1 Mn1''' 39.75(8) O7 Mn3 Mn1 78.35(14) O1''' Mn1 Mn1''' 39.75(8) O5 Mn3 Mn1 99.98(16) Mn3 Mn1 Mn1''' 178.25(3) Mn1 O1 Mn1''' 100.50(15) Mn1'' Mn1 Mn1''' 58.18(2) Mn1 O2 Mn2 134.03(16) O2 Mn2 O4' 96.10(18) Mn1 O2 Mn3 95.74(13) O2 Mn2 O4'' 88.22(17) Mn2 O2 Mn3 125.57(15) O2 Mn2 O6 94.36(15)

PAGE 185

185 APPENDIX E BOND VALENCE SUM CALCULATIONS Table E 1. Bond valence sum calculations and assignment for the Mn ions for [Mn 12 O 12 (O 2 CCH 3 ) 16 (CH 3 OH) 4 ] ( 2 3 ) a Atom Mn(II) Mn(III) Mn(IV) Mn 1 4.14 3.79 3.98 Mn2 3.28 3.00 3.15 Mn3 3.24 2.97 3.12 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 whole number nearest to the bond value. Table E 2 Bond valence sum calculations and assignment for the Mn ions for [Mn 12 O 12 ( p FC 6 H 4 CO 2 ) 16 (H 2 O) 4 ] 8MeCN ( 4 2 ) a Atom Mn(II) Mn(III) Mn(IV) Mn1 4.14 3.79 3.98 Mn2 3.34 3.06 3.21 Mn3 3.26 2.98 3.13 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 whole number nearest to the bond value.

PAGE 186

186 APPENDIX F PERMISSION TO REPRODUCE COPYRIGHTED MATERIAL Title: A Fourth Isolated Oxidation Level of the [Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ] Family of Single Molecule Magnets Author: Rashmi Bagai, George Christou Publication: Inorganic Chemistry Publisher: American Chemical Society Date: Dec 1, 2007 Copyright 2007, American Chemical Society PERMISSION/LICENSE IS GRANTED FOR YOUR ORDER AT NO CHARGE This type of permission/license, instead of the standard Terms & Conditions, is sent to you because no fee is being charged for your order. Please note the following: Permission is granted for your request in both print and electronic formats, and translations. If figures and/or tables were requested, they may be adapted or used in part. Please print this page for your records and send a copy of it to your pu blisher/graduate school. Appropriate credit for the requested material should be given as follows: "Reprinted (adapted) with permission from (COMPLETE REFERENCE CITATION). Copyright (YEAR) American Chemical Society." Insert appropriate information in place of the capitalized words. One time permission is granted only for the use specified in your request. No additional uses are granted (such as derivative works or other editions). For any other uses, please submit a new request. If credit is given to anoth er source for the material you requested, permission must be obtained from that source.

PAGE 187

187 Title: The Drosophila of single molecule magnetism: [Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ] Author: Rashmi Bagai,George Christou Publication: Chemical Society Reviews Publisher: Royal Society of Chemistry Date: Feb 23, 2009 Copyright 2009, Royal Society of Chemistry License Number 3815711021854 License date Feb 25, 2016 Licensed Content Publisher Royal Society of Chemistry Licensed Content Publication Chemical Society Reviews Licensed Content Title The Drosophila of single molecule magnetism: [Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ] Licensed Content Author Rashmi Bagai,George Christou Licensed Content Date Feb 23, 2009 Licensed Content Volume 38 Licensed Content Issue 4 Type of Use Thesis/Dissertation Requestor type academic/educational Portion figures/tables/images Number of figures/tables/images 2 Distribution quantity 10 Format print and electronic Will you be translating? no Order reference number None Title of the thesis/dissertation EXPLORATION OF ENVIRONMENTAL INFLUENCES ON THE PROPERTIES OF Mn 12 SINGLE MOLECULE MAGNETS Expected completion date Apr 2016 Estimated size 200 Requestor Location Adeline D Fournet PO BOX 117200 GAINESVILLE, FL 32611 United States Attn: Adeline D Fournet Billing Type Invoice Billing address Adeline D Fournet PO BOX 117200 GAINESVILLE, FL 32611 United States Attn: Adeline D Fournet Total 0.00 USD

PAGE 188

188 LIST OF REFERENCES 1. J. A. Bertrand, F. A. Cotton and W. A. Dollase, Inorganic Chemistry 1963, 2 1166 1171. 2. D. M. P. Mingos and D. J. Wales, Introduction to Cluster Chemistry Prentice Hall, 1990. 3. R. H. Crabtree, The Organometallic Chemistry of the Transition Metals John Wiley & Sons, Inc., 2005. 4. G. Schmid and D. Fenske, Metal clusters and nanoparticles 2010. 5. C. N. R. Rao and B. Raveau, Transition Metal Oxides VCH Publishers, Inc., 1995. 6. M. Moskovits, Annua l Review of Physical Chemistry 1991, 42 465 499. 7. A. J. Tasiopoulos, A. Vinslava, W. Wernsdorfer, K. A. Abboud and G. Christou, Angewandte Chemie 2004, 43 2117 2121. 8. X. Wang, T. Langetepe, C. Persau, B. Kang, G. M. Sheldrick and D. Fenske, Angewan dte Chemie 2002, 41 3818 3822. 9. A. Fernando, K. L. D. M. Weerawardene, N. V. Karimova and C. M. Aikens, Chemical Reviews 2015, 115 6112 6216. 10. S. J. Lippard and J. M. Berg, Principles of Bioinorganic Chemistry University Science Books, 1994. 11. Y. Umena, K. Kawakami, J. R. Shen and N. Kamiya, Nature 2011, 473 55 60. 12. J. S. Kanady, E. Y. Tsui, M. W. Day and T. Agapie, Science 2011, 333 733 736. 13. K. N. Ferreira, T. M. Iverson, K. Maghlaoui, J. Barber and S. Iwata, Science 2004, 303 1831 1838. 14. B. D. Cullity and C. D. Graham, Magnetic Materials, Second Edition Wiley edn., 2009. 15. J. S. Miller, Materials Today 2014, 17 224 235. 16. J. R. Friedman, M. P. Sarachik, J. Tejada and R. Ziolo, Physical Review Letters 1996, 76 3830 3833. 17. E. C. Publications, Global Magnetic Materials Maket Analysis Advanced Materials Research, 2014. 18. D. Gatteschi, A. Caneschi and R. Sessoli, in Inorganic and Organometallic Polymers with Special Properties ed. R. Laine, Springer Netherlands, 1992, vol. 206, pp. 147 160. 19. L. Bogani and W. Wernsdorfer, Nat Mater 2008, 7 179 186.

PAGE 189

189 20. T. Lis, Acta Crystallographica 1980, B36 2042 2046. 21. A. Caneschi, D. Gatteschi, R. Sessoli, A. L. Barra, L. C. Brunel and M. Guillot, Journal of the American Che mical Society 1991, 113 5873 5874. 22. P. D. W. Boyd, Q. Li, J. B. Vincent, K. Folting, H. R. Chang, W. E. Streib, J. C. Huffman, G. Christou and D. N. Hendrickson, Journal of the American Chemical Society 1988, 110 8537 8539. 23. H. J. Eppley, H. L. T sai, N. de Vries, K. Folting, G. Christou and D. N. Hendrickson, Journal of the American Chemical Society 1995, 117 301 317. 24. M. Soler, P. Artus, K. Folting, J. C. Huffman, D. N. Hendrickson and G. Christou, Inorganic Chemistry 2001, 40 4902 4912. 2 5. Physical Review B 2002, 66 054409. 26. O. Kahn, Chemical Physics Letters 1997, 265 109 114. 27. J. K. McCusker, J. B. Vincent, E. A. Schmitt, M. L. Mino, K. Shin, D. K. Coggin, P. M. Hagen, J. C. Huffman, G. Christou and D. N. Hendrickson, Journal of the American Chemical Society 1991, 113 3012 3021. 28. T. C. Stamatatos, D. Foguet Albiol, K. M. Poole, W. Wernsdorfer, K. A. Abboud, T. A. Inorganic Chemistr y 2009, 48 9831 9845. 29. R. Sessoli, D. Gatteschi, A. Caneschi and M. A. Novak, Nature 1993, 365 141 143. 30. S. M. J. Aubin, Z. Sun, D. N. Hendrickson, I. A. Guzei, A. L. Rheingold and G. Christou, Chemical Communications 1997, 2239 2240. 31. S. M. J. Aubin, Z. Sun, H. J. Eppley, E. M. Rumberger, I. A. Guzei, K. Folting, P. K. Gantzel, A. L. Rheingold, G. Christou and D. N. Hendrickson, Inorganic Chemistry 2001, 40 2127 2146. 32. R. Sessoli, H. L. Tsai, A. R. Schake, S. Wang, J. B. Vincent, K Folting, D. Gatteschi, G. Christou and D. N. Hendrickson, Journal of the American Chemical Society 1993, 115 1804 1816. 33. M. Soler, W. Wernsdorfer, K. A. Abboud, J. C. Huffman, E. R. Davidson, D. N. Hendrickson and G. Christou, Journal of the America n Chemical Society 2003, 125 3576 3588. 34. R. Bagai and G. Christou, Inorganic Chemistry 2007, 46 10810 10818. 35. R. Bagai and G. Christou, Chemical Society Reviews 2009, 38 1011 1026.

PAGE 190

190 36. M. Soler, W. Wernsdorfer, Z. Sun, D. Ruiz, J. C. Huffman, D N. Hendrickson and G. Christou, Polyhedron 2003, 22 1783 1788. 37. P. Artus, C. Boskovic, J. Yoo, W. E. Streib, L. C. Brunel, D. N. Hendrickson and G. Christou, Inorganic Chemistry 2001, 40 4199 4210. 38. J. T. Brockman, K. A. Abboud, D. N. Hendricks on and G. Christou, Polyhedron 2003, 22 1765 1769. 39. Bruker AXS, Madison, Wisconsin, USA, 2013. 40. E. R. Davidson, Indiana University, Bloomington, IN, 1999. 41. E. R. Davidson, Indiana University, Bloomington, IN, 1999. 42. A. Cornia, R. Sessoli, L. Sorace, D. Gatteschi, A. L. Barra and C. Daiguebonne, Physical Review Letters 2002, 89 257201. 43. G. Redler, C. Lampropoulos, S. Datta, C. Koo, T. C. Stamatatos, N. E. Chakov, G. Christou and S. Hill, Physical Review B 2009, 80 094408. 44. N. E. Chako v, S. C. Lee, A. G. Harter, P. L. Kuhns, A. P. Reyes, S. O. Hill, N. S. Dalal, W. Wernsdorfer, K. A. Abboud and G. Christou, Journal of the American Chemical Society 2006, 128 6975 6989. 45. C. Lampropoulos, M. Murugesu, A. G. Harter, W. Wernsdofer, S. Hill, N. S. Dalal, A. P. Reyes, P. L. Kuhns, K. A. Abboud and G. Christou, Inorganic Chemistry 2013, 52 258 272. 46. M. Soler, W. Wernsdorfer, Z. Sun, J. C. Huffman, D. N. Hendrickson and G. Christou, Chemical Communications 2003, 2672 2673. 47. Z. Sun, D. Ruiz, D. N. Hendrickson, N. R. Dilley, M. Brian Maple, M. Soler, K. Folting, G. Christou and J. Ribas, Chemical Com munications 1999, 1973 1974. 48. D. Gatteschi, R. Sessoli and J. Villain, Molecular Nanomagnets Oxford University Press, 2006. 49. G. Christou, D. Gatteschi, D. N. Hendrickson and R. Sessoli, MRS Bulletin 2000, 25 66 71. 50. R. Winpenny, Molecular Clus ter Magnets WORLD SCIENTIFIC, 2012. 51. Y. X. Wang, W. Shi, H. Li, Y. Song, L. Fang, Y. Lan, A. K. Powell, W. Wernsdorfer, L. Ungur, L. F. Chibotaru, M. Shen and P. Cheng, Chemical Science 2012, 3 3366 3370. 52. J. Bartolome, L. Fernando and J. F. Fern ndez, Molecular Magnets, Physics and Applications Springer, 2014.

PAGE 191

191 53. F. Luis, J. Bartolom, J. F. Fernndez, J. Tejada, J. M. Hernndez, X. X. Zhang and R. Ziolo, Physical Review B 1997, 55 11448 11456. 54. F. Bron, M. A. Novak, M. G. F. Vaz, G. P. Gu edes, M. Knobel, A. Caldeira and K. R. Pirota, Applied Physics Letters 2013, 103 052407. 55. T. Kuroda Sowa, T. Nogami, H. Konaka, M. Maekawa, M. Munakata, H. Miyasaka and M. Yamashita, Polyhedron 2003, 22 1795 1801. 56. G. Q. Bian, T. Kuroda Sowa, H. Konaka, M. Hatano, M. Maekawa, M. Munakata, H. Miyasaka and M. Yamashita, Inorganic Chemistry 2004, 43 4790 4792. 57. A. R. Schake, H. L. Tsai, R. J. Webb, K. Folting, G. Christou and D. N. Hendrickson, Inorganic Chemistry 1994, 33 6020 6028. 58. G. N. La Mar, W. D. Horrocks and R. H. Holm, NMR of Paramagnetic Molecules, Principles and Applications Academic Press, 1973. 59. I. Bertini, C. Luchinat and G. Parigi, Solution NMR of Paramagnetic Molecules Elsevier, 2001. 60. J. A. Orvis, B. Dimetry, J. Win ge and T. C. Mullis, Journal of Chemical Education 2003, 80 803. 61. I. Bertini, P. Turano and A. J. Vila, Chemical Reviews 1993, 93 2833 2932. 62. G. Otting, Annual Review of Biophysics 2010, 39 387 405. 63. M. Kruck, D. C. Sauer, M. Enders, H. Wadepohl and G. H. Lutz, Dalton Transactions 2011, 40 10406 10415. 64. R. H. Beer and S. J. Lippard, Inorganic Chemistry 1993, 32 1030 1032. 65. Z. Luz, B. L. Silver and D. Fiat, The Journal of Chemical Physics 196 7, 46 469 471. 66. R. Guilard, N. Jagrovic, J. M. Barbe, Y. H. Liu, I. Perrot, C. Naillon, E. Van Caemelbecke and K. M. Kadish, Polyhedron 1995, 14 3041 3050. 67. V. Tangoulis, D. A. Malamatari, G. A. Spyroulias, C. P. Raptopoulou, A. Terzis and D. P. Kessissoglou, Inorganic Chemistry 2000, 39 2621 2630. 68. T. Kurahashi and H. Fujii, Journal of the American Chemical Society 2011, 133 8307 8316. 69. H. S. Gutowsky and C. H. Holm, The Journal of Chemical Physics 1956, 25 1228 1234. 70. R. E. Connic k and R. E. Poulson, The Journal of Chemical Physics 1959, 30 759 761.

PAGE 192

192 71. Journal of the American Chemical Society 2002, 124 3951 3958. 72. D. G. Lonnon, G. E. Ball, I. Taylor, D. C. Craig and S. B. Colbran, Inor ganic Chemistry 2009, 48 4863 4872. 73. P. Roquette, A. Maronna, M. Reinmuth, E. Kaifer, M. Enders and H. J. Himmel, Inorganic Chemistry 2011, 50 1942 1955. 74. L. Fusaro, F. Mocci, R. N. Muller and M. Luhmer, Inorganic Chemistry 2012, 51 8455 8461. 75. S. McHUGH and M. P. SARACHIK, Modern Physics Letters B 2011, 25 1795 1807. 76. W. Wernsdorfer, N. Aliaga Alcalde, D. N. Hendrickson and G. Christou, Nature 2002, 416 406 409. 77. R. Bagai, W. Wernsdorfer, K. A. Abboud and G. Christou, Journal of th e American Chemical Society 2007, 129 12918 12919. 78. C. Boskovic, R. Bircher, P. L. W. Tregenna Piggott, H. U. Gdel, C. Paulsen, W. Wernsdorfer, A. L. Barra, E. Khatsko, A. Neels and H. Stoeckli Evans, Journal of the American Chemical Society 2003, 1 25 14046 14058. 79. L. M. Wittick, K. S. Murray, B. Moubaraki, S. R. Batten, L. Spiccia and K. J. Berry, Dalton Transactions 2004, 1003 1011. 80. A. Yamaguchi, N. Kusumi, H. Ishimoto, H. Mitamura, T. Goto, N. Mori, M. Nakano, K. Awaga, J. Yoo, David N. H endrickson and G. Christou, Journal of the Physical Society of Japan 2002, 71 414 417. 81. A. Bhattacharjee, Y. Miyazaki, M. Nakano, J. Yoo, G. Christou, D. N. Hendrickson and M. Sorai, Polyhedron 2001, 20 1607 1613. 82. M. Affronte, J. C. Lasjaunias, W. Wernsdorfer, R. Sessoli, D. Gatteschi, S. L. Heath, A. Fort and A. Rettori, Physical Review B 2002, 66 064408. 83. R. Tiron, W. Wernsdorfer, N. Aliaga Alcalde and G. Christou, Physical Review B 2003, 68 140407. 84. E. C. Yang, W. Wernsdorfer, S. Hil l, R. S. Edwards, M. Nakano, S. Maccagnano, L. N. Zakharov, A. L. Rheingold, G. Christou and D. N. Hendrickson, Polyhedron 2003, 22 1727 1733. 85. R. Inglis, L. F. Jones, K. Mason, A. Collins, S. A. Moggach, S. Parsons, S. P. Perlepes, W. Wernsdorfer and E. K. Brechin, Chemistry A European Journal 2008, 14 9117 9121.

PAGE 193

193 86. C. Coulon, H. Miyasaka and R. Clrac, in Single Molecule Magnets and Related Phenomena ed. R. Winpenny, Springer Berlin Heidelberg, Berlin, Heidelberg, 2006, pp. 163 206. 87. H. B. X u, B. W. Wang, F. Pan, Z. M. Wang and S. Gao, Angewandte Chemie International Edition 2007, 46 7388 7392. 88. H. Miyasaka and M. Yamashita, Dalton Transactions 2007, 399 406. 89. L. Bogani, A. Vindigni, R. Sessoli and D. Gatteschi, Journal of Materials Chemistry 2008, 18 4750 4758. 90. O. Roubeau and R. Clrac, European Journal of Inorganic Chemistry 2008, 2008 4325 4342. 91. H. Miyasaka, K. Nakata, L. Lecren, C. Coulon, Y. Nakazawa, T. Fujisaki, K. i. Sugiura, M. Yamashita and R. Clrac, Journal of the American Chemical Society 2006, 128 3770 3783. 92. S. K. Langley, N. F. Chilton, B. Moubaraki and K. S. Murray, Dalton Transactions 2011, 40 12201 12209. 93. T. Shiga, H. Miyasaka, M. Yamashita, M. Morimoto and M. Irie, Dalton Transactions 2011, 4 0 2275 2282. 94. T. N. Nguyen, W. Wernsdorfer, K. A. Abboud and G. Christou, Journal of the American Chemical Society 2011, 133 20688 20691. 95. A. M. Mowson, T. N. Nguyen, K. A. Abboud and G. Christou, Inorganic Chemistry 2013, 52 12320 12322. 96. T. N. Nguyen, W. Wernsdorfer, M. Shiddiq, K. A. Abboud, S. Hill and G. Christou, Chemical Science 2016, 7 1156 1173. 97. T. N. Nguyen, M. Shiddiq, T. Ghosh, K. A. Abboud, S. Hill and G. Christou, Journal of the American Chemical Society 2015, 137 7160 71 68. 98. P. Van Der Sluis and A. L. Spek, Acta Crystallographica Section A 1990, 46 194 201. 99. A. L. Spek, Acta Crystallographica 2009, D65 148 155. 100. R. S. Drago, Physical Methods in Chemistry Saunders Golden Sunburst Series, 1977. 101. S. M. J. Aubin, Z. Sun, E. M. Rumberger, D. N. Hendrickson and G. Christou, Journal of Applied Physics 2002, 91 7158 7160. 102. M. Balanda, Acta Physica Polonica A 2013, 124 964 976.

PAGE 194

194 103. J. Bartolom, G. Filoti, V. Kuncser, G. Schinteie, V. Mereacre, C. E. Anson, A. K. Powell, D. Prodius and C. Turta, Physical Review B 2009, 80 014430. 104. S. J. Liu, J. P. Zhao, W. C. Song, S. D. Han, Z. Y. Liu and X. H. Bu, Inorganic Chemistry 2013, 52 2103 2109. 105. K. J. Cho, D. W. Ryu, K. S. Lim, W. R. Lee, J. W. Lee, E. K. Koh and C. S. Hong, Dalton Transactions 2013, 42 5796 5804. 106. A. Grigoropoulos, M. Pissas, P. Papatolis, V. Psycharis, P. Kyritsis and Y. Sanakis, Inorganic Chemistry 20 13, 52 12869 12871. 107. M. R. Saber and K. R. Dunbar, Chemical Communications 2014, 50 2177 2179.

PAGE 195

195 BIOGRAPHICAL SKETCH Adeline D. Fournet was born in 1989 in Mamers, France, and was raised in the city of Angers, France She discovered a passion for science at an early age through science outreach events and games As early as she can remember, she was always curious about the world, dismantling objects to understand their mechanism, experimenting using diverse el ements of her environment to the delight of her parents, and always asking random "why questions". When, in high school, she had to choose a path for her studies, the choice of science courses was completely natural, and she had the most fun in p hysics and c hemistry classes. After receiving her high school diploma with honors, she enrolled in a two year intensive foundation degree in order to prepare for the national competitive exam which grants acceptance into French graduate schools of engineering. She then entered the engineering school of chemistry and physics of Bordeaux, France to pursue a m aster s degree. She graduated in 2010 and joined the chemistry graduate program at the Univeristy of Florida. For five years, she worked in the Christou group, investigating new physical proper ties of single molecule magnets, and received her Ph.D. in the spring of 2016