Citation
Supramolecular Aggregates of Single-Molecule Magnets

Material Information

Title:
Supramolecular Aggregates of Single-Molecule Magnets
Creator:
Nguyen, Tu N
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (1 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Chemistry
Committee Chair:
CHRISTOU,GEORGE
Committee Co-Chair:
TALHAM,DANIEL R
Committee Members:
WAGENER,KENNETH B
VEIGE,ADAM S
HAGELIN,HELENA AE
Graduation Date:
8/9/2014

Subjects

Subjects / Keywords:
Atoms ( jstor )
Ground state ( jstor )
Hysteresis ( jstor )
Ions ( jstor )
Magnetic fields ( jstor )
Magnetism ( jstor )
Magnetization ( jstor )
Magnets ( jstor )
Molecules ( jstor )
Solvents ( jstor )
Chemistry -- Dissertations, Academic -- UF
aggregate -- hfepr -- hysteresis -- magnet -- supramolecular
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Chemistry thesis, Ph.D.

Notes

Abstract:
This dissertation focuses on the syntheses and magnetic studies of covalently-bonded supramolecular aggregates of single-molecule magnets (SMMs). Employing 3-phenyl-1,5-di(pyridin-2-yl)pentane-1,5-dione dioxime (pdpdH2) in the reaction with the non-SMM [Mn3O(O2CMe)6(py)3](ClO4) complex yielded a novel rectangular [Mn3]4 supramolecule. Solid-state dc and ac magnetic susceptibility measurements revealed that each Mn3 unit is still an SMM with ground-state spin of S = 6. Magnetization vs. dc-field sweeps on a single crystal gave hysteresis loops below 1 K that exhibited exchange-biased quantum tunneling of magnetization (QTM) steps, confirming the rectangular [Mn3]4 complex to be a supramolecular aggregate of four weakly exchange-coupled SMM units. The inter-SMM coupling through pdpd2- groups was found to be very weak, J = -0.02 K. The organic compound 1,2-di(pyridin-2-yl)ethane-1,2-dione dioxime (dedH2) was synthesized and employed in a similar manner to pdpdH2. The reactions with the non-SMM [Mn3O(O2CR)6(py)3](ClO4), R= -Me, -Et, complexes afforded the corresponding tetrahedral [Mn3]4 supramolecules. These two molecules contain three and two MnII ions, respectively. Similar reactions employing excess of iodine gave two new [Mn3]4 tetrahedrons with all Mn ions in the +3 oxidation state. These tetrahedral supramolecules encapsulated solvent molecules inside the [Mn3]4 frames. Magnetic studies confirmed that the [Mn3]4 tetrahedrons are supramolecular aggregates of four exchange-coupled SMM units. Simulation of the spin state energies vs. applied magnetic field suggested the inter-SMM coupling between Mn3 units of -0.06 K. The QTM steps are broad due to the non-coplanarity of the Mn3 units. Employing the ligand 1,3-di(pyridin-2-yl)propane-1,3-dione dioxime (dpdH2) afforded a dimeric [Mn3]2 SMM. UV-Vis spectroscopy studies confirmed that the dimeric molecules are intact in an acetonitrile solution. Solid-state dc and ac magnetic susceptibility measurements suggested that Mn3 units have S = 6 ground-state, and they weakly ferromagnetically interact with each other. High-frequency electron paramagnetic resonance (HF-EPR) studies on a single crystal or a solution of the [Mn3]2 SMM in acetonitrile/toluene (1 : 1 v/v) confirmed the ferromagnetic inter-SMM interaction with J = +0.07 K. Trimesic acid, alpha-truxillic acid, and dimethylmalonic acid were employed to join [Mn3O(O2CMe)3(mpko)3](ClO4) SMMs by carboxylate substitution method. Toluene was added to the reaction mixtures and the acetic acid was removed as its toluene azeotrope by evaporation under vacuum. Trimesic acid gave a tetrahedral [Mn3]4 molecule while alpha-truxillic acid and dimethylmalonic acid result in two dimeric [Mn3]2 complexes. Solid-state dc and ac magnetic susceptibility measurements revealed that each Mn3 unit in all three complexes still preserves their SMM behavior with ground-state spin of S = 6. Hysteresis studies on a single crystal of the dimeric complex that was formed from the use of alpha-truxillic acid displayed the exchange-biased QTM steps. The inter-SMM coupling is very weak, J = -0.01 K. The dissertation demonstrates the feasibility of covalently connecting multiple Mn3 SMMs to give discrete supramolecular aggregates of SMMs from the use of a variety of ligands and represents a step toward practical applications of SMMs. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2014.
Local:
Adviser: CHRISTOU,GEORGE.
Local:
Co-adviser: TALHAM,DANIEL R.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2016-08-31
Statement of Responsibility:
by Tu N Nguyen.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
8/31/2016
Classification:
LD1780 2014 ( lcc )

Downloads

This item has the following downloads:


Full Text









PAGE 1

ASupramolecularAggregateofFourExchange-Biased Single-MoleculeMagnetsTuN.Nguyen, WolfgangWernsdorfer,!KhalilA.Abboud, andGeorgeChristou*, DepartmentofChemistry,UniversityofFlorida,Gainesville,Florida32611-7200,UnitedStates!InstitutNe el-CNRS,38042GrenobleCedex9,France*SSupportingInformationABSTRACT: Thereactionbetween3-phenyl-1,5-bis(pyridin-2-yl)pentane-1,5-dionedioxime(pdpdH2)and triangular[MnIII 3O(O2CMe)(py)3](ClO4)(1)affords [Mn12O4(O2CMe)12(pdpd)6)](ClO4)4(3).Complex 3 hasarectangularshapeandconsistsoffour[MnIII 3O]7+triangularunitslinkedcovalentlybythedioximateligands intoasupramolecular[Mn3]4tetramer.Solid-statedcand acmagneticsusceptibilitymeasurementsrevealedthat [Mn3]4containsfourMn3single-moleculemagnets (SMMs),eachwithan S =6groundstate.Magnetization versusdc-fieldsweepsonasinglecrystalgavehysteresisloops below1Kthatexhibitedexchange-biasedquantumtunneling ofmagnetizationsteps,confirming 3 tobeasupramolecular aggregateoffourweaklyexchange-coupledSMMunits.Single-moleculemagnets(SMMs)areindividualmolecules thatfunctionassingle-domainnanoscalemagneticparticles belowtheirblockingtemperature, TB.1Thisbehaviorarises fromthecombinationofalarge-spin(S)groundstateand Ising-typemagnetoanisotropy(negativezero-fieldsplitting parameter D),whichleadstofrequency-dependentout-ofphasealternatingcurrent(ac)magneticsusceptibilitysignals andhysteresisinaplotofmagnetizationversusapplieddirect current(dc)magneticfield.1SMMshavealsobeenshownto displayinterestingquantumphenomenasuchasquantum tunnelingofmagnetization(QTM)2andquantumphaseinterference(QPI).3Consequently,theyhavebeenproposedas qubitsforquantumcomputation4andascomponentsin molecularspintronicsdevices,5whichwouldexploittheir quantum-tunnelingproperties.Forsuchapplications,coupling oftwoormoreSMMstoeachotherortoothercomponentsof adeviceareessential,butthecouplingmustbeveryweakin ordertomaintaintheintrinsicsingle-moleculepropertiesof eachSMM.ThereportofsupramolecularCHClhydrogenbondedpairsof S =9/2[Mn4O3Cl4(O2CEt)3(py)3]SMMs demonstratedsuchcouplingforthefirsttime,manifestedas exchange-biasedQTMsteps,quantum-superpositionstates, andquantumentanglementofthetwoSMMs.6,7Several supramoleculardimers,chains,and3Dnetworksofweakly coupledSMMsconnectedbyH-bondshavesincebeen reported.8ThedisadvantagesoflinkagebyH-bonds,however, are(i)deaggregationintomonomericunitsupondissolution and(ii)majorlossofsyntheticcontrol,withalltheabove examplesofsupramolecularaggregationbyH-bondsinfact havingbeenobtainedserendipitously.Asuperiorapproachis connectionofSMMsviacovalentbonds.Suchcovalentlinkage ofSMMshasalreadybeenexploredextensivelyandusuallyhas beenfoundtoleadto1D,2D,or3Dpolymers.9Thecoupling betweentheseSMMsisoften(butnotalways10)strongenough toleadtolossofSMMidentity,giving1Dsingle-chainmagnets (SCMs)or2D-or3D-orderedmaterials.However,significant progresshasbeenmadeincovalentlinkageoftwonon-SMM unitsforquantum-computingapplications,suchaslinkingof twoCr7Niwheels11oroftwolanthanideions,12resultingin weakantiferromagnetic(AF)interactionsbetweenthem. Ourgrouphasthereforeinitiatedanewefforttolinktwoor moreMnSMMscovalentlytogivenonpolymeric,supramolecular clustersofSMMs showingveryweakinter-SMM interactions.Wehereinreportasupramolecularaggregateof fourMn3SMMsconnectedbyanewlydesigneddioximegroup, 3-phenyl-1,5-bis(pyridin-2-yl)pentane-1,5-dionedioxime (pdpdH2)(Figure1).Thestrategyisbasedontheobservation thatmethylpyridine-2-ylketoneoxime(mpkoH)reactswith thenon-SMMtriangularcomplex[Mn3O(O2CMe)6(py)3](ClO4)(1)toconvertittothe S =6SMM[Mn3O(O2CMe)3(mpko)3](ClO4)(2).13ThenewgrouppdpdH2consistsoftwompkoH2groupslinkedbyabenzylunitand isdesignedtoconnecttwo[Mn3O]7+units.Itwassynthesized intwostepsfrom2-acetylpyridine;theintermediate3-phenyl1,5-bis(pyridin-2-yl)pentane-1,5-dionewasobtainedaccording toaliteraturemethod14andtreatedwithhydroxylamineto formthecrudeproduct.Recrystallizationfromacetonegave purepdpdH2in55%overallyield.15Reactionof 1 with1.5equivofpdpdH2inCH2Cl2gavea dark-brownsolution.Thesolutionwasfiltered,andthefiltrate wasleftundisturbedatambienttemperature.X-ray-quality crystalsof[Mn12O4(O2CMe)12)(pdpd)6](ClO4)4xCH2Cl2(3xCH2Cl2)slowlyformedover2daysandwerecollected byfiltration;theyieldwas35%.Complex 3xCH2Cl2 Received: September16,2011 Published: December2,2011 Figure1. Structuresof(left)mpkoHand(right)pdpdH2. Communication pubs.acs.org/JACS 2011AmericanChemicalSociety20688dx.doi.org/10.1021/ja2087344 | J.Am.Chem.Soc. 2011,133,2068820691

PAGE 2

crystallizesinthemonoclinicspacegroup P21/c.16The asymmetricunitconsistsoftwoessentiallysuperimposable Mn12cations(oneisshowninFigure2),eightClO4 anions, andlargeamountsofdisorderedCH2Cl2solvent.16TheMn12cationconsistsoffour[Mn3(3-O)]7+unitslinkedbysixpdpd2groupstogiveasupramolecular[Mn3]4rectangle.Oneofthetwo 1:1:-MeCO2 ligandsbridgingeachedgeof 1 isreplacedbya bridgingoximatefromapdpd2group.Twoofthelatterbridgeto thesameneighboringMn3unit,andthethirdbridgestoadifferent one(Figure2,bottom).Inaddition,thepdpd2pyridylgroups replacetheterminalpyridineligandsof 1.Eachofthe 3-O2ions liesslightlyaboveitsMn3plane(r H 0.3),asin 2.Thus,thelocal structureofeachMn3unitof 3 isverysimilartothatof 2, comprisingatriangular[Mn3(3-O)]7+unitwhoseedgesareeach bridgedbyoneacetateandonepyridyloximategroup,thepyridyl groupofwhichbindsterminallytotheMn.Alsoasin 2,thethree bridgingoximategroupsareonthesamesideoftheMn3plane,and thisisonefactorfavoringtheformationofamoleculartetramer ratherthanapolymer.Infact,wehadanticipatedthattheproduct mightbean[Mn3]4tetrahedronwithabridgingpdpd2oneach edge,buttheobtainedrectangleisalsoalogicalarrangement.The cationhascrystallographic C1andvirtual D2symmetry.TheMnIIIoxidationstateswereconfirmedbybondvalencesumcalculations,17andtheirJahnTellerelongationaxes(greenbondsinFigure2, bottom)arealignedinapropellerfashion,againasin 2.The MnMnseparationsandMn(3-O)Mnanglesineachtriangle areslightlydifferent;thus,thetrianglesarescalenebutvirtually isosceleswithintheusual3 criterion.17Overall, 3 canaccuratelybe describedasatetramericversionofSMMcomplex 2,suggesting thateachMn3unitof 3 mightalsobeanSMM. Variable-temperaturedcmagneticsusceptibility(M)measurementswereperformedonapolycrystallinesampleof 32CH2Cl216inanappliedfieldof1000G(0.10T)overthe 5.0300Ktemperaturerange.Thesamplewasrestrainedin eicosanetopreventtorquing. MT increasedfrom48.25cm3K mol1at300Ktoaplateauvalueof76.55cm3Kmol1at20K andthendecreasedslightlyto70.62cm3Kmol1at5.0K (Figure3).The300Kvalueismuchlargerthanthespin-only (g =2)valuefor12MnIIIatoms(MT =36cm3Kmol1),and thepeakvalueatlow T isasexpectedforfournoninteracting S =6unitswith g slightlylessthan2.0(spin-only MT =84cm3Kmol1).Thedecreasein MT below20Kisassignedtozerofieldsplitting(ZFS),Zeemaneffectsfromtheappliedfield,and weakintermolecularinteractions.Theoverall MT versus T profileisextremelysimilartothatforcomplex 2 (S =6), indicatingthateachofthefourMn3unitsof 3 isalso ferromagneticallycoupledwithan S =6groundstate.Thedata werefittothetheoretical MT versus T expressionforfour independentandequivalentMnIII 3isoscelestrianglesper 3;13,17thespinHamiltonianisgivenineq1: (1)Onlydatafor T e 20Kwereusedbecausethelow-T decrease isduetofactorsnotincludedineq1.Thefitgave J =+16.8(6) cm1, J2 =+1.5(7)cm1,and g =1.91(1),withtemperatureindependentparamagnetism(TIP)keptconstantat600 106cm3mol1.SincethefourMn3unitsarecrystallographically inequivalent,the J and J2 areaveragevalues,buttheyaresimilar tothosefor 2 andanalogueswithothercarboxylates(J =+12.1 to+18.6cm1and J2 =+1.5to+6.7cm1). Magnetization(M)datawerecollectedoverthe0.17Tand 1.810Krangesandareplottedas M/NBversus H/T in Figure4,where N isAvogadrosnumberand BistheBohr magneton.ThedatawerefitusingtheprogramMAGNETby diagonalizationofthespinHamiltonianmatrixassumingthat onlythegroundstateispopulated,incorporatingaxial anisotropy(DS z 2)andZeemanterms,andemployingafull Figure2. (top)Completemolecularstructureofthecationof 3,with Hatomsomittedforclarity.(bottom)Structureofthecore, emphasizingtheconnectivitybetweentheMn3unitsandshowing theMn3planes(green-shadedtriangles)andJahnTelleraxes(green bonds).Colorcode:Mn,green;N,blue;O,red;C,gray. Figure3. Plotof MT vs T for 32CH2Cl2.Thesolidlineisthefitto thedata;seethetextforthefitparameters. JournaloftheAmericanChemicalSociety Communicationdx.doi.org/10.1021/ja2087344 | J.Am.Chem.Soc. 2011,133,206882069120689

PAGE 3

powderaverage.ThespinHamiltonianisgivenbyeq2, (2)where D istheaxialZFSparameter, 0isthevacuum permeability,and H istheappliedmagneticfield.Thefit(solid linesinFigure4)gave S =6, D = 0.30(2)cm1,and g = 1.92(1),againverysimilarto 2 [S =6, D = 0.34(2)cm1,and g =1.92(1)].13Thecombineddcdatathuscomplementthe structuraldatainsupportingtheconclusionthatcomplex 3 isa tetrameroffour S =6Mn3unitslikethatin 2 andthatthese unitsinteractwitheachotheronlyveryweakly(tooweaklyto affecttheabovefits,whichassumenoninteractingMn3units). ToprobetheSMMpropertiesof 32CH2Cl2,out-of-phaseac susceptibility(M3 )versus T datawerecollectedoverthe1.815K rangeusinga3.5Gacfieldoscillatingatfrequenciesofupto1000 Hz(Figure5). M3 signalsthataretailsofpeakslyingbelow1.8K wereobservedat<3K,andthesesuggestthat 3 mightbeatetramer ofSMMs.Therefore,dcmagnetizationversusfieldscansonsingle crystalsof 3xCH2Cl2werecarriedoutonamicro-SQUID.18Hysteresisloopswereseenbelow <1.0K(Figure6),andtheir coercivitiesincreasedwithdecreasingtemperatureandincreasing fieldsweeprate,asexpectedforSMMs.Wethusconcludethateach ofthefourMn3unitsin 3xCH2Cl2isanSMM,asisMn3complex 2 thattheycloselyresembleinstructureandmagneticproperties. WenowaddresswhetherthefourMn3SMMunitsin 3xCH2Cl2interactweaklywitheachother.Theansweris clearlyyes,becausethehysteresisloopsshowanexchangebias oftheQTMsteps.ThefirststepinthehysteresisloopofanSMM onscanningfromnegativetopositivefieldsisnormallyatzero field.Thisiswherethe MSlevelsoneithersideoftheanisotropy barrierareinresonanceandQTMcanoccur,reversingthe orientationofthemagnetizationvector.ThepresenceofanAF exchange-coupledneighborprovidesabiasfieldthatshiftsthe resonancetunneling(QTMstep)toanewpositionbeforezero field.Thiswasfirstseenforthehydrogen-bonded[Mn4]2dimerof S =9/2SMMs,6andarelatedexplanationcanbeprovidedfor 3, exceptthateachMn3SMMisnowexchange-coupledtotwo neighboringMn3SMMs.TheloopsofFigure6clearlyestablish weakAFinteractionsbetweentheMn3subunitsof 32CH2Cl2.As thefieldisscannedfrom 1T,wherethefourMn3spinvectors arepolarizedintothe MS= 6orientation,thefirststep correspondstotunnelingofaMn3vectorfrom MS= 6to MS= +6;thisoccursat 0.18T,whichthusequalsthetotalbiasfield fromtwo MS= 6neighbors.Intheformat(MS1, MS2, MS3, MS4), where i =14refertothefourMn3SMMsof 3,the 0.18Tstep isthe(6, 6, 6, 6)to(6,+6, 6, 6)tunnelingtransition (Forclarity,degeneratestatessuchas(6, 6,+6, 6),(+6, 6, 6, 6),etc.,arenotlisted).Since 3 isarectangular[Mn3]4aggregate,thereshouldbetwodifferentinter-Mn3interactions, J1and J2,whicharelikelycomparablebutnotidenticalinmagnitude; thediagonalinteractionshouldbemuchweakerandisignoredin thisdiscussion.Thesecondstepatzerofieldisassignableto tunnelingofmoleculeswith(6,+6)neighbors,yieldingzerobias if J1= J2orasmallbiasrelatedto |J1 J2| if J1` J2.Thestepat zerofieldisthe(6,+6, 6, 6)to(6,+6,+6, 6)transition, followedbythepossibilityofaflip-floprelaxationtothe(6,+6, 6,+6)groundstate.Aplotofspin-stateenergyversusapplied fieldshowingtheavoidedlevelcrossings,simulatedwith J1= J2= 0.02Kand D = 0.31cm1,isprovidedintheSI;17adetailed analysiswillbeprovidedinthefullpaperonthiswork. Furthermore,ifenoughvectorshavetunneledto+6inthefirst Figure4. Plotof M/NBvs H/T perMn3unitof 32CH2Cl2.The solidlinesarethefitstothedata;seethetextforthefitparameters. Figure5. Plotsoftheout-of-phaseacsusceptibility(M3 )vs T for 32CH2Cl2attheindicatedfrequencies. Figure6. Hysteresisloopsinplotsofmagnetizationvsdcfieldfora singlecrystalof 3xCH2Cl2(top)attheindicatedtemperaturesat0.28 T/sand(bottom)attheindicatedfieldscanratesat0.04K. M is normalizedtoitssaturationvalue, MS. JournaloftheAmericanChemicalSociety Communicationdx.doi.org/10.1021/ja2087344 | J.Am.Chem.Soc. 2011,133,206882069120690

PAGE 4

twosteps,thenastepisexpectedforMn3vectorstunnelinginthe presenceofa(+6,+6)bias.Thisshouldoccurat+0.18Tandis indeedseeninFigure6(bottom)at0.001T/s;atthislowscanrate, enoughspinshavehadtimetoreverseinthefirsttwostepstoallow the(+6,+6)situation.Atfasterscanrates,thisstepdisappears,and eventhezero-fieldstep[(6,+6)bias]becomessmallerbecause thetunnelingprobabilitydecreaseswithincreasingscanrate(i.e., fewermoleculesareinresonancelongenoughtotunnel).Athigh scanrates,atleasttwonewstepsappearinthe+0.3to+0.6Trange, alsoinvolving MS= 5levels.17Thesteppatternthusleadstothe conclusionthat 3 isasupramolecularaggregateoffourweakly interactingMn3SMMs,witheachMn3coupledtotwoneighbors. Insummary,anewlydesigneddioximehasyieldedasupramolecular[Mn3]4aggregateoffourcovalentlylinkedSMMs.Eachof theMn3subunitsof 3 isstructurallysimilartodiscrete 2,andthe magneticpropertiesarethereforealsonearlyidentical.Hysteresis studiesshowedthateachofthefourMn3SMMsisweaklycoupled totwoneighbors,leadingtoanexchangebiasoftheQTMsteps whosemagnitudedependsonthespinalignmentsofthese neighbors.Weassumeatthispreliminarystagethattheinter-Mn3interactionisviasuperexchangethroughthebridgingligands,since 2 showsnoexchangebiasofitsQTMstepsfromintermolecular dipolarinteractions.Unfortunately,theQTMstepsarerelatively broad,possiblyasaresultofthefollowing:(i)thetwoMn12cations intheasymmetricunithavedifferentorientations,andthefourMn3planeswithineachcationarenotcoplanar(Figure2,bottom).The appliedfieldwillthusbeatarangeofanglestotheeasy(z)axesof theeightMn3,leadingtostepbroadening;(ii) J1` J2,andnonzero diagonalinteractionswithin[Mn3]4orinteractionsbetweenseparate unitswillgivearangeofbiasfieldsforagiven(6, 6)situation. WearethusintroducingbulkiercarboxylatestoisolateMn12cations moreeffectivelyandseekingcrystalswithallcationsparallel.Notably, thecovalentlinkagewithinthe[Mn3]4assemblyresultsinretention ofthestructureupondissolution,allowingstudiesofanexchangebiasedsysteminfluidandfrozensolutionsforthefirsttime. Inconclusion, 3 confirmsthefeasibilityofcovalentlyconnecting multipleMn3SMMstogiveadiscretesupramolecular clusterof SMMs withonlyweakcouplingbetweenthem.Thisshouldlead totheiralsobeingquantum-mechanicallycoupled,asfoundfor [Mn4]2dimers,andrepresentsasteptowardthedevelopmentofa multiqubitsystembasedonSMMs.Suchstudiesarecurrentlyin progress,asareadditionalsyntheticeffortstoproduceother supramolecularaggregatesofweaklycoupledSMMs.ASSOCIATEDCONTENT*SSupportingInformation Crystallographicdetails(CIF),bondvalencesums,bonddistances andangles,NMRspectra,andmagnetismdata.Thismaterialis availablefreeofchargeviatheInternetathttp://pubs.acs.org.AUTHORINFORMATIONCorrespondingAuthor christou@chem.ufl.eduACKNOWLEDGMENTSThisworkwassupportedbyNSF(CHE-0910472).T.N.N.thanks theVietnamEducationFoundationforafellowship,andW.W. acknowledgestheERCAdvancedGrantMolNanoSpinNo.226558.REFERENCES(1)(a)Sessoli,R.;Gatteschi,D.;Caneschi,A.;Novak,M.A. Nature 1993, 365,141.(b)Sessoli,R.;Tsai,H.L.;Schake,A.R.;Wang,S.; Vincent,J.B.;Folting,K.;Gatteschi,D.;Christou,G.;Hendrickson,D. N. J.Am.Chem.Soc. 1993, 115,1804.(c)Bagai,R.;Christou,G. Chem. Soc.Rev. 2009, 38,1011. (2)Friedman,J.R.;Sarachik,M.P.;Tejada,J.;Ziolo,R. Phys.Rev. Lett. 1996, 76,3830. (3)Wernsdorfer,W.;Sessoli,R. Science 1999, 284,133. (4)Leuenberger,M.N.;Loss,D. Nature 2001, 410,789. (5)Bogani,L.;Wernsdorfer,W. Nat.Mater. 2008, 7,179. (6)Wernsdorfer,W.;Aliaga-Alcalde,N.;Hendrickson,D.N.; Christou,G. Nature 2002, 416,406. (7)(a)Hill,S.;Edwards,R.S.;Aliaga-Alcalde,N.;Christou,G. Science 2003, 302,1015.(b)Tiron,R.;Wernsdorfer,W.;FoguetAlbiol,D.;Aliaga-Alcalde,N.;Christou,G. Phys.Rev.Lett. 2003, 91, No.227203. (8)(a)Tiron,R.;Wernsdorfer,W.;Aliaga-Alcalde,N.;Christou,G. Phys.Rev.B. 2003, 68,No.140407.(b)Yang,E.-C.;Wernsdorfer,W.; Hill,S.;Edwards,R.S.;Nakano,M.;Maccagnano,S.;Zakharov,L.N.; Rheingold,A.L.;Christou,G.;Hendrickson,D.N. Polyhedron 2003, 22,1727.(c)Wittick,L.M.;Murray,K.S.;Moubaraki,B.;Batten,S. R.;Spiccia,L.;Berry,K.J. DaltonTrans. 2004,1003.(d)Bagai,R.; Wernsdorfer,W.;Abboud,K.A.;Christou,G. J.Am.Chem.Soc. 2007, 129,12918.(e)Inglis,R.;Jones,L.F.;Mason,K.;Collins,A.; Moggach,S.A.;Parsons,S.;Perlepes,S.P.;Wernsdorfer,W.;Brechin, E.K. Chem.’Eur.J. 2008, 14,9117.(f)Das,A.;Gieb,K.;Krupskaya, Y.;Demeshko,S.;Dechert,S.;Klingeler,R.;Kataev,V.;Buchner,B.; Muller,P.;Meyer,F. J.Am.Chem.Soc. 2011, 133,3433. (9)(a)Coulon,C.;Miyasaka,H.;Cle rac,R. Struct.Bonding(Berlin) 2006, 122,163.(b)Xu,H.-B.;Wang,B.-W.;Pan,F.;Wang,Z.-M.; Gao,S. Angew.Chem.,Int.Ed. 2007, 46,7388.(c)Miyasaka,H.; Yamashita,M. DaltonTrans. 2007,399.(d)Bogani,L.;Vindigni,A.; Sessoli,R.;Gatteschi,D. J.Mater.Chem. 2008, 18,4750.(e)Roubeau, O.;Cle rac,R. Eur.J.Inorg.Chem. 2008,4325. (10)(a)Miyasaka,H.;Nakata,K.;Lecren,L.;Coulon,C.;Nakazawa, Y.;Fujisaki,T.;Sugiura,K.;Yamashita,M.;Cle rac.,R. J.Am.Chem. Soc. 2006, 128,3770.(b)Langley,S.K.;Chilton,N.F.;Moubaraki,B.; Murray,K.S. DaltonTrans. 2011, 40,12201.(c)Shiga,T.;Miyasaka, H.;Yamashita,M.;Morimoto,M.;Irie,M. Dalton.Trans. 2011, 40, 2275. (11)Bellini,V.;Lorusso,G.;Candini,A.;Wernsdorfer,W.;Faust,T. B.;Timco,G.A.;Winpenny,R.E.P.;Affronte,M. Phys.Rev.Lett. 2011, 106,No.227205. (12)Arom1,G.;Aguila,D.;Gamez,P.;Luis,F.;Roubeau.O. Chem. Soc.Rev. [Onlineearlyaccess].DOI:10.1039/c1cs15115k.Published Online:Aug4,2011. (13)Stamatatos,T.C.;Foguet-Albiol,D.;Lee,S.-C.;Stoumpos,C. C.;Raptopoulou,C.P.;Terzis,A.;Wernsdorfer,W.;Hill,S.O.; Perlepes,S.P.;Christou,G. J.Am.Chem.Soc. 2007, 129,9484. (14)Constable,E.C.;Lewis,J.;Liptrot,M.C.;Raithby,P.R. Inorg. Chim.Acta 1990, 178,47. (15)NMRspectraofpdpdH2areshownintheSI. (16)Anal.Calcd(Found)for 32CH2Cl2(C152H148Cl8Mn12N24O56): C,43.99(43.58);H,3.59(3.53);N,8.10(7.75).Thecrystalstructure showsthattwoCH2Cl2moleculesareencapsulatedwithintheMn12cation(FigureS3intheSI).17Theelementalanalysisindicatestheir retentionevenafterdryinginvacuum.Crystalstructuredatafor 3xCH2Cl2:C150H144Cl4Mn12N24O56(excl.CH2Cl2):FW=3980.05; monoclinic,spacegroup P21/c; a =34.190(4), b =32.890(4), c = 44.013(5), =110.320(3); V =46413(9)3, Z =8; T =100(2)K; R1[I >2(I)]=0.0740, wR2(F2,alldata)=0.1886. (17)SeetheSupportingInformation(SI). (18)Wernsdorfer,W. Adv.Chem.Phys. 2001, 118,99.NOTEADDEDAFTERASAPPUBLICATIONThispaperwaspublishedontheWebonDecember2,2011, withanerrorineq1.Thecorrectedversionwasrepostedon December6,2011. JournaloftheAmericanChemicalSociety Communicationdx.doi.org/10.1021/ja2087344 | J.Am.Chem.Soc. 2011,133,206882069120691





PAGE 1

1 SUPRAMOLECULAR AGGREGATES OF SINGLE MOLECULE MAGNETS By TU NGOC NGUYEN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHI LOSOPHY UNIVERSITY OF FLORIDA 2014

PAGE 2

2 © 2014 Tu Ngoc Nguyen

PAGE 3

3 To my parents, my honey , and my sister for their wholehearted love and confidence in me

PAGE 4

4 ACKNOWLEDGMENTS Over my academic years , I have been blessed to work with so many smart, creative, fun, and passionate individuals, and this dissertation would not have been possible without their enormous support, in one way or another. I owe my deepest gratitude to my advisor, Professor George Christou, for his invaluable guidance, advice, support , and understanding all these years. Every discussion we have had and every meeting we have been through is my endless source of inspiration and ideas. His ingenuity and insight are extraordinary, but even more so are his dedication to science and his never ending support for his students. His unrelenting confidence in me has helped to transform me from the young boy, shy and purposeless, on the day I arrived in the States, to the one who ha s a much better idea o f what I want to do in life and what my strengths are as I am now. He is the role model for the type of advisor that I strive to become: knowledgeable, dedicated, persevering, inspiring, and full of generous spirit and curiosity. In addition, I ow e my debt to Dr. Daniel R. Talham, Dr. Adam S. Veige, Dr. Mark W. Meisel , Dr. Helena Hagelin Weaver, and Dr. Kenneth B. Wagener for serving as my graduate committee and following my research progress. I deeply thank Dr. Wolfgang Wernsdorfer for micro SQUID measurements, Dr. Stephen Hill and his student Muhandis Shiddiq for HF EPR studies. My special thanks go to Dr. Khalil A. Abboud for all crystal structure determinations and introduction to basics of X ray crystallography. I extend my gratitude to all the staff of the Center for X ray Crystallography at UF for helping to collect unit cells and instructing to work with the X ray instrument. I also warmly thank Dr. Spyros Perlepes, Dr. Anastasios Tasiopoulos, Dr. Theocharis Stamatatos, Dr. Naresh Dalal , and Dr. Enrique del Barco for interesting collaborations.

PAGE 5

5 I greatly thank Dr. Benjamin Smith and Lori Clark for helping me with all matters related to my VEF fellowship. The Grinter Award given to me by Dr. Smith really helped me a lot when I first ca me to the United States. I would also like to thank my teaching advisor Dr. James Horvath for his kindness and professional guidance. I would like to thank the entire Christou group, especially Dinos, Ninetta, Nemo, Galia, Arpita, Shreya, Antonio, Taketo, Linh, Andy, Yan, Andrew, Annaliese, Adeline, Daisuke, Kylie, Katye and Tuhin for their friendship and support. My special thanks go to Shreya, who helped and gave me invaluable suggestion s when I started my project, Tuhin, who helped me in the synt hesis of dpdH 2 , Andrew, Annaliese, and Kylie, who proof read this document . I also would like to thank Alice for helping me out with many official documents. During my time in the States, whenever I was nostalgic of the home that was thousands of miles fro m here, my Vietnamese friends here have made life a lot more fun and interesting for me, and I a m thankful for all the laughter , the dinners, the parties, the Vietnamese food they cooked, the opportunities to talk for hours in my mother tongue they playe d no small part in keeping me going. It has been my privilege to get to know you all. Studying in the United States is a dream come true for me but a sacrifice by my family. To em , my wife and best friend, who ha s patiently and repeatedly lent me y our ears to listen to my dreams, big and small alike, you have my deepest appreciation and affection. Thank you for making my life so beautiful with your caring and generous love. To my sister, I would like to thank you for taking care of my parents while I am away from home. Last but not least, to my remarkable parents, I am forever indebted to you for your lifetime of unconditional love, support , and faith in me. I owe so much to you all, my small but wonderful family.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 9 LIST OF FIGURES ................................ ................................ ................................ ....................... 12 LIST OF ABBREVIATIONS ................................ ................................ ................................ ........ 21 ABSTRACT ................................ ................................ ................................ ................................ ... 22 CHAPTER 1 GENERAL INTRODUCTION ................................ ................................ .............................. 24 1 .1 Supramolecular Chemistry ................................ ................................ ............................... 24 1.2 Single molecule Magnets ................................ ................................ ................................ . 32 2 SUPRAMOLECULAR AGGREGATES OF SINGLE MOLECULE MAGNETS FROM THE USE OF 3 PHENYL 1,5 DI(PYRIDIN 2 YL)PENTANE 1,5 DIONE DIOXIME: DEVELOPING A RATIONAL DESIGN APPROACH ................................ .... 54 2.1 Introduction ................................ ................................ ................................ ....................... 54 2.2 Experimental S ection ................................ ................................ ................................ ........ 57 2.2.1 Syntheses ................................ ................................ ................................ ................ 57 2.2.1.1 [Mn 6 O 2 (O 2 CMe) 8 (MeOH) 2 (pdpd) 2 ] (2 1) ................................ .................... 57 2.2.1.2 [Mn 6 O 2 (O 2 CMe) 8 (py) 2 (pdpd) 2 ](ClO 4 ) 2 (2 2) ................................ ............... 58 2.2.1.3 [Mn 12 O 4 (O 2 CMe) 12 (pdpd) 6 ](ClO 4 ) 4 (2 3) ................................ .................... 58 2.2.1.4 [Mn 12 O 4 (O 2 C t Bu) 12 (pdpd ) 6 ](ClO 4 ) 4 (2 4) ................................ .................... 59 2.2.2 X Ray Crystallography ................................ ................................ ........................... 59 2.2.3 Physical Measurements ................................ ................................ .......................... 61 2.3 Results and Discussion ................................ ................................ ................................ ..... 62 2.3.1 Syntheses ................................ ................................ ................................ ................ 62 2.3.2 Description of Structures ................................ ................................ ........................ 65 2.3.3 Magnetochemistry ................................ ................................ ................................ .. 67 2.3.3.1 Direct current magnetic susceptibility studies ................................ .............. 67 2.3.3.2 Magnetization versus D C magnetic field studies ................................ ......... 71 2.3.3.3 Alternating current magnetic susceptibility studies ................................ ..... 72 2.3.3.4 Magnetization decay studies ................................ ................................ ........ 74 2.3.3.5 Magnetization versus DC field hysteresis loops ................................ .......... 75 2.3.3.6 High frequency electron paramagnetic resonance (HF EPR) studies .......... 78 2.4 Summary and Conclusions ................................ ................................ ............................... 81

PAGE 7

7 3 SPHERICAL HOST GUEST SUPRAMOLECULAR AGGREGATES OF SINGLE MOLECULE MAGNETS FROM THE USE OF 1,2 DI(PYRIDIN 2 YL)ETHANE 1,2 DIONE DIOXIME ................................ ................................ ................................ ......... 117 3.1 Introduction ................................ ................................ ................................ ..................... 117 3.2 Experimental Section ................................ ................................ ................................ ...... 118 3.2.1 Syntheses ................................ ................................ ................................ .............. 118 3.2.1.1 [Mn III 9 Mn II 3 O 4 (O 2 CMe) 12 (ded) 6 (pyH)](ClO 4 ) 2 (3 1) .............................. 118 3.2.1.2 [Mn III 10 Mn II 2 O 4 (O 2 CEt) 10 (H 2 O) 2 Cl 2 ( ded) 6 (py)](ClO 4 ) 2 (3 2). ................. 119 3.2.1.3 [Mn III 12 O 4 (O 2 CMe) 12 (ded) 6 (CH 2 Cl 2 )](I 3 ) 3.5 I 0.5 (3 3) ............................... 119 3.2.1.4 [Mn III 12 O 4 (O 2 CEt) 12 (ded) 6 (EtOH )](I 3 ) 3.5 I 0.5 (3 4) ................................ .... 119 3.2.2 X Ray Crystallography ................................ ................................ ......................... 120 3.2.3 Physical Measurements ................................ ................................ ........................ 123 3.3 Results and Discussion ................................ ................................ ................................ ... 124 3.3.1 Syntheses ................................ ................................ ................................ .............. 124 3.3.2 Description of Structures ................................ ................................ ...................... 126 3.3.3 Supramolecular Chemistry ................................ ................................ ................... 128 3.3.4 Magnetochemistry ................................ ................................ ................................ 130 3.3.4.1 Direct current magnetic susceptibil ity studies ................................ ............ 130 3.3.4.2 Magnetization versus DC magnetic field studies ................................ ....... 132 3.3.4.3 Alternating current magnetic susceptibility studies ................................ ... 133 3.3.4.4 Magnetization versus DC field hysteresis loops ................................ ........ 134 3.4 Summary and Conclusions ................................ ................................ ............................. 137 4 FERROMAGNETIC EXCHANGE BIAS IN A DIMERIC SUPRAMOLECULAR AGGREGATE OF SINGLE MOLECULE MAGNETS ................................ ..................... 165 4.1 Introduction ................................ ................................ ................................ ..................... 165 4.2 Experimental Section ................................ ................................ ................................ ...... 167 4.2.1 Syntheses ................................ ................................ ................................ .............. 167 4.2.2 X Ray Crystallography ................................ ................................ ......................... 167 4.2.3 Physical Measurements ................................ ................................ ........................ 168 4.3 Results and Discussion ................................ ................................ ................................ ... 169 4.3.1 Syntheses ................................ ................................ ................................ .............. 169 4.3.2 Description of Structures ................................ ................................ ...................... 169 4.3.3 UV Vis Spectroscopy ................................ ................................ ........................... 170 4.3.4 Magnetochemistry ................................ ................................ ................................ 171 4.3.4.1 Direct current magnetic susceptibility studies ................................ ............ 171 4.3.4.2 Magnetization versus DC magnetic field studies ................................ ....... 173 4.3.4.3 Alternating current magnetic susceptibility studies ................................ ... 174 4.3.4.4 High frequency electron paramagnetic resonance (HF EPR) studies ........ 174 4.4 Summary and Conclusions ................................ ................................ ............................. 178 5 SUPRAMOLECULAR AGGREGATES OF SINGLE MOLECULE MAGNETS FROM THE USE OF DI AND TRICARBOXYLIC ACIDS ................................ ............. 195 5.1 Introduction ................................ ................................ ................................ ..................... 195

PAGE 8

8 5.2 Experimental Section ................................ ................................ ................................ ...... 196 5.2.1 Syntheses ................................ ................................ ................................ .............. 196 5.2.1.1 [Mn 12 O 4 (tma) 4 (mpko) 12 ](ClO 4 ) 4 (5 1) ................................ ....................... 197 5.2.1.2 [Mn 6 O 2 (dma) 3 (mpko) 6 ](ClO 4 ) 2 (5 2) ................................ ......................... 197 5.2.1 .3 [Mn 6 O 2 (txa) 4 (mpko) 6 ] (5 3) ................................ ................................ ....... 198 5.2.2 X Ray Crystallography ................................ ................................ ......................... 198 5.2.3 Physical Measurements ................................ ................................ ........................ 200 5.3 Results and Discussion ................................ ................................ ................................ ... 201 5.3.1 Syntheses ................................ ................................ ................................ .............. 201 5.3.2 Description of Structures ................................ ................................ ...................... 202 5.3.3 Magnetochemistry ................................ ................................ ................................ 203 5.3.3.1 Direct current magnetic susceptibility studies ................................ ............ 203 5.3.3.2 M agnetization versus DC magnetic field studies ................................ ....... 206 5.3.3.3 Alternating current magnetic susceptibility studies ................................ ... 207 5.3.3.4 Magnetizatio n versus DC field hysteresis loops ................................ ........ 207 5.4 Summary and Conclusions ................................ ................................ ............................. 209 APPENDIX A LIST OF COMPOUNDS ................................ ................................ ................................ ...... 234 B VAN VLECK EQUATIONS ................................ ................................ ............................... 235 C BOND DISTANCES AND ANGLES ................................ ................................ .................. 243 LIST OF REFERENCES ................................ ................................ ................................ ............. 265 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 279

PAGE 9

9 LIST OF TABLES Table page 2 1 Crystallographic data for complexes 2 1 ·xCH 3 OH·yCH 3 CN, 2 2 ·x C 4 H 10 O·yCH 2 Cl 2 , 2 3 · xCH 2 Cl 2 , and 2 4 ·xCH 2 Cl 2 . ................................ ................................ ....................... 83 2 2 Selected interatomic distances (Å) and angles ( o ) for complexes 2 1 and 2 2 . ................. 84 2 3 Selected interatomic distances (Å) and angles ( o ) for complexes 2 3 and 2 4 . ................. 85 2 4 Displacement of 3 oxide atoms from Mn 3 planes ( Å ) and Mn N O Mn torsion angles ( o ) of complexes 2 3 and 2 4 . ................................ ................................ ................. 86 2 5 Bond valence sum calculation for Mn a and selected O b atoms of complex 2 1 . ............... 87 2 6 Bond valence sum calculation for Mn a and s elected O b atoms of complex 2 2 . ............... 87 2 7 Bond valence sum calculation for Mn a and selected O b atoms of complex 2 3 . ............... 87 2 8 Bond valence sum calculation for Mn a and selected O b atoms of complex 2 4 . ............... 88 2 9 Parameters of the fits to Van Vleck equation. ................................ ................................ ... 88 2 10 Reduced ma gnetization fit parameters. ................................ ................................ .............. 88 2 11 Linear regression fit parameters. ................................ ................................ ........................ 88 3 1 Crystallographic data for complexes 3 1 ·xMeCN, 3 2 ·xMeCN, 3 3 ·xCH 2 Cl 2 ·yMeCN, and 3 4 ·xCH 2 Cl 2 . ................................ ................................ ............ 139 3 2 Selected interatomic distances (Å) and angles ( o ) for complexes 3 1 and 3 2 . ............... 140 3 3 Sele cted interatomic distances (Å) and angles ( o ) for complexes 3 3 and 3 4 . ............... 141 3 4 Displacement of 3 oxide atoms from Mn 3 planes ( Å ) and Mn N O Mn torsion angles ( o ) of complexes 3 3 and 3 4 . ................................ ................................ ............... 142 3 5 Bond valence sum calculation for Mn a and selected O b atoms of complex 3 1 . ............. 142 3 6 Bond valence sum calculation for Mn a and select ed O b atoms of complex 3 2 . ............. 143 3 7 Bond valence sum calculation for Mn a and selected O b atoms of complex 3 3 . ............. 143 3 8 Bond valence su m calculation for Mn a and selected O b atoms of complex 3 4 . ............. 143 4 1 Crystallographic data for complex 4 1 ·xCH 2 Cl 2 . ................................ ............................ 180

PAGE 10

10 4 2 Selected in teratomic distances (Å), bond angles ( o ), displacements (Å) of 3 oxide ions from Mn 3 planes, and Mn N O Mn torsion angles ( o ) of complex 4 1 . ................... 180 4 3 Several lowest eigenfunctions of the [Mn 3 ] 2 dimer. Only the lowest 5 multiplets are listed. ................................ ................................ ................................ ................................ 181 5 1 Crystallographic data for complexes 5 1 ·xCHCl 3, 5 2 ·xMeNO 2 ·yMeCN·zEt 2 O, and 5 3 ·xMe 2 CO·yMeCN. ................................ ................................ ................................ ..... 210 5 2 Selected interatomic distances (Å) and angles ( 0 ) for complex 5 1 . ................................ 211 5 3 Selected interatomic distances (Å) and angles ( o ) for complexes 5 2 and 5 3 . ............... 212 5 4 Displacement of 3 oxide atoms from Mn 3 planes (Å) and Mn N O Mn torsion angles ( o ) of complexes 5 1 , 5 2 and 5 3 . ................................ ................................ ........ 213 5 5 Bond valence sum calcu lation for Mn a and selected O b atoms of complex 5 1 . ............. 214 5 7 Bond valence sum calculation for Mn a and selected O b atoms of complex 5 3 . ............. 214 5 8 Parameters of the fits to Van Vleck equation. ................................ ................................ . 215 5 9 Reduced magnetization fit parameters. ................................ ................................ ............ 215 C 1 Selected interatomic dis tances (Å) and angles ( o ) for complex 2 1 . ................................ 243 C 2 Selected interatomic distances (Å) and angles ( o ) for complex 2 2 . ................................ 244 C 3 Selected interatomic distances (Å) and angles ( o ) for both cations in the asymmetric unit of complex 2 3 . ................................ ................................ ................................ ......... 245 C 4 Selected interatomic distances (Å) and angles ( o ) for complex 2 4 . ................................ 249 C 5 Selected interatomic distances (Å) and angles ( o ) for both cations in the asymmetric unit of complex 3 1 . ................................ ................................ ................................ ......... 250 C 6 Selected interatomic distances (Å) and angles ( o ) for complex 3 2 . ................................ 254 C 7 Selected interatomic distances (Å) and angles ( o ) for complex 3 3 . ................................ 255 C 8 Selected interatom ic distances (Å) and angles ( o ) for complex 3 4 . ................................ 256 C 9 Selected interatomic distances (Å) and angles ( o ) for complex 4 1 . ................................ 257 C 10 S elected interatomic distances (Å) and angles ( o ) for complex 5 1 . ................................ 258 C 11 Selected interatomic distances (Å) and angles ( o ) for both cations in the asymmetric unit of complex 5 2 . ................................ ................................ ................................ ......... 260

PAGE 11

11 C 12 Selected interatomic distances (Ã…) and angles ( o ) for complex 5 3 . ................................ 263

PAGE 12

12 LIST OF FIGURES Figure page 1 1 Examples of a crown ether (left), a cryptand (middle), and a spherand (right). ................ 44 1 2 DNA structure (left) with hydrogen bonding pattern (right) between nucleotides. 124 ....... 45 1 3 Prediction of two dimensional (top) and three dimensional architectures (bottom). Reprinted with permission from references 27 and 72. Copyright 2011 American Chemical Society and Copyright 2000 National A cademy of Sciences, U.S.A. ............... 46 1 4 Design strategy for (a) M 4 L 6 and (b) M 4 L 4 tetrahedra. Reprinted with permission from reference 27. Copyright 2011 American Chemical Society. ................................ ..... 47 1 5 Design strategy for M 6 L 3 trigonal prism and M 12 L 6 hexagonal prism. Reprinted with permission from reference 27. Copyright 2011 American Chemical Society. .................. 47 1 6 Weak link approach to access supramolecules. Reprinted with permission from reference 27. Copyright 2011 American Chemical Society. ................................ ............. 48 1 7 Schematic illustration of spin coupling be haviors in A) paramagnetic, B) ferromagnetic, C) antiferromagnetic, and D) ferromagnetic materials. ............................. 48 1 8 Schematic illustration of domains (left) and a domain wall (right). ................................ .. 48 1 9 Schematic illustration of a hysteresis curve for a typical ferromagnet. M s is saturated magnetization, M r is remnant magnetization and H c is coercive field. .............................. 49 1 10 The [Mn III 8 Mn IV 4 ( 3 O 2 ) 12 ] 16+ core (a) and the full structure of Mn 12 Ac complex (b). Color code: Mn IV green; Mn III blue; O red; C gray. H atoms have been omitted for clarity. ................................ ................................ ................................ ................................ 49 1 11 Plot of the orientations of the m s vectors along the z axis (left) and the double well potential showing the energy versus the m s sublevel for a complex with an S = 10 ground state spin, experiencing zero field splitting, z 2 (right). Reprinted with permission from reference 110. Copyright 2009 Royal Society of Chemistry. ................. 50 1 12 In phase ( M , plot as M T ) (top) and out of phase (as M ) (bottom) ac susceptibility signals for a dried, mic rocrystalline sample of [Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ] at the indicated oscillation frequencies. Reprinted with permission from reference 110. Copyright 2009 Royal Society of Chemistry. ................................ ................................ .... 50 1 13 Cole Col e plot of vs. for a sample of [Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ]. The dashed line is the fit of the data to a single relaxation process. The solid line is the fit to a distribution of single relaxation processes. Reprinted with permission from reference 110. Copyright 20 09 Royal Society of Chemistry. ................................ ............................ 51

PAGE 13

13 1 14 Magnetization hysteresis loops for [Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ] complex in the 1.3 3.6 K temperature range at a 4 mT/s field sweep rate. M is normalized to its s aturation value, M s . Reprinted with permission from reference 110. Copyright 2009 Royal Society of Chemistry. ................................ ................................ ................................ ......... 51 1 15 Schematic representation of the change in energy of m s sublevels as the magne tic field is swept from zero to a non zero value. Resonant magnetization tunneling occurs when the m s sublevels are aligned between the two halves of the diagram. .......... 52 1 16 Magnetization hysteresis loops of [Mn 4 ] 2 versus applied magnetic field at (A) different temperatures and (B) different field sweep rates. Reprinted with permission from reference 117. Copyright 2002 Nature Publishing Group. ................................ ....... 52 1 17 Energy diagrams showing the spin state energies of [Mn 4 ] 2 complex versus applied exchange coupled dimer of two spin S = 9/2 Mn 4 units. b) Enlargement of (a), showing 1.2 T to + 1.2 T, as indicated by green arrows. Dotted lines, labeled 1 to 5, indicate the strongest tunnel resonances. Reprinted with permission from reference 117. Copyright 200 2 Nature Publishing Group. ................................ ................................ ........ 53 2 1 Structures of the cation of complex 3 (top), mpkoH (bottom left) and pdpdH 2 (bottom right). Color code: Mn III green; N blue; O red; C gray. ................................ ....... 89 2 2 Complete molecular structure (top) and a stereopair (middle) of complex 2 1 with H atoms omitted for clarity, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits, and showing the Mn 3 planes and Jahn Teller axes (green bonds). Color code: Mn III green; Mn II sky blue; O red; N blue; C grey. ....... 90 2 4 (Top) Hydrogen bonds (black dotted bonds) connecting [Mn 3 ] 2 molecules of comp lex 2 1 [Mn 3 ] 2 molecules in complex 2 2 . ................................ ................................ ..................... 92 2 5 Structure of the cation (top) and a stereopair (middle) of complex 2 3 with H atoms omitt ed for clarity, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits, and showing the Mn 3 planes and Jahn Teller axes (green bonds). Color code: Mn III green; O red; N blue; C grey. ................................ .................. 93 2 6 Structure of the cation (top) and a stereopair (middle) of complex 2 4 with H atoms omitted for clarity, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits, and showing the Mn 3 planes and Jahn Teller ax es (green bonds). Color code: Mn III green; O red; N blue; C grey. ................................ .................. 94 2 7 Packing diagram of complex 2 4 viewing from c axis (top) and b axis (bottom). ............ 95 2 8 Encapsulation of two CH 2 Cl 2 molecules inside the Mn 12 cation of complex 2 3 . H atoms have been omitted for clarity. ................................ ................................ .................. 96

PAGE 14

14 2 9 Distances (Ã…) between Mn 3 units in complex 2 3 (top) comparing to those of complex 3 (bottom). pdpd 2 groups in complex 2 3 have been omitted for clarity. .......... 97 2 10 M T vs. T for complexes 2 1 (red) and 2 2 (blue). The solid lines are the fits to the data; see Table 2 5 for the fit parameters. ................................ ................................ .......... 98 2 11 M T vs. T for complexes 2 3 (pink) and 2 4 (green). The solid lines are the fits to the data; see Table 2 5 for the fit parameters. ................................ ................................ .......... 98 2 12 Two dimensional contour plot of the fitting errors vs. J and J' for complex 2 1 . ............. 99 2 13 Two dimensional contour plot of the fitti ng errors vs. J and J' for complex 2 2 . ............. 99 2 14 Two dimensional contour plot of the fitting errors vs. J and J' for complex 2 3 . ........... 100 2 15 Two dimensional contour plot of the fitting errors vs. J and J' for complex 2 4 . ........... 100 2 16 Energy ladder plot for complex 2 1 . Ground state is S T = 3/2. The first excited state S = 5/2 lies 49.7 cm 1 above the ground state. ................................ ................................ .... 101 2 17 Energy ladder plot for complex 2 2 . Ground state is S T = 2. The first excited state S = 3 lies 29.6 cm 1 above the ground state. ................................ ................................ ........... 101 2 18 Energy ladder plot for complex 2 3 . Ground state is S T = 6. The first excited state S = 5 lies 79.5 cm 1 above the ground state. ................................ ................................ ........... 102 2 19 Energy ladder plot fo r complex 2 4 . Ground state is S T = 6. The first excited state S = 5 lies 79.5 cm 1 above the ground state. ................................ ................................ ........... 102 2 20 M / N B (per Mn 3 ) vs. H / T for complex 2 1 . The solid lines are the fits to the data . See Table 2 10 for the fit parameters. ................................ ................................ .................... 103 2 21 M / N B (per Mn 3 ) vs. H / T for complex 2 2 . The solid lines are the fits to the data. See Table 2 10 for the fit parameters. ................................ ................................ .................... 103 2 22 M / N B (per Mn 3 ) vs. H / T for complex 2 3 . The solid lines are the fits to the data. See Table 2 10 for the fit parameters. ................................ ................................ .................... 104 2 23 M / N B (per Mn 3 ) v s. H / T for complex 2 4 . The solid lines are the fits to the data. See Table 2 10 for the fit parameters. ................................ ................................ .................... 104 2 24 Two dimensional contour plot of the errors vs. D and g for the fit of 2 1 . ..................... 105 2 25 Two dimensional contour plot of the errors vs. D and g for the fit of 2 2 . ..................... 105 2 26 Two dimensional contour plot of the errors vs. D and g for the fit of 2 3 . ..................... 106 2 27 Two dimensional contour plot of the errors vs. D and g for the fit of 2 4 . ..................... 106

PAGE 15

15 2 28 Plot of in ph ase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature for a micro crystalline sample of complex 2 1 in a 3.5G field oscillating at the indicated frequencies. ................................ ................................ ... 107 2 29 Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature for a micro crystalline sample of complex 2 2 in a 3.5G field oscillating at the indicated frequencies. ................................ ................................ ... 108 2 30 Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature for a micro crystalline sample of complex 2 3 in a 3.5G field oscillating at the indicated frequ encies. ................................ ................................ ... 109 2 31 Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature for a micro crystalline sample of complex 2 4 in a 3.5G field os cillating at the indicated frequencies. ................................ ................................ ... 110 2 32 Plot of ln( / ) vs. 1/ T for complex 2 3 . The lines are the linear regression fits to the data at the indicated frequencies. ................................ ................................ ............... 111 2 33 Magnetization decay measurements of complex 2 3 at zero applied field at several temperatures. ................................ ................................ ................................ .................... 111 2 34 Relaxation time ( ) vs. 1/ T using DC ma gnetization decay data. The dashed line is the fit of the thermally activated region to the Arrhenius equation. See the text for the fit parameters. ................................ ................................ ................................ .................. 112 2 35 Magnetization vs. dc field hysteresis loops for a single crystal of 2 3 ·xCH 2 Cl 2 at the indicated temperatures (top) and field scan rates at 0.04 K (bottom). M is normalized to its saturation value, M S . ................................ ................................ ............................... 113 2 36 (Top) Simulation of t he plot of spin state energies vs. applied magnetic field for a tetramer of four Mn 3 SMMs, each with S = 6. Red = spin states involving only the M s = ±6 states of the four Mn 3 sub units; Blue = spin states involving both M s = ±6 states and (only) one M s = ±5 state; Black = other states. (Bottom) Spin states involving only the M s = ±6 states. * = multiple spin flips (2, 3, or 4) at the same time. See the text for the spin states involved in the three QTM steps, which are at green arrows. ................................ ................................ ................................ .............................. 114 2 37 Temperature dependence of HF EPR spectra of complex 2 3 ·xCH 2 Cl 2 (experimental (a) and simulated (b)), recorded in field derivative mode at 217.6 GHz in the temperature range 2.5 20 K collected on a micro crystalline sample restrained in KBr. The features in (a) are labeled according to the scheme described in the main text. The top inset in (a) depicts a typical derivative mode powder spectrum for a biaxial system, illustrating the lineshapes expected for the x , y , and z components of the spectrum. ................................ ................................ ................................ ................ 11 5

PAGE 16

16 2 38 Easy axis ( z axis) frequency dependent EPR data for complex 2 3 ·xCH 2 Cl 2 . The solid lines are a simulation of the data employing the paramete rs: S = 6, D = 0.33 cm 1 , | E | = 0.03 cm 1 , = 8 × 10 5 cm 1 , and g x = g y = g z = 2.00. ................................ . 116 3 1 Structures of mpkoH (left), dedH 2 (middle) and pdpdH 2 (right). ................................ .... 144 3 2 Possible reactions that can occur in the formation of complex 3 1 . ................................ 144 3 3 Possible reactions that can occur in the formation of complex 3 2 . ................................ 145 3 4 Formation of complexes 3 3 and 3 4 . ................................ ................................ .............. 145 3 5 Structure of the cation (top) and a stereopair (middle) of complex 3 1 with H atoms omitted for clarity, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits, and showing the Mn 3 planes and Jahn Teller axes (green bonds). Color code: Mn III green; Mn II sky blue; O red; N blue; C grey. ......................... 146 3 6 Structure of the cation (top) and a stereopair (middle) of complex 3 2 with H atoms omitted for clarity, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits, and showing the M n 3 planes and Jahn Teller axes (green bonds). Color code: Mn III green; Mn II sky blue; O red; N blue; C grey; Cl pink. ........... 147 3 7 The structures of the cations of 3 1 (top) and 3 2 (bottom) showing t heir guests, pyridinium and pyridine, in space filling style, respectively. For complex 3 1 , due to crystallographic disorders in the aromatic ring, the N atom was not determined. For complex 3 2 , the pyridine guest is disordered in two positions. ................................ ...... 148 3 8 Structure of the cation (top) and a stereopair (middle) of complex 3 3 with H atoms omitted for clarity, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits, and sho wing the Mn 3 planes and Jahn Teller axes (green bonds). Color code: Mn III green; O red; N blue; C grey. ................................ ................. 149 3 9 Structure of the cation (top) and a stereopair (middle) of complex 3 4 with H atoms omitted for clarity, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits, and showing the Mn 3 planes and Jahn Teller axes (green bonds). Color code: Mn III green; O red; N blue; C grey. ................................ ................. 150 3 10 The structures of the cations of 3 3 (top) and 3 4 (bottom) showing their guests, CH 2 Cl 2 and EtOH, in space filling style, respectively. The guest molecules are disordered about the C 3 rotation axes. ................................ ................................ ............. 151 3 11 Determination of the center point (C) of the tetrahedron O14O1O1'O1''. ....................... 152 3 12 Racemic mixtures of complexes 3 3 (top) and 3 4 (bottom). ................................ .......... 153 3 13 Space filling presentation of the twelve O atoms of the [O 3 ] 4 tetrahedron from six ded 2 groups that form a cage like shell of O atoms about the central space in complex 3 4 . ................................ ................................ ................................ .................... 154

PAGE 17

17 3 14 M T vs. T for complexes 3 1 , 3 2 , 3 3 , and 3 4 . ................................ ............................... 154 3 15 Plot of the difference ( M T ) between the M T value of 3 4 and four times of M T value of 4 . ................................ ................................ ................................ ......................... 155 3 16 M / N B (per Mn 3 ) vs. H / T for complex 3 3 . The solid lines are the fits to the data. See the text for the fit parameters. ................................ ................................ .......................... 155 3 17 M / N B (per Mn 3 ) vs. H / T for complex 3 4 . The solid lines are the fits to the data. See the text for the fit parameters. ................................ ................................ .......................... 156 3 18 Two dimensional contour plot of the errors vs. D and g for t he fit of 3 3 . ..................... 156 3 19 Two dimensional contour plot of the errors vs. D and g for the fit of 3 4 . ..................... 157 3 20 Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature for a micro crystalline sample of complex 3 1 in a 3.5G field oscillating at the indicated frequencies. ................................ ................................ ... 158 3 21 Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature for a micro crystalline sample of complex 3 2 in a 3.5G field oscillating at the indicated frequencies. ................................ ................................ ... 159 3 22 Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature for a micro crystalline sample of complex 3 3 in a 3.5G field oscillating at the indicated frequencies. ................................ ................................ ... 160 3 23 Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature for a micro crystalline sample of complex 3 4 in a 3.5G field oscillatin g at the indicated frequencies. ................................ ................................ ... 161 3 24 Magnetization vs. dc field hysteresis loops for a single crystal of 3 1 ·xMeCN at the indicated temperatures (top) and field scan rates at 0.04 K (bottom). M is normalized to its saturation value, M S ................................ ................................ ................................ 162 3 25 Magnetization vs. dc field hysteresis loops for a single crystal of 3 4 ·xCH 2 Cl 2 at the indicated temperatures (top) and field scan rates at 0 .04 K (bottom). M is normalized to its saturation value, M S ................................ ................................ ................................ 163 3 26 Simulation of the plot of spin state energies vs. applied magnetic field for a tetramer of four Mn 3 SMMs, each with S = 6, spi n states involving only the M s = ±6 states. * = multiple spin flips (2, 3, or 4) at the same time. See the text for the spin states involved in the QTM steps, which are at green arrows. ................................ .................. 164 4 1 S tructures of mpkoH (top left), dedH 2 (top right), dpdH 2 (bottom left), and pdpdH 2 (bottom right). ................................ ................................ ................................ .................. 182

PAGE 18

18 4 2 (Top) Structure of the cation of complex 4 1 in side view (left) and top view (right), (m iddle) a stereopair, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits and showing the Mn 3 planes and Jahn Teller axes (green bonds). H atoms have been omitted for clarity. Color code: Mn III green; O red; N blue; C grey. ................................ ................................ ................................ ..................... 183 4 3 Packing of cations 4 1 within one unit cell in side view (top) and top view (bottom). ... 184 4 4 UV Vis spectra of complex 4 1 at di fferent concentrations and complex 3 in acetonitrile. ................................ ................................ ................................ ....................... 185 4 5 Plot of absorbance of the 360 nm band vs. concentration of complex 4 1 showing linear relationship. ................................ ................................ ................................ ............ 185 4 6 M T vs. T for complex 4 1 . ................................ ................................ ............................... 186 4 7 Ferromagnetic coupling by spin polarization mechanism through dpd 2 groups. The spin vectors were shown for one dpd 2 group. ................................ ................................ . 186 4 8 M T vs. T for complex 4 1 . The solid line is the fit to the model of two non interacting equilateral triangles. ................................ ................................ ....................... 187 4 9 M T v s. T for complex 4 1 . The solid line is the fit to the model of two non interacting isoceles triangles. ................................ ................................ ........................... 187 4 10 Root mean square errors vs. J and for the fit to Van Vlek equation. .......................... 188 4 11 Energy ladder plot for complex 4 1 . Ground state is S T = 6. The first excited state S = 5 lies 79.5 cm 1 above the ground state. ................................ ................................ ........... 188 4 12 M / N B (per Mn 3 ) vs. H / T for complex 4 1 . The solid lines are the fits to the data. See the text for the fit parameters. ................................ ................................ .......................... 189 4 13 Representations of the error surface for the D vs. g fit for comple x 4 1 . Top: two dimensional contour plot, bottom: three dimensional mesh plot. ................................ .... 190 4 14 Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus tempe rature for a micro crystalline sample of complex 4 1 in a 3.5G field oscillating at the indicated frequencies. ................................ ................................ ... 191 4 15 Simulations of the 5 Hz in phase AC magnetic susceptibility data by the prog ram MAGPACK. The g value was held at 1.91. ................................ ................................ ..... 192 4 16 Comparison between experimental (A) and simulated (B) EPR spectra (B || z) for the [Mn 3 ] 2 dimer at 148.67 GHz; the simulation assumed D = 0.22 cm 1 , = 7 × 10 5 cm 1 , g z = 2, and J z = J xy = J = 0.025 cm 1 . The resonances have been labeled to show the assigned transitions between states that are shown in schematic energy

PAGE 19

19 level diagram (C). The (A) inset is a simulated [Mn 3 ] monomer spectrum. The (B) inset shows the evolution of the spectrum as J increases. ................................ ............... 193 4 17 (A) Comparison between a spectrum of complex 4 1 in solution and a spectrum of a single crystal of 4 1 ·xCH 2 Cl 2 . (B) Frequency dependence of the main EPR peak positions (filled symbols: single crystal; unfilled symbols: solution) obtained at many microwave frequencies. The solid lines in the plot (B) are simulated energy transitions. ................................ ................................ ................................ ........................ 194 5 1 Several di , tri, and tetracaboxylic acids employed in metal organic framework chemistry. ................................ ................................ ................................ ......................... 216 5 2 The structures of trimesic acid (left), dime truxillic acid (right). ................................ ................................ ................................ ....................... 217 5 3 (Top) Structure of the cation of complex 5 1 , (middle) a stereopair, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits and showing the Mn 3 planes and Jahn Teller axes (green bonds). H atoms have been omitted for clarity. Color code: Mn III green; O red; N blue; C grey. ................................ .................. 218 5 4 (Top) Structure of the cation of complex 5 2 in side view and top view, (middle) a stereopair, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits and showing the Mn 3 planes and Jahn Teller axes (green bonds). H atoms have been omitted for cl arity. Color code: Mn III green; O red; N blue; C grey. ... 219 5 5 (Top) Structure of complex 5 3 , (middle) a stereopair, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subu nits and showing the Mn 3 planes and Jahn Teller axes (green bonds). H atoms have been omitted for clarity. Color code: Mn III green; O red; N blue; C grey. ................................ ................................ ........ 220 5 6 Packing of cations of 5 2 in top view showing the different orientations between two sets of Mn 6 cations. ................................ ................................ ................................ .......... 221 5 7 Packing of molecules of 5 3 showing parallel Mn 3 planes. ................................ ............. 222 5 8 M T vs. T for complexes 5 1 , 5 2 , and 5 3 . The solid lines are the fits to the data. See table 5 8 for the fit parameters. ................................ ................................ ........................ 223 5 9 Two dimensional contour plot of the fitting errors vs. J and J' for complex 5 1 . ........... 224 5 10 Two dimensional contour plot of the fitting errors vs. J and J' for complex 5 2 . ........... 224 5 11 Two dimensional co ntour plot of the fitting errors vs. J and J' for complex 5 3 . ........... 225 5 12 Energy ladder plot for complex 5 1 . Ground state is S T = 6. The first excited state S = 5 lies 67.5 cm 1 above the ground stat e. ................................ ................................ ........... 225

PAGE 20

20 5 13 Energy ladder plot for complex 5 2 . Ground state is S T = 6. The first excited state S = 5 lies 79.5 cm 1 above the ground state. ................................ ................................ ........... 226 5 14 Energy ladder plot for complex 5 3 . Ground state is S T = 6. The first excited state S = 5 lies 79.5 cm 1 above the ground state. ................................ ................................ ........... 226 5 15 M / N B (per Mn 3 ) vs. H / T for complex 5 1 . The so lid lines are the fits to the data. See the text for the fit parameters. ................................ ................................ .......................... 227 5 16 M / N B (per Mn 3 ) vs. H / T for complex 5 2 . The solid lines are the fits to the data. See the text for the fit param eters. ................................ ................................ .......................... 227 5 17 M / N B (per Mn 3 ) vs. H / T for complex 5 3 . The solid lines are the fits to the data. See the text for the fit parameters. ................................ ................................ .......................... 228 5 18 Two dimensional contour plot of the error surface for the D vs. g fit for complex 5 1 . . 228 5 19 Two dimensional contour plot of the error surface for the D vs. g fit for complex 5 2 . . 229 5 20 Two dimensional contour plot of the error surface for the D vs. g fit for complex 5 3 . . 229 5 21 Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature for a micro crystalline sample of complex 5 1 in a 3.5G field oscillating at the indicated frequencies. ................................ ................................ ... 230 5 22 Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature for a micro crystalline sample of complex 5 2 in a 3.5G field oscillating at the indicated frequencies. ................................ ................................ ... 231 5 23 Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature for a micro crystalline sample of complex 5 3 in a 3.5G field oscillating at the indicated frequ encies. ................................ ................................ ... 232 5 24 Magnetization vs. dc field hysteresis loops for a single crystal of 5 3 ·xMe 2 CO·yMeCN at the indicated temperatures (top) and field scan rates at 0.03 K (bottom). M is normalized to its saturation value, M S . ................................ .................... 233

PAGE 21

21 LIST OF ABBREVIATIONS AC alternating current AF antiferromagnetic BVS bond valence sum DC direct current dedH 2 1,2 di(pyridin 2 yl)ethane 1,2 dione dioxime dmaH 2 dimethylmalonic acid dpdH 2 1, 3 di(pyridin 2 yl) propane 1, 3 dione dioxime FM ferromagnetic HF EPR high frequency electron paramagnetic resonance IR infrared JT Jahn Teller mpkoH methyl 2 pyridyl ketone oxime pdpdH 2 3 phenyl 1,5 di(pyridin 2 yl)pentane 1,5 dione dioxime QPI quantum p hase interference QTM quantum tunneling of magnetization RM reduced magnetization SMM single molecule magnet SQUID superconducting quantum interference device TBA tetra n butyl ammonium TIP temperature independent paramagnetism tmaH 3 trimesic acid txaH 2 truxillic acid UV Vis u ltraviolet visible ZFS zero field splitting

PAGE 22

22 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 SUPRAMOLECULAR AGGREGATES OF SINGLE MOLECULE MAGNETS By Tu Ngoc Nguyen August 2 014 Chair: George Christou Major: Chemistry This dissertation focuses on the syntheses and magnetic studies of covalently bonded supramolecular aggregates of single molecul e magnets (SMMs) . Employing 3 phenyl 1, 5 di(pyridin 2 yl)pentane 1,5 dione dioxime (pdpdH 2 ) in the reaction with the non SMM [Mn 3 O(O 2 CMe ) 6 ( py ) 3 ] ( ClO 4 ) complex yield ed a novel rectangular [Mn 3 ] 4 supramolecule . Solid state dc and ac magnetic susceptibility m easurements revealed that each Mn 3 unit is still an SMM with ground state spin of S = 6. Magnetization vs. dc field sweeps on a single crystal gave hysteresis loops below 1 K that exhibited exchange biased quantum tunneling of magnetization (QTM) steps, co nfirming the rectangular [Mn 3 ] 4 complex to be a supramolecular aggregate of four weakly exchange coupled SMM units. The inter SMM c oupling through pdpd 2 groups was found to be very weak, J 0.02 K . The organic compound 1, 2 di(pyridi n 2 yl) ethane 1, 2 di one dioxime (d e dH 2 ) was synthesized and employed in a similar manner to pdpdH 2 . The reaction s with the non SMM [Mn 3 O(O 2 CR) 6 ( py ) 3 ] ( ClO 4 ) , R= M e, Et, complexes afford ed the corresponding tetrahedral [Mn 3 ] 4 supra molecules. These two molecules contain three and two Mn II ions , respectively . Similar reactions employing excess of iodine gave two new [Mn 3 ] 4 t etrahedrons with all Mn ions in the +3 oxidation state . T hese tetrahedral supra molecules encapsulated solvent molecules

PAGE 23

23 inside the [Mn 3 ] 4 frames. Magnetic st udies confirm ed that the [Mn 3 ] 4 tetrahedrons are supramolecular aggregate s of four exchange coupled SMM units. Simulation of the spin state energ ies vs . applied magnetic field suggest ed the inter SMM coupling between Mn 3 units of 0.06 K. The QTM steps a re broad due to the non coplanarity of the Mn 3 units. Employing the ligand 1, 3 di(pyridi n 2 yl)propane 1,3 dione dioxime (dpdH 2 ) afforded a dimeric [Mn 3 ] 2 SMM . UV Vis spectroscopy studies confirmed that the dimeric molecules are intact in an acetonitril e solution. Solid state dc and ac magnetic susceptibility measurements suggest ed that Mn 3 units have S = 6 ground state , and they weakly ferromagnetically interact with each other . H igh frequency electron paramagnetic resonance (HF EPR) studies on a single crystal or a solution of the [Mn 3 ] 2 SMM in acetonitrile/toluene (1 : 1 v/v) confirm ed the ferromagn etic inter SMM interaction with J +0. 0 7 K . T rimesic acid truxillic acid , and dimethylmalonic acid were employed to join [Mn 3 O(O 2 CMe) 3 (mpko) 3 ] ( ClO 4 ) SMMs by carboxylate substitution method . T oluene was added to the reaction mixture s and the acetic acid was removed as its toluene azeotrope by evaporation under vacuum . Trimesic acid gave a tetrahedral [Mn 3 ] 4 molecule while truxillic acid and dimethylmalonic acid result in two dimer ic [Mn 3 ] 2 complexes . Solid state dc and ac magnetic susceptibility measurements revealed that each Mn 3 unit in all three complexes still preserve s their SMM behavior with ground state spin of S = 6. Hysteresis studies on a single crystal of the dimeric complex that was fo truxillic acid displayed the exchange biased QTM steps . The inter SMM coupling is very weak, J 0.01 K . The dissertation demonstrates the f e asibility of covalently connecting multiple Mn 3 SMMs to give dis crete supramolecular aggregates of SMMs from the use of a variety of li gands and represents a step tow ard pr actical a pplication s o f SMMs.

PAGE 24

24 CHAPTER 1 GENERAL INTRODUCTION 1.1 Supramolecular Chemistry Supramolecular chemistry is the concept introduced by Jean Marie Lehn in 1978 , who later won the Nobel Prize in 1987 (shared with Donald J. Cram and Charles J. Pedersen) for his leading contribution in the area, as the chemistry of molecular assemblies and of the intermolecular bond 1 and the chemistry beyond the molecule . 2,3 The se defin ition s are somewhat mysterious and incomprehensible unless we look back at the history of supramolecular chemistry. T he first brick to build up the field was probably the discovery of cyclodextrins by Villiers and Hebd in 1891. 4,5 Soon after, n atural cyclodextrins exist ing in three were found to form many stable host guest complexes with hydrophobic molecules . 6 During the years 1893 1894 , two important core concepts of supramolecular chemistry were introduced : coordination chemistry of metal complexes by Alfred Werner , 7 and the lock and key model forming the basis of molecular recognition by Emil Fisher . 8 In 1906, Paul Ehrlich devised the concept of biological receptors . 9 As early as 1927, molecular association was well recognized and studied. 10 Even t Übermolekü le (supermolecule in German) had already appeared in the work of Karl Lothar Wolf in 1937 to describe high organization entities such as the gas phase dimer of acetic acid . 11 In 1939, the hydrogen bond Nature of the Chemical Bond: A Documentary History by Linus Pauling . Supramolecular organization has also been found in many biolog ical molecules . 12 The induced fit concept formulated by Daniel Koshland provide d a dynamic view of binding events to biomolecules in which phenomena such as cooperativity were discovered. In the late 1960 s and early 1970 s , supramolecular chemi st ry actually started as a new distinct field with vast studies of selective binding of alkali metal cations by either natural 13 20 or synthetic

PAGE 25

25 macrocyclic and macrop olycyclic ligands . Pedersen was the pioneer for the crown ethers and was followed by Lehn and Cram for c ryptands and spherands (Figure 1 1 ) . 21 25 Lehn and Cram have subsequently advanced the field with sophisticated organic compounds that contain cavities suita ble for binding of low molecular weight molecules with different types of geometry. Some host molecules that form particularly strong complexes of extremely high selectivity were engineered with the help of molecular mechanics calculation s . Cram coined the guest chemistry 26 , which seems to have a broa der concept. The definition of supramolecular chemistry by Lehn brought in a new philosophy where in molecules and systems that are not formed by traditional bonds, i . e . covalent bonds. The new point of view replaced the old paradigm in chemistry that molec ules possess properties by themselves, while the ir interactions with the environment are weak and approximately negligible. Chemists were suddenly given a new land of chemistry focusing on organized entities with non covalent forces, molecular recognition, self assembly , and then supramolecular reactivity and catalysis, carrier mediated transport, supramolecular devices, etc. For nearly 40 years of supramolecular research , a variety of molecular materials that exhibit unusual sensing, magnetic , catalytic , a nd optical properties has been synthesized . 27 Supr amolecular bonds are the mean s for molecules to recognize each other in a c omplexes are defined as two or more compounds bound to one another in a definable structural relationship by f orces such as hydrogen bonding, ion pairing, metal ion to acid to base attractions, van der Waals attractions . 28 These non covalent inte ract ions are worth briefly survey ing here. I on ion attractions are the strongest non covalent bonds , which dominate in ionic solids. Depending on the charge s of the ions and the distance b etween them, the strength of an

PAGE 26

26 ionic bond varies in the range of 100 350 kJ/mol and is comparable to those of covalent bonds. In non polar liquids having low dielectric constants, ions are not efficiently separated, thus i on pairs such as R 4 N + Br have b een known to exist and many of them were utilized as phase transfer catalysts, which facilitate the migration of a reactant from one phase into another phase. 29 An i onic solid such as sodium chloride , by the measure of many chemists, is not a supramolecular compound but it does show the feature that one Na + cation attracts six Cl anions in the lattice to maximize non covalent ionic interactions. A more supramolecular example is the solid formed from the binary assembly of molecular clusters such as [Co 6 Se 8 (PEt 3 ) 6 ][C 60 ] 2 in which electrons are transferred fr om the Co cluster to the fullerene molecules. 30 I ons that have the same sign of charge will repel each other but the y can be attract ed by the environment . Ion dipole attractions are the interaction s between an ion and neutral molecules that have a dipole , e.g. a sodium cation surrounded by six water molecules . This interaction is weaker than ionic bonds, the strength is in the range of 50 200 kJ/mol. The attraction become s stronger as the ionic charge increases and/or the magnitude of the dipole increases. A supramolecular illustration is already shown in the crown ether complex in Figure 1 1 although chelation and mac rocyclic preorganiz ation are the reasons for the stability of the complex. When there are no ions, the system can be stabilized by d ipole dipole attractions which arise w hen two dipoles come close toge ther. The interactive forces are between a positive pol e and a negative pole on two different polar molecules. This interaction is weak ( 5 50 kJ/mol ) but ver y common in all polar compounds. Dipole dipole interaction s are attributed to the fact that the highly polar iodine monochloride is a solid while Br 2 ha ving approximately the same molecular weight is a liquid. T he crystal structure of iodine monochloride shows chains of ICl molecules arrang ed in zig zag pattern . 31 Polar m olecules containing hydrogen might have another type of bonding. Hydrogen -

PAGE 27

27 bond is o ne of the most important non covalent bonds occurring between a H atom covalently bou nd to a highly electronegative atom such as fluorine (F), oxygen (O) , or nitrogen (N) and other electronegative atom s . The nature of h ydrogen bond s is still the dipole dipole interaction, thus they are weak bond s in general (5 50 kJ/mol) . However, a very strong hydrogen bond (161.5 kJ/mol) was found in the ion HF 2 . 32 Hydrogen bond s are responsible for the double helix st ructure of DNA (Figure 1 2), which is probably the best example of supramolecular assemblies in nature. H ydrogen bonding i s accountable for the high boiling point of water (100 °C), about 160 °C higher than the heavier H 2 S. A great number of supramolecules and assemblies based on hydrogen bonds have been reported. Two special non covalent interaction s are c ation anion attraction s . They are less known but there are several examples recognized . K + and ammonium ion s are noted to have strong interac tion s with benzene in the gas phase ( 80 kJ/mol). 33,34 In biology, it was found that 26% of all t ryptophan s are involved in a cation interaction. 35 The existence of a nion attraction seems to be intuitively questionable due to the repulsion between both species with negative electron density. However, attraction s between anion s such as Cl , Br or I and electron deficient aromatic center s were discovered recently. Such co mplexes having different color s depending on the nature of anion encourage their potential applications for anion sensing receptors. 36 Without ions, a romatic molecules are known to attract themselves. interaction s or stacking s occur when aromatic rings organize in proximity as in the benzene dimer whose binding energy has been calculated of 12 kJ/mol in the gas phase . 37 Although there have been debates on the nature of stacking , 38 t hese interactions are commonly found in crystal structures of many or ganic and inorganic compounds ; the most notable example is graphite, and also prevalent in protein crystal structures. 39 In supramolecular chemistry, lot s of efforts to design host molecules to encapsulate

PAGE 28

28 fullerene C 60 based on interactions have been reported. 40 When no ne of the interactions me ntio ned above exists , are the molecules free to move? This condition actually happen s only in ideal gases. In reality, there are still v an der Waal s (VDW) interaction s , named after Johannes Diderik van der Waals . They arise when the electron cloud of an atom is polarized by the attraction of an adjacent nucleus. This is the weakest non covalent interaction , usually about 2 4 kJ/mol per atom pair. Two types of VDW interaction are Debye force and London dispersion force. Debye force is the attraction between a permanent multipole and an induced multipole while London d ispersion force results from the interactions between spontaneous multipoles when there are no permanent multipole moments . These forces are seen in non polar molecules . It is interesting that gec kos can hang themselves on walls and ceilings because of VD W interaction . 41 A supramolecular example is the stable cage type clathrate that displays the encapsulation of a toluene molecule within the void of p tert butylcalix[4]arene molecule . 42 Finally, t he most important non covalent bonds for inorganic chemists are c oordinate bond s . They are the attraction s between a metal atom or ion and the coordinated ligands. Interestingly, they have the essence of ion ion, ion dipole, dipole dipole , or cation interaction s depending on the nature of metal and l igand . S upramolecular chemistry in this respect is a generalization of coordination chemistry. 43 The c helate effect is one of the most important factors contributing to the stability of metal based supramolcules. The stability of chelate complexes relate s to both thermodynamic and kinetic effects. In terms of thermodynamics , a chelating ligand , i . e . bi , tri , or polydentate ligands, react ing with a metal center will lead to an increase d number of free species, thus favoring the free energy of the reactio n by increasing entropy. The incre ase of the overall rate of reactions involving chelat ing ligands compared to monodentate ligands is due to the higher effective concentration of the c helate after the first coordination. For example, the

PAGE 29

29 reaction with the second donor atom of the chelat ing ligand that is tethered on the complex is much faster than with a second monodentate ligand. Coordination compounds are vital to the existence of living organisms in nature. In human s , hemoglobin containing iron porphyri n units are responsible for carrying oxygen while vitamin B 12 is essential for the brain and nervous system. T he f o llo wing discussion will focus on c oordination driven self assembly , which is currently one of the fastest growing bra nch es of supramolecula r chemistry. A comprehensive overview of supramolecu lar chemistry can be seen in Chemistry edited by Jerry L. Atwood and Jonathan W. Steed. D ue to the predictive nature of the metal ligand coordination sphere, greater contr ol is available to rationally design supramolecules and assemblies. Self correction based on t he kinetic reversibility between complementary building blocks, reaction intermediates, and self assembled architectures is the main mechanism leading to thermody namically stable products . 27 A wide range of diffe rent architectures have been reported such as infinite helicates, 44,45 grids, 46 catenates, rotaxanes, cylinders, knots, 47 53 and related species, 54 57 and a variety of two dimensional (2D) molecular ensembles, e.g. molecular dimers, triangles, rectangles, and three dimensional (3D) polyhedra and capsules. 27,58 63 Some research groups have appeared to utilize these 2D and 3D supra mole c ules in applications such as supramolecular catalysis 64,65 or nanomaterial s. 66 68 These supramolecular assemblies represent nanoscale scaffolds synthesized by bottom strategy and possess potential utili ty in mol ecular encapsulation and recognition, 69 cavity controlled reactions, 70 or drug delivery. 71 Various synthetic strategi es to construct discrete 2D and 3D supramolecules have been developed and can be summarized as follows. The d irectional bonding approach is based on the

PAGE 30

30 idea of retrosynthesis in which the supramolecule is divided into two complementary building blocks: th e metal containing species as the acceptors , and linkers as the donors. These are the precursors whose coded structur al information defines the geometry of the targets. The precursor units in t his approach are require d to be structural ly rigid . Considering the dynamic nature of molecules in reactions, only very small change s in binding angles might be allowed but should not change the general shapes of the building blocks. It is also necessary to conduct reaction with appropriate st o ichiometry. The geometr y of the two dimension al target s is predicted by the combination of the acceptor s and the donor s as shown in Figure 1 3 (top) . The three dimension al targets are formed when one of the building blocks h as three or more binding sites ( Figure 1 3 , bottom) . 72 For ins tance , a cube is built by the combination of eight 90 o tritopic subunits with twelve linear ditopic subunits 73 while a more complicated cuboctahedron can be assessed by consolidation of eight 12 0 o tritopic subunits and twelve 109 o ditopic subunits. 74 This rational strategy allows one to construct the appropriate bui lding blocks to achieve a particular discrete assembly. The second strategy worthy to mention is the s ymmetry interaction approach . This approach has been developed by Raymond and co workers for the synthesis of high ly symmetric coordination clusters using multi branched chelating ligands and naked metal ions . 58,75 It is based on the recognition that natur al supramolecular assemblies are formed in symmetry driven processes , e.g. in ferritin , simultaneous satisfaction of two incommensurate 3 fold and 4 fold symmetry axes can only be achieved by formation of a cluster with octahedral symmetry . 56 The symmetry of the target interprets the s ymmetry of the metal component and the ligand. For example, a regular tetrahedron has four C 3 axes connecting vertices with the centers of the opposite faces , and three C 2 axes connecting the midpoints of opposite edges . If t he metal ions take up the vertices, this requires the ligand to have C 2 symmetry on the edge or C 3 symmetry on

PAGE 31

31 the face of the tetrahedron (Figure 1 4). This method has been proved to be particular ly suited f o r build ing high ly symmetr ic simple supramolecul es like tetrahedrons or cubes. Another successful strategy is the p aneling approach . This approach was initiated by Fujita and co workers and was proved to be efficient for the synthes e s of various delicate architectures such as square pyramidal cones, 76 trigonal prisms , 7 7 or hexagonal prisms , 78 etc . The general idea is simplifying polyhedr al structures in to individual panels and ligand s are designed to match the shape of the panels. Figure 1 5 shows the layout of panels necessary to construct a trigonal prism and a hexagonal prism . It is noted that the corner unit s must be blocked in certain directions to force the formation of the desired product. The supramolecules obtained by this strategy often contain large interior cavities and are ideal for exploring guest exchange, 69 cavity driven catalysis , 79 or eve n studying secondary structure of oligopeptides. 80 The last strategy is t h e w eak link approach for which t he ground work was laid by Mirkin and coworkers who employed hemilabile ligands and transition metals to construct supramolecular assembl ies. 81 For this method, show n in Figure 1 6, it is critical that the bidentate ligand chelates the metal center with one coordinate bond weaker than the other. Th e condensed intermediate product can subsequently be transformed to the open product by the reaction with sma ll molecules or ions that can substitute the weak donor atoms . This strategy opens a strikingly distinct access to a great number of 2D dimetallic macrocylic rings with applications as on off catalysts. 82,83 The system can also be extend ed to tetrametallic macrocylic rings by a proper design of ligands. 84 The explosive development of coordination driven self assembly in particular and supramolecular chemistry in general has been galvanized by possible applications in many disciplines. Leonardo da Vinci once said Where Nature finishes producing its own species, man begins, using natural things and with the help of this nature, to create an infinity of species .

PAGE 32

32 The huge variety of supramolecular assem blies and supramolecules that ha ve been reported in the literature can surprise anyone and it reflects the fact that s upramolecular chemistry is now at the interface of many subjects : nanochemistry, material science, chemical synthesis, polymer science, et c. I t is also the core of growing research areas such as energy conversion or magnetism . The following section will be the introduction of single molecule magnetism and the contribution of supramolecular chemistry in the development of single molecule magn ets . 1.2 Single molecule Magnets Molecular magnetism and supramolecular chemistry developed as disjoint ed disc iplines but now adays each of them is the thrust of the other. The evolution of magnetism in general and molecular magnetism in particular can be traced back to ancient times when people were aware of magnetism in a magical stone that attracts elemental iron. T he stone is known to be composed of a mineral called magnetite whose chemical formula is Fe 3 O 4 . The name magnetite is name d after Magnes , a Greek shepherd, who discovered the mineral on mountain Ida in Greece. N ot all pieces of magnetite ore attract iron; only those which attract iron are called magnets. Thales of Miletus (634 546 B.C.) was said to attribute a soul to a magnet in order to explain its attractive nature . A magnet is also known to be called the lodestone or the leading stone because when it is suspended freely, it always orients itself in one direction , which is helpful in navigation. Throughout the ancient times and the Midd le Ages , magnet s were a source of many strange and curious properties. Magnet s w ere suppose d to give comfort and grace and believed to cure dropsy, hemorrhages, toothaches and many other diseases. In 1269 , Petrus Peregrinus de in which he described his own observations and experiments on magnet s . This can be considered the first scientific treatise on magnet ism in Europe. More than three centuries later, in 1600, a physician , physicist , and natural philosopher named William Gilb ert published his book De Magnete Magneticisque Corporibus et de Magno

PAGE 33

33 Magnete Tellure Physiologia Nova ( On the Magnet and Magnetic Bodies, and on the Great Magnet the Earth) and proposed a theory that likened the earth to a giant spherical magnet . 85 It is Isaac of gravity , which were demonstrated in book Principia Mathematica ( Mathematical Principles of Natural Philosophy ) published in 1687 . At on a magnetic needle, and especially Farad ies of magnetic properties of matter , set up fundamental concepts connected with magnetism. Magnetic field is one of the most original concepts in magnetism. In our everyday life, we can see magnetic field s everywhere. A door stop uses the magnetic force to keep the door open. W hen a magnet ic sticker is put on the fridge door, a magnetic field is seen as an invisible attractive force that holds it there. Electric motors work based on the interaction of magnetic force on an electric curr ent. The e arth possesses a weak magnetic field , which attracts and causes the floating compass to point in only one direction . But magnet ic field s do not have an effect on everything. A plastic chair , for example, does not respond to the magnetic field pro duced by a permanent magnet. In general, th e re are five main c ategories of magnetic materials: diamagnets, paramagnets , ferromagnets , ferrimagnets, and antiferromagnets . Mathematically, t he response of materials to a magnetic field is presented as magnetic susceptibility ( ). Plastic chairs are made from organic polymers whi ch contain no unpaired electron s ; they are diamagnetic material s . In diamagnetic materials, all electrons are paired and tend to be drawn towards the weakest part of the magnetic field. In other words, diamagnetic materials are repelled by magnetic field s although the repulsi ve force is very weak. Therefore , their value is small, negative (~ 10 5 ) , and independent of magnetic field. Organic radicals or salts of some of the first row tr ansition

PAGE 34

34 metals ha ve unpaired electron s and are paramagnetic material s . U npaired electrons in these materials cause the attract ion to the magnetic field , thus is positive . In paramagnets, the electron spins do not interact with each other and randomly or ient in the material (Figure 1 7 a). When a magnetic field is applied, electron spins will try to follow and align parallel to the field. However, the randomizing effect of spins by thermal energy is entropically favored and opposes the spin alignment. Cons equently, magnetic susceptibility of a paramagnet is positive, but still small (10 3 to 10 5 ). For paramagnets, varies inversely with temperature, = C/T, where C is the Curie constant . 86 Ferromagnetic, antiferromagnetic , and ferrimagnetic materials are specialized forms of paramagnet s. Apart f rom having unpaired electrons , these materials differ from paramagnets in that their electron spins of neighboring spin carriers interact with each alignm ent, magnetic susceptibility of the material can be increased or decreased. For weakly coupled spins, the susceptibility follows the Curie Weiss law, = C/ (T proportional to the coupling strength of adjacent spins . 87 When the spins are parallel , i.e. ferromagnetic coupling is positive , and when the spins are antiparallel , i.e. antiferromagen e t ic coupling is negative the coupling between a pair of spins , not a long range coupling. F in a series of pairs of spin can lead to long range ferromagnetic order (Figure 1 7 b). Examples of ferromagnetic m aterials are metallic iron, cobalt, nickel , and their alloy s . a ntiferromagnetic order when all spins oppose each other , which ultimately leads to zero magnetic moment (Figure 1 7 c). Common antiferrom agnetic materials are oxides of transition metal compounds such as nickel oxide (NiO), hematite, metals , and alloys such as chromium and iron manganese ( CrMn, FeMn). Ferrimagnetic order also results from pairwise antiferromagnetic

PAGE 35

35 coupling, but spins are n ot completely cancel ed out; therefore ferrimagnets have a net magnetic moment (Figure 1 7 d). This phenomenon occur s in magnetite (Fe 3 O 4 ), ferrites , and magnetic garnets . Theoretically, long range spin order in ferromagnets or ferrimagnets leads to spontane ous magnetization of these materials at zero magnetic field below a critical temperature, T c . However, in the absence of a magnetic field, zero net magnetization of ferromagnets is observed and can be explained as the magnetic moments align in randomly ori ented small domains (Figure 1 8). When a magnetic field is applied, the domain s whose magnetic moments are closest to the field direction start to grow at the expense of the other neighboring domains. The growth occurs by domain wall motion . 88 At a sufficient applied magnetic field, all domain walls are eliminated and that leaves a single domain with its magnetization orient ed in one dir ection, i.e. the magnetization is saturated . Removing the field allows spins to rotate back to their original direction and the sample initiates the formation of reverse magnetic domains. However, a complete reversal of domain walls back to their original position s requires an external magnetic field, coercive field, to overcome the energy barrier. Consequently, hysteresis is observed in the plot of magnetization vs . applied field and some magnetization remains even when the field is zero (Figure 1 9 ). This property of ferromagnetic materials is utilized in data storage of information. A multi domain material has a typical domain wall width around 1000Ã…. Qualitatively, if a particle is smaller than 1000Ã…, there is no domain wall , i.e. only a single domain e xists . This theory was predicted by Frenkel and Dorfman as early as 1930 and then experimentally demonstrated by Kittel and co workers in 1950 by studying Ni particles which are approximately 200Ã…. 89 Interestingly, magnetism of a single domain is quite different from multi domain materials. Since there is only one domain with all spins aligned in the same direction, the

PAGE 36

36 magnetization is always saturated and lies along an axis , which is called the easy axis . An applied field in the opposite direction makes the magnetization vector rotate to the new easy direction without any do main wall motion. However, there are still anisotropy forces providing an energy barrier to hold the magnetization in one direction; therefore , magnetization of a single domain still shows hysteresis with relatively large coercivity (but usually below a ce rtain temperature, the blocking temperature) . Another feature of hysteresis curve s of perfect single domains is that the hysteresis loop is square. In addition to the existence of two stable states of opposite magnetization, the field required to switch be tween them is well defined. This is ideal in applications for data storage in which the two states resemble 1 and 0 of binary mixture. The size of a single domain particle has a decided effect on its coercivity. It turns out that energy barrier due to ani sotropy forces is proportional to the volume of the particle ; therefore , small particles also have small coercivity. As the size is reduced to tens of nanometer s , anisotropy energy is comparable to or smaller than the thermal energy, k B T . In the absence of an applied magnetic field, the magnetization can still spontaneously switch from one direction to the reversed one . At equilibrium, each half of the particles in the material ha s magnetization in opposite direction s leading to no net magnetization . When t he system is cooled below a critical temperature , called blocking temperature (T B ), thermal energy now becomes smaller th an the energy barrier. Consequently, when applying a magnetic field , magnetization of the system stays in the direction parallel to the field. If the field is oscillated, the system will experience a slow relaxation for the magnetization to catch up with the oscillating field . This phenomenon is the ground concept of superparamagnetism. Depending on the material, magnetic particles smalle r than 30nm normally show superparamagetic behavior and are called superparamagnets. Interestingly, a lthough the energy barrier of superparamagnets is higher than thermal energy at

PAGE 37

37 temperature s below T B , and one should therefore expect no magnetization rev ersal in the absence of a magnetic field, it was theoretically predicted that magnetization of superparamagnets could reverse through quantum tunnel ing. 90 However, the observation of quantum tunneling of magnetization requires the experiments to be performed on individual particles , or, if an assembly of them is used, they must be absolutely identical in size and, preferably, orientation . 91 Magnetic nanoparticles have been prepared by different methods such as mechanical grinding or sol gel techniques 92 but po lydisperse particles were ob tained rather than monodisperse ones. For polydisperse magnetic nanoparticles, the large distribution of sizes and resulting energy barrier heights limits their use for studies of quantum properties . In a very few particular cas es, magnetic nano crystal s have been synthesized . 93 An alternative way to achieve identical nanoscale particles is using a molecular approach . 94 M olecules containing ions with unpaired electrons can be candidate s for magnetic monodisperse nano particles. The quest for an individual molecule that behav es as a superpara magnet was achieved when the first single molecule magnet [Mn 12 O 12 (O 2 C Me ) 16 (H 2 O) 4 ], abbreviated Mn 12 ac, was reported in 1993 (Figure 1 10 ). 95,96 Magnetic studies of Mn 12 ac showed slow magnetization relaxation and a hysteresis loop below its blocking temperature (T B ~ 4 K) , similar to a sup erparamagnet. In addition, the magnetic behavior of Mn 12 ac was found to be intrinsic to the molecule and not due to long range magnetic ordering as in traditional magnets. This was concluded from several experiments such as magnetization relaxation data fo r frozen solutions 97 or polymer doped samples, the absence of any anomaly in heat capacity measuremen ts , 98 and high frequency electron paramagnetic resonance (HF EPR) data . 99,100 The unique property of Mn 12 ac attracted attention in the chemistry and physics communit ies , and led to a burst of new interest in the synthesis of

PAGE 38

38 molecular nanoscale magnetic materials and the stu dy of their magnetic behavior. The term single molecule magnet (SMM) was coined by Christou and Hendrickson in 1996. 101 Over the following two decades, a variety of polynuclear metal complexes displaying SMM beh avior have been synthesized . Manganese clusters represent the majority of SMMs with different sizes and nuclearities. T he highest nuclearity manganese cluster has been known to show SMM behavior is Mn 84 . 102 There are also SMMs with Fe, 103 Ni, 104 Co, 105 V, 106 lanthanides , 107,108 and heterometal l ic clusters by combinations of 3d with 4f paramagnetic ions . 109 Single molecule magnets have several advantages over traditional magn etic nanoparticles : i) they can be synthesized with ease in solution at ambient temperature, high temperature s are not required, in contrast to many traditional nanomagnets; ii) they are truly soluble in organic liquids, not forming colloidal solutions as nanoparticles do , which allows deposition of the molecules evenly on a surface for potential appli cations ; iii) they can be crystallized, thus providing ordered arrays with (usually) a single orientation, that can be well defined by single crystal X ray crystallography, iv) they are a collection of truly monodisperse particles of nanoscale dimensions, v) magnetically, they possess a single, well defined ground state spin and often display well resolved quantum effects ( vide infra ) . Similar to s ingle domain superparamagnet s , SMMs also have an energy barrier for the magnetization reversal that leads to hy steresis at temperatures below T B . The maximum energy barrier is defined as U = S 2 | D | for integer spin and ( S 2 1/4)| D | for half integer spin systems , where S is the ground state spin and D is the axial zero field splitting (ZFS) parameter. For an SMM to ha ve a high energy barrier and blocking temperature T B , it needs to have a large S combined with a large and negative D . ZFS lead s to the splitting of the g round state S into 2 S +1 sublevels (m S states) , each of which has energy given by E = D m S 2 , where m S is a quantum

PAGE 39

39 number describ ing the spin p rojection onto the easy axis and has the value s S , S , S 1, S . D is negative as required for SMMs; therefore , the sublevel s having the highest absolute m S value ha ve the lowest energy. A s chematic illustration o f the spin projection and energy barrier for an S = 10 system is shown in Figure 1 11 . Two diagnostic properties of an SMM are the frequency dependence of alternating current (ac) magnetic susceptibility signals, and hysteresis in a plot of magnetization v s. applied direct current (dc) magnetic field. In the ac studies, a small ac magnetic field (3.5 G) is applied, oscillating at a particular frequency usually in the 5 1500 Hz range . Out of phase ac peaks ( ) as well as a decrease in the in phase ac susceptibility signal ( ) are observed when the magnetic moment of the molecule cannot relax fast enough to keep in phase with the oscillating field (Figure 1 12) . 110 At the M peak maximum, the magnetization relaxation rate ( 1/ ) is equal to the angular frequency of the oscillating field: 1/ = 2 , where is the relaxation time and is the frequency of the oscillating field . The magnetization relaxation rate (1/ ) also obeys the Arrhenius equation: 1/ = (1/ 0 )exp( U eff / kT ) (1 1) ln(1/ ) = ln(1/ 0 ) U eff / kT (1 2) where U eff is the effective energy barrier ( U ), and k is the Boltzmann constant. Therefore, ac measurements at different frequencies of the oscillating field provide relaxation rate vs. temperature, which allow determin ation of U eff and 0 . Ac susceptibility measurements can also provide information on the nature of the magnetization relaxation processes. 110 Casimir and Du Pré proposed that for a single relaxation process (i.e. a single relaxation barrier), the real ( ) and imaginary ( ) components of the susceptibility as a function of angular frequency ( = 2 ) are given by equations 1 3 and 1 4, respectively. 111 I f there is a distribution of single relaxation processes (i.e. a distribution of relaxation barriers due to a distribution of molecular

PAGE 40

40 environments in the crystal ), and are given by eq uation s 1 5 and 1 6, respectively. 112 (1 3) (1 4) ( 1 5) ( 1 6) where 0 1. When = 0, equations 1 5 and 1 6 become equations 1 3 and 1 4, respectively. The wider the distribution in relaxation times, the larger is . When << 1, i.e. very fast relaxation, the measured is the isothermal susceptibility ( T ), which obey s the Curie law. When >> 1, i.e. very slow relaxation, the system has no time to exchange energy with the external world, and the measured is thus the adiabatic susceptibility ( S ). A Cole Cole plot, which is a plot of M vs. M at a given tempera ture (Figure 1 13), often shows a semi circle . At the top of the semicircle, the relation 1/ = is satisfied, which allows the relaxation time to be determined . In practice, many SMMs do not show a full peak in the out of phase ac magnetic susceptibility , even at temperature s as low as 1.8 K ; therefore , it is not possible to determine U eff and 0 by the above method. A rough estimation of U eff and 0 can instead be obtained based on two assumptions: i) the relaxation is a single process, in other words, and obey equations 1 3 and 1 4, and ii) adiabatic susceptibility S is very small compared to T and is negligible. Consequently, the ratio / obtained from equations 1 3 and 1 4 is: (1 7)

PAGE 41

41 (1 8) U eff and 0 can therefore be obtained from the slope ( U eff / k ) and the intercept (ln(2 0 )) of a ln( / ) vs . 1/ T plot . SMMs display hysteresis in a plot of magnetization vs. applied dc magnetic field (Figure 1 14) . It is noted that the hysteresis loops of SMMs are markedly different from those of bulk traditional magnets in which coercive field increas es with decreasing temperature and with increasing field sweep rate. 110 In fact, this behavio r is similar to traditional superparamagnets; however, there are still some differences. The hysteresis loops of SMMs often display clear steps , which were explained by the effect of quantum tunneling of magnetization (QTM). 113 115 The QTM phenomenon is the ability to reverse magnetic moment without going over the energy barrier , i.e. it tunnels through the barrier. 116 This is a pure quantum property exhibited by quantum object s, which also have wave nature. It can be interpreted as the wavefunction of the spin down state extending to state; in other words, the system is a superposition of the two states . QTM only occurs when there is a coincidence of energy of m s levels, there fore QTM steps only occur a t some critical magnetic fields ( H = n D / g B , n = 1, 2 ) (Figure 1 15 ) . The steps represent an increase in the magnetization r elaxation rate , which would not be useful for traditional memory storage purposes because data will be lost when the magnetization is reversed rapidly. Interestingly, when two or more SMMs interact wi th each other, the system displays an exchange biased quantum tunneling of magnetization effect , which was first discovered in the supramolecular C H··Cl hydrogen bonded pairs of [Mn 4 O 3 Cl 4 (O 2 CEt) 3 (py) 3 ] ( S = 9/2). The exchange biased interaction between two Mn 4 units show ed a shift of the QTM steps in the hysteresis loops , with the first step when scanning from negative field to positive field being before zero field (Figure 1 16). 117 To explain this behavior, it is necessary to consider

PAGE 42

42 the spin Hami l tonian of the Mn 4 monomer (equation 1 9) and the dimer (equation 1 10) as follows : (1 9) (1 10 ) where i = 1 or 2 (referring to the two Mn 4 SMMs of the dimer), D is the axial ZFS parameter, B is the Bohr magneton, g is the electronic g factor, H z is the applied longitudinal magnetic field, J is the coupling parameter between the two Mn 4 units , and S 1 = S 2 = 9/2. Neglecting the transverse anisotropy terms, equation 1 10 simplifies to: (1 11) Thus, t he eigenvalue s of equation 1 1 1 are : E D (m 1 2 +m 2 2 ) g B H z (m 1 +m 2 ) 2 J (m 1 m 2 ) (1 12) with m 1 and m 2 having 7/2, 9/2. The energy diagram based on equation 1 12 is show n in Figure 1 17. The step pattern of the hysteresis loops can be explained by the leve l crossings , as show n by the dotted line s in Figure 1 17. At zero field, the possibility of tunneling from ( 9/2, 9/2) to (9/2, 9/2) is very low because this would require simultaneous tunneling of both Mn 4 molecules of the dimer , and no step is seen . 117 As mentioned above, QTM might not be usef ul for traditional memory storage; however, t he absence of tunneling at zero field due to exchange biased tunneling provides a realistic possibility to use these molecules in devices. Moreover, t he decoherence time of the quantum superposition states in th e Mn 4 dimer was found to be larger than 1 ns , which shed s light on developing quantum computers based on quantum coherence phenomenon. 118 SMMs have been proven to have intriguing properties and need to be explored further . Currently, there are several research approaches employed in the SMM field : i) synthesis of new

PAGE 43

43 SMMs with high energy barriers and blocking temperatures , 108,119 ii) synthesis of SMMs in a contro l l ed fashion to study different ideas of magnetism and quantum physics , 117,118,120,121 and iii) grafting of SMM on surfaces to investigate the possibility of applications . 122,123 This work focuses on the synthesis and study of the quantum properties of covalently bonded supramolecular aggregates of SMMs. Our objectives are: 1) to c onnect SMMs covalently to form discrete supramolecular aggregates , wh ich are expected to be stable in solu tion for deposition on surfaces; 2) to i nvestigate the structur al effects , e.g. flexibility, bulkiness , of ligands on the formation of these supramolecules and on their magnetic properties ; and 3) to i nvestigate the qua ntum properties of these supramolecules in the solid state and also in solution . Through this work, a variety of supramolecular aggregates of Mn 3 SMM ha s been synthesized and studied. The layout of the report is as follows: C hapter 2 presents non SMM [Mn 3 ] 2 and SMM [Mn 3 ] 4 supramolecular aggregates from the use of 3 phenyl 1,5 di(pyridin 2 yl)pentane 1,5 dione dioxime (pdpdH 2 ); C hapter 3 shows the employ ment of 1,2 di(pyridin 2 yl)ethane 1,2 dione dioxime (dedH 2 ) to form tetrahedral [Mn 3 ] 4 SMMs ; C hapter 4 e lucidates the importance of ligand flexibility by explor ing 1, 3 di(pyridin 2 yl)p ropane 1, 3 dione dioxime (dpdH 2 ) for the formation of a [Mn 3 ] 2 SMM; C truxillic acid , and dimethylmalonic acid to join Mn 3 SMMs by carboxylate substitution.

PAGE 44

44 Figure 1 1. Examples of a crown ether (left), a cryptand (middle), and a spher and (right) .

PAGE 45

45 Figure 1 2. DNA structure (left) with hydrogen bonding pattern (right) between nucleotides . 124

PAGE 46

46 Figure 1 3 . Prediction of two dimensional (top) and three dimensional architectures (bottom) . Reprinted with permission from ref erences 27 a nd 72 . Copyright 2011 American Chemical Society and Copyright 2000 National Academy of Sciences, U.S.A.

PAGE 47

47 Figure 1 4. Design strategy for (a) M 4 L 6 and (b) M 4 L 4 tetrahedra . Reprinted with permission from ref erence 27. Copyright 2011 American Chemical Societ y . Figure 1 5. Design strateg y for M 6 L 3 trigonal prism and M 12 L 6 hexagonal prism . Reprinted with permission from ref erence 27. Copyright 2011 American Chemical Society .

PAGE 48

48 Figure 1 6. Weak link approach to access supramolecules . Reprinted with permission from ref erence 27. Copyright 2011 American Chemical Society . Figure 1 7. Schematic illustration of spin coupling behaviors in A ) paramagnetic, B ) ferromagnetic, C ) antiferromagnetic, and D ) ferromagnetic materials . Figure 1 8. Schematic illustration of domains (left) and a domain wall (right) . A) Paramagnetic B) Ferromagnetic C) Antiferrom agnetic D) Ferrimagnetic

PAGE 49

49 Figure 1 9. Schematic illustration of a hysteresis curve for a typical ferromagnet. M s is saturated magnetization, M r is remnant magnetization and H c is coercive field . Figure 1 10. The [Mn III 8 Mn IV 4 ( 3 O 2 ) 12 ] 16+ core (a) and the full structure of Mn 12 Ac complex (b). Color code: Mn IV green; Mn III blue; O red; C gray. H atoms have been omitted for clarity .

PAGE 50

50 Figure 1 11 . Plot of the orientations of the m s vectors along the z axis ( left) and the double well potential showing the energy versus the m s sublevel for a complex with an S = 10 ground state spin , experiencing zero field splitting, z 2 (right) . Reprinted with permission from ref erence 110 . Copyright 2009 Royal Society of Che mistry . Figure 1 12 . In phase ( M , plot as M T ) (top) and out of phase (as M ) (bottom) ac susceptibility signals for a dried, microcrystalline sample of [Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ] at the indicated oscillation frequencies . Reprinted with permission from ref erence 110 . Copyright 2009 Royal Society of Chemistry .

PAGE 51

51 Figure 1 13. Cole Cole plot of " vs. for a sample of [Mn 12 O 12 (O 2 CR) 16 (H 2 O) 4 ]. The dash ed line is the fit of the data to a single relaxation process. The solid line is the fit to a distribution of single relaxation processes. Rep rinted with permission from reference 110 . Copyright 2009 Royal Society of Chemistry . Figure 1 1 4 . Magnetization hysteresis loops for [Mn 12 O 12 (O 2 C R) 16 (H 2 O) 4 ] complex in the 1.3 3.6 K temperature range at a 4 mT/s field sweep rate. M is normalized to its saturation value, M s . Reprinted with permission from ref erence 110 . Copyright 2009 Royal Society of Chemistry .

PAGE 52

52 Figure 1 1 5 . Schematic represent ation of the change in energy of m s sublevels as the magnetic field is swept from zero to a non zero value. Resonant magnetization tunneling occurs when the m s sublevels are aligned between the two halves of the diagram . Figure 1 16. Magnetization hyster esis loops of [Mn 4 ] 2 versus applied magnetic field at (A) different temperatures and (B) different field sweep rates . Reprinted with permission from ref erence 117 . Copyright 200 2 Nature Publishing Group . QTM occurs QTM is not allowed QTM occurs

PAGE 53

53 Figure 1 17. Energy diagrams showing the spin sta te energies of [Mn 4 ] 2 complex versus applied magnetic field . a) exchange coupled dimer of two spin S = 9/2 Mn 4 units. b) Enlargement of ( a ) , showing only levels populated at very low temperature wh 1.2 T to + 1.2 T, as indicated by green arrows. Dotted lines, labeled 1 to 5, indicate the strongest tunnel resonances . Reprinted with permission from ref erence 117. Copyright 2002 Nature Publishing Group .

PAGE 54

54 CHAPTER 2 SUPRAMOLECU LAR AGGREGATES OF SINGLE MOLECULE MAGNETS FROM THE USE OF 3 PHENYL 1,5 DI(PYRIDIN 2 YL)PENTANE 1,5 DIONE DIOXIME: DEVELOPING A RATIONAL DESIGN APPROACH 2.1 Introduction Single molecule magnets (SMMs) are individual molecules that function as single domain nanoscale magnetic particles , 95,96,100,110,125,126 and are considered to be th e smallest species able to store one bit of magnetic information. Different from molecule based magnets which require intermolecular in teractions and long range ordering, 127 129 these molecules are a class of transition metal clusters that display intrinsic superparamagnetic behavior on the molecular scale below their blocking temperatur e, T B . For 3d transition metal molecules, this behavior arises from the combination of a large ground state spin ( S ) and Ising type magnetoanisotropy (negative zero field splitting parameter, D ), which leads to a significant energy barrier (vs. kT ) to the thermal equilibration of the molecular magnetic moment. The maximum energy barrier can be calculated by S 2 | D | or ( S 2 1/4)| D | for integer and half integer spin systems , respectively. Experimentally, an SMM exhibits frequency dependent out of phase ac magnet ic susceptibility signals, and hysteresis in a plot of magnetization vs. applied dc magnetic field. The significant advantage of the single molecule approach to nano magnetism is which are trully identical and have we ll defined sizes and electronic lev els. Moreover, the possibility o f control over their chemical reactions allows for exquisite tun ing of structures and magnetic properties of the molecule s . Various famili es of SMMs have been discovered, the majority of wh ich are Mn based complexes with Mn 84 being the largest among them . 102 SMMs have also been shown to display interesting quantum phenomena such as quantum tunneling of magnetization (QTM) 113,115 , quantum phase interference (QPI) 130 132 , spin spin cross relaxation, 133 and quantum entanglement. 118,121,134

PAGE 55

55 Consequently, they have been proposed as qubits for quantum computation 135 138 and as components in molecular spintronics devices, 139,140 which would exploit their quantum properties. For s uch applications, weak coupling of two or more SMMs to each other or to other components of a device are essential, while maintaining the intrinsic single molecule prop erties of each SMM. The report of supramolecular C H·· · Cl hydrogen bonded pairs of [Mn 4 O 3 Cl 4 (O 2 CEt) 3 (py) 3 ] ( S = 9/2) demonstrated such coupling between two SMMs for the first time, manifested as exchange biased QTM steps, quantum superposition states, and quantum entanglement of the two SMMs . 117,118,121 Several supramolecular dimers, chains and 3 D networks of weakly coupled SMMs connected by hydrogen bonds have since been reported. 120,141 145 Disadvantages of linka ge by hydrogen bonds, however, are (i) de aggregation into monomeric units on dissolution, and (ii) the major loss of synthetic control, with all the previously ment ioned supramolecular aggregates being obtained serendipitously. A superior approach is conn ection of SMMs via covalent bonds. Such covalent linkage of SMMs has already been explored extensively, invariably leading to a 1 D chain or 2 or 3 D networks. 146 151 The coupling between these SMMs is often (but not always 152 154 ) strong enough to lead to a loss of SMM identity and gives 1 D single chain magnets (SCMs) or 2 or 3 D ordered materials. Discrete molecules having covalent linkage of two or more subunits are re latively rare 155 162 , and among them very few compounds exhibit SMM behavior. 163,164 However, none of these SMM supramolecules display exchanged biased coupling, probably due to the following reasons: i) the subunits of the supramolecule interact too strongly causing the supramolecule to behave like a single molecule, ii) when the interactions between subunits are weak, each individual subunit might have low lying excited states that interfere with the quantum tunneling process, which results in stepless hysteresis loops. A n ideal supramolecular aggregate of SMM s should possess

PAGE 56

56 two features: i) each subunit is a "good" SMM showing hysteresis loops with clear QTM steps , ii) SMM subunits weakly co uple to each other through connecting ligands but still maintain their intrinsic single molecule magnetic behavior. Moreover, better understanding is achieved by comparing the subunit to the corresponding supramolecular aggregate, thus it requires each SMM subunit monomer to be well studied structurally and magnetically . The first feature indicates tha t SMM supramolecular aggregates are rather derived from known SMM building blocks. Serendipitous syntheses using the "right" ligand can sometimes give SMM sup ramolecular aggregates , but there is limited guarantee that the desired magnetic properties will be obtained. 164 The latter feature can be obt ained if the ligands connecting those SMM subunits create relatively large separation between them. Several studies have provided interesting insight into tailoring exchange interactions when connecting non SMM subunits, such as the linking of two Cr 7 Ni wh eel molecules 165 or of two lanthanide ions, 166 resulting in weak antiferromagnetic (AF) interactions. Our first goal is to covalently link two or more Mn SMMs to give non polymeric, of SMMs sho wing very weak inter SMM interactions. In the present work, we report the syntheses and magnetic properties of four supramolecular aggregates , two of which are non SMM dimers [Mn 3 ] 2 , and the other two are SMM tetramers [Mn 3 ] 4 . The connecting ligand is deri ved from a designed dioxime group, 3 phenyl 1,5 di(pyridin 2 yl)pentane 1,5 dione dioxime (pdpdH 2 ) (Figure 2 1). It was constructed based on the chemistry of the previously known methyl 2 pyridyl ketone oximes (mpkoH), part of a longstanding interest for many research groups to study structures and properties of transition metal oxime compounds. 167 170 We shall first present the use of pdpdH 2 in a direct approach and the subsequent process leading to the isolation of rectang u l ar supramolecular aggregates of Mn 3

PAGE 57

57 SMMs in a controlled fashion starting with the well known SMM building block [Mn 3 O(O 2 CMe) 3 (mpko) 3 ](ClO 4 ) ( 3 ) (Figure 2 1 ). 171 We shall also demonstrate the exchange biased interaction resulting from weak couplings between Mn 3 units in the SMM tetramers [Mn 3 ] 4 . HF EPR stud ies on a po wder sample the SMM tetramer [Mn 3 ] 4 will also be discussed. 2.2 Experimental Section 2.2.1 Synthese s All manipulations were performed under aerobic conditions using chemical s and solvents as received unless otherwise stated . [Mn 3 O(O 2 CMe) 6 ( py ) 3 ] ( ClO 4 ) ( 1 ) w as prepared as described elsewhere. 172 The dioxime pdpdH 2 was synthesized in two steps : f rom the condensation of 2 acetylpyridine and benzaldehyde, the intermediate 3 phenyl 1,5 bis(pyridin 2 yl)pentane 1,5 dione was obtained according to a literature method 173 and treated with hydroxylamine to form the crude product. Recrystallization from acetone gave pure pdpdH 2 in 55% overall yield. 174 Caution! Although no such behavior was observed during the present work, perchl orate compounds are potentially explosive; such compounds should be synthesized and used in small quantities, and treated with utmost care at all times. 2.2.1.1 [Mn 6 O 2 (O 2 CMe) 8 (MeOH) 2 (pdpd) 2 ] (2 1) A solution of Mn(O 2 CMe) 2 · 4H 2 O (0.25 g, 1 .0 mmol) in 15 m L of MeCN : MeOH (2 : 1 v/v) was treated with triethylamine (0.14 m L , 1 .0 mmol) and pdpdH 2 (0.18 g, 0.5 0 mmol). The solution was stirred for 1 hour at room temperature, filtered and the black filtrate was left undisturbed to concentrate slowly by evaporation . X ray quality crystals of 2 1 formed after one week in 75% yield. These were collected by filtration, washed with Et 2 O, and dried under vacuum. Anal. Calcd (found)% for 2 1 · 4H 2 O (Mn 6 O 28 C 60 H 76 N 8 ): C 42.72 (42.39); H 4.54 (4.18);

PAGE 58

58 N 6.64 (6.25). Selected IR data (KBr, cm 1 ): 1600 (s), 1471 (m), 1397 (s), 1338 (m), 1182 (m), 1107 (w), 1035 (w), 784 (w), 760 (w), 700 (m), 658 (m), 614 (m), 542 (w), 421 (w). 2.2.1.2 [Mn 6 O 2 (O 2 CMe) 8 (py) 2 (pdpd) 2 ](ClO 4 ) 2 (2 2) A brown solution of complex 1 (0.44 g, 0.5 0 mmol) in 1 5 m L of MeCN : MeOH (2 : 1 v/v) was treated with pdpdH 2 (0.18 g, 0.5 0 mmol). The solution was stirred for 1 hour at room temperature and then evaporated under vacuum to obtain a black powder. The powder was redissolved in CH 2 Cl 2 , and the resulting solution was layered with Et 2 O . X ray quality crystals of 2 2 formed after 2 days in 80% yield. The crystals were collected by filtration, washed with Et 2 O , and dried under vacuum. Anal. Calcd (found)% for 2 2 · 3H 2 O (Mn 6 O 33 C 68 H 76 N 10 Cl 2 ): C 41.63 (41.79); H 3.90 (3 .62); N 7.14 (6.77). Selected IR data (KBr, cm 1 ): 1600 (s), 1474 (m), 1392 (s), 1339 (m), 1182 (w), 1108 (m), 1030 (w), 937 (w), 784 (w), 760 (w), 698 (w), 661(m), 625 (w). 2.2.1. 3 [Mn 12 O 4 (O 2 CMe) 12 (pdpd) 6 ](ClO 4 ) 4 (2 3) A brown solution of complex 1 (0.04 4 g, 0.05 0 mmol) in 25 m L of CH 2 Cl 2 was treated with pdpdH 2 (0.036 g, 0.1 0 mmol). The solution was stirred for 1 hour at room temperature, filtered and the black filtrate was left undisturbed. X ray quality crystals of 2 3 formed after 2 days in 35% yield. The crystals were collected by filtration, washed with CH 2 Cl 2 , and dried under vacuum. Anal. Calcd (found)% for 2 3 · 2CH 2 Cl 2 (Mn 12 O 56 C 152 H 148 N 24 Cl 8 ): C 43.99 (43.58); H 3.59 (3.54); N 8.10 (7.75). Selected IR data (KBr, cm 1 ): 1600 (s), 1574 (s), 1474 (m), 1388 (s), 1337 (m), 1183 (m), 1162 (m), 1109 (s), 1088 (m), 1047(w), 781 (w), 758 (w), 700 (m), 661(m), 623 (m).

PAGE 59

59 2.2.1.4 [Mn 12 O 4 (O 2 C t Bu) 12 (pdpd) 6 ](ClO 4 ) 4 (2 4) A brown solution of complex 2 3 (0.21 g, 0.05 0 mmol) in 30 m L of MeCN : EtOH (2 : 1 v/v) was t reated with pivalic acid (0.12 g, 0.12 mmol). The solution was stirred overnight at room temperature. Toluene was added and the solvent was removed in vacuo. After three cycles of toluene addition and solvent removal, the residue was redissolved in 25 m L o f CH 2 Cl 2 and the resulting solution was layered with hexane s . X ray quality crystals of 2 4 formed after 2 3 days in 50% yield. The crystals were collected by filtration, washed with hexane s, and dried under vacuum. Anal. Calcd (found)% for 2 4 · 2CH 2 Cl 2 (Mn 12 O 56 C 188 H 220 N 24 Cl 8 ) C 48.51 (48.83); H 4.76 (4.92); N 7.22 (6.95). Selected IR data (KBr, cm 1 ): 1601 (s), 1584 (s), 1561 (s), 1516 (w), 1478 (s), 1440 (w), 1400 (s), 1384 (m), 1348 (s), 1219 (s), 1188 (m), 1165 (w), 1096 (s), 783 (m), 756 (w), 700 (m), 6 63(m), 622 (s), 449 (m). 2.2.2 X Ray Crystallography X Ray Intensity data were collected at 100 K on a Bruker DUO diffractometer using MoK radiation ( = 0.71073 Å) and an APEXII CCD area detector. Suitable crystals of 2 1 · xCH 3 OH · yCH 3 CN, 2 2 · xC 4 H 10 O · yCH 2 Cl 2 , 2 3 · xCH 2 Cl 2 and 2 4 · xCH 2 Cl 2 were attached to glass fibers using silicone grease and transferred to a goniostat where they were cooled to 100 K for data collection. Raw data frames were read by the program SAINT and integrated using 3D profiling algor ithms. The resulting data were reduced to produce hkl reflections, their intensities and estimated standard deviations. The data were corrected for Lorentz and polarization effects, and numerical absorption corrections were applied based on indexed and mea sured faces. The structures were solved and refined in SHELXTL6.1, 175 using full matrix least squares refinement. The non H atoms were refined with anisotropic thermal parameters and a ll of the H atoms were calculated in idealized positions and refined riding on their parent atoms.

PAGE 60

60 For 2 1 · xCH 3 OH · yCH 3 CN, the asymmetric unit consists of a half Mn 6 cluster, one and a half methanol and one acetonitrile solvent molecules. The solvent molec ules were disordered and could not be modeled properly; thus the program SQUEEZE, a part of the PLATON package of crystallographic software, was used to calculate the solvent disorder area and remove its contribution to the overall intensity data. Atom O12 is protonated; its H atoms were obtained from a different Fourier map and refined freely. The phenyl ring C9 disordered and refined in two parts whose site occupation factors were dependently refined. In the final cycle of refinement, 9 225 reflections (of which 6757 are observed with I > 2 (I)) were used to refine 423 parameters and the resulting R 1 , wR 2 , and S (goodness of fit) were 4.60%, 12.76% , and 1.080, respectively. For 2 2 · xC 4 H 10 O · yCH 2 Cl 2 , the asymmetric unit consists of a Mn 6 cluster cation, two perchlorate anions, two ether and four dichloromethane solvent molecules. Both anions were disordered. In the first case it was resolved in two parts while in the second case the minor and major parts shared atom O23. The solvent molecu les were disordered and could not be modeled properly; thus the program SQUEEZE, a part of the PLATON package of crystallographic software, was used to calculate the solvent disorder area and remove its contribution to the overall intensity data. In the fi nal cycle of refinement, 10839 reflections (of which 7710 are observed with I > 2 (I)) were used to refine 1014 parameters and the resulting R 1 , wR 2 , and S (goodness of fit) were 7.32%, 20.28% , and 1.112, respectively. For 2 3 · xCH 2 Cl 2 , the asymmetric un it consists of two Mn 12 cluster cations, eight perchlorate anions, and 52 dichloromethane solvent molecules. Three perchlorate anions were disordered along one of the Cl O bonds, thus the three O atoms were refined in two parts whose site occupation factor s were dependently refined. Those are on Cl2, Cl6 and Cl7. The Cl8

PAGE 61

61 perchlorate anion was completely refined in two parts. In this case there was a ¼ dichloromethane associated with each disordered part of the Cl8 anion. Additionally, most of the 52 dichlor omethane solvent molecules were significantly disordered and could not be modeled properly; thus the program SQUEEZE, a part of the PLATON package of crystallographic software, was used to calculate the solvent disorder area and remove its contribution to the overall intensity data. In the final cycle of refinement, 81724 reflections (of which 18367 are observed with I > 2 (I)) were used to refine 2905 parameters and the resulting R 1 , wR 2 , and S (goodness of fit) were 7.40%, 16.17% , and 0.624, respectively. For 2 4 · xCH 2 Cl 2 , t he asymmetric unit consists of a half Mn 12 cluster cation, two perchlorate anions and ten dichloromethane solvent molecules. The solvent molecules were disordered and could not be modeled properly, thus the program SQUEEZE, a part of the PLATON package of crystallographic software, was used to calculate the solvent disorder area and remove its contribution to the overall intensity data. The methyl groups on C102 and C122 were disordered and each set was refined in two parts whose site occupation factors fixed at values of 0.6 and 0.4. They were also constrained to maintain equivalent geometries using SADI and EADP in the final refinement model. In the final cycle of refinement, 31525 reflections (of which 15982 are observed with I > 2 ( I)) were used to refine 1275 parameters and the resulting R 1 , wR 2 , and S (goodness of fit) were 5.05 %, 11.25 % , and 0.840 , respectively. Unit cell data and structural refinement details for the four compounds are listed in Table 2 1. 2.2.3 Physical Measure ments Infrared spectra were recorded in the solid state (KBr pellets) on a Nicolet Nexus 670 FTIR spectrometer in the 400 4000 cm 1 range. Elemental analyses (C, H, and N) were

PAGE 62

62 performed by the in house facilities of the University of Florida, Chemistry Department. Variable temperature direct current (dc) and alternating current (ac) magnetic susceptibility data were collected at the University of Florida using a Quantum Design MPMS XL SQUID magnetometer equipped with a 7 T magnet and operating in the 1. 8 300 K range. Samples were embedded in solid eicosane to prevent torquing. Magnetization vs. field and temperature data were fit using the program MAGNET. 17 6 diamagnetic correction, which was subtracted from the experimental susceptibility to give the molar paramagnetic susceptibility ( M ). Low temperature (<1.8 K) hysteresis loop and dc relaxation measurements we re performed at Institut Néel using an array of microSQUIDS. 177 The high sensitivity of this magnetometer al lows the study of single crystals of SMMs of the order of 10 500 m. The field can be applied in any direction by separately driving three orthogonal coils. High frequency electron paramagnetic resonance (HF EPR) data were collected at the U.S. National High Magnetic Field Laboratory Electron Magnetic Resonance facility. 2.3 Results and Discussion 2.3.1 Synthes e s In the direct approach, the reaction can occur between the ligand and a simple metal salt. In our case, different reactions between pdpdH 2 and a series of manganese salts were screened u n der different conditions such as a variety of solvent s , or chemical ratios, or the use of different base s, etc. Complex 2 1 formed when pdpdH 2 reacted with Mn(O 2 CMe) 2 · 4H 2 O in MeCN : MeOH (2:1 v/v) . T he yield of the product was improved when a weak base, triethylamine, was used. The formation of complex 2 1 is summarized in equation 2 1: 6Mn(O 2 CMe) 2 · 4H 2 O + 2pdpdH 2 + 2MeOH + 4Et 3 N + O 2 6 O 2 (O 2 CMe) 8 (MeOH) 2 (pdpd) 2 ] + 4MeCO 2 + 4Et 3 NH + + 4H 2 O ( 2 1)

PAGE 63

63 Methanol is essential in this reaction since it helps to completely dissolve all the reactants . The dioxime pdpdH 2 was used in slight excess , and it was noted that increasing the amount of pdpdH 2 in the reaction by changing the Mn : ligand ratio from 3 : 1 to 1 : 1 s till leads to the same product. The optimized Mn : ligand ratio to give a clean and high yield product is 2 : 1. The presence of Mn II ions in complex 2 1 , which often exhibit weak exchange interactions ( vide infra ) , is unfortunate, but logical since the st arting material was a Mn II salt and not all the Mn II ions were oxidized by O 2 from air, which is the only source of oxidizing agent in the reaction. To obtain all Mn III products , starting from an all Mn III preformed complex, e . g . complex 1, would be prefer red in order to prevent the formation of Mn II containing compounds. Complex 2 2 , an all Mn III version of complex 2 1 , was formed when one equivalent of 1 was t reat ed with one equivalent of pdpdH 2 . The formation of complex 2 2 is summarized in equation 2 2 : 2[Mn 3 O(O 2 CMe) 6 (py) 3 ](ClO 4 ) + 2pdpdH 2 6 O 2 (O 2 CMe) 8 (py) 2 (pdpd) 2 ](ClO 4 ) 2 + 4MeCO 2 + 4py + 4H + ( 2 2) Four MeCO 2 and four pyridine molecules were replaced by two pdpd 2 groups in the reaction. Additional base was not essential because the reaction pr oduces pyridine, which is also a relatively strong base. Although complexes 2 1 and 2 2 are compri sed of supramolecular aggregates of Mn 3 units, they are uninterest ing in terms of single molecule magnetism ( vide infra ). The direct approach to form 2 1 or improved one 2 2 does not give products with des ired properties . Reconsidering our approach , we remembered that the well known Mn 3 SMM complex 3 is formed from the reaction between 1 and the monooxime mpkoH (Fig ure 2 1) . A source of structural inter est which should be noted is that the three oximate groups all lie on one side of the Mn 3 plane. This is a necessary condition for the formation of a discrete supramolecular

PAGE 64

64 aggregate. The o xim ate ligand , however, is a tridentate ligand and cannot easily b e substituted by a dioxime. To target the supramolecular aggregate of complex 3 , we used pdpdH 2 in place of mpkoH in a similar reaction. 171 The reaction gave complex 2 3 and its formation is summarized in equation 2 3: 4[Mn 3 O(O 2 CMe) 6 (py) 3 ](ClO 4 ) + 6pdpdH 2 12 O 4 (O 2 CMe) 12 (pdpd) 6 ](ClO 4 ) 4 +12MeCO 2 + 12py + 12H + ( 2 3) Six pdpd 2 groups replaced twelve MeCO 2 and twelve pyridine molecules. F or the reaction to proceed , pdpdH 2 was used in large excess , with the initial ratio [Mn 3 ] : pdpdH 2 being 1 : 3. The reaction was further optimized by systematically decreasing the amount of pdpdH 2 . The reaction with a [Mn 3 ] : pdpdH 2 ratio of 2 : 3 gave a very low yield , while the 1 : 2 ratio ga ve a clean product in high yield . Similar to 2 2 , additional base is not necessary because pyridine is produced in the reaction. It is worth noting that the [Mn 3 ] : pdpdH 2 ratio is crucial in the formation of 2 2 and 2 3 . The next step was to modify complex 2 3 in a control led fashion. Usually the organic ligands can be m odified by using standard ligand substitution methods, which have been crucial in allowing targeted modification of SMMs such as the Mn 4 and Mn 12 families, 178,179 with profound impact on many studies. The first carboxylate substitution on complex 2 3 was performed with pivalic acid and this successfully gave complex 2 4 . Its formation is summarized in equation 2 4: [Mn 12 O 4 (O 2 CMe) 12 (pdpd) 6 ](ClO 4 ) 4 + 12 t BuC O 2 12 O 4 (O 2 C t Bu) 12 (pdpd) 6 ](ClO 4 ) 4 +12MeCO 2 H ( 2 4) To drive the reaction completely to the product, toluene was added to the reaction mixture and the acetic acid was removed as its toluene azeotrope by evaporation under vacuum. The success

PAGE 65

65 of this c arboxylate substitution is the premise for the isolation of [Mn 3 ] 4 supramolecular aggregates contain ing functional carboxylate ligands des igned for material applications. 2.3.2 Description of Structure s Partially labeled structures of complex es 2 1 and 2 2 are shown in Figure 2 2 and Figure 2 3, respectively. Some important i nteratomic distances and angles of these two complexes are listed in Table 2 2; a more comple te listing is available in the A ppendix C . Complex es 2 1 ·xMeOH·yMeCN and 2 2 ·xC 4 H 10 O·yCH 2 Cl 2 crystallize in the monoclinic space group C2/c and triclinic space group , respectively. The cores ( Fig ure s 2 2 and 2 3 ) can be described as consisting of two [Mn III 2 Mn II ( 3 O)] 6+ or [Mn III 3 ( 3 O)] 7+ triangular subunits linked by two pdpd 2 groups to give dimer s [Mn 3 ] 2 0/+ . There is one bridging 3 oxide atom in each Mn 3 subunit. The central oxide atoms lie close to the Mn 3 planes in both complexes (d 0.01Å). Two edges of a triangle are bridged by four 1 : 1 : acetate groups; the other edge is br idged by two 1 : 1 : 1 : pyridyloximate groups. One methanol molecule bin ds terminally to each of the Mn II ions in complex 2 1 . The MnIII ions at similar positions in complex 2 2 are bound terminally by pyridine molecules. The Mn ··· Mn separations and Mn ( 3 O) Mn angles in each triangle are different; the Mn ··· Mn distance and the angle between two Mn ions bridged by two oxime groups are much smaller than the distances and angles between Mn ions bridged by two acetates . This is a characteristic of the o ximate group since a diatomic N O group replaces a triatomic O C O carboxylate bridge and shortens the Mn ··· Mn separation. The triangles are scalene, but almost isosceles within the 3 criterion (Table 2 2 ). In complex 2 1 , the Mn ions in the two triangles are crystallographically equivalent; complex 2 1 has C 2 symmetry. The Mn ions in complex 2 2 are not crystallographically symmetric, thus the molecule has C 1 symmetry but shows virtual C 2 symmetry. The oxidation states of the Mn atoms were determined base d on bond valence sum

PAGE 66

66 (BVS) calculations, charge balance considerations , and the presence of JT distortion of the Mn III ions. BVS calculations also confirm that the 3 O atoms are O 2 ions. Inspection of short contacts reveals that in complex 2 1 each [Mn 3 ] 2 molecule connects to its neighbors by two weak hydrogen bonds ( 1.98 Ã…) to form a virtual {[Mn 3 ] 2 } n chain , while in complex 2 2 there are interactions between the molecules (Figure 2 4) . Partially labeled structures of the cations of 2 3 and 2 4 are shown in Figure s 2 5 and 2 6, respectively . Selected interatomic distances and angles are listed in Table 2 3; a complete listing is available in the A ppendix C . Complex es 2 3 and 2 4 crystallize in the monoclinic space group P2 1 /c and C2/c, respe ctively. For complex 2 3 , the asymmetric unit consists of two essentially superimposable Mn 12 cations (one is shown in Figure 2 5 ), eight perchlorate anions, and large amounts of disordered CH 2 Cl 2 solvent. Complex 2 4 has only one type of Mn 12 cation in th e asymmetric unit ; the cations are arrang ed parallel to each other in the crystal (Figure 2 7). The Mn 12 cation in both complexes consists of four [Mn 3 ( 3 O)] 7+ units linked by six pdpd 2 groups to give a rectangular [Mn 3 ] 4 supramolecule . One of the two 1 : 1 : MeCO 2 ligands bridging each edge of complex 1 has been replaced by a bridging oximate from a pdpd 2 group. Two of the latter bridge to the same neighboring Mn 3 unit, and the third bridges to a different one (Figure s 2 5 and 2 6 , bottom). In addi tion, the pdpd 2 pyridyl groups have replaced the terminal pyridine ligands of 1 . Each of the 3 O 2 ions lies slightly above the Mn 3 plane (d ~ 0.3 Ã…), and all the Mn N O Mn torsion angles are relatively large ( 8 20 o ) (Table 2 4), as in 3 . Thus, the lo cal structure of each Mn 3 unit of 2 3 and 2 4 is very similar to that of 3 , comprising a [Mn 3 ( 3 O)] 7+ triangular unit whose edges are each bridged by one acetate and one pyridyloximate group, and whose pyridyl group binds terminally to the Mn. Also as in 3 , the three bridging oximate groups are on the same side of the Mn 3 plane, and this is crucial to the formation of a molecular

PAGE 67

67 tetramer rather than a polymer. In fact, it can be anticipated that the product might be an [Mn 3 ] 4 tetrahedron with a bridging p dpd 2 on each edge, but the obtained rectangle is also a logical arrangement due to the flexibility of the carbon backbone of pdpd 2 group. T wo CH 2 Cl 2 molecules are encapsulated within the central cavity of the Mn 12 cations, with no opening big enough for them to escape (Figure 2 8 ). The Mn···Mn separations and Mn ( 3 O) Mn angles in each triangle are slightly different; thus the triangles are scalene but virtually isosceles within the usual 3 criterion. The cations of 2 3 and 2 4 have crystallographic C 1 and C 2 symmetry, respectively, and virtual D 2d symmetry in both cases. The Mn III oxidation states were confirmed by bond valence sum calculations, and their Jahn Teller elongation axes (green bonds in Figure s 2 5 and 2 6 , bottom) are aligned in a propeller fashion, again as in 3 . Inspection of short contacts and packing diagrams of 2 3 and 2 4 reveals that there is no essential interaction between the [Mn 3 ] 4 molecules. The shortest distances between Mn 3 units in 2 3 and 2 4 are ~6 and ~7Å, respectively, whi ch are similar to the one s observed in 3 (Figure 2 9 ) . Overall, 2 3 and 2 4 can accurately be described a s a tetrameric version of complex 3 SMM . 2.3.3 Magnetochemistry 2.3.3.1 Direct current magnetic susceptibility studies Variable temperature, dc magneti c susceptibility ( M ) measurements were performed on vacuum dried polycrystalline samples of complex es 2 1 · 4H 2 O, 2 2 · 3H 2 O, 2 3 · 2CH 2 Cl 2 and 2 4 · 2CH 2 Cl 2 in an applied field of 1000 G (0.10 T) and the 5.0 300 K temperature range. The samples were restrained in eicosane to prevent torquing. Figure 2 10 shows the molar magnetic susceptibility ( M ) of complex es 2 1 and 2 2 as a M T vs. T plot. The M T value for complex 2 1 is 15.51 cm 3 K mol 1 at 300 K, and steadily decreases with decreasing temperature to 3.32 cm 3 K mol 1 at 5K. Complex 2 2 displays a similar behavior, with a M T value of 19.76 cm 3 K mol 1 at

PAGE 68

68 300 K, which decreases to 5.31 cm 3 K mol 1 at 5 K. The value at 300 K for complex 2 1 is much smaller than the spin only (g = 2) value for four Mn III and two Mn II non interacting ions ( M T = 20.75 cm 3 K mol 1 ), indicating dominant antiferromagnetic interactions within the molecule. Complex 2 2 , however, has a M T value at 300 K slightly higher than the spin only ( g = 2) value for six Mn III non interacting i ons ( M T = 18.00 cm 3 K mol 1 ), suggesting that there are competing ferromagnetic and antiferromagnetic interactions. The M T values at 5 K are expected for two non interacting S = 3/2 and S = 2 subunits for complex 2 1 and 2 2, respectively, with g slightl y less than 2 (spin only, g = 2, 2 × M T S=3/2 = 3.75 cm 3 K mol 1 , 2 × M T S=2 = 6 cm 3 K mol 1 ). In both cases, t he decrease of M T at low temperature is also attributed to intermolecular interactions and zero field splitting (ZFS) effects. For complex es 2 3 a nd 2 4 , as expected from their structural similarity, their magnetic properties are nearly identical , as can be seen from the plot M T vs. T in Figure 2 11 . The M T value of complex 2 3 increases from 48.25 cm 3 K mol 1 at 300 K to a plateau value of 76.55 cm 3 K mol 1 at 20 K, and then decreases slightly to 70.62 cm 3 K mol 1 at 5 K. Complex 2 4 exhibits a nearly superimposable profile with a M T value of 46.34 cm 3 K mol 1 at 300 K, which increases to 76.57 cm 3 K mol 1 at 25 K before slightly decreasing to 70.97 cm 3 K mol 1 at 5 K. The 300 K values are much larger than the spin only ( g = 2) value for twelve Mn III atoms ( M T = 36 cm 3 K mol 1 ), and the peak values at low T are as expected for four non interacting S = 6 units with g slightly less than 2.0 (spin only , g = 2, 4 × M T S=6 = 84 cm 3 K mol 1 ). The decrease of M T below 20 K can be attributed to zero field splitting (ZFS), Zeeman effects from the applied field, and weak intermolecular interactions. The overall M T v s . T profile of 2 3 and 2 4 is extremel y similar to that of complex 2 ( S = 6), indicating each of the four Mn 3 units of 2 3 and 2 6 also to be ferromagnetically coupled with S = 6 ground states. Exchange interaction parameters J ij between Mn i Mn j pairs could be determined when the data in each

PAGE 69

69 c omplex were fit to the theoretical M T vs . T expression. For complex es 2 1 and 2 2 , the theoretical M T is the sum of M T f or two independent Mn 3 subunits, assuming the interaction through the bridging pdpd 2 is insignificant at temperatures above 5 K . Similarly, the theoretical M T for com plex es 2 3 and 2 4 is the sum of M T of four Mn 3 subunits. We used a Mn III 2 Mn II isosceles triangle model for each subunit of complex 2 1 and a Mn III 2 Mn III isosceles triangle model for each subunit of complex es 2 2 , 2 3 , and 2 4 . The models are based on bon d distances and angles in those Mn 3 subunits; in any case, even crystallographically symmetric [Mn 3 O] triangles are known to have magnetic Jahn Teller distortion, 180 therefore an isosceles model will give a higher quality fit. There are two exchange parameters J in an isosceles triangle, e.g. Mn1Mn2Mn2 ' ; the HDVV spin Hamiltonian is given by equation 2 5 = 2 J ( 1 · 2 + 1 · ) 2 J 2 · ( 2 5) where J = J 12 =J 12 ' , J ' =J 22 ' . Using A = 2 + 2 ' and T = A + 1 in the Kambe method, 181 where T is the total spin of the Mn 3 , the spin Hamiltonian becomes = J ( T 2 A 2 1 2 ) J ' ( A 2 2 2 2 ' 2 ) ( 2 6) The eigenvalues of the spin Hamiltonian are given by equation 2 7 , E( S T , S A ) = J [ S T ( S T +1) S A ( S A +1)] J ' [ S A ( S A +1)] ( 2 7) where E( S T , S A ) is the energy of state S T arising from S A . For complex 2 1 , S 2 = S 2 ' =2, S 1 = 5/2, S T ranges from 1/2 to 13/2. For complexes 2 2 , 2 3 , and 2 4 , S 2 = S 2 ' = S 1 = 2, S T ranges from 0 to 6. A theoretical M vs. T expression was derived from the use of the Van Vleck equation 182 and was modified to include temperature i ndependent paramagnetism (TIP), which was ke pt constant at 600 × 10 6 cm 3 K mol 1 . Good fits of M T vs. T data for all four complexes were obtained and are shown as solid lines in Figure s 2 10 and 2 11 . The goodness of the fit s is

PAGE 70

70 evident from the low statistical error, R 2 > 0.99. The data at low te mperature s (e.g. below 20K) were omitted because they are affected by factors not included in equation 2 5. The global minimum in the fit s for 2 1 and 2 2 was proven in the error surface of the J vs. J ' fit ( Figure s 2 12 and 2 13 ). For complexes 2 3 and 2 4 , the error surface plots of the J vs. J' fit show two comparable minimum ( Figure s 2 14 and 2 15 ). The fit parameters given in Table 2 9 are associated with the low est root mean square error . The presence of both ferromagnetic and antiferromagnetic inte ractions in complex 2 1 and 2 2 can be explained when considering bond angles between Mn ions in the triangles. In both complexes, the Mn 3 O Mn angle between two Mn III ions bridged by oximate groups is much smaller than the angles between Mn ions bridged by acetate groups. In addition, oximate groups have a twist, which also promotes ferromagnetic interactions. 183,184 The small Mn 3 O Mn angle when Mn ions are bridged by oximate group in complex 2 2 leads to the strong ferromagnetic interacti on and high J ' value. For complex es 2 2 , 2 3 , and 2 4 , s ince the re are at least two Mn 3 subunits crystallographically inequivalent, the J and J ' must be taken as average values. Interestingly, J and J ' values of complex es 2 3 and 2 4 are similar to those f ound for 3 and analogs with other carboxylates ( J = +12.1 to +18.6 cm 1 and J ' = +1.5 to +6.7 cm 1 ). 171 To further characterize the spin states of these molecules, energy ladder plots for the Mn 3 subunits were determined ( Figure s 2 16, 2 17, 2 18, and 2 19 ). For complex es 2 1 and 2 2 , the ground state and the first excited state spin s are S gs = 3/2, S es = 5/2 and S gs = 2, S es = 3, w here the energy differences between the first excited state and the ground state are 49.7 cm 1 and 29.6 cm 1 , respectively. Both complex es 2 3 and 2 4 have an S = 6 ground st ate and S = 5 first excited state. T he first excited state lies 79.5cm 1 above the ground state in both cases . It is safe to conclude that each of these complexes has a well isolated ground state .

PAGE 71

71 2.3.3.2 Magnetization versus DC magnetic field studies To confirm the ground state spi n and determine the magnitude of the zero field splitting parameter D , magnetization ( M ) data were collected in the 0.1 7 T and 1.8 10 K ranges, and are plotted as M / N B vs. H / T in Figure s 2 20, 2 21, 2 22, and 2 23, where N number and B is the Bohr magneton. For complexes populating only the ground state and possessin g no axial zero field splitting (ZFS) , i.e. D = 0, the magnetization versus field plot follows the Brillouin function and the isofield lines superimpose and saturate at a va lue of gS . However, the experimental data for all complexes clearly do not superimpose, indicating magnetic anisotropy in the ground state. The data were fit, using the program MAGNET, 176 by diagonalization of the spin Hamiltonian matrix assuming only the ground state is populated, incorporating axial anisotropy and Zeeman terms, and employing a full powder average. 185,186 The spin Hamiltonian is given by equation 2 8 , = z 2 + g B 0 · H ( 2 8) where D is the axial ZFS parameter, 0 is the vacuum permeability, and H is the a pplied field. Interaction between [Mn 3 ] units through pdpd 2 group in all complexes is negligible in these studies . The fit parameters are shown in Table 2 10 . The goodness of the fit is evident from the low statistical error, R 2 > 0.99. Alternative fits w ith different S values were discarded because of unreasonable g values. For all complexes, fits can also be obtained for positive values of D . To determine the global minimum in each fit , we established error surfaces for the D vs. g fits by using the prog ram GRID, 176 which calculated the root mean square errors between the experimental M/N B data and those calculated for various combinations of D and g . In a ll cases, the error surface displays a double minimum with positive and negative D values. The error

PAGE 72

72 values of the two minimum are quite comparable in complex es 2 1 and 2 2 ; however the error minimum associated with the negative D is harder , lower, and lik ely more reliable. For complexes 2 3 and 2 4 , the error minimum with negative D is clearly harder and has significantly smaller error value than the one with positive D ( Figure s 2 24, 2 25, 2 26, and 2 27 ) . Considering the Jahn Teller axes in 2 1 and 2 2 , the absolute values of ZFS parameter D are relatively large and might not realistic. In fact, the simplistic model employed does not allow for an accurate determination of D . M any factors such as intermolecular interactions, disorder s which cause a distr ibution in molecular environments, the presence of weak exchange interaction , and the likely population of excited states in the temperature range studied are all ignored in this model . Interestingly, several Mn 3 molecules with oximate ligands that have an S = 2 ground state and comparably large D values ( D ranges from 2.33 to 3.77 cm 1 ) were reported in the literature . 187 For complex es 2 3 and 2 4 , the g and D values obtained are very similar to complex 3 ( S = 6, D = 0.34 cm 1 , and g = 1.92). 171 Thus, the combined dc data complement the structural data in supporting the conclusion that complex es 2 3 and 2 4 are tetramer s of four S = 6 Mn 3 units like that in 3 , and that they must be interacting with each other only ver y weakly to affect the above fits which assume non interacting Mn 3 subunits. 2.3.3.3 Alternating current magnetic susceptibility studies Alternating current (ac) magnetic susceptibility measurements were performed in the 1.8 15 K temperature range in a 3 .5 G ac field oscillating at 5 15 00 Hz. In the ac susceptibility experiment, the ac magnetic field is oscillating at a particular frequency and a peak in the out of phase M versus T plot is observed when the magnetic moment of the molecule cannot relax (reorient) fast enough to keep in phase with the oscillating field. The plots of in phase ( M ,

PAGE 73

73 plotted as M T ) and out of phase ( M ) ac susceptibility data are shown in Figure s 2 28, 2 29, 2 30, and 2 31. The in phase M T values above approximately 7 K are almost constant . At lower temperature, M T drops due to the effect of zero field splitting and intermolecular interactions . For SMMs, the field dependent dropping of M T is due to slow relaxation with a concomitant increase in the out of phase M signal. Extrapolating the in phase ac data to 0 K allows f o r confirm ation of the ground state of the molecule. For complex es 2 1 and 2 2 , extrapolation of the data above 7 K to 0 K gives values of 3.6 cm 3 K mol 1 and 5.3 cm 3 K mol 1 , respectively, which correspond to the sum M T of two independent S = 3/2 and S = 2 Mn 3 ( spin only, g = 2, 2 × M T S =3/2 = 3.75 cm 3 K mol 1 , 2 × M T S =2 = 6 cm 3 K mol 1 ), with g slightly smaller than 2. Unfortunately, no out of phase signals were observed indicating that both complexes 2 1 and 2 2 are not single molecule magnets. For complex es 2 3 and 2 4 , extrapolation of the data above 7 K to 0 K leads to a value of 76 cm 3 K mol 1 , which correspond s to four non interacting S = 6 Mn 3 with a g value slightly less than 2 ( g = 2, 4 × M T S =6 = 84.00 cm 3 K mol 1 ). This is consistent with the result s obtained from the dc magnetization studies. The out of phase M signals of complex es 2 3 and 2 4 that are tails of peaks lying below 1.8 K were observed at < 3 K suggesting 2 3 and 2 4 might be a tetramer of Mn 3 SMMs. For SMMs that only exhibit tails of the ac out of phase peak s, t he effective energy barrier can be roughly estimated based on the relation: ln ( / )= ln(2 0 ) + U eff / kT (2 9) where is the frequency of the oscillating field, and k is the Boltzmann constant. U eff / k and ln(2 0 ) are the slope and the intercept of the line ln( ) vs. 1/ T . Figure 2 32 shows the plot of ln( ) vs. 1/T and the linear regression fits for complex 2 3 . U eff and 0 values are collected in T able 2 11. It is noted that U eff is an intrinsic characteristic of an SMM and not depend ent on

PAGE 74

74 the frequency of the applied field. The different values of U eff in T able 2 11 (8.7 13.3 cm 1 ) present a rough view of the energy barrier of the complex. A more precise value of U eff requires a magnetization decay study at temperatures < 1.8 K . 2.3.3.4 Magnetization decay studies To achieve a deeper insight into the magnetization relaxation dynamics of 2 3 , DC relaxat ion decay measurements were performed. A large magnetic field was applied to the crystal at ~ 5 K to saturate the magnetization in one direction, and the temperature decreased to a chosen value in the 1 0.03 K range. The field was then set to 0 T and the magnetization in zero field was measured as a function of time (Figure 2 3 3 ). From these magnetization decay data, relaxation times could be extracted to construct an Arrhenius plot (Figure 2 3 4 ). The relaxation 3 K (quantum regime). Above 0.3 K, the magnetization relaxation rate (1/ ) obeys the Arrhenius equation , which is given in equation s 2 1 0 and 2 11 . (1/ ) = (1/ 0 )exp( U eff /kT ) ( 2 1 0 ) ln(1/ ) = ln(1/ 0 ) U eff /kT ( 2 1 1 ) where U eff is the effective energy barrier and k is the Boltzmann constant. The fit to the data above 0.3 K (dashed line in Figure 2 34 ) gave 0 = 5 × 10 9 s and U eff = 12.2 K (8.5 cm 1 ) . The effective barrier observed is smaller than the theoretical upper limit ( U = S 2 | D | = 10.8 cm 1 = 15.5 K) calculated using the D value obtained from the powder dc magnetization measurements, as expected for QTM via a thermally (phonon) assisted pathway involving higher energ y m S levels. At temperature below 0.3 K, the magnetization relaxation is only between the lowest energy m S = ±6 levels, indicating pure ground state QTM.

PAGE 75

75 2.3.3. 5 Magnetization versus DC field hysteresis loops The ac out of phase signals themselves do not p rove the presence of SMM behavior , 188,189 therefore, dc magnetization vs. field scans on single crystals of 2 3 ·xCH 2 C l 2 were carried out on a micro SQUID. 177 Hysteresis loops were observed below ~1.0 K (Figure 2 35 ) whose coe rcivities increase with decreasing temperature and increasing field sweep rate, as expected for SMMs. The blocking temperature is 1.0 K, above which there is no hysteresis because the spin relaxes faster than the time scale of the hysteresis loop measurem ent. We can thus conclude that each of the four Mn 3 units in 2 3 ·xCH 2 Cl 2 is an SMM, as is the corresponding Mn 3 complex 3 that they closely resemble in both structure and magnetic properties. We now address whether the four Mn 3 SMM units in 2 3 ·xCH 2 Cl 2 are weakly interacting with each oth er. The answer is clear because the hysteresis loops show an exchange bias of the QTM steps. The first step in the hysteresis loop of an SMM on scanning from negative to positive fields is normally at zero field. This is where the M S levels on either side of the anisotropy barrier are in resonance and QTM can occur, reversing the orientation of the magnetization vector. The presence of an AF exchange coupled neighbor provides a bias field that shifts the resonan t tunnelin g (QTM step) to a new position before zero field. This was first seen for the hydrogen bonded [Mn 4 ] 2 dimer of S = 9 / 2 SMMs, 117 and a related explanation can be provided for 2 3 , except that each Mn 3 SMM is now exchange coupled to two neighboring Mn 3 SMMs through the pdpd 2 groups. The loops of Fig ure 2 35 clearly establish weak AF interactions between the Mn 3 subunits of 2 3 ·xCH 2 Cl 2 . To obtain a simplified spin Hamiltonian describing the [Mn 3 ] 4 tetramer, each Mn 3 SMM can be modeled as a giant spin of S = 6 with Ising like anisotropy. The spin Ha miltonian of each Mn 3 is given by equation 2 12 :

PAGE 76

76 i = D z,i 2 + trans,i + g B 0 z i . H z ( 2 12 ) where i = 1 4 (referring to the four Mn 3 SMMs of the tetramer). The general Hamiltonian for the tetramer is as follows: = i 2 J ij i j ( 2 1 3 ) where J ij are superexchange parameters between subunits i and j, S i = S j = 6, assuming superexchange interactions are isotropic ( J z = J xy ). It is worth noting that the second t erm in equation 2 13 ( 2 J ij i j ) is the standard chemistry Hamiltonian for the coupling between the spin operators i and j . Several physics colleagues prefer to use a different convention , J ij i j , thus J phys = 2 J chem . In the pioneer work of SMM supramo lecular aggregates by Wernsdor fer et al., the coupling parameter J between two [Mn 4 ] units was reported in physics convention. 190 For our case, the value of J will be reported in both conventions to relieve the confusion. Since 2 3 is a rectangular [Mn 3 ] 4 aggregate, there should be two different in ter Mn 3 interactions, J 1 and J 2 , which are likely to be comparable but not identical in magnitude; the diagonal interaction should be much weaker and can be ignored. Therefore, the spin Hamiltonian is given in equation 2 14 = 1 + 2 + 3 + 4 2 J 1 ( 1 2 + 3 4 ) 2 J 2 ( 1 4 + 2 3 ) ( 2 14 ) Tunneling among the (2S+1) 4 = 28561 energy states is allowed by the small transverse anisotropy trans,i and the transverse coupling terms containing xi and yi operators. The energy states of [Mn 3 ] 4 can be calculated by exact diagonalization but it requires a large amount of calculation. A more simple approach can be obtained by neglecting all transverse anisotropy terms. The spin Hamiltonian in equation 2 14 can be re written as 1 + 2 + 3 + 4 2 J 1 ( z 1 z 2 + z 3 z 4 ) 2 J 2 ( z 1 z 4 + z 2 z 3 ) ( 2 1 5 ) Since J 1 J 2 , we can assume J 1 = J 2 = J . The eigenvalue s of equation 2 15 is given as follows

PAGE 77

77 E = D (m 1 2 + m 2 2 + m 3 2 + m 4 2 ) g B 0 H z (m 1 + m 2 + m 3 + m 4 ) 2 J (m 1 m 2 + m 3 m 4 + m 1 m 4 + m 2 m 3 ) ( 2 1 6 ) A spin state energy vs. applied magnetic field plot based on equation 2 16 showing the avoided level crossings is provided in Figure 2 36 , simulated with J 1 = J 2 = 0.0 1 0 K = 0.0070 cm 1 ( J 1 = J 2 = 0.02 0 K = 0.014 cm 1 in physics convention) and D = 0.31 cm 1 , which gives predicted QTM steps at the experimentally observed position . As the field is scanned from 1 T, where the four Mn 3 magnetization vectors will have all been polarized into the M S = 6 orientation, the first step corresponds t o tunneling of the Mn 3 vector from the M S = 6 to the M S = +6 state; this occurs at 0.18 T, which thus represents the total bias field from two M S = 6 neighbors. In the format (M 1 , M 2 , M 3 , M 4 ), where i = 1 4 are the four Mn 3 SMMs within 2 3 , the 0.18 T step corresponds to the ( 6, 6, 6, 6) to ( 6, +6, 6, 6) tunneling transition. The second step at zero field is assignable to tunneling of molecules with 6, +6 neighbors, yielding a zero bias (if J 1 = J 2 , or a small bias related to | J 1 J 2 | if J 1 J 2 ). The step at zero field is the ( 6, +6, 6, 6) to ( 6, +6, +6, 6) tunneling transition, followed by relaxation to the ( 6, +6, 6, +6) ground state. Further, if enough vectors have tunneled in the first two steps to +6, then a step is expected for Mn 3 vectors tunneling in the presence of a +6, +6 bias field. This should occur at +0.18 T, and is indeed seen in Figure 2 35 (bottom) at 0.001 T/s scan rate; at this slow scan rate, enough spins have had time to reverse in the first two steps to allow th e +6, +6 situation to be possible. At faster scan rates, this step disappears, and even the zero field step ( 6, +6 bias field) becomes smaller since the tunneling probability decreases with increasing scan rate, i.e. fewer molecules are in the resonance s ituation long enough to tunnel. At fast scan rates, at least two new steps appear in the 0.3 0.6 T range, involving also M S = ± 5 levels. There are several level crossings (marked * in Fig ure 2 36 , bottom) that require multiple spin flips at the same tim e.

PAGE 78

78 Their tunneling probability is very small and can be neglected. The step pattern of the hysteresis loops leads to the conclusion that 2 3 is a supramolecular aggregate of four weakly interacting Mn 3 SMMs . 2.3.3. 6 High frequency electron paramagnetic re sonance (HF EPR) s tudies High frequency electron paramagnetic resonance (HF EPR) data were collected for a finely ground sample of 2 3 · x CH 2 Cl 2 that was incorporated into a KBr pellet in order to avoid field alignment of the micro crystallites within the p owder. Measurements have been performed at high frequencies of 52.4 422.4 GHz in the temperature range of 2.5 20 K. HF EPR spectra were collected using a transmission probe in which microwaves are propagated through cylindrical light pipes. High freque ncy microwaves were generated by a phase locked Virginia Diodes sol id state source operating at 13 ±1 GHz, followed by a chain of multipliers and amplifiers. Microwave detection was provided by a bolometer. High magnetic fields were provided by a 17 T super conducting magnet. 191 The t emperature dependence of EPR spectra collected at 217.6 GHz is presented in Figure 2 37 sign) at 3.2 and 5.5 T have been well characterized and are attributed to paramagnetic oxygen impurities trapped in the KBr pellet. The sharp feature observed around the g sign, corresponds to an impurity phase within the powder, po ssibly containing isotropic Mn II ; this is often found to be the case in Mn containing polynuclear clusters. The spectra were recorded in field derivative mode ( dI/dB , where I denotes the absorption intensity), making it relatively easy to determine which p eaks correspond to which components of the spectrum. The inset to Figure 2 37 (a) displays a trivial example of a T = 0 derivative mode powder spectrum for a biaxial system for which three features are observed: the z component corresponds to the onset of a bsorption and appears as a peak in the derivative; the x component corresponds to the cessation

PAGE 79

79 of absorption and thus appears as a dip in the derivative; meanwhile, the y component occurs at the maximum in absorption and therefore looks like the derivativ e of z component. For an easy axis ( D < 0) system at low temperatures, one expects that the parallel ( B || z ) components will be at the low field side of the g = 2.00, while the stronger perpendicular components should occur on the high field, as is clea rly the case in Fig ure 2 37 . Examination of the resonances at the low field side of the g = 2.00 in Fig ure 2 37 (a) indicates that they are indeed peaks, confirming their assignment to the parallel part of the spectrum. For an axial system, the parallel ( B || z ) component of the powder spectrum typically extends about twice as far from the g = 2.00 position, compared to the perpendicular ( ) component, with the extent of the spread being directly proportional to the magnitude of the axial D parameter. On this basis, the multiple low field features can be attributed to resolved parallel excitations. Careful inspection of the 20 K spectrum reveals six such peaks ( Z 1 to Z 6 ) on the low field side of g = 2. This is an indication of ground state S = 6 for the Mn 3 triangle units. The peaks correspond to the following fine structure transitions within the ground S = 6 spin multiplet: m s = 5 (Z 6 ), m s = 4 (Z 5 ), m s = 1 ). The fact that the spectral weight associated with the parallel spectrum shifts to the low field m s = 5 (Z 6 ) transition upon cooling provides confirmation of the negative sign of D , i.e., th e m s = 6 state is lowest in energy when B || z. On the other hand, the fine structure transitions within the perpendicular spectrum were not clearly resolved. We cannot assign a definite transition to each feature in perpendicular spectrum. There are sev eral possible reasons for this. First, the relevant peaks should have been much more closely spaced (i.e., two perpendicular components, each with half the separation of the parallel spectrum) and, therefore, more difficult to resolve clearly. Nevertheless , this alone cannot fully account for the observed spectrum. A likely explanation is molecular disorder and/or

PAGE 80

80 strains in the sample that affect the transverse zero field splitting parameters (e.g., E ). This type of disorder is known to occur in the extens ively studied Mn 12 acetate SMM , 192 194 where such strains have been shown to wash out the perpendicular spectrum without affecting the parallel spectrum significantly. It is also well known that the crystallites in powdered SMM samples are subject to substantial torques that can cause minor reorientations of individual particles, even in well constrained samples. Again, this effect can lead to a substantial broadening of the perpendicular spectrum and almost no effe ct on the parallel component. All of these influences can be taken into consideration when simulating spectra using a program such as EasySpin . We note that the temperature dependent studies (Figure 2 37 ) do not provide any strong indications that there ar e low lying states with significant spin values (i.e., no additional peaks emerged upon raising the temperature that cannot be accounted for via simulations that assume an isolated S = 6 ground state). Figure 2 3 7(b) displays a simulation, obtained by emp loying the following Hamiltonian: (2 17) where , , , and are spin operators, is the applied magnetic field vector, is the Landé g tensor, is the Bohr magneton, D and E are the second order axial and rhombic zero field splitting parameter respectively, and the final term represents axial fourth order zero field splitting . 195 The parameters used for the simulation were S = 6, D = 0.33 cm 1 , | E | = 0.03 cm 1 , = 8 × 10 5 cm 1 , and g x = g y = g z = 2.00. As can be seen, the relevant parallel and perpendicular portions of the simulations agree with the experiments in every respect, and the obtained D value is fully consistent with that deduced from reduced magnetization studies. We employed Gaussian distributions of the D and E parameters ( ful l width at half maximum ~ 0.027 cm 1 ) in these simulations.

PAGE 81

81 In order to obtain tighter constraints on the spin Hamiltonian parameter for complex 2 3 ·x CH 2 Cl 2 , frequency dependent powder EPR experiments were carried out with frequencies in the range 52.4 422.4 GHz. The positions of the parallel component peaks were then plotted versus frequency, as seen in Figure 2 38 , and the data simulated with the Hamiltonian of eq uation 2 1 7 . The solid lines in Figure 2 38 represent the best simulation of the frequency dependent data using the same parameters used to simulate Figure 2 37 . We note that the fourth order term is absolutely essential to the simulations, accounting for the uneven spacing of the resonances. The obtained parameters agree well with those extrac ted from fits of reduced magnetization data. In addition, it is possible to calculate the kinetic barrier to magnetization relaxation of 11.88 cm 1 from these parameters ( U = DS 2 ) . Again, this value is slightly larger than that deduced from magnetization d ecay studies, with the difference attributable to the tunneling near the top of the barrier. We conclude this section by discussing the interaction between the four Mn 3 units. EPR spectra of the powder sample of 2 3 ·x CH 2 Cl 2 suggests that the couplings bet ween Mn 3 units are too weak to be detected by EPR. This weak coupling conclusion was supported by magnetization versus DC field hysteresis loop measurements. Hysteresis loop data suggest antiferromagnetic coupling of 0.02 0 K = 0.014 cm 1 between the four Mn 3 units which is too small to be detected by EPR. 2. 4 Summary and C onclusions In summary, employment of a newly designed dioxime group has allowed the isolation of two non SMM [Mn 3 ] 2 and two SMM [Mn 3 ] 4 supramolecular aggregates. The formation of 2 1 by a direct approach is interesting, as it shows the power of the method, which has been used for more than two decades to produce almost all known SMMs up to now. Its disadvantage is the major loss of control in making desired supramolecular aggregates of S MM. Complex 2 3 w as

PAGE 82

82 synthesized based on the understanding of 3 with the aim to build a supramolecular aggregate of this building block. Thus, each of the Mn 3 subunits is extremely similar to that of the discrete monomeric 3 , and as a result, the magnetic properties are also nearly identical. Hysteresis loops of complex 2 3 at very low T established that the four Mn 3 SMMs are each weakly coupled to two neighbors, and this interaction is manifested as an exchange bias of the QTM steps, whose magnitude de pends on the spin alignments at the two nearest neighbors. The inter Mn 3 interaction is mainly via superexchange mechanism through the bridging ligands since discrete molecules of complex 3 show no exchange bias of its QTM steps from similar intermolecular dipolar interactions. Unfortunately, the QTM steps are relatively broad, which could be due to the following reasons 196 : (i) the two Mn 12 cations in the asymmetric unit of complex 2 3 have different orientations, and the four Mn 3 planes within each cation are not coplanar. The applied field will thus be at a range of angles to the easy ( z ) axes of the eight Mn 3 , leading to step broadening; and (ii) J 1 J 2 , and non zero diagonal interactions within [Mn 3 ] 4 or interactions between separate units will give a range of bias fields for a given ±6, ±6 situation. Furthermore, the inter Mn 3 intera c tion through pdpd 2 group is too weak to be detected by HF EPR d ue to the superexchange coupling through three sp 3 C atoms. The success of the carboxylate substitution on 2 3 to form 2 4 can lead to the isolation of new [Mn 3 ] 4 molecules with different carboxylate ligands designed for material applications. 139 The covalent linkage within the [Mn 3 ] 4 supramolecular assemblies also allows for retention of the str ucture on dissolution and offer promise for studies in fluid and frozen solution for the first time of an exchange biased system. In conclusion, complex 2 3 confirms the feasibility of covalently connecting multiple Mn 3 SMMs to give a discrete supramolecu of SMMs with only weak coupling between them. This should lead to them also being quantum mechanically coupled, as found for

PAGE 83

83 [Mn 4 ] 2 dimers, and represents a step towards developing a multi qubit system based on SMMs. Additional synthetic effo rts to produce other supramolecular aggregates of weakly coupled SMMs will be presented in the next chapters. Table 2 1. Crystallographic data for complexes 2 1 · xCH 3 OH · yCH 3 CN, 2 2 · xC 4 H 10 O · yCH 2 Cl 2 , 2 3 · xCH 2 Cl 2 , and 2 4 · xCH 2 Cl 2 . Parameters 2 1 · xCH 3 OH · yCH 3 C N 2 2 · xC 4 H 10 O · yCH 2 Cl 2 2 3 · xCH 2 Cl 2 2 4 · xCH 2 Cl 2 Formula a C 65 H 86 Mn 6 N 10 O 26 C 82 H 98 Cl 10 Mn 6 N 10 O 32 C 150.25 H 144 Cl 4.50 Mn 12 N 24 O 56 C 206 H 252 Cl 44 Mn 12 N 24 O 48 FW, g/mol a 1753.08 2419.84 4000.70 6051.40 Crystal system Monoclinic Triclinic Monoclinic Monoclinic S pace group C2/c P2 1 /c C2/c a, Å 16.0232(17) 16.7797(18) 34.190(4) 42.112(3) b, Å 41.093(4) 16.9615(18) 32.890(4) 26.4980(16) c, Å 13.0506(14) 20.538(4) 44.013(5) 30.5334(18) , ° 90 93.849(3) 90 90 ,° 110.908(2) 105.068(3) 110.320(3) 127.707(1) ,° 90 116.263(2) 90 90 V, Å 3 8027.2(15) 4948.4(11) 46413(9) 26956(3) Z 4 2 8 4 T, K 100(2) 100(2) 100(2) 100(2) , Å b 0.71073 0.71073 0.71073 0.71073 cal, Mg/m 3 1.451 1.624 1.145 1.491 , mm 1 0.995 1.097 0.748 1.048 R 1 c,d 0.0460 0.0732 0.074 0 0.0505 wR 2 e 0.1276 0.2028 0.1617 0.1218 a Including solvate molecules. b Graphite monochromator. c I>2 (I). d R 1 = (||F o | |F c ||) / |F o |. e wR 2 = [ w(F o 2 F c 2 ) 2 ] / w ( F o 2 ) 2 ]] 1/2 where w = 1/[ 2 (F o 2 )+(m × p) 2 + n × p], p = [max(F o 2 ,0)+ 2 × F c 2 ]/3, m & n are constants.

PAGE 84

84 Table 2 2. Selected interatomic distances (Ã…) and angles ( o ) for complexes 2 1 and 2 2 . Complex 2 1 Complex 2 2 Bond d istances Mn1 Mn2 3.1082(6) Mn1 Mn2 3.1610(15) Mn1 Mn3 3.3940(2) Mn1 Mn3 3.2913(19) Mn2 Mn3 3.3730(3) Mn2 Mn3 3.2922(15) Mn1 O1 1.8262(18) Mn4 Mn5 3.1641(21) Mn1 O4 2.1910(19) Mn4 Mn6 3.2817(16) Mn1 O3 2.2134(19) Mn5 Mn6 3.2864(18) Mn2 O1 1.8238(18) Mn1 O3 2.154(5) Mn2 O10 2.163(2) Mn1 O4 2.185(5) Mn2 O2 2.3115(19) Mn2 O10 2.161(6) Mn3 O1 2.0500(18) Mn2 O2 2.221(5) Mn3 O7 2.182(6) Mn3 O9 2.194(5) Mn4 O17 2.167(6) Mn4 O14 2.200(5) Mn5 O19 2.150(5) Mn5 O13 2.180(5) Mn6 O22 2.181(5) Mn6 O16 2.215(5) Bond a ngles Mn2 O1 Mn1 116.76(10) Mn1 O1 Mn2 113.6(2) Mn2 O1 Mn3 121.02(9) Mn1 O1 Mn3 123.2(2) Mn1 O1 Mn3 122.20(9) Mn2 O1 Mn3 123.2(3) Mn4 O12 Mn5 114.1(2) Mn4 O12 Mn6 123.3(3) Mn5 O 12 Mn6 122.6(3)

PAGE 85

85 Table 2 3. Selected interatomic distances (Ã…) and angles ( o ) for complexes 2 3 and 2 4 . Complex 2 3 Complex 2 4 Bond distances Mn1 Mn2 3.212(2) Mn4 O12 2.179(6) Mn1 Mn2 3.1982(6) Mn5 O2 1.8706(19) Mn1 Mn3 3.1842(19) Mn4 O23 2.235(6) Mn1 Mn3 3.1970(6) Mn5 O8 2.1741(19) Mn2 Mn3 3.2012(19) Mn5 O9 2.174(5) Mn2 Mn3 3.2024(7) Mn5 O16 2.184(2) Mn4 Mn5 3.1887(18) Mn5 O21 2.249(6) Mn4 Mn5 3.1990(6) Mn6 O2 1.8758(19) Mn4 Mn6 3.1726(18) Mn6 O20 2.165(6) Mn4 Mn6 3.1971(7 ) Mn6 O6 2.1743(19) Mn5 Mn6 3.216(2) Mn6 O8 2.195(5) Mn5 Mn6 3.2055(6) Mn6 O20 2.2203(19) Mn7 Mn8 3.180(2) Mn7 O5 2.171(6) Mn1 O1 1.8716(19) Mn7 Mn9 3.206(2) Mn7 O25 2.198(7) Mn1 O4 2.1705(19) Mn8 Mn9 3.205(2) Mn8 O7 2.151(5) Mn1 O11 2.20 4(2) Mn10 Mn11 3.1850(18) Mn8 O29 2.191(6) Mn2 O1 1.8809(19) Mn10 Mn12 3.197(2) Mn9 O6 2.199(6) Mn2 O5 2.1969(19) Mn11 Mn12 3.218(2) Mn9 O28 2.213(6) Mn2 O10 2.211(2) Mn1 O1 2.156(5) Mn10 O3 2.197(6) Mn3 O1 1.8668(19) Mn1 O15 2.214(6) Mn10 O31 2.259(6) Mn3 O3 2.1919(19) Mn2 O11 2.163(6) Mn11 O4 2.162(5) Mn3 O13 2.177(2) Mn2 O14 2.225(6) Mn11 O33 2.258(6) Mn4 O2 1.872(2) Mn3 O10 2.178(5) Mn12 O2 2.176(7) Mn4 O7 2.2071(19) Mn3 O18 2.204(6) Mn12 O36 2.185(6) Mn4 O17 2.186(2) Bond angles Mn2 O37 Mn1 118.3(2) Mn8 O39 Mn7 116.3(3) Mn1 O1 Mn2 116.92(10) Mn4 O2 Mn5 117.46(10) Mn3 O37 Mn1 115.3(3) Mn9 O39 Mn7 118.0(3) Mn2 O1 Mn3 117.41(10) Mn5 O2 Mn6 117.66(10) Mn2 O37 Mn3 118.6(3) Mn9 O39 Mn8 118.7(3) Mn3 O1 Mn1 117.56(10) Mn6 O2 Mn4 117.08(10) Mn4 O38 Mn5 116.5(3) Mn11 O40 Mn10 118.5(3) Mn6 O38 Mn4 116.9(3) Mn10 O40 Mn12 116.6(3) Mn6 O38 Mn5 118.2(3) Mn11 O40 Mn12 119.3(3)

PAGE 86

86 Table 2 4. Displacement of 3 oxide atoms from Mn 3 planes ( Ã… ) and Mn N O Mn torsion an gles ( o ) of complex es 2 3 and 2 4 . Complex 2 3 Complex 2 4 Triangle d ( Ã… ) Torsion a ngles ( o ) Triangle d ( Ã… ) Torsion a ngles ( o ) Mn1Mn2Mn3 0.31 Mn1 N19 O10 Mn3 17.98 Mn1Mn2Mn3 0.31 Mn1 N2 O3 Mn3 14.38 ` Mn2 N1 O1 Mn1 14.49 Mn2 N4 O4 Mn1 10.6 Mn3 N2 1 O11 Mn2 10.90 Mn3 N6 O5 Mn2 13.3 Mn4Mn5Mn6 0.32 Mn4 N17 O9 Mn5 19.82 Mn4Mn5Mn6 0.31 Mn5 N10 O7 Mn4 18.3 Mn5 N15 O8 Mn6 10.75 Mn6 N12 O8 Mn5 12.6 Mn6 N28 O12 Mn4 18.28 Mn4 N8 O6 Mn6 10.9 Mn7Mn8Mn9 0.29 Mn7 N13 O7 Mn8 15.88 Mn8 N11 O6 Mn9 9.97 Mn9 N9 O5 Mn7 15.73 Mn10Mn11Mn12 0.26 Mn10 N7 O4 Mn11 15.57 Mn11 N3 O2 Mn12 8.09 Mn12 N5 O3 Mn10 12.59

PAGE 87

87 Table 2 5. Bond valence sum calculation for Mn a and selected O b atoms of complex 2 1 . Mn II Mn III Mn IV Assignment BVS Assignment Mn1 3.29 3.07 3.12 Mn III O1 2.17 O 2 Mn2 3.21 3.00 3.04 Mn III O12 1.31 OH Mn3 2.25 2.10 2.14 Mn II a The bold value is the one closest to the charge for which it was calculated. The oxidation state of a particular atom can be taken as the nearest whole number to the bold value. b A BVS in the ~ 1.8 2. 2 , ~1.0 1. 4 , and ~0.2 0.4 ranges for an O atom is indicative of non , single , and double protonation, respectively, but can be altered somewhat by hydrogen bonding. Table 2 6. B ond valence sum calculation for Mn a and selected O b atoms of complex 2 2 . Mn II Mn III Mn IV Assignment BVS Assignment Mn1 3.29 3.07 3.12 Mn III O1 2.17 O 2 Mn2 3.29 3.07 3.12 Mn III O12 2.19 O 2 Mn3 3.29 3.06 3.13 Mn III Mn4 3.26 3.03 3.09 Mn III Mn5 3.25 3.03 3.09 Mn III Mn6 3.30 3.06 3.14 Mn III a See footnote a of table 2 5. b See footnote b of table 2 5. Table 2 7. Bond valence sum calculation for Mn a and selected O b atoms of complex 2 3 . Mn II Mn III Mn IV Assignment BVS Assignment Mn 1 3.31 3.10 3.14 Mn III O37 2.20 O 2 Mn2 3.26 3.04 3.09 Mn III O38 2.21 O 2 Mn3 3.30 3.07 3.13 Mn III O39 2.22 O 2 Mn4 3.35 3.13 3.17 Mn III O40 2.25 O 2 Mn5 3.20 2.99 3.03 Mn III Mn6 3.31 3.10 3.14 Mn III Mn7 3.30 3.09 3.13 Mn III Mn8 3.37 3.12 3 .16 Mn III Mn9 3.28 3.07 3.11 Mn III Mn10 3.28 3.07 3.11 Mn III Mn11 3.35 3.13 3.18 Mn III Mn12 3.43 3.22 3.25 Mn III a See footnote a of table 2 5. b See footnote b of table 2 5.

PAGE 88

88 Table 2 8. Bond valence sum calculation for Mn a and selecte d O b atoms of complex 2 4 . Mn II Mn III Mn IV Assignment BVS Assignment Mn1 3.32 3.11 3.15 Mn III O1 2.19 O 2 Mn2 3.24 3.03 3.07 Mn III O12 2.19 O 2 Mn3 3.26 3.03 3.09 Mn III Mn4 3.28 3.07 3.11 Mn III Mn5 3.31 3.10 3.14 Mn III Mn6 3.26 3.05 3.0 9 Mn III a See footnote a of table 2 5. b See footnote b of table 2 5. Table 2 9. Parameters of the fits to Van Vleck equation . Complex J (cm 1 ) J' (cm 1 ) g 2 1 9.94 ± 0.46 10.42 ± 3.87 1.96 ± 0.04 2 2 4.93 ± 0.11 28.41 ± 1.98 1.99 ± 0.01 2 3 16.78 ± 0.61 1.54 ± 0.74 1.91 ± 0.01 2 4 13.48 ± 0.76 3.78 ± 1.00 1.91 ± 0.01 Table 2 10. Reduced magnetization fit parameters . Complex S g D (cm 1 ) 2 1 3/2 1.95 ± 0.01 3.09 ± 0.12 2 2 2 1.98 ± 0.01 1.61 ± 0.06 2 3 6 1.92 ± 0.01 0.30 ± 0.01 2 4 6 1.89 ± 0.01 0.30 ± 0.01 Table 2 11. Linear regression fit parameters . Frequency U eff (cm 1 ) 0 (s) R 2 1000Hz 13.26 9.57x10 10 0.9779 500Hz 13.63 8.29x10 10 0.9712 250Hz 13.77 7.94x10 10 0.9809 50Hz 10.75 8.820x10 9 0.9977 25Hz 8.72 4.48x10 8 0.9950

PAGE 89

89 Figure 2 1 . Structure s of the cation of complex 3 (top), mpkoH (bottom left) and pdpdH 2 (bottom right). Color code: Mn III green; N blue; O red; C gray .

PAGE 90

90 Figure 2 2. Complete molecular structure (top) and a stereopair (m iddle) of complex 2 1 with H atom s omitted for clarity, ( b ottom) the core emphasizing the corresponding connectivity between Mn 3 subunits, and showing the Mn 3 planes and Jahn Teller axes (green bonds). Color code: Mn III green; Mn II sky blue; O red; N blue ; C grey .

PAGE 91

91 Figure 2 3. S tructure of the cation (top) and a stereopair (middle) of complex 2 2 with H atom s omitted for clarity, ( b ottom) the core emphasizing the corresponding connectivity between Mn 3 subunits, and showing the Mn 3 planes and Jahn Tell er axes (green bonds). Color code: Mn III green; O red; N blue; C grey .

PAGE 92

92 Figure 2 4. (Top) Hydrogen bonds (black dotted bonds) connecting [Mn 3 ] 2 molecules of complex 2 1 3 ] 2 molecules in complex 2 2 .

PAGE 93

93 Figure 2 5. S tructure of the cation (top) and a stereopair (middle) of complex 2 3 with H atom s omitted for clarity, ( b ottom) the core emphasizing the corresponding connectivity between Mn 3 subunits, and showing the Mn 3 planes and Jahn Teller axes (green bonds). Color code: Mn III green; O red; N blue; C grey .

PAGE 94

94 Figure 2 6. S tructure of the cation (top) and a stereopair (middle) of complex 2 4 with H atom s omitted for clarity, (b ottom) the core emphasizing the corresponding connectivity between Mn 3 subunits, and showing the Mn 3 planes and Jahn Teller axes (green bonds). Color code: Mn III green; O red; N blue; C grey .

PAGE 95

95 Figure 2 7 . Packing diagram o f complex 2 4 viewing from c axis (top) and b axis (bottom) .

PAGE 96

96 Figure 2 8 . Encapsulation of two CH 2 Cl 2 molecules inside the Mn 12 cation of complex 2 3 . H atoms have been omitted for clarity .

PAGE 97

97 Figure 2 9. Distances (Ã…) between Mn 3 units in complex 2 3 (top) comparing to those of complex 3 (bottom). pdpd 2 groups in complex 2 3 have been omitted for clarity .

PAGE 98

98 Figure 2 10. M T vs. T for complex es 2 1 (red) and 2 2 (blue). The solid lines are the fits to the data; see Table 2 5 for the fit parameters . Figure 2 11. M T vs. T for complex es 2 3 (pink) and 2 4 (green). The solid lines are the fits to the data; see Table 2 5 for the fit parameters .

PAGE 99

99 Figure 2 12. Two dimensional contour plot of the fitting errors vs. J and J' for complex 2 1 . Figure 2 13. Two dimensional contour plot of the fitting errors vs. J and J' for complex 2 2 . * *

PAGE 100

100 Figu re 2 14. Two dimensional contour plot of the fitting errors vs. J and J' for complex 2 3 . Figure 2 15. Two dimensional contour plot of the fitting errors vs. J and J' for complex 2 4 . * *

PAGE 101

101 Figure 2 16. Energy ladder plot for complex 2 1 . Ground state i s S T = 3 /2. The f irst excited state S = 5/2 lies 49.7 cm 1 above the ground state . Figure 2 17. Energy ladder plot for complex 2 2 . Ground state i s S T = 2. The f irst excited state S = 3 lies 29.6 cm 1 above the ground state .

PAGE 102

102 Figure 2 18. Energy ladder plot for co mplex 2 3 . Ground state i s S T = 6. The f irst excited state S = 5 lies 79.5 cm 1 above the ground stat e. Figure 2 19. Energy ladder plot for complex 2 4 . Ground state i s S T = 6. The f irst excited state S = 5 lies 79.5 cm 1 above the ground state .

PAGE 103

103 Figure 2 20. M / N B (per Mn 3 ) vs. H / T for complex 2 1 . The solid lines are the fits to the data. See T able 2 10 for the fit parameters . Figure 2 21. M / N B (per Mn 3 ) vs. H / T for complex 2 2 . The solid lines are the fits to the data. See T able 2 10 for the fit paramete rs .

PAGE 104

104 Figure 2 22. M / N B (per Mn 3 ) vs. H / T for complex 2 3 . The solid lines are the fits to the data. See T able 2 10 for the fit parameters . Figure 2 23. M / N B (per Mn 3 ) vs. H / T for complex 2 4 . The solid lines are the fits to the data. See T able 2 10 fo r the fit parameters .

PAGE 105

105 Figure 2 24. Two dimensional contour plot of the errors vs. D and g for the fit of 2 1 . Figure 2 25. Two dimensional contour plot of the errors vs. D and g for the fit of 2 2 . * *

PAGE 106

1 06 Figure 2 26. Two dimensional contour plot of the err ors vs. D and g for the fit of 2 3 . Figure 2 27. Two dimensional contour plot of the errors vs. D and g for the fit of 2 4 . * *

PAGE 107

107 Figure 2 28. Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signal s versus temperature for a micro crystalline sample of complex 2 1 in a 3.5G field oscillating at the indicated frequencies .

PAGE 108

108 Figure 2 29. Plot of in phase ( M , as M T ) (top) a nd out of phase ( ) (bottom) ac susceptibility signal s versus temperature for a micro crystalline sample of complex 2 2 in a 3.5G field oscillating at the indicated frequencies .

PAGE 109

109 Figure 2 30. Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signal s versus temperature for a micro crystalline sample of complex 2 3 in a 3.5G field oscillating at the indicated frequencies .

PAGE 110

110 Figure 2 31. Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptib ility signal s versus temperature for a micro crystalline sample of complex 2 4 in a 3.5G field oscillating at the indicated frequencies .

PAGE 111

111 Figure 2 32. Plot of ln( / ) vs. 1/ T for complex 2 3 . The lines are the linear regression fits to the data at the indicated frequencies . Figure 2 33. Magnetization decay measurements of complex 2 3 at zero applied field at several temperatures .

PAGE 112

112 Figure 2 34. Relaxation time ( ) vs. 1/ T using DC magnetization decay data. The dashed line is the fit of the thermally activated region to the Arrhenius equation. See the text for the fit parameters .

PAGE 113

113 Figure 2 35. Magnetization vs. dc field hysteresis loops for a single crystal of 2 3 ·xCH 2 Cl 2 at the indicated temperatures (top) and field scan rates at 0.04 K (bot tom). M is normalized to its saturation value, M S .

PAGE 114

114 Figure 2 36. (Top) Simulation of the plot of spin state energies vs. applied magnetic field for a tetramer of four Mn 3 SMMs, each with S = 6. Red = spin states involving only the M s = ±6 states of the four Mn 3 sub units; Blue = spin states involving both M s = ±6 states and (only) one M s = ±5 state; Black = other states. (Bottom) Spin states involving only the M s = ±6 states. * = multiple spin flips (2, 3 , or 4) at the same time. See t he text for the spin states involved in the three QTM steps , which are at green arrows. * * * * *

PAGE 115

115 Fi gure 2 37. Temperature dependence of HF EPR spectra of complex 2 3 · x CH 2 Cl 2 (experimental (a) and simulated (b)), recorded in field derivative mode at 217.6 GHz in t he temperature range 2.5 20 K collected on a microcrystalline sample restrained in KBr. The features in (a) are labeled according to the scheme described in the main text. The top inset in (a) depicts a typical derivative mode powder spectrum for a biaxi al system, illustrating the lineshapes expected for the x , y , and z components of the spectrum.

PAGE 116

116 Figure 2 3 8 . Easy axis ( z axis) frequency dependent EPR data for complex 2 3 · x CH 2 Cl 2 . The solid lines are a simulation of the data employing the parameters: S = 6, D = 0.33 cm 1 , | E | = 0.03 cm 1 , = 8 × 10 5 cm 1 , and g x = g y = g z = 2.00.

PAGE 117

117 CHAPTER 3 SPHERICAL HOST GUEST SUPRAMOLECULAR AGGREGATES OF SINGLE MOLECULE MAGNETS FROM THE USE OF 1,2 DI(PYRIDIN 2 YL)ETHANE 1,2 DIONE DIOXIME 3.1 Introduction W e initiated a new approach in which SMMs are covalently connected by organic linkers into a discrete supramolecule whose subunits weakly interact with each other. There are several advantages of linking SMMs covalently: i) the number of potential supramolecular by products in a particular reaction is reduced compared to the supramolecules formed by H bonds; ii) the formed supramolecules are still stable upon dissolution in solvents, which allows for further manipulation such as depositio n on surfaces; iii) the ligand structure may allow for prediction of the shape of the forming supramolecules. In fact, there is a very rich chemistry of coordination driven self assembly in which various synthetic methods have been explored to synthesize a wide range of different architectures such as infinite helicates, 44,45 catenates, rotaxanes, cylinders, knots, 47 49 and related species, 54 56 and a variety of two dimensional (2D) molecular ensembles, e.g. molecular dimers, triangles, rectangles, and three dimensi onal (3D) polyhedra and capsules. 27,58 61 Our general aim is to apply principles and methods of supramolecular self assembly in single molecule magnetism in order to obtain unprecedented structures and magnetic beh aviors. In C hapter 2, we built up the strategy to link Mn 3 SMMs into a rectangular [Mn 3 ] 4 supramolecular aggregate by using the dioximate pdpdH 2 . The antiferromagnetic coupling between Mn 3 units through the pdpd 2 groups, however, was found to be very we ak, J 0.0 1 K ( 0.02 K in physics convention) . To increase the strength of the exchange interaction , reducing the separation between oximate groups is a possible approach. E mploying the dioxime , 1,2 di(pyridin 2 yl)ethane 1,2 dione dioxime (dedH 2 ) ( Figur e 3 1 ), of which oxime groups are

PAGE 118

118 directly connected , was attempted . It is worth noting that this dioxime has never been used in inorganic chemistry. In this present work, we report the syntheses and structures of four new tetrahedral [Mn 3 ] 4 supramolecular aggregates from the use of dedH 2 . These supramolecules have different carboxylate ligands, charge s, and e ncapsulated guests . We will also discuss the ma gnetic properties of these supramolecules and the exchange biased interaction resulting from weak coupl ings between Mn 3 units. 3.2 Experimental Section 3.2.1 Syntheses All preparations were performed under aerobic conditions using reagents and solvents as received , unless otherwise stated . [Mn 3 O(O 2 CR) 6 (py) 3 ](ClO 4 )(R = Me( 1 ), Et( 2 )) complexes were synth esized as reported elsewhere. 172,197 The known organic molecule 1,2 di(pyridin 2 yl)ethane 1,2 dione dioxime (dedH 2 ) was prepared based on a reported procedure involving the pyridil and hydroxylamine. 198 Caution! Although no such behavior was observed during the present work, perchlorate compounds are potentially explosive; such compounds should be synthesized and used in small quantities, and treated with utmost care at all ti mes. 3.2.1.1 [Mn III 9 Mn II 3 O 4 (O 2 CMe) 12 (ded) 6 (pyH)](ClO 4 ) 2 (3 1) A brown solution of [Mn 3 O(O 2 CMe) 6 (py) 3 ](ClO 4 ) (0.44 g, 0.5 0 mmol) in 15 m L of MeCN : MeOH (2 : 1 v/v) was treated with dedH 2 (0.36 g, 1.5 mmol). The solution was stirred for 1 h ou r at room te mperature and then evaporated under vacuum to obtain a black powder. The powder was redissolved in 15 m L of DMF/MeCN (2 : 1 v/v) and the resulting solution was layered with Et 2 O . X ray quality crystals of 3 1 formed after 4 days, and these were collected b y filtration, washed with Et 2 O and dried under vacuum. The yield was 30%. Anal.Calcd (found)% for 3 1 · 5H 2 O (Mn 12 O 53 C 101 H 100 N 25 Cl 2 ): C 37.42 (37.12); H 3.11 (3.01); N 10.80

PAGE 119

119 (10.75); Cl 2.19 (2.10). Selected IR data (cm 1 ): 1663 (m), 1600 (s), 1571 (s), 14 59 (s), 1388 (s), 1337 (s), 1252 (w), 1211 (w), 1182 (m), 1147 (m), 1097 (s), 1044 (m), 1026 (m), 943 (w), 778 (m), 745 (w), 709 (m), 661(s), 622 (m), 558 (m). 3.2.1.2 [Mn III 10 Mn II 2 O 4 (O 2 CEt) 10 (H 2 O) 2 Cl 2 (ded) 6 (py)](ClO 4 ) 2 (3 2). A brown solution of [Mn 3 O(O 2 CEt) 6 (py) 3 ](ClO 4 ) (0.45 g, 0.5 0 mmol) in 15 m L of MeCN was treated with dedH 2 (0.36 g, 1.5 mmol). The solution was stirred for 1 h ou r at room temperature, filtered , and then layered with Et 2 O . X ray quality crystals of 3 2 formed after 4 5 days, and these were collected by filtration, washed with Et 2 O , and dried under vacuum. The yield was 20%. Anal.Calcd (found)% for 3 2 · 3H 2 O (Mn 12 O 49 C 107 H 113 N 25 Cl 4 ): C 38.54 (38.35); H 3.42 (3.22); N 10.50 (10.32). Selected IR data (cm 1 ): 1600 (s), 1572 (s), 1460 (s), 1 391 (s), 1360 (s), 1280 (m), 1213 (w), 1181 (s), 1164 (m), 1092 (s), 944 (w), 884(w), 811 (w), 777 (m), 745 (w), 709 (m), 698 (w), 664(m), 623 (s), 559 (m). 3.2.1.3 [Mn III 12 O 4 (O 2 CMe) 12 (ded) 6 (CH 2 Cl 2 )](I 3 ) 3.5 I 0.5 (3 3) A brown solution of [Mn 3 O(O 2 CMe) 6 (p y) 3 ](ClO 4 ) (0.44 g, 0.5 0 mmol) in 25 m L of CH 2 Cl 2 : MeCN (4 : 1 v/v) was treated with dedH 2 (0.36 g, 1.5 mmol) and I 2 (0.38 g, 1.5 mmol). The solution was stirred for 1 h ou r at room temperature, filtered , and kept undisturbed. X ray quality crystals of 3 3 formed after 2 days, and these were collected by filtration, washed with Et 2 O , and dried under vacuum. The yield was 55%. Anal.Calcd (found)% for 3 3 (Mn 12 O 40 C 97 H 86 N 24 Cl 2 I 11 ): C 26.76 (26.50); H 1.99 (2.06); N 7.72 (7.25) . Selected IR data (cm 1 ): 1600 (m), 1571 (s), 1463 (m), 1385 (s), 1335 (s), 1291 (w), 1181 (w), 1162 (w), 1109 (w), 1044 (w), 943(w), 776 (m), 710 (w), 662(m), 647 (w), 613(m), 561(w). 3.2.1. 4 [Mn III 12 O 4 (O 2 CEt) 12 (ded) 6 (EtOH)](I 3 ) 3.5 I 0.5 (3 4) A brown solution of [Mn 3 O(O 2 CEt) 6 (py) 3 ](ClO 4 ) (0.45 g, 0.5 0 mmol) in 25 m L of CH 2 Cl 2 : EtOH (4 : 1 v/v) was treated with dedH 2 (0.36 g, 1.5 mmol) and I 2 (0.38 g, 1.5 mmol).

PAGE 120

120 The solution was stirred for 1 hour at room temperature, filtered , and the black filtrate left undisturbed. X ray quality cry stals of 3 4 formed after 2 days, and these were collected by filtration, washed with Et 2 O , and dried under vacuum. The yield was 35%. Anal.Calcd (found)% for 3 4 (Mn 12 O 41 C 110 H 114 N 24 I 1 1 ): C 29.47 (29.25); H 2.56 (2.47); N 7.50 (7.35). Selected IR data ( cm 1 ): 1600 (m), 1572 (s), 1459 (s), 1386 (s), 1356 (m), 1276 (m), 1183 (m), 1162 (w), 1109 (m), 943 (w), 809(w), 773 (m), 708 (m), 663(m), 647 (w), 615 (m), 559 (w). 3.2.2 X Ray Crystallography X Ray Intensity data were collected at 100 K on a Bruker DUO diffractometer using MoK radiation ( = 0.71073 Å) and an APEXII CCD area detector. Suitable crystals of 3 1 · xMeCN, 3 2 · xMeCN, 3 3 · xCH 2 Cl 2 · yMeCN, and 3 4 · xCH 2 Cl 2 were attached to glass fibers using silicone grease and transferred to a goniostat where the y were cooled to 100 K for data collection. Raw data frames were read by the program SAINT and integrated using 3D profiling algorithms. The resulting data were reduced to produce hkl reflections and their intensities and estimated standard deviations. The data were corrected for Lorentz and polarization effects and numerical absorption corrections were applied based on indexed and measured faces. The structures were solved and refined in SHELXTL6.1, using full matrix least squares refinement. The non H ato ms were refined with anisotropic thermal parameters and all of the H atoms were calculated in idealized positions and refined riding on their parent atoms. For 3 1 · xMeCN, the asymmetric unit consists of two Mn 12 clusters, 2.5 perchlorate anions, 1.5 pyrid ine solvent molecules, a half pyridinium cation and an estimated 25 acetonitrile solvent molecules. The acetonitrile solvent molecules were disordered and could not be modeled properly, thus the program SQUEEZE, a part of the PLATON package of crystallogra phic software, was used to calculate the solvent disorder area and remove its contribution to the

PAGE 121

121 overall intensity data. The charge of the asymmetric unit could not be defined with high certainty because of the dis order in the perchlorate anions, and the pyridine/ pyridinium molecules/cations . The refinement model which yielded the best results consists of one perchlorate anion in general position, another completely disordered perchlorate anion successfully refined in two parts, one partial perchlorate in general position , and another partial anion located on an inversion center. The best refinement of the partial perchlorate anions was refined and gave a sum of a half occupancy unit. The pyridine molecules and pyridinium cations were refined as benzene mo lecules because of their disorder. The N atom could not be distinguished in the rings. In the final cycle of refinement, 47316 reflections (of which 18753 are observed with I > 2 (I)) were used to refine 3297 parameters and the resulting R 1 , wR 2 , and S (g oodness of fit) were 5.71 %, 12.35 % , and 0.728, respectively. For 3 2 · xMeCN, t he asymmetric unit consists of a half Mn 12 cluster cation, a perchlorate anion and 5 acetonitrile solvent molecules. The solvent molecules were disordered and could not be mode led properly, thus the program SQUEEZE, a part of the PLATON package of crystallographic software, was used to calculate the solvent disorder area and remove its contribution to the overall intensity data. The perchlorate anion is disordered about a pseudo 3 fold rotation around bond Cl1 O20. The three disordered oxygen atoms were refined in two parts with their site occupation factors dependently refined. There are also disorders in the cluster itself, specifically atoms C20 C21, C24, C27, C48 , and C51. In each case , the disorder was refined in two parts with their site occupation factors were also dependently refined. All disordered parts geometries were constrained to remain ideal during the final refinement model along with restraining their displacement parameters, using line commands DFIX and EADP. In the final cycle of refinement, 13948 reflections (of which 10308 are observed with I > 2 (I))

PAGE 122

122 were used to refine 890 parameters and the resulting R 1 , wR 2 , and S (goodness of fit) were 5.36 %, 15.93 % , and 1 .083 , respectively. For 3 3 · xCH 2 Cl 2 .yMeCN, t he asymmetric unit consists of a one third Mn 12 cluster located on a 3 fold rotation symmetry element, one I 3 1 in general position, an I 1 disordered against an I 3 1 around a 3 symmetry element . The asymmetr ic unit also has a one third dichloromethane located on a 3 fold rotation symmetry element at the center of each of the Mn 12 clusters. Also in the asymmetric unit is a channel of disordered dichloromethane and acetonitrile solvent molecules. The minor part of the I 3 1 general position and around the 3 symmetry element (I4 I5 I6), were restrained to mimic the geometry of the major part (I1 I2 I3). The displacement parameters of I4 I 5, as well as I7 I 9, were restrained to remain equivalent during refinement. Similarly treated were all atoms of the disordered solvent molecules. In the final cycle of refinement, 12413 reflections (of which 8314 are observed with I > 2 (I)) were used to refine 666 parameters and the resulting R 1 , wR 2 , and S (goodness of fit) were 8.55 %, 26.97 % , and 1.982 , respectively. For 3 4 · xCH 2 Cl 2 , t he asymmetric unit consists of a one third Mn 12 cluster, one I 3 1 , an I 1 and a one third ethanol molecule disordered around the 3 fold rotatio n symmetry, and three dichloromethane solvent molecules disordered in general positions as well as around symmetry elements. The solvent molecules could not be modeled properly, thus the program SQUEEZE, a part of the PLATON package of crystallographic sof tware, was used to calculate the solvent disorder area and removed its contribution to the overall intensity data. There are two disorders in the Mn 12 clusters, one has the ethyl group C26 and the other has me thyl group C33 disordered along with the C32 H atoms. This disorder was In the

PAGE 123

123 final cycle of refinement, 10684 reflections (of which 8538 are observed with I > 2 ( I)) were used to refine 587 parameters and the resulting R 1 , wR 2 , and S (goodness of fit) were 5.38 %, 15.90 % , and 1.068 , respectively. Unit cell data and structural refinement details for the four compounds are listed in Table 3 1. 3.2.3 Physical M easur ements Infrared spectra were recorded in the solid state (KBr pellets) on a Nicolet Nexus 670 FTIR spectrometer in the 400 4000 cm 1 range. Elemental analyses (C, H, and N) were performed by the in house facilities of the University of Florida, Chemistry Department. Variable temperature direct current (dc) and alternating current (ac) magnetic susceptibility data were collected at the University of Florida using a Quantum Design MPMS XL SQUID magnetometer equipped with a 7 T magnet and operating in the 1. 8 300 K range. Samples were embedded in solid eicosane to prevent torquing. Magnetization vs. field and temperature data were fit using the program MAGNET. 17 6 diamagnetic correction, which was subtracted from the experimental susceptibility to give the molar paramagnetic susceptibility ( M ). Low temperature (<1.8 K) hysteresis loop and dc relaxation measurements we re performed at Institut Néel using an array of microSQUIDS. 177 The high sensitivity of this magnetometer al lows the study of single crystals of SMMs of the order of 10 500 m. The field can be applied in any direction by separately driving three orthogonal coils.

PAGE 124

124 3.3 Results and Discussion 3.3.1 Syntheses The syntheses of complexes 3 1 , 3 2 , 3 3 , and 3 4 were modified from the reactions to synthesize complexes [Mn 3 O(O 2 CR) 3 (mpko) 3 ]ClO 4 (R= Me ( 3 ), Et ( 4 ), mpkoH= methyl 2 pyridyl ketone oxime) 171 but with the dioxime dedH 2 in place of mpkoH. For complex 3 1 , the reaction was conducted in MeCN/MeOH as in the synthesis of complex 3 . The mixture was then evaporated under vacuum to remo ve all solvents. The black solid obtained was not soluble in CH 2 Cl 2 , which is different from 3 , and therefore it could not be redissolved in this solvent. It was, however, very soluble in DMF in which poor quality crystals of 3 1 came out by layering with diethyl ether. Although 3 1 was also not soluble in MeCN, better crystals were obtained when a mixture of MeCN/DMF (1 : 1 v/v) was used. Complex 3 1 has nine Mn III and three Mn II ions while the starting material (complex 1 ) has all Mn III ions. The monooxim ate based complex 3 also has all Mn ions in the +3 oxidation state . It is not clear what acted as the reducing agent responsible for the reduction of the Mn III ions to Mn II ions in the reaction. Comparing the reactions to make 3 and 3 1 suggests that the d ioxime dedH 2 and/or DMF are/is the possible reductant(s). Unfortunately, both of them are necessary for the reaction and it is impossible to conduct a control reaction without either of them. From the literature, glyoximes can be oxidized to furoxan s in th e presence of oxidizing agents. 199 201 DMF has been used widely as a solvent in manganese chemistry . A similar reaction was performed with dry DMF used in the crystallization process ; however, the obtained product was still complex 3 1 . It is possible that the dioxime dedH 2 itself is the reducing agent in the reaction. The formation of complex 3 1 is rather complicated with the dioxime dedH 2 involve d in two reactions: i) substitution of three pyridine and three acet ate groups in complex 1 ; ii) redox reaction with some Mn III ions, which can happen with complex 1 before the substitution and/or with the formed

PAGE 125

125 Mn III 12 right after the substitution. With the assumption that furoxan is the only product of the oxidation of glyoxime, the possible reactions that can occur in the formation of 3 1 are summarized in Figure 3 2 . The reaction of 3 moles of Mn III with 3/2 moles of dedH 2 to form 3 moles of Mn II and 3/2 moles of furoxan will also release three protons H + . Since the re action was conducted without extra base, the protons released were captured by pyridine molecules to form pyridinium cations. One pyridinium cation is trapped inside the cage of 3 1 to make the complex total charge of +2. T his was confirmed by elemental an alysis in which the percentage of Cl corresponds to two perchlorate anions. For complex 3 2 , the reaction occurred between the starting material , complex 2 , and the dioxime dedH 2 . At first, we attempted to follow the procedure of making complex 3 1 but no success was achieved. However, it was found that the black solid obtained was soluble not only in DMF but also in many other solvents such as CH 2 Cl 2 or MeCN, probably due to the higher solubility of propionate complexes compared to acetate ones. The reacti on was then modified in which MeCN was the only solvent. The product obtained has ten Mn III and two Mn II ions; moreover, there are two chloride anions bound to the two Mn II ions. The chloride anions can only be formed from the reduction of perchlorate, whi ch is the only source of Cl in the reaction . This again proves the reductive ability of the dioxime dedH 2 , although the reason for the redox reactions involving two Mn III and two perchlorate anions (not three Mn III as in 3 1 ) is not clear. The formation of 3 2 is complicated , which also involves the redox reaction of perchlorate anions and the substitution of water and Cl for propionate ligands. One way to summarize the possible reactions that occurred is shown in Figure 3 3 . The reaction of two moles of M n III with one mole of dedH 2 to form two moles of Mn II and one mole of furoxane release s two moles of protons H + . However, the substitution of water and Cl will also provide two propionates to

PAGE 126

126 capture these protons. Therefore, t he reaction medium for the e ntire process is not acidic, which can be the reason for the fact that pyridine was found as the guest molecule in the cavity of 3 2 , in contrast to the pyridinium cation in the case of 3 1 . In light of the reductive ability of dedH 2 , f or complexes 3 3 a nd 3 4 , we added an excess amount of iodine into the reaction mixture. The effectiveness of using iodine on the formation of 3 3 and 3 4 is evidenced by : i) all Mn ions in 3 3 and 3 4 are Mn III , ii) no Cl can be detected, iii) reduced forms of iodine, I 3 and I 1 , are present as the counter anions. The reaction of dedH 2 with the starting material 1 or 2 to form 3 3 and 3 4 only yield the substitution of dioximate ligands for carboxylates and pyridines (Figure 3 4). Dichloromethane and ethanol are the guest molecules inside the cavity of complex 3 3 and 3 4 , respectively. Attempts to access possible host guest chemistry are in progress. The main challenge is the difficulty in obtain ing high quality single crystals and , in fact , complexes 3 1 to 3 4 were able to be formed in different solvents; the synthese s reported above gave the best crystals for X ray crystallography. 3.3.2 Description of Structures Labeled structures of complex 3 1 and 3 2 are shown in Figure s 3 5 and 3 6, respectively. S ome selected i nteratomic distances and angles are listed in Table 3 2. Complex 3 1 ·MeCN crystallizes in triclinic space group with the asymmetric unit co nsisted of two essentially identical Mn 12 clusters. Complex 3 2 ·xMeCN crystallizes in monoclinic space group C2/c. The core of 3 1 and 3 2 is a supramolecular tetrahedron [Mn 3 ] 4 +/2+ , which consists one [Mn 3 ( 3 O)] 7+ and th ree [Mn 3 ( 3 O)] 6+ units in case of 3 1 , or two [Mn 3 ( 3 O)] 7 + and two [Mn 3 ( 3 O)] 6 + units in case of 3 2 . These Mn 3 units are linked by six de d 2 groups. There is one 3 oxide atom bridging in each Mn 3 unit. The central oxide atoms lie slightly above the Mn 3 plane in all triangles ( d ~ 0.3Å). For complex 3 1 , each edge of a triangle is bridged by an 1 : 1

PAGE 127

127 : acetate group and an 1 : 1 : 1 : pyridyloximate group in which the pyridyl and oximate nitrogen atoms are chelating a Mn atom and form a five me mbered ring. Complex 3 2 has similar binding modes of the carboxylate and pyridyloximate groups; the only difference is in the two symmetric triangles Mn4Mn5Mn6 of which the edge Mn4Mn6 has water and chloride ligands instead of a carboxylate . In both compl exes, the Mn···Mn separations and Mn ( 3 O) Mn angles in triangles are different; thus the triangles are scalene. In complex 3 1 , all Mn ions are not symmetrically equivalent; therefore, it has crystallographic C 1 symmetry but virtual C 3 symmetry. In compl ex 3 2 , the Mn 3 triangles are symmetrically related by a C 2 rotation axis, thus the complex has a crystallographic C 2 symmetry. The oxidation states of Mn atoms were established by bond valence sum (BVS) calculations. In complex 3 1 , due to crystallographi c disorders, BVS values of some Mn ions are close to 2.5 rather than the usual integer values for Mn ions, i.e. 3 for Mn III and 2 for Mn II . Nevertheless, we still can assign the oxidation states as the nearest integer to the calculated values. BVS calculat ions also confirm that all central 3 O atoms in both complexes are O 2 ions, and confirm the H 2 O ligand of complex 3 2 with a BVS value of 0.29. Both complexes 3 1 and 3 2 show a host guest feature in which either a pyridinium cation or a pyridine molecu le is encapsulated in the cavity (Figure 3 7 ). The pyridine molecule of 3 2 is disordered in two positions. There are weak H bonds between the O atoms of the ded 2 groups and the H atoms of pyridinium or pyridine ( 2.1Å). Both complexes 3 3 ·xCH 2 Cl 2 ·y MeCN and 3 4 ·yCH 2 Cl 2 crystallize in rhombohedral space group with nearly identical unit cell data. Their selected bond distances and angles are provided in Table 3 3 . They have similar structures; the only difference s are the size of the carboxylate liga nds and the guest molecule inside the cage. The structures of 3 3 and 3 4 consist of one Mn 12 4+ cation, which is shown in Figure s 3 8 and 3 9 . The Mn 12 cation composed of four

PAGE 128

128 [Mn 3 ( 3 O)] 7+ units linked by a total of six ded 2 groups to give a supramolecu lar [Mn 3 ] 4 sphere. One of the two 1 : 1 : RCO 2 ligands bridging each edge of the Mn 3 triangle in 1 and 2 has been replaced by a bridging oximate from a ded 2 group (Figure s 3 8 and 3 9 , bottom). In addition, the ded 2 pyridyl groups have replaced the term inal pyridine ligands of 1 and 2 . Each of the 3 O 2 ions lies slightly above the Mn 3 plane ( d ~ 0.3 Å), as in 3 and 4 . Thus, the local structure of each Mn 3 unit of 3 3 and 3 4 is very similar to that of 3 and 4 , comprising a [Mn 3 ( 3 O)] 7+ triangular unit whose edges are each bridged by one acetate and one pyridyloximate group, and whose pyridyl group binds terminally to the Mn. The complete cation s of 3 3 and 3 4 has crystallographic C 3 and virtual T point group symmetry. The Mn III oxidation states were c onfirmed by bond valence sum (BVS) calculations , and their Jahn Teller elongation axes (green bonds in Fig ures 3 8 and 3 9 , bottom) are aligned in a propeller fashion, again as in 3 and 4 . Except for the crystallographically equilateral triangle Mn4 Mn4 ' M n4 " , the Mn···Mn separations and Mn ( 3 O) Mn angles in triangles Mn1 Mn2 Mn3 are slightly different; thus the triangles are scalene but virtually equilateral within the usual 3 criterion. A solvent molecule, CH 2 Cl 2 for 3 3 and EtOH for 3 4 , is encapsulat ed inside the cage of these spheres to form host guest complexes (Figure 3 10) . The diameter of the spherical central cavity can be determined as the shortest distance from the center point of the tetrahedron to atoms nearby and was found as ~8.0 Å for com plex 3 4 (Figure 3 11) , giving a volume of 270 Å 3 (~0.27 nm 3 ). The solvent guest molecules are much smaller (2.9 Å for CH 2 Cl 2 and 4.0 Å for EtOH) and fit easily into this space, and are found to be stati sti cally disordered about the C 3 axis. 3.3.3 Supramo lecular C hemistry The tetrahedral cage employ ing six bis chelate ligands is familiar in supramolecular chemistry. A great number of tetrahedral M 4 L 6 cages have been reported in the literature. 202 205 In

PAGE 129

129 our system , generalized as [M 3 ] 4 L 6 , we still follow the principles that facilitate the formation of cages 75,206 : i) high spin octahedr al Mn III ions (t 2g 3 e g 1 ) are kinetically labile allowing initial kinetic products such as dimers, polymers, etc . to rearrange into the final tetrameric thermodynamic products; ii) the ded 2 bridging groups are rigid and have steric constraints that drive th e formation of a globular complex with a tetrahedral [Mn 3 ] 4 topology rather than a more 2D one such as a rectangular topology seen with the less rigid linker , pdpd 2 , in C hapter 2 . One particularly important factor favoring the formation of the molecular [ Mn 3 ] 4 L 6 nanocapsule in 3 3 and 3 4 is that the three bridging oximate groups at each Mn 3 unit are on the same side of the Mn 3 plane, as they also are in 3 and 4 , giving a tripod shape to the oxime dispositions. While this does not rule out the formation of polymers, it does favor a closo molecular product. As often seen in other M 4 L 6 tetrahedral n molecules , the bridging ded 2 groups are achiral, but the crystallographic C 3 symmetry means the [Mn 3 ] 4 L 6 products are chiral, and in fact, complexes 3 3 and 3 4 c rystallize as a racemic mixture ( Figure 3 1 2 ). In terms of host guest chemistry, two important factors that affect the selectivity of a guest are: i) physical characteristics of the guests, e.g. shape, size , and charge, and ii) interactions between the ho st and the guest, e.g. electrostatic, H bonds, charge dipole, dipole dipole. It is reasonable to have pyridinium and pyridine as the guests of 3 1 and 3 2 since they are pretty large (diameter d 4.6Ã…) and there are weak H bonds between hydrogen atoms on the aromatic ring and oxygen atoms of the ded 2 groups. For 3 3 and 3 4 , a pyridinium cation was also expected as the guest . However, CH 2 Cl 2 in 3 3 and EtOH in 3 4 were in fact more favorable to be encapsulated. One possible explanation is that both 3 3 an d 3 4 are highly positive ly charged (4+) and the repuls ive force between two positive ly charged species may restrict the encapsulation of pyridinium cation (but it is not always the case 207 ). The favorability of EtOH

PAGE 130

130 over CH 2 Cl 2 in 3 4 can be explained upon closer inspection of the structure of 3 4 which reveals that the EtOH guest is anchored to the host by an O H···O hydrogen bond (2.830(7) Å) to the O atom of the nearby oximate NO unit. In fact, the twelve total oximate O atoms of the compl ete [Mn 3 ] 4 cage are the nearest to the center of the molecule and thus form the inner boundary with in the central cavity. They occur as a [O 3 ] 4 tetrahedron of four O 3 triangles, one from each Mn 3 unit, and thus form a roughly spherical ball of twelve O ato ms (Figure 3 1 3 ) that would be the nearest atoms to guest molecules. This suggests that a variety of small organic or inorganic molecules could be introduced as gues ts . The host guest chemistry provides another motivation for ongoing work since many exampl es of metal organic framework materials encapsulating movable polar guest molecules such as ethanol, methanol or water have been found to exhibit ferroelectric responses. 208 211 3.3. 4 Magnetochemistry 3.3. 4 .1 Direc t current magnetic susceptibility studies Variable temperature, dc magnetic susceptibility ( M ) measurements were performed on vacuum dried polycrystalline samples of complexes 3 1 , 3 2 , 3 3 , and 3 4 in an applied field of 1000 G (0.10 T) and the 5.0 300 K temperature range. The samples were restrained in eicosane to prevent torquing. Figure 3 1 4 shows the molar magnetic susceptibility ( M ) of complexes 3 1 to 3 4 as a M T versus T plot. For complex 3 1 , the M T value is 41.59 cm 3 K mol 1 at 300 K, sligh tly increasing with decreasing temperature to 43.33 cm 3 K mol 1 at 175 K and then decreasing to 25.86 cm 3 K mol 1 at 5 K. The value at 300 K is close to the spin only ( g = 2) value for nine Mn III and three Mn II non interacting ions ( M T = 40.13 cm 3 K mol 1 ). The data suggest the presence of predominant antiferromagnetic interactions within the molecule. For complex 3 2 , t he M T value is 43.04 cm 3 K mol 1 at 300 K, which increas es with decreasing temperature to

PAGE 131

131 51.24 cm 3 K mol 1 at 50 K before decreas ing to 24.47 cm 3 K mol 1 at 5 K. The value at 300 K is higher than the s pin only ( g = 2) value for ten Mn III and two Mn II non interacting ions ( M T = 38.75 cm 3 K mol 1 ). The data suggest t he competing ferromagnetic and antiferromagnetic interactions with predomi nant antiferromagnetic interactions that lead to the decrease of M T at low temperature. In both complexes 3 1 and 3 2 , the antiferromagnetic interaction s can be assigned to Mn II O Mn III couplings. Complex 3 2 has ten Mn III ions while complex 3 1 has nine Mn III ions, thus complex 3 2 has less antiferromagnetic Mn II O Mn III couplings, which explains for the larger M T values of 3 2 compared to those of 3 1 (Figure 3 14) . For complexes 3 3 and 3 4 , as expected from the structural similarity, their magnetic pr operties are nearly identical, as can be seen in Figure 3 1 4 . The M T value of complex 3 3 increases from 50.69 cm 3 K mol 1 at 300 K to a plateau value of 68.69 cm 3 K mol 1 at 7 0 K, and then decreases to 30.16 cm 3 K mol 1 at 5 K. Complex 3 4 exhibits a ver y similar profile with M T value of 51.39 cm 3 K mol 1 at 300 K, increasing to 66.48 cm 3 K mol 1 at 70 K and then decreasing to 28 . 5 7 cm 3 K mol 1 at 5 K. The 300 K value s are larger than the spin only ( g = 2) value for twelve non interacting Mn III ions ( M T = 36 cm 3 K mol 1 ) indicating ferromagnetic interactions within each Mn 3 molecule, as previously observed for monomeric Mn 3 complexes 3 and 4 . 171 As the temperature is decreased, M T increases steadily, but the maximum value at 70 K is lower than that calculated for four non interacting S = 6 Mn 3 units of M T = 84 cm 3 K mol 1 wit h g = 2. Comparing the M T of 3 4 and four times of M T of 4 also shows substantial difference at temperature s below 70 K (Figure 3 1 5 ). Investigat ing the packing of complex 3 4 in the crystal rules out the possibility of intermolecular interaction s throug interaction s . 212,213 This indicates the presence of antiferromagnetic inter Mn 3 exchange interactions within each [Mn 3 ] 4 molecule that are weaker than the intra Mn 3 ferromagnetic ones

PAGE 132

132 and do not significantly affect the high temperature data. Such antiferro magnetic coupling between the four Mn 3 2 groups and providing a superexchange pathway for magnetic orbital overlap and antiferromagnetic exchange. Fitting M T vs . T data to the Van Vlec k equation by the Kambe vector coupling approach requires that the molecule does not have more than two magnetic coupling J parameters. In all complexes, in addition to the interactions of Mn ions in each triangle, there are also six weak inter Mn 3 interac tions between the triangles. An attempted fitting for complex 3 3 or 3 4 with the assumption that there are no interactions through the ded 2 group, i.e. four independent Mn III 3 triangles, gave a very poor quality fit. A simulation using the MAGPACK progra m could not be perform ed due to the large number of total spin multiplets ( (2S i +1) = 5 12 ), which is beyond the computing capacity of the program. Nevertheless, the structural similarity between Mn 3 units of 3 3 or 3 4 and complex es 3 and 4 indicates that each of these Mn 3 subunits has a n S = 6 ground state as in 3 and 4 . The strength of the antiferromagnetic interaction through ded 2 group can be estimated by analyzing the exchange biased field extracted from the hysteresis study ( vide infra ). 3.3.4.2 Ma gnetization versus DC magnetic field studies To obtain the gro und state spin of a molecule, a usual method is to perform a magnetization vs . dc field measuremen t . For a supramolecular aggregate, a ground state spin can be assigned to its subunits if all t he subunits are magnetically identical. Complexes 3 1 and 3 2 have different types of Mn 3 subunits and are not egligible for this study. C omplexes 3 3 and 3 4 meet the requirement. Thus, magnetization (M) data were collected in the 0.1 7 T and 1.8 10 K ranges, and were plotted for complex es 3 3 and 3 4 as M/N B vs. H/T in Figure s 3 1 6 and 3 1 7 , where N B is the Bohr magneton. The data were fit, using the

PAGE 133

133 program MAGNET, by diagonalization of the spin Hamiltonian matrix assumin g only the ground state is populated, incorporating axial anisotropy ( z 2 ) and Zeeman terms, and employing a full powder average. The spin Ha miltonian is given by equation 3 1 , where D is the axial ZFS parameter, 0 is the vacuum permeability and H is the applied field. z 2 + g B 0 ( 3 1) Due to the antiferromagnetic coupling between Mn 3 triangles through ded 2 group s , only high magnetic field data (4 7 T) were used in order to overcome this weak interaction. The fit gave quite similar results for both comple xes, with S = 6, g = 1.96 0.09, D = 0.76 0.16 cm 1 for 3 3 and S = 6, g = 1.90 0.06, D = 0.78 0.10 cm 1 for 3 4 . The quality of the fits is not very g ood with high standard errors of g . The error surfaces in Figure s 3 1 8 and 3 19 also show severa l minimum points, thus the obtained values of D and g are of high uncertainty. In fact, the simplistic model employed does not allow for an accurate determination of D and g since factors such as inter Mn 3 interactions or intermolecular interactions were i gnored. Nevertheless, these fits still prove that complexes 3 3 and 3 4 are tetrameric versions of S = 6 SMMs. 3.3. 4 .3 Alternating current magnetic susceptibility studies Since complexes 3 1 to 3 4 have at least one Mn 3 unit structurally similar to the SM M s 3 and 4 , they are all expected to exhibit SMM behavior. Therefore, alternating current (ac) magnetic susceptibility measurements were performed on their microcrystalline samples in the 1.8 15 K temperature range in a 3.5 G ac field oscillating at 50 1000 Hz. The in phase ( M , plotted as M T ) and out of phase ( M ) ac plots of all four complexes are shown in Figure s 3 2 0 , 3 2 1 , 3 2 2 , and 3 2 3 . The M T vs. T profiles of complex es 3 3 and 3 4 are nearly superimposable as expected for their structural similarity. The inter M n 3 antiferromagnetic interactions significantly affect the M T values at very low temperature s ; for 3 3 and 3 4 they

PAGE 134

134 quickly decrease from 50 cm 3 K mol 1 to 12 cm 3 K mol 1 at 1.8 K. All complexes exhibit small M signals which are the tail of peaks lyin g below 1.8K indicating the characteristic of slow magnetization relaxation. 3.3.4.4 Magnetization versus DC field hysteresis loops To confirm the SMM behavior of 3 1 and 3 4 , and to investigate what extent the four constituent SMMs in 3 4 are exchange cou pled to each other and whether they thus exhibit exchange biasing of their quantum tunneling of magnetization ( QTM ) , magnetization vs. dc magnetic field scans on single crystals of 3 1 · xMeCN and 3 4 · xCH 2 Cl 2 were carried out on a micro SQUID. 177 The crystals were maintained in mother liquor to prevent degradation by solvent loss until a suitable crystal was chos en and transferred to the instrument. Hysteresis loops were observed below ~1.0 K (Figure s 3 2 4 and 3 2 5 for 3 1 and 3 4 , respectively) whose coercivities increase with decreasing temperature and increasing field sweep rate, as expected for SMMs. Complex 3 1 has one Mn III 3 unit and three Mn III 2 Mn II units. The presence of Mn II io ns affects the magnetic properties in two aspects: i) Mn II ions weakly antiferromagnetically couple with Mn III ions in each Mn III 2 Mn II unit; consequently, these subunits themselves h ave low lying excited states typical for Mn II containing species; 214 ii) Mn III 3 and Mn III 2 Mn II units possess different magnetization M upon applying magnetic field. These effects can be seen clearly in the hysteresis loops of complex 3 1 , where there are no clear steps of QTM. For normal SMMs, the first step in the h ysteresis loop on scanning from negative to positive fields is at zero field. This is where the M S levels on either side of the anisotropy barrier are in resonance and QTM can occur, reversing the orientation of the magnetization vector. The presence of an AF exchange coupled neighbor provides a bias field that shifts the resonant tunneling (QTM step) to a new position before zero field. For complex 3 4 , a prominent feature of these hysteresis loops is that

PAGE 135

135 the first step appears before zero field. This rar e phenomenon is diagnostic of systems having inter SMM interactions. To obtain a simplified spin Hamiltonian describing the [Mn 3 ] 4 tetramer, each Mn 3 SMM can be modeled as a giant spin of S = 6 with Ising like anisotropy. The spin Hamiltonian of each Mn 3 is given by equation 3 5 , i = z,i 2 + trans,i + g B 0 zi . H z ( 3 2 ) where i = 1 4 (referring to the four Mn 3 SMMs of the tetramer). The general Hamiltonian for the tetramer is as follows: = i 2 J ij i j ( 3 3 ) where J ij are superexchange coupling s between subunits i and j, S i = S j = 6, assuming superexchange interactions are isotropic ( J z = J xy ). Since 3 4 is a tetrahedral [Mn 3 ] 4 supramolecular aggregate, there is only one inter Mn 3 interaction J ; the dipolar (through space) interaction s should be much weaker and can be ignored. Therefore, the spin Hamiltonian is given in equation 3 4 = 1 + 2 + 3 + 4 2 J ( 1 2 + 1 3 + 1 4 + 2 3 + 2 4 + 3 4 ) ( 3 4 ) Tunneling among the (2S+1) 4 = 28561 energy states is allowed by the small transverse anisotropy trans,i and the transverse coupling terms containing xi and y i operators. The energy states of [Mn 3 ] 4 can be calculated by exact diagonalization but it requires a large amount of calculation. A more simple approach can be obtained by neglecting all transverse anisotropy terms. The spin Hamiltonian in equation 3 4 ca n be rewritten as 1 + 2 + 3 + 4 2 J ( z 1 z 2 + z 1 z 3 + z 1 z4 + z2 z3 + z2 z4 + z3 z4 ) ( 3 5 ) The eigenvalue s of equation 3 5 are given as follows :

PAGE 136

136 E = D (m 1 2 + m 2 2 + m 3 2 + m 4 2 ) g B 0 H z (m 1 + m 2 + m 3 + m 4 ) 2 J (m 1 m 2 + m 1 m 3 + m 1 m 4 + m 2 m 3 + m 2 m 4 + m 3 m 4 ) ( 3 6 ) Simulation of spin state energies versus applied field based on equation 3 6 can be done when D , g , and J values are input. Since the value of D obtained from the fit of reduced magnetization is of high uncertainty, we assume D = 0.3 1 cm 1 ( 0.45 K), which is a common value for this type of Mn 3 as in complex 3 and 4 . 171 Figure 3 2 6 shows the plot of spin state energies vs. applied magnetic field for a tetramer of four Mn 3 SMMs ; the plot was simulated with D = 0.3 1 cm 1 = 0.45 K, J = 0.0 2 1 cm 1 = 0.0 30 K ( J = 0.04 2 cm 1 = 0.0 60 K in physics convention) , an d g = 2. For a system of four interacting Mn 3 SMMs in which each SMM couples with three other SMMs, it is expected to have two steps before/after zero field and no step at zero field. As the field is scanned from 1.5 T, where the four Mn 3 spin vectors are polarized into the M S = 6 orientation, the first step corresponds to tunneling of a Mn 3 vector from M S = 6 to M S =+6; this occurs at 0. 7 T, which equals the total bias field from three M S = 6 neighbors , thus the first step biased field is proportion al to 3 J ( J < 0). The second step corresponds to tunneling of a second Mn 3 vector from M S = 6 to M S = +6; it can be at any of the remaining three Mn 3 . For this Mn 3 , the interactions are 2 J from two M s = 6 and J from one M s = +6 neighbors. Therefore, the b iased field is proportional to 1 J , which is consistent with the hysteresis loop da ta where the second step is at 0.2 T. Although the steps are broad because the four Mn 3 planes are not coplanar and the applied field will thus be at a range of angles to the easy axis, the step pattern confirms that 3 3 and 3 4 are supramolecular aggregates of four exchange biased Mn 3 SMMs.

PAGE 137

137 3. 4 Summary and Conclusions In summary, four supramolecular [Mn 3 ] 4 complexes with different charges and carboxylate ligands were synt hesized in high yields from the use of dedH 2 . This study reveals the reductive ability of dedH 2 and the effectiveness of using iodine to protect Mn III containing species from reduction. In regards to supramolecular chemistry, each complex encapsulates a gu est molecule depending on the reaction conditions: acidity, protic and aprotic solvents. The formation of these tetrahedral supramolecules in racemic mixtures is as expected. Magnetochemistry studies on these complexes confirmed their SMM behavior. The pre sence of Mn II ions in 3 1 and 3 2 leads to antiferromagnetic interactions with other Mn III ions and probably the loss of SMM behavior of Mn III 2 Mn II units. The weak inter Mn 3 antiferromagnetic coupling through ded 2 group s manifested as exchange biased QTM steps was demonstrated in the hysteresis studies of 3 4 . The work described in this report represents the amalgamation of single molecule magnetism and supramolecular chemistry. Dioxime (pdpdH 2 and dedH 2 ) have been demonstrated to be able to connect Mn 3 S MMs into supramolecular aggregates. Additional synthetic effort to synthesize more di/trioximes is necessary to further explore different supramolecular structures and consequent magnetic behavior of these aggregates. The supramolecules contain solvent mol ecules inside their central cavity, which also gives a host guest chemistry angle to this work. These complexes can accurately be described as structurally supramolecular, as a tetramer of four distinct inorganic units linked by organic groups. They can al so be described as magnetically supramolecular, with each Mn 3 unit being an SMM and the weak exchange interactions between them representing a small exchange biasing perturbation of their single molecule properties. Finally we note that bringing together h ost guest chemistry and single molecule magnetism offers a wealth of possible future studies, such as placing magnetic

PAGE 138

138 guests inside the nanocapsules to explore their effect on (i) the properties of the Mn 3 SMMs, such as their relaxation barrier heights; ( ii) the interactions between the separate SMMs and modulation of the exchange bias; and (iii) the quantum physics of the QTM processes, such as altering the quantum tunneling rates, which will show up as larger steps at a given field position. Thus, we fee l optimistic that amalgamation of supramolecular chemistry and single molecule magnetism will prove a powerful new way to use chemistry to modulate the quantum physics of nanoscale magnets.

PAGE 139

139 Table 3 1. Crystallographic data for complexes 3 1 · xMeCN, 3 2 · xM eCN, 3 3 · xCH 2 Cl 2 · yMeCN, and 3 4 · xCH 2 Cl 2 . Parameters 3 1 · xMeCN 3 2 · xMeCN 3 3 · xCH 2 Cl 2 · yMeCN 3 4 · xCH 2 Cl 2 Formula a C 252 H 253.50 Cl 2.50 Mn 24 N 75 O 90 C 127 H 137 Cl 4 Mn 12 N 35 O 46 C 104.20 H 97.70 Cl 5.60 I 11 Mn 12 N 26.70 O 40 C 122 H 137.5 Cl 24 I 1 1 Mn 12 N 24 O 41 FW, g/mol a 7179.98 3690. 80 4617.69 5375.14 Crystal system Triclinic Monoclinic Rhombohedral Rhombohedral Space group C2/c a, Å 17.2032(17) 27.7813(8) 27.498(4) 28.2700(6) b, Å 30.598(3) 17.8178(6) 27.498(4) 28.2700(6) c, Å 31.984(3) 32.1834(10) 37.144(5) 36.8596(9) , ° 96.226(2) 90 90 90 ,° 102.701(3) 95.155(2) 90 90 ,° 102.238(2) 90 120 120 V, Å 3 15837(3) 15866.4(9) 24324(6) 25511.3(10) Z 2 4 6 6 T, K 100(2) 100(2) 100(2) 100(2) , Å b 0.71073 0.71073 0.71073 0.71073 cal , mg/m 3 1.506 1.506 1.891 2.099 , mm 1 1.031 1.076 3.161 3.13 R 1 c,d 0.0571 0.0536 0.0855 0.0538 wR 2 e 0.1235 0.1593 0.2 697 0.1590 a Including solvate molecules. b Graphite monochromator. c I>2 (I). d R 1 o | |F c o |. e wR 2 o 2 F c 2 ) 2 o 2 ) 2 ]] 1/2 2 (F o 2 )+(m × p) 2 + n × p], p = [max(F o 2 ,0)+ 2 × F c 2 ]/3, m & n are constants.

PAGE 140

140 Table 3 2. Selected interatomic distances (Ã…) and angles ( o ) for complexes 3 1 and 3 2 . Com p lex 3 1 Complex 3 2 Bond distances Mn1 Mn2 3.1728(15) Mn7 O37 1.875(4) Mn1 Mn2 3.2358(9) Mn1 Mn3 3.3243(15) Mn7 O32 2.186(6) Mn1 Mn3 3.1940(9) Mn2 Mn3 3.3069(15) Mn7 O11 2.212(5) Mn2 Mn3 3.2421(8) Mn4 Mn5 3.2154(17) Mn8 O37 1.900(4) Mn4 M n5 3.2969(9) Mn4 Mn6 3.2665(16) Mn8 O36 2.223(5) Mn4 Mn6 3.4381(10) Mn5 Mn6 3.2921(17) Mn8 O10 2.249(4) Mn5 Mn6 3.1235(9) Mn7 Mn9 3.2114(16) Mn9 O37 1.868(5) Mn1 O1 1.857(3) Mn7 Mn8 3.2298(15) Mn9 O33 2.179(5) Mn1 O3 2.184(3) Mn8 Mn 9 3.2307(17) Mn9 O1 2.183(4) Mn1 O7 2.200(3) Mn10 Mn11 3.2866(15) Mn10 O39 1.953(4) Mn2 O1 1.915(3) Mn10 Mn12 3.2748(16) Mn11 O39 1.923(4) Mn2 O6 2.177(3) Mn11 Mn12 3.2337(14) Mn11 O24 2.194(5) Mn2 O4 2.207(3) Mn1 O38 1.832(4) Mn11 O6 2.241(4 ) Mn3 O1 1.875(3) Mn1 O2 2.207(4) Mn12 O39 1.838(4) Mn3 O10 2.182(4) Mn1 O26 2.237(4) Mn12 O9 2.196(4) Mn3 O2 2.193(3) Mn2 O38 1.823(4) Mn12 O19 2.218(4) Mn4 O11 2.115(3) Mn2 O28 2.209(4) Mn5 O11 1.836(3) Mn2 O8 2.224(4) Mn5 O12 2.181(3) Mn3 O38 2.050(4) Mn5 O17 2.239(3) Mn4 O40 1.829(5) Mn6 O11 1.827(3) Mn4 O12 2.181(5) Mn6 O13 2.200(3) Mn4 O18 2.207(6) Mn6 O19 2.221(3) Mn5 O40 1.905(5) Mn5 O13 2.197(5) Mn5 O3 2.234(4) Mn6 O40 1.979(4) Bond angles Mn1 O38 Mn2 120.5(2) Mn1 O1 Mn3 117.68(14) Mn1 O38 Mn3 117.7(2) Mn1 O1 Mn2 118.14(15) Mn2 O38 Mn3 117.2(2) Mn3 O1 Mn2 117.59(15) Mn4 O40 Mn5 118.9(2) Mn6 O11 Mn5 117.02(14) Mn4 O40 Mn6 118.1(2) Mn6 O11 Mn4 121.29(14) Mn5 O40 Mn6 115.9(2) Mn5 O11 M n4 112.91(13) Mn7 O37 Mn8 117.7(2) Mn7 O37 Mn9 118.2(2) Mn8 O37 Mn9 118.1(2) Mn10 O39 Mn11 116.0(2) Mn10 O39 Mn12 119.5(2) Mn11 O39 Mn12 118.6(2)

PAGE 141

141 Table 3 3. Selected interatomic distances (Ã…) and angles ( o ) for complexes 3 3 and 3 4 . Complex 3 3 Complex 3 4 Bond distances Mn1 Mn2 3.202(2) Mn1 Mn2 3.2073(14) Mn1 Mn3 3.206(2) Mn1 Mn3 3.2059(12) Mn2 Mn3 3.199(2) Mn2 Mn3 3.2036(13) Mn4 Mn4' a 3.207(2) Mn4 Mn4' a 3.2094(12) Mn1 O4 1.859(7) Mn1 O1 1.873(4) Mn 1 O5 2.157(7) Mn1 O6 2.166(5) Mn1 O1 2.194(6) Mn1 O4 2.182(4) Mn2 O4 1.877(7) Mn2 O1 1.877(4) Mn2 O7 2.131(7) Mn2 O8 2.163(5) Mn2 O3 2.166(7) Mn2 O2 2.185(4) Mn3 O4 1.877(6) Mn3 O1 1.869(4) Mn3 O2 2.196(7) Mn3 O11 2.208(4) Mn3 O9 2.197(7) Mn3 O3 2.2 21(4) Mn4 O12 1.876(2) Mn4 O14 1.8732(11) Mn4 O14 2.153(6) Mn4 O12 2.167(4) Mn4 O11 2.179(6) Mn4 O5 2.191(4) Bond angles Mn1 O4 Mn2 118.0(3) Mn1 O1 Mn2 117.6(2) Mn1 O4 Mn3 118.2(3) Mn1 O1 Mn3 117.9(2) Mn2 O4 Mn3 116.9(4) Mn2 O1 Mn3 117.6(2) Mn4 O12 Mn4' 117.42(15) Mn4 O14 Mn4' 117.89(9) a Primed and unprimed atoms are related by symmetry.

PAGE 142

142 Table 3 4. Displacement of 3 oxide atoms from Mn 3 planes ( Ã… ) and Mn N O Mn torsion angles ( o ) of complex es 3 3 and 3 4 . T riangle d ( Ã… ) Torsion Angles ( o ) Complex 3 3 Mn1Mn2Mn3 0.29 Mn1 N5 O3 Mn2 15. 6(8) Mn2 N3 O2 Mn3 13.3(8) Mn3 N1 O1 Mn1 19.5(8) Mn4Mn4'Mn4'' 0.31 Mn4 N7 O11 Mn4 ' 11.7(7) Complex 3 4 Mn1Mn2Mn3 0.29 Mn1 N2 O2 Mn2 16.6(5) Mn2 N4 O3 Mn3 1 4.3(5) Mn3 N6 O4 Mn1 15.3(5) Mn4Mn4'Mn4'' 0.28 Mn4 N8 O5 Mn4 ' 17.3(4) Table 3 5. Bond valence sum calculation for Mn a and selected O b atoms of complex 3 1 . Mn II Mn III Mn IV Assignment BVS Assignment Mn1 3.13 2.92 2.97 Mn III O37 2.14 O 2 M n2 3.23 3.02 3.06 Mn III O38 2.15 O 2 Mn3 2.12 1.97 2.01 Mn II O39 2.07 O 2 Mn4 3.24 3.03 3.08 Mn III O40 2.07 O 2 Mn5 2.86 2.67 2.71 Mn III Mn6 2.54 2.37 2.41 Mn II Mn7 3.25 3.03 3.08 Mn III Mn8 3.11 2.91 2.95 Mn III Mn9 3.33 3.11 3.16 Mn III Mn10 2.55 2.38 2.42 Mn II Mn11 2.77 2.59 2.63 Mn III Mn12 3.18 2.97 3.02 Mn III a The bold value is the one closest to the charge for which it was calculated. The oxidation state of a particular atom can be taken as the nearest whole number to the bold value. b A BVS in the ~ 1.8 2. 2 , ~1.0 1. 4 , and ~0.2 0.4 ranges for an O atom is indicative of non , single , and double protonation, respectively, but can be altered somewhat by hydrogen bonding.

PAGE 143

143 Table 3 6. Bond valence sum calculation for Mn a a nd selected O b atoms of complex 3 2 . Mn II Mn III Mn IV Assignment BVS Assignment Mn1 3.27 3.05 3.10 Mn III O1 2.14 O 2 Mn2 3.27 3.05 3.10 Mn III O11 2.06 O 2 Mn3 3.19 2.97 3.03 Mn III O19 0.29 H 2 O Mn4 2.15 2.05 2.06 Mn II Mn5 3.07 2.88 2. 92 Mn III Mn6 3.21 3.00 3.05 Mn III a See footnote a of table 3 5. b See footnote b of table 3 5. Table 3 7. Bond valence sum calculation for Mn a and selected O b atoms of complex 3 3 . Mn II Mn III Mn IV Assignment BVS Assignment Mn1 3.24 3.02 3 .07 Mn III O4 2.21 O 2 Mn2 3.27 3.05 3.10 Mn III O12 2.17 O 2 Mn3 3.23 3.01 3.07 Mn III Mn4 3.30 3.08 3.13 Mn III a See footnote a of table 3 5. b See footnote b of table 3 5. Table 3 8. Bond valence sum calculation for Mn a and selected O b atoms of complex 3 4 . Mn II Mn III Mn IV Assignment BVS Assignment Mn1 3.27 3.05 3.10 Mn III O1 2.19 O 2 Mn2 3.29 3.08 3.13 Mn III O14 2.19 O 2 Mn3 3.19 2.97 3.03 Mn III Mn4 3.25 3.04 3.09 Mn III a See footnote a of table 3 5. b See footnote b of table 3 5.

PAGE 144

144 Figure 3 1. Structure s of mpkoH (left) , dedH 2 (middle) and pdpdH 2 (right ) . Figure 3 2. Possible reactions that can occur in the formation of complex 3 1 .

PAGE 145

145 Figure 3 3. Possible reactions that can occur in the formatio n of complex 3 2 . Figure 3 4. Formation of complex es 3 3 and 3 4 .

PAGE 146

146 Figure 3 5. Structure of the cation (top) and a stereopair (middle) of complex 3 1 with H atom s omitted for clarity, ( b ottom) t he core emphasizing the corresponding connectivity between Mn 3 subunits, and showing the Mn 3 planes and Jahn Teller axes (green bonds). Color code: Mn III green; Mn II sky blue; O red; N blue; C grey .

PAGE 147

147 Figure 3 6. Structure of the cation (top) and a stereopair (middle) of complex 3 2 with H atoms omitted for clarity, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits, and showing the Mn 3 planes and Jahn Teller axes (green bonds). Color code: Mn III green; Mn II sky blue; O red; N blue; C grey; C l pink.

PAGE 148

148 Figure 3 7 . The structures of the cations of 3 1 (top) and 3 2 (bottom) showing their guest s , pyridin i u m and pyridine , in space filling style , respectively . For complex 3 1 , d ue to crystallographic disorders in the aromatic ring, the N atom was not determined. For complex 3 2 , the pyridine guest is disordered in two positions.

PAGE 149

149 Figure 3 8 . Structure of the cation (top) and a stereopair (middle) of complex 3 3 with H atoms omitted for clarity, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits, and showing the Mn 3 planes and Jahn Teller axes (green bonds). Color code: Mn III green; O red; N blue; C grey.

PAGE 150

150 Figure 3 9. Structure of the cation (top) and a stereopair (middle) of complex 3 4 with H atoms omitted for clarity, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits, and showing the Mn 3 planes and Jahn Teller axes (green bonds). Color code: Mn III green; O red; N blue; C grey.

PAGE 151

151 Figure 3 10. The structures of the cations of 3 3 (top) and 3 4 (bottom) showing their guests, CH 2 Cl 2 and EtOH, in space filling style, respectively. T he guest molecules are disordered about the C 3 rotation axes.

PAGE 152

152 Figure 3 1 1 . Determination of the center point (C) of the tetrahedron O14O1O1'O1'' .

PAGE 153

153 Figure 3 1 2 . Racemic mixtures of complex es 3 3 (top) and 3 4 (bottom) .

PAGE 154

154 Figure 3 1 3 . Space filling presentation of the twelve O atoms of the [O 3 ] 4 tetrahedron from six ded 2 groups that form a cage like shell of O atoms about the central spac e in complex 3 4 . Figure 3 1 4 . M T vs. T for complex es 3 1 , 3 2 , 3 3 , and 3 4 .

PAGE 155

155 Figure 3 1 5 . Plot of the difference ( M T ) between the M T value of 3 4 and four times of M T value of 4 . Figure 3 1 6 . M / N B (per Mn 3 ) vs. H / T for complex 3 3 . The solid lines are the fits to the data. See the text for the fit parameters .

PAGE 156

156 Figure 3 1 7 . M / N B (per Mn 3 ) vs. H / T for complex 3 4 . The solid lines are the fits to the data. See the text for the fit parameters . Figure 3 1 8 . Two dimensional contour plot of the e rrors vs. D and g for the fit of 3 3 .

PAGE 157

157 Figure 3 19 . Two dimensional contour plot of the errors vs. D and g for the fit of 3 4 .

PAGE 158

158 Figure 3 2 0 . Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signal s versus temperature for a micro crystalline sample of complex 3 1 in a 3.5G field oscillating at the indicated frequencies .

PAGE 159

159 Figure 3 2 1 . Plot of in phase ( M , as M T ) (top) a nd out of phase ( ) (bottom) ac susceptibility signal s versus temperature for a micro crystalline sample of complex 3 2 in a 3.5G field oscillating at the indicated frequencies .

PAGE 160

160 Figure 3 2 2 . Plot of in phase ( M , as M T ) (top) and out of phase ( ) ( bottom) ac susceptibility signal s versus temperature for a micro crystalline sample of complex 3 3 in a 3.5G field oscillating at the indicated frequencies .

PAGE 161

161 Figure 3 2 3 . Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibil ity signal s versus temperature for a micro crystalline sample of complex 3 4 in a 3.5G field oscillating at the indicated frequencies .

PAGE 162

162 Figure 3 2 4 . Magnetization vs. dc field hysteresis loops for a single crystal of 3 1 ·xMeCN at the indicated temperat ures (top) and field scan rates at 0.04 K (bottom). M is normalized to its saturation value, M S

PAGE 163

163 Figure 3 2 5 . Magnetization vs. dc field hysteresis loops for a single crystal of 3 4 ·xCH 2 Cl 2 at the indicated temperatures (top) and field scan rates at 0.0 4 K (bottom). M is normalized to its saturation value, M S

PAGE 164

164 Figure 3 2 6 . Simulation of the plot of spin state energies vs. applied magnetic field for a tetramer of four Mn 3 SMMs, each with S = 6, s pin states involving only the M s = ±6 states. * = multipl e spin flips (2, 3 , or 4) at the same time. See the text for the spin s tates involved in the QTM steps , which are at green arrows. * * * * * *

PAGE 165

165 CHAPTER 4 FERROMAGNETIC EXCHANGE BIAS IN A DIMERIC SUPRAMOLECULAR AGGREGATE OF SINGLE MOLECULE MAGNET S 4.1 Introduction In C hapter s 2 and 3, we reported the rectangular and tetrahedral [Mn 3 ] 4 covalently bonded supramolecular aggregate s of SMMs . In both cases, Mn 3 SMM units weakly antiferromagnetically interact with each other through the connecting ligands at very low tem perature. The success of this work suggested the feasibility to develop a multi qubit system based on SMMs in which the quantum mechanically coupled aggregates are stable in solution and ready for deposition on surfaces, unlike the hy drogen bonded supramo lecules, p olymers, or networks. 120,142,144,145 It also encourage d efforts to modify organic ligands not only to provide control over the formation of supramolecular structures but also to finely tune the nature, i. e. ferromagnetic or antiferromagnetic, and the strength of inter SMM interaction s within SMM supramolecular aggregates. Such manipulation has rarely been reported in discrete SMM supramolecular aggregate systems although it is known in some non SMM supramo lecules. For example, a ferromagnetic coupling between dimeric or trimeric units can sometimes be achieved even when the metal ions are quite isolated from each other by long chain organic ligands; some of them were explained through spin polarization mech anism. 156,215 219 Tuning the strength of the magnetic interaction was demonstrated in a series of Cr 7 Ni dimers by systematically modifying the aromatic linker. 165 Reconsidering the structures and magnetic properties of complexes 2 3 , 2 4 , 3 3 , and 3 4 , what was learned from these studies? Firstly, the flexibility/rigidity of the dioximate ligand affects the form ation the supramolecules. A very flexible group pdpd 2 has both extended and folded conformations in complexes 2 3 and 2 4 , thus a rectangular supramolecule is a reasonable product. O n the other hand, the rigid ded 2 group form s the tetrahedral supramolecu le s 3 3 and 3 -

PAGE 166

166 4 , which seem to be the only possible product s . App arently, an intermediate dioxime that is less flexible than pdpdH 2 and less rigid than dedH 2 would be interesting to further explore the structural effect of the dioximes on the formation of the se Mn 3 supramolecules . Secondly, the structure of the dioximate ligand also affects the strength of the exchange interaction between Mn 3 units. The pdpd 2 group has three sp 3 C atoms separating the oxim ate groups; therefore , the interaction through the ligand is very weak , probably too weak for HF EPR study. The ded 2 group has the direct connection between oxim ate groups; consequently, the interaction between Mn 3 units through the ded 2 group is stronger than the one through the pdpd 2 group as expected . However, it also leads to a significant d ecrease of the magnetic moment at low temperature when these exchange interactions start to show up. Based on thes e considerations, a new dioxime that has one or two sp 3 C atoms between oxime groups was considered a good next choice for the study. In addition, we also targeted a dimer of Mn 3 SMMs because a dimer has the highest possibility to have parallel Mn 3 planes, which is crucial in single crystal hysteresis and HF EPR studies. Retrosynthetic analysis based on t he directi onal bonding approach 27 reveals the re quirement of a 109 o ditopic ligand to form a dimer of tritopic metal complexes. Thus, the dioxime 1,3 di(pyridin 2 yl)propane 1,3 dione dioxime (dpdH 2 ), which has one sp 3 C atom between oxime groups, wa s identified as suitable . In the present work, w e p resent the formation of a dimer of Mn 3 SMMs from the use of dpdH 2 . We also demonstrate that the inter SMM exchange interaction can be seen in solution, which paves the way to deposit these supramolecular aggregates on surface s for potential applications.

PAGE 167

167 4.2 Experimental Section 4.2.1 Syntheses All manipulations were performed under aerobic conditions using chemicals and solvents as received , unless otherwise stated. [Mn 3 O(O 2 CMe) 6 (py) 3 ]ClO 4 ( 1 ) was prepared as described elsewhere. 172 The dioxime dpdH 2 was synthesized following a reported procedure. 220 Caution! Although no such behavior was observed during the present work, perchlorate compounds are potentially explosive; such compounds s hould be synthesized and used in small quantities, and treated with utmost care at all times. [Mn 6 O 2 (O 2 CMe) 6 (dpd) 3 )](I 3 ) 2 ( 4 1 ) . The reaction of 0.044 g (0.05 0 mmol) of 1 with 0.019 g (0.075 mmol) of dpdH 2 and 0.038 g (0.15 mmol) of I 2 in CH 2 Cl 2 /EtOH (25: 1 v/v) gave a dark brown solution. The solution w as filtered and the filtrate was left to concentrate by slow evaporation at ambient temperature. X ray quality crystals of 4 1 ·xCH 2 Cl 2 slowly formed over a few days and were collected by filtration; the yiel d was ~22% based on Mn. Anal. Calcd. (found) for 4 1 · H 2 O (C 51 H 50 Mn 6 N 12 O 21 I 6 ) C 27.13 (27.38); H 2.23 (1.89); N 7.44 (6.96). I 33.72 (34.34). Selected IR data (cm 1 ):1601 (s), 1570 (s), 1521 (m) 1473 (s), 1384 (vs), 1332 (s), 1186 (s), 1159 (m), 1110 (s), 1 063 (w), 1046(w), 1032 (w), 934 (w), 770 (m), 742 (w), 698 (m), 658(m), 608 (m), 563 (w). 4.2.2 X Ray Crystallography X Ray Intensity data were collected at 100K on a Bruker DUO diffractometer using MoK radiation ( = 0.71073 Å) and an APEXII CCD area d etector. A suitable crystal of 4 1 · xCH 2 Cl 2 was attached to a glass fiber using silicone grease and transferred to a goniostat where they were cooled to 100K for data collection. Raw data frames were read by the program SAINT and integrated using 3D profili ng algorithms. The resulting data were reduced to produce hkl reflections, their intensities and estimated standard deviations. The data were corrected for

PAGE 168

168 Lorentz and polarization effects, and numerical absorption corrections were applied based on indexed and measured faces. The structures were solved and refined in SHELXTL6.1, using full matrix least squares refinement. The non H atoms were refined with anisotropic thermal parameters and all of the H atoms were calculated in idealized positions and refine d riding on their parent atoms. The asymmetric unit consists of a 1/6 Mn 6 cluster located on a symmetry element, and two 1/6 I 3 counterions also located on symmetry elements . The I 3 ions are very disordered where one is disordered along a sy mmetry element and the ions form a channel of electron density, while the second I 3 is disordered along another symmetry element but they are perpendicular to it in this case. Because the I 3 counterions were heavily disordered and could not be mode led properly, the program SQUEEZE, a part of the PLATON package of crystallographic software, was used to calculate the solvent and anion disorder area and remove its contribution to the overall intensity data. The correct stoichiometry of one Mn 6 cluster with a +2 charge and two I 3 counterions is also supported by other experiments since the xray data did not enable us to refine the counterions. In the final cycle of refinement, 3349 reflections (of which 2818 are observed with I > 2 (I)) were used to refine 136 parameters and the resulting R 1 , wR 2 and S (goodness of fit) were 4.96 %, 13.15 % and 1.063 , respectively. Unit cell data and structural refinement details for compound 4 1 are listed in Table 4 1. 4.2.3 Physical M easurements Infrared spectra were recorded in the solid state (KBr pellets) on a Nicolet Nexus 670 FTIR spectrometer in the 400 4000 cm 1 range. Elemental analyses (C, H, and N) were performed by the in house facilities of the University of Florida, Chemistry Departme nt. UV Vis spectra were obtained on a Jasco V570 spectrometer in the 200 800 nm. Variable temperature

PAGE 169

169 direct current (dc) and alternating current (ac) magnetic susceptibility data were collected at the University of Florida using a Quantum Design MPMS XL S QUID magnetometer equipped with a 7 T magnet and operating in the 1.8 300 K range. Samples were embedded in solid eicosane to prevent torquing. Magnetization vs. field and temperature data were fit using the program MAGNET. 176 constants were used to estimate the diamagnetic correction, which was subtracted from the experimental susceptibility to give the molar paramagnetic susceptibility ( M ). High f requency electron paramagnetic resonance (HF EPR) data were collected at the U.S. National High Magnetic Field Laboratory Electron Magnetic Resonance facility. 4.3 Results and Discussion 4.3.1 Syntheses The synthesis of complex 4 1 is very similar to the p reparations of complexes 3 3 and 3 4 but with d p dH 2 in place of ded H 2 . The presence of I 2 is crucial for this reaction , i n fact, the reaction bleaches within 5 minutes without using I 2 . As in C hapter 3, I 2 has been proved to be a good oxidizing agent to pr event Mn III i o n s from the reduction to Mn II . The effectiveness of using I 2 on the formation of 4 1 ensures all Mn ions are Mn III while the reduced form of iodine, I 3 , is presented as the counter anion. The formation of complex 4 1 is summarized in equatio n 4 1. 2 [Mn 3 O(O 2 CMe) 6 (py) 3 ] + + 3 dpdH 2 6 O 2 (O 2 CMe) 6 (dpd) 3 ] 2+ + 6 MeCO 2 H + 6 py (4 1) 4.3.2 Description of Structures Complex 4 1 ·xCH 2 Cl 2 crystallizes in trigonal space group P 1c (see Supporting Information). The Mn 6 cation consists of two [Mn 3 ( 3 O)] 7+ units linked by three dpd 2 gr oups to give a supramolecular dimer [Mn 3 ] 2 (Fig ure 4 2). Similar to the previously reported [Mn 3 ] 4 molecules , the local structure of each Mn 3 unit of 4 1 is very similar to that of 3 , comprising a [Mn 3 ( 3 O)] 7+ triangular unit whose edges are each bridged by one acetate and one

PAGE 170

170 pyridyloximate group , and each of the 3 O 2 ions lies slightly above its Mn 3 plane ( d ~ 0.29 Å ), as in 3 . The Mn III oxidation states were confirmed by bond valence sum (BVS) calculation s for the crystallographically equivalent Mn io ns giving the value of 3.07 , and their Jahn Teller elongation axes (green bonds in Figure 4 2, bottom) are aligned in a propeller fashion, again as in 3. The cation has crystallographic D 3 symmetry with all Mn···Mn separations and Mn ( 3 O) Mn angles in ea ch triangle exactly the same , thus these Mn 3 triangles are equilateral and different from the virtually isosceles triangles in 3 and the rectangular 2 3 . The Mn 3 planes are parallel in each [Mn 3 ] 2 molecule and throughout the unit cell (Fig ure 4 3 ). T he sho rtest distance between Mn ions of two neighboring Mn 6 molecules is ~8.2 Å, at which the dipole dipole interaction can be neglected , and thus provides a promising candidate for the study of quantum mechanical coupling. 4.3.3 UV Vis S pectroscopy UV Vis spec tra of t hree solutions of complex 4 1 in acetonitrile with concentration M, M/2 and M/4 were obtain ed in the 200 800 nm range (M is an arbitrary concentration , the solution is dilute d to obtain peaks with peak height below the maximum threshold ) . The UV Vi s spectrum of complex 3 in acetonitrile was also recorded for comparison (Figure 4 4) . According to the Beer Lambert law, the absorbance of a solution is directly proportional to the concentration ( C ) of the absorbing species in the solution and the path l ength ( l ), 221 (4 2) where A is the measured absorbance, I 0 is the intensity of the incident light at a given wavelength, I is the transmitted intensity, and is the molar absorptivity (or ex tinction coefficient). Figure 4 5 shows the linear relationship between the absorbance and the concentration of complex 4 1 and thus confirm s that complex 4 1 is stable and not undergoing

PAGE 171

171 any reaction in solution. The distinct band around 300 nm can be ass igned to I 3 in 4 1 , which is not present ed in complex 3 . 4.3.4 Magnetochemistry 4.3. 4 .1 Direct current magnetic susceptibility studies Variable temperature, direct current (dc) magnetic susceptibility ( M ) measurements were performed on a vacuum dried microcrystalline sample of 4 1 , restrained in eicosane to prevent torquing, in an applied field of 1000 G (0.10 T) and in the 5.0 300 K temperature range. M T increases from 25.93 cm 3 K mol 1 at 300 K to a plateau value of 40.53 cm 3 K mol 1 at 15 K, and then slightly increase s to 45.70 cm 3 K mol 1 at 5.0 K (Figure 4 6). The 300 K value is much larger than the spin only ( g = 2) value for six Mn III atoms ( M T = 18 cm 3 K mol 1 ), and the plateau value at around 15K is as expected for two non interacting S = 6 units with g slightly less than 2.0 (spin only M T = 42 cm 3 K mol 1 ). At T < 15 K, the weak ferromagnetic coupling between two S = 6 spin units causes the observed increase of M T. The low temperature limit at 5 K is far from an S = 12 ground state spin ( M T = 78 cm 3 K mol 1 with g =2.0) due to the very weak inter Mn 3 ferromagnetic coupling , which caus es the presence of low lying spin states. A similar situation was found in a trimer of [MnCu] complex. 156 The mechanism for this ferromagnetic coupling is not clear yet but we believe it is due to the spin polarization in which the metallic centers are separated by an odd number of atoms (Figure 4 7) . The data from 300 15 K can be fit to the theoretical M T vs. T expression for two independent and equivalent Mn III 3 triangles . Since the Mn 3 units are crystallographically equilateral triangles , the first attempt is the fit for two non interacting equilateral triangles. The HDVV spin Hamiltonian for an equi lateral Mn 3 is given in equation 4 3 , which can be converted into the equivalent form in equation 4 4 using the Kambe method,

PAGE 172

172 = 2 J ( 1 · 1' + 1 · + 1' · ) (4 3) = J ( T 2 1 2 1' 2 1' 2 ) (4 4) where A = 1' + , and T = A + 1 ; S T is the total spin of the Mn 3 , taking values of S T = 0 6 since S 1 = S 1' = S 1'' =2. The eigenvalues of equation 4 4 are given by equation 4 5. E( S T ) = J [ S T ( S T + 1)] (4 5) Data below 15 K were neglected s ince the weak ferromagnetic interaction at very low temperature is not accommodated by eq uation 4 3 . T he fit gave J = +11.8(2) cm 1 , and g = 1.93(1), with TIP held constant at 600 x 10 6 cm 3 mol 1 . The fit quality for this model of two equilateral triangle s is not high as there are discrepancies between the data and the fit (Figure 4 8) . It is known that even crystallographically equ ilateral [M 3 O] triangles are known to undergo the magnetic Jahn Teller distortion resulting in an isosceles situation. 171 Thus, the model can be modified to two non interacting isosceles triangles. The HDVV spin Hamiltonian for an isosceles Mn 3 is given in equation 4 6 and its equivalent form in equation 4 7 = 2 J ( 1 · 2 + 1 · ) 2 J 2 · ( 4 6 ) = J ( T 2 A 2 1 2 ) J A 2 2 2 2 ) ( 4 7 ) where A = 2 + , T = A + 1 ; S T is the total spin of the Mn 3 unit , taking values of S T = 0 6 since S 1 = S 2 = S 2' =2. The eigenvalues of equation 4 7 are given by equation 4 8 . E( S T , S A ) = J [ S T ( S T + 1) S A ( S A + 1)] [ S A ( S A + 1)] (4 8 ) T he fit for data from 300 15 K is shown in Fi gure 4 9. It gave J = +6.8(1) cm 1 , J' = +26.3(4) cm 1 , and g = 1.95(1), with TIP held constant at 600 x 10 6 cm 3 mol 1 . These values are comparable to those for 3 and other relat ed Mn 3 supramolecular aggregates. To confirm t he global minimum of the fit , r oot mean square errors, which measure the differences between the

PAGE 173

173 M T values predicted by the model and the values actually observed, were calculated for different values of J and J' and are shown in Figure 4 10 . J and J' values reported above correspond to t he minimum having the lowest error value . To further investigat e the spin states of the molecule , an energy ladder plot was established (Figure 4 11). The ground state and the first excited state spins are S gs = 6 and S es = 5 with the energy difference of 79.4 cm 1 , again similar to 3 and other relat ed Mn 3 supramolec ular aggregates. 4.3.4.2 Magnetization versus DC magnetic field studies Magnetization ( M ) data were collected in the 0.1 7 T and 1.8 10 K ranges, and are plotted as M / N B vs. H / T in Figure 4 1 2 , where N B is the Bohr magneto in order to overcome the weak interactions between Mn 3 units . The data were fit, using the program MAGNET, by diagonalization of the spin Hamiltonian matrix assuming only the ground state is populated, incorporating axial anisotropy ( z 2 ) and Zeeman terms, and employing a full powder average. The spin Hamiltonian is given by equation 4 9 , = z 2 + g B 0 · H (4 9 ) where D is the axial ZFS paramet er, z is the easy axis spin operator, g is the electronic g factor, B is the Bohr magneton, 0 is the vacuum permeability, and H is the applied field. The fit (solid lines in Fig ure 4 1 2 ) gave S = 6, D = 0.24(1) cm 1 , and g = 1.89(1) with the goodness o f the fit R 2 = 0.998. The magnetization data at 1 T are placed slightly above the Brillouin curve indicating the presence of the weak ferromagnetic interaction between two Mn 3 units even at 1 T . 156 A similar fit can also be obtained with a positive value of D . To determine the global minimum of the fit, root mean square errors for the D vs. g fit were calculated by using the program GRID, 176 and are shown as an error surface in Figure 4 1 3 . The error values of the two

PAGE 174

174 minimum are quite comp arable; however, the error minimum associated with the negative D is lower and the fit is better . It is also noted that for an SMM, the D value must be negative. 4 .3.4.3 Alternating current magnetic susceptibility studies Alternating current (ac) suscepti bility data ( M and M ) for 4 1 were collected using a 3.5 G ac field oscillating at frequencies up to 1500 Hz and in the 1.8 15 K range. The increase of M T below 15 K (Figure 4 14) is again due to the weak ferromagnetic coupling between two Mn 3 unit s, which leads to a value higher than the spin only M T value for two non interacting S = 6 units but much lower than the spin only M T value for an S = 12 spin. At temperatures below 3 K, there are frequency dependent drops in the M T vs. T plots and con comitant appearance of out of phase ( M ) signals (Fig ure 4 1 4 ), which indicat e the slow relaxation of magnetic moment , the SMM behavior, when the magnetic moment cannot relax (reorient) fast enough to keep in phase with the oscillating field. An attempt to estimate the magnitude of the ferromagnetic interaction through dpd 2 groups can b e performed by simulating the in phase AC data using the MAGPACK program. 222 The model of two interacting S = 6 units and the simulation are given in Figure 4 1 5 . The g value was held at 1.91 . The simulations with different J values clearly show that the interaction through dpd 2 groups is weakly ferromagnetic with J 0.025 cm 1 = 0.035K ( J 0.050cm 1 = 0.070K in physics convention) . 4.3.4.4 High frequency electro n paramagnetic resonance (HF EPR) studies In order to determine the spin Hamiltonian parameters for each [Mn 3 ] cluster and the magnitude of the exchange interaction between [Mn 3 ] building blocks, we carried out HF EPR studies on a single crystal of compl ex 4 1 ·xCH 2 Cl 2 . Furthermore, we also carried out an HF EPR measurement on a solution of 4 1 to know if the dimers, and the exchange interaction, are still

PAGE 175

175 intact in solution. Single crystal EPR data were obtained from carefully oriented single crystals. Th e sensitivity of the measurements was enhanced via the use of a unique cavity, which additionally permitted sample rotation relative to the applied field direction. 223 The quality factor of the cavity at its fundamental TE 011 mode frequency of 51 GHz is 20 ,000. However, measurements were also feasible to frequencies well above 300 GHz on high order modes. The use of a cavity provides additional benefits in terms of being able to control the polarization of the microwave field (b 1 ) at the position of the sam ple, and in reducing instrumental problems that can give rise to distorted lineshapes. The use of a network analyzer further enables measurement s of both the in phase (absorption) and out of phase (dispersion) response of the sample . 223,224 T o avoid solvent loss from the samples , the single crystals were fished out directly from their mother liquor and immediately protected with paratone N oil before cooling under 1 atm of He gas. For solution samples, HF EPR spect ra were collected using a transmission probe in which microwaves are propagated through cylindrical lightpipes. High frequency microwaves were generated by a phase locked Virginia Diodes solid state source operating at 13 ±1 GHz, followed by a chain of mul tipliers and amplifiers. Microwaves detection was provided by a bolometer. High magnetic fields were provided by a 17 T superconducting magnet . 225 A s ingle crystal of 4 1 ·xCH 2 Cl 2 was oriented carefu lly such as the easy axis of the crystal was parallel to the external magnetic field B 0 . Assuming a ground state S = 6 and neglecting off diagonal crystal field terms, the effective spin Hamiltonian (to fourth order) for an external magnetic field applied parallel to the easy ( z ) axis of a single isolated SMM has the form in equation 4 10, (4 10)

PAGE 176

176 where is the z axis spin projection operator and the index i (= 1, 2) will be used to label the two Mn 3 molecules in the dimer for the interacting case; D (< 0) is t he uniaxial anisotropy constant, characterizes the fourth order axial anisotropy , and g z is the z component of the Land é g tenso r. The o mission of transverse terms in equation 4 10 does not affect the EPR spectra (they merely result in weak avoided level crossings, which cause the magnetic quantum tunneling). For the case of two exchange coupled SMMs, the effective dimer Hamiltonia n ( ) is given by equation 4 11, (4 11) where and are g iven by e q uation 4 10 , the last term describe s the exchange couplin g between the two SMMs within the dimer, and the J value characterize s the strength of this coupling. Eq uation 4 11 can be separated into the diagonal and off diagonal terms of equation 4 12 (4 12) Here the assumed isotropic Heisenberg exchange interaction of the form is separated into the exchange bias , which commutes with and , and a pe rturbation which is written in terms of the raising and lowering operators and . The eigenvectors for may be written as products of the single molecul e eigenvectors and ( abbreviated ), where m 1 and m 2 represent the spin projections of the two molecules within the dimer. The eigenvalues are t hen easily obtained by solving e q ua tion 4 10 separately for molecules 1 and 2, and adding the diagonal exchange term 2 J z m 1 m 2 , which serves as an exchange bias.

PAGE 177

177 As noted above, the diagonal exchange term does not alter the character of th e uncoupled SMM eigenstates of e q uation 4 10 . It sim ply causes additional contributions to the zfs which serve to translate the EPR peak positions. Since the eigenstates can still be labeled according to quantum numbers m 1 and m 2 , it is obvious that there is no coupling or entanglement of the spins. The ent anglement is brought abou t by the off diagonal terms in e q uation 4 12 , i.e. by the perturbation of . This interaction conserves the angular momentum of the dimer. Consequently, it operates only between states having the same total angular momentum M = m 1 + m 2 . Accordingly, the uncoupled basis states may be grouped into multiplets of M = m 1 + m 2 . In first order, raises the angular momentum of one spin within dimer by . Consequently, it connects states such as and and lifts the degeneracy between them. The induced splittings produce additional fine structure s in each HF EPR spectrum seen in Fig ure 4 16 (A) , which sh ows temperature dependen t spectra of a single crystal of comp lex 4 1 ·xCH 2 Cl 2 obtained at 148.67 GHz with the magnetic field applied parallel to the easy axis of the crystal. The i nset of Fig ure 4 16 (B) shows simulat ed spectra at T = 6 K with varying J (we use an isotropic J in this case). The top spectrum in Fig ure 4 16 (B) inset was generated assumin g that there was no interaction between the Mn 3 building blocks ( J = 0), i.e. it is essentially an S = 6 monomer spectrum. The simulated monomer spectrum is a typical spectrum for most SMMs, showing a series of more or less evenly spaced resonance s and a smooth variation in intensity from one peak to the next. In contrast, the experimental dimer spectra in Fig ure 4 1 6 (A) exhibit considerable complexity. Despit e this complexity, the simulated dimer spectra in Fig ure 4 16 (B) show re mark able agreement with the experimental data, both in peak positions and relative intensities. The parameters used to generate the simulat ed spectra are: D = 0.22 cm 1 , = 7 × 10 5 cm 1 , g z = 2, and J z = J xy = J = 0.0 2 5 cm -

PAGE 178

178 1 ( J z = J xy = J = 0.05 0 cm 1 in physics convention ) . Because of the off diagonal term s of the exchang e coupling, the eigenstates of e q uation s 4 11 or 4 12 can no longer be written as si mple products of the uncoupled basis states; they now correspond to real mixtures, or entangled superpositions. Some of these eigenfunctions are listed in Table 4 3 . Fig ure 4 16 ( C ) gives the assigned transitions between states (listed in Table 4 3 ) for re sonances in the spectr a . We noted that the positions of transitions d, e, and f are really close together and they formed a broad dip in the spectrum. This also happens with transitions h and i. HF EPR measurements on a toluene/acetonitrile (1 : 1 v/v) so lution of comp lex 4 1 were also performed. Fig ure 4 17 (A) shows the spectrum of 4 1 in solution recorded at 112 GHz and 2.5 K; for comparison it also shows the spectrum of a single crystal of 4 1 ·xCH 2 Cl 2 recorded at the same temperature of 2.5 K and a sl ightly different frequency 111.65 GHz. The spectrum of the solution is upside down compared to the spectrum of the single crystal because the solution spectrum was recorded in absorption mode while the single crystal spectrum was recorded in transmission m ode. Aside from this inversion , the solution spectrum is very similar to the single crystal spectrum. The splitting of resonance b and c indicates that there is interaction between [Mn 3 ] monomers even in solution. Figure 4 17 (B) shows the frequency depend ence of the main EPR peak positions (symbols) obtained at many microwave frequencies. The unfilled symbols represent EPR peak positions of a single crystal sample; meanwhile the filled symbols represent EPR peak positions of a solution sample. The solid li nes are simulated energy transitions. From Figure 4 17 (A) and (B), it can be concluded that the dimers are still intact in solution. 4.4 Summary and Conclusions In summary, the employment of the dpdH 2 dioxime has allowed for the isolation of a dimeric s upramol e cular aggregate of two exchange coupled SMMs . The formation of 4 1 as a Mn 3 dimer is expected from the retrosynthetic analysis based on the directional bonding

PAGE 179

179 approach . All physical studies confirm that each of the Mn 3 subunits is extremely simila r to that of the discrete 3 , and as a result, the magnetic properties are also nearly identical. T he inter Mn 3 coupling through the dpd 2 group s is ferromagnetic , which is seen for the first time in discrete supramolecular aggregates of exchange coupled S MMs. The coupling i s visible in both single crystal and solution HF EPR s pectra. In conclusion, complex 4 1 confirms the feasibility to direct and control the formation of supramolecular aggregates of SMMs with only weak coupling between them. Th e [Mn 3 ] 2 dimer represents the next generation of quantum mechanically coupled supramolecular SMMs and is ready for deposition on surfaces for device studies. T he success of this study also encourages us to reconsider other well studied SMM families such as Mn 4 or Mn 12 to obtain new supramolecular aggregates of SMM s.

PAGE 180

180 Table 4 1. Crystallographic data for complex 4 1 · x CH 2 Cl 2 . Parameters 4 1 · xCH 2 Cl 2 Formula C 51 H 48 I 6 Mn 6 N 12 O 20 FW, g/mol 2240.05 Crystal system Trigonal Space group P 1 c a, Å 16.5692(10) b, Å 16.5692(10) c, Å 18.2847(11) ,° 90 ,° 90 ,° 120 V, Å 3 4347.3(6) Z 2 T, K 100(2) , Å a 0.71073 cal , mg/m 3 1.711 , mm 1 3.036 R 1 b,c 0.0496 wR 2 d 0.1383 a Graphite monochromator. b I>2 (I). c R 1 o | |F c o |. d wR 2 o 2 F c 2 ) 2 ] / o 2 ) 2 ]] 1/2 2 (F o 2 )+(m × p) 2 + n × p], p = [max(F o 2 ,0)+ 2 × F c 2 ]/3, m & n are constants. Table 4 2 . Selected interatomic distances (Å), bond angles ( o ), displacements (Å) of 3 oxide ions from Mn 3 planes, and Mn N O Mn torsion angles ( o ) of complex 4 1 . Bond distances (Å) Bond angles ( o ) Mn1 ··· Mn1' 3.2005(7) Mn1 Mn1' Mn1" 60.0 Mn1 O3 1.8704(6) Mn1 O3 Mn1' 117.64(5) Mn1 O2 1.935(2) Mn1 N2 2.012(2) Mn1 N1 2.033(2) Mn1 O1 2.1670( 19) Mn1 O4 2.200(2) d (Å) Torsion angles ( o ) O3 to Mn1Mn1'Mn1" plane 0.29 Mn1' N2 O2 Mn1 3.134(4)

PAGE 181

181 Table 4 3. Several lowest eigenfunctions of the [Mn 3 ] 2 dimer. Only the lowest 5 multiplets are listed . Label M (= m 1 + m 2 ) Eigen function (1) S 1 2 (2) S 11 (3) A 11 (4) S 10 (5) A 10 (6) S 10 (7) S 9 (8) A 9 (9) S 9 (10) A 9 (11) S 8 (12) A 8

PAGE 182

182 Figure 4 1. S tructure s of mpkoH (top le ft) , dedH 2 (top right) , dpdH 2 (bottom left) , and pdpdH 2 (bottom right).

PAGE 183

183 Figure 4 2 . (Top) Structure of the cation of complex 4 1 in side view (left) and top view (right), (middle) a stereopair, (bottom) the core emphasizing the corresponding con nectivity between Mn 3 sub units and showing the Mn 3 planes and Jahn Teller axes (green bonds). H atoms have been omitted for clarity. Color code: Mn III green; O red; N blue; C grey.

PAGE 184

184 Figure 4 3. Packing of cations 4 1 within one unit cell in side view ( top) and top view (bottom).

PAGE 185

185 Figure 4 4. UV Vis spectra of complex 4 1 at different concentrations and complex 3 in acetonitrile. Figure 4 5 . Plot of absorbance of the 360 nm band vs. concentration of complex 4 1 showing linear relationship .

PAGE 186

186 Figure 4 6. M T vs. T for complex 4 1 . Figure 4 7 . Ferromagnetic coupling by spin polarization mechanism through dpd 2 group s . The spin vectors were shown for one dpd 2 group.

PAGE 187

187 Figure 4 8 . M T vs. T for complex 4 1 . The solid line is the fit to the model of two no n interacting equilateral triangles. Figure 4 9 . M T vs. T for complex 4 1 . The solid line is the fit to the model of two non interacting isoceles triangles.

PAGE 188

188 Figure 4 10. Root mean square errors vs. J and for the fit to Van Vlek equation. Figure 4 11. Energy ladder plot for complex 4 1 . Ground state is S T = 6. The first excited state S = 5 lies 79.5 cm 1 above the ground state. *

PAGE 189

189 Figure 4 1 2 . M / N B (per Mn 3 ) vs. H / T for complex 4 1 . The solid lines are the fits to the data. See the text for the fit parameters.

PAGE 190

190 Figure 4 1 3 . Representations of the error surface for the D vs. g fit for complex 4 1 . Top: t wo dimensional contour plot , bottom: three dimensional mesh plot. *

PAGE 191

191 Figure 4 1 4 . Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature for a micro crystalline sample of complex 4 1 in a 3.5G field oscillating at the indicated frequencies.

PAGE 192

192 Figure 4 15. Simulations of the 5 Hz in phase AC magne tic susceptibility data by the program MAGPACK. The g value was held at 1.91. The last two data points were not included in the simulation s . S = 6 S = 6 J

PAGE 193

193 Figure 4 16. Comparison between experimental (A) and simulated (B) EPR spectra (B || z) for the [Mn 3 ] 2 dimer at 148.67 GHz; the simulation assumed D = 0.22 cm 1 , = 7 × 10 5 cm 1 , g z = 2, and J z = J xy = J = 0.025 cm 1 . The resonances have been labeled to show the assigned transitions between states that are shown in schematic energy level diagram (C). The (A) inset is a simulated [Mn 3 ] monomer spe ctrum. The (B) inset shows the evolution of the spectrum as J increases.

PAGE 194

194 Figure 4 1 7 . (A) Comparison between a spectrum of complex 4 1 in solution and a spectrum of a single crystal of 4 1 ·xCH 2 Cl 2 . (B) Frequency dependence of the main EPR peak posi tions (filled symbols: single crystal; unfilled symbols: solution) obtained at many microwave frequencies. The solid lines in the plot (B) are simulated energy transitions.

PAGE 195

195 CHAPTER 5 SUPRAMOLECULAR AGGREGATES OF SINGLE MOLE CULE MAGNETS FROM THE USE OF DI AND TRICARBOXYLIC ACIDS 5.1 Introduction In C hapters 2, 3, and 4, the discrete supramolecular aggregates of Mn 3 SMMs were formed from the use of dioximes , which was based on the observation that the three mono oximate groups in complex 3 stay on the same side of the Mn 3 plane. The formation of these supramolecules and their magnetic properties were found to depend on the structure and the flexibility/rigidity of the ligands. The same strategy can be appl ied to the three mono carboxylate groups on the other side of the Mn 3 plane, which require s the employment of di , tri , or tetracarboxylic acids . Many of them , some are shown in Figure 5 1, are commercially available or can be synthesized, and have been widely employed in metal organic framework (MOF) chemistry. 226,227 It is worth noting that more than 20,000 MOFs have been reported within the past decade , 228 and the library of di , tri , or tetracarbo xylic acids employed in MOF chemistry can amaze anyone. Ca r boxylate substitution is well SMMs such as the Mn 4 and Mn 12 families . 178,179 For example, the radical 1 [N t butyl N (oxyl)amino] 4 benz oic acid was used to introduce paramagnetic carboxylate groups onto the Mn 12 molecule ; 229 substitution of acetate with stearate groups allows the fo rmation of a strongly hydrophobic Mn 12 derivative ; 230 the trifluoroacetate Mn 12 derivative was found to be more volatile than the original Mn 12 Ac , thus it is superior for thin film preparation on surfaces; 231 when acetate groups are replaced by naphthalene carboxylate groups, the obtained Mn 12 derivati ve assemble s into a two dimensional network via interactions; 232 deposition of Mn 12 molecules onto gold surfaces or insertion between gold electrodes was performed by employ ing sulfur based carb oxylate groups. 233,234 In

PAGE 196

196 principle, carboxylate ligands on Mn molecules are labile and can be substituted by other carboxylates; however , the s ubstitution is an equilibrium reaction that its e xtent depends on the relative acidity of the incoming and outgoing carboxylic acids. If the incoming acid is significantly more acidic than the outgoing acid, multiple treatments employing an excess of the incoming acid can drive the substitution to comple tion. In case that the incoming acid has a pKa comparable to, or higher than that of the outgoing acid, the equilibrium will favor the reactan t s and can only be driven to the product s by mechanically removing the outgoing acid. It can be done for acetic ac id (boiling point 118.5 o C ) as its azeotrope with toluene has significantly lower boiling point, 105.0 o C, and is easily evaporated under vacuum. Thus, the substitution of acetate groups on Mn molecules by other carboxylate groups can be performed for almo st all cases. In this pres ent work, the possibility of di and tricarboxylates as linkers to join Mn 3 SMMs was investigated. We will discuss the use of trimesic acid (tmaH 3 ) , dimethylmalonic acid (dmaH 2 ) truxillic acid (txaH 2 ) (Figure 5 2) and their structural features leading to different supramolecular aggregates of Mn 3 SMMs. The exchange biased interaction resulting from weak couplings between Mn 3 units will also be discussed. 5.2 Experimental Section 5.2.1 Syntheses All preparations were perform ed under aerobic conditions using reagents and solvents as received , unless otherwise stated. [Mn 3 O(O 2 CMe) 3 (mpko) 3 ](ClO 4 ) ( 3 ) was synthesized as reported elsewhere. 171 T rimesic acid and dimethylmalonic acid are commercially available; truxillic acid was prepared following a reported procedure based on the photodimerization of trans cinnamic acid . 235

PAGE 197

197 Caution! Although no such behavior was observed during the present work, perchlorate compounds are potentially explosive; su ch compounds should be synthesized and used in small quantities, and treated with utmost care at all times. 5.2.1.1 [Mn 12 O 4 (tma) 4 (mpko) 12 ] (ClO 4 ) 4 (5 1) A dark brown solution of [Mn 3 O(O 2 CMe) 3 (mpko) 3 ](ClO 4 ) (0.17 g, 0.2 0 mmol) in 15 m L of MeCN : MeOH (2 : 1 v/v) was treated with tmaH 3 (0.042 g, 0 . 2 0 mmol). The solution was stirred for overnight at room temperature and then evaporated under vacuum to obtain a black powder. Toluene (10 mL) was added and the solution was again evaporated to dryness. The additi on and removal of toluene w ere repeated for total three times. The resulting dark brown solid was redissolved in a mixture of MeCN/EtOH/ MeNO 2 (30 m L , 1 : 2 : 2 v/v/v ) ; the obtained solution was filtered, and the filtrate was layered with CHCl 3 ( 60 m L ). Aft er few days, X ray quality crystals of 5 1 · x CHCl 3 were collected by filtration, washed with CHCl 3 (2 x 5 m L ), and dried under vacuum; the yield was 4 0%. Anal.Calcd (found)% for 5 1 · 3CHCl 3 (Mn 12 O 5 6 C 1 23 H 99 N 2 4 Cl 13 ): C 37. 60 (37. 54 ); H 2 . 54 ( 2 . 34 ); N 8 . 55 ( 8 . 38 ); Cl 11 . 73 ( 11 .1 3 ). Selected IR data (cm 1 ): 1616 ( s ), 160 4 (s), 157 2 (s), 14 76 ( m ), 1440 (m), 138 3 ( m ), 13 44 (s), 118 0 (m), 11 62 (m), 1109 (m), 10 86 (s), 104 8 ( w ), 93 5 (w), 845 (s), 77 4 ( w ), 75 1 ( m ), 7 15 (m), 699 (m), 663 ( m ), 62 0 (m), 557 (m), 489 ( w), 452 (w), 413 (w). 5.2.1.2 [Mn 6 O 2 (dma) 3 (mpko) 6 ](ClO 4 ) 2 (5 2) A dark brown solution of [Mn 3 O(O 2 CMe) 3 (mpko) 3 ](ClO 4 ) (0.17 g, 0.2 0 mmol) in 15 m L of MeCN : MeOH (2 : 1 v/v) was treated with dmaH 2 (0. 040 g, 0.3 0 mmol). The solution was stirred for overnigh t at room temperature and then evaporated under vacuum to obtain a black powder. Toluene (10 mL) was added and the solution was again evaporated to dryness. The addition and removal of toluene w ere repeated for total three times. The resulting dark brown s olid was redissolved in a MeNO 2 / MeCN mixture (30 m L 1 : 1 v/v ) ; the obtained solution was

PAGE 198

198 filtered, and the filtrate was layered with Et 2 O ( 60 mL ). After three days, X ray quality crystals of 5 2 · x MeNO 2 · yMeCN · zEt 2 O were collected by filtration, washed with Et 2 O (2 x 5 m L ), and dried under vacuum; the yield was 2 0 %. Anal.Calcd (found)% for 5 2 · 2H 2 O (Mn 6 O 30 C 57 H 64 N 12 Cl 2 ): C 3 8 .0 8 (37. 6 4); H 3 .5 9 ( 3 .3 7 ); N 9 . 3 5 (8. 95). Selected IR data (cm 1 ): 1601 ( s ), 1 532 ( w ), 14 74 ( w ), 1403 (w), 13 57 ( w ), 13 11 ( m ), 11 74 (w ), 11 56 ( w ), 1107 (s), 10 82 (s), 1044 (m), 10 48 ( w ), 885 (w), 78 2 ( w ), 698 (w), 664 ( w ), 62 7 (m), 562 ( w ). 5.2.1. 3 [Mn 6 O 2 (txa) 4 (mpko) 6 ] (5 3) A dark brown solution of [Mn 3 O(O 2 CMe) 3 (mpko) 3 ](ClO 4 ) (0.17 g, 0.2 0 mmol) in 15 m L of MeCN : MeOH (2 : 1 v/v) was treated with txaH 2 (0.12 g, 0.4 0 mmol). The solution was stirred overnight at room temperature and then evaporated under vacuum to obtain a black powder. Toluene (10 mL) was added and the solution was again evaporated to dryness. The addition and removal of toluene were repeated for total three times. The resulting dark brown solid was redissolved in a MeCN/ EtOH mixture (30 m L 2 : 1 v/v); the obtained solution was filtered, and the filtrate was layered with Me 2 CO/ hexane ( 60 mL , 1 : 1 v/v ). After few day s , X ray quality crystals of 5 3 · x Me 2 CO · yMeCN were collected by filtration, washed with Et 2 O (2 x 5 m L ), and dried under vacuum; the yield was 3 5%. Anal.Calcd (found)% for 5 3 (Mn 6 O 24 C 114 H 98 N 12 ): C 58.27 (58.44); H 4.20 (4.07); N 7.15 (7.44). Selected I R data (cm 1):1601 (s), 1578 (s), 1496 (w), 1477 (m), 1385 (s), 1336 (m), 1262 (w), 1182 (m), 1162 (m), 1108 (m), 1084 (w), 1046(w), 775 (w), 756 (w), 700 (m), 663(w), 621(m), 604 (m). 5.2.2 X Ray Crystallography X Ray Intensity data were collected at 100 K on a Bruker DUO diffractometer using MoK radiation ( = 0.71073 Å) and an APEXII CCD area detector. Suitable crystals of 5 1 · xCHCl 3 , 5 2 · xMeNO 2 · yMeCN · zEt 2 O, and 5 3 · xMe 2 CO · yMeCN were attached to glass fibers

PAGE 199

199 using silicone grease and transferred to a g oniostat where they were cooled to 100 K for data collection. Raw data frames were read by the program SAINT and integrated using 3D profiling algorithms. The resulting data were reduced to produce hkl reflections, their intensities and estimated standard deviations. The data were corrected for Lorentz and polarization effects, and numerical absorption corrections were applied based on indexed and measured faces. The structures were solved and refined in SHELXTL6.1, 175 using full matrix least squares refinement. The non H atoms were refined with anisotropic thermal parameters and all of the H atoms were calculated in idealized positions and refined riding on their parent atoms. For 5 1 · xCH Cl 3 , t he asymmetric unit consists of a Mn 12 cluster, four perchlorate anions , and twenty four chloroform solvent molecules. The solvent molecules were estimated based on the electron density map and the SQUEEZE total electron prediction. All solvent mol ecules as well as the perchlorate anions could not be modeled and refined properly , thus the program SQUEEZE, a part of the PLATON package of crystallographic software, was used to calculate the solvent and anion disorder area and remove its contribution t o the overall intensity data. T he disordered chloroform molecule at the center of the Mn 12 tetrahedral structure was kept in the final refined model. In the final cycle of refinement, 19060 reflections (of which 13452 are 890 parameters and the resulting R 1 , wR 2 , and S (goodness of fit) were 18.17 %, 41.91 % , and 2.430 , respectively. For 5 2 . xMeNO 2 . yMeCN . zEt 2 O , the asymmetric unit consists of two Mn 6 clusters , four perchlorate anions , and an estimated one diethylether, twelve nitromethane, and two acetonitrile solvent molecules. Most of the solvent mol ecules were disordered and could not be modeled properly, thus the program SQUEEZE, a part of the PLATON package of crystallographic software, was used to calculate the solvent disorder area and remove its contribution to the

PAGE 200

200 overall intensity data. Each M n 6 molecule exhibits significant disordered reg ions. The three Mn centers Mn4Mn5 Mn6 and Mn10Mn11Mn12 are disordered along with all ligands attached to them. After refining all disordered parts, their site occupation factors refined to 0.5 , thus this value was fixed in the final cycles of refinement. In the final cycle of refinement, 33299 reflections (of which 22849 are observed with I > 2 (I)) were used to refine 1665 parameters and the resulting R 1 , wR 2 , and S (goodness of fit) were 5.79 %, 15.68 % , and 1.041 , respectively. For 5 3 . xMe 2 CO · yMeCN , the asymmetric unit consists of a half Mn 6 cluster, four acetone , and four acetonitrile solvent molecules. All solvent molecules were disordered and could not be modeled properly, thus the program SQUEEZE, a part of the PLATON package of crystallographic software, was used to calculate the solvent disorder area and remove its contribution to the over all intensity data. The number and the identity of the solvent molecules were determined as a compromise between the observed electron density and the total number of electrons calculated by the SQUEEZE program. In the final cycle of refinement, 12209 refl ections (of which 10780 are observed with I > 2 (I)) were used to refine 703 parameters and the resulting R 1 , wR 2 , and S (goodness of fit) were 3.37 %, 9.34 % , and 1.084 , respectively. Unit cell data and structural refinement details for the three compounds are listed in Table 5 1. 5.2.3 Physical Measurements Infrared spectra were recorded in the solid state (KBr pellets) on a Nicolet Nexus 670 FTIR spectrometer in the 400 4000 cm 1 range. Elemental analyses (C, H, and N) were performed by the in house fa cilities of the University of Florida, Chemistry Department. Variable temperature direct current (dc) and alternating current (ac) magnetic susceptibility data were collected at the University of Florida using a Quantum Design MPMS XL SQUID magnetometer eq uipped with a 7 T magnet and operating in the 1.8 300 K range. Samples were

PAGE 201

201 embedded in solid eicosane to prevent torquing. Magnetization vs. field and temperature data were fit using the program MAGNET. 176 diamagnetic correction, which was subtracted from the experimental susceptibility to give the molar paramagnetic susceptibility ( M ). Low temperature (<1.8 K) hysteresis loop and dc relaxation measurements were performed at Institut Néel using an array of microSQUIDS. 177 The high sensitivity of this magnetometer allows the study of single crystals of SMMs of the order of 10 500 m. The field can be applied in any direction by separately driving three orthogonal coils. 5.3 Results and D iscussion 5.3.1 Syntheses The syntheses of complexes 5 1 , 5 2 , and 5 3 were based on the substitution of acetate s on complex 3 by the corresponding di or tricarboxylate groups. Toluene was used to form the azeotrope with acetic acid, which wa s removed by evaporation under vacuum to drive the equilibrium to the product. The formation s of these complexes are summarized in equation s 5 1, 5 2 , and 5 3 : 4[Mn 3 O(O 2 CMe) 3 (mpko) 3 ] + + 4tmaH 3 12 O 4 (tma) 4 (mpko) 12 ] 4+ + 12MeCO 2 H (5 1) 2[Mn 3 O(O 2 CMe) 3 (mpko) 3 ] + + 3dm aH 2 6 O 2 (dma) 3 (mpko) 6 ] 2+ + 6MeCO 2 H (5 2) 2[Mn 3 O(O 2 CMe) 3 (mpko) 3 ] + + 4txaH 2 6 O 2 (txa) 4 (mpko) 6 ] + 6MeCO 2 H + 2H + (5 3 ) The stoichiometric ratio of [Mn 3 ] : acid was strictly followed in the experiments. I n fact, it is noted that using an excess amount of di or tricarboxylic acids significantly reduces the solubility of the black powder obtained after evapo ra tion under vacuum. EtOH was used along MeCN as the solvent for these reactions since it helps to dissolve the di or tricarboxylic acids co mpletely . Cho o sing the right combination of sol vents for crystalization appeared to be the most challenging part to obtain X ray quality crystals, especially for complex 5 1 . It is

PAGE 202

202 interesting that the direct reaction between Mn(ClO 4 ) 2 .xH 2 O, mpkoH, TBAMnO 4 , and tmaH 3 can also lead to the isolation of complex 5 1 but there are Mn II ions cocrytstallized with 5 1 in the crystal. The carboxylate substitution on the Mn 3 starting material has the advantage over the direct reaction in which it prevents the presenc e of par amagnetic impurities such as Mn II ions in the product. Finally, it is worth noting that the success of these carboxylate substitution s is also due to the robustness of the Mn 3 molecule as the three tridentate mpko ligand s keep the molecule stable in the condition of the substitution reactions. 5.3.2 Description of Structures Labeled structures of complexes 5 1 , 5 2 , and 5 3 are shown in Figure s 5 3, 5 4, and 5 5, respectively. Some selected interatomic distances and angles are listed in Table s 5 2 and 5 3 . Complex 5 1 ·xCHCl 3 crystallizes in monoclinic space group C 2/c; complex 5 2 ·xMeNO 2 ·yMeCN·zEtO 2 crystallizes in triclinic space group with the asymmetric unit composed of two essentially identical Mn 6 clusters; complex 5 3 ·xMe 2 CO·yMeCN crystallizes in triclinic space group . The structure of 5 1 consists of one Mn 12 4+ cation, which is shown in Figure 5 3. The Mn 12 cation compo sed of four [Mn 3 ( 3 O)] 7+ units linked by four tma 3 groups to give a supramolecular [Mn 3 ] 4 4+ tetrahedron. The structure of 5 2 consists of one Mn 6 2+ cation ( Figure 5 4), which is a dimeric supramolecular aggregate of two [Mn 3 ( 3 O)] 7+ units linked by thr ee dma 2 groups. The structure of 5 3 is also a dimeric [Mn 3 ] 2 supramolecular aggregate (Figure 5 5) but the two [Mn 3 ( 3 O)] 7+ units are bridged by only two txa groups due to the steric effect of two bulky phenyl groups , with two additional txa binding only through one carboxylate at each Mn 3 . In all three complexes, each edge of the [Mn 3 ( 3 O)] 7+ triangular unit is bridged by one 1 : 1 : carboxylate grou p and one 1 : 1 : 1 : pyridyloximate groups. Each of the 3 O 2 ions lies slightly above its Mn 3 plane (d ~ 0.3 Å). Thus, the local structure of each

PAGE 203

203 Mn 3 unit is very similar to that of 3 , comprising a [Mn 3 ( 3 O)] 7+ triangular unit whose edges are eac h bridged by one acetate and one pyridyloximate group. The structural affects such as the flexibility/rigidity or bulkiness of the ligands dictate the formation of these supramolecules. The rigid tma 3 and the 109 o ditopic dma 2 forced the formation of the only possible products, a tetrahedral molecule for the former and a dimer for the latter . With the bulky txa 2 , we would expect to see a rectangular topology, as we had previously seen for the dioximate bridged [Mn 3 ] 4 , or something else. A dimeric supramo lecule is a reasonable product , certainly due to the bulkiness of the phenyl groups. Both the cation s of 5 1 and 5 2 ha ve crystallographic C 1 but they possess virtual T and D 3 point group symmetry , respectively . The complete molecule of 5 3 has crystallogr aphic C 2 point group symmetry. The Mn III oxidation states were confirmed by bond valence sum (BVS) calculations, and their Jahn Teller elongation axes (green bonds in Figures 5 3, 5 4, and 5 5 ) are aligned in a propeller fashion, again as in 3 . In all thre e complexes, the Mn···Mn separations and Mn ( 3 O) Mn angles in Mn 3 triangles are slightly different , thus the triangles are scalene but virtually isosceles within the usual 3 criterion. In complexes 5 2 and 5 3 , the Mn 3 units are parallel in each Mn 6 cat ion/molecule. However, i nvestigating the packing of Mn 6 cations in the unit cell of 5 2 reveals that there are two sets of Mn 6 cations ori ented in different di rections (Figure 5 6). Only in complex 5 3 the Mn 3 planes are parallel throughout the crystal ( Fi gure 5 7 ). 5.3.3 Magnetochemistry 5.3.3.1 D irect current magnetic susceptibility studies Variable temperature, dc magnetic susceptibility ( M ) measurements were performed on vacuum dried polycrystalline samples of complexes 5 1 · 3CHCl 3 , 5 2 · 2H 2 O, and 5 3 in an applied field of 1000 G (0.10 T) and the 5.0 300 K temperature range. The samples were

PAGE 204

204 restrained in eicosane to prevent torquing. Fi gure 5 8 shows the molar magnetic susceptibility ( M ) of complexes 5 1 , 5 2 , and 5 3 as M T vs. T plots. The M T value for complex 5 1 is 43.98 cm 3 K mol 1 at 300 K, which steadily increases with decreasing temperature to73.19 cm 3 K mol 1 at 10 K and sligh tly decreases to 72.07 cm 3 K mol 1 at 5 K. The value at 300 K is larger than the spin only ( g = 2) value for twelve Mn III non interacting ions ( M T = 36 cm 3 K mol 1 ), indicating dominant ferromagnetic interactions within the molecule. T he plateau values at low T are as expected for four non interacting S = 6 units with g slightly less than 2.0 (spin only, g = 2, 4× M T S=6 = 84 cm 3 K mol 1 ). Both complexes 5 2 and 5 3 are dimers of Mn 3 ; therefore, their M T vs. T profiles are similar. The M T value of comple x 5 2 increases from 23.54 cm 3 K mol 1 at 300 K to a plateau value of 38.29 cm 3 K mol 1 at 20 K, and then slightly decreases to 36.66 cm 3 K mol 1 at 5 K. For 5 3 , t he M T value increases from 23.56 cm 3 K mol 1 at 300 K to a plateau value of 38.27 cm 3 K mol 1 at 25 K, and then slightly decreases to 35.31 cm 3 K mol 1 at 5 K. The M T values at 300 K for both complexes are much larger than the spin only ( g = 2) value for six Mn III non interacting ions ( M T = 18 cm 3 K mol 1 ), indicating dominant ferromagnetic in teractions within the molecules . The plateau M T values at low temperature suggest the behavior of two non interacting S = 6 units with g slightly less than 2.0 (spin only, g = 2, 2× M T S=6 = 42 cm 3 K mol 1 ). Exchange interaction parameters J ij between Mn i Mn j pairs could be determined when the data in each complex were fit to the theoretical M T vs . T expression. For complex 5 1 , the theoretical M T is the sum of M T of four independent Mn 3 subunits, assuming the interaction through ligand tma 3 is insigni ficant at temperatures above 5K . Similarly, the theoretical M T for complex 5 2 and 5 3 is the sum of M T of two Mn 3 subunits with no interaction through dma 2 and txa 2 groups. We used a Mn III 2 Mn III isosceles triangle model for each Mn 3 subunit in all thr ee

PAGE 205

205 com p lexes. There are two exchange parameters J in an isosceles triangle, e.g. Mn1Mn2Mn2'; the HDVV spin Hamiltonian is given by equation 5 4 , 2 J ( 1 · 2 + 1 · ) 2 J 2 · (5 4) where J = J 12 = J 12' , J' = J 22' . Using A = 2 + 2' and T = A + 1 in the Kambe method, 181 where T is the total spin of the Mn 3 , the spin Hamiltonian becomes = J ( T 2 A 2 1 2 ) J' ( A 2 2 2 2' 2 ) (5 5) The eigenvalues of the spin Hamiltonian are determined using equation 5 6 , E( S T , S A ) = J [ S T ( S T +1) S A ( S A +1)] J' [ S A ( S A +1)] (5 6) where E( S T , S A ) is the energy of state S T arising from S A . For all three complexes, S 2 = S 2' = S 1 = 2, S A ranges from 0 to 4, and S T ranges from 0 to 6. A theoretical M vs. T expression was derived from the use of t he Van Vleck equation 182 and was modified to include temperature independent paramagnetism (TIP), which was kept constant at 600 × 10 6 cm 3 K mol 1 . Good fits of M T vs. T data for all three complexes are shown as solid lines in Figure 5 8. The goodness of the fit s is evident from the low statistical error, R 2 > 0.99. The data at low temperature (e.g. below 20 K) were omitted because the low T values are affected b y factors not included in equation 5 4. The fit parameters are given in Table 5 8. To investigate whether each fit reach ed the global minimum, root mean square errors, which measure the differences between the M T values predicted by the model and the va lues actually observed, were calculated for different values of J and J' , and are shown in Figure s 5 9, 5 10, and 5 11 as error surface plots for complexes 5 1 , 5 2 , and 5 3 , respectively. In all three complexes, the error surfaces show two comparable mini mum. The reported J and J' values in T able 5 8 are associated with the minimum that has the lowest root mean square error. To further inves tigate the spin states of each complex , energy ladder plot s were established (Figure s

PAGE 206

206 5 1 2 , 5 13, and 5 14 ). The gro und state and the first excited state spins for complexes 5 1 , 5 2 , and 5 3 are S gs = 6 and S es = 5 with the energy difference s of 67.5 cm 1 , 79.5 cm 1 , and 79.5cm 1 , respectively. 5.3.3.2 M agnetization versus DC magnetic field stud ies M agnetization (M) data were collected in the 0.1 7 T and 1.8 10 K ranges, and were plotted for complex es 5 1 , 5 2 , and 5 3 as M/N B vs. H/T in Figure s 5 1 5, 5 16, and 5 1 7 , repec tively, where N B is the Bohr magneton. The data were fit, using t he program MAGNET, by diagonalization of the spin Hamiltonian matrix assuming only the ground state is populated, incorporating axial anisotropy ( z 2 ) and Zeeman terms, and employing a full powder average. The spin Hamiltonian is given by equation 5 7 , wh ere D is the axial ZFS parameter, 0 is the vacuum permeability and H is the applied field. z 2 + g B 0 ( 5 7 ) The fit parameters are given in Table 5 9 . The goodness of the fit s is evident f rom the low statistical error, R 2 > 0.99. Alternative fits with different S values were discarded because of unreasonable g values. For all complexes, similar fits can also be obtained with positive values of D . To determine the global minimum in each fit, we established error surfaces for the D vs. g fits by using the program GRID, 176 which calculated the root mean square errors between the experimental M/N B data and those calculated for various combinations of D and g . In all cases, the error surface displays a double minimum with positive and negative D values. The error minimum associated with the negative D is substantially lower than the one with posit ive D (Figure s 5 18, 5 19, and 5 20) , which confirms that the fits for negative D are better. T hese fits prove that complexes 5 1 , 5 2 , and 5 3 are tetrameric and dimeric versions of S = 6 SMMs.

PAGE 207

207 5.3.3.3 A lternating current magnetic susceptibility studies Alternating current (ac) magnetic susceptibility measurements were performed in the 1.8 15 K temperature range in a 3.5 G ac field oscillating at 50 1500 Hz. In the ac susceptibility experiment, the ac magnetic field is oscillating at a particular freq uency and a peak in the out of phase M versus T plot is observed when the magnetic moment of the molecule cannot relax (reorient) fast enough to keep in phase with the oscillating field. The plots of in phase ( M , plotted as M T ) and out of phase ( M ) ac susceptibility data are shown in Figure s 5 21, 5 22, and 5 23. The in phase M T values above approximately 7 K are almost constant , which can be e xtrapolat ed to 0 K to determine the ground state of the molecule. For complex 5 1 , extrapolation of the data above 7 K to 0 K gives the value of 7 3 cm 3 K mol 1 , which correspond s to the sum M T of four independent S = 6 Mn 3 ( spin only, g = 2, 4 × M T S = 6 = 84 cm 3 K mol 1 ), with g slightly smaller than 2. For complex es 5 2 and 5 3 , extrapolation of the data above 7 K to 0 K leads to a value of 38 cm 3 K mol 1 , which corresponds to two non interacting S = 6 Mn 3 with a g value slightly less than 2 .0 ( g = 2, 2 × M T S =6 = 42 cm 3 K mol 1 ). Th ese results are consistent with those obtained from the dc magnetization studies. The out of phase M signals of a ll three complexes that are tails of peaks lying below 1.8 K were observed at < 3 K suggesting that they are supramolecular aggregates of Mn 3 SMMs. 5.3.3.4 Magnetization versus DC field hysteresis loops Complex 5 3 has all Mn 3 planes parallel throughout th e crystal and is ideal for studies of quantum properti es. To investigate the exchange interaction between Mn 3 units in complex 5 3 , dc magnetization vs. field scans on single crystals of 5 3 ·x Me 2 CO · yMeCN were carried out on a micro SQUID. 177 Hysteresis loops were observed below ~1.0 K (Figure 5 24 ) , whose coercivities increase with decreasing temperature and inc reasing field sweep rate, as expected for

PAGE 208

208 SMMs. The blocking temperature is 1.0 K, above which there is no hysteresis because the spin relaxes faster than the time scale of the hysteresis loop measurement. We can thus conclude that each of the two Mn 3 uni ts in 5 3 · xMe 2 CO · yMeCN is an SMM, as is the corresponding Mn 3 complex 3 that they closely resemble in both structure and magnetic properties. For normal SMMs, the first step in the hysteresis loop on scanning from negative to positive fields is at zero field , where the M S levels on either side of the anisotropy barrier are in resonance and QTM can occur, reversing the orientation of the magnetization vector. The presence of an AF exchange coupled neighbor provides a bias field that shifts the resonan t tu nneling (QTM step) to a new position before zero field. Considering a dimer of S = 6 SMMs and applying a similar mathematical treatment used in C hapter s 2 and 3, the spin Hamiltonian can be estimated as given in equation 5 8 , 1 + 2 2 J ( z1 z2 ) (5 8) i = z,i 2 + g B 0 zi . H z (5 9) where i = 1 2 (referring to the two Mn 3 SMMs of the dimer), J is the exchange biase d coupling parameter . The eig envalue s of equation 5 8 are given as follows E (m 1 , m 2 ) = D (m 1 2 + m 2 2 ) g B 0 H z (m 1 + m 2 ) 2 J (m 1 m 2 ) ( 5 1 0 ) T he hysteresis loops of complex 5 3 · xMe 2 CO · yMeCN show a step before zero field but the exchange biased field is weak, 0.0 5 T. The first step occurs when E( 6, 6) = E( 6, 6). Providing D = 0.30 cm 1 = 0.4 3 K , g =1.90, B = 0.67 K T 1 , and 0 H z = 0.05 0 T, J can be calculated to give a value of 0.0055 K ( J = 0.011 K in physics convention) . Thus, the exchange interaction be tween the two Mn 3 units in complex 5 3 · xMe 2 CO · yMeCN is very weak and comparable to the couplings via dipolar (through space) interaction . As a consequence, the

PAGE 209

209 hysteresis loops displayed by complex 5 3 · xMe 2 CO · yMeCN a r e very similar to the ones of 3 · 3CH 2 Cl 2 with only a small perturbation. 171 5.4 Summary and Conclusions In summ ary, three new supramolecular aggregates of Mn 3 SMMs were prepared by carboxylate substitution. The formation of these dimeric and tetrameric supramolecules was predicted from the analysis of the directional bonding approach. Magnetic studies confirmed tha t all three complexes exhibit SMM behavior , w hich results from the SMM behavior of each Mn 3 units. The inter Mn 3 exchange biased coupling in complex 5 3 is very weak and only causes a small shift of QTM steps. In conclusion, this work establish es an addi tional method to obtain supramolecular aggregates of Mn 3 SMMs, complementing the previous method using dioxime s . W e now have the option of site selectively aggregating at the carboxylate or ox i m ate positions or indeed sequentially at both. M any dicarboxyli c acids and dioximes with a variety of lengths, shape, and bulk that are available commercially or easily synthesized promise access to various oligomers of SMMs with di ering topologies and strengths of inter Mn 3 interactions.

PAGE 210

210 Table 5 1. Crystallographic data for complexes 5 1 · xCHCl 3, 5 2 · xMeNO 2 · yMeCN · zEt 2 O, and 5 3 · xMe 2 CO · yMeCN . Parameters 5 1 · xCHCl 3 5 2 · xMeNO 2 · yMeCN · zEt 2 O 5 3 . xMe 2 CO.yMeCN Formula a C 120 H 96 Mn 12 N 24 O 56 Cl 4 C 57 H 60 Mn 6 N 12 O 28 Cl 2 C 114 H 98 Mn 6 N 12 O 24 FW, g/mol 3571.30 1761.71 2349.75 Crystal syst em Monoclinic Triclinic Triclinic Space group C2/ c a, Å 33.3478(8) 20.4042(9) 13.5272(5) b, Å 19.2129(5) 20.9896(9) 15.9384(5) c, Å 70.2465(18) 25.7470(12) 17.8515(6) ,° 90 103.6311(11) 76.015(1) ,° 103.5210(10) 91.2279(12) 80.849(1) ,° 90 116.8120(11) 80.895(1) V, Å 3 43760.1(19) 9459 .9(7) 3658.3(2) Z 76 2 1 T, K 100(2) 100(2) 100(2) , Å b 0.71073 0.71073 0.71073 cal , mg/m 3 1.703 1.479 1.427 , mm 1 11.954 0.896 4.777 R 1 c,d 0.1817 0.0579 0.0337 wR 2 e 0.4301 0.1665 0.0955 a Ex cluding solva t e molecules . b Graphite monochromator. c I>2 (I). d R 1 o | |F c o |. e wR 2 o 2 F c 2 ) 2 o 2 ) 2 ]] 1/2 2 (F o 2 )+(m × p) 2 + n × p], p = [max(F o 2 ,0)+ 2 × F c 2 ]/3, m & n are constants.

PAGE 211

211 Table 5 2. Selected interatomic distances (Ã…) and angles ( 0 ) for complex 5 1 . Com p le x 5 1 Bond distances Bond angles Mn1 Mn2 3.198(4) Mn5 O2 1.907(12) Mn1 O1 Mn2 119.2(6) Mn1 Mn3 3.214(5) Mn5 O18 2.176(15) Mn1 O1 Mn3 118.6(6) Mn2 Mn3 3.200(5) Mn5 O19 2.266(12) Mn2 O1 Mn3 115.6(5) Mn4 Mn5 3.195(5) Mn6 O2 1.893(11) Mn4 O2 Mn6 118.0(6) Mn4 Mn6 3.212(5) Mn6 O20 2.140(16) Mn4 O2 Mn5 116.4(6) Mn5 Mn6 3.223(5) Mn6 O22 2.211(13) Mn6 O2 Mn5 116.1(6) Mn7 Mn8 3.211(5) Mn7 O3 1.916(14) Mn8 O3 Mn9 120.6(8) Mn7 Mn9 3.193(5) Mn7 O25 2.106(14) Mn8 O3 Mn7 117.4(7) Mn8 Mn9 3.210(5) Mn7 O23 2.283(14) Mn9 O3 Mn7 115.8(7) Mn10 Mn11 3.211(5) Mn8 O3 1.842(14) Mn10 O4 Mn11 119.1(6) Mn10 Mn12 3.192(5) Mn8 O28 2.189(14) Mn10 O4 Mn12 117.8(6) Mn11 Mn12 3.215(5) Mn8 O26 2.247(12) Mn11 O4 Mn12 117.9(5) Mn1 O1 1.831(11) Mn9 O 3 1.853(14) Mn1 O5 2.177(13) Mn9 O31 2.174(13) Mn1 O6 2.232(12) Mn9 O29 2.259(12) Mn2 O1 1.876(11) Mn10 O4 1.850(11) Mn2 O11 2.182(14) Mn10 O32 2.179(14) Mn2 O12 2.196(13) Mn10 O34 2.265(11) Mn3 O1 1.907(11) Mn11 O4 1.875(10) Mn3 O8 2. 214(12) Mn11 O38 2.192(13) Mn3 O10 2.221(13) Mn11 O40 2.274(13) Mn4 O2 1.854(12) Mn12 O4 1.878(11) Mn4 O14 2.201(14) Mn12 O35 2.228(13) Mn4 O16 2.202(13) Mn12 O37 2.236(14)

PAGE 212

212 Table 5 3. Selected interatomic distances (Ã…) and angles ( o ) for complexes 5 2 and 5 3 . Complex 5 2 Complex 5 3 Bond distances Mn1 Mn2 3.1951(5) Mn1 Mn2 3.2011(4) Mn1 Mn3 3.2073(6) Mn1 Mn3 3.2132(4) Mn2 Mn3 3.2134(5) Mn2 Mn3 3.2168(4) Mn4 Mn5 3.1978(10) Mn4 Mn6 3.1978(11) Mn5 Mn6 3.2175(11) Mn1 O1 1.8742(17) Mn1 O1 1.8719(12) Mn1 O5 2.1951(14) Mn1 O5 1.9304(13) Mn1 O9 2.2108(14) Mn1 N5 2.0051(16) Mn2 O1 1.8724(16) Mn1 N6 2.0181(16) Mn2 O3 2.1671(16) Mn1 O2 2.1596(13) Mn2 O13 2.2356(16) Mn1 O10 2.1815(13) Mn3 O1 1.8646( 15) Mn2 O1 1.8790(12) Mn3 O4 2.1619(19) Mn2 O7 1.9172(13) Mn3 O12 2.2052(19) Mn2 N1 2.0014(15) Mn4 O2 1.876(3) Mn2 N2 2.0363(16) Mn4 O7 2.160(3) Mn2 O3 2.1641(13) Mn4 O17 2.228(3) Mn2 O6 2.2388(13) Mn5 O2 1.860(3) Mn3 O1 1.8785(12) Mn5 O8 2.193(3) Mn3 O9 1.9393(13) Mn5 O8 2.251(3) Mn3 N3 2.0057(15) Mn6 O2 1.891(3) Mn3 N4 2.0300(16) Mn6 O6 2.181(3) Mn3 O4 2.2050(13) Mn6 O20 2.281(3) Mn3 O8 2.2157(13) Bond angles Mn1 O1 Mn2 117.03(8) Mn1 O1 Mn2 117.17(6) Mn1 O1 Mn3 118.15(8) M n1 O1 Mn3 117.91(6) Mn2 O1 Mn3 118.61(9) Mn2 O1 Mn3 117.76(6) Mn4 O2 Mn5 117.70(18) Mn4 O2 Mn6 116.17(15) Mn5 O2 Mn6 118.14(17)

PAGE 213

213 Table 5 4. Displacement of 3 oxide atoms from Mn 3 planes (Ã…) and Mn N O Mn torsion angles ( o ) of complex es 5 1 , 5 2 and 5 3 . Triangle d (Ã…) Torsion Angles ( o ) Complex 5 1 Mn1Mn2Mn3 0.2 8 Mn1 N 1 O 12 Mn2 24 . 579 ( 3 ) Mn2 N 5 O 10 Mn3 20 . 158 ( 3 ) Mn3 N 3 O 5 Mn1 17 .5 55 ( 3 ) Mn4Mn5Mn6 0.34 Mn4 N9 O18 Mn5 23.301(3) Mn5 N7 O20 Mn6 17.920(3) Mn6 N 11 O16 Mn4 18.014(3) Mn7Mn8Mn9 0.27 Mn7 N13 O31 Mn9 19.686(3) Mn9 N17 O28 Mn8 20.357(3) Mn8 N15 O25 Mn7 18.106(3) Mn10Mn11Mn12 0.25 Mn10 N21 O35 Mn12 21.144(3) Mn12 N23 O38 Mn11 18.545 (3) Mn11 N19 O32 Mn10 16.280(3) Complex 5 2 Mn 1Mn2Mn3 0.27 Mn1 N2 O3 Mn3 6.230(7) Mn2 N4 O4 Mn3 8.192(7) Mn3 N6 O5 Mn1 22.652(7) Mn4Mn5Mn6 0.31 Mn4 N8 O6 Mn6 20.397(7) Mn6 N12 O8 Mn5 3.076(7) Mn5 N10 O7 Mn4 1 6 .557(7) Complex 5 3 Mn1Mn2Mn3 0.29 Mn1 N5 O4 Mn3 8 .688(158) Mn3 N3 O3 Mn2 15.176 ( 154 ) Mn2 N1 O2 Mn1 13.994 ( 15 5)

PAGE 214

214 Table 5 5. Bond valence sum calculation for Mn a and selected O b atoms of complex 5 1 . Mn II Mn III Mn IV Assignment BVS Assignment Mn1 3.34 3.12 3.17 Mn III O1 2.21 O 2 Mn2 3.43 3.21 3.25 Mn III O2 2.12 O 2 Mn3 3.23 3.01 3.07 Mn III O3 2.22 O 2 Mn4 3.31 3.10 3.14 Mn III O4 2.22 O 2 Mn5 3.12 2.93 2.97 Mn III Mn6 3.28 3.07 3.11 Mn III Mn7 3.11 2.91 2.95 Mn III Mn8 3.41 3.20 3.24 Mn III Mn9 3.40 3.18 3.22 Mn III Mn10 3.32 3.10 3.15 Mn III Mn11 3.38 3.17 3.20 Mn III Mn12 3.18 2.97 3.01 Mn III a The bold value is the one closest to the charge for which it was calculated. The oxidation state of a particular atom can be taken as the nearest whole number to the bold val ue. b A BVS in the ~ 1.8 2. 2 , ~1.0 1. 4 , and ~0.2 0.4 ranges for an O atom is indicative of non , single , and double protonation, respectively, but can be altered somewhat by hydrogen bonding. Table 5 6. Bond valence sum calculation for Mn a and selected O b atoms of complex 5 2 . Mn II Mn III Mn IV Assignment BVS Assignment Mn1 3.25 3.04 3.08 Mn III O1 2.21 O 2 Mn2 3.29 3.07 3.12 Mn III O2 2.18 O 2 Mn3 3.28 3.05 3.12 Mn III Mn4 3.31 3.10 3.14 Mn III Mn5 3.30 3.09 3.13 Mn III Mn6 3.15 2.94 2.99 Mn III a See footnote a of table 5 5. b See footnote b of table 5 5. Table 5 7. Bond valence sum calculation for Mn a and selected O b atoms of complex 5 3 . Mn II Mn III Mn IV Assignment BVS Assignment Mn1 3.35 3.13 3.17 Mn III O1 2.17 O 2 Mn2 3.28 3.07 3.11 Mn III Mn3 3.17 2.96 3.02 Mn III a See footnote a of table 5 5. b See footnote b of table 5 5.

PAGE 215

215 Table 5 8. Parameters of the fits to Van Vleck equation. Complex J (cm 1 ) J' (cm 1 ) g 5 1 5.63 ± 0. 12 19.45 ± 0.57 1. 87 ± 0.0 1 5 2 5.14 ± 0.09 20.78 ± 0.36 1.9 1 ± 0.01 5 3 14.36 ± 0.71 4.61 ± 0.97 1.9 1 ± 0.01 Table 5 9 . Reduced magnetization fit parameters. Complex S g D (cm 1 ) 5 1 6 1.8 6 ± 0.01 0.3 3 ± 0.01 5 2 6 1.90 ± 0.01 0.35 ± 0.01 5 3 6 1.90 ± 0.01 0.30 ± 0.01

PAGE 216

216 Figure 5 1. Sev eral di , tri, and tetracaboxylic acids employed in metal organic framework chemistry.

PAGE 217

217 Figure 5 2 . truxillic acid (right) .

PAGE 218

218 Figure 5 3. (Top) Structure of the ca tion of complex 5 1 , (middle) a stereopair, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits and showing the Mn 3 planes and Jahn Teller axes (green bonds). H atoms have been omitted for clarity. Color code: Mn III green; O r ed; N blue; C grey.

PAGE 219

219 Figure 5 4. (Top) Structure of the cation of complex 5 2 in side view and top view, (middle) a stereopair, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits and showing the Mn 3 planes and Jahn Tell er axes (green bonds). H atoms have been omitted for clarity. Color code: Mn III green; O red; N blue; C grey.

PAGE 220

220 Figure 5 5 . (Top) Structure of complex 5 3 , (middle) a stereopair, (bottom) the core emphasizing the corresponding connectivity between Mn 3 subunits and showing the Mn 3 planes and Jahn Teller axes (green bonds). H atoms have been omitted for clarity. Color code: Mn III green; O red; N blue; C grey.

PAGE 221

221 Figure 5 6. Packing of cations of 5 2 in top view showing the different orientations between two sets of Mn 6 cations.

PAGE 222

222 Figure 5 7 . Packing of molecules of 5 3 showing parallel Mn 3 planes .

PAGE 223

223 Figure 5 8 . M T vs. T for complex es 5 1 , 5 2 , and 5 3 . The solid line s are the fit s to the data. See table 5 8 for the fit parameters.

PAGE 224

224 Figure 5 9. Two di mensional contour plot of the fitting errors vs. J and J' for complex 5 1 . Figure 5 10 . Two dimensional contour plot of the fitting errors vs. J and J' for complex 5 2 . * *

PAGE 225

225 Figure 5 11. Two dimensional contour plot of the fitting errors vs. J and J' for co mplex 5 3 . Figure 5 1 2 . Energy ladder plot for complex 5 1 . Ground state is S T = 6. The first excited state S = 5 lies 67 .5 cm 1 above the ground state. *

PAGE 226

226 Figure 5 13. Energy ladder plot for complex 5 2 . Ground state is S T = 6. The first excited state S = 5 li es 79.5 cm 1 above the ground state. Figure 5 14. Energy ladder plot for complex 5 3 . Ground state is S T = 6. The first excited state S = 5 lies 79.5 cm 1 above the ground state.

PAGE 227

227 Figure 5 15. M / N B (per Mn 3 ) vs. H / T for complex 5 1 . The solid lines are th e fits to the data. See the text for the fit parameters. Figure 5 16. M / N B (per Mn 3 ) vs. H / T for complex 5 2 . The solid lines are the fits to the data. See the text for the fit parameters.

PAGE 228

228 Figure 5 1 7 . M / N B (per Mn 3 ) vs. H / T for complex 5 3 . The soli d lines are the fits to the data. See the text for the fit parameters. Figure 5 18. Two dimensional contour plot of the error surface for the D vs. g fit for complex 5 1 . *

PAGE 229

229 Figure 5 19. Two dimensional contour plot of the error surface for the D vs. g f it for complex 5 2 . Figure 5 20 . Two dimensional contour plot of the error surface for the D vs. g fit for complex 5 3 . * *

PAGE 230

230 Figure 5 21. Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature f or a micro crystalline sample of complex 5 1 in a 3.5G field oscillating at the indicated frequencies.

PAGE 231

231 Figure 5 22. Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature for a micro crystallin e sample of complex 5 2 in a 3.5G field oscillating at the indicated frequencies.

PAGE 232

232 Figure 5 2 3 . Plot of in phase ( M , as M T ) (top) and out of phase ( ) (bottom) ac susceptibility signals versus temperature for a micro crystalline sample of complex 5 3 in a 3.5G field oscillating at the indicated frequencies.

PAGE 233

233 Figure 5 24. Magnetization vs. dc field hysteresis loops for a single crystal of 5 3 ·xMe 2 CO · yMeCN at the indicated temperatures (top) and field scan rates at 0.03 K (bottom). M is normalized to its saturation value, M S .

PAGE 234

234 APPENDIX A LIST OF COMPOUNDS [Mn 3 O(O 2 CMe) 6 (py) 3 ]ClO 4 ( 1 , ref . 172) [Mn 3 O(O 2 CEt) 6 (py) 3 ]ClO 4 ( 2 , ref . 197) [Mn 3 O(O 2 CR) 6 (mpko) 3 ]ClO 4 (R= Me ( 3 ), Et ( 4 ), ref . 171) [Mn 6 O 2 (O 2 CMe) 8 (MeOH) 2 (pdpd) 2 ] ( 2 1 ) [Mn 6 O 2 (O 2 CMe) 8 (py) 2 (pdpd) 2 ]( ClO 4 ) 2 ( 2 2 ) [Mn 12 O 4 (O 2 CMe) 12 (pdpd) 6 (CH 2 Cl 2 ) 2 ](ClO 4 ) 4 ( 2 3 ) [Mn 12 O 4 (O 2 C t Bu) 12 (pdpd) 6 (CH 2 Cl 2 ) 2 ](ClO 4 ) 4 ( 2 4 ) [Mn 12 O 4 (O 2 CMe) 12 (ded) 6 (pyH)](ClO 4 ) 2 ( 3 1 ) [Mn 12 O 4 (O 2 CEt) 10 (H 2 O) 2 Cl 2 (ded) 6 (py)](ClO 4 ) 2 ( 3 2 ) [Mn 12 O 4 (O 2 CMe) 12 (ded) 6 (CH 2 Cl 2 )](I 3 ) 3.5 I 0.5 ( 3 3 ) [M n 12 O 4 (O 2 CEt) 12 (ded) 6 (EtOH)](I 3 ) 3.5 I 0.5 ( 3 4 ) [Mn 6 O 2 (O 2 CMe) 6 (dpd) 3 ] (I 3 ) 2 ( 4 1 ) [Mn 12 O 4 (tma) 4 (mpko) 12 ](ClO 4 ) 4 ( 5 1 ) [Mn 6 O 2 (dma) 3 (mpko) 6 ](ClO 4 ) 2 ( 5 2 ) [Mn 6 O 2 (txa) 4 (mpko) 6 ] ( 5 3 )

PAGE 235

235 APPENDIX B VAN VLECK EQUATIONS p = paramagnetic impurity c = N B 2 / 3k N = Avo g k = Boltzmann constant T = Temperature TIP = Temperature independent paramagnetism B 1. The Van Vleck equation for a dimer of isosceles triangles [ Mn III 2 Mn II ] 2 . [Mn 6 O 2 (O 2 CMe) 8 (MeOH) 2 (pdpd) 2 ] ( 2 1 ) J inter = 0 cm 1 m = J/k/T n = J'/k/T numerator = + 52.5000 × exp(8.7500 × m + 0.0000 × n) + 15.0000 × exp(1.7500 × m + 2.0000 × n) + 52.5000 × exp(6.7500 × m + 2.0000 × n) + 126.0000 × exp(13. 7500 × m + 2.0000 × n) + 1.5000 × exp( 5.2500 × m + 6.0000 × n) + 15.0000 × exp( 2.2500 × m + 6.0000 × n) + 52.5000 × exp(2.7500 × m + 6.0000 × n) S = 5/2 S = 2 S = 2 J J J' J inter S=5/2 S = 2 S = 2 J J J'

PAGE 236

236 + 126.0000 × exp(9.7500 × m + 6.0000 × n) + 247.5000 × exp(18.7500 × m + 6.0000 × n) + 1.5000 × exp( 11 .2500 × m + 12.0000 × n) + 15.0000 × exp( 8.2500 × m + 12.0000 × n) + 52.5000 × exp( 3.2500 × m + 12.0000 × n) + 126.0000 × exp(3.7500 × m + 12.0000 × n) + 247.5000 × exp(12.7500 × m + 12.0000 × n) + 429.0000 × exp(23.7500 × m + 12.0000 × n) + 15.000 0 × exp( 16.2500 × m + 20.0000 × n) + 52.5000 × exp( 11.2500 × m + 20.0000 × n) + 126.0000 × exp( 4.2500 × m + 20.0000 × n) + 247.5000 × exp(4.7500 × m + 20.0000 × n) + 429.0000 × exp(15.7500 × m + 20.0000 × n) + 682.5000 × exp(28.7500 × m + 20.0000 × n) denominator = + 6.0000 × exp(8.7500 × m + 0.0000 × n) + 4.0000 × exp(1.7500 × m + 2.0000 × n) + 6.0000 × exp(6.7500 × m + 2.0000 × n) + 8.0000 × exp(13.7500 × m + 2.0000 × n) + 2.0000 × exp( 5.2500 × m + 6.0000 × n) + 4.0000 × exp( 2.2500 × m + 6 .0000 × n) + 6.0000 × exp(2.7500 × m + 6.0000 × n) + 8.0000 × exp(9.7500 × m + 6.0000 × n) + 10.0000 × exp(18.7500 × m + 6.0000 × n) + 2.0000 × exp( 11.2500 × m + 12.0000 × n) + 4.0000 × exp( 8.2500 × m + 12.0000 × n) + 6.0000 × exp( 3.2500 × m + 12. 0000 × n) + 8.0000 × exp(3.7500 × m + 12.0000 × n) + 10.0000 × exp(12.7500 × m + 12.0000 × n) + 12.0000 × exp(23.7500 × m + 12.0000 × n) + 4.0000 × exp( 16.2500 × m + 20.0000 × n) + 6.0000 × exp( 11.2500 × m + 20.0000 × n) + 8.0000 × exp( 4.2500 × m + 20.0000 × n) + 10.0000 × exp(4.7500 × m + 20.0000 × n) + 12.0000 × exp(15.7500 × m + 20.0000 × n) + 14.0000 × exp(28.7500 × m + 20.0000 × n)

PAGE 237

237 B 2. The Van Vleck equation for a dimer of isosceles triangles [Mn III 3 ] 2 . [Mn 6 O 2 (O 2 CMe) 8 (py) 2 (pdpd) 2 ](ClO 4 ) 2 ( 2 2 ) [Mn 6 O 2 (O 2 CMe) 6 (dpd) 3 ](I 3 ) 2 ( 4 1 ) [Mn 6 O 2 (dma) 3 (mpko) 6 ](ClO 4 ) 2 ( 5 2 ) [Mn 6 O 2 (txa) 4 (mpko) 6 ] ( 5 3 ) J inter = 0 cm 1 m = J/k/T n = J'/k/T numerator = + 30.0000 × exp( 6.0000 × m + 0.0000 × n) + 6.0000 × exp( 0.0000 × m + 2.0000 × n) + 30.0000 × exp( 4.0000 × m + 2.0000 × n) + 84.0000 × exp( 10.0000 × m + 2.0000 × n) + 0.0000 × exp( 6.0000 × m + 6.0000 × n) + 6.0000 × exp( 4.0000 × m + 6.0000 × n) + 30.000 0 × exp( 0.0000 × m + 6.0000 × n) + 84.0000 × exp( 6.0000 × m + 6.0000 × n) + 180.0000 × exp( 14.0000 × m + 6.0000 × n) + 6.0000 × exp( 10.0000 × m + 12.0000 × n) + 30.0000 × exp( 6.0000 × m + 12.0000 × n) + 84.0000 × exp( 0.0000 × m + 12.0000 × n) + 180.0000 × exp( 8.0000 × m + 12.0000 × n) + 330.0000 × exp( 18.0000 × m + 12.0000 × n) + 30.0000 × exp( 14.0000 × m + 20.0000 × n) + 84.0000 × exp( 8.0000 × m + 20.0000 × n) + 180.0000 × exp( 0.0000 × m + 20.0000 × n) + 330.0000 × exp( 10.0000 × m + 20.0000 × n) + 546.0000 × exp( 22.0000 × m + 20.0000 × n) S = 2 S = 2 S = 2 J J J' J inter S = 2 S = 2 S = 2 J J J'

PAGE 238

238 denominator = + 5.0000 × exp( 6.0000 × m + 0.0000 × n) + 3.0000 × exp( 0.0000 × m + 2.0000 × n) + 5.0000 × exp( 4.0000 × m + 2.0000 × n) + 7.0000 × exp( 10.0000 × m + 2.0000 × n) + 1.0000 × exp( 6.0000 × m + 6.0000 × n) + 3.0000 × exp( 4.0000 × m + 6.0000 × n) + 5.0000 × exp( 0.0000 × m + 6.0000 × n) + 7.0000 × exp( 6.0000 × m + 6.0000 × n) + 9.0000 × exp( 14.0000 × m + 6.0000 × n) + 3.0000 × exp( 10.0000 × m + 12.0000 × n) + 5.00 00 × exp( 6.0000 × m + 12.0000 × n) + 7.0000 × exp( 0.0000 × m + 12.0000 × n) + 9.0000 × exp( 8.0000 × m + 12.0000 × n) + 11.0000 × exp( 18.0000 × m + 12.0000 × n) + 5.0000 × exp( 14.0000 × m + 20.0000 × n) + 7.0000 × exp( 8.0000 × m + 20.0000 × n) + 9.0000 × exp( 0.0000 × m + 20.0000 × n) + 11.0000 × exp( 10.0000 × m + 20.0000 × n) + 13.0000 × exp( 22.0000 × m + 20.0000 × n) B 3. The Van Vleck equation for a dimer of equilateral triangles [Mn III 3 ] 2 . [Mn 6 O 2 (O 2 CMe) 6 (dpd) 3 ](I 3 ) 2 ( 4 1 ) J inter = 0 cm 1 m = J/k/T numerator = + 150.0000 × exp(6.0000 × m) + 18.0000 × exp(2.0000 × m) + 336.0000 × exp(12.0000 × m) + 0.0000 × exp(0.0000 × m) + 540.0000 × exp(20.0000 × m) + 660.0000 × exp(30.0000 × m) + 546.0000 × exp(42.0000 × m) S = 2 S = 2 S = 2 J J J J inter S = 2 S = 2 S = 2 J J J

PAGE 239

239 denominator = + 25.0000 × exp(6.0000 × m) + 9.0000 × exp(2.0000 × m) + 28.0000 × exp(12.0000 × m) + 1.0000 × exp(0.0000 × m) + 27.0000 × exp(20.0000 × m) + 22.0000 × exp(30.0000 × m) + 13.0000 × exp(42.0000 × m) B 4 . The Van Vleck equation for a rectangular tetramer of isosceles triangles [Mn III 3 ] 4 [Mn 12 O 4 (O 2 CMe) 12 (pdpd) 6 (CH 2 Cl 2 ) 2 ](ClO 4 ) 4 ( 2 3 ) [Mn 12 O 4 (O 2 C t Bu) 12 (pdpd) 6 (CH 2 Cl 2 ) 2 ](ClO 4 ) 4 ( 2 4 ) J inter = 0 cm 1 m = J/k/T n = J'/k/T numerator = + 30.0000 × exp( 6.0000 × m + 0.0000 × n) + 6.0000 × exp( 0.0000 × m + 2.0000 × n) + 30.0000 × exp( 4.0000 × m + 2.0000 × n) + 84.0000 × exp( 10.0000 × m + 2.0000 × n) J inter J inter J inter J inter S = 2 S = 2 S = 2 J J J' S = 2 S = 2 S = 2 J J J' S = 2 S = 2 S = 2 J J J' S = 2 S = 2 S = 2 J J J'

PAGE 240

240 + 0.0000 × exp( 6.0000 × m + 6.0000 × n) + 6.0000 × exp( 4.0000 × m + 6.0000 × n) + 30.0000 × exp( 0.0000 × m + 6.0000 × n) + 84.0000 × exp( 6.0000 × m + 6.0000 × n) + 180.0000 × exp( 14.0000 × m + 6.0000 × n) + 6.0000 × exp( 10.00 00 × m + 12.0000 × n) + 30.0000 × exp( 6.0000 × m + 12.0000 × n) + 84.0000 × exp( 0.0000 × m + 12.0000 × n) + 180.0000 × exp( 8.0000 × m + 12.0000 × n) + 330.0000 × exp( 18.0000 × m + 12.0000 × n) + 30.0000 × exp( 14.0000 × m + 20.0000 × n) + 84.00 00 × exp( 8.0000 × m + 20.0000 × n) + 180.0000 × exp( 0.0000 × m + 20.0000 × n) + 330.0000 × exp( 10.0000 × m + 20.0000 × n) + 546.0000 × exp( 22.0000 × m + 20.0000 × n) denominator = + 5.0000 × exp( 6.0000 × m + 0.0000 × n) + 3.0000 × exp( 0.0000 × m + 2.0000 × n) + 5.0000 × exp( 4.0000 × m + 2.0000 × n) + 7.0000 × exp( 10.0000 × m + 2.0000 × n) + 1.0000 × exp( 6.0000 × m + 6.0000 × n) + 3.0000 × exp( 4.0000 × m + 6.0000 × n) + 5.0000 × exp( 0.0000 × m + 6.0000 × n) + 7.0000 × exp( 6.0000 × m + 6.0000 × n) + 9.0000 × exp( 14.0000 × m + 6.0000 × n) + 3.0000 × exp( 10.0000 × m + 12.0000 × n) + 5.0000 × exp( 6.0000 × m + 12.0000 × n) + 7.0000 × exp( 0.0000 × m + 12.0000 × n) + 9.0000 × exp( 8.0000 × m + 12.0000 × n) + 11.0000 × exp( 18.0 000 × m + 12.0000 × n) + 5.0000 × exp( 14.0000 × m + 20.0000 × n) + 7.0000 × exp( 8.0000 × m + 20.0000 × n) + 9.0000 × exp( 0.0000 × m + 20.0000 × n) + 11.0000 × exp( 10.0000 × m + 20.0000 × n) + 13.0000 × exp( 22.0000 × m + 20.0000 × n)

PAGE 241

241 B 5. The Van Vleck equation for a tetrahedral tetramer of isosceles triangles [Mn III 3 ] 4 [Mn 12 O 4 (tma) 4 (mpko) 12 ](ClO 4 ) 4 ( 5 1 ) J inter = 0 cm 1 m = J/k/T n = J'/k/T numerator = + 30.0000 × exp( 6.0000 × m + 0.0000 × n) + 6.0000 × exp( 0.0000 × m + 2.0000 × n) + 30.0000 × exp( 4.0000 × m + 2.0000 × n) + 84.0000 × exp( 10.0000 × m + 2.0000 × n) + 0.0000 × exp( 6.0000 × m + 6.0000 × n) + 6.0000 × exp( 4.0000 × m + 6.0000 × n) + 30.0000 × exp( 0.0000 × m + 6.0000 × n) + 84.0000 × exp( 6.0000 × m + 6.0000 × n) + 180.0000 × exp( 14.0000 × m + 6.0000 × n) + 6.0000 × exp( 10.0000 × m + 12.0000 × n) J inter J inter J inter J inter S = 2 S = 2 S = 2 J J J' S = 2 S = 2 S = 2 J J J' S = 2 S = 2 S = 2 J J J' S = 2 S = 2 S = 2 J J J' J inter J inter

PAGE 242

242 + 30.0000 × exp( 6.0000 × m + 12.0000 × n) + 84.0000 × exp( 0.00 00 × m + 12.0000 × n) + 180.0000 × exp( 8.0000 × m + 12.0000 × n) + 330.0000 × exp( 18.0000 × m + 12.0000 × n) + 30.0000 × exp( 14.0000 × m + 20.0000 × n) + 84.0000 × exp( 8.0000 × m + 20.0000 × n) + 180.0000 × exp( 0.0000 × m + 20.0000 × n) + 330. 0000 × exp( 10.0000 × m + 20.0000 × n) + 546.0000 × exp( 22.0000 × m + 20.0000 × n) denominator = + 5.0000 × exp( 6.0000 × m + 0.0000 × n) + 3.0000 × exp( 0.0000 × m + 2.0000 × n) + 5.0000 × exp( 4.0000 × m + 2.0000 × n) + 7.0000 × exp( 10.0000 × m + 2.0000 × n) + 1.0000 × exp( 6.0000 × m + 6.0000 × n) + 3.0000 × exp( 4.0000 × m + 6.0000 × n) + 5.0000 × exp( 0.0000 × m + 6.0000 × n) + 7.0000 × exp( 6.0000 × m + 6.0000 × n) + 9.0000 × exp( 14.0000 × m + 6.0000 × n) + 3.0000 × exp( 10.0000 × m + 12.0000 × n) + 5.0000 × exp( 6.0000 × m + 12.0000 × n) + 7.0000 × exp( 0.0000 × m + 12.0000 × n) + 9.0000 × exp( 8.0000 × m + 12.0000 × n) + 11.0000 × exp( 18.0000 × m + 12.0000 × n) + 5.0000 × exp( 14.0000 × m + 20.0000 × n) + 7.0000 × exp( 8.0 000 × m + 20.0000 × n) + 9.0000 × exp( 0.0000 × m + 20.0000 × n) + 11.0000 × exp( 10.0000 × m + 20.0000 × n) + 13.0000 × exp( 22.0000 × m + 20.0000 × n)

PAGE 243

243 APPENDIX C BOND DISTANCES AND A NGLES Table C 1. Selected interatomic distances (Ã…) and angles ( o ) f or complex 2 1 . Parameters Parameters Mn1 Mn2 3.1082(6) Mn2 O1 Mn1 116.76(10) Mn1 Mn3 3.3940(2) Mn2 O1 Mn3 121.02(9) Mn2 Mn3 3.3730(3) Mn1 O1 Mn3 122.20(9) Mn1 O1 1.8262(18) Mn1 Mn2 Mn3 63.016(13) Mn1 O6 1.950(2) Mn2 Mn3 Mn1 54.668(12) Mn1 N 2 2.021(2) Mn3 Mn1 Mn2 62.316(13) Mn1 N1 2.052(2) O1 Mn1 Mn2 31.60(6) Mn1 O4 2.1910(19) O3 Mn1 Mn2 64.65(5) Mn1 O3 2.2134(19) O4 Mn1 Mn2 117.30(5) Mn2 O1 1.8238(18) O6 Mn1 Mn2 119.39(6) Mn2 O8 1.9464(19) N1 Mn1 Mn2 137.75(6) Mn2 N3 2.040(2) N2 Mn1 Mn 2 66.62(6) Mn2 N4 2.042(2) O1 Mn2 Mn1 31.64(6) Mn2 O10 2.163(2) O2 Mn2 Mn1 63.70(5) Mn2 O2 2.3115(19) O8 Mn2 Mn1 124.17(6) Mn3 O1 2.0500(18) O10 Mn2 Mn1 108.24(5) Mn3 O12 2.162(2) N3 Mn2 Mn1 64.96(6) Mn3 O11 2.163(2) N4 Mn2 Mn1 138.85(7) Mn3 O9 2.17 6(2) Mn3 O12 H12 112(3) Mn3 O7 2.195(2) Mn3 O5 2.199(2)

PAGE 244

244 Table C 2. Selected interatomic distances (Ã…) and angles ( o ) for complex 2 2 . Parameters Parameters Mn1 Mn2 3.1610(15) Mn4 N6 2.019(6) Mn1 Mn3 3.2913(19) Mn4 O17 2.167(6) Mn2 Mn3 3.2922(15) Mn4 O14 2.200(5) Mn4 Mn5 3.1641(21) Mn5 O12 1.894(5) Mn4 Mn6 3.2817(16) Mn5 O21 1.913(5) Mn5 Mn6 3.2864(18) Mn5 N8 2.009(6) Mn1 O1 1.889(5) Mn5 N9 2.044(6) Mn1 O6 1.919(5) Mn5 O19 2.150(5) Mn1 N2 2.015(6) Mn5 O13 2.180(5) Mn1 N1 2.031(6) Mn6 O12 1.852(4) Mn1 O3 2.154(5) Mn6 O18 1.915(6) Mn1 O4 2.185(5) Mn6 O20 1.942(6) Mn2 O1 1.890(5) Mn6 N10 2.092(6) Mn2 O8 1.924(5) Mn6 O22 2.181(5) Mn2 N4 2.001(6) Mn6 O16 2.215(5) Mn2 N3 2.023(6) Mn1 O1 Mn2 113.6(2) Mn2 O10 2.161(6 ) Mn1 O1 Mn3 123.2(2) Mn2 O2 2.221(5) Mn2 O1 Mn3 123.2(3) Mn3 O1 1.853(5) Mn4 O12 Mn5 114.1(2) Mn3 O11 1.930(5) Mn4 O12 Mn6 123.3(3) Mn3 O5 1.942(5) Mn5 O12 Mn6 122.6(3) Mn3 N5 2.091(6) Mn1 Mn2 Mn3 61.290(35) Mn3 O7 2.182(6) Mn2 Mn3 Mn1 57.392(34) M n3 O9 2.194(5) Mn3 Mn1 Mn2 61.318(38) Mn4 O12 1.877(5) Mn4 Mn5 Mn6 61.125(37) Mn4 O15 1.922(5) Mn5 Mn6 Mn4 57.599(34) Mn4 N7 2.016(6) Mn6 Mn4 Mn5 61.276(36)

PAGE 245

245 Table C 3. Selected interatomic distances (Ã…) and angles ( o ) for both cations in the asymmetri c unit of complex 2 3 . Parameters Parameters Mn1 Mn2 3.212(2) Mn3 N21 2.014(7) Mn1 Mn3 3.1842(19) Mn3 O10 2.178(5) Mn2 Mn3 3.2012(19) Mn3 O18 2.204(6) Mn4 Mn5 3.1887(18) Mn4 O38 1.862(5) Mn4 Mn6 3.1726(18) Mn4 O19 1.936(6) Mn5 Mn6 3.216(2) Mn4 N17 1.985(7) Mn7 Mn8 3.180(2) Mn4 N18 2.010(6) Mn7 Mn9 3.206(2) Mn4 O12 2.179(6) Mn8 Mn9 3.205(2) Mn4 O23 2.235(6) Mn10 Mn11 3.1850(18) Mn5 O38 1.887(5) Mn10 Mn12 3.197(2) Mn5 O24 1.919(6) Mn11 Mn12 3.218(2) Mn5 N15 2 .004(7) Mn1 O37 1.894(5) Mn5 N16 2.063(7) Mn1 O13 1.924(6) Mn5 O9 2.174(5) Mn1 N19 1.984(7) Mn5 O21 2.249(6) Mn1 N20 2.026(7) Mn6 O38 1.860(5) Mn1 O1 2.156(5) Mn6 O22 1.941(5) Mn1 O15 2.214(6) Mn6 N23 2.008(7) Mn2 O37 1.848(5) Mn6 N24 2.029(7) Mn2 O17 1.936(5) Mn6 O20 2.165(6) Mn2 N2 2.037(7) Mn6 O8 2.195(5) Mn2 N1 2.038(6) Mn7 O39 1.879(5) Mn2 O11 2.163(6) Mn7 O27 1.936(6) Mn2 O14 2.225(6) Mn7 N14 2.005(8) Mn3 O37 1.875(5) Mn7 N13 2.017(7) Mn3 O16 1.906(6) Mn7 O5 2 .171(6) Mn3 N22 2.011(7) Mn7 O25 2.198(7)

PAGE 246

246 Mn8 O39 1.865(5) Mn12 O36 2.185(6) Mn8 O26 1.942(6) Mn13 Mn15 3.1887(17) Mn8 N11 1.990(8) Mn13 Mn14 3.2046(19) Mn8 N12 2.037(7) Mn14 Mn15 3.1995(17) Mn8 O7 2.151(5) Mn16 Mn18 3.194(2) Mn8 O29 2.191(6) Mn16 Mn17 3.204(2) Mn9 O39 1.860(6) Mn17 Mn18 3.205(2) Mn9 O30 1.915(6) Mn19 Mn20 3.187(2) Mn9 N9 2.015(7) Mn19 Mn21 3.2008(19) Mn9 N10 2.041(7) Mn20 Mn21 3.191(2) Mn9 O6 2.199(6) Mn22 Mn24 3.1946(17) Mn9 O28 2. 213(6) Mn22 Mn23 3.2109(17) Mn10 O40 1.867(5) Mn23 Mn24 3.2116(19) Mn10 O35 1.899(6) Mn13 O78 1.879(5) Mn10 N7 2.011(7) Mn13 O76 1.961(5) Mn10 N8 2.038(6) Mn13 N28 2.006(6) Mn10 O3 2.197(6) Mn13 N27 2.009(6) Mn10 O31 2.259(6) Mn13 O42 2.161(6) Mn11 O40 1.838(5) Mn13 O56 2.182(6) Mn11 O32 1.928(6) Mn14 O78 1.867(5) Mn11 N3 2.006(8) Mn14 O53 1.938(5) Mn11 N4 2.025(7) Mn14 N44 1.988(7) Mn11 O4 2.162(5) Mn14 N43 2.034(7) Mn11 O33 2.258(6) Mn14 O41 2.178(5) Mn12 O40 1.8 91(5) Mn14 O75 2.225(6) Mn12 O34 1.918(5) Mn15 O78 1.864(5) Mn12 N6 1.957(8) Mn15 O55 1.925(6) Mn12 N5 2.000(7) Mn15 N31 2.023(6) Mn12 O2 2.176(7) Mn15 N32 2.025(7)

PAGE 247

247 Mn15 O43 2.181(5) Mn20 N39 2.015(8) Mn15 O54 2.194(6) Mn20 O50 2.164(5) Mn16 O77 1.869(6) Mn20 O67 2.208(6) Mn16 O57 1.921(6) Mn21 O80 1.866(5) Mn16 N25 1.993(7) Mn21 O68 1.926(5) Mn16 N26 2.018(7) Mn21 N46 2.014(7) Mn16 O45 2.170(6) Mn21 N45 2.021(7) Mn16 O59 2.229(6) Mn21 O52 2.181(5) Mn17 O77 1.888(5) Mn 21 O64 2.207(5) Mn17 O62 1.910(7) Mn22 O79 1.872(4) Mn17 N29 1.993(7) Mn22 O71 1.903(6) Mn17 N30 2.057(7) Mn22 N37 2.015(7) Mn17 O46 2.162(5) Mn22 N38 2.040(7) Mn17 O58 2.198(5) Mn22 O48 2.202(5) Mn18 O77 1.873(5) Mn22 O70 2.252(6) Mn18 O60 1.939(6) Mn23 O79 1.875(5) Mn18 N34 2.017(7) Mn23 O73 1.931(5) Mn18 N33 2.019(6) Mn23 N47 2.000(6) Mn18 O44 2.197(6) Mn23 N48 2.005(7) Mn18 O61 2.226(6) Mn23 O47 2.161(6) Mn19 O80 1.853(5) Mn23 O72 2.181(6) Mn19 O63 1.957(5) Mn24 O7 9 1.878(5) Mn19 N42 2.015(8) Mn24 O69 1.914(6) Mn19 N41 2.021(7) Mn24 N35 2.008(6) Mn19 O51 2.167(6) Mn24 N36 2.040(7) Mn19 O65 2.235(7) Mn24 O49 2.162(5) Mn20 O80 1.880(5) Mn24 O74 2.272(5) Mn20 O66 1.937(7) Mn2 O37 Mn1 118.3(2) Mn20 N40 2.013(7) Mn3 O37 Mn1 115.3(3)

PAGE 248

248 Mn2 O37 Mn3 118.6(3) Mn4 O38 Mn5 116.5(3) Mn6 O38 Mn4 116.9(3) Mn6 O38 Mn5 118.2(3) Mn8 O39 Mn7 116.3(3) Mn9 O39 Mn7 118.0(3) Mn9 O39 Mn8 118.7(3) Mn11 O40 Mn10 118.5(3) Mn10 O40 Mn12 116.6(3) Mn1 1 O40 Mn12 119.3(3) Mn14 O78 Mn13 117.6(3) Mn15 O78 Mn13 116.8(3) Mn15 O78 Mn14 118.1(3) Mn16 O77 Mn17 117.1(3) Mn16 O77 Mn18 117.3(3) Mn18 O77 Mn17 116.9(3) Mn19 O80 Mn20 117.2(3) Mn19 O80 Mn21 118.8(3) Mn21 O80 Mn20 116.8(3) Mn22 O79 Mn23 117.9(3) Mn22 O79 Mn24 116.8(3) Mn23 O79 Mn24 117.7(2)

PAGE 249

249 Table C 4. Selected interatomic distances (Ã…) and angles ( o ) for complex 2 4 . Parameters Parameters Mn1 Mn2 3.1982(6) Mn4 N8 2.018(2) Mn1 Mn3 3.1970(6) Mn4 N7 2.019(2) Mn2 Mn3 3.2024(7) Mn4 O17 2.186(2) Mn4 Mn5 3.1990(6) Mn4 O7 2.2071(19) Mn4 Mn6 3.1971(7) Mn5 O2 1.8706(19) Mn5 Mn6 3.2055(6) Mn5 O19 1.933(2) Mn1 O1 1.8716(19) Mn5 N10 2.008(2) Mn1 O9 1.919(2) Mn5 N9 2.026(2) Mn1 N2 2. 011(2) Mn5 O8 2.1741(19) Mn1 N1 2.019(2) Mn5 O16 2.184(2) Mn1 O4 2.1705(19) Mn6 O2 1.8758(19) Mn1 O11 2.204(2) Mn6 O18 1.9255(19) Mn2 O1 1.8809(19) Mn6 N12 2.015(2) Mn2 O14 1.938(2) Mn6 N11 2.032(2) Mn2 N4 2.011(2) Mn6 O6 2.1743(19) Mn2 N3 2.020(2) Mn6 O20 2.2203(19) Mn2 O5 2.1969(19) Mn3 O1 Mn1 117.56(10) Mn2 O10 2.211(2) Mn3 O1 Mn2 117.41(10) Mn3 O1 1.8668(19) Mn1 O1 Mn2 116.92(10) Mn3 O12 1.931(2) Mn6 O2 Mn4 117.08(10) Mn3 N5 2.020(3) Mn6 O2 Mn5 117.66(10) Mn3 N6 2.021(2 ) Mn4 O2 Mn5 117.46(10) Mn3 O13 2.177(2) Mn3 O3 2.1919(19) Mn4 O2 1.872(2) Mn4 O15 1.927(2)

PAGE 250

250 Table C 5. Selected interatomic distances (Ã…) and angles ( o ) for both cations in the asymmetric unit of complex 3 1 . Parameters Parameters M n1 Mn2 3.1728(15) Mn3 O29 2.188(5) Mn1 Mn3 3.3243(15) Mn3 N4 2.231(5) Mn2 Mn3 3.3069(15) Mn3 N3 2.247(5) Mn4 Mn5 3.2154(17) Mn3 O4 2.257(5) Mn4 Mn6 3.2665(16) Mn4 O40 1.829(5) Mn5 Mn6 3.2921(17) Mn4 O15 1.947(5) Mn7 Mn9 3.2114(16) M n4 N6 2.046(6) Mn7 Mn8 3.2298(15) Mn4 N5 2.048(6) Mn8 Mn9 3.2307(17) Mn4 O12 2.181(5) Mn10 Mn11 3.2866(15) Mn4 O18 2.207(6) Mn10 Mn12 3.2748(16) Mn5 O40 1.905(5) Mn11 Mn12 3.2337(14) Mn5 O17 1.998(6) Mn1 O38 1.832(4) Mn5 N9 2.084(6) Mn1 O30 1.963(5) Mn5 N10 2.121(5) Mn1 N16 2.063(5) Mn5 O13 2.197(5) Mn1 N15 2.069(5) Mn5 O3 2.234(4) Mn1 O2 2.207(4) Mn6 O40 1.979(4) Mn1 O26 2.237(4) Mn6 O14 2.029(7) Mn2 O38 1.823(4) Mn6 N24 2.144(8) Mn2 O25 1.953(4) Mn6 N23 2.162(6) Mn2 N8 2.040(6) M n6 O16 2.203(6) Mn2 N7 2.055(5) Mn6 O5 2.240(5) Mn2 O28 2.209(4) Mn7 O37 1.875(4) Mn2 O8 2.224(4) Mn7 O34 1.928(5) Mn3 O38 2.050(4) Mn7 N20 2.016(6) Mn3 O27 2.111(5) Mn7 N19 2.035(5) Mn7 O32 2.186(6) Mn12 N11 2.049(5)

PAGE 251

251 Mn7 O11 2.212(5) Mn12 O9 2.196( 4) Mn8 O37 1.900(4) Mn12 O19 2.218(4) Mn8 O31 1.937(5) Mn13 Mn15 3.2177(18) Mn8 N2 2.037(5) Mn13 Mn14 3.2391(17) Mn8 N1 2.041(6) Mn14 Mn15 3.3278(17) Mn8 O36 2.223(5) Mn16 Mn17 3.3052(18) Mn8 O10 2.249(4) Mn17 Mn18 3.1871(16) Mn9 O37 1.868(5) Mn16 Mn18 3.2956(18) Mn9 O35 1.910(6) Mn19 Mn21 3.2194(14) Mn9 N22 1.991(6) Mn19 Mn20 3.2271(15) Mn9 N21 2.029(7) Mn20 Mn21 3.2320(16) Mn9 O33 2.179(5) Mn22 Mn23 3.3120(17) Mn9 O1 2.183(4) Mn22 Mn24 3.1800(16) Mn10 O39 1.953( 4) Mn23 Mn24 3.3075(15) Mn10 O23 2.014(5) Mn13 O78 1.833(5) Mn10 N17 2.111(6) Mn13 O68 1.938(6) Mn10 N18 2.172(6) Mn13 N37 2.017(7) Mn10 O21 2.235(4) Mn13 N38 2.057(6) Mn10 O7 2.245(4) Mn13 O71 2.202(5) Mn11 O39 1.923(4) Mn13 O46 2.205(4) Mn 11 O20 1.994(5) Mn14 O78 1.959(5) Mn11 N13 2.098(6) Mn14 O72 2.032(6) Mn11 N14 2.133(5) Mn14 N45 2.142(7) Mn11 O24 2.194(5) Mn14 N46 2.161(6) Mn11 O6 2.241(4) Mn14 O70 2.177(6) Mn12 O39 1.838(4) Mn14 O47 2.245(5) Mn12 O22 1.960(4) Mn15 O78 1. 927(4) Mn12 N12 2.039(6) Mn15 O69 1.961(7) Mn15 N35 2.103(6) Mn20 O56 1.934(4)

PAGE 252

252 Mn15 N36 2.116(7) Mn20 N47 2.036(5) Mn15 O51 2.192(5) Mn20 N48 2.054(6) Mn15 O67 2.209(6) Mn20 O44 2.205(5) Mn16 O79 1.997(5) Mn20 O73 2.226(5) Mn16 O62 2.028 (6) Mn21 O77 1.883(4) Mn16 N43 2.168(6) Mn21 O75 1.943(5) Mn16 O66 2.205(6) Mn21 N32 2.047(6) Mn16 N44 2.209(7) Mn21 N31 2.067(5) Mn16 O41 2.249(4) Mn21 O49 2.218(4) Mn17 O79 1.842(4) Mn21 O55 2.230(5) Mn17 O65 1.978(6) Mn22 O80 1.838(4) Mn17 N40 2.040(6) Mn22 O54 1.959(4) Mn17 N39 2.065(6) Mn22 N28 2.049(6) Mn17 O50 2.169(5) Mn22 N27 2.052(5) Mn17 O64 2.186(5) Mn22 O43 2.206(5) Mn18 O79 1.879(4) Mn22 O59 2.244(5) Mn18 O63 1.996(5) Mn23 O80 2.025(4) Mn18 N25 2.054(6) M n23 O57 2.064(6) Mn18 N26 2.094(5) Mn23 N29 2.198(6) Mn18 O61 2.180(6) Mn23 O53 2.228(5) Mn18 O48 2.204(5) Mn23 N30 2.264(6) Mn19 O77 1.890(4) Mn23 O45 2.271(5) Mn19 O74 1.914(5) Mn24 O80 1.848(4) Mn19 N42 2.041(6) Mn24 O60 1.952(5) Mn1 9 N41 2.045(5) Mn24 N33 2.053(6) Mn19 O76 2.172(5) Mn24 N34 2.056(6) Mn19 O52 2.195(5) Mn24 O58 2.181(5) Mn20 O77 1.875(4) Mn24 O42 2.232(4) Mn1 O38 Mn2 120.5(2)

PAGE 253

253 Mn1 O38 Mn3 117.7(2) Mn2 O38 Mn3 117.2(2) Mn4 O40 Mn5 118.9(2) Mn4 O40 Mn6 118.1(2) Mn5 O40 Mn6 115.9(2) Mn7 O37 Mn8 117.7(2) Mn7 O37 Mn9 118.2(2) Mn8 O37 Mn9 118.1(2) Mn10 O39 Mn11 116.0(2) Mn10 O39 Mn12 119.5(2) Mn11 O39 Mn12 118.6(2) Mn13 O78 Mn15 117.6(2) Mn13 O78 Mn14 117.3(2) Mn15 O78 Mn14 117.8(2) Mn17 O79 Mn18 117.8(3) Mn17 O79 Mn16 118.8(2) Mn18 O79 Mn16 116.4(2) Mn20 O77 Mn21 118.6(2) Mn20 O77 Mn19 118.0(2) Mn21 O77 Mn19 117.2(2) Mn22 O80 Mn24 119.3(2) Mn22 O80 Mn23 118.0(2) Mn24 O80 Mn23 117.2(2)

PAGE 254

254 Table C 6. Selected interatomic distances (Ã…) and angles ( o ) for complex 3 2 . Parameters Parameters Mn1 Mn2 3.2358(9) Mn4 N7 2.243(4) Mn1 Mn3 3.1940(9) Mn4 N8 2.252(3) Mn2 Mn3 3.2421(8) Mn4 Cl1 2.4333(12) Mn4 Mn5 3.2969(9) Mn5 O11 1 .836(3) Mn4 Mn6 3.4381(10) Mn5 O16 1.936(3) Mn5 Mn6 3.1235(9) Mn5 N9 2.040(3) Mn1 O1 1.857(3) Mn5 N10 2.048(3) Mn1 O5 1.942(3) Mn5 O12 2.181(3) Mn1 N1 2.022(4) Mn5 O17 2.239(3) Mn1 N2 2.028(3) Mn6 O11 1.827(3) Mn1 O3 2.184(3) Mn6 O18 1. 941(3) Mn1 O7 2.200(3) Mn6 N12 2.038(4) Mn2 O9 1.911(4) Mn6 N11 2.041(4) Mn2 O1 1.915(3) Mn6 O13 2.200(3) Mn2 N4 2.051(4) Mn6 O19 2.221(3) Mn2 N3 2.062(3) Mn1 O1 Mn3 117.68(14) Mn2 O6 2.177(3) Mn1 O1 Mn2 118.14(15) Mn2 O4 2.207(3) Mn3 O1 Mn2 1 17.59(15) Mn3 O1 1.875(3) Mn6 O11 Mn5 117.02(14) Mn3 O8 1.933(3) Mn6 O11 Mn4 121.29(14) Mn3 N5 2.030(4) Mn5 O11 Mn4 112.91(13) Mn3 N6 2.039(4) Mn1 Mn2 Mn3 59.082(19) Mn3 O10 2.182(4) Mn2 Mn3 Mn1 60.362(19) Mn3 O2 2.193(3) Mn3 Mn1 Mn2 60.556(20) Mn4 O11 2.115(3) Mn4 Mn5 Mn6 64.692(20) Mn4 O15 2.204(3) Mn5 Mn6 Mn4 60.098(20) Mn4 O14 2.223(3) Mn6 Mn4 Mn5 55.210(18)

PAGE 255

255 Table C 7. Selected interatomic distances (Ã…) and angles ( o ) for complex 3 3 . Parameters Parameters Mn1 Mn2 3.202(2) Mn4 O14 2.153(6) Mn1 Mn3 3.206(2) Mn4 O11 2.179(6) Mn2 Mn3 3.199(2) Mn1 O4 Mn2 118.0(3) Mn4 Mn4' (a) 3.207(2) Mn1 O4 Mn3 118.2(3) Mn1 O4 1.859(7) Mn2 O4 Mn3 116.9(4) Mn1 O10 1.939(7) Mn4 O12 Mn4' 117.42(15) Mn1 N5 2.038(9) Mn2 Mn1 Mn 3 59.88(5) Mn1 N6 2.053(9) Mn3 Mn2 Mn1 60.12(5) Mn1 O5 2.157(7) Mn2 Mn3 Mn1 59.99(5) Mn1 O1 2.194(6) Mn4 Mn4' Mn4'' 60.0 Mn2 O4 1.877(7) Mn2 O6 1.945(8) Mn2 N3 2.023(8) Mn2 N4 2.047(9) Mn2 O7 2.131(7) Mn2 O3 2.166(7) Mn3 O4 1.877(6) Mn3 O8 1.923(6) Mn3 N1 2.016(7) Mn3 N2 2.018(7) Mn3 O2 2.196(7) Mn3 O9 2.197(7) Mn4 O12 1.876(2) Mn4 O13 1.926(6) Mn4 N7 2.023(7) Mn4 N8 2.036(7) a) Primed and unprimed atoms are related by symmetry.

PAGE 256

256 Table C 8. Selected inter atomic distances (Ã…) and angles ( o ) for complex 3 4 . Parameters Parameters Mn1 Mn2 3.2073(14) Mn4 O12 2.167(4) Mn1 Mn3 3.2059(12) Mn4 O5 2.191(4) Mn2 Mn3 3.2036(13) Mn1 O1 Mn2 117.6(2) Mn4 Mn4' (a) 3.2094(12) Mn1 O1 Mn3 117 .9(2) Mn1 O1 1.873(4) Mn2 O1 Mn3 117.6(2) Mn1 O10 1.931(5) Mn4 O14 Mn4' 117.89(9) Mn1 N2 2.027(5) Mn3 Mn1 Mn2 59.94(3) Mn1 N1 2.037(6) Mn3 Mn2 Mn1 60.01(3) Mn1 O6 2.166(5) Mn2 Mn3 Mn1 60.05(3) Mn1 O4 2.182(4) Mn4 Mn4' Mn4'' 60.0 Mn2 O1 1.877(4) Mn2 O7 1.920(5) Mn2 N4 2.026(5) Mn2 N3 2.028(6) Mn2 O8 2.163(5) Mn2 O2 2.185(4) Mn3 O1 1.869(4) Mn3 O9 1.927(4) Mn3 N6 2.024(5) Mn3 N5 2.026(5) Mn3 O11 2.208(4) Mn3 O3 2.221(4) Mn4 O14 1.8732(11) Mn4 O13 1.934(4) Mn4 N8 2.026(4) Mn4 N7 2.038(4) a) Primed and unprimed atoms are related by symmetry.

PAGE 257

257 Table C 9. Selected interatomic distances (Ã…) and angles ( o ) for complex 4 1 . Parameters Parameters Mn1 Mn1' 3.2005(7) O1' Mn1 Mn1'' 104.84 ( 5 ) Mn1 O3 1.8704(6) O4 Mn1 Mn1'' 71.58 ( 6 ) Mn1 O2 1.935(2) O3 Mn1 Mn1' 31.18 ( 2 ) Mn1 N2 2.012(2) O2 Mn1 Mn1' 82.25 ( 7 ) Mn1 N1 2.033(2) N2 Mn1 Mn1' 106.69 ( 6 ) Mn1 O1 2.1670(19) N1 Mn1 Mn1' 147.15 ( 7 ) Mn1 O4 2.200(2) O1' Mn1 62.49 ( 5 ) O3 Mn1 O2 96.99 ( 1 0) O4 Mn1 Mn1' 116.1 5 ( 6 ) O3 Mn1 N2 90.91 ( 9 ) Mn1 Mn1' Mn1" 60.0 O2 Mn1 N2 171.06 ( 9 ) Mn1 O3 Mn1' 117.64(5) O3 Mn1 N1 167.66 ( 11 ) O2 Mn1 N1 94.17 ( 1 0) N2 Mn1 N1 78.4 ( 1 ) O3 Mn1 O1' 89.36 ( 8 ) O2 Mn1 O1' 91.44 ( 8 ) N2 Mn1 O1' 92.83 ( 8 ) N1 Mn1 O1' 85.08 ( 8 ) O3 Mn1 O 4 88.31 ( 8 ) O2 Mn1 O4 92.13 ( 9 ) N2 Mn1 O4 83.90 ( 9 ) N1 Mn1 O4 96.56 ( 9 ) O1' Mn1 O4 175.95 ( 8 ) O3 Mn1 Mn1'' 31.18 ( 2 ) O2 Mn1 Mn1'' 122.59 ( 7 ) N2 Mn1 Mn1'' 63.69 ( 6 ) N1 Mn1 Mn1'' 140.96 ( 7 ) a) Primed and unprimed atoms are related by symmet ry.

PAGE 258

258 Table C 10. Selected interatomic distances (Ã…) and angles ( o ) for complex 5 1 . Parameters Parameters Mn1 Mn2 3.198(4) Mn3 N3 2.033(17) Mn1 Mn3 3.214(5) Mn3 O8 2.214(12) Mn2 Mn3 3.200(5) Mn3 O10 2.221(13) Mn4 Mn5 3.195(5) Mn4 O2 1. 854(12) Mn4 Mn6 3.212(5) Mn4 O15 1.972(13) Mn5 Mn6 3.223(5) Mn4 N10 1.992(15) Mn7 Mn8 3.211(5) Mn4 N9 2.001(17) Mn7 Mn9 3.193(5) Mn4 O14 2.201(14) Mn8 Mn9 3.210(5) Mn4 O16 2.202(13) Mn10 Mn11 3.211(5) Mn5 O2 1.907(12) Mn10 M n12 3.192(5) Mn5 O17 1.930(13) Mn11 Mn12 3.215(5) Mn5 N8 1.988(15) Mn1 O1 1.831(11) Mn5 N7 2.083(18) Mn1 O7 1.922(12) Mn5 O18 2.176(15) Mn1 N2 2.001(15) Mn5 O19 2.266(12) Mn1 N1 2.053(16) Mn6 O2 1.893(11) Mn1 O5 2.177(13) Mn6 O21 1.965( 13) Mn1 O6 2.232(12) Mn6 N11 1.996(16) Mn2 O1 1.876(11) Mn6 N12 2.004(16) Mn2 O13 1.908(12) Mn6 O20 2.140(16) Mn2 N5 1.965(16) Mn6 O22 2.211(13) Mn2 N6 2.010(15) Mn7 O3 1.916(14) Mn2 O11 2.182(14) Mn7 O24 1.944(12) Mn2 O12 2.196(13) Mn7 N14 2.025(18) Mn3 O9 1.904(13) Mn7 N13 2.063(16) Mn3 O1 1.907(11) Mn7 O25 2.106(14) Mn3 N4 1.961(17) Mn7 O23 2.283(14)

PAGE 259

259 Mn8 O3 1.842(14) Mn12 O37 2.236(14) Mn8 O27 1.949(12) Mn1 O1 Mn2 119.2(6) Mn8 N15 1.975(14) Mn1 O1 Mn3 118.6(6) Mn8 N1 6 1.984(14) Mn2 O1 Mn3 115.6(5) Mn8 O28 2.189(14) Mn4 O2 Mn6 118.0(6) Mn8 O26 2.247(12) Mn4 O2 Mn5 116.4(6) Mn9 O3 1.853(14) Mn6 O2 Mn5 116.1(6) Mn9 O30 1.920(13) Mn8 O3 Mn9 120.6(8) Mn9 N17 1.990(16) Mn8 O3 Mn7 117.4(7) Mn9 N18 1.995(17) Mn9 O3 Mn7 115.8(7) Mn9 O31 2.174(13) Mn10 O4 Mn11 119.1(6) Mn9 O29 2.259(12) Mn10 O4 Mn12 117.8(6) Mn10 O4 1.850(11) Mn11 O4 Mn12 117.9(5) Mn10 O33 1.928(12) Mn1 Mn2 Mn3 60.31(10) Mn10 N21 1.972(16) Mn2 Mn3 Mn1 59.82(10) Mn10 N20 2.053(16) Mn3 Mn1 Mn2 59.88(10) Mn10 O32 2.179(14) Mn4 Mn5 Mn6 60.05(10) Mn10 O34 2.265(11) Mn5 Mn6 Mn4 59.55(10) Mn11 O4 1.875(10) Mn6 Mn4 Mn5 60.40(11) Mn11 N19 1.906(16) Mn7 Mn8 Mn9 59.65(11) Mn11 O39 1.950(13) Mn8 Mn9 Mn7 60.20(11) Mn11 N24 2.023(15) Mn9 Mn 7 Mn8 60.15(11) Mn11 O38 2.192(13) Mn10 Mn11 Mn12 59.57(10) Mn11 O40 2.274(13) Mn11 Mn12 Mn10 60.15(10) Mn12 O4 1.878(11) Mn12 Mn10 Mn11 60.28(10) Mn12 O36 1.952(11) Mn12 N22 1.995(15) Mn12 N23 2.040(16) Mn12 O35 2.228(13)

PAGE 260

260 Table C 11. Selected interatomic distances (Ã…) and angles ( o ) for both cations in the asymmetric unit of complex 5 2 . Parameters Parameters Mn1 Mn2 3.1951(5) Mn4 N7 2.048(4) Mn1 Mn3 3.2073(6) Mn4 O7 2.160(3) Mn2 Mn3 3.2134(5) Mn4 O17 2.228(3) Mn4 Mn5 3.1978(10) Mn4' O2' 1.892(3) Mn4 Mn6 3.1978(11) Mn4' O15' 1.928(3) Mn5 Mn6 3.2175(11) Mn4' N8' 2.018(4) Mn1 O1 1.8742(17) Mn4' N7' 2.021(4) Mn1 O11 1.9362(17) Mn4' O7' 2.201(3) Mn1 N2 2.009(2) Mn4' O17' 2.232(3) Mn1 N1 2.029(2) Mn5 O2 1.860(3) Mn1 O5 2.1951(14) Mn5 O19 1.928(3) Mn1 O9 2.2108(14) Mn5 N10 2.005(4) Mn2 O1 1.8724(16) Mn5 N9 2.010(4) Mn2 O10 1.9275(14) Mn5 O8 2.193(3) Mn2 N4 2.0024(18) Mn5 O16 2.251(3) Mn2 N3 2.029(2) Mn5' O2' 1.849(3) Mn2 O3 2.1671( 16) Mn5' O19' 1.929(3) Mn2 O13 2.2356(16) Mn5' N10' 1.990(4) Mn3 O1 1.8646(15) Mn5' N9' 2.014(4) Mn3 O14 1.9196(16) Mn5' O8' 2.147(4) Mn3 N6 2.018(2) Mn5' O16' 2.264(3) Mn3 N5 2.028(2) Mn6 O2 1.891(3) Mn3 O4 2.1619(19) Mn6 O18 1.924(3) Mn3 O12 2.2052(19) Mn6 N12 2.016(4) Mn4 O2 1.876(3) Mn6 N11 2.053(4) Mn4 O15 1.915(3) Mn6 O6 2.181(3) Mn4 N8 1.986(3) Mn6 O20 2.281(3)

PAGE 261

261 Mn6' O2' 1.891(3) Mn9 O32' 1.778(3) Mn6' O18' 1.926(3) Mn9 O21 1.8648(18) Mn6' N12' 2.005(5) Mn9 N26 1.888(3) Mn6' N11' 2.050(5) Mn9 N25 1.956(4) Mn6' O6' 2.168(3) Mn9 O32 2.086(3) Mn6' O20' 2.284(3) Mn9 N25' 2.128(4) Mn7 Mn8 3.1959(5) Mn9 N26' 2.154(3) Mn7 Mn9 3.2095(6) Mn9 O23 2.182(2) Mn8 Mn9 3.1980(6) Mn9 O34 2.277(2) Mn10 M n11 3.1996(11) Mn10 O22 1.872(3) Mn10 Mn12 3.207(1) Mn10 O37 1.931(3) Mn11 Mn12 3.2013(10) Mn10 N28 2.005(3) Mn20 Mn22 3.1893(12) Mn10 N27 2.029(4) Mn20 Mn21 3.2023(12) Mn10 O28 2.124(3) Mn21 Mn22 3.2143(12) Mn10 O35 2.213(3) Mn 7 O21 1.8718(17) Mn11 O22 1.869(3) Mn7 O29 1.9216(17) Mn11 O36 1.953(3) Mn7 N22 2.002(2) Mn11 N30 1.989(4) Mn7 N21 2.036(2) Mn11 N29 2.031(4) Mn7 O24 2.1597(14) Mn11 O26 2.186(3) Mn7 O31' 2.174(3) Mn11 O39 2.202(3) Mn7 O31 2.349(3) Mn12 O22 1.870(3) Mn8 O21 1.8756(16) Mn12 O40 1.948(3) Mn8 O33 1.9184(17) Mn12 N31 2.024(4) Mn8 N24 2.009(2) Mn12 N32 2.031(4) Mn8 N23 2.0276(18) Mn12 O27 2.178(3) Mn8 O25 2.096(3) Mn12 O38 2.232(3) Mn8 O25' 2.311(3) Mn8 O30 2.2120(16)

PAGE 262

262 Mn20 O22' 1.871(4) Mn9 O21 Mn7 118.39(9) Mn20 O37' 1.931(3) Mn9 O21 Mn8 117.52(9) Mn20 N28' 2.013(4) Mn7 O21 Mn8 117.04(8) Mn20 N27' 2.034(4) Mn11 O22 Mn12 117.78(16) Mn20 O28' 2.139(4) Mn11 O22 Mn10 117.61(15) Mn20 O35' 2.207(3) Mn12 O22 Mn10 1 17.95(17) Mn21 O22' 1.883(4) Mn22 O22' Mn20 117.10(19) Mn21 O36' 1.913(3) Mn22 O22' Mn21 117.97(19) Mn21 N30' 2.003(4) Mn20 O22' Mn21 117.10(18) Mn21 N29' 2.033(5) Mn21 O26' 2.147(4) Mn21 O39' 2.243(3) Mn22 O22' 1.868(4) Mn22 O40' 1 .932(3) Mn22 N31' 2.031(4) Mn22 N32' 2.044(4) Mn22 O27' 2.170(3) Mn22 O38' 2.197(3) Mn1 O1 Mn2 117.03(8) Mn1 O1 Mn3 118.15(8) Mn2 O1 Mn3 118.61(9) Mn4 O2 Mn5 117.70(18) Mn4 O2 Mn6 116.17(15) Mn5 O2 Mn6 118.14(17) Mn5' O2' Mn6' 118.61(18) Mn5' O2' Mn4' 117.47(18) Mn6' O2' Mn4' 115.62(15) a) Primed and unprimed atoms are related by symmetry.

PAGE 263

263 Table C 12. Selected interatomic distances (Ã…) and angles ( o ) for complex 5 3 . Parameters Parameters Mn1 Mn2 3.2 011(4) O1 Mn1 Mn2 31.48(4) Mn1 Mn3 3.2132(4) O5 Mn1 Mn2 81.88(4) Mn2 Mn3 3.2168(4) N5 Mn1 Mn2 105.58(5) Mn1 O1 1.8719(12) N6 Mn1 Mn2 146.93(5) Mn1 O5 1.9304(13) O2 Mn1 Mn2 62.19(3) Mn1 N5 2.0051(16) O10 Mn1 Mn2 115.49(4) Mn1 N6 2.0181(16) O1 Mn1 Mn3 31.11(4) Mn1 O2 2.1596(13) O5 Mn1 Mn3 118.55(4) Mn1 O10 2.1815(13) N5 Mn1 Mn3 64.49(5) Mn2 O1 1.8790(12) N6 Mn1 Mn3 142.50(5) Mn2 O7 1.9172(13) O2 Mn1 Mn3 105.92(4) Mn2 N1 2.0014(15) O10 Mn1 Mn3 71.27(3) Mn2 N2 2.0363(16) O1 Mn2 Mn1 31.35(4) Mn2 O3 2.1641(13) O7 Mn2 Mn1 123.30(4) Mn2 O6 2.2388(13) N1 Mn2 Mn1 63.51(4) Mn3 O1 1.8785(12) N2 Mn2 Mn1 140.33(5) Mn3 O9 1.9393(13) O3 Mn2 Mn1 101.65(4) Mn3 N3 2.0057(15) O6 Mn2 Mn1 72.00(3) Mn3 N4 2.0300(16) O1 Mn2 Mn3 31.12(4) Mn3 O4 2.2050(13) O7 M n2 Mn3 80.39(4) Mn3 O8 2.2157(13) N1 Mn2 Mn3 110.19(5) Mn1 O1 Mn2 117.17(6) N2 Mn2 Mn3 152.70(5) Mn1 O1 Mn3 117.91(6) O3 Mn2 Mn3 62.26(3) Mn2 O1 Mn3 117.76(6) O6 Mn2 Mn3 112.00(3) Mn2 Mn1 Mn3 60.198(9) O1 Mn3 Mn1 30.98(4) Mn1 Mn2 Mn3 60.087(9) O9 Mn3 Mn1 82.02(4) Mn1 Mn3 Mn2 59.715(8) N3 Mn3 Mn1 105.92(4)

PAGE 264

264 N4 Mn3 Mn1 151.83(5) O4 Mn3 Mn1 61.80(3) O8 Mn3 Mn1 112.23(3) O1 Mn3 Mn2 31.13(4) O9 Mn3 Mn2 125.09(4) N3 Mn3 Mn2 63.14(5) N4 Mn3 Mn2 139.48(5) O4 Mn3 Mn2 101.11(4) O8 Mn3 M n2 72.48(3)

PAGE 265

265 LIST OF REFERENCES (1) Lehn, J. M. Pure Appl. Chem. 1978 , 50 , 871. (2) Lehn, J. M. Angew. Chem. 1988 , 100 , 91. (3) Lehn, J. M. Angew. Chem. Int. Ed. Engl. 1988 , 27 , 89. (4) Villiers, A.; Hebd, C . R. Seances Acad. Sci. 1891 , 112 , 435. (5) Villiers, A.; Hebd, C. R. Seances Acad. Sci. 1891 , 112 , 536. (6) Szejtli, J. Cyclodextrin Technology ; Springer: New York, 1988; Vol. 1. (7) Werner, A. Zeitschr. Anorg. Chem. 1893 , 3 , 267. (8) Fischer, E. Ber. Deutsch. Chem. Ges. 1894 , 27 , 2985. (9) Ehrlich, P. Studies on Immunity ; Wiley: New York, 1906. (10) Pfeiffer, P. Organische Molekulverbindungen ; Enke: Stutgart, 1927. (11) Wolf, K. L.; Frahm, H.; Harms, H. Z. Phys. Chem. (B) 1937 , 36 , 237. (1 2) Lehninger, A. L. Naturwiss. 1966 , 53 , 57. (13) Pressman, B. C. Proc. Natl. Acad. Sci. USA 1965 , 53 , 1077. (14) Mueller, P.; Rudin, D. O. Biochem. Biophys. Res. Commun. 1967 , 26 , 398. (15) Andreoli, T. E.; Tieffenberg, M.; Tosteson, D. C. J. Gen. B iol. 1967 , 50 , 2527. (16) Pressman, B. C.; Harris, E. J.; Jagger, W. S.; Johnson, J. H. Proc. Natl. Acad. Sci. USA 1967 , 58 , 1949. (17) Kilbourn, B. T.; Dunitz, J. D.; Pioda, L. A. R.; Simon, W. J. Mol. Biol. 1967 , 30 , 559. (18) Stefanac, Z.; Simon, W . Microchem. J. 1967 , 12 , 125. (19) Ovchinnikov, Y. A.; Ivanov, V. T.; Skrob, A. M. Membrane Active Complexones ; Elsevier: New York, 1974. (20) Pressman, B. C. Annu. Rev. Biochem. 1976 , 45 , 501. (21) Pedersen, C. J. J. Am. Chem. Soc. 1967 , 89 , 7017. (22) Dietrich, B.; Lehn, J. M.; Sauvage, J. P. Tetrahedron Lett. 1969 , 2885.

PAGE 266

266 (23) Dietrich, B.; Lehn, J. M.; Sauvage, J. P.; Blanzat, J. Tetrahedron 1973 , 29 , 1629. (24) Lehn, J. M. Struct. Bonding 1973 , 16 , 1. (25) Pedersen, C. J. Angew. Chem. Int. Ed. Engl. 1988 , 27 , 1053. (26) Cram, D. J.; Cram, J. M. Science 1974 , 183 , 803. (27) Chakrabarty, R.; Mukherjee, P. S.; Stang, P. J. Chem. Rev. 2011 , 111 , 6810. (28) Cram, D. J. Angew. Chem. Int. Ed. Engl. 1986 , 25 , 1039. (29) Stark, C. M. J. Am. Chem. Soc 1971 , 93 , 195. (30) Roy, X.; Lee, C. H.; Crowther, A. C.; Schenck, C. L.; Besara, T.; Lalancette, R. A.; Siegrist, T.; Stephens, P. W.; Brus, L. E.; Kim, p.; Steigerwald, M. L.; Nuckolls, C. Science 2013 , 341 , 157. (31) Capenter, G. B.; Richa rd, S. M. Acta. Cryst. 1962 , 15 , 360. (32) Emsley, J. Chem. Soc. Rev. 1980 , 9 , 91. (33) Sunner, J.; Nishizawa, K.; Kebarle, P. J. Phys. Chem. 1981 , 85 , 1814. (34) Deakyne, C. A.; Meot Ner, M. J. Am. Chem. Soc. 1985 , 107 , 474. (35) Gallivan, J. P.; Dougherty, D. A. Proc. Natl. Acad. Sci. USA 1999 , 96 , 9459. (36) Rosokha, Y. S.; Lindeman, S. V.; Rosokha, S. V.; Kochi, J. K. Angew. Chem., Int. Ed. 2004 , 43 , 4650. (37) Park, Y. C.; Lee, J. S. J. Phys. Chem. A 2006 , 110 , 5091. (38) Martinez, C. R.; Iverson, B. L. Chem. Sci. 2012 , 3 , 2191. (39) McGaughey, G. B.; Gagné, M.; Rappé, A. K. J. Biol. Chem. 1998 , 273 , 15458. (40) Xia, J.; Bacon, J. W.; Jasti, R. Chem. Sci. 2012 , 3 , 3018. (41) Autumn, K.; Sitti, M.; Liang, Y. A.; Peattie, A. M.; Hansen , W. R.; Sponberg, S.; Kenny, T. W.; Fearing, R.; Israelachvili, J. N.; Full, R. J. Proc. Natl. Acad. Sci. USA 2002 , 99 , 12252. (42) Andreetti, G. D.; Ungaro, R.; Pochini, A. J. Chem. Soc., Chem. Commun. 1979 , 1005.

PAGE 267

267 (43) Lehn, J. M.; Editors: Williams, A. F.; Floriani, C.; Merbach, A. E. Perspectives in Coordination Chemistry VCH: Weinheim, 1992. (44) Piguet, C.; Bernardinelli, G.; Hopfgartner, G. Chem. Rev. 1997 , 97 , 2005. (45) Albrecht, M. Chem. Rev. 2001 , 101 , 3457. (46) Ruben, M.; Rojo, J.; Rom ero Salguero, F. J.; Uppadine, L. H.; Lehn, J. M. Angew. Chem., Int. Ed. 2004 , 43 , 3644. (47) Amabilino, D. B.; Stoddart, J. F. Chem. Rev. 1995 , 95 , 2725. (48) Lukin, O.; Vogtle, F. Angew. Chem., Int. Ed. 2005 , 44 , 1456. (49) Forgan, R. S.; Sauvage, J. P.; Stoddart, J. F. Chem. Rev. 2011 , 111 , 5434. (50) Sauvage, J. P.; Dietrich, B., Eds. Molecular Catenanes, Rotaxanes, and Knots: A Journey Through the World of Molecular Topology ; Wiley VCH: Weinheim, Germany, 1999. (51) Faiz, J. A.; Heitz, V.; Sa uvage, J. P. Chem. Soc. Rev. 2009 , 38 , 422. (52) Crowley, J. D.; Goldup, S. M.; Lee, A. L.; Leigh, D. A.; McBurney, R. T. Chem. Soc. Rev. 2009 , 38 , 1530. (53) Balzani, V.; Credi, A.; Venturi, M. Chem. Soc. Rev. 2009 , 38 , 1542. (54) Swiegers, G. F.; M alefetse, T. J. Chem. Rev. 2000 , 100 , 3483. (55) Fang, L.; Olson, M. A.; Benítez, D.; Tkatchouk, E.; Goddard, W. A., III; ; Stoddart, J. F. Chem. Soc. Rev. 2010 , 39 , 17. (56) Beves, J. E.; Leigh, D. A. Nat. Chem. 2010 , 2 , 708. (57) Champin, B.; Mobia n, P.; Sauvage, J. P. Chem. Soc. Rev. 2007 , 36 , 358. (58) Caulder, D. L.; Raymond, K. N. Acc. Chem. Res. 1999 , 32 , 975. (59) Han, Y. F.; Jia, W. G.; Yu, W. B.; Jin, G. X. Chem. Soc. Rev. 2009 , 38 , 3419. (60) Ward, M. D. Chem. Commun. (Review) 2009 , 4 487. (61) Amouri, H.; Desmarets, C.; Moussa, J. Chem. Rev. 2012 , 112 , 2015. (62) Cotton, F. A.; Lin, C.; Murillo, C. A. Acc. Chem. Res. 2001 , 34 , 759. (63) Fujita, M. Chem. Soc. Rev. 1998 , 27 , 417.

PAGE 268

268 (64) van Leeuwen, P. W. N. M., Ed. Supramolecular C atalysis ; Wiley VCH Weinheim, 2008. (65) Meeuwissen, J.; Reek, J. N. H. Nat. Chem. 2010 , 2 , 615. (66) Hupp, J. T. Struct. Bonding 2006 , 121 , 145. (67) Li, S. S.; Northrop, B. H.; Yuan, Q. H.; Wan, L. J.; Stang, P. J. Acc. Chem. Res. 2009 , 42 , 249. (68) Kudernac, T.; Lei, S.; Elemans, J. A. A. W.; De Feyter, S. Chem. Soc. Rev. 2009 , 38 , 402. (69) Meng, W.; Breiner, B.; Rissanen, K.; Thoburn, J. D.; Clegg, J. K.; Nitschke, J. R. Angew. Chem. Int. Ed. 2011 , 50 , 3479 . (70) Vriezema, D. M.; Aragones, M. C.; Elemans, J. A. A. W.; Cornelissen, J. J. L. M.; Rowan, A. E.; Nolte, R. J. M. Chem. Rev. 2005 , 105 , 1445. (71) Therrien, B.; Suss Fink, G.; Govindaswamy, P.; Renfrew, A. K.; Dyson, P. J. Angew. Chem., Int. Ed. 2008 , 47 , 3773. (72) Leininger, S. ; Fan, J.; Schmitz, M.; Stang, P. J. Proc. Natl. Acad. Sci. USA 2000 , 97 , 1380. (73) Roche, S.; Haslam, C.; Adams, H.; Heath, S. L.; Thomas, J. A. Chem. Commun. 1998 , 1681. (74) Olenyuk, B.; Whiteford, J. A.; Fechtenkötter, A.; Stang, P. J. Nature 1999 , 398 , 796. (75) Caulder, D. L.; Bruckner, C.; Powers, R. E.; Konig, S.; Parac, T. N.; Leary, J. A.; Raymond, K. N. J. Am. Chem. Soc. 2001 , 123 , 8923. (76) Yu, S. Y.; Kusukawa, T.; Biradha, K.; Fujita, M. J. Am. Chem. Soc. 2000 , 122 , 2665. (77) Fujit a, N.; Biradha, K.; Fujita, M.; Sakamoto, S.; Yamaguchi, K. Angew. Chem., Int. Ed. 2001 , 40 , 1718. (78) Bar, A. K.; Chakrabarty, R.; Mostafa, G.; Mukherjee, P. S. Angew. Chem., Int. Ed. 2008 , 47 , 8455. (79) Karthikeyan, S.; Ramamurthy, V. J. Org. Chem. 2006 , 71 , 6409.

PAGE 269

269 (80) Tashiro, S.; Tominaga, M.; Yamaguchi, Y.; Kato, K.; Fujita, M. Angew. Chem., Int. Ed. 2006 , 45 , 241. (81) Oliveri, C. G.; Ulmann, P. A.; Wiester, M. J.; Mirkin, C. A. Acc. Chem. Res. 2008 , 41 , 1618. (82) Gianneschi, N. C.; Berti n, P. A.; Nguyen, S. T.; Mirkin, C. A. J. Am. Chem. Soc. 2003 , 125 , 10508. (83) Gianneschi, N. C.; Masar, M. S., III;; Mirkin, C. A. Acc. Chem. Res. 2005 , 38 , 825. (84) Yoon, H. J.; Heo, J.; Mirkin, C. A. J. Am. Chem. Soc. 2007 , 129 , 14182. (85) Gilbe rt, W. De Magnete ; Mottelay P. F., English trans., Dover Publications, Inc., 1958. (86) Kahn, O. Molecular Magnetism ; VCH: Weinheim, Germany, 1993. (87) Turnbull, M. M.; Sigimoto, T.; Thompson, L. K. Molecule based Magnetic Materials ; ACS: USA, 1996. (88) Spaldin, N. A. Magnetic Materials Fundamentals and Applications ; Cambridge University Press: New York, 2010. (89) Kittel, C.; Galt, J. K.; Campbell, W. E. Phys. Rev. 1950 , 77 , 725. (90) Leggett, A. J. Macroscopic quantum effects in magnetic system s: an overview ; Kluwer: Dordrecht, 1995. (91) Gatteschi, D.; Sessoli, R.; Villain, J. Molecular Nanomagnets ; Oxford University Press: New York, 2006. (92) Sugimoto, T. Fine Particles: Synthesis, Characterization, and Mechanism of Growth ; Marcel Dekker: New York, 2000. (93) Redl, F. X.; Cho, K. S.; Murray, C. B.; O'brien, S. Nature 2003 , 423 , 968. (94) Gatteschi, D.; Canneschi, A.; Pardi, L.; Sessoli, R. Science 1994 , 265 , 1054. (95) Sessoli, R.; Tsai, H. L.; Schake, A. R.; Wang, S.; Vincent, J. B. ; Folting, K.; Gatteschi, D.; Christou, G.; Hendrickson, D. N. J. Am. Chem. Soc. 1993 , 115 , 1804. (96) Sessoli, R.; Gatteschi, D.; Canneschi, A.; Novak, M. A. Nature 1993 , 365 , 141. (97) Sessoli, R. Mol. Cryst. Liq. Cryst. 1995 , 274 , 145.

PAGE 270

270 (98) Novak, M. A.; Sessoli, R.; Canneschi, A.; Gatteschi, D. J. Magn. Magn. Matter. 1995 , 146 , 211. (99) Aubin, S. M. J.; Sun, Z.; Parsi, L.; Krzystek, J.; Folting, K.; Brunel, L. C.; Rheigold, A. A.; Christou, G.; Hendrickson, D. N. Inorg. Chem. 1999 , 38 , 5329. (100) Christou, G.; Gatteschi, D.; Hendrickson, D. N.; Sessoli, R. MRS Bulletin 2000 , 25 , 66. (101) Aubin, S. M. J.; Wemple, M. W.; Adams, D. M.; Tsai, H. L.; Christou, G.; Hendrickson, D. N. J. Am. Chem. Soc. 1996 , 118 , 7746. (102) Tasiopoulos, T. J.; Vinslava, A.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. Angew. Chem., Int. Ed. 2004 , 43 , 2117. (103) Barra, A. L.; Debrunner, P.; Gatteschi, D.; Schulz, C. E.; Sessoli, R. Europhys. Lett. 1996 , 35 , 133. (104) Cadiou, C.; Murrie, M.; Paulsen, C.; Vi llar, V.; Wernsdorfer, W.; Winpenny, R. E. P. Chem. Commun. 2001 , 2666. (105) Yang, E. C.; Hendrickson, D. N.; Wernsdorfer, W.; Nakano, M.; Sommer, R.; Rheingold, A. L.; Ledezma Gairaud, M.; Christou, G. J. Appl. Phys. 2002 , 91 , 7382. (106) Sun, Z.; Gr ant, C. M.; Castro, S. L.; Hendrickson, D. N.; Christou, G. Chem. Commun. 1998 , 721. (107) Ishikawa, N.; Sugita, M.; Ishikawa, t.; Koshihara, S.; Kaizu, Y. J. Am. Chem. Soc. 2003 , 125 , 8694. (108) Rinehart, J. D.; Fang, M.; Evans, W. J.; Long, J. R. J. Am. Chem. Soc. 2011 , 133 14236. (109) Papatriantafyllopoulou, C.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. Inorg. Chem. 2011 , 50 , 421. (110) Bagai, R.; Christou, G. Chem. Soc. Rev. 2009 , 38 , 1011. (111) Casimir, H. B. J.; Du Pre, F. K. Physica 1 938 , V , 507. (112) Cole, K. S.; Cole, R. H. J. Chem. Phys. 1941 , 9 , 341. (113) Thomas, L.; Lionti, F.; Ballou, R.; Gatteschi, D.; Sessoli, R.; Barbara, B. Nature 1996 , 383 , 145.

PAGE 271

271 (114) Friedman, J. R. E.; Sarachik, M. P.; Tejada, J.; Maciejewski, J.; Ziolo, R. J. Appl. Phys. 1996 , 79 , 6031. (115) Friedman, J. R. E.; Sarachik, M. P.; Tejada, J.; Ziolo, R. Phys. Rev. Lett. 1996 , 76 , 3830. (116) Gatteschi, D.; Sessoli, R. Angew. Chem. Int. Ed. 2003 , 42 , 268. (117) Wernsdorfer, W.; Aliaga Alcade, N.; Hendrickson, D. N.; Christou, G. Nature 2002 , 416. (118) Hill, S.; Edwards, R. S.; Aliaga Alcade, N.; Christou, G. Science 2003 , 302 , 1015. (119) Holynska, M.; Premuzic, D.; Jeon, I.; Wernsdorfer, W.; Clerac, R.; Dehnen, S. Chem. Eur. J. 2011 , 17 , 960 5 . (120) Bagai, R.; Wersdorfer, W.; Abboud, K. A.; Christou, G. J. Am. Chem. Soc. 2007 , 129 , 12918. (121) Tiron, R.; Wernsdorfer, W.; Foguet Albiol, D.; Aliaga Alcade, N.; Christou, G. Phys. Rev. Lett. 2003 , 91 , 227203 . (122) Saywell, A.; Britton, A . J.; Taleb, N.; Gimenez Lopez, M. C.; Champness, N. R.; Nanotechnology 2011 , 22 , 075704. (123) Mannini, M.; Pineider, F.; Danieli, C.; Totti, F.; Sorace, L.; Sainctavit, P.; Arrio, M. A.; Otero, E.; Joly, L.; Cezar, J. C.; Co rnia, A.; Sessoli, R. Nature 2010 , 468 , 417. (124) http://en.wikipedia.org/wiki/DNA . (125) Christou, G. Polyhedron 2005 , 24 , 2065. (126) Aromi, G.; Brechin, E. K. Struct. Bonding (Berlin) 2006 , 122 , 1. (127) Miller, J. S.; Epstein, A. J. MRS Bull. 2000 , 25 , 21. (128) Blundell, S. J.; Pratt, F. L. J. Phys.: Condens. Matter. 2004 , 16 , R771. (129) Miller, J. S. Chem. Soc. Rev. 2011 , 40 , 3266. (130) Wernsdorfer, W.; Sessoli, R. Science 1999 , 284 , 133. (131) Wernsdorfer, W.; Soler, M.; Christou, G.; Hendrickson, D. N. J. Appl. Phys. 2002 , 91 , 7164. (132) Wernsdorfer, W.; Chakov, N. E.; Christou, G. Phys. Rev. Lett. 2005 , 95 , 037203.

PAGE 272

272 (133) Wernsdorfer, W.; Bhaduri, S.; Tiron, R.; Hendrickson, D. N.; Christou, G. Phys. Rev. Lett. 2002 , 89 , 197201. (134) Wilson, A.; Hill, S.; Edwards, R. S.; Aliaga Alcalde, N.; G., C. AIP Conf. Proc. 2006 , 850 , 1141. (135) Leuenberger, M. N.; Loss, D. Nature 2001 , 410 , 789. (136) Zhou, B.; Tao, R.; Shen, S. Q.; L iang, J. Q. Phys. Rev. A. 2002 , 66 , 010301(R). (137) Affronte, M.; Troiani, F.; Ghirri, A.; Candini, A.; Evangelisti, M.; Corradini, V.; Carretta, S.; Santini, P.; Amoretti, G.; Tuna, F.; Timco, G.; Winpenny, R. E. P. J. Phys. D: Appl. Phys. 2007 , 40 , 29 99. (138) Vincent, R.; Klyatskaya, S.; Ruben, M.; Wernsdorfer, W.; Balestro, F. Nature 2012 , 488 , 357. (139) Bogani, L.; Wernsdorfer, W. Nature Mater. 2008 , 7 , 179. (140) Katoh, K.; Isshiki, H.; Komeda, T.; Yamashita, M. Chem. Asian. J. 2012 , 7 , 1154 . (141) Tiron, R.; Wernsdorfer, W.; Aliaga Alcade, N.; Christou, G. Phys. Rev. B. 2003 , 68 , 140407. (142) Yang, E.; Wernsdorfer, W.; Hill, S.; Edwards, R. S.; Nakano, M.; Maccagnano, S.; Zakharov, L. N.; Rheingold, A. N.; Christou, G.; Hendrickson, D. N. Polyhedron 2003 , 22 , 1727. (143) Wittick, L. M.; Murray, K. S.; Moubaraki, B.; Batten, S. R.; Spiccia, L.; Berry, K. J. Dalton Trans. 2004 , 1003. (144) Inglis, R.; Jones, L. F.; Mason, K.; Collins, A.; Moggach, S. A.; Parsons, S.; Perlepes, S. P.; W ernsdorfer, W.; Brechin, E. K. Chem. Eur. J. 2008 , 14 , 9117. (145) Das, A.; Gieb, K.; Krubskaya, Y.; Demeshko, S.; Dechert, S.; Klingeler, R.; Kataev, V.; Buchner, B.; Muller, P.; Meyer, F. J. Am. Chem. Soc. 2011 , 133 , 3433. (146) Coulon, C.; Miyasaka, H.; Clérac, R. Struct. Bonding 2006 , 122 , 163. (147) Xu, H. B.; Wang, B. W.; Pan, F.; Wang, Z. M.; S., G. Angew. Chem. Int. Ed. 2007 , 46 , 7388. (148) Miyasaka, H.; Yamashita, M. Dalton Trans. 2007 , 399. (149) Bogani, L.; Vindigni, A.; Sessoli, R.; G atteschi, D. J. Mater. Chem. 2008 , 18 , 4750.

PAGE 273

273 (150) Roubeau, O.; Clérac, R. Eur. J. Inorg. Chem. 2008 , 4325. (151) Jeon, I. R. C., R. Dalton Trans. 2012 , 41 , 9569. (152) Miyasaka, H.; Nakata, K.; Lecren, L.; Coulon, C.; Nakazawa, Y.; Fujisaki, T.; Sugi ura, K.; Yamashita, M.; R., C. J. Am. Chem. Soc. 2006 , 128 , 3770. (153) Langley, S. K.; Chilton, N. F.; Moubaraki, B.; Murray, K. S. Dalton Trans. 2011 , 40 , 12201. (154) Shiga, T.; Miyasaka, H.; Yamashita, M.; Morimoto, M.; Irie, M. Dalton Trans. 2010 , 40 , 2275. (155) Grillo, V. A.; Knapp, M.; Bollinger, J. C.; Hendrickson, D. N.; Christou, G. Angew. Chem. Int. Ed. Engl. 1996 , 35 , 1818. (156) Pascu, M.; Lloret, F.; Avarvari, N.; Julve, M.; Andruh, M. Inorg. Chem. 2004 , 43 , 5189. (157) Wu, G.; Cler ac, R.; Wersdorfer, W.; Qiu, S.; Anson, C. E.; Hewitt, I. J.; Powell, A. K. Eur. J. Inorg. Chem. 2006 , 1927. (158) Inglis, R.; Katsenis, A. D.; Collins, A.; White, F.; Milios, C. J.; Papaefstathiou, G. S.; Brechin, E. K. CrystEngCom. 2010 , 12 , 2064. (1 59) Sanudo, E. C.; Uber, J. S.; Balague, A. P.; Roubeau, O.; Aromi, G. Inorg. Chem. 2012 , 51 , 8441. (160) Albores, P.; Plenk, C.; Renschler, E. Inorg. Chem. 2012 , 51 , 8373. (161) Kumari, H.; Dennis, C. L.; Mossine, A. V.; Deakyne, C. A.; Atwood, J. L. ACS Nano 2012 , 6 , 272. (162) Whitehead, G. F. S.; Moro, F.; Timco, G. A.; Wernsdorfer, W.; Teat, S. T.; Winpenny, R. E. P. Angew. Chem. Int. Ed. 2013 , 52 , 9932. (163) Stamatatos, T. C.; Abboud, K. A.; Wernsdorfer, W.; Christou, G. Angew. Chem., Int. Ed . 2008 , 47 , 6694. (164) Costa, J. S.; Barrios, L. A.; Craig, G. A.; Teat, S. J.; Luis, F.; Roubeau, O.; Evangelisti, M.; Camon, A.; Aromi, G. Chem. Commun. 2012 , 48 , 1413. (165) Bellini, V.; Lorusso, G.; Candini, A.; Wernsdorfer, W.; Faust, T. B.; Timc o, G. A.; Winpenny, R. E. P.; Affronte, M. Phys. Rev. Lett. 2011 , 106 , 227205. (166) Chem. Soc. Rev. 2012 , 41 , 537.

PAGE 274

274 (167) Chakravorty, A. Coord. Chem. Rev. 1974 , 13 , 1. (168) Chaudhuri, P. Coord. Chem . Rev. 2003 , 243 , 143. (169) Miyasaka, H.; Clerac, R. Bull. Chem. Soc. Jpn. 2005 , 78 , 1725. (170) Milios, C. J.; Stamatatos, T. C.; Perlepes, S. P. Polyhedron 2006 , 25 , 134. (171) Stamatatos, T. C.; Foguet Albiol, D.; Lee, S. C.; Stoumpos, C. C.; Rap topoulou, C. P.; Terzis, A.; Wersdorfer, W.; Hill, S. O.; Perlepes, S. P.; Christou, G. J. Am. Chem. Soc. 2007 , 129 , 9484. (172) Vincent, J. B.; Chang, H. R.; Folting, k.; Huffman, J. C.; Christou, G.; Hendrickson, D. N. J. Am. Chem. Soc. 1987 , 109 , 5703 . (173) Constable, E. C.; Lewis, J.; Liptrot, M. C.; Raithby, P. R. Inorganica Chimica Acta. 1990 , 178 , 47. (174) Nguyen, T. N.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. J. Am. Chem. Soc. 2011 , 133 , 20688. (175) SHELXTL6; Bruker AXS: Madison, Wis consin, USA, 2008. (176) Davidson, E. R.; Indiana University: Bloomington, IN, 1999. (177) Wernsdorfer, W. Adv. Chem. Phys. 2001 , 118 , 99. (178) Wemple, M. W.; Tsai, H. L.; Folting, K.; Hendrickson, D. N.; Christou, G. Inorg. Chem. 1993 , 32 , 2025. (179) Eppley, H. J.; Tsai, H. L.; de Vries, N.; Folting, K.; Christou, G.; Hendrickson, D. N. J. Am. Chem. Soc. 1995 , 117 , 301. (180) Cannon, R. D.; Jayasooriya, U. A.; Wu, R.; arapKoske, S. K.; Stride, J. A.; Nielsen, O. F.; White, R. P.; Kearley, G. J. ; Summerfields, D. J. Am. Chem. Soc. 1994 , 116 , 11869. (181) Kambe, K. J. Phys. Soc. Jpn. 1950 , 5 , 48. (182) Van Vleck, J. H. The Theory of Electric and Magnetic Susceptibilities ; Oxford University Press, 1932. (183) Cano, J.; Cauchy, T.; Ruiz, E.; M ilios, C. J.; Stoumpos, C. C.; Stamatatos, T. C.; Perlepes, S. P.; Christou, G.; Brechin, E., K. Dalton Trans. 2008 , 234.

PAGE 275

275 (184) Atanasov, M.; Delley, B.; Neese, F.; Tregenna Piggott, P. L.; Sigrist, M. Inorg. Chem. 2011 , 50 , 2112. (185) Aliaga Alcade, N.; Edwards, R. S.; Hill, S. O.; Wernsdorfer, W.; Folting, K.; Christou, G. J. Am. Chem. Soc. 2004 , 126 , 12503. (186) Yoo, J.; Yamaguchi, A.; Nakano, M.; Krzystek, J.; Streib, W. E.; Brunel, L. C.; Ishimoto, H.; G.;, C.; Hendrickson, D. N. Inorg. Chem. 2 001 , 40 , 4604. (187) Inglis, R.; Taylor, S. M.; Jones, L. F.; Papaefstathiou, G. S.; Perlepes, S. P.; Datta, S.; Hill, S.; Wernsdorfer, W.; Brechin, E. K. Dalton Trans. 2009 , 9157. (188) Chakov, N. E.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. Inorg . Chem. 2004 , 43 , 5919. (189) Mishra, A.; Tasiopoulos, A. J.; Wernsdorfer, W.; Abboud, K. A.; Christou, G. Inorg. Chem. 2007 , 46 , 3105. (190) Wernsdorfer, W.; Aliaga Alcade, N.; Hendrickson, D. N.; Christou, G. Nature 2002 , 406. (191) Hassan, A. K.; Pardi, L. A.; Krzystek, J.; Sienkiewicz, A.; Goy, P.; Rohrer, M.; Brunel, L. C. J. Magn. Reson. 2000 , 142 , 300. (192) Hill, S.; Edwards, R. S.; Jones, S. I.; Dalal, N. S.; North, J. M. Phys. Rev. Lett. 2003 , 90 , 217204. (193) Hill, S.; Maccagnano, S.; Achey, R.; Dalal, N.; Park, K. Int. J. Mod. Phys. B. 2002 , 16 , 3326. (194) Park, K.; Novotny, M. A.; Dalal, N. S.; Hill, S.; Rikvold, P. A. Phys. Rev. B. 2001 , 65 , 014426. (195) Stoll, S.; Schweiger, A. J . Magn . Reson . 2006 , 178 , 42. (196) , N. V.; Stamp, P. C. E. Phys. Rev. Lett. 1998 , 80 , 5794. (197) Vincent, J. B.; Christmas, C.; Chang, H. R.; Li, Q.; Boyd, P. D. W.; Huffman, J. C.; Hendrickson, D. N.; Christou, G. J. Am. Chem. Soc. 1989 , 111 , 2086. (198) Richardson, C.; Steel, P. J.; D'Alessandro, D. M.; Junk, P. C.; Keene, F. R. Dalton Trans. 2002 , 2775. (199) Kalinina, M. I.; Moiseev, K. Chem. Heterocycl. Compd. 1988 , 24 , 217.

PAGE 276

276 (200) Godovikova, T. I.; Golova, S. P.; Strelenko, Y. A.; Antipin, M. Y.; Struchkov, Y. ii, L. I. Mendeleev Commun. 1994 , 7. (201) Pasinszki, T.; Havasi, B.; Hajgató, B.; P., W. N. J. Phys. Chem. A. 2009 , 113 , 170. (202) Beissel, T.; Powers, R. E.; Raymond, K. N. Angew. Chem. Int. Ed. Engl. 1996 , 35 , 1084. (203) Caulder, D. L.; Powers, R. E.; Parac, T. N.; Raymond, K. N. Angew. Chem., Int. Ed. 1998 , 37 , 1840. (204) Mal, P.; Schultz, D.; Beyeh, K.; Rissanen, K.; Nitschke, J. R. Angew. Chem. Int. Ed. 2008 , 47 , 8297. (205) Clegg, J. K.; Li, F.; Jolliffe, K. A.; Meehan, G. V.; Lindoy, L. F. Chem. Commun. 2011 , 47 , 6042. (206) Caulder, D. L.; Raymond, K. N. Dalton Trans. 1999 , 1185. (207) Bourgeois, J. P.; Fujita, M.; Kawano, M.; Sakamoto, S.; Yamaguchi, K. J. Am. Chem. Soc. 2003 , 125 , 9260. (208) Cui, H. B.; Takahashi, K.; Okano, Y. ; Wang, Z. M.; Kobayshi, H.; Kobayashi, A. Angew. Chem., Int. Ed. 2005 , 44 , 6508. (209) Cui, H. B.; Wang, Z. M.; Takahashi, K.; Okano, Y.; Kobayshi, H.; Kobayashi, A. J. Am. Chem. Soc. 2006 , 128 , 15074. (210) Cui, H. B.; Zhou, B.; Long, L. S.; Okano, Y .; Kobayashi, H.; Kobayashi, A. Angew. Chem., Int. Ed. 2008 , 47 , 3376. (211) Zhang, W.; Xiong, R. G. Chem. Rev. 2012 , 112 , 1163. (212) Vázquez, M.; Taglietti, A.; Gatteschi, D.; Sorace, L.; Sangregorio, C.; González, A. M.; Maneiro, M.; Pedrido, R. M.; Bermejo, M. R. Chem. Commun. 2003 , 1840. (213) Nguyen, T. N.; Abboud, K. A.; Christou, G. Polyhedron 2013 , 66 , 171. (214) Saha, A.; Abboud, K. A.; Christou, G. Inorg. Chem. 2012 , 50 , 12774. (215) Bouwman, E.; Huffman, J. C.; Lobkovsky, E. B.; Christ ou, G.; Tsai, H. L.; Hendrickson, D. N. Inorg. Chem. 1992 , 31 , 4436. (216) Ung, V. A.; Thompson, A. M. W. C.; Bardwell, D. A.; Gatteschi, D.; McCleverty, J. A.; Totti, F.; Ward, M. D. Inorg. Chem. 1997 , 36 , 3447.

PAGE 277

277 (217) Fernadez, I.; Ruiz, R.; Faus, J.; Julve, M.; Lloret, F.; Cano, J.; Ottenwaelder, X.; Journaux, Y.; Munoz, M. C. Angew. Chem., Int. Ed. 2001 , 40 , 3129. (218) Glaser, T.; Heidemeier, M.; Grimme, S.; Bill, E. Inorg. Chem. 2004 , 43 , 5192. (219) Dul, M. C.; Ottenwaelder, x.; Pardo, E.; Lesc ouezec, R.; Journaux, Y.; Chamoreau, L. M.; Ruiz Inorg. Chem. 2009 , 48 , 5244. (220) Babska, A.; Bielski, L.; Kuczynski, L.; Respond, S.; Witek, H. Pol. J. Pharmacol. Phar. 1973 , 25 , 175. (221) Ingle, J. D. J. ; Crouch, S. R. Spectrochemical Analysis ; Prentice Hall: New Jersey, 1988. (222) Borrás Almenar, J. J.; Clemente Juan, J. M.; Coronado, E.; Tsukerblat, B. S. J. Comput. Chem. 2001 , 22 , 985. (223) Takahashi, S.; Hill, S. Rev . Sci . Instrum . 2005 , 76 . ( 224) Mola, M.; Hill, S.; Goy, P.; Gross, M. Rev . Sci . Instrum . 2000 , 71 , 186. (225) Hassan, A. K.; Pardi, L. A.; Krzystek, J.; Sienkiewicz, A.; Goy, P.; Rohrer, M.; Brunel, L. C. J . Magn . Reson . 2000 , 142 , 300. (226) Eddaoudi, M.; Kim, J. R., N.;; Voda k, D.; Wachter, J.; O'Keeffe, M.; Yaghi, O. M. Science 2002 , 295 , 469. (227) Chem. Rev. 2014 , 114 , 1343. (228) Science 2013 , 341 , 974. (229) Gerbier, P.; Ruiz Molina, D.; Domingo, N.; Amabilino, D. B.; Vidal Gancedo, J.; Tejada, J.; Hendrickson, D. N.; Veciana, J. Monatsh. Chem. 2003 , 134 , 265. (230) Park, C. D.; Jung, D. Y. Bull. Korean Chem. Soc. 2001 , 22 , 611. (231) Zhao, H. H.; Berlinguette, C. P.; Bacsa, J.; Prosvirin, A. V.; Bera, J. K.; Tichy, S. E.; Schelter, E. J.; Dunbar, K. R. Inorg. Chem. 2004 , 43 , 1359. (232) Bian, G. Q.; Kuroda Sowa, T.; Nogami, T.; Sugimoto, K.; Maekawa, M.; Munakata, M.; Miyasaka, H.; Yamashita, M. Bull. Chem. Soc. Jpn. 2005 , 78 , 1032. (233) Cornia, A.; Fabretti, A. C.; Pacchioni, M.; Zobbi, L.; Bonacchi, D.; Caneschi, A.; Gatteschi, D.; Biagi, R.; Del Pennino, U.; De Renzi, V.; Gurevich, L.; Van der Zant, H. S. J. Angew. Chem., Int. Ed. 2003 , 42 , 1645.

PAGE 278

278 (234) Heer sche, H. B.; De Groot, Z.; Folk, J. A.; van der Zant, H. S. J.; Romeike, C.; Wegewijs, M. R.; Zobbi, L.; Barreca, D.; Tondello, E.; Cornia, A. Phys. Rev. Lett. 2006 , 96 , 206801. (235) Yang, H.; Jia, L.; Wang, Z.; Di Cicco, A.; Levy, D.; Keller, P. Macrom olecules 2011 , 44 , 159.

PAGE 279

279 BIOGRAPHICAL SKETCH Tu Ngoc Nguyen was born in Ha Noi, Viet Nam in 1984. He attended the University of Natural Sciences in Ho Chi Minh City and earned his Bachelor of Science degree in 2006 with valedictorian. During his last year o f undergraduate studies, he joined the research group of Dr. Ha Thuc Huy and performed synthesis and structural analysis of polylactide/clay nanocomposites. After graduat ion , he stayed in school as a teaching assistant and was then awarded a graduate fellow ship from National University of Singapore in 2007 to pursue a master degree. Under the supervision of Dr. Sim Wee Sun at the Department of Chemistry, he explored chiral modification of metal nanoparticle surfaces and was conferred the Master of Science de gree in 2009. In the fall of 2009, he received a PhD fellowship from the Vietnam Education Foundation (a fellowship granted by US Congress) and joined the group of Distinguished Professor George Christou at the Chemistry Department of University of Florida. Hi s doctoral research primarily focuses on exploring the chemistry and physics of supramolecular aggregates of single molecule magnets. He h as been the recipient of several awards and fellowships for research excellence , including the Grinter Award (2009 2011), Procto r & Gamble Award (2012) , Crow Award (2012), Eastman Chemical Summer Fellowship (2013) , and travel awards from American Chemical Society (2013) and the UF Graduate Student Council (2014).