UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Synthesis and characterization of oligometallic and polymetallic transition metal azolates Summers, David Alexander 1996

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1997-251713.pdf [ 26.97MB ]
Metadata
JSON: 831-1.0061691.json
JSON-LD: 831-1.0061691-ld.json
RDF/XML (Pretty): 831-1.0061691-rdf.xml
RDF/JSON: 831-1.0061691-rdf.json
Turtle: 831-1.0061691-turtle.txt
N-Triples: 831-1.0061691-rdf-ntriples.txt
Original Record: 831-1.0061691-source.json
Full Text
831-1.0061691-fulltext.txt
Citation
831-1.0061691.ris

Full Text

SYNTHESIS A N D C H A R A C T E R I Z A T I O N OF O L I G O M E T A L L I C A N D P O L Y M E T A L L I C TRANSITION M E T A L A Z O L A T E S by DAVID A L E X A N D E R SUMMERS B.Sc.H., Queen's University, 1992 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Chemistry) We accept this thesis as conforming to th£ required standard THE UNIVERSITY OF BRITISH C O L U M B I A August 1997 © David Alexander Summers, 1997 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library, shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ^ T f t Y The University of British Columbia Vancouver, Canada Date DE-6 (2/88) A B S T R A C T Several divalent transition metal azolate compounds have been prepared and characterized. Amongst these is a series of (substituted)pyrazolate transition metal polymers of general formulation [M(pz*)2]x ( M = N i (seven compounds), M = Mn (four compounds); and pz* = pyrazolate and C-substituted pyrazolates). Two manganese(II) polymeric materials with slightly different formulation, [Mn(4-Xpz) 2(4-XpzH)] x (X = CI, Br), were also prepared. Indirect evidence suggests that these materials consist of linear chains of metal(II) ions doubly bridged by pyrazolate ligands. Four of the nickel(II) complexes and all of the manganese(II) compounds are paramagnetic and their magnetic susceptibilities were measured over the 2 to 300 K range. In each case the materials exhibit antiferromagnetic exchange coupling with J values ranging from -14 to -16 cm"1 for the nickel compounds, -1.2 to -2.1 cm"1 for the [Mn(pz*)2]x materials, and a/value of -0.4 cm"1 for the two [Mn(4-Xpz) 2(4-XpzH)] x complexes. In addition to the polymeric complexes, a number of oligometallic transition metal (substituted)pyrazolate compounds were prepared and characterized. Six dimetallic nickel(II) complexes of the general formula [CpNi(pz*)]2 (where Cp is the cyclopentadienyl ion and pz* = C-substituted pyrazolates), and four trimetallic nickel(II) materials of formulation [CpNi(4-X-3,5-diMepz)2]2Ni (X = H , C H 3 , CI, and Br), were prepared and examined. Three of the dimetallic and two of the trimetallic compounds have been characterized by single crystal X-ray diffraction which revealed them to consist of nickel(II) ions doubly bridged by pyrazolate ligands and end-capped with Cp groups. A green, mixed valence (Cu(II)/Cu(I)), trimetallic copper compound of formulation [Cu3(3,5-F6diMepz)5] ii (3.5-F6diMepz = bis(trifluoromethyl)pyrazolate) was also prepared and examined. A single crystal X-ray diffraction study revealed that this complex consists of a triangular ring of copper ions bridged by the 3,5-F6diMepz groups. The two copper(II) ions in this complex are connected by three pyrazolate bridges and one pyrazolate-Cu(I)-pyrazolate bridge. Variable temperature magnetic measurements (2 to 300 K) revealed strong antiferromagnetic coupling between the two Cu(II) ions (J =-235 cm"1). A three dimensional iron(II) imidazole/imidazolate polymer was prepared and characterized. A single crystal X-ray diffraction study revealed that this material consists of linear chains of tetrahedral Fe(II) ions singly-bridged by imidazolate groups. The individual chains of tetrahedral iron centers are cross-linked to six parallel chains via octahedral iron(II) centers. Neutral imidazole groups occupy two (trans) coordination sites on the octahedral centers. Variable temperature magnetic measurements revealed antiferromagnetic exchange along the tetrahedral chains (J = -2.4 cm"1). A canted spin structure leads to weak ferromagnetism at temperatures below a magnetic phase transition temperature of 17 K. Two transition metal complexes containing triazolate, [M(trz) 2] x ( M = Cu(II), Mn(II)), were also prepared. Variable temperature magnetic studies revealed that both of these materials exhibit antiferromagnetic exchange. The copper complex exhibits long-range magnetic ordering and becomes a weak ferromagnet below a magnetic phase transition temperature of 30 K. This magnetic ordering is thought to be a result of a canted spin structure. iii T A B L E OF CONTENTS ABSTRACT ii LIST OF TABLES xvii LIST OF FIGURES xx LIST OF ABBREVIATIONS A N D S Y M B O L S . xxxi A C K N O W L E D G M E N T S xxxvi Chapter 1 INTRODUCTION 1 1.1 DIMENSIONALITY 2 1.2 M A G N E T I S M 3 1.2.1 M A G N E T I C E X C H A N G E 8 1.2.2 M A G N E T I C M O D E L I N G OF MAGNETIC E X C H A N G E INTERACTIONS 14 1.2.3 M O L E C U L A R M A G N E T S 16 1.3 AZOLES 18 1.4 AZOLES A N D AZOLATES AS LIGANDS IN TRANSITION M E T A L COMPOUNDS 20 1.5 OBJECTIVES A N D ORGANIZATION OF THIS THESIS 21 1.6 P H Y S I C A L METHODS OF CHARACTERIZATION 23 1.6.1 MAGNETIC PROPERTY M E A S U R M E N T S 23 I V 1.6.2 X - R A Y DIFFRACTION 25 1.6.3 SPECTROSCOPY 26 1.6.3.1 INFRARED 26 1.6.3.2 UV-VIS-NIR SPECTROSCOPY 26 1.6.3.3 N M R 27 1.6.4 OTHER METHODS 28 1.6.4.1 E L E M E N T A L ANALYSIS 28 1.6.4.2 T G A 28 1.6.4.3 DSC 29 1.6.4.4 MASS SPECTROMETRY 29 1.6.4.5 SCANNING E L E C T R O N MICROSCOPY 30 Chapter 2 OLIGOMETALLIC N I C K E L P Y R A Z O L A T E S 31 2.1 INTRODUCTION 31 2.2 RESULTS A N D DISCUSSION 41 2.2.1 [CpNi(4-X-3,5-DIMETHYLPYRAZOLATE)] 2 and [CpNi(3,5-BIS(TRIFLUOROMETHYL)PYRAZOLATE)] 2 41 2.2.1.1 SYNTHESIS, PHYSICAL AND T H E R M A L PROPERTIES 41 2.2.1.2 X - R A Y DIFFRACTION 48 2.2.1.3 SPECTROSCOPIC BEHAVIOUR 54 v 2.2.1.3.1 INFRARED SPECTROSCOPY 54 2.2.1.3.2 N M R SPECTROMETRY 55 2.2.1.4 M A G N E T I C PROPERTIES 59 2.2.2 [CpNi(4-X-3,5-DIMETHYLPYRAZOLATE) 2 ] 2 Ni 61 2.2.2.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPERTIES 61 2.2.2.2 X - R A Y DIFFRACTION STUDIES 67 2.2.2.3 SPECTROSCOPIC B E H A V I O U R 72 2.2.2.3.1 INFRARED SPECTROSCOPY 72 2.2.2.3.2 N M R SPECTROSCOPY 73 2.2.2.4 MAGNETIC B E H A V I O U R 75 2.3 S U M M A R Y A N D CONCLUSIONS 76 Chapter 3 POLY-(NICKEL(II) AZOLATES) \" 78 3.1 INTRODUCTION. 78 3.2 RESULTS A N D DISCUSSION 80 3.2.1 DIAMAGNETIC NICKEL(II) P O L Y M E R S 80 3.2.1.1 NICKEL(II) 4 -X-PYRAZOLATES (X = H , CI) 80 3.2.1.1.1 SYNTHESIS, STRUCTURE, A N D T H E R M A L A N D MAGNETIC PROPERTIES 80 3.2.1.1.2 X - R A Y DIFFRACTION STUDIES 92 vi 3.2.1.1.3 SPECTROSCOPIC B E H A V I O U R 93 3.2.1.1.3.1 INFRARED SPECTROSCOPY 93 3.2.1.1.3.2 ELECTRONIC SPECTROSCOPY 94 3.2.1.1.4 PROPOSED STRUCTURES 96 3.2.1.2 NICKEL(II) INDAZOLATE, 99 3.2.1.2.1 SYNTHESIS, PHYSICAL, A N D T H E R M A L A N D MAGNETIC PROPERTIES OF [Ni(indz)2]x 100 3.2.1.2.2 INFRARED SPECTROSCOPY 102 .2.2 P A R A M A G N E T I C NICKEL(II) P O L Y M E R S 106 3.2.2.1 NICKEL(II) 4-X-3 ,5-DIMETHYLPYRAZOLATES (X = H , C H 3 , Br, CI) 106 3.2.2.1.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPET1ES 106 3.2.2.1.2 X - R A Y DIFFRACTION STUDIES 114 3.2.2.1.3 SPECTROSCOPIC B E H A V I O U R 117 3.2.2.1.3.1 INFRARED SPECTROSCOPY 117 3.2.2.1.3.2 ELECTRONIC SPECTROSCOPY 120 3.2.2.1.4 M O L E C U L A R M O D E L I N G STUDIES A N D PROPOSED STRUCTURES 122 3.2.2.1.5 MAGNETIC PROPERTIES 131 vii 3.3 S U M M A R Y A N D CONCLUSIONS 140 Chapter 4 POLY-(MANGANESE(II) AZOLATES) 142 4.1 INTRODUCTION 142 4.2 RESULTS A N D DISCUSSION .143 4.2.1 MANGANESE(II) 4 -X-PYRAZOLATES (X = CI, Br) 143 4.2.1.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPERTIES 143 4.2.1.2 X - R A Y DIFFRACTION STUDIES 150 4.2.1.3 INFRARED SPECTROSCOPY 151 4.2.1.4 ELECTRONIC SPECTROSCOPY 152 4.2.1.5 MAGNETIC PROPERTIES A N D PROPOSED STRUCTURE 153 4.2.2 MANGANESE(II) 4-X-3 ,5-DIMETHYLPYRAZOLATES (X = H , C H 3 , Br, CI) 164 4.2.2.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPERTIES 164 4.2.2.2 INFRARED SPECTROSCOPY 171 4.2.2.3 PROPOSED STRUCTURES A N D M A G N E T I C PROPERTIES 173 4.3 MANGANESE(II)TRIAZOLATE 180 viii 4.3.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPERTIES 184 4.3.2 INFRARED SPECTROSCOPY 186 4.3.3 M A G N E T I C PROPERTIES A N D PROPOSED STRUCTURE 186 4.4 S U M M A R Y AND CONCLUSIONS 188 Chapter 5 A M I X E D V A L E N C E COPPER 3,5-BIS(TRIFLUOROMETHYL)PYRAZOLATE C O M P L E X 191 5.1 INTRODUCTION 191 5.2 RESULTS A N D DISCUSSION 192 5.2.1 [Cu3(3,5-F6diMepz)5] 192 5.2.1.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPERTIES 192 5.2.1.2 X - R A Y DIFFRACTION STUDIES 194 5.2.1.3 SPECTROSCOPIC STUDIES 198 5.2.1.3.1 INFRARED SPECTROSCOPY 198 5.2.1.3.2 ELECTRONIC SPECTROSCOPY 198 5.2.1.3.3 MASS SPECTROMETRY 199 5.2.1.4 MAGNETIC PROPERTIES 199 5.2.2 STUDIES O N THE PURPLE DECOMPOSITION PRODUCT 205 5.2.2.1 PREPARATION, PHYSICAL A N D T H E R M A L PROPERTIES 205 I X 5.2.2.2 SPECTROSCOPIC STUDIES 206 5.2.2.2.1 INFRARED SPECTROSCOPY 206 5.2.2.2.2 ELECTRONIC SPECTROSCOPY 207 5.2.2.2.3 MASS SPECTROMETRY 208 5.2.2.4 MAGNETIC PROPERTIES A N D PROPOSED STRUCTURAL FEATURES 209 5.3 S U M M A R Y A N D CONCLUSIONS 212 Chapter 6 A N IRON(II)IMIDAZOLE EVUDAZOLATO COMPOUND 214 6.1 INTRODUCTION 214 6.2 RESULTS A N D DISCUSSION 216 6.2.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPERTIES 216 6.2.2 X - R A Y DIFFRACTION STUDIES 219 6.2.3 SPECTROSCOPIC STUDIES 222 6.2.3.1 INFRARED SPECTROSCOPY 222 6.2.3.2 ELECTRONIC SPECTROSCOPY 223 6.2.4 M A G N E T I C BEHAVIOUR 224 6.2.4.1. PROTOCOL FOR OBTAINING M A G N E T I C D A T A FOR M O L E C U L A R M A G N E T S 235 6.3 S U M M A R Y A N D CONCLUSIONS 236 Chapter 7 BIS(l,2,4-TRIAZOLATO)COPPER(II) [Cu(trz)2]x 238 x 7.1 INTRODUCTION 23 8 7.2 RESULTS A N D DISCUSSION 238 7.2.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPERTIES 238 7.2.2 SPECTROSCOPIC STUDIES 242 7.2.2.1 INFRARED SPECTROSCOPY 242 7.2.2.2 ELECTRONIC SPECTROSCOPY 242 7.2.3 PROPOSED STRUCTURE A N D M A G N E T I C B E H A V I O U R 244 7.3 S U M M A R Y A N D CONCLUSIONS 254 Chapter 8 G E N E R A L S U M M A R Y AND SUGGESTION FOR FUTURE WORK 255 8.1 G E N E R A L S U M M A R Y 255 8.2 SUGGESTIONS FOR FUTURE WORK 261 Chapter 9 E X P E R I M E N T A L .262 9.1 INTRODUCTION 262 9.2 SYNTHESES 262 9.2.1 A Z O L E DERIVATIVES 263 9.2.1.1 4-CHLOROPYRAZOLE, (4-ClpzH) 263 9.2.1.2 4-BROMOPYRAZOLE, (4-BrpzH) 264 9.2.1.3 4-CHLORO-3,5-DIMETHYLPYRAZOLE, (4-Cl-3,5-diMepzH) 265 xi 9.2.1.4 4-BROMO-3,5-DIMETHYLPYRAZOLE, (4-Br-3,5-diMepzH) 265 9.2.1.5 3,4,5-TRIMETHYLPYRAZOLE, (4-CH3-3,5-diMepzH) 266 9.2.1.6 4-NITRO-3,5-DJMETHYLP Y R A Z O L E , (4-N0 2-3,5-diMepzH) 267 9.2.1.7 3,5-BIS(TRIFLUOROMETHYL)PYRAZOLE, (3,5-F 6diMepzH) 268 9.2.2 M E T A L L O C E N E S 268 9.2.2.1 BISCYCLOPENTADIENYLNICKEL(II ) (NICKELOCENE), (Ni(Cp)2) 269 9.2.2.2 B I S C Y C L O P E N T A D I E N Y L M A N G A N E S E (MANGANOCENE) , (Mn(Cp)2) 270 9.2.3 NICKEL(II) P Y R A Z O L A T E S 271 9.2.3.1 POLY-BIS((a-PYRAZOLATO-N,N')NICKEL(II), ([Ni(4-Hpz)2]x) 271 9.2.3.2 POLY-BIS(u-4-CHLOROPYRAZOLATO-N,N')NICKEL(II), ([Ni(4-Clpz)2]x) 272 9.2.3.3 POLY-BIS( |a-3,5-DIMETHYLPYRAZOLATO-N,N')NICKEL(II), ([Ni(4-H-3,5-diMepz)2]x) 272 xii 9.2.3.4 POLY-BIS(M-CHLORO-3 ,5 -D I M E T H Y L P Y R A Z O L A T O - N , N ' )NICKEL(II), ([Ni(4-Cl-3,5-diMepz)2]x) 273 9.2.3.5 POLY-BIS(M-BROMO-3 ,5 -D I M E T H Y L P Y R A Z O L A T O - N , N ' )NICKEL(II), ([Ni(4-Br-3,5-diMepz)2]x) 274 9.2.3.6 POLY-BIS(u.-3,4,5-TRTMETHYLPYRAZOLATO-N,N')NICKEL(II), ([Ni(4-CH3-3,5-diMepz)2]x)...... 275 9.2.3.7 POLY-BIS(u.-rNDAZOLATO-N,N')NICKEL(II), ([Ni(indz)2]x) 275 9.2.3.8 Di(r | -CYCLOPENTADIENYL-u-3,5-D I M E T H Y L P Y R A Z O L ATO-N,N ' -NICKEL(II)), ([CpNi(4-H-3,5-diMepz)]2) 276 9.2:3.9 Di(T]-CYCLOPENTADIENYL-p:-4-CHLORO-3,5-D I M E T H Y L P Y R A Z O L ATO-N,N ' -NICKEL(II)), ([CpNi(4-Cl-3,5-diMepz)]2) 277 9.2.3.10 Di(r)-C YCLOPENT ADIENYL-u,-4-BROMO-3,5 -D I M E T H Y L P Y R A Z O L ATO-N,N ' -NICKEL(II)), ([CpNi(4-Br-3,5-diMepz)]2) 277 xiii 9.2.3.11 Di(r | -CYCLOPENTADIENYL-u-3,4,5-TRTMETHYLP Y R A Z O L A T O - N , N ' -NICKEL(II)), ([CpNi(4-CH3-3,5-diMepz)]2) 278 9.2.3.12 Di(r |-CYCLOPENTADIENYL-u-4-NITRO-3,5-D I M E T H Y L P Y R A Z O L ATO-N,N ' -NICKEL(II)), ([CpNi(4-N02-3,5-diMepz)]2) 278 9.2.3.13 Di(ri-CYCLOPENTADIENYL-ix-3,5-BIS(TRJTLOUROMETHYL)P Y R A Z O L ATO-N,N ' -NICKEL(II)), ([CpNi(3,5-F6diMepz)]2) 279 9.2.3.14 [CpNi(4-Cl-3,5-diMepz)2]2Ni 280 9.2.3.15 [CpNi(4-H-3,5-diMepz)2]2Ni 281 9.2.3.16 [CpNi(4-Br-3,5-diMepz)2]2Ni 282 9.2.3.17 [CpNi(4-CH3-3,5-diMepz)2]2Ni 282 9.2.4 MANGANESE(II) AZOLATES 283 9.2.4.1 POLY-BIS(^-3 ,5-DIMETHYLPYRAZOLATO-N,N')MANGANESE(II) , ([Mn(4-H-3,5-diMepz)2]x) 283 9.2.4.2 POLY-BIS(ji-4-CHLORO-3,5-DJMETHYLPYRAZOLATO-N,N' )MANGANESE(I I ) , (|Mn(4-Cl-3,5-diMepz)2]x) 284 xiv 9.2.4.3 POLY-BIS(u-4-BROMO-3,5-DIMETHYLPYRAZOLATO-N,N ' )MANGANESE(I I ) , ([Mn(4-Br-3,5-diMepz)2]x) 285 9.2.4.4 POLY-BIS( |a-3,4,5-TRIMETHYLPYRAZOLATO-N,N')MANGANESE(II) , ([Mn(4-CH3-3,5-diMepz)2]x) 286 9.2.4.5 POLY-BIS(ix-4-CHLOROPYRAZOLATO-N,N')(4-CHLOROPYRAZOLE)MANGANESE(II) , ([Mn(4-Clpz)2(4-ClpzH)]x) 286 9.2.4.6 POLY-BIS(p:-4-BROMOPYRAZOLATO-N,N')(4-BROMOPYRAZOLE)MANGANESE(II ) , ([Mn(4-Brpz)2(4-BrpzH)]x) 287 9.2.4.7 POLY-BIS(u-TRIAZOLATO)MANGANESE(H), (|Mn(trz)2]x) 288 9.2.5 COPPER(II) AZOLATES 289 9.2.5.1 POLY-BIS(u.-TRIAZOLATOCOPPER(II), ([Cu(trz)2]x) 289 9.2.5.2 A M I X E D V A L E N C E COPPER(II)/COPPER(I) COMPOUND, (Cu3(3,5-F6diMepz)5) 290 9.3 P H Y S I C A L METHODS 291 9.3.1 M A G N E T I C M E A S U R E M E N T S A N D P A R A M E T E R CALCULATIONS 291 xv 9.3.2 SINGLE C R Y S T A L X - R A Y DIFFRACTION 295 9.3.3 POWDER X - R A Y DIFFRACTION 295 9.3.4 E L E M E N T A L ANALYSIS 296 9.3.5 INFRARED SPECTROSCOPY 296 9.3.6 ELECTRONIC SPECTROSCOPY 296 9.3.7 N M R SPECTROSCOPY 297 9.3.8 T G A 297 9.3.8 DSC 297 9.3.9 S E M 298 9.3.10 MASS SPECTROMETRY. : 298 REFERENCES 299 APPENDIX I SINGLE C R Y S T A L X - R A Y DIFFRACTION D A T A 318 APPENDIX II M A G N E T I C D A T A 345 APPENDIX III INFRARED D A T A 3 77 APPENDIX IV MASS SPECTRA 383 APPENDIX V POWDER X - R A Y DIFFRACTION D A T A 385 xvi LIST OF T A B L E S Number Page Table 1.1. List of abbreviations for the azole derivatives used in this thesis 19 Table 2.1. T G A results for the [CpNi(4-X-3,5-diRpz)]2 dimers 47 Table 2.2. DSC results for the [CpNi(4-X-3,5-diRpz)]2 materials 48 Table 2.3. Ni—Ni non-bonded distances in dimeric N i systems 53 Table 2.4. X H nmr data for dimeric systems 57 Table 2.5. T G A results for the [CpNi(4-X-3,5-diMepz)2]2Ni materials 66 Table 2.6. DSC results for the [CpNi(4-X-3,5-diMepz)2]2Ni materials 67 Table 2.7. Ni—Ni non-bonded distances in trimetallic nickel complexes 71 Table 2.8. TT nmr data for trimetallic complexes 74 Table 3.1. Theoretical molecular weights and % compositions for various chain lengths of Cp[Ni(4-Hpz)2]xNiCp, 89 Table 3.2. Theoretical molecular weights and % compositions for various chain lengths of Cp[Ni(4-Clpz)2]xNiCp, 90 Table 3.3. T G A results for the [Ni(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI and Br) and [Ni(4-Xpz) 2] x (X = H , CI) systems I l l Table 3.4. Positions of the in-plane vring band positions in the [Ni(4-X-3,5-diMepz) 2] x complexes and 4-X-3,5-diMepzH compounds 118 xvii Table 3.5. Comparison of the pc-cro band positions in [Ni(4-X-3,5-diMepz)2]xand 4-X-3,5-diMepzH 120 Table 3.6. Calculated magnetic parameters for the [Ni(4-X-3,5-diMepz)2]x polymers 135 Table 3.7. \j\ and \4JS2\ ranges for polymeric transition metal (Ni(II), Cu(II), Co(II)) C-substituted pyrazolates 139 Table 4.1. Magnetic parameters for the [Mn(4-Xpz)2(4-XpzH)]x (X = Br, CI) materials 159 Table 4.2. Comparison of the elemental compositions, molecular weights and calculated magnetic moments for [Mn(4-Xpz) 2(4-XpzH) n] x (X = CI, Br) for two values of n (1, 0.9) with the experimental values 161 Table 4.3. Comparison of the magnetic parameters for substituted-pyrazolate polymers of Mn(II), Ni(II), Cu(II) and Co(II) 163 Table 4.4. Elemental analyses of the [Mn(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI, and Br) complexes before and after brief exposure to the atmosphere 166 Table 4.5. T G A results for the [Mn(4-X-3,5-diMepz)2]x materials 167 Table 4.6. Magnetic parameters for the [Mn(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI, Br) materials 178 xviii Table 4.7. \4JS2\ values for polymeric manganese(II), nickel(II), cobalt(II) and copper(II) systems in which pyrazolate or substituted pyrazolate ligands bridge the metal centers. \4JS2\ values shown here have been calculated for data over the 2 to 300 K temperature range 179 Table 7.1. Selected magnetic parameters for canted-spin magnetic materials 252 Table 9.1. Diamagnetic corrections of ligands and metal ions 292 xix LIST OF FIGURES Number Page Figure 1.1. A schematic of the d-orbital energy diagrams for a d 8 system (a) as a free ion, and (b) in an octahedral crystal field. The diagram helps to illustrate how the orbital angular momentum can be quenched by the introduction of a crystal field 5 Figure 1.2. Zeeman splitting of an S = V2 doublet in an applied magnetic field 6 Figure 1.3. Spin orientations in a 2-D planar array resulting from various types of magnetic exchange interactions 10 Figure 1.4. Temperature dependence of the magnetic susceptibilities for (a) a paramagnetic material, (b) a ferromagnetic material and (c) an antiferromagnetic material 11 Figure 1.5. Representation of the two types of superexchange; (a) kinetic exchange, and (b) potential exchange 13 Figure 1.6. Atom numbering scheme for (a) pyrazole, (b), 1,2,4-triazole and (c) imidazole 19 Figure 2.1. (a) T G A and (b) DSC plots for [CpNi(4-H-3,5-diMepz)]2 43 Figure 2.2. (a) T G A and (b) DSC plots for [CpNi(4-Br-3,5-diMepz)]2 44 Figure 2.3. (a) T G A and (b) DSC plots for [CpNi(4-N02-3,5-diMepz)]2 45 xx Figure 2.4. (a) T G A and (b) DSC plots for [CpNi(4-H-3,5-F6diMepz)]2 46 Figure 2.5. The molecular structure of [CpNi(4-H-3,5-diMepz)]2, with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms 49 Figure 2.6. The molecular structure of [CpNi(4-N02-3,5-diMepz)]2, with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms 50 Figure 2.7. The molecular structure of [CpNi(3,5-F6diMepz)]2, with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms 51 Figure 2.8. Stereoscopic ORTEP diagram of [CpNi(4-H-3,5-diMepz)]2, with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms 52 Figure 2.9. 200 M H z *H nmr spectrum for [CpNi(4-H-3,5-diMepz)]2 in C 6 D 6 solution 56 Figure 2.10. Space-filling and "ball and stick" representations of a modeled planar Ni-(N-N) 2 -Ni ring system for [CpNi(4-H-3,5-diMepz)]2. There are considerable steric interactions between the methyl protons on the pyrazolate rings with the protons on the capping Cp groups 58 xxi Figure 2.11. Space-filling and "ball and stick" representations for the boat conformation of the Ni-(N-N)2-Ni ring system for [CpNi(4-H-3,5-is noticeably less steric interactions between the methyl protons on the pyrazolate rings with the protons on the capping Cp groups for this conformation 58 Figure 2.12. Powder magnetic susceptibility and magnetic moment plot for [CpNi(4-H-3,5-diMepz)]2 60 Figure 2.13. (a) T G A and (b) DSC plots for [CpNi(4-H-3,5-diMepz)2]2Ni 64 Figure 2.14. (a) T G A and (b) DSC plots for [CpNi(4-Cl-3,5-diMepz)2]2Ni 65 Figure 2.15. The molecular structure of one of the two independent molecules of [CpNi(4-H-3,5-diMepz)2]2Ni, with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms 68 Figure 2.16. The molecular structure of [CpNi(4-Cl-3,5-diMepz)2]2Ni, with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms 69 Figure 2.17. Stereoscopic ORTEP diagram of [CpNi(4-Cl-3,5-diMepz)2]2Ni, with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms 70 Figure 2.18. 200 M H z *H nmr spectrum of [CpNi(4-H-3,5-diMepz)2]2Ni in C 6 D 6 solution 74 xxii Figure 2.19. Powder magnetic susceptibility and magnetic moment plots for [CpNi(4-Cl-3,5-diMepz)2]2Ni 76 Figure 3.1. Schematic representation of the polymerization of metallocenes with azoles 82 Figure 3.2. Thermal decomposition of [Ni(4-Hpz)2]x by (a) T G A and (b) DSC 84 Figure 3.3. Thermal decomposition of [Ni(4-Clpz)2]x by (a) T G A and (b) DSC 85 Figure 3.4. S E M image of [Ni(4-Hpz)2]x. The white bar represents a length of 5 um.... 91 Figure 3.5. Powder X-ray diffractogram of [Ni(4-Hpz)2]x 93 Figure 3.6. Electronic spectra of (a) [Ni(4-Hpz)2]x, (b) [Ni(4-Clpz)2]x and (c) a lower mull concentration of [Ni(4-Clpz)2]x showing the bands at 250 and 325 nm 95 Figure 3.7. Possible trimetallic structure of [Ni(4-Hpz)2]x 96 Figure 3.8. Proposed linear chain structure, with square planar nickel centers, for the [Ni(4-Hpz)2]x material 98 Figure 3.9. The potential bridging ligand indazolate 99 Figure 3.10. Thermal decomposition of [Ni(indz)2]x studied by (a) T G A and (b) DSC 101 Figure 3.11. Four possible arrangements of the indazolate bridging groups for a [M(indz)2]x material 104 xxiii Figure 3.12. A linear chain structure in which some steric interactions similar to those in the [Ni(4-X-3,5-diMepz)2]x materials, in this case between the bridging indazolate groups, is present 105 Figure 3.13. Thermal decomposition of fNi(4-H-3,5-diMepz)2]x as studied by (a) T G A and (b) DSC 109 Figure 3.14. Thermal decomposition of [Ni(4-Cl-3,5-diMepz)2]x as studied by (a) T G A and (b) DSC 110 Figure 3.15. S E M images of (a) [Ni(4-H-3,5-diMepz)2]x, and (b) [Ni(4-CH3-3,5-diMepz)2]x. The white bar represents a length of 5 urn 113 Figure 3.16. Powder diffraction patterns for the [Ni(4-X-3,5-diMepz)2]x materials, where (a) X = H, (b) X = C H 3 , (c) X = CI, and (d) X = Br 115 Figure 3.17. Powder diffraction patterns for the [M(4-H-3,5-diMepz)2]x materials, where (a) M = N i (this work), (b) M = Co (199), (c) M = Zn(199), and (d )M = Cu(193) 116 Figure 3.18. Electronic spectra of (a) [Ni(4-Cl-3,5-diMepz)2]x, (b) [Ni(4-H-3,5-diMepz)2]x, (c) [Ni(4-Br-3,5-diMepz)2]x, (d) [Ni(4-CH3-3,5-diMepz) 2] x and (e) a more concentrated mull of [Ni(4-H-3,5-diMepz) 2] x 121 Figure 3.19. Crystal structure of [CpNi(4-H-3,5-diMepz)2]2Ni showing the square planar geometry about the central nickel(II) ion 123 xxiv Figure 3.20. Modeled polymer segment for [Ni(4-H-3,5-diMepz)2]x in which the metal centers are present in a square planar geometry 124 Figure 3.21. Crystal structure of [Co(4-H-3,5-diMepz)2Cl(4-H-3,5-diMepzH)]2Co (113) showing the tetrahedral geometry about the central metal ion 125 Figure 3.22. Modeled polymer segment for [Ni(4-H-3,5-diMepz)2]x in which the metal centers are present in a tetrahedral geometry 126 Figure 3.23. Pictorial representations showing the factors influencing Ni—Ni separations in doubly pyrazolate bridged nickel systems for (a) a tetrahedral metal geometry and (b) a square planar metal geometry 128 Figure 3.24. Steric interaction resulting from the elongation of a linear chain of square planar nickel(II) ions doubly bridged by 4-H-3,5-diMepz ligands. The arrows labeled "a" represent the elongation along the chain and the arrows labeled "b" represent the resulting steric interaction between methyl groups and ring moieties 130 Figure 3.25. Powder magnetic susceptibility plots for the [Ni(4-X-3,5-diMepz)2]x materials: (a) X = H, (b) X = C H 3 , (c) X = Br, and (d) X = CI. The circles are the experimental data and the calculated values are represented by the curved lines. The insert plots show the data fitted to the low temperature range (10 - 100 K for (a), (c), and (d) and 30 -xxv 100 K for (b)). Parameters for the calculated susceptibilities are those in Table 3.6 133 Figure 4.1. Thermal decomposition of [Mn(4-Clpz)2(4-ClpzH)]x as measured by (a) T G A and (b) DSC 145 Figure 4.2. S E M images of [Mn(4-Clpz)2(4-ClpzH)]2. The white bar represents a length of (a) 5 urn, (b) 50 urn, and (c) 5 urn 149 Figure 4.3. Powder diffraction pattern for [Mn(4-Clpz)2(4-ClpzH)]x 151 Figure 4.4. Possible linear chain structure of [Mn(4-Xpz) 2(4-XpzH)] x (X = CI, Br). In this case the manganese(II) ions are five coordinate and present in a square pyramidal chromophore geometry 154 Figure 4.5. Linear chain structure of [Mn(4-Xpz)2(4-XpzH)]x in which the five-coordinate manganese(II) centers are present in a trigonal bipyramidal M n N 5 arrangement 155 Figure 4.6. Linear chain structure for [Mn(4-Xpz)2(4-XpzH)]x (X = CI, Br). In this case there are alternating octahedral and square planar manganese(II) centers 156 Figure 4.7. Linear chain structure for [Mn(4-Xpz)2(4-XpzH)]x (X = CI, Br). In this case there are alternating octahedral and tetrahedral manganese(II) centers 156 xxvi Figure 4.8. Magnetic susceptibility versus temperature plots for (a) [Mn(4-Clpz)2(4-ClpzH)]x and (b) Mn(4-Brpz)2(4-BrpzH)]x. Experimental points are shown as circles and theoretical fits (calculated as described in the text) are represented as solid lines 158 Figure 4.9. Thermal analysis of [Mn(4-CH3-3,5-diMepz)2]x as monitored by (a) T G A and (b) DSC 168 Figure 4.10. Thermal analysis of [Mn(4-Cl-3,5-diMepz)2]x as monitored by (a) T G A and (b) DSC 169 Figure 4.11. Powder magnetic susceptibility plots for the [Mn(4-X-3,5-diMepz) 2] x materials: (a) X = H , (b) X = C H 3 , (c) X = Br, and (d) X = CI. The circles are the experimental data and the calculated values (as described in the text) are represented by the curved lines. Parameters for the calculated susceptibilities are those in Table 4.5 177 Figure 4.12. Three different possibilities for a triazolate ion bridging metal centers. (A) is analogous to pyrazolate bridging, (B) is similar to imidazolate bridging, and (C) represents the case in which both types of bridging occur simultaneously 180 Figure 4.13. Magnetic susceptibility and moment plots for [Mn(trz)2]x 187 Figure 5.1. The molecular structure of Cu3(3,5-F6diMepz)5, with atom numbering scheme 195 xxvii Figure 5.2. The molecular structure of Cu3(3,5-F6diMepz)5, with 50% probability thermal ellipsoids shown for all non-hydrogen atoms ...196 Figure 5.3. Magnetization plots for Cu3(F6dmpz)5 at 300 K (A), 100 K ( • ) and 50 K (O). The lines have been calculated using the data above 10 000 G and were used to determine the magnetization intercept at zero applied field ....200 Figure 5.4. Magnetic susceptibilities (per mole of copper(II)) versus temperature plot. • , A, and O, are data points for run #1 at 10 000 Gauss, run #2 at 10 000 Gauss and one run at 20 000 Gauss, respectively. The solid line is calculated from theory as described in the text 202 Figure 5.5. Variable temperature magnetic susceptibility and magnetic moment plots for the purple decomposition product of [Cu3(3,5-F6diMepz)5]. The values have been calculated per mole of Cu(II), assuming that two thirds of the copper ions are in the 2+ oxidation state, the molecular weight is 1206 gmol"1 210 Figure 6.1. Thermal decomposition of [Fe3(imid)6(imidH)2]x as studied by (a) T G A and (b) DSC 218 Figure 6.2. The molecular structure of the repeat unit of [Fe3(imid)6(imidH)2]x with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms 220 xxviii Figure 6.3. Stereo view of the unit cell of [Fe3(imid)6(imidH)2]x. For clarity all hydrogen atoms, carbons 4 and 5 of imid and all atoms except the coordinating nitrogen of imidH have been removed 221 Figure 6.4. UV-Vis-NIR spectrum for [Fe3(imid)6(imidH)2]x. (The sharp peak at 2950 nm results from the V N - H IR band (-3390 cm"1) and the sharp peaks at about 2500 nm arise from the mulling agent, Nujol) 224 Figure 6.5. Plot of the magnetic susceptibility (O) and xT ( • ) versus temperature at 500 G for a powdered sample of [Fe3(imid)6(imidH)2]x 226 Figure 6.6. Plot of magnetization versus applied magnetic field for [Fe3(imid)6(imidH)2]x at 300 K (red O), 100 K ( • ) , 50 K (blue A), 20 K (V), 13 K (O), 4.8 K (green O), and 2 K (yellow A) 227 Figure 6.7. Field dependence of [Fe3(imid)6(imidH)2]x at 4.8 K 228 Figure 6.8. Plots of %™ versus temperature at 500 G. Magnetic contribution from subtracted out as described in text for v = 0 ( • ) and v = +10 (O). The lines are calculated from theory as described in the text 231 Figure 7.1. T G A and DSC plots for [Cu(trz)2]x 240 Figure 7.2. UV-Vis-NIR spectrum of a concentrated mull (a) and a dilute mull (b) of[Cu(trz) 2] x .,...243 xxix Figure 7.3. Magnetic susceptibility and magnetic moment versus temperature data at 500 Gauss for [Cu(trz)2]x 245 Figure 7.4. Plot of magnetization versus applied magnetic field for [Cu(trz)2]x at 300 K (red O), 100 K (green • ) , 50 K (A), 25 K (V), 13 K (O), 4.8 K (green O), and 2 K (yellow • ) 246 Figure 7.5. Field dependence of [Cu(trz)2]x at 4.8 K 248 Figure 7.6. Possible 3-dimensional structure of [Cu(trz)2]x in which each copper(II) ion is six coordinate 249 Figure 7.7. Structure of [Cu(trz)2]x proposed by Inoue and Kubo (298) to account for two different proton environments in this material 251 Figure 7.8. Two potential alignments for canting of magnetic orbitals leading to (a) a small net magnetization and (b) a large net magnetization 253 Figure 9.1. Arbitrary plot illustrating the significance of the standard error values of the calculated magnetic parameters. The circles represent experimental data, the dashed line represents the theoretical values generated by the computed magnetic parameters, and the two solid lines represent other equally "good" fits generated with the extreme parameter values (parameters ± standard error) 294 xxx LIST OF A B B R E V I A T I O N S A N D S Y M B O L S % V X M-•n a (3 ± o 5 v /?, B . M . or u,B °C o rC M.eff 0 A H Xcalc percent principle quantum number magnetic susceptibility bridging hapticity angle out-of-plane angle plus or minus degree(s) chemical shift vibrational or electronic band Bohr magneton degree(s) Celsius effective magnetic moment diffraction angle change in enthalpy calculated magnetic susceptibility xxxi X M molar magnetic susceptibility u,m micrometer(s) Hs.o. spin-only magnetic moment approximately < less than > greater than H applied magnetic field sh shoulder 0-D zero-dimensional (a point) 1-D one-dimensional (a line) 2-D two-dimensional (a sheet) 3-D three-dimensional (a cube) acac acetylacetone Anal. analysis A Angstrom(s) br broad calcd. calculated cm centimeter(s) cos cosine DSC differential scanning calorimetry xxxii E M U electromagnetic unit(s) F goodness-of-fit parameter fw formula weight g Lande splitting factor G Gauss g gram(s) imid imidazolate indz indazolate IR infrared J magnetic exchange coupling constant J Joule(s) k Boltzmann's constant K Kelvin L orbital angular momentum I orbital angular momentum quantum number lit. literature M molar m medium m t magnetic quantum number m/e mass over charge ratio xxxiii methyl milligram(s) megahertz milliliter(s) millimole(s) mole(s) melting point magnetic saturation molecular weight Avogadro's number not applicable near infrared nanometer(s) nuclear magnetic resonance Oakridge thermal ellipsoid plot relative proportion of paramagnetic impurity phenyl part(s) per million pyrazolate room temperature xxxiv s singlet or strong s solid S or S total spin S E M scanning electron microscopy SQUID superconducting quantum interference device T temperature Tc magnetic long-range ordering temperature T G A thermogravimetric analysis THF tetrahydrofuran TIP temperature independent paramagnetism trz triazolate U V ultraviolet Vis visible V S M vibrating sample magnetometer vw very weak w weak W Watt(s) cm"1 wavenumber(s) xxxv A C K N O W L E D G M E N T S I would like to thank sincerely Drs. A. Storr and R.C. Thompson for their tremendous support and assistance during the past five years. I would also like to thank the members of my guidance committee, Drs. J. Trotter, F. Aubke and F. G. Herring, for their helpful suggestions throughout this work. Next I would like to express my thanks to Drs. S. J. Rettig and J. Trotter of this department for their rapid crystal structure determinations, and to Dr. V. G. Young at the University of Minnesota for a crystal diffraction study. I would like to thank Mr. P. Borda of this department for microanalysis determinations. I also extend my gratitude to the experts in the electronics, mechanical and glassblowing shops for their assistance. I am grateful to Drs. M . Ehlert and T. Otieno for their insight and advice at the beginning of this work, and to Mr. S. Xia for many useful discussions. I would like to thank my friends, fellow members of various chemistry sporting teams, attendees of the group trips to professional hockey and basketball games, and all the participants of the numerous sports pools, for making the past several years all the more enjoyable. I wish to. extend special thanks to my parents for their everlasting support and understanding, and to my little brother, Kenjinator, for countless hours of fun. Finally, I thank my wife, Mia, for her love, support, contagious sense of humour, and for making the past few years truly wonderful. xxxvi Chapter 1 INTRODUCTION This thesis consists of work potentially interesting to chemists, physicists, material scientists and technologists. The molecular materials described herein have been prepared with an anticipation of correlating their physical properties to their molecular structures. One particular aim was to synthesize complexes exhibiting interesting and, possibly, useful magnetic properties. The target compounds chosen for these studies are oligometallic and polymeric transition metal azolates. The role of the chemist in magnetism, a field most often associated with physicists, has increased dramatically over the past few decades. This increased interest is due, in part, to recent discoveries that certain molecular materials exhibit potentially useful magnetic properties. Among the potential applications for these molecular materials are information storage and components of display devices (1-4). This chapter has been designed to introduce the reader to the concepts of dimensionality in matter and its pertinence to magnetic properties, fundamental magnetism and theories of magnetic exchange, the role that azoles and azolates play in transition metal chemistry and to discuss the organization of this thesis and the methods employed in physical characterization. 1 1.1 DIMENSIONALITY Dimensionality, in the context of this dissertation, refers to the extent that a material is chemically bonded in physical space. Four different possibilities of dimensionality exist: zero-dimensional (0-D, a point), one-dimensional (1-D, a line), two dimensional (2-D, a sheet), and three-dimensional (3-D, a solid block or sphere). In chemistry the dimensionality is best represented by four known types of carbon. Methane, like most compounds studied by chemists, has only local connectivity and is an example of a 0-D material. Fligh density polyethylene, which is chemically bonded along an extended chain is a good example of a 1-D material, while graphite, which has an extended sheet structure, and diamond, which has chemical connectivity in three dimensions, are termed 2-D and 3-D materials respectively. It is important to note that regardless of the dimensionality of a material it usually exists physically as a three-dimensional aggregate held together by the relatively weak van der Waals interactions. It is, however, still valid to regard these compounds in terms of their lower-dimensionality as the strength of the chemical bonds far exceeds the strength of the forces holding separate molecular species together. It is also important to note that materials of higher dimensionality (1-D, 2-D and 3-D) often exhibit properties associated with the lower-dimensional building blocks of that material (points in chains, points or chains in sheets, and points, chains or sheets in cubes). 2 1.2 M A G N E T I S M Magnetism and its usefulness to chemists is an area of ever increasing interest and many detailed theoretical descriptions of the magnetic behaviour of molecular materials (magnetochemistry) have been published (5-9). Magnetism is a property that, for most materials, exists only in the presence of an external magnetic field. In the presence of such a field, two distinct types of magnetism can be induced. Diamagnetism, which results from the presence of paired electrons in a magnetic field gradient, will cause a substance to tend towards a low field region. Paramagnetism, on the other hand, typically results from unpaired electrons in an applied field and forces substances to be attracted to a high field region. Almost any molecular material containing both paired and unpaired electrons will exhibit net paramagnetism as the effect of paramagnetism is much stronger than that of diamagnetism. Of primary interest here is paramagnetism which, more precisely, results from the angular momentum of unpaired electrons. This angular momentum can come from two sources: orbital angular momentum, L , and a property known as electron spin, S. The concept of electron spin was developed by Uhlenbach and Goudsmit in 1926 out of a need for a fourth quantum number (in addition to n, £, and m t ) in order to account for details of the emission spectra of atoms (10). Their spectroscopy experiments showed 3 every electron has a magnetic moment which has two possible orientations when an atom is placed in a magnetic field. The term 'spin' was chosen for this phenomenon as initially it was thought to be a result of the electron spinning, which in classical mechanics would produce a magnetic moment. It should be noted that although electrons possess this quantized angular momentum, the physical picture of a spinning sphere is not accurate. Electron spin is purely quantum mechanical in nature with magnitude Vfft and possesses the quantum number ± 5 . As stated earlier, paramagnetism is a result of the angular momentum of unpaired electrons in compounds. However, many paramagnetic transition metal complexes possess electronic configurations such that the orbital component to this angular momentum is partially or fully quenched. As an example of this orbital quenching, consider the case, illustrated in Figure 1.1, for a d 8 system as both a free ion and in an octahedral crystal field. A d 8 free ion has a 3 F ground state with a total orbital angular momentum, L , of 3. By introducing an octahedral crystal field, the ground state term becomes 3 A 2 and the degeneracy is removed. In such complexes the paramagnetism observed is due primarily to the spin angular momentum. 4 A A A A (a) Free ion case, L = 3 (b) Octahedral crystal field case, L = 0 Figure 1.1. A schematic of the d-orbital energy diagrams for a d 8 system (a) as a free ion, and (b) in an octahedral crystal field. The diagram helps to illustrate how the orbital angular momentum can be quenched by the introduction of a crystal field. In order to simplify the development of the models used to describe and predict magnetic behaviour, the models are often based solely on spin properties and spin-spin interactions. It is of course important to realize that blindly interpreting the magnetic properties of complexes based on these simplified models can lead to erroneous conclusions, especially if the unpaired electrons in the materials do happen to possess a significant orbital angular momentum. Additionally, the presence of electron orbital angular momentum in materials being modeled can result in poorer agreement between the experimental data and theoretical values. Magnetism is a bulk property and is observed or measured as induced magnetism. Induced magnetism arises from a population difference in 2 or more energy states, in which individual magnetic dipoles adopt different quantized orientations in an applied magnetic 5 field. As an example, consider the S = Vi state, composed of two quantum levels, ms = ± Vi. As illustrated in Figure 1.2, in the presence of an applied magnetic field this doublet splits (Zeeman splitting) into the -Vi and +V2 components. . + 1/2 S = 1/2 - 1/2 H = 0 H Figure 1.2. Zeeman splitting of an S = V2 doublet in an applied magnetic field. The population difference between the -Vi and +V2 energy levels is then governed by the Boltzmann distribution. The slight excess of magnetic dipoles aligning with the applied magnetic field results in the induced magnetism. A convenient way of expressing this magnetic induction has been developed. Magnetic susceptibility, %, is related to the ratio of the induced magnetic field to the applied magnetic field. The value for molar magnetic susceptibility, % M , for the Zeeman splitting of the S = V2 state can be expressed as, 4kT 5CM A L-T ' [1-1] 6 where NA is Avogadro's number, g is a proportionality constant (equal to -2.0 for an electron with no orbital angular momentum), P is the Bohr magneton of the electron, and k is the Boltzmann constant. A more general expression for S > XA is: (S(S + 1)) [1.2] Another common quantity used to describe magnetic materials is the effective magnetic moment, \xes, which is defined as: [1.3] For systems with no orbital angular momentum this value is termed the spin-only magnetic moment, (is.o., and can be expressed as: ixs0 =2[S(S + 1)]X B . M . [1.4] It is important to note that systems in which there is an orbital contribution to the angular momentum often possess effective magnetic moments above the spin-only value. Among the paramagnetic materials described in this thesis, many possess ground state electron configurations expected to have such an orbital contribution. Complexes of tetrahedral nickel(II), and high spin octahedral iron(II) have orbitally degenerate ground states and first order orbital contributions to the magnetic moment. The complexes of these 7 X M • n • 3kT ions have measured room temperature magnetic moments well above the corresponding spin-only values. 1.2.1 M A G N E T I C E X C H A N G E Magnetic exchange is a phenomenon in which the electron spins on one atom influence the electron spins on a neighbouring atom. This interaction can be either intermolecular or intramolecular in nature depending on the structure of the material in question. A 2-dimensional schematic representation of the various types of magnetic exchange described below, along with a representation of a system in which no exchange is occurring (paramagnetism, Figure 1.3a) is shown in Figure 1.3(a-e). There are two main types of interactions; one in which the spins tend to line up in a parallel sense, termed ferromagnetism (Figure 1.3b), and one in which the spins tend to line up in an antiparallel alignment, termed antiferromagnetism (Figure 1.3 c). From these two main forms of exchange many possible net results can be predicted. In systems in which the electrons line up ferromagnetically a net magnetic moment, greater in magnitude than the corresponding moment for an identical system in which no exchange is permitted, is observed. On the other hand, an antiferromagnetic system will possess a net magnetic moment below that of a 8 corresponding non-exchanging (magnetically dilute) material. In fact, for the limiting case of a complete antiferromagnetic coupling, all of the spins are lined up in an antiparallel fashion and the system behaves like a diamagnetic (spin-paired) material. Other possibilities are also conceivable. In a system in which the unpaired electrons of adjacent atoms line up an antiparallel but canted manner as shown in Figure 1.3e, a net moment of magnitude between the ferromagnetic and antiferromagnetic extremes is predicted. This phenomenon is known as canted-spin antiferromagnetism (Figure 1.3e). In addition, if interacting neighbour atoms contain differing numbers of unpaired spins the net result of both ferromagnetic and antiferromagnetic exchange coupling will be net ferromagnetism. This type of exchange is termed ferrimagnetism (Figure 1.3d). As mentioned, interacting paramagnetic centers can give rise to several different types of magnetic behaviour. The temperature dependence of the magnetic moments and magnetic susceptibilities of these exchange coupled systems are characteristic of the type of exchange occurring. To a first approximation and in the absence of an orbital angular momentum contribution, the magnetic moment of a paramagnetic material is independent of temperature. This leads to a magnetic susceptibility that is inversely related to temperature. This behaviour is shown in a graphical representation in Figure 1.4(a). In systems in which there is a magnetic exchange interaction between the paramagnetic ions, there is a competition between the thermal randomization of the individual dipoles and the ordering due to the exchange. In a ferromagnetic material this competition leads to a magnetic 9 AT ^ 4 7 ^ 4 -U (a) paramagnetism 4—J-A f-f-t-i—JL—A 7 ¥ U l 7 f / (b) ferromagnetism 7 ^ 7^ ^ 4 "7 (d) ferrimagnetism l ' 7 -f-—/ 7 ^ \ V 71 if • h -it r—f—f—f—f—f (c) antiferromagnetism IT 71 V 71 V 4, / ^ / H. > V (e) canted spin antiferromagnetism ST 1 Figure 1.3. Spin orientations in a 2-D planar array resulting from various types of magnetic exchange interactions. 10 Temperature Figure 1.4. Temperature dependence of the magnetic susceptibilities for (a) a paramagnetic material, (b) a ferromagnetic material and (c) an antiferromagnetic material. moment that increases as the temperature is reduced since removing thermal energy shifts this competition to the ferromagnetic state. Therefore a ferromagnetically coupled system will have a magnetic susceptibility that increases (as the temperature is reduced) more 1 1 rapidly than a corresponding paramagnetic material (Figure 1.4(b)). Conversely, an antiferromagnetically coupled system has a magnetic moment that decreases with decreasing temperature. The temperature dependence of the magnetic susceptibility is very distinctive in this case as there is a temperature at which a maximum value for the magnetic susceptibility is obtained (Figure 1.4(c)). This maximum occurs at the temperature at which the strength of the antiferromagnetic exchange becomes more significant than the thermal randomization of the magnetic dipoles. There have been several proposed mechanisms that account for exchange interactions between paramagnetic centers including dipolar coupling, direct exchange and superexchange. A correlation of paramagnetic centers may be rationalized in terms of the local magnetic fields associated with spin dipoles influencing neighbouring spin dipoles. This coupling phenomenon is termed dipolar coupling, and the effects are proportional to the square of the magnetic moment and inversely proportional to the cube of the spin-spin separation. Direct exchange involves a direct overlap of orbitals containing unpaired electrons on neighboring centers. This effect is therefore only applicable in systems in which the paramagnetic centers are in close (less than 3 A) proximity. In the paramagnetic materials described in this thesis, the magnetic centers are separated by diamagnetic organic ligands and are not in close enough proximity for the above effects to make a substantial contribution to any exchange interactions. However, there is an exchange pathway that is capable of spanning these relatively large paramagnetic 12 center separations. Superexchange involves the favourable overlap of occupied orbitals of bridging ligands with the magnetic orbitals on metal centers. Two general possibilities are available depending on the orientation of the bridging ligand orbital. In the case when a non-orthogonal orbital pathway spans adjacent magnetic centers the interaction is said to be kinetic exchange, and antiferromagnetic coupling is observed. This exchange is illustrated graphically in Figure 1.5(a). Conversely, if there is orthogonality in the orbital pathway the interaction is termed potential exchange, and ferromagnetic coupling is obtained. Potential exchange is represented in Figure 1.5(b). Figure 1.5. Representation of the two types of superexchange; (a) kinetic exchange, and (b) potential exchange. By simple inspection of Figure 1.5 it is apparent that the molecular structure and composition of paramagnetic compounds can greatly influence the magnetic properties. It is thus the challenge of correlating the molecular structures of materials with their observed magnetic properties that motivates magnetochemists. This motivation is very evident by the 13 vast number of reports devoted to the magnetic properties of transition metal complexes, and several in depth reviews on these studies have been published (11-15). 1.2.2 M A G N E T I C M O D E L I N G OF MAGNETIC E X C H A N G E INTERACTIONS In real systems the spin vectors on individual paramagnetic centers can orient in three dimensions, giving rise to S X , S Y , and S Z Cartesian components. Several models have been developed that account for a number of the possible correlations between magnetically coupled spin vector components. The Heisenberg model predicts the isotropic case in which all three of the spin vector components couple equally. Other models assume anisotropy of the spin vector coupling. The Ising model predicts the case in which only the S Z spin vector components of neighbouring ions couple. A third situation in which only the S X and S Y components couple is described by the X - Y model. In all cases the magnetic exchange is represented by the coupling constant, J. The sign of J indicates the nature of the exchange while the strength of the coupling is represented by the magnitude of J. According to the definition used in this thesis, where H = -2JSi'S2 for an antiferromagnetically coupled system, antiferromagnetic systems have negative J values and ferromagnetically coupled systems have positive J values. Another important factor influencing real magnetic systems is the nature of the exchange. Firstly, and fortunately for the model developers, exchange interactions between 14 paramagnetic centers are proportional to f , where r is the interspin separation (12). Therefore the most important magnetic interactions occur between nearest neighbours. A second factor that greatly influences the magnetic behaviour is the dimensionality of the exchange. Systems in which magnetic ions couple in one dimension (along a chain) have very different properties from systems in which the magnetic coupling occurs in two or three dimensions. Most of the conceivable spin exchange interactions in real systems have not been modeled as the mathematics involved in accurately predicting the behaviour is too complex, or no analytic solution exists. There have, however, been several expressions developed to predict magnetic exchange coupling in one dimensional systems. Detailed explanation of the development of these models is beyond the scope of this thesis. Instead, references to the published reports on any model used are given at the first occurrence of each such model. It is important at this point to consider some of the limitations of the models used in this work. Again due to the complexity of many real systems the models do not account for all the factors influencing the magnetic behaviour. For example, there have been no models developed that can simultaneously predict the temperature dependence of both the orbital angular momentum, a single ion effect, and the magnetic exchange coupling. For this reason the results of modeling the magnetic data for systems in which an orbital angular momentum contribution is present tend to be less accurate than similar results for fully 15 orbitally quenched systems. Secondly, the models assume that only nearest neighbours interact and thus these models are not suited to systems exhibiting long range ordering (described later). Another real system factor that is not accounted for in the developed models is spin canting. As shown above, if the system being modeled has magnetic orbitals that align in a non-parallel manner, the lowest energy state of the system has a corresponding magnetic moment somewhere between the ferromagnetic and antiferromagnetic extremes. This may again detrimentally influence the agreement between experimental data and theoretical values. Finally, in order to model meaningfully, and interpret the magnetic data for a real system, it is useful to have at least some structural information. This perhaps is the largest hurdle in many of the molecular complexes akin to the ones discussed in this thesis. Due to the physical properties of many polymeric materials, it is often extremely difficult to isolate them in a form suitable for structure determination. In addition, some of the typical indirect methods of structurally characterizing molecular compounds cannot be used due to these same physical properties. 1.2.3 M O L E C U L A R M A G N E T S The distinguishing feature of a magnet is the presence of long-range, three dimensional magnetic ordering. In any assembly of a material with a magnetic ground state 16 there is a finite temperature, Tc, at which long-range magnetic ordering is expected. However, for most molecular materials this temperature is very low, often ~ 10"2 K (5). Molecular materials in which ferromagnetic, ferrimagnetic, or canted spin antiferromagnetic exchange is present often have a relatively high Tc. In theory, at temperatures below T0 spontaneous magnetization should occur. As the temperature is decreased (below Tc) this magnetization should increase to a saturation level, Ms = NA&gS, where Ms is the saturation magnetization and S is the spin associated with the molecular ground state (5). In practice however, the magnetization requires the presence, at least initially, of an external magnetic field, and magnetic saturation is only achieved in the presence of a moderate applied field. It should be noted that even paramagnetic materials, or ferromagnetic materials above Tc, will achieve magnetic saturation in the presence of sufficiently high applied magnetic fields. Two properties of many molecular-based ferromagnets (and ferrimagnets and canted spin magnets) are remnance and hysteresis. Remnance, or remnant magnetization, is the field induced magnetization that remains after the external field has been removed. In order to remove this remnant magnetization a coercive field in the opposite direction must be applied. It is the combination of remnant magnetization and coercive field that determine the boundaries of the hysteresis loop. In the past 10 years there have been several of reports on molecular based ferromagnets (16-26). In particular there are the decamethylmetallocenium charge transfer 17 salts, one of which is [Fe(Cp*)2][TCNE], where Cp* is the pentamethylcyclopentadienyl ion and TCNE is tetracyanoethylene (19-21). This material has a Tc value of 4.8 K and exhibits hysteresis behaviour below this temperature. Subsequently, related manganese materials were reported with Tc values of 6.2 K (for [Mn(Cp*)2][TCNQ], where TCNQ is 7,7,8,8-tetracyano-/?-quinodimethanide) (22) and 8.8 K (for [Mn(Cp*)2][TCNE]) (23). In 1991, Manriquez et al. reported the preparation of a vanadium complex, V(TCNE) 2- 1/ 2(CH 2C1 2), for which the Tc was over 35OK! In this case the exchange is thought to be ferrimagnetic resulting from 3-D antiferromagnetic exchange (27). There have also been recent reports on the long range 3-D ordering of ferrimagnetic chain compounds (28-32). Net magnetism arising from the antiferromagnetic exchange coupling of structurally canted molecular-based systems has also received some recent attention (5, 32-40). 1.3 AZOLES Azoles are aromatic, five-membered, nitrogen containing heterocycles. The azoles of interest in this work are pyrazole, 1,2,4-triazole, imidazole and some 3,4,5-substituted pyrazoles. The atom numbering schemes for these azoles are shown in Figure 1.6 and a list of the abbreviations used throughout this thesis is given in Table 1.1. 18 H H H H (a) (b) (c) Figure 1.6. Atom numbering scheme for (a) pyrazole, (b), 1,2,4-triazole and (c) imidazole. Table 1.1. List of abbreviations for the azole derivatives used in this thesis. Azole3 Abbreviation11 pyrazole 4-HpzH 4-chloropyrazole 4-ClpzH 4-bromopyrazole 4-BrpzH 3,5-dimethypyrazole 4-H-3,5-diMepzH 4-chloro-3,5-dimethylpyrazole 4-Cl-3,5-diMepzH 4-bromo-3,5-dimethylpyrazole 4-Br-3,5-diMepzH 4-nitro-3,5-dimethylpyrazole 4-N02-3,5-diMepzH 3,4,5-trimethylpyrazole 4-CH3-3,5-diMepzH 3,5-bis(trifluoromethyl)pyrazole 3,5-F6diMepzH indazole indzH imidazole imidH 1,2,4-triazole trzH a the abbreviation pz* is sometimes used to represent several different pyrazolate derivatives b In the abbreviations for the corresponding azolates the terminal "HP is omitted. 19 1.4 AZOLES A N D AZOLATES AS LIGANDS IN TRANSITION M E T A L COMPOUNDS Azoles have long been known to coordinate to transition metal centers. Due, in part, to the vast amount of work published on transition metal azolyl complexes, the reviews of recent or relevant reports on these materials are presented in three separate introductory sections, later in this thesis. A review of transition metal pyrazolate complexes appears in Chapter 2, Section 2.1., while the reviews of metal triazole and metal imidazolate compounds appear in Chapter 4, Section 4.3, and in Chapter 6, Section 6.1 respectively. There are several conceivable ways in which these azoles and azolates can bind to metal centers. The most common of these includes coordination as a neutral ligand through the N(2) (pyrazole), N(3) (imidazole), or N(2) and/or N(4) (triazole) nitrogen atoms. Another very common binding mode, and the one primarily observed in the compounds described in this work, involves the deprotonated form of the azole acting as an exobidentate bridging ligand. 20 1.5 OBJECTIVES A N D ORGANIZATION OF THIS THESIS There have been many previous studies on the magnetic properties of transition metal complexes. The present work has, in part, been performed to extend these studies to include several new paramagnetic 1-D complexes. Among the materials prepared were a series of nickel(II) and a series of manganese(II) polymeric materials in which pyrazolate and substituted pyrazolate ions bridge adjacent metal centers. A goal of this work was to correlate the observed magnetic properties of these complexes with their compositions and molecular structures. Several 0-D nickel(II) complexes incorporating the same or similar pyrazolate bridging groups as the 1-D nickel(II) complexes were also prepared. These oligometallic materials, although themselves not magnetically interesting, provided insight into the structures of the less easily characterized polymeric analogues. A second goal of this study was to prepare magnetically interesting complexes of higher dimensionality by utilizing closely related, but structurally different, bridging ions. The ligand precursors chosen for this study were imidazole and triazole, which, due to the arrangement of the donor nitrogen atoms in the ring systems, are expected to generate 2-D or 3-D polymeric systems. Complexes of iron(II) with imidazolate, copper(II) with triazolate and manganese(II) with triazolate have been prepared and characterized. Lastly, during the course of this work a mixed valence 0-D copper(II)/copper(I) pyrazolate compound was prepared and characterized. The magnetic behaviour of this complex was also investigated. 21 This thesis is divided into nine chapters and an Appendix. This chapter is intended to introduce the reader to the concepts important in this work and to provide some general background information in regards to the physical methods of characterization used herein. Chapter 2 details the preparation, physical properties and structural characterization of several dimetallic and trimetallic nickel(II) complexes. These materials are composed of nickel centers bridged by substituted-pyrazolate ions and end-capped with cyclopentadienyl groups. Chapter 3 describes the preparation and characterization of a series of nickel(II) (substituted)pyrazolate linear chain polymers. The chapter is divided into two general portions; the first, in which the complexes are diamagnetic and the second, in which the materials are paramagnetic. In Chapter 4, the synthesis and characterization of a series of manganese(II) substituted-pyrazolate polymers is reported. This chapter has been divided into three main portions. The first portion is devoted to chains of five coordinate manganese(II) ions in which one equivalent (per Mn(II)) of coordinated neutral substituted-pyrazole is present. The second portion describes four manganese(II) polymers analogous to the nickel(II) 4-substituted-3,5-dimethylpyrazolate polymers of Chapter 3. In the final portion the synthesis and characterization of a manganese(II) triazolate complex is described. Chapter 5 details the preparation and characterization of a mixed valence copper(II)/copper(I) trimetallic ring complex. Some magnetic studies of a decomposition product of this material are also given. A 3-D iron(II) imidazolate polymer is the focus of Chapter 6. Chapter 7 describes the low temperature magnetic behaviour and some characterization of a polymeric copper(II) triazolate compound. A general summary of this 22 work, along with some suggestions for related future studies, is presented in Chapter 8. Finally, Chapter 9 presents the experimental details of the syntheses and the physical methods used in characterizations. 1.6 PHYSICAL METHODS OF CHARACTERIZATION All of the physical methods of characterization used in this work have been previously described, by others, in comprehensive detail. This section is presented to describe briefly the principles of each technique, and to explain the usefulness to the current work. References to more in-depth descriptions for the methods used are given in the appropriate sections. With the exception of the elemental analyses and the mass spectral and single crystal X-ray diffraction studies, all of the physical measurements described in this thesis were performed by the author. Acknowledgments regarding the other three methods will be made in the appropriate sections. 1.6.1 M A G N E T I C PROPERTY M E A S U R M E N T S Measuring the magnetic properties of materials is typically accomplished in one of two general ways. The first general technique involves measuring the force exerted on a 23 material that has been placed in an applied magnetic field gradient. In the presence of an external magnetic field gradient a paramagnetic material experiences a force towards a higher field region. In typical experimental setups this leads to an apparent increase in weight. Conversely, a diamagnetic material experiences an apparent reduction in weight. With knowledge of the magnitude of the applied magnetic field gradient, the volume occupied by the sample and the resulting relative apparent weight change, it is possible to calculate the volume magnetic susceptibility of the material. This is the fundamental basis of the Gouy method, which was used to determine the room temperature magnetic susceptibilities of some of the compounds described in this work. The second general technique in measuring magnetic properties is an induction method. The induction method works by measuring an induced change of voltage in a detection coil, in which a magnetic material has been placed. This is the detection method utilized by a superconducting quantum interference device (SQUID magnetometer) and a vibrating sample magnetometer (VSM). All of the variable temperature magnetic studies described in this work were performed on a SQUID magnetometer. In order to obtain magnetic measurements over a large temperature range (between 2 K and 300 K) the system utilizes liquid helium and the temperature of the sample area is controlled with cryostats. These techniques and others are described in detail in Gerloch's text (41). 24 1.6.2 X - R A Y DIFFRACTION Two types of X-ray diffraction studies were utilized in this work; single crystal diffraction and powder diffraction. Single crystal X-ray diffraction is probably the most desired means of obtaining solid state structural information on molecular compounds. However, as the name implies, single crystal X-ray diffraction requires macroscopic single crystals. In the current work, all of the 0-D materials were obtained as single crystals. However, only one of the polymeric complexes described herein was prepared in a form suitable for single crystal X-ray diffraction studies, and therefore this technique could not be used for the majority of the polymeric materials. Single crystal X-ray diffraction utilizes an X-ray beam that is directed at a crystal. Depending on the orientation of the crystal relative to the beam, the crystal will diffract the X-rays at fixed angles. A detector is used to measure the intensities of the diffracted beam and the structure is subsequently determined by mathematical treatment of the data. A powder is often comprised of a vast number of microcrystals which, like a macroscopic single crystal, diffract X-ray beams. However, the diffraction pattern obtained from a powder sample provides much less information than a corresponding single crystal pattern. In this work, powder diffraction patterns are used as fingerprints to compare 25 various related compounds. More in-depth descriptions of X-ray diffraction techniques are readily available (42, 43). 1.6.3 SPECTROSCOPY 1.6.3.1 INFRARED Infrared (IR) spectroscopy is used to measure specific molecular vibrations within materials. This technique is useful for confirming the presence or absence of neutral azoles in the compounds isolated in this work. IR spectra can also provide some limited information on the nature of the bridging ligands present in these complexes. Fairly detailed explanations of the principles of IR, the usefulness of the technique to inorganic chemists, and a review of some complexes exhibiting interesting IR behaviour can be found in Nakamoto's text (44). 1.6.3.2 UV-VIS-NTR SPECTROSCOPY UV-Vis-NTR (electronic spectroscopy) uses electromagnetic radiation to excite a material out of its electronic ground state and into an excited state. Depending on the system, several absorptions may occur which provide information on the relative energies of 26 the various electronic states. In the case of transition metal complexes these relative energies can provide quantitative information about the strength of the ligand field (45) and, more importantly in this work, provide evidence for specific metal chromophore geometries in complexes. A comprehensive treatment of the electronic spectroscopy of transition metal compounds is presented in Lever's text (46). 1.6.3.3 N M R Nuclear magnetic resonance (nmr) is normally a powerful spectroscopic tool for determining molecular structures in solution. It must be noted however, that unlike X-ray crystallography, it cannot normally provide any detailed information regarding bond angles or distances. In the current work nmr had a very limited role. Even if the polymeric complexes of interest in this work were soluble in a suitable nmr solvent (which they are not), the fact that most of them are paramagnetic makes this technique essentially irrelevant for these materials. In nmr the local magnetic fields about magnetic nuclei (ones with non-zero nuclear spin) influence the frequency of the nmr transitions. These local fields, and thus the transition frequencies, are greatly affected by the presence of unpaired electrons, and in many cases expected transitions are broadened out of observable existence. In this study nmr was primarily used to confirm the purity of the ligands. It was also used to investigate the solution structures of the oligometallic nickel complexes. (47, 48) 27 1.6.4 OTHER METHODS 1.6.4.1 E L E M E N T A L ANALYSIS Elemental analysis is an essential tool used to confirm the purity and composition of many of the materials described in this dissertation. It provides the relative percentages of analyzed elements from which an empirical formula can be determined. The technique normally does not provide any structural information, but in some cases in this dissertation, it has been used to estimate the length of some oligometallic chains. (49) 1.6.4.2 T G A Thermogravimetric analysis (TGA) can provide information about the thermal decomposition of a material. The sample platform is a sensitive analytical balance and the weight of a given sample is monitored as a function of increasing temperature. In some instances T G A results can provide support for the presence or absence of certain components. In other instances it may be possible, for example, to determine how strongly a particular component is interacting with the rest of the molecule by measuring the temperature at which that component is liberated. (50) 28 1.6.4.3 DSC Differential scanning calorimetry (DSC), like TGA, can provide information about the thermal decomposition of a material. Unlike T G A it can also be used to determine the temperature at which a material melts or freezes, exhibits a phase transition, or any other event in which thermal energy is either released or absorbed. The technique essentially measures the difference in the amount of energy required to heat two identical cells at the same rate. One of the cells contains the sample being analyzed while the second empty cell is the reference. During an occurrence of a thermal event the compound either releases or absorbs energy and thus a temperature differential occurs between the two cells. The amount of energy needed to compensate for this temperature differential is measured, thus providing thermodynamic information. For the most part the compounds studied in the current work decompose before melting and thus only decomposition transition events were observed. (50) 1.6.4.4 MASS SPECTROMETRY Mass spectrometry can be used to determine the molecular formulae of complexes. Additional information regarding certain components, or fragments, of the complexes can 29 also be obtained. In mass spectrometry, a material is volatilized and ionized. The ionization of the material often results in partial fragmentation into smaller charged species. The resulting ions are accelerated across a voltage potential into a mass analyzer. The mass analyzer separates the fragments based on their mass-to-charge ratios (m/z) and the relative number of each ion is plotted as a function of m/z. (51) 1.6.4.5 SCANNING E L E C T R O N MICROSCOPY Scanning electron microscopy (SEM) can be used to obtain information about the morphology and size distributions of particles in powdered materials. Images are obtained by first scanning a focused electron beam across a sample. This electron beam causes the emission of a secondary electron from the surface, which is in turn collected and focused by electromagnetic lenses resulting in the generation of a topographical map with a maximum resolution down to 100 A. (50) 30 Chapter 2 OLIGOMETALLIC NICKEL PYRAZOLATES 2.1 INTRODUCTION Pyrazole and pyrazolate containing transition metal coordination complexes have been extensively investigated and several review papers have been devoted to the topic. Trofimenko has contributed two such papers (52, 53) and more recently a review article by La Monica and Ardizzoia (54) has been written on the topic. In addition, a review of aromatic nitrogen heterocycles as bridging ligands covers some recent developments in pyrazolate transition metal chemistry (55). The first published report of a transition metal pyrazolate complex came in 1889 when Biichner described the preparation of silver(I) pyrazolate, [Ag(4-Hpz)]x (56). Interestingly, the structure of this material has only recently been determined (57), and has been shown to consist of infinite linear chains of silver(I) centers, singly-bridged by pyrazolate ions. Another crystalline phase of silver(I) pyrazolate, the trimetallic [Ag(4-Hpz)] 3, has also recently been prepared and characterized (54). Shortly after Buchner's initial publication, he reported the preparation of a series of [Ag(4-Xpz)]x (X = Br, I, N 0 2 ) compounds (58). The copper(I) analogue of the polymeric [Ag(4-Xpz)]x was first reported 31 in 1972 (52) and, depending on the synthetic method used, appears in two distinct crystalline phases (57). Other singly-bridged transition metal pyrazolate compounds that have been reported in the literature include [Ag(4-CH3pz)]x (59), [M(pz*)]x ( M = Ag, Cu; pz* = 4-CH 3 CH 2 pz, 4-Ipz, 4-Brpz, 4-N0 2pz, and 3-CH3pz) (60), and [Cu(3-CH3pz)]x (61). The above materials are all reported as polymeric compounds. However, some recent X -ray crystallographic studies have shown other complexes that were previously believed to be linear chain polymers are in fact trimetallic, [M(pz*)]3, complexes. Specifically, the reported copper(I) polymeric [Cu(4-H-3,5-diMepz)]x material (62) has now been shown to be the trimetallic [Cu(4-H-3,5-diMepz)]3 (63). Several other singly-bridged trimetallic materials have also appeared in the literature. The gold(I) complexes, [Au(pz*)]3, where pz* can be a variety of substituted pyrazolate groups, have been reported (64). The derivative in which pz* is 3,5-F6diMepz is appreciably volatile (65) and its X-ray structure has been determined (66). This compound consists of linearly-coordinated gold(I) centers linked by 3,5-F6diMepz groups to form a nine-membered ring. Similar compounds possessing the same structural motif include [Cu(3,5-diPhpz)]3 (Ph = phenyl) (67), [Ag(3,5-diPhPz)]3 (52), and [Au(3,5-diPhpz)]3 (68). In addition to the trimetallic [Au(3,5-diPhpz)]3 complex, the hexamer [Au(3,5-diPhpz)]6 has also been crystallographically characterized (68). Other trimeric copper(I) complexes reported include [Cu(4-CH3-3,5-diMepz)]3 (69), [Cu(4-N0 2-3,5-diMepz)]3 (70, 71), [Ag(4-N02-3,5-diMepz)]3, [Cu(3,5-di(pClPh)pz)]3, and [Cu(3,5-di(pCH3Ph)pz)]3 (54). 32 After Biichner's initial contribution, the next publication on transition metal pyrazolates did not appear for 35 years. In 1925 Fischer reported that precipitates formed upon adding 3,5-dimethylpyrazole to solutions of cobalt, iron or zinc (72). This discovery led to Heim's work, published in 1930 (73), in which he reported the use of 3,5-dimethylpyrazole in determinations of cobalt from basic aqueous solutions. The use of 3,5-dimethylpyrazole as a gravimetric determination reagent was more thoroughly examined in the 1950s by Pflaum and Dieter (74). The authors concluded that the purple precipitate formed in these reactions had the formulation [Co(4-H-3,5-diMepz)2]x, and suggested, based on the insolubility of the complex, that the material possessed a sandwich structure similar to cobaltocene. No substantiating evidence for this proposed structure was given and, although the structure has still not been definitively determined, a sandwich type structure seems very unlikely. The area of transition metal pyrazolate compounds remained dormant for the next 20 years until Seel and Sperber (75) reported the synthesis of Fe(II) pyrazolate, [Fe(4-Hpz) 2] x and Fe(III) pyrazolate, [Fe(4-Hpz)3]x. The materials were prepared by reactions of pyrazole with a variety of iron derivatives such as Fe(CO) 5 or [CpFe(CO) 2] 2 in benzene or toluene solvent. No analytical or spectroscopic data were reported for these materials and no subsequent report on these compounds by the authors has been published to date. The interest in binary transition metal pyrazolates increased considerably in the early 1970s with a renewed interest in cobalt(II) pyrazolates. Bagley, Nicholls, and Warburton 33 reported the preparation of [Co(4-H-3,5-diMepz)2]x from Co(II) acetylacetonate and hydrazine in isopropanol (76). The authors also described the preparation of [Co(4-Hpz) 2] x, [Co(3-CH 3pz) 2] x, and [Co(4-H-3,5-diMepz)2]x from cobalt(II) salts and the appropriate pyrazole in basic solutions. In addition, infrared and electronic spectroscopic studies, along with room temperature magnetic moment measurements, were performed on these compounds. The results of these studies led the authors to propose that the compounds possess polymeric structures in which the pyrazolate ligands act as bridging groups and the cobalt(II) centers are tetrahedrally coordinated. A few years later Trofimenko reported, as unpublished results, the preparation of nickel(II) and copper(II) pyrazolate by pyrolysis of the appropriate M(4-Hpz) 4(OAc) 2 (OAc = acetate) complex or by reacting pyrazole with metal salts in basic solutions (52). Trofimenko also reported the preparation of other copper(II) complexes incorporating 4-bromopyrazolate, 3,4-dibromopyrazolate, 3,4,5-tribromopyrazolate, 4-methylpyrazolate, 4-isopropylpyrazolate, and 4-cyanopyrazolate. In addition, the [Cu(3,4,5-tribromopyrazolate)2]x material was shown to exist as a tetramer in benzene solution (52). In 1973, along with their report of the copper(I) complex, [Cu(4-H-3,5-diMepz)]x, Singh, Satpathy and Sahoo described the syntheses and characterizations of [Co(4-H-3,5-diMepz)2]x and [Ni(4-H-3,5-diMepz) 2-2H 20] x (77). These materials were characterized by infrared and electronic spectroscopy, elemental analysis, and, for the Ni(II) and Co(II) materials, by variable temperature magnetic susceptibility data. 34 A series of reports by Vos and Groeneveld was published between 1977 and 1979 on divalent and trivalent metal pyrazolates (59, 78-81). In these publications, the authors describe the synthesis, and characterization of 54 metal pyrazolate complexes. The complexes described consist of Mn, Co, Ni , Cu, Zn and Cd metal ions with 4-Xpz (X = H, C H 3 , CI, Br, I, N O 2 ) and 4-H-3,5-diMepz anions. The materials were prepared by reacting aqueous solutions of metal salts with basic aqueous solutions of pyrazoles. The resulting compounds were characterized by infrared, Raman and electronic spectroscopy, thermogravimetric studies and elemental analyses. However, the purities of the complexes prepared by Vos and Groeneveld are suspect. Firstly, for the majority of the materials described, only metal analysis results were reported. In these cases roughly only one quarter of the materials have values presented that are within 0.3% (absolute deviation) of the theoretical values. Secondly, in the instances in which the analytical results for carbon, hydrogen and nitrogen are reported along with the metal analyses (13 compounds), only two complexes have results that fall within the acceptable ranges. As described in more detail later in this chapter, in 1977 Blake et al. prepared some nickel(II) pyrazolates by reacting nickelocene with pyrazoles in benzene solutions (82). In 1982, Strahle et al. reported the synthesis of [Fe(4-Hpz)2]x from F e C l 2 and lithium pyrazolate in toluene (83). Concurrently, Seelig described the results of extended Hiickel calculations on polymeric chains of iron(III) ions triply-bridged by pyrazolate ligands (84). 35 Between 1989 and 1993 Ehlert et al. reported on the synthesis and characterization of a series of polymeric copper(II), [Cu(4-X-3,5-diRpz)2]x (X = H , C H 3 , Br, CI; R = H , CH 3 ) (85-87), and cobalt(II), [Co(4-Hpz)2]x, [Co(3-CH 3pz) 2] x, and [Co(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI, Br) (88), substituted-pyrazolate complexes. Several of these polymers were prepared by reacting copper or cobalt metal with the appropriate molten pyrazole under an atmosphere of dioxygen gas. The [Co(4-Hpz)2]x complex was synthesized from CoCl 2 and Hpz in aqueous NaOH under ambient conditions. Three of the [Cu(4-X-3,5-diMepz) 2] x materials (X = CI, Br, CH 3 ) were prepared by reacting the corresponding trimetallic copper(I) complex, [Cu(4-X-3,5-diMepz)]3, with the appropriate molten pyrazole under a dioxygen atmosphere. Crystal structures of [Cu(4-Clpz)2]x (green form) and [Cu(4-CH 3pz) 2] x confirmed that the molecules consist of linear chains of metal ions doubly-bridged by pyrazolate ligands. Magnetic studies on these systems revealed strong antiferromagnetic coupling between metal centers in the chains. Polymeric complexes of formulation [Cu 2(OH)(4-Xpz) 3(4-XpzH) 2] x (X = CI or Br) and [Cu 2(OH)(4-Xpz) 3] x (X = H, C H 3 , CI, Br) have been recently reported as products of reactions between Cu(OH) 2 and molten 4-XpzH (89). The authors propose structures in which the copper(II) centers are alternately linked by a pair of 4-Xpz bridges and a combination of O H and 4-Xpz groups resulting in linear chains. The X-ray crystal structure of a linear trimetallic copper(II) complex, [Cu(4-H-3,5-dcmpz)2]3 (where dcmpz = dicarbomethoxypyrazolate), has been reported (54). Each 36 copper center is roughly square planar with some of the oxygen atoms on the dmcpz ligands occupying the remaining coordination sites on the external copper(II) ions. Another method of producing binary metal pyrazolates, [M(pz*)2]x ( M = Zn, Pt, Pd), involves removing thermally the neutral coordinated pz*H from complexes of formulation [M(pz*)2(pz*H)x]2. For example, heating [Zn(3,5-diMepz)2(3,5-diMepzH)]2 at 240°C causes the loss of 3,5-diMepzH and the formation of the complex [Zn(3,5-diMepz)2]x (90). A mixture of polymeric [Pt(4-Hpz)2]x and trimeric [Pt(4-Hpz)2]3 was prepared upon decomposition of the dimeric complex [Pt(4-Hpz)2(4-HpzH)2]2 (91, 92). Similar reactions have been reported for the formation of the polymeric [Pd(4-Hpz)2]x and trimeric [Pd(4-H-3,5-diMepz)2]3 by heating [Pd(4-Hpz)2(4-HpzH)2]2 and [Pd(4-H-3,5-diMepz)2(4-H-3,5-diMepzH) 2] 2 respectively (93). In 1995, Vecchio-Sadus reported on the electrochemical synthesis and magnetic studies of a series of transition metal pyrazolates, [M(4-Xpz) 2] x ( M = Ni(II), Cu(II), Zn(II); X = H , CI, I), [M(4-X-pz) 3] x (M = Co(III); X = H , CI, I), [M(4-X-3,5-diMepz)2]x (M = Co(II), Zn(II); X = H , Br), [Ni(4-Br-3,5-diMepz)2]x, and [Cu(4-X-3,5-diMepz)]x (X = H , Br) (94). In addition to the reports on homoleptic pyrazolate complexes, there has been a large number of publications on complexes containing a M(u,-pz*)nM core with additional neutral or charged ligands. These metal pyrazolates are most often dimeric. However, there 37 is a growing number of reports on homo-trimetallic, M(u,-pz )2M(p.-pz ) 2 M , and hetero-trimetallic, M(u.-pz*)2M'(u-pz*)2M, compounds. Recent reports on doubly-pyrazolate bridged homo-dimetallic systems include publications on zinc compounds (90, 95), several copper systems (96-102), nickel, palladium and platinum complexes (103-112), a cobalt dimer (113), rhodium (114-120) and iridium (121-126) compounds, and a few ruthenium complexes (127, 128). The six-membered M ( N - N ) 2 M ring in these materials is most often found in a boat-like conformation. However, there are examples of both planar rings (96, 112) and rings in a chair conformation (99). There are also several reports on hetero-bimetallic complexes in which a pair of pyrazolate bridges link two different metal centers (129-133). Reports on linear trimetallic systems with double-pyrazolate-bridges include some nickel and cobalt materials (111, 112, 134). Considerably less common than the doubly-bridged complexes described above are materials in which more than two pyrazolate groups link metal centers. Reports on triply-bridged systems include a nickel dimer (135), a manganese dimer (136) and a hetero-bimetallic (iridium, rhodium) complex (137). A rhodium compound in which four pyrazolate bridges link the metal centers has also been reported (138). Another related class of complexes is one in which, in addition to the pyrazolate bridges, other bridging groups are also present. A series of dimetallic nickel, palladium and platinum complexes in which the metal centers are linked by both pyrazolate links and bridging O H groups has been reported (103-109). A series of platinum complexes with 38 mixed pyrazolate and chloride bridges has also been reported (139). Several authors have published reports on homo-dimetallic complexes in which pyrazolates and other dinucleating ligands bridge metal (copper, palladium and nickel) centers (140-149). Other homo-dimetallic systems with mixed bridges include an osmium compound (150), and a number of rhodium (117, 126, 151-158), iridium (156-160), and ruthenium (127, 161-165) complexes. There have also been reports on hetero-dimetallic (Ru-M, where M = Pd or Rh) complexes (133, 166, 167), and hetero-trimetallic (M-Pd-M, where M = Rh or Ir) complexes (167-170) that incorporate mixed bridges. Larger clusters in which pyrazolate and other bridging groups are present include a tetrameric copper complex (171), a hexacopper compound (172), an octacopper material (173), and an octamolybdenum complex (174). Others in the field have concentrated more on multi-metallic complexes in which dinucleating functionalized pyrazolates bridge the metal centers. The bridging ligands used most often have the additional donor groups linked to either the 3 position, or to the 3 and 5 positions on the pyrazolate ring. Many homo-dimetallic and hetero-dimetallic complexes of this sort have recently been reported (175-190). On the route to producing the desired polymeric nickel materials described in Chapter 3, several oligomeric nickel azolates were isolated and characterized. Most of these compounds, although themselves not interesting magnetically, were of particular interest as a basis of comparison with the polymeric materials. A series of oligometallic 39 nickel(II) 4-substituted-3,5-dimethylpyrazolate compounds end-capped with cyclopentadienyl groups, and two 3,5-bis(trifluoromethyl)pyrazolate analogues, was prepared and examined. These short chain species were more amenable to crystallization than the polymeric nickel pyrazolates, and through X-ray crystallographic studies, some insight into the molecular structures of this class of material was obtained. Blake et al. have previously reported nickel complexes incorporating bridging 3,5-dimethylpyrazolate ligands with end-capping cyclopentadienyl groups. No X-ray determined structures were reported in this early work in which nickelocene was used as a precursor to form a dimetallic nickel pyrazolate complex (82). In the current work the synthesis of that compound, [CpNi(4-H-3,5-diMepz)]2, has been reproduced and its X-ray structure has been determined. In addition, reactions of nickelocene with this and other 3,5-disubstituted pyrazoles have been found to give a series of dimetallic and trimetallic products containing bridging pyrazolate ligands and end-capping cyclopentadienyl groups. Definitive X-ray structural studies of these compounds are presented and the effect on structure and properties of substitution on the bridging pyrazolate ligands is discussed. 40 2.2 RESULTS A N D DISCUSSION The reactions of nickelocene with pyrazoles under mild conditions has provided not only a route to the dimeric complexes of formulation [CpNi(4-X-3,5-diMepz)]2 as reported earlier by Blake et al. (82) but also to trimetallic species of formulation [CpNi(4-X-3,5-diMepz) 2]2Ni by altering the reaction conditions. Both types of complexes can be isolated as crystalline materials and several have been subjected to X-ray crystallographic study. The single crystal X-ray diffraction experiments reported in this chapter were all performed by SJ . Rettig of this Department. 2.2.1 [CpNi(4-X-3,5-DIMETHYLPYRAZOLATE)] 2 and [CpNi(3,5-BIS(TRIFLUOROMETHYL)PYRAZOLATE)] 2 2.2.1.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPERTIES Complete synthetic details are given in Chapter 9, Sections 9.2.3.8 through 9.2.3.13. In each case, nickelocene and the appropriate azole were combined in equimolar amounts and the resulting mixture was dissolved in dry and dioxygen-free benzene. The initial colour of the solutions was green due to the presence of dissolved nickelocene. This green 41 solution would typically turn red after approximately 20 minutes. In some cases a red powder would precipitate out of the solution within a few hours. Isolation of the products was achieved by removing the benzene solvent by flash evaporation followed by removal of any remaining nickelocene and azole by vacuum sublimation. The compounds were purified by first redissolving the solid residue in a minimum amount of dry and dioxygen-free benzene, followed by slow evaporation of the solvent under a dinitrogen atmosphere. The red crystalline products were subsequently collected by filtration and dried under vacuum for 12 hours. The syntheses involve an acid-base reaction in which the pyrazole N H proton is transferred to a Cp ring (from the nickelocene). The resulting pyrazolate binds to the nickel liberating a cyclopentadiene molecule in the process. The equation for this reaction is shown below, 2(NiCp 2 ) + 2(pz'H) ) [ C p N i ( p z ' ) ] 2 + 2 ( C p H ) [2.1] where pz' is 4-X-3,5-diMepz (X = H , .GH 3 , CI, Br, N 0 2 ) or 3,5-F6diMepz, and Cp is the cyclopentadienyl ion. The products are red, air-stable, crystalline solids. The complexes are soluble in a variety of polar and non-polar organic solvents, but decompose if exposed to moisture when in solution. 42 A u — o c 60 i o -0.5 300 400 500 Temperature (C) 800 Figure 2.1. (a) T G A and (b) DSC plots for [CpNi(4-H-3,5-diMepz)]2. T G A and DSC studies show that the complexes are thermally robust, withstanding temperatures above 150°C without any evidence for decomposition. The T G A and DSC data are presented in Tables 2.1 and 2.2 respectively. Four representative T G A and DSC plots are shown in Figures 2.1 through 2.4. 43 Figure 2.2. (a) T G A and (b) DSC plots for [CpNi(4-Br-3,5-diMepz)]2. The T G A and DSC results provide evidence for the thermal stabilities of the dimeric complexes. With the exception of the X = N 0 2 derivative, which rapidly volatilizes at 280°C, each of the materials loses approximately 80% of its weight upon being heated to 800°C. Interestingly the ~ 20% - 30% of weight that remains is very close to the weight of 44 12 200 300 400 500 600 700 800 Temperature (C) Figure 2.3. (a) TGA and (b) DSC plots for [CpNi(4-N02-3,5-diMepz)]2. the nickel component in each of these compounds. The theoretical weight percentages due to the nickel component are: 27% for the X = H derivative; 25% for the Me derivative; 23% for the CI derivative; 20% for the Br derivative and 18% for the 3,5-F6diMepz complex. 45 Figure 2.4. (a) T G A and (b) DSC plots for [CpNi(4-H-3,5-F6diMepz)]2. The DSC results indicate that these materials decompose without melting at the temperatures indicated (event 1) in Table 2.2. For the most part, the events observed in the DSC experiments occur at temperatures at which a thermal weight loss step is observed in the TGA. For the compounds of formulation [CpNi(4-X-3,5-diMepz)]2 (X = H, C H 3 , CI, and Br) there are typically three DSC events and three steps in the thermal weight loss. 46 Table 2.1. T G A results for the [CpNi(4-X-3,5-diRpz)]2 dimers. Ligand Step 1 Step 2 Step 3 Step 4 substituents X R Temp range (°C) Weight left (%) Temp range (°C) Weight left (%) Temp range (°C) Weight left (%) Temp range (°C) Weight left (%) H C H 3 212 -267 78 267 - 312 65 312 - 570 17 570 - 800 22 C H 3 C H 3 188 -213 87 213 -277 86 277 -460 29 - -CI C H 3 180 -220 86 220 -285 85 285 - 821 20 - -Br C H 3 185 -224 73 224 -443 67 443 - 800 22 - -N 0 2 C H 3 279 -292 5* 292 -405 4 405 - 800 2 - -H CF 3 163 -213 74 213 -283 71 283 -655 17 655 - 800 19 Material rapidly volatilizes One trend that is apparent here, and is seen again in the trimetallic and polymeric nickel species described later, is the relationship between the nature of the 4-X substituent and the A H values. The DSC events are found to be predominantly exothermic for the complexes in which a halogen is present in the 4-position of the bridging pyrazolates, and endothermic in the compounds with no halogen substituent suggesting that this 4-position substituent plays an important role in the thermal decomposition. 47 Table 2.2. DSC results for the [CpNi(4-X-3,5-diRpz)]2 materials. Ligand substituents Event 1 Event 2 Event 3 X R Temp (°C) A H (kJ mol"1) Temp (°C) A H (kJ mol"1) Temp (°C) A H (kJ mol"1) H C H 3 254 18.5a 313 26.2a 377 95.5a C H 3 C H 3 192 33.9b 314 18.5a 353 89.7a CI C H 3 177 44.0b 269 201b 400 162b Br C H 3 185 95.4b 265 81.0b'c 395 45.0b N 0 2 C H 3 280 ~ 40 b ' c 300 ~ 40b ' c 300 -340 shoulderb H CF 3 Exhibits peaks at: 193a, 197b, 234b, 278b, 329b, 372b, and 575°C b a Event is endothermic b Event is exothermic c Two peaks that are too close to integrate separately 2.2.1.2 X - R A Y DIFFRACTION The X-ray crystal structures of three of these dimeric complexes have been obtained and all of them display a boat-like conformation for the Ni-(N-N) 2 -Ni ring system. The molecular arrangements are shown in Figures 2.5 - 2.7 and a stereoscopic ORTEP view of [CpNi(4-H-3,5-diMepz)]2 is shown in Figure 2.8. Crystallographic data, atomic coordinates, bond lengths and bond angles are tabulated in Appendix I, Tables 1-1, 1-3, 1-4 and 1-5 respectively. 48 Figure 2.5. The molecular structure of [CpNi(4-H-3,5-diMepz)]2, with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms. The internal bond angles and bond lengths in the cyclopentadienyl and pyrazolate ligands are normal for all three complexes. Examination of the nickel(II) coordination spheres of the X = H and X = N 0 2 derivatives reveals no significant differences in bond 49 C13 Figure 2.6. The molecular structure of [CpNi(4-N02-3,5-diMepz)]2, with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms. parameters and hence no measurable structural effect, on substitution of the hydrogen in the 4-position of the pyrazolate ring by the nitro group. However, comparisons of the average Ni -N bond lengths in the [CpNi(4-H-3,5-diMepz)]2 and [CpNi(3,5-F6diMepz)]2 complexes 50 Figure 2.7. The molecular structure of [CpNi(3,5-F6diMepz)]2, with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms. (1.891 A and 1.926 A respectively) indicates a slight lengthening in the complex with the CF 3 electron withdrawing substituents. This is accompanied by a small increase in the Ni-N-C bond angles and decrease in the Ni-N-N bond angles in [CpNi(3,5-F6diMepz)]2 51 Figure 2.8. Stereoscopic ORTEP diagram of [CpNi(4-H-3,5-diMepz)]2, with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms. compared to [CpNi(4-H-3,5-diMepz)]2, a consequence perhaps of increased steric repulsion between the CF 3 groups and the cyclopentadienyl ligands in the former. This also appears to result in a slightly shorter non-bonded Ni—Ni distance in [CpNi(3,5-F6diMepz)]2 compared to [CpNi(4-H-3,5-diMepz)]2 (Table 2.3). 52 It is interesting to compare the non-bonded Ni—Ni separations in these complexes with previously reported values for similar doubly pyrazolate bridged dinickel systems (111, 112). These values are listed in Table 2.3. R Table 2.3. Ni—Ni non-bonded distances in dimeric N i systems L N i R-Compound Ni—Ni Distances (A) Reference L R X Cp C H 3 H 3.159 This work Cp C H 3 N 0 2 3.156 This work Cp CF 3 H 3.128 This work ri 3-Allyl C H 3 H 3.00 (111) NO C H 3 H 3.673 (112) In the three structures reported here the Ni—Ni separations vary very little with only a slight reduction in the distance observed on substituting C H 3 groups by the CF 3 groups on the bridging u-pz' moieties. All three of these dimeric compounds are formally 18 electron Ni complexes in contrast to the r|3-allyl and linear nitrosyl complexes listed in Table 2.3. These 53 latter complexes are both 16-electron N i systems and display widely different Ni—Ni separations. In contrast the very long Ni—Ni separation (3.673 A) in the nitrosyl derivative results from the planar Ni-(N-N) 2-Ni six-membered ring. This arrangement is not disadvantaged sterically since the NO group, unlike the r|3-allyl group, is sterically non-demanding. In addition it has been suggested that the planar arrangement in the nitrosyl complex may arise from favorable 7r-bonding between the pyrazolate 7i-system and the filled d-orbitals on the N i atoms (111). 2.2.1.3 SPECTROSCOPIC BEHAVIOUR 2.2.1.3.1 INFRARED SPECTROSCOPY Unassigned IR band frequencies with relative intensities for the [CpNi(4-X-3,5-diMepz)]2 (X = H , C H 3 , CI, Br, N 0 2 ) and [CpNi(3,5-F6diMepz)]2 appear in Appendix III, Tables III-1. The IR spectra of these complexes do not provide any detailed information regarding their structures but may be used as a "fingerprint" for individual compounds. Infrared spectroscopy was primarily used to determine if any neutral azole was present in the isolated samples by the presence or absence of bands arising from N - H stretching and bending vibrations. In the case of the six dimeric nickel complexes discussed here, no N - H bands were observed indicating that only deprotonated forms of the azoles were present in 54 all complexes. Also observable in the IR spectra of these dimetallic complexes are two bands attributed to stretches arising from the Cp groups. The importance of the presence of these stretches is described in greater detail later in the thesis. 2.2.1.3.2 N M R SPECTROMETRY The TT nmr spectra were recorded in C 6 D 6 solution and the data are presented in Table 2.4. A representative spectrum, for [CpNi(4-H-3,5-diMepz)]2, is shown in Figure 2.9. The nmr results are consistent with a symmetrical structure with equivalent \x-pz' and Cp groups, the latter exhibiting but one sharp singlet signal indicative of free rotation about the Ni-Cp centroid axis. A planar six-membered Ni-(N-N) 2-Ni central ring system with end-capping, freely rotating Cp groups could explain these lH nmr results, but molecular modeling studies show that this leads to severe steric interaction between the methyl substituents on the |J.-pz' rings and the Cp groups, making such an arrangement unlikely (see Figure 2.10). The model shown was generated by manipulating the solid state molecular structure of [CpNi(4-H-3,5-diMepz)]2 (obtained by X-ray crystallography) 55 a . o 7.0 e.o 5.0 4.0 3.0 2.0 1.0 0.0 P P M Figure 2.9. 200 M H z *H nmr spectrum for [CpNi(4-H-3,5-diMepz)]2 in C 6 D 6 solution. wherein the Ni-(N-N) 2 -Ni ring system is in the boat conformation. Al l bond lengths and angles in the figure modeled are identical to the bond lengths and angles found for the molecular structure of [CpNi(4-X-3,5-diMepz)]2. 56 X Table 2.4. lH nmr data for dimeric systems R — — R C p N i < X ^ N i C p R—" — R R Compound X Cp 8 (CeD6 solution) C H 3 X C H 3 H 5.20 s 2.20 s 5.55 s C H 3 CI 5.00 s 2.10 s N A C H 3 Br 4.95 s 2.10 s • N A C H 3 Me 5.22 s 2.15 s 1.62 s C H 3 N 0 2 5.20 s 2.20 s N A CF 3 H 3.37 s N A 6.67 s * 8(C„H6) = 7.20 ppm NA = Not applicable The boat conformation for the Ni-(N-N) 2-Ni ring reduces these steric interactions considerably and maintains planarity about the nitrogen atoms of the u,-pz' rings thereby retaining the resonance derealization energy of the system (see Figure 2.11). 57 Figure 2.10. Space-filling and "ball and stick" representations of a modeled planar Ni-(N-N) r Ni ring system for [CpNi(4-H-3,5-diMepz)]2. There are considerable steric interactions between the methyl protons on the pyrazolate rings with the protons on the capping Cp groups. Figure 2.11. Space-filling and "ball and stick" representations for the boat conformation of the Ni-(N-N) 2-Ni ring system for [CpNi(4-H-3,5-diMepz)]2 (atom positions obtained by X-ray crystallography). There is noticeably less steric interactions between the methyl protons on the pyrazolate rings with the protons on the capping Cp groups for this conformation. 58 The : H nmr results are therefore consistent with the solid state molecular structures obtained by X-ray crystallography in which the three complexes display the boat-like conformation for the Ni-(N-N) 2 -Ni ring system. This suggests that there are no structural changes on dissolution. One notable feature in the X H nmr results is the effect of substituting CF 3 for C H 3 groups on the bridging pyrazolates. This substitution leads to a marked shift in the signals for both the 4-H proton on the pyrazolate rings and the Cp ring protons. The latter signal moves to a higher field position at 5 3.37 whereas the former signal shifts to a lower field position at 8 6.67. Similar shifts in the 4-H proton signal have been reported by Dias et al. for a number of HB(3,5-F 6diMepz) 3 containing complexes. (191, 192). 2.2.1.4 M A G N E T I C PROPERTIES The nmr spectra for the six dimetallic complexes showed no unusual chemical shifts or peak broadening often seen for paramagnetic materials. These results are consistent with spin paired d 8 Ni(II) configurations. In order to verify this the magnetic susceptibilities of two representative dimeric compounds, [CpNi(4-X-3,5-diMepz)]2 (X = H , N 0 2 ) were measured from 2 to 300 K on a SQUID magnetometer. The powder magnetic susceptibility and magnetic moment versus temperature data are tabulated in Appendix II, Table II-2. 59 c s B o o u o SB 400 300 H 200 H 0.25 U 03 Temperature (K) Figure 2.12. Powder magnetic susceptibility and magnetic moment plot for [CpNi(4-H-3,5-diMepz)]2. Both compounds gave evidence for weak paramagnetism. In the high temperature region the magnetization could not be measured accurately as it was of the order of that obtained for the background signal. The susceptibilities measured below 50 K showed an increase in magnitude with decreasing temperature as expected for paramagnetism. The magnitudes of the measured susceptibilities were small, typically of the order of 3% or less than is expected for S=l high spin nickel(II). The % versus temperature and magnetic moment versus temperature plots for the X = H derivative over the low temperature region are shown in Figure 2.12. 60 These results are consistent with diamagnetic spin paired nickel(II) centers with small amounts of high spin nickel(II) impurity. Blake et al. (82) concluded that the ligand environment of Ni(II) in the complexes resulted in a low spin d 8 configuration. Moreover, these authors observed weak paramagnetism in their complexes which they ascribed to either the presence of small amounts of paramagnetic pseudotetrahedral monomers or the presence of a thermally accessible excited S=l state in the dimers. The second possibility can now be ruled out as the magnetic susceptibility decreases with increasing temperature, behaviour inconsistent with a ground S = 0 and an excited (but thermally accessible) S = 1 state. 2.2.2 [CpNi(4-X-3,5-DIMETHYLPYPvAZOLATE)2]2Ni 2.2.2.1 SYNTHESIS, PHYSICAL AND THERMAL PROPERTIES Details of the preparations of these trimetallic materials are given in Chapter 9, Sections 9.2.3.14 through 9.2.3.17. Initially, the [CpNi(4-Cl-3,5-diMepz)2]2Ni complex was isolated during an attempt to make the polymeric [Ni(4-Cl-3,5-diMepz)2]x. An excess of the 4-chloro-3,5-dimethylpyrazole was mixed with nickelocene under an atmosphere of dry dinitrogen. The mixture was heated to 140°C, causing the pyrazole to melt and the nickelocene to dissolve in the molten azole. Within a few minutes the initial green colour of 61 the solution turned red and shortly thereafter a red solid began .to precipitate out of the solution. The reaction vessel was allowed to cool after approximately three hours of heating and the excess 4-Cl-3,5-diMepzH was removed by washing with dry and dioxygen-free benzene. At this point it was noticed that the red solid obtained was slightly soluble in the benzene solvent. Following the removal of the excess azole the a small amount of the red powder was dissolved in a minimum amount of benzene. The solution was placed in a sealed tube and red plate crystals formed over an eight month period. The crystals were isolated by filtration, dried under vacuum for 12 hours and subsequently determined to be the trimetallic [CpNi(4-Cl-3,5-diMepz)2]2Ni. A rational preparation for the [CpNi(4-X-3,5-diMepz)2]2Ni compounds was then developed. Like the dimeric species described above, these trimetallic nickel complexes were prepared by reacting nickelocene with the appropriate pyrazole in dry and dioxygen-free benzene solution. However, to obtain the trimetallic materials an excess of the pyrazole was required and the benzene solutions needed heating (~ 60°C, 18 hours). Upon heating, the initial green colour of the solution (dissolved nickelocene in benzene) typically would turn red within 15 minutes. Within 18 hours an orange or red solid would precipitate out of the solution either as a powder or as single crystals. These solids were collected by filtration and purified by washing with additional dry and dioxygen-free benzene. The desired trimetallic complexes were then dried under vacuum for 12 hours. 62 The syntheses again involve an acid-base reaction in which the pyrazole N H proton is transferred to a Cp ring (from the nickelocene). The resulting pyrazolate binds to the nickel liberating a cyclopentadiene molecule in the process. The equation for this reaction is shown below, 3(NiCp2) + 4(pz''H) ~70°c'48hou" >[CpNi(pz'%]2Ni + MCpH) [2.2] where pz" is 4-X-3,5-diMepz (X = H , C H 3 , CI, Br,) and Cp is the cyclopentadienyl ion. The products are red, air stable, crystalline solids. The complexes are slightly soluble in a variety of polar and non-polar organic solvents, but decompose if exposed to moisture when in solution. T G A and DSC were performed in order to determine the thermal stability of the individual compounds. The T G A and DSC data are summarized in Tables 2.5 and 2.6 respectively, and plots of both T G A and DSC are shown for two representative samples in Figures 2.13 and 2.14. 63 300 400 500 Temperature (C) c U a U 0-—< Figure 2.13. (a) T G A and (b) DSC plots for [CpNi(4-H-3,5-diMepz)2]2Ni. The T G A results reveal that each of the trimetallic complexes is thermally robust, withstanding temperatures over 200°C without decomposing. The weight remaining after each of the complexes had been heated to 800°C varied from 22% for the brominated complex to 42% for the X = H derivative. Unlike the dimetallic complexes discussed earlier, the weight remaining after the maximum temperature had been reached was above that of the nickel content alone. The theoretical weight percentages for the nickel 64 — i 1 1 1 1 1 1 h 0 100 200 300 400 500 600 700 800 Temperature (C) Figure 2.14. (a) T G A and (b) DSC plots for [CpNi(4-Cl-3,5-diMepz)2]2Ni. component in these trimetallic complexes are: 26% for the X = H derivative; 24% for the X = C H 3 derivative; 18% for the X = Br derivative; and 21% for the X = CI derivative. Other than the X = H derivative, the final weight remaining after the materials had been heated to 800°C were, however, quite close to the value that would be expected if the final decomposition product was NiO. Assuming none of the nickel component was lost during the T G A experiment, the weight percentages expected for NiO as the final decomposition 65 product are: 33% for the X = H derivative; 30% for the X = C H 3 derivative; 22% for the X = Br derivative; and 27% for the X = CI derivative. Further evidence that the thermal decomposition product of the nickel pyrazolates may be NiO, and the potential source of oxygen, is given later in the thesis. Table 2.5. T G A results for the [CpNi(4-X-3,5-diMepz)2]2Ni materials. Ligand Step 1 Step 2 Step 3 substituent X Temp Weight Temp Weight Temp Weight Range Left (%) Range Left (%) Range Left (%) (°C) (°C) (°C) H 235 -250 88 250 -421 42 - -C H 3 245 - 335 82 335 -425 52 425 - 656 28 Br 265 -310 78 310-590 22 - -CI 300 - 350 80 350-660 23 660 - 750 27 DSC experiments show that, in most cases, the DSC events occur at temperatures associated with steps in the thermal weight loss. For example, the first DSC event occurs at the temperature at which each complex begins to lose weight. The relationship between the A H values and the 4-X substituent of the pyrazolate group that was observed for the nickel dimeric complexes, was again seen here. The events for the two complexes in which a halogen was present were found to be exothermic while the events for the non-halogenated 66 complexes are primarily endothermic There was one exception to this general trend. The first event (274°C) observed for [CpNi(4-CH3-3,5-diMepz)2]2Ni is exothermic. Table 2.6. DSC results for the [CpNi(4-X-3,5-diMepz)2]2Ni materials. Ligand Event 1 Event 2 Event 3 substituent X T(°C) A H (kJmol1) T(°C) A H (kJmof1) T(°C) A H (kJmol1) H 240 32.7a 307 39.0a 372 261a C H 3 274 16.8b 309 158a - -Br 280 354b - - - -CI 301 336b 385 312b a Event is endothermic b Event is exothermic 2.2.2.2 X - R A Y DIFFRACTION STUDIES The trimetallic materials were obtained in forms suitable for single crystal X-ray diffraction studies. The X-ray determined molecular structures of [CpNi(4-H-3,5-67 Figure 2.15. The molecular structure of one of the two independent molecules of [CpNi(4-H-3,5-diMepz)2]2Ni, with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms. diMepz) 2]2Ni and [CpNi(4-Cl-3,5-diMepz)2]2Ni are shown in Figures 2.15 and 2.16 along with a stereoscopic view of the latter in Figure 2.17. Crystallographic data, atomic coordinates, bond lengths and bond angles are listed in Appendix I, Tables 1-2,1-3,1-4 and 68 C14 Figure 2.16. The molecular structure of [CpNi(4-Cl-3,5-diMepz)2]2Ni, with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms. 1-5 respectively. The complexes again display the boat-like conformation for the Ni-(N-N) 2 -Ni ring systems. These are favoured over a planar conformation on two counts, to relieve steric interactions between the Cp rings and adjacent Me substituents on the 69 Figure 2.17. Stereoscopic ORTEP diagram of [CpNi(4-Cl-3,5-diMepz)2]2Ni, with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms. pyrazolate ligands, and also to relieve steric interaction between the Me substituents of the pyrazolate rings which are adjacent to the central N i atom. As observed for the dimeric complexes, the internal bond lengths and angles of the ligands in [CpNi(4-H-3,5-diMepz)2]2Ni and [CpNi(4-CI-3,5-diMepz)2]2Ni are normal. Moreover, bond parameters associated with the nickel coordination sphere are not significantly different in the two complexes showing, again, that substitution of the hydrogen in the 4-position by a more electron withdrawing substituent (CI) has no 70 measurable structural consequence. The Ni—Ni non-bonded distance is slightly greater in the X = CI derivative than in the X = H analogue and this is accompanied by marginally greater N i -N-N and smaller Ni-N-C bond angles in the former complex. There seems no obvious intramolecular steric interactions which could cause these small differences suggesting that intermolecular packing forces are responsible. In Table 2.7 the Ni—Ni non-bonded distances observed in the two complexes are listed together with those for related trinickel complexes (111, 112, 134). Table 2.7. Ni—Ni non-bonded distances in trimetallic nickel complexes Compound Ni—Ni Distances (A) Reference l l R X Cp CFfe H Cp C H 3 CI ri 3-Allyl C H 3 H NO C H 3 H acac C H 3 H 3.2739 3.2949 3.2314 3.3770 3.1972 This work This work (111) (112) (134) 71 Unlike the similarly substituted dimeric systems discussed earlier, each of the five complexes shown in Table 2.7 has been shown to adopt the boat-like Ni-(N-N) 2 -Ni ring systems in the solid state. This results in a much smaller range of Ni—Ni non-bonded distances than is observed for the dimetallic systems, in which both planar and boat-like Ni-(N-N) 2 -Ni ring arrangements were observed. The result is not unexpected since a totally planar structure, which might lead to a longer non-bonded nickel-nickel distance (as observed for the dimeric nitrosyl derivative (111)), is not favoured since this would lead to severe steric interactions around the central N i atom. 2.2.2.3 SPECTROSCOPIC B E H A V I O U R 2.2.2.3.1 INFRARED SPECTROSCOPY The IR band frequencies with relative intensities for the [CpNi(4-X-3,5-diMepz) 2] 2Ni (X = H , C H 3 , CI, Br) appear in Appendix III, Table III-2. As mentioned earlier, infrared spectroscopy was primarily used to determine if any neutral azole was present in the isolated samples by the presence or absence of bands arising from N - H stretching and bending vibrations. For the dimetallic complexes discussed previously and the trimetallic materials described here, the infrared spectra were used to ensure that the removal, by washing or sublimation, of the excess azole was complete The IR spectra of 72 the four trimetallic nickel complexes exhibited no N - H bands confirming that only the deprotonated forms of the azoles were present in these samples. Lastly, like the dimetallic nickel(II) systems, two bands (described in more detail in the next chapter) attributed to the Cp groups were observable in these trimetallic systems. 2.2.2.3.2 N M R SPECTROSCOPY The X H nmr spectra were recorded in C 6 D6 solution and the data are presented in Table 2.8. A representative spectrum, for [CpNi(4-H-3,5-diMepz)2]2Ni, is shown in Figure 2.18. The nmr data for the complexes in CeD6 solution are consistent with the solid state structures described above. In solution, the two end-capping Cp rings are equivalent and freely-rotating about the Ni-Cp axis giving rise to one sharp singlet at 8 5.6. The 3-Me and 5-Me groups on the pyrazolate ligands are non-equivalent in these structures and give rise to two closely separated singlets at ~ 8 2.1 - 2.2. The 4-X substituents, where applicable, give signals which are easily located in the lH nmr spectra on the basis of both anticipated intensities and predicted chemical shifts. 73 s . o s . o 7,0 0 . 0 a . o 4 . 0 3 . o a.a 1.0 Figure 2.18. 200 M H z *H nmr spectrum of [CpNi(4-H-3,5-diMepz)2]2Ni in CeD 6 solution. Table 2.8. *H nmr data for trimetallic complexes Compound 8 (C 6 D6 solution) X R Cp C H 3 X H C H 3 5.63 s 2.15,2.20 s 5.40 s Br C H 3 5.15 s 1.95, 2.10 s N A C H 3 C H 3 5.53 s 2.05, 2.23 s 1.55 s CI C H 3 5.65 s 2.15, 2.18 s N A 74 2.2.2.4 M A G N E T I C B E H A V I O U R The magnetic susceptibilities at temperatures from 2 to 300 K of two representative trimetallic compounds (X = H , CI) were measured using a SQUID magnetometer. The powder magnetic susceptibility and magnetic moment versus temperature data are tabulated in Appendix II, Table II-3. Both of these compounds, like the dimeric complexes described earlier in this chapter, gave evidence for weak paramagnetism due to small amounts of paramagnetic impurity. Once again the magnetization in the high temperature region could not be measured accurately as it was of the order of that obtained as the background signal for the sample container. Plots of % versus temperature and magnetic moment versus temperature for the [CpNi(4-Cl-3,5-diMepz) 2]2Ni complex are shown in Figure 2.19. The similarities between the magnetic behaviour of the trimetallic compounds and the dimeric complexes described earlier are not surprising as the outer Ni(II) ions have the same environment as the Ni(II) ions in the dimeric systems, whereas the central Ni(II) ion is in a regular square planar environment which is expected to give rise to a low spin (diamagnetic) d 8 configuration. 75 o £ o up O 3000 2500 H 2000 g- 1500 00 •2 1000 —I <u c DO ^ 500 1 i i i i 1 r 10 15 20 25 30 35 Temperature (K) 40 45 0.6 h 0.5 a 0.4 £ o o © 0.3 g) 03 0.2 0.1 50 Figure 2.19. Powder magnetic susceptibility and magnetic moment plots for [CpNi(4-Cl-3,5-diMepz)2]2Ni. 2 3 S U M M A R Y AND CONCLUSIONS Reactions between pyrazoles and nickelocene under controlled temperature conditions have produced a series of dimetallic, [CpNi(4-X-3,5-diMepz)]2 (X = H, C H 3 , CI, Br, N0 2 ) , [CpNi(3,5-F6diMepz)]2, and trimetallic, [CpNi(4-X-3,5-diMepz)2]2Ni (X = H, C H 3 , CI, Br) nickel pyrazolate complexes. A representative number of these materials 76 (three dimetallic and two trimetallic) has been examined by single crystal X-ray diffraction studies. The materials consist of nickel(II) centers linked by double pyrazolate bridges and end-capped with cyclopentadienyl groups. The determined solid state structures show the Ni-(N-N) 2 -N rings to be in a boat-like arrangement due to steric interactions involving the 3-Me and 5-Me substituents on the pyrazolate groups. *H nmr spectra of samples in CeD6 solution are consistent with the solid state molecular structures. Magnetic measurements from 2-300 K on both dimetallic and trimetallic complexes indicate that the materials are diamagnetic, consistent with a low-spin d 8 electron configuration. In each, however, there is some structural paramagnetic impurity resulting in a small, but measurable, magnetic moment at low temperatures. 77 Chapter 3 P O L Y - (NI C K E L (11) AZOLATES) 3.1 INTRODUCTION A major goal of this work was to produce a series of linear chain polymeric compounds of the form [Ni(azolate)2]x. These target compounds were studied with an anticipation of being able to explain the magnetic properties of the materials based on their molecular structures. It was also a goal to compare subsequently the magnetic properties of this series of nickel compounds with similar materials involving other transition metals, some new and some thoroughly characterized previously. Several of these targeted polymers with the general formula [Ni("azolate")2]x (where "azolate" = 4-Hpz, 4-Clpz, 4-H-diMepz, 4-Br-diMepz, 4-Cl-diMepz, 3,4,5-triMepz and indz) have been prepared successfully. Although none of the polymeric materials was isolated in a form suitable for single crystal X-ray diffraction studies, structures based on indirect spectroscopic and X-ray powder diffraction evidence and magnetic data are proposed. Subsequently, the variable temperature magnetic data have been rationalized in terms of the proposed structures. 78 The idea to synthesize the target transition metal polymers from reactions involving nickelocene came from a 1977 report by Blake et al. (82), in which the successful preparations of polymeric nickel complexes obtained by reacting nickelocene with pyrazole and 3-methylpyrazole were reported. In their report the authors also described the isolation of dimeric nickel complexes containing doubly-bridging 3,5-dialkylpyrazolate ligands and end-capping r | - C 5 H 5 groups. No polymeric complexes involving 3,5-disubstituted pyrazoles were prepared and steric arguments were used in an attempt to explain this lack of polymerization. No structures were confirmed by X-ray diffraction studies in this earlier work. The complexes discussed in this chapter have been divided into two categories based on their magnetic properties. Diamagnetic compounds constitute the first category while the second category is comprised of paramagnetic complexes. Initially in this chapter three diamagnetic nickel(II) polymers are described and some representative characterization of these systems is discussed. These diamagnetic polymers were of interest primarily as a basis of comparison with the paramagnetic systems obtained, and as such have not been as thoroughly characterized as the paramagnetic polymers. 79 3.2 RESULTS A N D DISCUSSION 3.2.1 DIAMAGNETIC NICKEL(II) P O L Y M E R S 3.2.1.1 NICKEL(II) 4 -X-PYRAZOLATES (X = H , CI) 3.2.1.1.1 SYNTHESIS, STRUCTURE, A N D T H E R M A L A N D M A G N E T I C PROPERTIES Details of the syntheses employed in these preparations are given in Chapter 9, Sections 9.2.3.1 through 9.2.3.7. [Ni(4-Hpz)2]x was successfully prepared using two synthetic approaches. The complex was first produced using a method developed by Ehlert (193) in which nickel powder was reacted with an excess of molten pyrazole at 110°C under normal atmospheric conditions for 48 hours. The mixture was allowed to cool and the excess pyrazole was removed by washing with dichloromethane. The product was then isolated as an insoluble orange powder by physical separation using a magnet to remove the unreacted nickel powder. A second synthetic approach was used to produce both [Ni(4-Hpz)2]x and [Ni(4-Clpz) 2] x in which nickelocene was reacted with an excess amount of the appropriate pyrazole under inert conditions. A convenient and clean pathway to the desired products was developed wherein the reactants were placed in a Carius tube which was subsequently evacuated and sealed. The mixture was then heated above the melting temperature of the pyrazole (90 - 150°C) giving rise to a green solution of dissolved nickelocene in the molten 80 ligand. For the reactions described in this section, the solution would typically change colour to orange or yellow, accompanied by the precipitation of a yellow solid within 30 minutes. This method is a variation of the one reported by Blake et al. in which metal pyrazolate complexes were obtained by reacting nickelocene with pyrazoles in benzene solutions at room temperature (82). In the metallocene synthetic approach, the polymerization involves an acid-base reaction in which the pyrazole N H proton is transferred to a Cp ring (from the nickelocene). The resulting pyrazolate binds to the nickel metal liberating cyclopentadiene in the process. x(MCp2) + xs(pz"H) 90°-150°48h°urs >[Ni(pz")2]x(s) + 2x(CpH) [3.1] In reaction [3.1] pz" is 4-Hpz or 4-Clpz and CpH is cyclopentadiene. The desired products were obtained and purified by first allowing the solution to cool and solidify. The mixtures were subsequently washed with a variety of solvents in order to dissolve and remove the excess pyrazole and any other soluble by-products from the desired products. Reaction [3.1] can be generalized to include other metallocenes and azoles, and was the primary synthetic approach used to obtain polymeric materials described in this thesis (Figure 3.1). 81 poly(metal(II))pyrazolate Figure 3.1. Schematic representation of the polymerization of metallocenes with azoles. One difficulty encountered using this synthetic approach was the lack of control over reaction conditions. In the absence of a solvent the reaction temperature needed to be above that of the melting temperature of the reacting pyrazole, typically over 100°C and the rate of the reactions tended to be too fast for the formation of any crystals suitable for X-ray diffraction studies. Attempts to reduce the reaction rate, and thus increase the likelihood of growing single crystals, were made by including small amounts of various solvents and lowering the reaction temperature. It was found, however, that in the case of reactions between nickelocene with 4-Hpz and 4-Clpz, lowering the reaction temperature did not produce macroscopic single crystals. In addition, i f the temperature was lowered too far, the powder obtained did not analyze as pure polymer, typically analyzing high in both carbon and hydrogen, and low in nitrogen. These elemental analysis results can be 82 attributed to the presence of a mixture of shorter chain species in which the end-capping cyclopentadienyl groups are detectable in the elemental analysis (see Table 3.1 and 3.2 later in this section). In another effort to grow crystals suitable for X-ray diffraction studies, the reactions were again carried out in sealed Carius tubes, but below the melting point of the pyrazoles and with no solvent present. It was hoped that the significant vapour pressure of both nickelocene and pyrazole, especially at elevated temperatures, would permit either a gas-solid or gas-gas reaction to proceed at a suitable rate for polymer crystal formation. However, not only did these reactions fail to yield suitable crystals, the materials obtained were again mixtures of the shorter chain species described above. Finally, and again in an attempt to grow single crystals, the previously developed synthetic approach in which the elemental metal is reacted directly with molten pyrazole under a dioxygen atmosphere was employed. Unfortunately, the reaction between nickel powder and the ligands invariably yielded an amorphous powder product. Attempts to react nickel metal shot (in place of the metal powder) with pyrazoles, both in the presence of and in the absence of an organic solvent, failed to yield any product under all conditions attempted. The [Ni(4-Xpz) 2] x materials are air stable and involatile solids. They are insoluble in water and all common organic solvents. The compounds dissolve with decomposition in 83 Figure 3.2. Thermal decomposition of [Ni(4-Hpz)2]x by (a) T G A and (b) DSC. mineral acids. The materials were found to be thermally quite robust, decomposing without melting at temperatures above 350°C. T G A and DSC plots for the X = H and X = CI derivatives are shown in Figures 3.2 and 3.3 respectively. 84 Figure 3.3. Thermal decomposition of [Ni(4-Clpz)2]x by (a) T G A and (b) DSC The thermal decomposition of these polymeric materials occurs over a broad temperature range; 350 - 515°C for [Ni(4-Hpz)2]x and 360 - 545°C for [Ni(4-Clpz)2]x. TGA studies in a dinitrogen atmosphere reveal that weight loss for both compounds takes place in two steps as the temperature is ramped slowly upward. For [Ni(4-Hpz)2]x there is roughly a 50% loss in mass between 350°C and 445°C. The material then loses an additional 20% of its total mass between 445°C and 515°C. Similarly for [Ni(4-Clpz)2]x roughly 53% of its mass is 85 lost in a continuous fashion between 360°C and 425°C, followed by an additional loss of 20% between 425°C and 545°C. One unusual occurrence in the T G A results of [Ni(4-Hpz) 2] x is an apparent increase in mass as the temperature is increased above 515°C. The mass of the material increases from roughly 30% of the polymer's initial mass at 515°C to approximately 37% by 800°C. The actual weight increase (7% of the initial 17.5 mg) is roughly 1.25 mg. This peculiar behaviour, possibly a result of the formation of nickel oxide, is also observed for several other polymeric materials described in other sections of the thesis. The masses remaining after the compounds had been heated to temperatures in excess of 700°C was only slightly higher than the mass due to the nickel component originally present. In fact, the observed weights remaining at the highest temperatures studied are consistent with the weights expected for a final decomposition product of NiO. From the initial amount of nickel available, and assuming that no nickel is lost during the T G A experiment, the weight percent expected for NiO is 39% for the X = H derivative and 28% for the X = CI derivative. Also consistent with this decomposition product are the physical properties of the remaining solids obtained following T G A experiments. Like NiO, the solids are green materials that are soluble in acid, and insoluble in water and common organic solvents. In order to explain the presence of dioxygen in the sample chamber (obviously needed if this decomposition product is in fact NiO) some details of the T G A apparatus need to be described in more detail. Although the T G A experiment is said to be performed under a dinitrogen atmosphere, in reality there is only a flow of N 2 gas passing through the sample chamber. It is not possible to completely prevent dioxygen gas from 86 entering the sample area as an exhaust pipe is required in order to prevent pressure from building up in the apparatus. It is not practical to place a gas bubbler on the exhaust pipe as this could lead to fluctuations in pressure within the sample area. Fluctuations in pressure would in turn disrupt the very sensitive analytical balance used to determine the weight of the material. Further evidence that the thermal decomposition products of similar materials contain oxygen is provided later in the thesis. Differential Scanning Calorimetry (DSC) studies give some insight into the manner in which these materials decompose. The thermal decomposition of [Ni(4-Hpz)2]x involves a major endothermic process (AH = 72.9 kJmol"1) between 350°C and 475°C, peaking at about 460°C. Conversely, the thermal decomposition of [Ni(4-Clpz)2]x involves a significant and fairly symmetrical exothermic event between 380°C and 425°C (AH = -84.8 kJmol"1), followed by a smaller broad exothermic event between 425°C and 460°C. These results again suggest that the 4-X (X = CI, H) substituents of the pyrazolate ligands play an important role in the thermal decomposition of these polymeric materials. A possible explanation for these results involves a decomposition pathway in which nickel-X bonds are formed giving rise to some undetermined intermediate decomposition product. It is difficult to determine an approximate length (or molecular weight) for these materials. Typical methods for such determinations on polymeric materials require that the material be at least partially soluble. As no suitable solvent was found for these compounds, only a minimum chain length, based on elemental analysis, could be 87 determined. As can be seen in Table 3.1, in order for the elemental analysis result for [Ni(4-Hpz) 2] x to be consistent with a polymer of infinite chain length (within a 0.3% error per element determined), approximately 100 Ni(4-Hpz) 2 repeat units are necessary. Even then the analysis will not necessarily fall within the accepted range for the "pure" polymer. Similar calculations, results of which are shown in Table 3.2, for [Ni(4-Clpz)2]x again indicate a minimum chain length of approximately 100 Ni(4-Clpz) 2 repeat units. The experimentally determined values for C, H , and N are 37.4%, 3.3% and 28.9% respectively for [Ni(4-Hpz)2]x and 27.8%, 1.4%, and 21.2% respectively for [Ni(4-Clpz)2]x. One potential method for determining the approximate molecular weight or chain length of these polymers would be to incorporate heavier end-capping groups. If, for example, pentamethylcyclopentadienyl groups were used, the minimum chain length that would result in an expected elemental analysis consistent with [Ni(4-Clpz)2]x would increase to 200 repeat [Ni-(N 2C 3H 2C1) 2] units. If the actual chains are shorter than this, the elemental analysis would give a very rough estimate of that chain length. Alternatively, if the elemental analysis is still within the 0.3% absolute error for the infinite chain material it could be established that the chains are likely at least 200 repeat units in length. 88 Table 3.1. Theoretical molecular weights and % compositions for various chain lengths of Cp[Ni(4-Hpz)2]xNiCp, Chain length Mol. Weight (g/mol) % Carbon % Hydrogen % Nitrogen x = 0 188.899 63.6 5.3 0.0 x= 1 381.7492 50.3 4.2 14.7 x = 2 574.5994 46.0 3.9 19.5 x = 5 1153.15 41.7 3.5 24.3 x= 10 2117.401 39.7 3.3 26.5 x = 20 4045.903 38.6 3.2 27.7 x = 50 9831.409 37.9 3.2 28.5 x= 100 19473.919 37.6 3.2 28.8 x = 500 96604.10 37.4 3.1 29.0 x = 0 0 (192.8502)x 37.4 3.1 29.1 89 Table 3.2. Theoretical molecular weights and % compositions for various chain lengths of Cp[Ni(4-Clpz)2]xNiCp, CI Chain length Mol . Weight (g/mol) % Carbon % Hydrogen % Nitrogen x = 0 188.899 63.6 5.3 0.0 x= 1 381.7492 42.7 3.1 12.4 x = 2 574.5994 37.1 2.6 15.7 x = 5 1153.15 32.1 2.0 18.7 x= 10 2117.401 30.0 1.8 20.0 x = 20 4045.903 28.8 1.7 20.7 x = 50 9831.409 28.1 1.6 21.1 x= 100 19473.919 27.8 1.6 21.3 x = 500 96604.10 27.6 1.6 21.4 X = 00 (261.72)x 27.5 1.5 21.4 90 Figure 3.4. S E M image of [Ni(4-Hpz)2]x. The white bar represents a length of 5 urn. The particle morphology of [Ni(4-Hpz)2]x was examined by S E M and an image of the complex is shown in Figure 3.4. It is clear from this image that the compound consists of very small particles which aggregate in an apparent random mass. The room temperature magnetic moments of the [Ni(4-Xpz) 2] x compounds were recorded on a Johnson-Matthey magnetic susceptibility balance. In both cases the materials were found to be diamagnetic, consistent with a square planar nickel(II) d 8 configuration. 91 3.2.1.1.2 X - R A Y DIFFRACTION STUDIES As mentioned earlier, no macroscopic single crystals of nickel (II) polymeric materials were successfully prepared. However, some powder diffraction studies were undertaken in order to compare these materials with previously studied systems involving other transition metals. The <i-spacings and relative intensities for the peaks in the pattern are listed in Appendix V, Table V - l . In the case of [Ni(4-Hpz)2]x (prepared via the nickelocene synthetic procedure) the material is evidently mostly amorphous as can be seen by the poor diffraction pattern in Figure 3.5. This result is consistent with the S E M examinations in which no microcrystalline material was observable under the low resolution settings of the microscope. In addition, by comparing the diffractogram with the diffraction patterns of [Co(4-Hpz)2]x and [Cu(4-Hpz)2]x (193, 194) it is apparent that [Ni(4-Hpz)2]x is isomorphous neither with the copper or cobalt analogue. No quantitative information was determined from the diffraction pattern obtained for this material. 92 a O U Figure 3.5. Powder X-ray diffractogram of [Ni(4-Hpz)2]x. 3.2.1.1.3 SPECTROSCOPIC BEHAVIOUR 3.2.1.1.3.1 INFRARED SPECTROSCOPY The infrared data for the polymers can be found in Appendix III, Table III-3. Infrared spectroscopy was used primarily to ensure that no neutral pyrazole molecules remained in the polymeric materials. In the spectra obtained, the absence of the characteristic N - H stretch was a good indication that all of the ligand was present as deprotonated pyrazolate in these materials. Assignments of the vibrational spectra of pyrazole and 4-chloropyrazole have been made previously (59, 195). Assignments of the vibrational spectra of pyrazolate and 4-chloropyrazolate have also been made (79). 93 Another potential use of IR is to ensure that no bands emanating from Cp groups are present in the polymeric materials. Theoretically, the presence of any end-capping Cp groups, akin to the ones in the dimetallic and trimetallic nickel complexes already discussed, would be detectable by IR. Conversely, if the polymers are indeed "infinite" chains, no signals corresponding to Cp groups would be expected. Qualitatively, only two bands attributed to Cp stretches in -q-CpNi containing materials (nickelocene for example) are distinctive enough to be identified in these nickel pyrazolate systems. In nickelocene, a band attributed to v(CH) occurs at 3075 cm"1, and another band, 7i(CH), is present at 800cm"1 (196, 197). Appropriately, bands close in energy to both of these absorptions are observed in all of the dimetallic (3095 cm"1, 785-795cm"1) and trimetallic (3095cm'1, 785-790cm"1) nickel complexes discussed in Chapter 2. However, consistent with the proposed polymeric nature of the [Ni(4-Xpz) 2] x materials, these bands are not observed in the IR spectra of these materials. 3.2.1.1.3.2 ELECTRONIC SPECTROSCOPY The electronic spectra of the two [Ni(4-Xpz) 2] x polymers are shown in Figure 3.6. In both cases the spectra consist of a symmetrical band centered at about 450 nm (22 000 cm"1). This result is consistent with a N i N 4 square-planar chromophore (198). 94 (a) I I l I 1 1 1 0 500 1000 1500 2000 2500 3000 Wavelength (nm) Figure 3.6. Electronic spectra of (a) [Ni(4-Hpz)2]x, (b) [Ni(4-Clpz)2]x and (c) a lower mull concentration of [Ni(4-Clpz)2]x showing the bands at 250 and 325 nm. Two higher energy bands, centered at 250 nm and 325 nm, are visible in the spectra of samples run at lower concentrations (shown in Figure 3.6 (c) for [Ni(4-Clpz)2]x) and are likely charge transfer in origin. 95 3.2.1.1.4 PROPOSED STRUCTURES Throughout this section it has been assumed that the [Ni(4-Xpz)2]x materials are linear chain polymers. However, in the absence of single crystal X-ray diffraction data this conclusion must be examined more closely and alternate possible structures must be considered or discounted. The most probable alternate structure that is consistent with the elemental analyses, infrared data and electronic spectra of these materials would be a ring structure. Blake et al. (82) suggested a trimetallic ring, shown in Figure 3.7, in which three square planar nickel (II) ions are connected by a total of six pyrazolate groups. Figure 3.7. Possible trimetallic structure of [Ni(4-Hpz)2]x. 96 Larger and more complicated ring structures can also be imagined that would yield similar experimental results. Consistent with both linear chain polymers and large ring polymers, these materials are insoluble in all solvents in which they do not decompose or react. The materials are involatile and do not melt up to the temperatures at which they decompose, in both cases above 350°C. Although there is no conclusive proof to discount these large ring structures, there is increasing evidence that the materials are indeed linear chain polymers. Incomplete reactions between 4-ClpzH and nickelocene yielded products with elemental analysis data (C: 28.5%, H : 1.8%, N : 21.0%) consistent with oligomeric chains of [Ni(4-Clpz) 2] x (x = 20 - 50) capped by cyclopentadienyl groups (see Table 3.2). In addition, structurally characterized copper(II) pyrazolate complexes have been shown to adopt rigid linear chain structures (85). If these materials of composition [Ni(4-Xpz) 2] x are indeed linear chain polymers they are most likely to adopt a structure in which each nickel(II) ion is linked to two other nickel centers via a total of four 4-X-pyrazolate bridges. The coordination geometry about the nickel ions is thought to be square planar giving rise to a series of Ni-(N-N) 2 -Ni six membered rings in a boat conformation as shown in Figure 3.8. The boat conformation is more likely than a totally planar structure as the latter would be less favored sterically. 97 Nil Figure 3.8. Proposed linear chain structure, with square planar nickel centers, for the [Ni(4-Hpz)2]x material. Evidence for a square planar N i N 4 chromophore includes the electronic and magnetic data. The electronic spectroscopic data show a single band attributable to spin-allowed transition at an energy consistent with other square planar nickel(II) systems (198). Compounds in which four-coordinate nickel(II) chromophores are present in other geometries have markedly different electronic spectra - both in terms of the number of visible electronic transitions and the energies at which these compounds absorb (198). Magnetic data indicate the compounds are diamagnetic, also consistent with a square planar nickel(fl) geometry. 98 3.2.1.2 NICKEL(II) INDAZOLATE, Later in this chapter it is shown that the substitution of relatively bulky substituents at the 3 and 5 positions on pyrazolate bridging ligands force the otherwise square planar nickel centers in bispyrazolato-nickel complexes into a tetrahedral geometry. Attempts were made to see if substitution at only one of the 3 or 5 positions would lead to the same nickel chromophore distortion. For this purpose indazole was chosen. In its deprotonated form, indazolate, shown in Figure 3.9, can bridge transition metal centers in much the same way as the pyrazolate ion. N N Figure 3.9. The potential bridging ligand indazolate. 99 3.2.1.2.1 SYNTHESIS, PHYSICAL, A N D T H E R M A L A N D M A G N E T I C PROPERTIES OF [Ni(indz)2]x A detailed description of the preparation of [Ni(indz)2]x can be found in Chapter 9, Section 9.2.3.7. The target complex was obtained by heating nickelocene with excess indazole in a sealed and evacuated tube. In this case the initial green solution of nickelocene in molten indazole turned orange within about an hour. Over the course of 2 days an orange powder precipitated out of the solution and the intensity of colour in the molten ligand decreased. The mixture was subsequently allowed to cool and the tube was opened under normal atmospheric conditions. The orange powder was isolated by removal of any remaining soluble starting materials by washing with THF, xylenes and hexane. The powder was then dried under vacuum for 12 hours. The product was insoluble in all common solvents. The complex is involatile and thermally robust. T G A and DSC results are shown in Figure 3.10. T G A results indicate the material loses weight in two steps as the temperature is continuously raised. The first step involves a loss of roughly 40% of the initial mass and this weight loss occurs between 400°C and 440°C. The second step in thermal decomposition, as monitored by T G A involves a loss of approximately 38%. This second step occurs between 440°C and 595°C. Following this second step there is an apparent 100 Figure 3.10. Thermal decomposition of [Ni(indz)2]x studied by (a) TGA and (b) DSC. weight gain of 5% as the temperature is raised to the maximum temperature studied of 800°C. This weight gain may be the result of the formation of nickel(II) oxide, as there was likely dioxygen gas present in the sample chamber. The final weight percent recorded (27%) is again consistent with a final decomposition product of NiO. DSC results show an exothermic event (AH = 56.8 kJmor1) occurring between 400°C and 480°C accompanying 101 the thermal decomposition. The room temperature magnetic moment of [Ni(indz)2]x was measured on a Johnson-Matthey magnetic susceptibility balance and the compound was found to be diamagnetic, consistent with a square planar spin-paired d 8 configuration. 3.2.1.2.2 INFRARED SPECTROSCOPY Infrared data for this complex are tabulated in Appendix III, Table III-4. The infrared spectrum of [Ni(indz)2]x shows the characteristic absence of the V N - H stretch indicating that all of the indazole present is in its deprotonated form, consistent with the proposed formulation. 3.2.1.2.3 POSSIBLE STRUCTURES Due to the asymmetrical nature of substitution on the indazolate moiety, there are several different conceivable ways the individual indazolate ligands can align when incorporated into a linear chain polymer. A few of the extreme possibilities are represented in Figure 3.11. 102 V ^ • < > iW" ".1 VT^  Figure 3.11 continued overleaf 103 Figure 3.11. Four possible arrangements of the indazolate bridging groups for a [M(indz)2]x material. Interestingly, in each of these illustrated molecular arrangements (Figure 3.11) a similar steric interaction to the one responsible for the distortion to tetrahedral nickel centers in the [Ni(4-X-3,5-diMepz)2]x materials (discussed in detail later in this chapter) is absent. It is possible to imagine a structure in which there are steric interactions similar to the ones present in the tetrahedral polymers and one such arrangement is shown in Figure 3.12. To illustrate this repulsion better some of the hydrogen atoms likely to interact sterically are shown. 104 . N N . M NL N N '"• x x y x X X x N V Yrfl Figure 3.12. A linear chain structure in which some steric interactions similar to those in the [i\i(4-X-3,5-diMepz)2]x materials, in this case between the bridging indazolate groups, is present. It should be noted however that even in this case the steric interaction can only be present in short segments of the linear chain. Without a bulky group in both the 3 and 5 position there is the likelihood that some or all of the nickel centers can obtain a square planar conformation. Consistent with this are the magnetic properties of [Ni(indz)2]x. The compound is diamagnetic suggesting the nickel chromophore geometry is square planar throughout the material. Therefore the molecular structure of this complex is likely one of, or a combination of, the structures represented in Figure 3.11. 105 3.2.2 P A R A M A G N E T I C NICKEL(II) P O L Y M E R S 3.2.2.1 NICKEL(II) 4-X-3,5 -D1METHYLP Y R A Z O L A T E S (X = H , C H 3 , Br, CI) 3.2.2.1.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPETIES Detailed descriptions of the syntheses can be found in Chapter 9, Sections 9.2.3.3 through 9.2.3.6. Attempts to produce [Ni(4-H-3,5-diMepz)2]x using the nickel powder in molten pyrazole reaction that was successful for the [Ni(4-Hpz)2]x preparation (described earlier in this chapter) did not yield the product in a pure form. However, the other successful synthetic approach for the preparation of the [Ni(4-Xpz) 2] x (X = H, CI) materials, in which nickelocene was reacted with excess amounts of the appropriate pyrazole, led to the successful preparations of the 4-X-3,5-diMepz (X = H , C H 3 , CI, Br) analogues. Blake et al. had reported in 1977 that unlike 4-H-pyrazole and 3-CH3-pyrazole, 3,5-dialkylated pyrazoles would not form polymeric materials due to the steric bulkiness of the 3,5-substituents (82). The authors were able only to obtain dimeric complexes of general formula [CpNi(3,5-dialkylpz)]2. However it seems that in their attempts to prepare polymeric 4-X-3,5-dialkylpyrazolate nickel(II) materials the authors did not employ temperatures high enough to generate the polymeric compounds. Most of the reactions attempted by Blake et. al (82) were performed in benzene solution and at room temperature. Under these mild conditions the polymeric materials do not form. Instead, dimeric materials (as found by Blake et al. (82)) and trimetallic compounds of general formula [CpNi(4-X-3,5-diMepz)2]2Ni (discussed previously in Chapter 2) are generated. In 106 the present work it was found that in order to form polymeric chains the reaction had to be performed at temperatures above 100°C and for at least 48 hours (reaction [3.2]). x(Cp2M)+xs(4-X-3,5-diMepzH) ' ^ ' ^ > [ M ( 4 - X - 3 , 5 - d i M e p z ) 2 ] x + 2x(CpH) [3.2] In these reactions nickelocene was intimately mixed with an excess of the desired substituted pyrazole. The mixture was placed in a dry and dioxygen-free Carius tube and the tube flame-sealed under vacuum. The reaction vessel was then placed in an oven (100°C -150°C) causing the pyrazole to melt. Initially a green solution of nickelocene dissolved in the molten pyrazole was observed. After about ten minutes the green solution would change to a deep red and a few hours later a coloured powder (puce for X = H , C H 3 and brown for X = CI, Br) would begin to precipitate out of the melt. Upon prolonged reaction (typically 48 hours) the intensity of the colour in solution would decrease and the quantity of solid present would increase. At this point the mixture was allowed to cool and the solid product was purified by removal of the excess 4-X-3,5-diMepzH by washing with dry and dioxygen-free THF. Like the [Ni(4-Xpz) 2] x counterparts, the [Ni(4-X-3,5-diMepz)2]x materials were isolated as coloured powders. No single crystals suitable for X-ray diffraction studies were obtained for any of the compounds prepared. In an effort to grow X-ray quality crystals, 107 similar attempts to those tried for the [Ni(4-Xpz) 2] x systems (described in Section 3.2.1.1) were performed but were ultimately unsuccessful. There are many physical similarities between the pSTi(4-X-3,5-diMepz)2]x systems and the previously discussed [Ni(4-Xpz) 2] x analogues. The 3,5-disubstituted systems are air-stable, coloured powders insoluble in water and all common organic solvents. They are also involatile and thermally quite robust. The thermal decomposition of these polymers was studied by both T G A and DSC and representative plots are shown in Figures 3.13 and 3.14. The T G A results show that in general these compounds are slightly less thermally stable than the [Ni(4-Xpz) 2] x and [Ni(indz)2]x systems. Depending on the 4-X substituent on the bridging 3,5-dimethylpyrazolate groups (see Table 2.3) the materials begin to decompose without melting at temperatures between 300°C and 345°C. As mentioned earlier in this chapter the [Ni(4-Xpz) 2] x systems begin to decompose at 350°C (X = CH 3 ) and 360°C (X = CI). Also described earlier in this chapter is the unusual occurrence in the T G A results in which an increase in weight is observed for both the [Ni(4-Hpz)2]x and [Ni(indz)2]x materials. A similar increase in weight is again observed, this time for the two non-halogen containing [Ni(4-X-3,5-diMepz)2]x polymers. The [Ni(4-CH3-3,5-diMepz)2]x material has its minimum weight percent (25% of the initial mass) at 480°C. The weight of the decomposition material then increases with rising temperature up to a value of 30% of 108 Figure 3.13. Thermal decomposition of [Ni(4-H-3,5-diMepz)2]x as studied by (a) T G A and (b) DSC. the initial mass at 550°C. There is no additional weight change in the material as the temperature is increased to the maximum value studied of 800°C. The actual measured weight gain of the material (5% of the initial 5.9 mg) is nearly 0.3 mg. Similarly for the [Ni(4-H-3,5-diMepz)2]x material, an apparent weight gain of roughly 5% is observed 109 100 200 300 400 500 600 700 800 Temperature (C) Figure 3.14. Thermal decomposition of [Ni(4-Cl-3,5-diMepz)2]x as studied by (a) T G A and (b) DSC. between 520°C and 650°C. No additional weight change is observed as the temperature is increased from 650°C to 800°C. Again it is possible that this weight gain is the result of the formation of nickel oxide. 110 Table 3.3. T G A results for the [Ni(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI and Br) and [Ni(4-Xpz) 2] x (X = H , CI) systems. Material 1st step Weight left 2 n a step Weight left Weight temp, range after 1st temp, range after 2 n d step expected for (°C) step(%) (°C) (%) NiO(%) [Ni(4-H-3,5-diMepz)2]x 345 -430 35 430 - 520 30 30 [Ni(4-CH3-3,5 -diMepz) 2] x 335 -480 30 N A N A 27 [Ni(4-Cl-3,5-diMepz)2]x 300 - 395 35 395 - 495 25 24 [Ni(4-Br-3,5-diMepz)2]x 315 - 395 50 395 - 595 20 18 rNi(4-Hpz)2]x 350 -445 50 445 - 515 39 39 [Ni(4-Clpz)2]x 360 -425 45 445 - 545 27 28 [Ni(indz)2]x 400 -440 40 440 - 595 27 26 * Weight increases slightly as temperature is raised further The thermal stabilities and decomposition trends of the [Ni(4-X-3,5-diMepz)2]x systems were investigated further by DSC. Two markedly different results were obtained for the systems with, and without, halogen substituents. In the case of the polymers with the. brominated and chlorinated bridging 4-X-3,5-diMepz ligands, each two step decomposition was accompanied by two separate exothermic events. Firstly the 4-Br-3,5-diMepz containing polymer has a symmetrical and sharp exothermic peak centered at 375°C (AH = -167 kJmol"1), followed by a broad, less well-defined exothermic event spanning 400°C to 550°C. Similarly the 4-Cl-3,5-diMepz containing polymer has an initial sharp, 111 symmetrical exothermic event centered at 390°C (AH = -79.5 kJmol"1), followed by a less well defined and much broader exothermic peak from 400°C to 550°C. In the two systems in which no halogen substituents are present an endothermic event is observed. The thermal decomposition of [Ni(4-H-3,5-diMepz)2]x has, associated with it, a broad symmetrical endothermic peak from 415°C to 460°C (AH = 34.1 kJmol"1). Similarly there is an endothermic event spanning 415°C to 460°C (AH =100 kJmol"1) accompanying the thermal decomposition of the ^i(4-CH 3-3,5-diMepz) 2] x material. Unfortunately there was no event associated with the weight increase for the two non-halogenated compounds and thus no information about this increase in mass with temperature was obtained. Comparing the DSC data for the four polymers discussed here with the data obtained for the [Ni(4-Xpz) 2] x (X = H , CI) materials, a trend becomes apparent. In each system in which a halogen is occupying the 4 position of the bridging pyrazolate ligand the thermal decomposition is exothermic. For the systems in which a methyl group or proton occupies the 4-position carbon of the pyrazolate bridge the thermal decomposition is endothermic in nature. This same trend is also present for most of the events observed in the thermal decomposition of the nickel dimetallic and trimetallic complexes described in Chapter 2. This is very good evidence that the 4-substituent of the pyrazolate plays a major role in the decomposition of these polymeric materials. 112 0 k 9 6 6 i E0W Figure 3.15. S E M images of (a) [Ni(4-H-3,5-diMepz)2]x, and (b) [Ni(4-CH 3-3,5-diMepz)2]x. The white bar represents a length of 5 um. The particle morphologies of two of the [Ni(4-X-3,5-diMepz)2]x materials (X = H, CH 3 ,) were examined by S E M and images of these are shown in Figure 3.15(a)-(b). The compounds appear to be present as very small crystallites. From the images obtained it is difficult to gain any conclusive insight as to the detailed morphologies. However, as can be seen in the image of [Ni(4-CH3-3,5-diMepz)2]x, there is a tendency for the microcrystallites to be elongated in nature. This, of course, would be consistent with an internal structure consisting of extended chains. 113 3.2.2.1.2 X - R A Y DIFFRACTION STUDIES Single crystals suitable for X-ray diffraction studies were not prepared for the [Ni(4-X-3,5-diMepz) 2] x materials and thus only powder diffractograms could be recorded. The resulting patterns are shown in Figure 3.16(a-d) and the af-spacings and relative intensities for the peaks in these patterns are listed in Appendix V, Table V-2. The compounds are not isomorphous with each other and neither are they isomorphous with [Ni(4-Hpz)2]x (Figure 3.5). In addition, the X = Br derivative has a significant amorphous component and thus does not diffract as well as the other three derivatives. However, as can be seen in Figure 3.17, the X = H derivative is isomorphous with the previously studied [Co(4-H-3,5-diMepz) 2] x and [Zn(4-H-3,5-diMepz)2]x, but not with [Cu(4-H-3,5-diMepz)2]x (193, 199). Unfortunately, the structures of the cobalt and zinc analogues have not been determined. However, in the case of the cobalt compound strong evidence in favour of a linear chain polymer with tetrahedral metal centers has been presented (88). 114 115 Figure 3.17. Powder diffraction patterns for the [M(4-H-3,5-diMepz)2]x materials, where (a) M = N i (this work), (b) M = Co (199), (c) M = Zn (199), and (d) M = Cu(193). 116 3.2.2.1.3 SPECTROSCOPIC BEHAVIOUR 3.2.2.1.3.1 INFRARED SPECTROSCOPY The infrared data for the polymers can be found in Appendix III, Table III-5. As mentioned earlier in this chapter, infrared spectroscopy was used primarily to ensure that no neutral pyrazole molecules were remaining in the polymeric complexes. In the four systems studied here, there is again the absence of the characteristic N - H stretch expected for neutral pyrazoles indicating that all of the ligand is present as deprotonated pyrazolate. Further evidence supporting the presence of the deprotonated ligand, most likely coordinated to metal centers, can be obtained by looking at the in-plane VHn g mode of the ligands. It has been reported that the in-plane Vn n g mode of 4-H-3,5-diMepzH which occurs at 1596 cm"1 in both solid and CC1 4 solutions of the neutral ligand (200) shifts to lower frequencies when coordinated to a metal. For example ,^ the band shifts to: 1573 cm"1 in Au(4-H-3,5-diMepzH)(PPh3)(BF4) (201), 1575 cm"1 in Au(4-H-3,5-diMepzH)Cl (201), and 1556 cm"1 in Co(4-H-3,5-diMepzH)2Cl2 (77). However, a shift of this peak to lower energy does not necessarily confirm the coordination of the ligand to a metal center. In the ionic compound, Na[4-H-3,5-diMepz], the in-plane Vrin g band of the ligand occurs at 1512 cm"1 (193). This latter case shows that the deprotonated form of the ligand can give rise to a similar shift as is seen for the coordination of the neutral ligand to a metal center. Perhaps the most pertinent complexes for a comparison with the [Ni(4-X-3,5-diMepz)2]x 117 compounds studied here, are the polymeric transition metal species, [M(4-X-3,5-diMepz)2]x ( M = Co, Cu and X = H , C H 3 , CI, Br), studied by M . K. Ehlert et al. (85-87, 193). In these complexes the 4-X-3,5-diMepz ligands are both deprotonated and coordinated to metal centers. Ehlert et al. reported that the methylated pyrazole in-plane Vhng band is shifted 60 -75 cm"1 lower in energy when the deprotonated ligand is coordinated to a metal. In the [Ni(4-X-3,5-diMepz)2]x (X = H , C H 3 CI and Br) complexes, like the previously reported copper and cobalt analogues, it is believed that the ligand is present in its deprotonated form and is coordinated to metal centers. Examining the IR spectra of these four complexes reveals that the expected shift is indeed observed (Table 3.4). Table 3.4. Positions of the in-plane Vn n g band positions in the [Ni(4-X-3,5-diMepz)2]x complexes and 4-X-3,5-diMepzH compounds. Compound band position (cm"1) 4-H-3,5-diMepzH 1596 [Ni(4-H-3,5-diMepz)2]x 1526 4-CH 3-3,5-diMepzH 1596 [Ni(4-CH3-3,5-diMepz)2]x 1512 4-Cl-3,5-diMepzH 1584 [Ni(4-Cl-3,5-diMepz)2]x 1516 4-Br-3,5-diMepzH 1577 [Ni(4-Br-3,5-diMepz)2]x 1508 118 Another correlation between IR spectral data and the coordination mode of 4-X-3,5-diMepz groups has been reported by Vos and Groeneveld (59, 79). They found that complexes containing only bridging 4-X-3,5-diMepz ligands exhibit a shift in frequency of the PC-CH3 mode of the methyl groups to higher values (compared to the corresponding free ligand) by 35-50 cm"1. It is thought that the shift to higher energy is due to increased steric interactions between these methyl groups and other adjacent groups upon coordination. In the case of 4-CH3-3,5-diMepz there is an additional Pc-ci* mode the frequency of which is less dependent on the coordination mode of the ligand (193). The PC-CHS band positions of the [Ni(4-X-3,5-diMepz)2]x complexes and those of the corresponding neutral, free ligands are listed in Table 3.5. These observed shifts of the pc-cra band positions provide good evidence for the presence of coordinated 4-X-3,5-diMepz anions as the uncoordinated 4-H-3,5-diMepz anion, found in the ionic Na[4-H-3,5-diMepz], does not give rise to a PC-CHS band position shift when compared to the free neutral ligand (193). As is the case for the nickel polymers of formulation [Ni(4-Xpz) 2] x, no bands attributed to cyclopentadienyl groups are observed. This result is consistent with the proposed polymeric nature of these materials. 119 Table 3.5. Comparison of the PC-CHS band positions in [Ni(4-X-3,5-diMepz)2]xand 4-X-3,5-diMepzH. Compound 3C-CH3 band position (cm"1) 4-H-3,5-diMepzH 416 [Ni(4-H-3,5-diMepz)2]x 450 4-CH 3-3,5-diMepzH 482, 581 [Ni(4-CH3-3,5-diMepz)2]x 503,571 4-Cl-3,5-diMepzH 479 [Ni(4-Cl-3,5-diMepz)2]x 515 4-Br-3,5-diMepzH 473 [Ni(4-Br-3,5-diMepz)2]x 509 3.2.2.1.3.2 ELECTRONIC SPECTROSCOPY The electronic absorption spectra of the pSTi(4-X-3,5diMepz)2]x (X = CI, H , C H 3 , Br) compounds are shown in Figure 3.18 (a) to (d). These compounds have spectra with three observable spin allowed d-d transitions as is expected for tetrahedral Ni(II) and one spin forbidden d-d transition. The three spin allowed transitions, Vi: 3Ti g(F)—» 3T 2 g(F), v 2: 3 T l g (F) -^ 3 A 2 g (F) , and v 3: 3Ti g(F)-> 3Ti g(P) are observed in the ranges of 1665-1695 nm, 770-780 nm and 520-525 nm respectively. The remaining observed transition has been assigned to the 3Tig(F)—»*D and lies in the range 860-870 nm. Comparison of the data with 120 200 400 600 800 1000 1200 1400 1600 1800 2000 Wavelength (nm) Figure 3.18. Electronic spectra of (a) [Ni(4-Cl-3,5 -diMepz)2]x, (b) [Ni(4-H-3,5-diMepz)2]x, (c) [Ni(4-Br-3,5-diMepz)2]x, (d) [Ni(4-CH3-3,5-diMepz)2]x and (e) a more concentrated mull of [Ni(4-H-3,5-diMepz)2]x. the correlation diagram developed by Tanabe and Sugano (45) for tetrahedral systems with d 8 electron configurations show that these transitions are consistent with Dq and B values of 700 cm"1 and 970 cm"1. Finally the compounds exhibit strong charge transfer bands in the ranges 350-360 nm and 240-250 nm. 121 3.2.2.1.4 M O L E C U L A R M O D E L I N G STUDIES A N D PROPOSED STRUCTURES Like the [Ni(4-Xpz)2]x compounds the 4-X-3,5-dimethylpyrazolate analogues have properties consistent with polymeric structures. However, the increased bulk due to the additional methyl substituents on the bridging pyrazolate ligands is thought to account for a change in the geometry of the nickel chromophore. Evidence for this changing N i N 4 chromophore geometry comes primarily through the analysis of the electronic spectra described above. There is also a basic change in the magnetic properties (to be discussed below), consistent with the chromophore change. Molecular modeling studies, to be described next, support the contention that the change to a tetrahedral N1N4 chromophore in the 3,5-dimethylpyrazolate complexes is a consequence of steric effects. In order to generate reasonable representations of the proposed linear chain structures of the nickel polymers, crystallographic data from related compounds were utilized. Firstly, a polymer incorporating square planar nickel ions doubly-bridged by 3,5-dimethylpyrazolate moieties was modeled by manipulating the crystal structure data of the trimetallic compound [(CpNi(4-H-3,5-diMepz)2]2Ni (figure 3.19). 122 Figure 3.19. Crystal structure of [CpNi(4-H-3,5-diMepz)2]2Ni showing the square planar geometry about the central nickel(II) ion. As can be seen, this structure involves a square planar central nickel ion linked to two other nickel ions via four 4-H-3,5-diMepz bridges. Cyclopentadienyl groups cap the two terminal nickel ions. To simulate the polymer, the Cartesian coordinates from the crystal structure data were imported into a 3-dimensional molecular modeling program on a Silicon Graphics computer. The data points representing the capping cyclopentadienyl groups were removed from the data set and a copy of this "bare" oligomer was made. The first nickel from one oligomer was then superimposed on the third nickel of the second identical oligomer to mimic a short portion of a polymeric species. Although this modeling method allows for no additional bending or stretching of the nickel-nitrogen bonds it does give a reasonable representation of how much room will be available for successive methyl groups on the 123 Figure 3.20. Modeled polymer segment for [Ni(4-H-3,5-diMepz)2]x in which the metal centers are present in a square planar geometry. bridging pyrazolate ligands. As is evident from the resulting modeled structure, shown in Figure 3.20, there is considerable steric interactions between the methyl substituents on every third pyrazolate ring. In order to show the reduced steric interactions for a compositionally identical polymer incorporating tetrahedral nickel ions in place of square planar ones, the crystal structure data of a different trimetallic species was utilized. [Co(4-H-3,5-diMepz)2Cl(4-H-3,5-diMepzH)]2Co, prepared by Ehlert in our laboratory (113) and shown in Figure 3.21, was used in the model building. 124 Figure 3.21. Crystal structure of [Co(4-H-3,5-diMepz)2Cl(4-H-3,5-diMepzH)]2Co (113) showing the tetrahedral geometry about the central metal ion. This compound has been chosen as the starting point for this model for 2 reasons: the metal ions present are in a regular tetrahedral configuration and the bridging ligands are the desired 4-H-3,5-diMepz moieties. The atomic x, y and z coordinates from the structural data were imported into the modeling software and the capping CI and 4-H-3,5-diMepzH groups were deleted from the data set. The remaining portion of the trimetallic molecule was subsequently copied and translated on to itself in the same manner as described above for the square planar modeled system. This process gives rise to the simulated polymer segment incorporating 5 metal ion centers and 8 pyrazolate bridges with each metal ion in a tetrahedral geometry and is shown in Figure 3.22. 125 Figure 3.22. Modeled polymer segment for [Ni(4-H-3,5-diMepz)2]x in which the metal centers are present in a tetrahedral geometry. It is apparent from the two modeled systems that there is significantly more room for the methyl substituents on the pyrazolate rings in the system in which the nickel ions are present in the tetrahedral configuration. In the two modeled systems above, the C—C distances between the sterically hindered methyl substituents are calculated at ~ 2.29 A for the tetrahedral system and 1.42 A for the square planar arrangement. It can also be seen that upon removal of the two methyl substituents the steric interaction is eliminated from both geometries discussed, and electronic effects would likely control the nickel geometry. These models are consistent with the electronic spectral and magnetic data (to be presented later in the case of the 3,5-dimethylpyrazolate complexes) obtained for the four polymeric species discussed here and the two [Ni(4-Xpz) 2] x materials described earlier. In the 126 materials incorporating the more bulky 4-X-3,5-diMepz ligands the spectral and magnetic data suggest tetrahedral nickel ions. However, in the polymeric [Ni(4-Xpz) 2] x in which no bulky groups are present in the critical 3,5 positions of pyrazolate bridging ligands, the data support square planar nickel ions. Interestingly, although attempts have been made, no linear chain polymeric materials have been prepared involving any of the metals studied in our laboratory thus far (Cu, Co (199)) and Mn, N i (this work)) with the even bulkier 3,5-bis-(trifluoromethyl)pyrazolate (3,5-F6diMepz) ligand. This suggests that the additional steric bulkiness of the CF 3 groups is sufficient to prevent polymer formation. At first it may not be obvious as to why the tetrahedral system may give a reduced steric interaction between the methyl substituents on the pyrazolate rings. The reason for this is most easily understood by first looking at the metal to metal separation in the two possibilities. Figure 3.23(a) shows a Ni-(N-N) 2-Ni polymer segment for the case in which the nickel ions are present in a tetrahedral geometry and Figure 3.23(b) shows the same segment in a square planar geometry. The factors influencing the metal to metal separation include the N i - N (distance a) and N - N (distance b) bond lengths and the angle N - N i - N (angle 2a). For the square planar system an additional 'out of plane' angle (P) must be considered. In both cases the distance between two adjacent nickel ions (x) can be described by the equation x = 2a' + b [3.3] 127 where a' is the component of a, along the internickel axis and b is the N - N bond length. In the tetrahedral case a' can be determined by simple trigonometry a' = a(cosa) [3.4] and in the square planar case by a'= a(cosa)(cosP). [3.5] where the angles a and P are those shown in Figure 3.23 (a) and (b). b (a) Tetrahedral Nickel Ions (b) Square Planar Nickel Ions Figure 3.23. Pictorial representations showing the factors influencing Ni—Ni separations in doubly pyrazolate bridged nickel systems for (a) a tetrahedral metal geometry and (b) a square planar metal geometry. 128 In the tetrahedral system, the bridging pyrazolate rings are in a common plane between any two neighbouring nickel ions. In the square planar conformation the bridging ligands are in a boat conformation which, depending on the boat angle P, can force two successive nickel ions, and thus every third 4-X-3,5-diMepz ligand, closer to each other. Assuming there is some steric interaction between the methyl groups on the pyrazolate rings for both of the idealized geometries, it is useful to consider the possibility of reducing these steric interactions by distorting the metal chromophore geometries. The simplest way to create more room for the methyl substituents is to elongate the chain, or in other words, to increase the distance between each metal center. By considering the effects of elongating the chain it becomes evident that a distorted tetrahedral system has more potential for increasing the separation of the methyl groups than does a distorted square planar arrangement. In the distorted tetrahedral system a reduction in the angle N-Ni-N, or 2a as it is labeled in Figure 3.23(a), gives rise to greater separation between nickel centers (and therefore methyl substituents) without any significant increased steric interactions elsewhere. However, in the case of the square planar nickel geometry, reducing the angle 2a does not have the same impact as a comparable reduction of 2a for the tetrahedral system. The reason for this is the Ni—Ni separation in the square planar system is also dependent on the out-of-plane angle P (Figure 3.23(b)). By decreasing angle P (squashing the boat) it would be possible to increase the Ni—Ni separation. However in order to 129 Figure 3.24. Steric interaction resulting from the elongation of a linear chain of square planar nickel(II) ions doubly bridged by 4-H-3,5-diMepz ligands. The arrows labeled "a" represent the elongation along the chain and the arrows labeled "b" represent the resulting steric interaction between methyl groups and ring moieties. achieve Ni—Ni separations comparable to the tetrahedral system, the angle P must be decreased to the extent wherein a new steric interaction between the methyl substituents on pyrazolate rings with the ring portion of adjacent pyrazolate groups is introduced (Figure 3.24). 130 3.2.2.1.5 M A G N E T I C PROPERTIES Magnetic susceptibilities of the four [Ni(4-X-3,5-diMepz) 2] x compounds were measured from 2 to 3 0 0 K using a SQUID magnetometer. The powder magnetic susceptibilities and magnetic moments at different temperatures are tabulated in Appendix II, Table II-1. Firstly it is important to note that unlike the [Ni(4-Xpz) 2 ] x materials, these compounds are paramagnetic. This, as mentioned in the previous section, is attributed to a tetrahedral or distorted tetrahedral geometry about the nickel(II) ions as a square planar N i N 4 geometry typically leads to a diamagnetic compound. The temperature dependent magnetic studies show that the polymers exhibit fairly strong antiferromagnetic coupling. In all cases the magnetic moment values decrease with decreasing temperature. Plots of powder magnetic susceptibility, X M , versus temperature (Figures 3 . 2 5 (a) to (d)) reveal that each complex exhibits a maximum in susceptibility at - 6 0 K with the exception of the [Ni(4-Br-3,5-diMepz) 2] x material which exhibits a maximum at - 5 5 K . Al l of the compounds show an increase in susceptibility with decreasing temperature at the lowest temperatures studied consistent with small amounts of structural paramagnetic impurities in the samples. There is an observable discontinuity in the susceptibility vs. temperature data for the [Ni(4-CH 3-3,5-diMepz) 2] x occurring at 3 0 K suggesting a structural phase change for this polymer. Similar discontinuities have been observed previously in the copper systems [Cu(4-Clpz) 2] x and [Cu(4-Brpz 2] x ( 8 6 ) . 131 0.012 0.012 .2 0.006 y d So 0.004 0 50 100 150 200 Temperature (K) 250 300 50 100 150 200 Temperature (K) 250 300 Figure 3.25 continued overleaf 132 o E E u o u s ao C3 E u 0.012 0.010 (c) a 0.008 -4 0.006 0.004 0.012 0.010 H a 0.008 U u 93 ed 0.006 0.004 0.010 0.009 0.008 -0.007 0.006 -0.005 - 1 — r ~ 0 20 40 60 80 100 50 il I f— 100 150 200 Temperature (K) 250 300 (d) 0.010 0.009 • • • 0 50 100 150 200 Temperature (K) 250 300 Figure 3.25. Powder magnetic susceptibility plots for the [Ni(4-X-3,5-diMepz) 2] x materials: (a) X = H, (b) X = C H 3 , (c) X = Br, and (d) X = CI. The circles are the experimental data and the calculated values are represented by the curved lines. The insert plots show the data fitted to the low temperature range (10 - 100 K for (a), (c), and (d) and 30 - 100 K for (b)). Parameters for the calculated susceptibilities are those in Table 3.6. 133 In order to quantify better the extent of antiferromagnetic coupling in these systems the magnetic data were modeled based on the linear chain structures proposed for these materials. This was accomplished by analyzing the magnetic susceptibility data using the isotropic Heisenberg model for exchange coupled linear chains developed by Weng (202) and Hiller et al. (203). This model is described by the equation, 0.6667 + 2.5823(| J\/kTf k T [ l + 3.603 5(\j\/kT) + 39.558(| j\/kTf NgW [3.6] where N is Avogadro's number, g is the Lande splitting factor, u.B is the Bohr magneton, k is Boltzmann's constant, T is the absolute temperature, and J is the exchange coupling constant. Equation [3.6] as it is written does not take into account any trace paramagnetic impurity. To generate better correlation between the experimental and modeled data a correction for paramagnetic impurity should be made. Assuming the paramagnetic impurity behaves according to Curie law, the term A para *$kT can be added to [3.6] where S is the spin of the magnetic ion (£ = 1 in the tetrahedral nickel(II) systems described here). The resulting equation 134 Xc •il-p) Ng2^ kT 0.6667 + 2.5823(| J\/kTf 1 + 3.6035(| j\/kf) + 39.558(| j\/kTf J r + P J V Ng2nBS(S + l)* 3kT [3.8] where P is the relative proportion of paramagnetic impurity present in the sample, was used to model the experimental magnetic data. The calculated susceptibilities were obtained by minimizing the function F = i ^ ( x rb s - x r , c ) 2 1=1 fabs)2 [3.9] where n is the number of data points. This minimized value of F was also used to indicate the goodness-of-fit between the experimental and calculated data. The parameters J, \AJS2\, g, P, and F are shown in Table 3.6. Table 3.6. Calculated magnetic parameters for the [Ni(4-X-3,5-diMepz)2]x polymers. Compound Temperature range (K) /(cm" 1) \4JS2\ g P F [Ni(4-H-3,5-diMepz)2]x 2-300 - 19.3(7) 76 2.48(3) 0.0043(6) 0.062 10-100 - 16.2(8) 64 2.33(6) 0 (fixed) 0.031 [Ni(4-CH3-3,5-diMepz)2]x 2-300 - 16.2(3) 64 2.43(2) 0.0018(3) 0.028 30-100 - 14.5(3) 56 2.28(2) 0 (fixed) 0.0073 [Ni(4-Br-3,5-diMepz)2]x 2-300 - 16.4(2) 64 2.41(1) 0.0032(2) 0.027 10-100 - 14.2(2) 56 2.28(1) 0 (fixed) 0.0095 [Ni(4-Cl-3,5-diMepz)2]x 2-300 * - 18.9 76 * 2.50 0.0010* 0.058 10-100 - 16.6(2) 64 2.40(1) 0 (fixed) 0.055 Computer unable to calculate error in parameter. 135 It is obvious from the plots comparing the experimental and calculated magnetic susceptibilities (Figures 3.25 (a) to (d)), as well as by considering the values of F (Table 3.6), that there is not an excellent agreement between the observed and theoretical data. The primary difficulty in modeling these systems involves the orbitally degenerate ground state for the tetrahedral nickel(II) centers. There have been no models developed that can simultaneously predict the effects (on magnetic properties) of spin-orbit coupling in orbitally degenerate ground term configurations and magnetic exchange coupling between metal centers as a function of temperature. For this reason, the magnetic data have also been analyzed over the temperature range 30 - 100 K for the [Ni(4-CH3-3,5-diMepz)2]x compound and 10 - 100 K for the other three compounds. It was hoped that by considering only the data in the temperature regions near the maximum susceptibility values that better agreement between the experimental data and theoretical values could be obtained. As is evident from the fits shown in Figure 3.21 and from the F values listed in Table 3.6, there is better agreement between the experimental and calculated data when only the lower temperature data is utilized. The two most noticeable, albeit not very significant, differences between the full temperature range fits and the lower temperature range fits is the decrease in the calculated coupling constant, J , by 2 to 3 cm"1 and a decrease in the calculated g by roughly 0.15. From the derived magnetic parameters it would appear that the magnitude of the exchange constant is not very dependent on the electronic contribution from the 4-X 136 substituent on the pyrazolate ring, as the magnitudes of J for all four compounds are the same within experimental error. In addition, the magnetic results suggest that the four complexes are structurally very similar as significant deviations in the N i N 4 chromophore or N i - N bond lengths would be expected to have a more profound effect on the magnitude of the exchange coupling. Even though the magnetic parameters calculated for the [Ni(4-X-3,5-diMepz)2]x materials have been based on a less than perfect model, the results do give a fair indication of the magnitude of the exchange coupling in these systems. It is therefore interesting to compare the results to those for similar systems involving other transition metals. In particular the copper(II) and cobalt(II) analogues to the nickel polymers studied here have previously been characterized (87, 193). The [Cu(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI, and Br) materials exhibited fairly strong antiferromagnetic coupling showing maxima in the susceptibility between 105 and 120 K. The calculated values of the exchange constant for the copper polymers were found to be; -56 cm"1 for the X = H derivative, -61 cm"1 for X = C H 3 , -81 cm"1 for X = CI, and -66 cm"1 for X = Br. Comparisons can also be made between the current nickel polymers and other copper systems. [Cu(4-Xpz)2]x (X = H , C H 3 , Br, CI (green) and CI (brown)), [Cu(3-CH3pz)2]x, and Cu(indz)2]x have all been prepared and characterized (85, 86). Once again, like the nickel polymers, each of these complexes exhibited antiferromagnetic exchange. The magnitude of the exchange coupling was again large for the copper systems with susceptibility maxima occurring at 143, 180, 137 200, 195, 160, 110 and 160 K for the 4-Hpz, 4-CH 3pz, 4-Brpz, 4-Clpz (green), 4-Clpz (brown), 3-CH 3pz and indz compounds, respectively. The calculated values of J for the series of copper polymers were determined to be -82, -99, -106, -99, -90, -62 and - 93cm"1 for the 4-Hpz, 4-CH 3pz, 4-Brpz, 4-Clpz (green), 4-Clpz (brown), 3-CH 3pz, and indz compounds, respectively. The [Co(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI, and Br) polymers as well as [Co(4-Hpz)2]x and [Co(3-CH 3pz) 2] x also exhibited antiferromagnetic exchange between metal centers. Susceptibility maxima were found to be between 10 and 15 K for the four 4-X-3,5-diMepz derivatives, 38 K for the 4-Hpz complex and 25 K for [Co(3-CH 3pz) 2] x . Attempts by Ehlert et al. to model quantitatively the magnetic data revealed that no single model available could adequately describe the magnetic exchange of these Co(II) pyrazolates over the entire temperature range (2 to 300 K) (88). However, approximate values of -J were obtained for these systems and typically fell in the 2 - 7 cm"1 range. Qualitatively, the magnitude of - J for these complexes was found to follow the trend: 4-Hpz > 3-CH 3pz » 4-X-3,5-diMepz (X = C H 3 , CI, Br) > 4-H-3,5-diMepz In order to compare the magnitude of exchange coupling in systems with different total spin, S, it is appropriate to compare \AJS2\ values (204) instead of simply comparing values 138 of J. Therefore, the range of \j\ and \4JS2\ values for the nickel(II) systems have been tabulated (Table 3.7) along with the corresponding values for the copper(II) and cobalt(II) systems described above. Table 3.7. \j\ and \4JS2\ ranges for polymeric transition metal (Ni(II), Cu(II), Co(II)) C-substituted pyrazolates. Metal | J\ values (cm"1) 4JS2 values (cm 1) Ni(II) 1 Cu(II) Vi Co(II) 3/2 14-16 56-1063 2-7b 56-64 56-106 16-63 a references: (85-87) b references: (88) By considering the \4JS2\ values listed in Table 3.7, it is evident that the magnitude of the exchange coupling is comparable in all three metal systems. In fact the four nickel(II) complexes have values of \4JS2\ that fall within the ranges obtained for both the copper and cobalt systems. However, the much smaller variation in the magnitude of exchange coupling observed for the nickel(II) polymers suggests that there is less dependence on the substitution of the bridging pyrazolate groups in these nickel systems than there is for the cobalt and copper analogues. 139 Another comparison can be made between the magnetic behaviour of the four nickel polymers described here and the related nickel complex, pSfi(4-H-3,5-diMepz)2-2H20]x, reported by Singh, Satpathy and Sahoo (77). Interestingly, the above paramagnetic complex has a temperature independent magnetic moment of ~3.2 B . M . (over the temperature range of 100 - 300 K), indicating the absence of magnetic exchange. 3.3 S U M M A R Y A N D CONCLUSIONS A series of seven nickel(II) polymeric materials has been successfully prepared and characterized. Although no crystals suitable for single crystal X-ray diffraction studies were obtained, there is considerable evidence for the proposed linear chain structures in which the azolate ligands doubly-bridge successive nickel centers. Magnetic data and molecular modeling studies indicate that the N i N 4 chromophore geometry in these systems is affected by the bulkiness of the substituents in the 3 and 5 positions on the pyrazolate bridges. The nickel centers tend to attain a square planar N i N 4 chromophore geometry unless there is sufficient steric interactions between the bridging ligands to force the N i N 4 chromophore into a tetrahedral arrangement. In the seven materials studied here the distortion to tetrahedral nickel occurred only in the complexes that incorporate 4-X-3,5-diMepz (X = H , C H 3 , CI, Br) bridges. The two methyl groups in the 3 and 5 positions are thought to be 140 responsible for the steric interaction and thus the distortion. In polymers in which the 3 and 5 positions of the pyrazolate bridges were unsubstituted, the N i N 4 chromophore geometry was able to acquire a square planar geometry. Substitution of a relatively bulky group at only the 3 position did not provide enough steric interaction to force the nickel chromophore geometry into the tetrahedral conformation. Studies of the magnetic properties of the four tetrahedral systems indicate the presence of antiferromagnetic coupling between nickel(II) ions in the chains. The calculated /values ranged from -16 cm"1 to -19 cm"1, and the electronic effects from the 4-X (X = H, C H 3 , CI, and Br) substituent did not seem to influence the magnitude of the exchange. An anomaly in the % versus temperature plot at 30 K for the [Ni(4-CH3-3,5-diMepz)2]x material suggests a structural phase change for this polymer. Finally, nickelocene has been shown to be an excellent source of nickel(II) for preparing divalent nickel polymers. Reactions between nickelocene and azoles can be carried out in the absence of solvent by heating the mixture above the melting temperature of the azole in a sealed and evacuated tube. 141 Chapter 4 POLY-(MANGANESE (II) AZOLATES) 4.1 INTRODUCTION In order to extend further the work done on linear chain poly-(metal(II) azolates) some novel manganese complexes of the form [Mn(azolate)2]x were targeted. As with the polymeric nickel compounds described earlier, the goal was to prepare a series of azolate derivatives and attempt to explain the magnetic properties of the materials based on their molecular structures. It was also a goal to compare the magnetic properties with those found for other, similar transition metal systems. Four manganese pyrazolate materials of the formulation [Mn(4-X-3,5-diMepz)2]x (where X = H , C H 3 , CI and Br) and two compounds of formulation [Mn(4-Xpz)2(4-XpzH)] x (where X = CI, Br) were successfully prepared. Although none of the materials were obtained in a form suitable for single crystal X-ray diffraction studies, proposed structures based on indirect evidence are given. 142 4.2 RESULTS A N D DISCUSSION 4.2.1 MANGANESE(II) 4 -X-PYRAZOLATES (X = CI, Br) 4.2.1.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPERTIES Details of the synthetic procedures used are given in Chapter 9, Section 9.2.4.5 and 9.2.4.6. The success of the reactions of nickelocene with pyrazoles described in Chapters 2 and 3 led to a similar approach for the syntheses of manganese complexes. Manganocene was used as the manganese(II) source and was reacted with an excess of the appropriate pyrazole in a sealed and evacuated tube. The manganocene tended not to be very soluble in the molten pyrazoles giving rise to a mixture of beige powder ( M n C p 2 ) and a colourless liquid in the tubes. This poor solubility made the reactions difficult to monitor as the final polymeric products were also insoluble in the molten azoles and very similar in appearance to the starting manganocene. As described earlier, the extent of the polymerization reactions between nickelocene and molten pyrazoles could be crudely monitored by visual inspection of both the intensity of the colour present in the molten pyrazole solution, and the amount of coloured solid product present in the reaction vessels. Since this procedure could not be followed for the manganocene reactions, the reactants were often heated for at least a week to ensure complete formation of suitable amounts of the polymeric complexes. The benefits of this synthetic approach should be noted. Firstly, by reacting the manganocene directly with the molten pyrazoles the possibility of side reactions involving 143 solvent is eliminated. Secondly, by placing the reactants in a sealed and evacuated tube no side reactions involving atmospheric components could interfere with the desired reactions. This was particularly important as both manganocene and many of the isolated manganese polymers (described here and later in this chapter in Section 4.2.2) were found to be quite sensitive to air and moisture. Finally, the sealed tube enables the reactants to be heated well above the melting temperature of the pyrazole without fear of losing either of the volatile starting materials. The materials obtained were white ([Mn(4-Clpz)2(4-ClpzH)]x) or near-white ([Mn(4-Brpz)2(4-BrpzH)]x) powders insoluble in all common organic solvents. They were involatile and thermally robust. DSC and T G A results on [Mn(4-Clpz)2(4-ClpzH)]x are shown in Figure 4.1 and indicate the material decomposes without melting at temperatures over 220°C. Unfortunately neither thermal gravimetric analysis or differential scanning calorimetry data were obtained on the X = Br derivative because of its air sensitivity. T G A data collected on a sample of [Mn(4-Clpz)2(4-ClpzH)]x showed the sample to lose mass in several stages as the temperature was gradually increased between 220°C and 700°C. No further loss in weight was detected up to the maximum temperature studied of 800°C. The T G A results indicate that the one equivalent of neutral 4-chloropyrazole present in [Mn(4-Clpz)2(4-ClpzH)]x is indeed coordinated to metal centers as no weight loss is observed at 144 —r 1 1 1 r— 1 1 h- 0 100 200 300 400 500 600 700 800 Temperature (C) Figure 4.1. Thermal decomposition of [Mn(4-Clpz)2(4-ClpzH)]x as measured by (a) T G A and (b) DSC. temperatures below 220°C. Had the neutral 4-ClpzH been present as a molecule of crystallization or as an impurity, weight loss would be expected to be observed at much lower temperatures as the ligand precursor itself is considerably volatile even below its melting temperature of 90°C. By examining the percent weight loss in the complex as temperature is increased it is conceivable that the first step of thermal decomposition 145 involves the dissociation of the neutral 4-ClpzH moiety. Between temperatures of 220°C and 340°C roughly 27% of the mass is released. The neutral 4-ClpzH accounts for 28% of the molecular weight of [Mn(4-Clpz)2(4-ClpzH)]x. Having said this, it is unlikely that the [Mn(4-Clpz)2]x complex could be isolated by removing thermally the neutral pyrazole portion as there is no extensive plateau in the T G A plot of temperature versus weight percent. In other words, immediately following the initial 27% weight loss the remaining material continues to lose mass due to additional decomposition. DSC data collected by increasing the temperature slowly between 30°C and 600°C on the chlorinated derivative showed several poorly defined events with maximum heat flow occurring at roughly 300°C, 350°C, 420°C and 495°C. The endothermic nature of the first two events in the thermal decomposition of [Mn(4-Clpz)2(4-ClpzH)]x is in contrast to the pattern observed for the nickel polymeric complexes in which the halogen-substituted derivatives exhibited an exothermic event upon thermal decomposition. However, the remaining observed events are all exothermic which suggests that these events may, like the nickel complexes, involve some sort of metal halide formation as an intermediate decomposition product. The first endothermic peak (300°C) correlates well with the first step in weight loss (27%) observed in the T G A results. This result is also consistent with the proposed first stage of thermal decomposition being due to the removal of the neutral 4-ClpzH from the complex. Interestingly, the two compounds of formulation [Mn(4-Xpz) 2(4-XpzH)] x (X = CI, Br) showed markedly different stability towards atmospheric dioxygen and moisture. The 146 chlorinated derivative was found to be very stable with respect to atmospheric conditions. [Mn(4-Clpz)2(4-ClpzH)]x is almost pure white in colour and portions of this material were subjected to extended exposure to the atmosphere both as a solid and in wet solvents without any visible changes. More importantly, the C, H and N elemental analysis results were unaffected by these lengthy exposures to dioxygen and moisture. Conversely the brominated derivative discoloured upon relatively brief exposure to air. The initial off-white material would typically turn brown within several hours, and much faster upon exposure to wet solvents. Unfortunately elemental analysis of this brown decomposition material was not obtained. It should be noted here that all attempts to isolate the white product that was formed in sealed-tube reactions between MnCp 2 and 4-HpzH were unsuccessful. The product, possibly [Mn(4-Hpz)2(4-HpzH)]x or [Mn(4-Hpz)2]x, would turn brown during any attempts at purification, either by washing with dry and dioxygen-free solvents in a dinitrogen atmosphere or by removal of the excess 4-HpzH and MnCp 2 by sublimation, followed by exposure to a dinitrogen atmosphere. The unpurified material was also found to decompose instantly (turning a dark brown or black) if exposed to normal atmospheric conditions. The material is evidently extremely sensitive to dioxygen and/or moisture, and appears to decompose spontaneously if removed from excess neutral 4-HpzH. There is no obvious trend in the relative stabilities towards dioxygen and moisture of the two [Mn(4-Xpz) 2(4-XpzH)] x (X = CI, Br) materials and the white product that was not isolated in a pure form from attempts to produce the X = H derivative. It is however possible to speculate at the relatively high stability of the 4-Clpz derivative and even the 4-Brpz 147 complex when compared to the four compounds of formulation [Mn(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI, Br) which will be discussed in more detail later in this chapter. The presence of the one equivalent of coordinated neutral 4-XpzH possibly prevents (X = CI) or reduces (X = Br) the likelihood of attack on the metal centers by dioxygen or moisture by sterically blocking the access to the manganese sites. Again, it is unclear, and possibly counterintuitive, as to why the neutral chlorinated pyrazole would provide better protection than the brominated pyrazole in these materials. The presence of this additional coordinated neutral substituted pyrazole in the 4-Xpz Mn(II) derivatives, but not in the 4-X-3,5-diMepz Mn(II) derivatives, is likely a result of less steric crowding in the former complexes. It is unclear, however, as to why the manganese(II) pyrazolates form these five-coordinated complexes while the nickel(II), cobalt(II) (88), and most of the copper(II) (85-87, 193) systems do not. The particle morphology of [Mn(4-Clpz)2(4-ClpzH)]x was investigated by SEM. Three S E M images of this material are shown in Figure 4.2. The first two reveal the material to be microcrystalline with elongated rigid rod crystallites. The third image reveals that, in some instances, microcrystallites of [Mn(4-Clpz)2(4-ClpzH)]x tend to aggregate in a "disco-dancing" arrangement. 148 (c ) x 5 * 0 k 3 • V Figure 4.2. SEM images of [Mn(4-Clpz)2(4-ClpzH)] represents a length of (a) 5 um, (b) 50 um, and (c) 5 urn. 149 4.2.1.2 X - R A Y DIFFRACTION STUDIES Unfortunately no single crystals suitable for X-ray diffraction studies were obtained for either the brominated or chlorinated polymer. However, a powder diffraction study was performed on [Mn(4-Clpz)2(4-ClpzH)]x and the diffraction pattern is shown in Figure 4.3. The <3?-spacings and relative intensities of the peaks in this diffraction pattern are given in Appendix V, Table V-3. No structural information was determined from the diffraction pattern. However, the presence of the pattern does reveal that the material is microcrystalline, consistent with the S E M images obtained and shown earlier in this chapter. No powder diffraction pattern is available for the air-sensitive [Mn(4-Brpz)2(4-BrpzH)]x as the material decomposed before a pattern could be obtained. 150 JJU W i * * ^ ^ ^ ^ . T i '• i i i i i i r 5 10 15 20 25 30 35 40 45 50 20 Figure 4.3. Powder diffraction pattern for [Mn(4-Clpz)2(4-ClpzH)]x. 4.2.1.3 INFRARED SPECTROSCOPY The infrared data for the [Mn(4-Xpz)2(4-XpzH)]x (X = CI, Br) materials are listed in Appendix III, Table III-6. The spectra revealed an intense and sharp band in the v N H 151 region for both complexes. These bands, at 3347 cm"1 for the X = CI complex and 3337 cm"1 for the X = Br compound, are consistent with the proposed formulation of these materials. The sharpness and intensity of the N - H stretches imply that these protons are not involved in any hydrogen bonding interactions. Additional qualitative evidence for the presence of both 4-Xpyrazoles and 4-Xpyrazolates can be seen in the fingerprint region of the spectra. Approximately twice as many peaks are observed in this region of the spectra than are found in the infrared spectra of free neutral 4-Xpyrazole, coordinated neutral 4-Xpyrazole, or coordinated 4-Xpyrazolate. The number of peaks and peak positions observed for these manganese complexes are also quite similar to those observed in a similar copper(II) complex, [Cu(4-Ipz)2(4-IpzH)0.5]x, isolated by Martin Ehlert (193). Assignments of the vibrational spectra of 4-BrpzH and 4-ClpzH, as well as 4-bromopyrazolate and 4-chloropyrazolate, have been made previously (79). 4.2.1.4 ELECTRONIC SPECTROSCOPY The [Mn(4-Clpz)2(4-ClpzH)]x complex was examined by UV-Vis-NTR spectroscopy in an attempt to obtain some structural information on the [Mn(4-Xpz)2(4-XpzH)]x materials. UV-Vis-NTR spectra were obtained on mulled samples (in Nujol) of various concentrations. From the magnetic data discussed below it is evident that the 152 manganese(II) ions are d 5 high spin in these complexes and thus there should be no "spin-allowed" d to d transitions observed. It was hoped, however, that at high enough concentrations (of mulled samples) some spin-forbidden transitions would be detected and perhaps this would give some insight into the metal chromophore geometries. Unfortunately even at the highest concentrations that were used in this study no transitions were observed other than an intense and very sharp peak at slightly under 3000 cm"1 (3333 nm). This peak is attributed to the N - H stretch which is also observed in the IR spectrum of this complex at 3347 nm. 4.2.1.5 M A G N E T I C PROPERTIES A N D PROPOSED STRUCTURE There are a few possible linear chain structures which should be considered in attempting to account for the proposed empirical formulae and magnetic data (discussed in detail later in this chapter). The first possibility would have all of the manganese(II) ions in a five-coordinate geometry. Representations of the two extreme cases for five-coordinate geometry are shown in Figure 4.4 (square pyramidal) and Figure 4.5 (trigonal bipyramidal). An alternative to this would be a chain in which the geometry about the manganese ions alternates between four coordinate and six coordinate. The two extreme 153 possibilities for this alternating system are shown in Figure 4.6 (octahedral and square planar metal centers) and Figure 4.7 (octahedral and tetrahedral centers). Figure 4.4. Possible linear chain structure of [Mn(4-Xpz)2(4-XpzH)]x (X = CI, Br). In this case the manganese(IT) ions are five coordinate and present in a square pyramidal chromophore geometry. Al l of the potential structures shown are consistent with the empirical formulae of the complexes in question. However, previous examples of square planar manganese(n) complexes possess intermediate spin with a 4 A l g ground state (205). Magnetic data obtained on these complexes (discussed below) are consistent with high spin Mn(II) centers throughout the materials. Any intermediate or low spin square planar manganese(II) centers in the polymers discussed here would significantly decrease the overall magnetic 154 X Figure 4.5. Linear chain structure of [Mn(4-Xpz)2(4-XpzH)]x in which the five-coordinate manganese(II) centers are present in a trigonal bipyramidal M n N 5 arrangement. moment from the observed values of approximately 5.9 B.M. In addition, with the presence of intermediate or low spin d 5 centers, spin-allowed transitions would be expected in the electronic spectra, but none are observed. These observations suggest that the potential structure in which manganese chromophore geometries alternate between square planar and 155 Figure 4.6. Linear chain structure for [Mn(4-Xpz)2(4-XpzH)]x (X = CI, Br). In this case there are alternating octahedral and square planar manganese(II) centers. Figure 4.7. Linear chain structure for [Mn(4-Xpz)2(4-XpzH)]x (X = CI, Br). In this case there are alternating octahedral and tetrahedral manganese(II) centers. 156 octahedral is unlikely for the two [Mn(4-Xpz)2(4-XpzH)]x compounds. On the other hand, many previous examples of five coordinate manganese(II) complexes have been shown to be high spin. There are examples of high spin Mn(II) complexes in which the metal geometry is trigonal bipyramidal, square pyramidal and severely distorted between these two extremes. Specific examples include the high spin trigonal bipyramidal MnBr2(N,N'-dimethylurea)3 (206), the square pyramidal MnI(OPPh 3) 4 (207), and the distorted MnCl 2(2-MeimidH) 3 in which the metal geometry is slightly closer to trigonal bipyramidal than to square pyramidal (208). The precedent for pentacoordinate high spin manganese complexes in a variety of metal ion geometries gives rise to the possibility of similar chromophore geometries in the manganese polymers discussed here. Unfortunately without crystals suitable for X-ray diffraction studies it is not possible to determine the geometry for the [Mn(4-Xpz)2(4-XpzH)]x materials. Magnetic susceptibilities of the two [Mn(4-Xpz)2(4-XpzH)]x (X = CI, Br) compounds were measured from 2 to 300 K using a SQUID magnetometer. The powder magnetic susceptibilities and magnetic moments measured at different temperatures are tabulated in Appendix II, Table II-4. The room temperature magnetic moments of both complexes indicate that the manganese(II) ions are high spin with magnetic moments of 5.93 B . M . and 5.90 B . M . for the X = CI and Br derivatives respectively. The two complexes studied here are magnetically very similar. Both materials exhibit antiferromagnetic coupling between manganese centers. The magnetic moments 157 50 i r 100 150 Temperature (K) 250 300 100 150 200 Temperature (K) 250 300 Figure 4.8. Magnetic susceptibility versus temperature plots for (a) [Mn(4-Clpz)2(4-ClpzH)]x and (b) Mn(4-Brpz)2(4-BrpzH)]x. Experimental points are shown as circles and theoretical fits (calculated as described in the text) are represented as solid lines. decrease significantly with decreasing temperature in both cases. In plots of X M versus temperature (Figures 4.8 (a) and (b)) a susceptibility maximum is observed at 5 K for both the chlorinated and brominated derivative. 158 Having proposed a linear chain structure for these systems, the magnetic data were modeled using the isotropic Heisenberg model for exchange coupled linear chains developed by Weng (202) and Hiller et al (203). For modeling these systems the equation X chain Nghh kT 2.9167+ 208.04(|7|/A:r)2 1 + 15.543(|y|/*r) + 2707.2(| j\/krf [4.1] is used, where all of the terms are the same as described earlier. It should be noted that the value of g is fixed at 2.00 for the calculations discussed here. The resulting fits are in good agreement with the experimental data and are represented as solid lines in the respective plots. The calculated best fit values of J, along with the values of \4JS2\ and the corresponding F values are listed for the two compounds in Table 4.1. Table 4.1. Magnetic parameters for the [Mn(4-Xpz)2(4-XpzH)]x (X = Br, CI) materials. Compound J (cm1) 4JS2 F [Mn(4-Brpz)2(4-BrpzH)]x -0.412(4) 10 0.028 [Mn(4-Clpz)2(4-ClpzH)]x -0.408(4) 10 0.030 159 A curious feature in the experimental magnetic data is the apparent inconsistency between the room temperature magnetic moments obtained and the observed antiferromagnetic coupling in these systems. The presence of antiferromagnetic exchange coupling in linear chains of manganese(II) ions should result in an observed room temperature moment slightly below that expected (5.92 B.M.) for a magnetically dilute high spin Mn(II) ion. As mentioned above the room temperature magnetic moments were calculated to be 5.93 and 5.90 for the X = CI and Br derivatives respectively. To illustrate this discrepancy further, the magnetic parameters obtained by utilizing the model discussed above generate expected room temperature magnetic moments of 5.83 B . M . for each complex. In other words considering the magnitude of the magnetic exchange observed, the measured magnetic moments appear to be approximately 1.5% too high. A 1.5% error in a magnetic moment corresponds to roughly a 3% error in magnetic susceptibility, which is larger than the experimental error typically associated with such measurements (< ± 1%). One possibility, that would account for this 3% error in the magnetic susceptibility, involves a slight variation on the formulation for these complexes. Since the calculated molar magnetic susceptibility is directly proportional to the molecular weight of a material, any error in the molecular weight (used in the calculations) will carry over as the same relative error in the molar magnetic susceptibility. Therefore, if the molecular weight used in these calculations is 3% higher than the actual molecular weight of the material, the calculated molar magnetic susceptibility will also be 3% too high. Thus far, it has been assumed that the formulation of these materials is [Mn(4-Xpz) 2(4-XpzH)i]x. If however, the actual 160 formulation is closer to [Mn(4-Xpz)2(4-XpzH)o.9]x, the molecular weights of these materials, and thus the calculated molecular magnetic susceptibilities, will be approximately 3 percent lower. Interestingly, by utilizing the formulations in which there are only 9 neutral pyrazole moieties for every 10 manganese ions, there is an improvement in the agreement not only between the experimental and theoretical magnetic moments, but in the elemental analyses results as well (Table 4.2). It should also be noted that this adjusted formulation (and reduced molecular weight) does not significantly change the calculated values of /obtained for the two materials. Table 4.2. Comparison of the elemental compositions, molecular weights and calculated magnetic moments for [Mn(4-Xpz) 2(4-XpzH) n] x (X = CI, Br) for two values of n (1, 0.9) with the experimental values. X n C (%) H (%) N (%) M W (g/mol) \x (B.M.)* CI 1 30.0 2.0 22.9 360.69 5.93 0.9 29.8 1.9 23.2 350.11 5.84 experimental 29.4 1.9 23.3 - ** 5.83 Br 1 21.9 1.4 17.1 493.84 5.90 0.9 21.8 1.4 17.0 478.64 5.81 experimental 21.6 1.2 16.8 - 5.83** these magnetic moments are room temperature moments. expected value considering the observed antiferromagnetic exchange in these materials It is useful to compare the magnitude of the magnetic coupling calculated for the [Mn(4-Xpz) 2(4-XpzH)] x (X = CI, Br) systems with the corresponding coupling values obtained for similar substituted-pyrazolate bridged linear chain transition metal systems. 161 However, as discussed earlier, in order to compare systems with different total spin, S, it is appropriate to compare \4JS2\ values (204) instead of simply comparing values of J. Therefore, the \4JS2\ values of these two manganese systems, four additional manganese polymers discussed next, the four nickel complexes described in Chapter 3, and the previously studied substituted-pyrazolates of copper(II) and cobalt(II) studied by Ehlert (69, 85-87, 113, 193) are listed in Table 4.3, along with the corresponding g and F values. As is evident from the table, the magnitude of the coupling in the [Mn(4-Xpz)2(4-XpzH)]x systems is significantly lower than the other systems studied. It is important to note that, with the exception of the [Cu(4-Ipz)2(4-IpzH)o.5]x, none of the other systems contain any neutral substituted-pyrazoles. It is possible that the presence of these additional neutral ligands may weaken the Mn-bridging ligand interaction, and therefore the exchange. The metal chromophore geometry may also influence the extent of the magnetic exchange, but without an X-ray crystallographically determined structure any speculation as to the nature of this effect would be premature. 162 Table 4.3. Comparison of the magnetic parameters for substituted-pyrazolate polymers of Mn(II), Ni(Ii), Cu(II) and Co(II). Material -J (cm1) 4JS2 g F Reference [Mn(4-Clpz)2(4-ClpzH)]x 0.41 10 * 2.00 0.028 This work [Mn(4-Brpz)2(4-BrpzH)]x 0.41 10 2.00* 0.028 This work [Mn(4-H-3,5-diMepz)2]x 1.2 30 * 2.00 0.010 This work [Mn(4-CH3-3,5 -diMepz) 2] x 2.1 53 2.00* 0.005 This work [Mn(4-Cl-3,5-diMepz)2]x 1.5 38 2.00* 0.045 This work [Mn(4-Br-3,5 -diMepz) 2] x 1.6 41 2.00* 0.031 This work [Ni(4-H-3,5-diMepz)2]x 19 76 2.36 0.092 This work [Ni(4-CH3-3,5 -diMepz) 2] x 16 64 2.43 0.028 This work [Ni(4-Cl-3,5-diMepz)2]x 18 76 2.50 0.058 This work [Ni(4-Br-3,5-diMepz)2]x 16 64 2.41 0.027 This work [Cu(4-H-3,5-diMepz)2]x 56 56 2.22 0.023 (87) [Cu(4-CH3-3,5 -diMepz) 2] x 61 61 2.26 0.012 (87) [Cu(4-Cl-3,5-diMepz)2]x 67 67 2.24 0.005 (87) [Cu(4-Br-3,5 -diMepz) 2] x 65 65 2.21 0.014 (87) [Cu(4-Hpz)2]x 81 81 2.12 0.001 (85) [Cu(4-CH 3pz) 2] x 96 96 2.15 0.003 (86) [Cu(4-Brpz)2]x 105 105 2.20 0.004 (86) [Cu(4-Clpz)2]x (green) 104 104 2.13 0.002 (86) [Cu(4-Clpz)2]x (brown) 88 88 2.06 0.001 (86) [Cu(4-Ipz)2(4-I-pzH)0.5]x 82 82 2.33 0.014 (193) [Cu(3-CH 3pz) 2] x 62 62 2.12 0.025 (193) [Cu(indz)2]x 92 92 2.12 0.095 (193) [Co(4-Hpz)2]x 5 45 2.1 0.16 (88) [Co(3-CH 3pz) 2] x 4.4 40 2.29 0.057 (88) [Co(4-H-3,5-diMepz)2]x 3.3 30 2.15 0.062 (88) [Co(4-CH3-3,5-diMepz)2]x 4.4 40 2.24 0.051 (88) [Co(4-Cl-3,5-diMepz)2]x 4.3 39 2.17 0.043 (88) [Co(4-Br-3,5-diMepz)2]x 3.8 34 2.11 0.041 (88) * g values fixed at 2.00 for the manganese complexes 163 4.2.2 MANGANESE(II) 4-X-3,5-DTMETHYLPYRAZOLATES (X = H , C H 3 , Br, CI) 4.2.2.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPERTIES Details of the synthetic procedures are given in Chapter 9, Section 9.2.4.1 through 9.2.4.4. Manganocene was again used as the manganese(II) source for these preparations. In this case the manganocene was mixed with an excess of the appropriate 4-X-3,5-diMepzH (X - H , C H 3 , CI, Br) under a dry dinitrogen atmosphere and this mixture was placed in a Carius tube. Each of the tubes was subsequently sealed under vacuum and heated above the corresponding melting temperature of the ligand. As was the case with the syntheses of the [Mn(4-Xpz) 2(4-XpzH)] 2 (X = CI, Br) materials, the progress of these reactions was difficult to monitor. Again neither the manganocene nor the resulting polymers were soluble in the molten ligand solvent. The polymeric materials produced were visually very similar to the manganocene starting material (off-white powders) and the mixture in the Carius tube appeared not to change throughout the course of the reaction. Largely for this reason the reaction vessels were heated for at least a week before the products were collected and purified. Purification was achieved by removing any excess neutral 4-X-3,5-diMepzH and manganocene by washing with dioxygen and moisture-free solvents (THF, hexanes, benzene), and collecting the remaining insoluble material by filtration. These products were then dried by heating under vacuum for several hours. 164 Consistent with the proposed polymeric nature of these complexes the four products isolated are involatile powders and are insoluble in all common organic solvents. The materials are, however, quite susceptible to normal atmospheric conditions and have to be handled only in the absence of dioxygen and moisture. Upon exposure to either water or oxygen the off-white materials turn brown. No attempts to identify these brown decomposition materials were made. However each of the complexes was exposed to air for approximately five seconds and subsequently re-analyzed for weight percent of carbon, hydrogen and nitrogen. The brief exposure of the [Mn(4-X-3,5-diMepz)2]x (X = H , CH 3 ) complexes to this air resulted in a colour change with the powders becoming noticeably darker (beige-brown). The C, H , and N elemental analyses of the resulting dark beige materials were determined and the weight percentages of the carbon and nitrogen had decreased, as shown in Table 4.4. Conversely, the chlorinated and brominated derivatives showed almost no colour change resulting from the brief exposure to the air. In addition, the C, H , and N weight percentages for these two derivatives did not change significantly following this exposure and, although in both cases the measured C, H , and N percentages were lower than in the initial compounds (Table 4.4), the two sets of measurements for each complex were the same within error. Possible explanations for a reduction of carbon and nitrogen weight percentages in the materials following exposure to air include the formation of metal oxides or hydroxides, and adsorption or coordination of water from the air. It is unclear why the halogenated derivatives appear to be less sensitive to air than the non-halogenated complexes. 165 Table 4.4. Elemental analyses of the [Mn(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI, and Br) complexes before and after brief exposure to the atmosphere. X Element Theory Weight percent without Weight percent after 5 air exposure (%) seconds of air exposure (%) C 49.0 49.03 47.95 H H 5.8 5.65 5.66 N 22.9 23.00 22.47 C 52.8 52.70 51.79 C H 3 H 6.6 6.59 6.48 N 20.5 20.25 19.80 C 38.2 38.54 38.00 CI H 3.9 3.80 3.72 N 17.8 17.68 17.55 C 29.8 30.05 29.81 Br H 3.0 3.04 2.98 N 13.9 14.02 13.92 The thermal stabilities of the [Mn(4-X-3,5-diMepz)2]x materials were investigated by T G A and DSC. Attempts to ensure minimum exposure to the atmosphere were made for each material, but inevitably there was brief exposure of all compounds to dioxygen and water vapour. In all, T G A was performed on three of these materials (X = C H 3 , CI, Br) and DSC was obtained for two of the complexes (X = C H 3 , CI). The T G A results indicate that these polymers lose weight in roughly three or four separate stages as the compounds are heated from room temperature to 800°C. The T G A and DSC plots for [Mn(4-X-3,5-diMepz) 2] x (X = C H 3 and CI) are shown in Figures 4.9 and 4.10 respectively. The actual relative weight changes with respect to increasing temperature are summarized in Table 4.5. Table 4.5. T G A results for the [Mn(4-X-3,5-diMepz)2]x materials. X Temperature range (°C) Weight Loss (%) Weight remaining (%) C H 3 20-80 0 100 80-145 9 91 145-250 1 90 250-330 39 51 330-400 11 40 400-505 15 25 505-800 0 25 CI 20-250 0 100 250-340 58 42 340-440 10 32 440-480 7 25 460-800 0 25 Br 20-75 0 100 75-145 10 90 145-265 2 88 265-350 46 42 350-480 10 32 480-660 10 22 660-800 0 22 167 Figure 4.9. Thermal analysis of [Mn(4-CH3-3,5-diMepz)2]x as monitored by (a) T G A and (b) DSC. Interestingly, in two cases (X = Br, CH 3 ) the material lost roughly 10% of its initial mass as the temperature was raised slowly between 75°C and 145°C. The weight loss over this temperature range is a bit surprising as the reaction temperature for the formation of these complexes was typically over 125°C. The loss in weight over this temperature range is therefore thought to be a result of the brief exposure of these materials to the atmosphere. 168 Figure 4.10. Thermal analysis of [Mn(4-Cl-3,5-diMepz)2]x as monitored by (a) T G A and (b) DSC. It should be noted that the T G A apparatus, although operated with a flow of dinitrogen gas passing over the sample, has no rigorous means to remove dioxygen or water vapour from the sample chamber. Possible scenarios for the loss of weight in this temperature range include loss of adsorbed water (if the complexes are at all hydroscopic) and loss of neutral 4-X-3,5-diMepzH (if a reaction involving an atmospheric constituent displaces the ligands). 169 The next and most significant loss of weight as the temperature was increased further occurred between roughly 240°C and 330°C. Approximately 40% - 50% of the complexes initial weight was lost over this temperature range. The final steps in the thermal decomposition of the materials proved to be the only stages in which the decomposition temperature was dependent on the substituents on the bridging ligands. In each case the final steps resulted in the loss of 15% - 25% of the complexes initial weight. However, the chlorinated complex lost the weight between 335°C and 460°C. The [Mn(4-CH3-3,5-diMepz) 2] x complex lost this weight between 335°C and 505°C and the brominated analogue lost the weight between 350°C and 660°C. As mentioned earlier in the section, DSC was only performed on the X = CI and C H 3 derivatives. The chlorinated derivative shows only one event between 20°C and 400°C; a broad endothermic peak between 350°C and 395°C and centered at 375°C. A second, much more intense (exothermic) event occurs at 495°C. Conversely, the X = C H 3 derivative exhibits four endothermic events between room temperature and 400°C. The first event is a sharp peak centered at 139°C. The second peak is a broad event located between 145 °C and 210°C. The final two events overlap between 335°C and 375°C with local maxima at 340°C and 360°C respectively. The additional events occurring for the X = C H 3 complex may be the result of exposing this material to the atmosphere. As was determined by elemental analysis, the X = C H 3 derivative is particularly sensitive to brief exposure to atmospheric conditions (see Table 4.2). The peak at 139°C may be due to the 170 melting of neutral 4-CH 3-3,5-diMepzH (literature melting temperature = 138°C), which could be released upon the complex reacting with atmospheric gases or moisture. Unfortunately, due to their sensitivity towards atmospheric conditions, powder X -ray diffraction studies were not performed on any of the [Mn(4-X-3,5-diMepz)2]x materials. 4.2.2.2 INFRARED SPECTROSCOPY The infrared data for the [Mn(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI, Br) polymers are listed in Appendix III, Table 111-7. Infrared spectroscopy was again primarily used to ensure that no neutral pyrazole molecules remained in the products. In the four systems discussed here there is an absence of the characteristic N - H stretch expected for neutral pyrazoles indicating that in each case all of the ligand present is in its deprotonated form. The infrared spectra obtained for the four manganese polymers were very similar to those obtained for the analogous nickel complexes discussed earlier. As is the case in the nickel systems, the in-plane Vring band positions resulting from the 4-X-3,5-dimethylpyrazolate ligands occur between 60 - 75 cm"1 lower in energy than those reported for the corresponding non-coordinated pyrazolates (193). This observed shift provides good evidence for the presence of coordinated 4-X-3,5-diMepz anions in these complexes. 171 One notable difference in the IR spectra of the nickel and manganese systems is the presence of a broad band of medium intensity at ~ 610 cm"1 in the spectra of the manganese complexes. IR spectra obtained on brown, partially decomposed, samples of some of the [Mn(4-X-3,5-diMepz)2]x complexes show that this broad peak intensifies with exposure of the polymers to dioxygen and moisture. Unfortunately, these broad bands obstructed the (3C-CH3 bands expected for the pyrazolate ligands, and thus no comparisons could be made with the 3C-CH 3 bands of the free neutral and pyrazoles. Comparisons of the IR spectra obtained on the [Mn(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI, Br) complexes with those obtained on the [Mn(4-Xpz) 2(4-XpzH)] x (X = CI, Br) materials, show that there are significant differences between these systems. Most notably, the spectra obtained on samples of the [Mn(4-Xpz) 2(4-XpzH)] x complexes have approximately twice as many observable peaks in the fingerprint region as do the [Mn(4-X-3,5-diMepz)2]x materials. This observation is not unexpected as the former complexes have both bridging pyrazolate ligands and coordinated neutral pyrazole ligands while the latter complexes have only bridging pyrazolates. 172 4.2.2.3 PROPOSED STRUCTURES A N D M A G N E T I C PROPERTIES Once again, these materials have properties consistent with polymeric structures. The lack of any evidence for capping groups in both the elemental analyses and infrared spectra rule out short chain oligomers. The materials are insoluble, involatile and quite thermally robust, all consistent with polymeric species. The possibility of a trimetallic ring system, similar to the one shown in Chapter 3 (Figure 3.5), can be ruled out on the basis of magnetic data. The magnetic data obtained from these and similar paramagnetic compounds [M(u-pz')2]x (where M = Cu, Co, Mn, N i and u-pz' = 4-Xpz or 4-X-3,5-diMepz (X = H , C H 3 , CI, Br)) show antiferromagnetic coupling between metal centers. The magnetic data eliminate the trimetallic possibility (and any ring system with an odd number of metal centers) as the unpaired electrons on three metal centers linked in the fashion shown in Figure 3.5 cannot all align in an antiparallel manner. The closest such a system can get to complete antiparallel alignment is to have the spins on two of the metal centers aligned in one direction while the unpaired electron spin on the remaining metal center aligns in the opposite direction. Such a situation has been termed "spin frustration" (209) and leads to characteristic magnetic properties not seen in the systems under consideration here. Larger ring systems (with an even number of metal centers) cannot be eliminated as a possibility, but X-ray determined structures of the related copper(II) systems (85, 86) are a good precedent for the infinite linear chain polymer structures proposed It would have 173 been useful to compare powder diffraction patterns of these manganese materials with the previously studied copper, cobalt and nickel systems. However, due to the high sensitivity of the four [Mn(4-X-3,5-diMepz)2]x complexes to air, powder diffraction experiments were not attempted. Magnetic susceptibilities of the four [Mn(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI, Br) compounds were measured from 2 to 300 K on a SQUID magnetometer. The powder magnetic susceptibility and magnetic moment values at various temperatures are tabulated in Appendix II, Table II-5. The room temperature magnetic moments of the complexes indicate that the manganese(II) ions are high spin with magnetic moments of 5.75 B . M . (X = H), 5.47 B . M . (X = CH 3 ) , 5.85 B . M . (X = CI) and 5.74 B . M . (X = Br). The 2:1 ratio of the substituted bridging pyrazolate ligands to manganese centers gives rise to a four coordinate metal chromophore for these systems. The two extreme possible metal chromophore geometries are therefore square planar and tetrahedral. As was mentioned earlier in this chapter, square planar manganese is relatively rare. Additionally, the manganese ions in these materials are high spin and no high spin square planar manganese compounds have been reported to date. This evidence, along with the molecular modeling studies done on the analogous nickel complexes (described in Chapter 3, Section 3.2.2.4), strongly supports a tetrahedral or distorted tetrahedral manganese chromophore geometry for these materials. 174 All of these materials exhibit antiferromagnetic coupling between manganese centers. In three of the complexes an obvious maximum can be seen in the x versus temperature plots. These plots are shown in Figure 4.11 (a), (b), and (c), and the maxima occur at 13 K, 25 K and 20 K for the X = H, C H 3 and Br derivatives respectively. The maximum in the x versus temperature plot for the [Mn(4-Cl-3,5-diMepz)2]x material (Figure 4.11 (d)) is more difficult to see as there is a relatively large amount of paramagnetic impurity in this complex. There is, however, an inflection point in the plot at approximately 20 K which can be used for a basis of comparison with the maxima found in the plots of the other three samples. The magnetic data were modeled using the isotropic Heisenberg model for exchange coupled linear chains described earlier in equation [4.1]. However, the four complexes considered here were observed to contain varying amounts of paramagnetic impurity. In order to generate better agreement between experimental data and modeled fits, this impurity was assumed to obey Curie-law paramagnetism and the equation calc (i-p) 2 .. 2 kT 2.9161 + 208.04(| J\/klf 1 + 15.543(|y|/ArT) + 2707.2^j\/kTf f + p J V Ng2MBS(S + \) 3kT [4.2] where the parameters are the same as were described for equation [3.6] and [3.8], was used to fit the data. As was the case for the [Mn(4-Xpz)2(4-XpzH)]x materials, g was set to 2.00 and the only parameters allowed to vary were J and P. The resulting fits are represented in Figure 4.11 (a-d) as solid lines. Calculated values of J and P, along with the corresponding F values, are listed in Table 4.6. 175 Figure 4.11 continued overleaf 176 0.25 *f 0.20 -\ 3 0.15 H 0.10 H 0.05 H 0.00 50 100 150 Temperature (K) 200 250 300 50 150 200 Temperature (K) 250 300 Figure 4.11. Powder magnetic susceptibility plots for the [Mn(4-X-3,5-diMepz) 2] x materials: (a) X = H, (b) X = C H 3 , (c) X = Br, and (d) X = CI. The circles are the experimental data and the calculated values (as described in the text) are represented by the curved lines. Parameters for the calculated susceptibilities are those in Table 4.5. 177 Table 4.6. Magnetic parameters for the [Mn(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI, Br) materials. Compound J (cm1) 4JS2 (cm"1) P F [Mn(4-H-3,5 -diMepz) 2] x -1.206(8) 30 0.0097(8) 0.023 [Mn(4-CH3-3,5 -diMepz) 2] x -2.12(2) 52.5 0.0051(5) 0.028 [Mn(4-Cl-3,5-diMepz)2]x -1.53(3) 37.5 0.090(3) 0.045 [Mn(4-Br-3,5-diMepz)2]x -1.64(1) 40 0 (fixed) 0.031 At this point it is useful to compare the magnitude of magnetic exchange as a function of both the metal ion and bridging ligand in the transition metal azolates described thus far. Comparisons can also be made with the analogous systems of the previously prepared copper(II) and cobalt(II) materials. As mentioned earlier, for meaningful comparisons between systems with differing total spin, the values of \4JS2\ for each material should be used as a measure of the extent of magnetic coupling. The values of \4JS2\ have therefore been tabulated (Table 4.7) for the nickel and manganese polymers discussed in this dissertation and the closely related copper(II) and cobalt(II) polymers studied previously (85-87). 178 Table 4.7. \4JS2\ values for polymeric manganese(II), nickel(II), cobalt(II) and copper(II) systems in which pyrazolate or substituted pyrazolate ligands bridge the metal centers. \4JS2\ values shown here have been calculated for data over the 2 to 300 K temperature range. Pyrazolate 4-Clpz Metal Cu(II) (green) Cu(II) (brown) Mn(II)* 4JS2 (cm"1) 99 90 10 4-Brpz Cu(II) 106 Mn(II)* 10 4-H-3,5-diMepz Cu(II) 56 Co(II) 30 Mn(II) 30 Ni(II) 76 4-CH3-3,5-diMepz Cu(II) 61 Co(II) 40 Mn(II) 52.5 Ni(II) 64 4-Cl-3,5-diMepz Cu(II) 67 Co(II) 39 Mn(II) 37.5 Ni(II) 76 4-Br-3,5-diMepz Cu(II) 65 Co(II) 34 Mn(II) 40 Ni(II) 64 These two polymers are five-coordinate, with one neutral pyrazolate per metal center. As can be seen, the magnitude of the magnetic exchange is not vastly different for the polymers of form [M(4-X-3,5-diMepz)2]x in the four metals systems studied thus far. It 179 does however appear that the magnetic exchange observed for the manganese and cobalt systems is slightly weaker than that for the copper and nickel complexes. Another interesting observation can be made in regards to the effect of substitution at the 4-X position on the pyrazolate ring. It appears that varying the nature of X has no significant effect on the magnitude of the magnetic exchange. 4.3 MANGANESE(II)TRIAZOLATE 1,2,4-triazolate has the ability to bridge metal centers in more orientations than either pyrazolate or imidazolate (described in Chapter 6). In fact, both the M-N-N-M linkages (seen in binary transition metal pyrazolate compounds) and the M-N-C-N-M linkages (imidazolate complexes) are possible for metal triazolate complexes (Figure 4.12). Figure 4.12. Three different possibilities for a triazolate ion bridging metal centers. (A) is analogous to pyrazolate bridging, (B) is similar to imidazolate bridging, and (C) represents the case in which both types of bridging occur simultaneously. 180 Transition metal complexes of 1,2,4-triazole and 1,2,4-triazolate based ligands have been extensively studied. In oligometallic and polymeric complexes triazole most often acts as a bridging ligand through its N - l and N-2 atoms (210-234). Structurally characterized examples of triazole bridging two metal centers through its N - l and N-4 positions are much rarer (235, 236), as are cases in which the triazolate ligand bridges three metal centers through its three nitrogen atoms (237-240). There have been a few reports published on structurally characterized polymeric metal triazole complexes. Among the first to be studied by single crystal X-ray crystallography was [Cu(trzH)Cl2]x. This material was originally described by Paolini and Goria in 1932 (241), and the unit cell dimensions were published by Sanero in 1936 (242). In 1962, the crystal structure was reported by Jarvis, and the compound was shown to consist of linear chains of copper(II) centers linked by single triazole (N- l , N-2), and double chloride bridges (211). This material has subsequently been shown to exhibit antiferromagnetic coupling between copper(II) ions in the chains (243-246). The related [Cu(trzH)Br2]x material was also shown to exhibit antiferromagnetic exchange coupling (245), but the structure of this complex has not yet been reported. In 1987, the crystal structure of a copper(II) linear chain polymer in which 4-amino- 1,2,4-triazole groups triply bridge the metal centers was reported (213). More recently a series of metal triazole compounds of formulation [M(trzH)X 2] x ( M = Fe, Co, Ni , Cu; X = CI, Br) and 181 [M(trzH) 2X 2]x ( M = Mn, Co, Ni , Cu; X = CI, Br) has been characterized by electronic and magnetic data (247). Polymeric structures in which triazole ligands bridge the metal centers have been proposed for these materials. The iron, cobalt and copper compounds all exhibit antiferromagnetic interactions. However, the [Ni(trzH)X 2] x complexes show a ferromagnetic exchange coupling (247). Another linear chain compound in which triazole groups bridge metal centers has been structurally characterized by single crystal X-ray diffraction studies. [Cd2(NCS) 4(butrzH)3]x (butrzH = 4-t-butyl-l,2,4-triazole) consists of a chain of cadmium centers with alternating bridges of three butrz groups and two N-bonded isothiocyanate anions (248). Recently the structures of the 1-D linear chain polymeric iron compounds, [Fe(trzH)2(trz)](BF4) and [Fe(trzH)3](BF4)2, have been shown to consist of linear chains of iron(II) centers triply bridged by triazole and triazolate groups (219). Examples of structurally characterized 2-D polymeric metal triazole compounds have also been reported. In 1979, Engelfriet et al. published the structure of a sheet polymer in which each triazole ligand bridges two metal centers through its NI and N4 nitrogen atoms (235). The first and only report of a structurally characterized polymeric complex in which triazolate ligands bridge metal centers through all three nitrogen atoms appeared in 1995 (237). The compound obtained, [Zn(trz)Cl]x, consists of tetrahedral zinc(II) ions singly-bridged to three other metal centers resulting in a 2-D sheet. Chloride ions occupy the fourth coordination site on each zinc center. 182 There have also been a number of reports on oligometallic metal triazole complexes. Most common are linear trimetallic compounds in which triazole, or C- or N-substituted triazole groups triply-bridge the metal centers (217-230). Other reports have been made on dimetallic systems triply-bridged by triazole ligands (230-234). Many of these compounds have been studied magnetically and several reveal antiferromagnetic exchange coupling between the metal centers (226-233). Other oligometallic complexes in which triazole groups link metal centers include several mixed bridged trimetallic materials (214-216), and dimetallic compounds in which dinucleating functionalized triazole groups bridge the metal centers (249-251). Polymeric metal complexes with dinucleating functionalized triazole bridges have also been reported (252). There has been considerable interest of late in iron(II)-1,2,4-triazole (and triazolate) spin transition molecular materials (212, 219-223, 253). Many of these systems exhibit abrupt spin transitions upon heating or cooling of the material. Such materials are potentially very useful as components of memory devices, especially when a thermal hysteresis loop is observed near room temperature. The most spectacular examples of this effect published to date involve the spin transitions observed in the polymeric [Fe(trzH)2 8 5(4-NH2trz)o.i5](C104)2«H20), [Fe(trzH)2(trz)](BF4) and [Fe(trzH)3](BF4)2 complexes (220, 222). In the current work, the polymeric complex of formulation [Mn(trz)2]x has been prepared by reacting manganocene with excess molten 1,2,4-triazole. Although the 183 structure of this material has not been determined, the physical and thermal properties, spectroscopic data and magnetic behaviour of this complex are described. 4.3.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPERTIES Details of the synthesis can be found in Chapter 9, Section 9.2.4.7. The reaction is analogous to the ones used for the preparation of the manganese pyrazolates discussed above. An excess of 1,2,4-triazole was mixed with manganocene and a small amount of xylenes in a dry and dioxygen-free Carius tube. The tube was sealed under vacuum in the usual fashion and was placed in an oven (140°C) for approximately 3 days followed by additional heating at 180°C for two days. At the elevated temperatures the triazole either melted or completely dissolved in the xylene solution resulting in a mixture of solid manganocene and molten, or dissolved, azole. As with the manganese pyrazolate reactions, this reaction was difficult to monitor due to the similarities in appearance between the reactant, manganocene, and the polymeric product. One difference, however, was the presence of a slight orange tint in the liquid phase, the intensity of which decreased as the 184 reaction continued. After the intensity of the orange colour in the liquid phase had nearly completely vanished (5 days), the tube was cooled and subsequently opened under a dry dinitrogen atmosphere. The excess triazole was removed by washing with acetonitrile and the light beige powder product was collected by filtration. The material is insoluble in all solvent systems attempted, involatile, and thermally very robust. However, like many of the manganese pyrazolate polymers the compound is sensitive to dioxygen or water. The thermal stability was investigated by T G A and DSC. T G A results indicate that the complex is stable, with no weight loss observed, up to 360°C. As the temperature is increased further, the compound loses mass in a single step until roughly 40% of the initial mass remains at 525°C. The mass remains constant as the temperature is increased to the maximum studied value of 800°C. The 40% of the initial mass remaining above 525°C is roughly 10% more than the manganese content alone. The DSC results are consistent with the T G A data as only one event is observed, centered at 465°C. The event is endothermic (AH = 121 kJmol"1) and spans approximately the same temperature range (360°C - 500°C) in which the complex loses mass. 185 4.3.2 INFRARED SPECTROSCOPY Band frequencies and relative intensities of [Mn(trz)2]x are listed in Appendix III, Table III-8. The characteristic N - H stretching bands of neutral triazole are absent indicating that all of the ligand present is in its deprotonated form. 4.3.3 M A G N E T I C PROPERTIES AND PROPOSED STRUCTURE Variable temperature magnetic susceptibilities (2 to 300 K) were measured on a SQUID magnetometer. The magnetic susceptibility and moment data are listed in Appendix II, Table II-6. The temperature dependence of the magnetic susceptibility and moment are shown in Figure 4.13. The magnetic moment decreases with temperature, consistent with antiferromagnetic exchange between the high spin Mn(II) centers. Attempts to fit the data to linear chain S=5/2 models (one of which was described earlier in this chapter), and to models developed for two dimensional S=5/2 systems yielded unsatisfactory results. Without knowing the structure it is not possible to explain this lack of agreement with theory. 186 Figure 4.13. Magnetic susceptibility and moment plots for [Mn(trz)2]x. The magnitude of the exchange can, however, be qualitatively analyzed by visual inspection of the plots shown. Since no maximum is observed in the magnetic susceptibility data above 2 K, (the lowest temperature studied here) the magnitude of the magnetic exchange is likely weak. Recall that a maximum in the susceptibility at ~ 5 K (for the [Mn(4-Xpz)2(4-XpzH)]x materials) resulted in a calculated J value of -0.41cm"1. It is therefore probable that the coupling constant, J, for the [Mn(trz)2]x system is lower than this value. However, since the structure of [Mn(trz)2]x may be radically different from that 187 of the linear chain materials described above, not much weight should be put into comparisons of this sort. Due to the relatively large number of ways the triazolate ligand can bridge metal centers (Figure 4.12), proposing a structure for this complex is not straightforward. No evidence regarding the M n N n (n = 4-6) chromophore geometry could be obtained from U V -Vis-NTR as no transitions were observed at even the highest mulled concentrations attempted. [Mn(trz)2]x is however relatively air stable and considerably more thermally robust than the other manganese pyrazolate polymers. These properties suggest that this material is different from the 1-D linear chain manganese(II) polymers. As is described in more detail in Chapter 7, the presence of three potentially donating nitrogen atoms on the triazolate ring make two or three dimensional structures a distinct possibility. 4.4 S U M M A R Y A N D CONCLUSIONS A series of manganese(II) azolate polymers has been prepared by reacting manganocene with appropriate azoles at elevated temperatures. The complexes have been characterized and the magnetic behaviour of these systems has been investigated. Although no crystals suitable for single crystal X-ray diffraction studies were obtained, there is good 188 evidence that the manganese(II) pyrazolate complexes are 1-dimensional chain materials with double pyrazolate bridges linking adjacent metal centers. The compounds in which the 4-X-3,5-dimethylpyrazolates (X = H, C H 3 , CI, Br) bridge the manganese centers are thought to be analogous to the nickel(II) chain polymers described earlier. However, the complexes that formed upon reacting the less sterically bulky 4-Xpyrazolates (X = CI and Br) with manganocene yielded polymers of formulation [Mn(4-Xpz) 2(4-XpzH)] 2. The presence of the coordinated neutral 4-XpzH ligands on these systems is thought to be responsible for a higher stability towards dioxygen and moisture. A reaction between 1,2,4-triazole and manganocene produced a polymer of formulation [Mn(trz)2]x. Evidence suggests that [Mn(trz)2]x in not analogous to the metal pyrazolate linear chain polymers. The presence of three potential donor nitrogen atoms on the triazolate ring make more complicated 2-D or 3-D structures possible for this material. Studies of the magnetic properties of the six manganese pyrazolate systems indicate the presence of antiferromagnetic coupling between the Mn(II) centers in the chains. The two systems in which neutral pyrazolates were coordinated to the metal centers ([Mn(4-Xpz) 2(4-XpzH)] x) exhibited considerably weaker magnetic exchange coupling than the four polymers of formulation [Mn(4-X-3,5-diMepz)2]x. The calculated J values for the four [Mn(4-X-3,5-diMepz)2]x complexes ranged from -1.2 to -2.1 cm - 1. The \AJS2\ values for these Mn(II) compounds are comparable to those for the nickel(II) (and other metal(II)) pyrazolate polymers suggesting that the magnitude of exchange is not significantly 189 influenced by the nature of the metal ions present. The weaker antiferromagnetic exchange observed for the [Mn(4-Xpz)2(4-XpzH)]x complexes indicates that the coordination about the metal center can significantly affect the magnitude of the exchange coupling in these systems. The magnetic properties of [Mn(trz)2]x suggest antiferromagnetic coupling is present in this system. Finally, manganocene has been shown to be a good source of Mn(II) ions for preparing divalent manganese azolate polymers. 190 Chapter 5 A MIXED VALENCE COPPER 3,5-BIS(TRIFLUOROMETHYL)PYRAZOLATE COMPLEX 5.1 INTRODUCTION As discussed earlier, copper(II) pyrazolate and substituted pyrazolates have been shown to adopt structures in which the copper ions are doubly bridged by pyrazolate ligands to form extended linear chain polymers. The compounds show relatively strong antiferromagnetic coupling with values of J, the exchange coupling constant, in the range -58 to -105 cm"1. These values vary with the substitution on the pyrazolate bridging ligands and attempts have been made to correlate structural variations with these numbers. An attempt to extend these previous studies to include the 3,5-bis(trifluoromethyl)pyrazolate, 3,5-F6diMepz, bridging ligand, with highly electron withdrawing trifluoromethyl groups on the heterocyclic rings was made. The anticipated polymeric species, [Cu(3,5-F6diMepz)2]x, was not obtained but instead a dark green mixed Cu(I)/Cu(II) compound, Cu3(3,5-F 6diMepz) 5, was isolated. An X-ray determined structure of this complex shows the molecular unit to consist of a triangular arrangement of three copper ions held together by bridging pyrazolate ligands. The magnetic behaviour of the green complex over the 2 to 300 K temperature range was also examined. 191 Upon prolonged exposure to moist air the green material slowly changes in colour to a bluish purple. This material is compositionally quite similar to the parent green complex, but has vastly different magnetic properties. 5.2 RESULTS A N D DISCUSSION 5.2.1 [Cu3(3,5-F6diMepz)5] 5.2.1.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPERTIES Detailed descriptions of the preparation of this compound can be found in Chapter 9, Section 9.2.5.2. The mixed valence copper complex described here was isolated during attempts to produce the 3,5-F6diMepz analogue to a series of copper(II) pyrazolate polymers previously prepared in the laboratory by M . K. Ehlert (87). Attempts to produce the desired polymeric species were made utilizing several different synthetic routes, but this target polymer has yet to be prepared. The mixed valence Cu3(3,5-F6diMepz)5 was prepared by heating a mixture of clean copper shot with 3,5-F6diMepzH under a dry dioxygen atmosphere and at a temperature above the melting point of the neat azole (79-81°C). The 3,5-F6diMepzH is very volatile above its melting point and thus from time to time it was necessary to reintroduce the material to the reaction by physically scraping or melting the sublimed azole out of the condenser and back into the reaction flask. It was 192 determined that the easiest way to isolate the pure trimetallic compound was to continue heating the reaction mixture, once sufficient product has formed, until all of the unreacted 3,5-F6diMepzH has sublimed away from the remaining copper shot and green product mixture. Once the excess ligand precursor was removed the green compound was physically separated from the relatively large pieces of copper metal in the reaction vessel. The green compound is soluble in dry solvents with a wide range of polarities. Cu3(3,5-F6diMepz)5 melts at 236°C without decomposition and is appreciably volatile at this temperature even at 1 atmosphere. Initially it was thought that the compound was quite stable as a solid under atmospheric conditions. The colour of the product appears not to change, even after weeks of exposure to dioxygen and moisture. In addition, the elemental analysis results are unaffected by these lengthy exposures to normal atmospheric conditions. However, after several months of exposure to moist air, Cu3(3,5-F6diMepz)5 slowly changes into a purple material, which is discussed in more detail later in this chapter. Interestingly, this purple material has elemental analysis results nearly identical to those of the parent green complex, and thus elemental analysis cannot be used to determine the purity of the green compound. This purple compound can be formed more rapidly by dissolving the green complex in a wet, non-coordinating solvent, the initial green colour of the solution turning blue with the deposition of a purple solid. 193 5.2.1.2 X - R A Y DIFFRACTION STUDIES Macroscopic crystals of Cu3(3,5-F6diMepz)5 can be obtained either directly from the reaction mixture, by sublimation at 1 atmosphere or from a saturated solution in dry CHC1 3 in an evacuated tube. Samples of crystals were sent to V. J. Young Jr., of the X-ray Crystallographic Laboratory at the University of Minnesota, and the structure was determined. Crystallographic data, atomic coordinates, bond lengths and bond angles are given in Appendix I, Tables 1-6 through 1-9. The molecular arrangement and atom numbering scheme is shown in Figure 5.1 and an ORTEP diagram is presented in Figure 5.2. The X-ray crystal structure determination shows the three copper ions (2 Cu(II) and 1 Cu(I)) to be arranged in an isosceles triangle with each side of the triangle being occupied by an exobidentate 3,5-F6diMepz ligand. The 2 Cu(II) ions, which form the short base of the triangle, are bridged by two additional 3,5-F6diMepz ligands, one above and one below the trigonal planar array. The structural array adopted is similar to those arrangements found in the all-Cu(I) trimers of formula [Cu(4-CH3-3,5-diMepz)]3 and [Cu(4-H-3,5-diMepz)]3 (63, 69) with the significant difference arising from the two types of Cu ions found in the present structure. These differences are exemplified by comparing the non-bonded Cu w *Cu distances in the trigonal structures. 194 Figure 5.1. The molecular structure of Cu3(3,5-F6diMepz)5, with atom numbering scheme. In the compound discussed here the unique Cu(II)'"Cu(II) distance is 2.975 A (sum of van der Waals radii for Cu is 2.86 A (254)) with the Cu(II)---Cu(I) distances being much longer at 3.236 and 3.250 A. In the all-Cu(I) trimer species, [Cu(4-CH3-3,5-diMepz)]3 (69), the Cu---Cu distance is 3.212 A and in [Cu(4-H-3,5-diMepz)]3 (63) the distance is 3.218 A. Two other structures having the basic central trigonal planar array of Cu ions bridged by 195 Figure 5.2. The molecular structure of Cu3(3,5-F6diMepz)5, with 50% probability thermal ellipsoids shown for all non-hydrogen atoms. pyrazolate ions have been reported. Both are all-Cu(II) systems. In [Cu3(OH)(4-Hpz)3py2Cl2]py (255) there is one unique Cu^ 'Cu distance at 3.112 A and two Cu---Cu distances at 3.321 A. In this structure there is significant deviation from planarity displayed by the Cu 3(4-Hpz) 3 trigonal core. On the other hand, the central feature of the complex [Cu 3(OH)(4-Hpz) 3(4-HpzH) 2(N0 3) 2]H 20 (256) is essentially planar with Cu---Cu distances at 3.324, 3.364 and 3.365 A. The present structure is completed by two additional bridging 3,5-bis(trifluoromethyl)pyrazolate, 3,5-F6diMepz, ligands above and below the trigonal 196 planar array and bridging the 2 Cu(II) ions. These ligands are tilted towards the Cu(I) ion thereby relieving steric interactions with the pyrazolate ion already bridging the two Cu(II) ions in the trigonal plane. The angle between the least squares planes of these two bridging groups is -154° (see least-squares planes given in the Appendix). The coordination about the Cu(I) ion is almost linear (N(52)-Cu(3)-N(41)) at 176.2° whereas the Cu(II) ions display a very distorted tetrahedral environment, (N(42)-Cu(2)-N(31), 171.2°, N ( l l ) -Cu(2)-N(21), 121.7°, N(31)-Cu(2)-N(ll), 89.1°, N(31)-Cu(2)-N(21), 89.3°, N(42)-Cu(2)-N ( l l ) , 94.7°, N(42)-Cu(2)-N(21), 94.3°). The non-bonded Cu ' "Cu distances of the present structure may be compared also with those found in the Cu(II) pyrazolate polymer, [Cu(4-Hpz)2]x. In the latter structure (85) the Cu(II) ions are in a distorted tetrahedral environment created by the two bridging pyrazolate ligands between the Cu centers but the overall linear chain structure allows a much longer Cu*"Cu non-bonded distance of 3.889A The 2 Cu(II) ions of the present structure are in a constrained trigonal environment with the Cu(I) ion and are essentially triply bridged by 3,5-F6diMepz ligands. The coordination sphere of each of these Cu(II) ions is completed by the 3,5-F6diMepz ligands being shared with the Cu(I) ion. 197 5.2.1.3 SPECTROSCOPIC STUDIES 5.2.1.3.1 INFRARED SPECTROSCOPY Infrared data are tabulated in Appendix III, Table III-9. The IR data confirmed that no neutral 3,5-F6diMepzH remained in the product, bands attributable to the N - H function of the 3,5-F6diMepzH starting material being notably absent from the spectrum. 5.2.1.3.2 ELECTRONIC SPECTROSCOPY The spectrum consists of an asymmetric absorption band with a maximum at ~ 14,000 cm"1 with a shoulder to higher wave numbers. According to Hathaway (257), this absorption is consistent with a compressed tetrahedral CuN 4 chromophore geometry. This is therefore consistent with the structure determined by single crystal X-ray crystallography. 198 5.2.1.3.3 M A S S SPECTROMETRY The low resolution electron impact ionization spectrum of Cu3(3,5-F6diMepz)5 was obtained using a KRATOS MS50 mass spectrometer. The spectrum is shown in Appendix IV, Figure IV-1, and exhibits a parent ion peak, P + , at m/e -1206 along with strong signals due to ions arising from consecutive loss of two 3,5-F6diMepz ligands, L , from the parent species at m/e -1003 (P-L +) and m/e -800, (P-2L+). The intensities of the individual lines in these multi-line signals are those expected from the presence of three Cu atoms and the naturally occurring isotopic abundance for the element. 5.2.1.4 M A G N E T I C PROPERTIES Magnetic susceptibilities were measured using a SQUID magnetometer. Susceptibility versus temperature behaviour (2 to 300 K) was measured at 10 000 Gauss and 20 000 Gauss. The powder magnetic susceptibility and magnetic moment versus temperature data for two runs at 10 000 G and the one at 20 000 G are tabulated in Appendix II, Tables II-7 and II-8 respectively. The presence of two copper(II) ions triply-bridged by pyrazolate ligands leads to the possibility of strong antiferromagnetic coupling between the two paramagnetic centers. Magnetic susceptibility measurements on powdered samples of Cu3(3,5-F6diMepz)5 revealed some field dependence over the entire temperature range studied. Due to this field 199 1.5e-3 0 10000 20000 30000 40000 50000 60000 Field (Gauss) Figure 5.3. Magnetization plots for Cu3(F6dmpz)5 at 300 K (A), 100 K ( • ) and 50 K (O). The lines have been calculated using the data above 10 000 G and were used to determine the magnetization intercept at zero applied field. dependence, magnetization versus field data were collected at four different temperatures. Magnetization measurements at applied fields ranging from 0 to 55 000 Gauss were made at 2, 50, 100 and 300 K, and the data is tabulated in Appendix II, Table II-9. These studies revealed linear magnetization versus field behaviour above -3000 G at all temperatures as shown in Figure 5.3. Extrapolation of the linear regions to zero field gave a non-zero 200 intercept indicative of ferromagnetic impurity, the presence of which made a temperature and field independent contribution to the susceptibility data. In order to determine accurately the magnetic susceptibilities arising from the pure green complex, the magnetic contribution from this ferromagnetic impurity had to first be subtracted. Fortunately, the magnetization versus field plots, at all temperatures studied, showed linear behaviour above 3000 Gauss with a common magnetization intercept when these linear portions were extrapolated to zero field. This suggested that any magnetic data collected above 3000 G would have a fixed magnetic contribution from the impurity. This contribution could then be subtracted from the overall magnetic signal in a similar manner as corrections are made for the sample holders. Correcting the data for this impurity gave field independent susceptibilities. The source of the ferromagnetic impurity is not certain; however, as described above, Cu3(3,5-F6diMepz)5 is very susceptible to moisture and traces of impurities should not be unexpected. The magnetic susceptibilities (per mole of copper(II)) are plotted versus temperature for three runs in Figure 5.4. The susceptibility decreases from 300 K to about 80 K, then at lower temperatures shows an increase as the temperature is decreased further. This is consistent with strong antiferromagnetic coupling between the copper(II) centers with the upturn in susceptibility at the lowest temperatures arising from trace paramagnetic impurity. To obtain a measure of the strength of the magnetic coupling the data obtained (for run #2 at 10 000 G) was fitted over the entire temperature range to the equation for 201 0 50 100 150 200 250 300 Temperature (K) Figure 5.4. Magnetic susceptibilities (per mole of copper(II)) versus temperature plot. • , A, and O, are data points for run #1 at 10 000 Gauss, run #2 at 10 000 Gauss and one run at 20 000 Gauss, respectively. The solid line is calculated from theory as described in the text. 202 antiferromagnetic dimeric copper(II) systems developed by Bleaney and Bowers (258), with allowance for a paramagnetic component and for temperature independent paramagnetism, TIP. The equation used is: kT X = 2J/kT [l-p] + P/4 _ l + 3e 2 y / k T_ + TIP where J is the exchange coupling constant. This gave best fit values of -J, g, P and TIP of 235(2) cm"1, 2.112(6), 0.00245(6) and 29(1) xlO"6 cm'mol"1, respectively, with the fitting function F = 0.063. The function F was defined earlier (Chapter 3, Equation [3.9]). The value of TIP is somewhat low for copper(II); however it must be realized that the strong coupling in this compound renders the paramagnetism very weak. As a consequence diamagnetism and TIP contribute a significant amount to the overall susceptibilities, each making a temperature independent contribution. Diamagnetic corrections were estimated in the usual way (259) and an estimate only 10% greater than the values used here would, for example, not affect the quality of the fit nor the other parameters but would lead to a 75% increase in TIP to a value of 51 x 10"6 cm3mol"1. The value of -J is significantly greater than that observed in the compounds that have doubly 203 pyrazolate bridged extended chain structures (58 to 105 cm"1) and this may be a consequence of the fact the metal centers are triply-bridged in the compound studied here. More relevant, perhaps, are comparisons with dimetallic copper(II) systems. Structural and magnetic studies have been reported on copper(II) complexes containing binucleating ligands and incorporating singly bridging pyrazolate (54, 140). These compounds, however, also contain other bridging groups which are involved in the exchange, making comparisons with the present system inappropriate. Finally we refer to a class of compounds that contain dinucleating pyrazolate ligands which incorporate chelating arms in the 3,5-positions on the pyrazolate ring (177, 260, 261). In these compounds two copper(II) centers are .bridged by two ligands; each ligand provides one pyrazolate bridge (di-|j,-pz bridging) and chelates to each metal via the substituents in the 3,5-positions. The metal-ligand geometry in these systems is square pyramidal and the values of the exchange integral, J, vary from -159 cm"1 to -214 cm"1. Hence, for these systems the coupling approaches values of the same order of magnitude as that observed for the title compound in spite of the structural differences, including the number of bridging pyrazolate ligands. 204 5.2.2 STUDIES O N THE PURPLE DECOMPOSITION PRODUCT 5.2.2.1 PREPARATION, PHYSICAL A N D T H E R M A L PROPERTIES Prolonged exposure of the mixed Cu(I)/Cu(II) complex Cu3(3,5-F6diMepz)5 to moisture results in the formation of a purple powder. This purple material can be prepared intentionally, and much more rapidly, by dissolving the green parent compound in wet (H 2 0 containing) organic solvents. In one preparation wet toluene was used resulting in an initially green solution due to the dissolved trimetallic complex. The solution turned blue within a few minutes and a purple solid deposited from the liquid. After several hours the colour in the solution became significantly less intense and the purple solid precipitate was collected by filtration. As mentioned earlier, the purple material has a elemental composition nearly identical to that of the parent green complex. The elemental analysis results are the same, within error, as those for Cu3(3,5-F6diMepz)5. This purple powder is however remarkably different, physically, from the green complex. The purple material is air-stable, involatile at temperatures below 100°C, and insoluble in most common organic solvents and water. However, heating the purple solid above 100°C causes some of the purple material to apparently sublime and convert back into the green form. This combination of properties make characterization of the purple material extremely difficult. The insolubility and involatility hamper any attempt to grow X-ray quality single crystals of the material. The 205 tendency for the material to at least partially covert back into the green form upon heating at high temperatures results in T G A and DSC analyses that are drastically inconsistent. This inconsistency is thought to result from the high volatility of the green material, to which the purple material seemingly converts, at temperatures above 100°C. This same tendency for the purple material to change back into the green form upon heating can also be seen in the mass spectrometry results described later. Due to the insolubility of the purple material, the only way to obtain a mass spectrum is to volatilize the solid. All attempts to obtain a mass spectrum of the purple complex by this process yielded results identical to the mass spectra obtained on samples of the pure green material. Infrared and electronic spectroscopy results are however different for the purple and green materials and are discussed next. 5.2.2.2 SPECTROSCOPIC STUDIES 5.2.2.2.1 INFRARED SPECTROSCOPY The infrared data are listed in Appendix III, Table III-10. The bands attributed to the N - H function of the 3,5-F6diMepzH starting material were notably absent from the spectra of the purple product. The most dramatic difference between the infrared spectra of the green and purple materials is a sharp absorption found at 3680 cm"1 in the spectrum of the purple material. This band most likely arises from coordinated non-hydrogen bonded 206 OH groups (262). The sharpness and location of this band suggest an absence of hydrogen bonding. Other copper(II) polymers that have both bridging pyrazolate ligands and bridging O H groups have previously been reported (255, 256). Angaroni et al. report a broad peak at 3240 cm"1 attributed to an overlap of the OH an N H stretching vibrations in the IR spectrum of the distorted trimetallic [Cu3(OH)(4-Hpz)3(4-HpzH)2(Cl)2]solv (solv = H 2 0 or THF) (255). Hulsbergen et al. obtained a related compound, [Cu3(OH)(4-Hpz)3(4-HpzH) 2 (N0 3 ) 2 ], the structure of which has been determined (256). In this material the O H group bridges three copper(II) centers, and the infrared band for the O H stretching vibration occurs at 3600 cm"1. 5.2.2.2.2 ELECTRONIC SPECTROSCOPY The UV-Vis-NTR spectrum of the purple material exhibits a broad absorption centered at 600 nm (16 700 cm"1) with a shoulder at 800 nm (12 500 cm"1) , and a sharp absorption at 2750 nm (3640 cm"1). The sharp peak at 2750 nm corresponds to the O H stretch also observed in the infrared spectrum. The absorption at 600 nm, would be consistent with a square pyramidal or octahedral copper(II) chromophore geometry (257). The shoulder at 800 nm is also consistent with both 5 and 6 coordinate copper(II) centers (257). 207 5.2.2.2.3 MASS SPECTROMETRY The low resolution electron impact ionization spectrum of the purple decomposition product of Cu3(3,5-F6diMepz)5 was obtained using a KRATOS MS50 mass spectrometer. The spectrum is shown in Appendix IV, Figure IV-2. Interestingly, the spectrum is nearly identical to that of the green, [Cu3(3,5-F6diMepz)5], complex exhibiting a parent ion peak, P + , at m/e -1206 along with strong signals presumably due to ions arising from consecutive loss of two 3,5-F6diMepz ligands, L , from the parent species at m/e - 1003 (P-L +) and m/e -800, (P-2L+). The intensities of the individual lines in these multi-line signals are again those expected from the presence of three Cu atoms and the naturally occurring isotopic abundance for the element. The similarities in the spectra obtained for the two materials suggest that either some of the parent green [Cu3(3,5-F6diMepz)5] complex was present as an impurity in the purple solid sample, or that by heating the purple material the parent green compound is reformed. The presence of some of the green complex as an impurity could result in the nearly identical spectra as, unlike the green compound, the purple material is not appreciably volatile, and therefore may not exhibit any peaks of its own. Also consistent with this is the relatively low intensities of the peaks observed in the spectrum of the purple material. The other possibility involves the reformation of [Cu3(3,5-F 6diMepz) 5] upon heating the purple material. As mentioned earlier, the purple material 208 does appear to at least partially convert back into the parent green complex upon heating above 100°C. The mass spectra were obtained on samples heated to 150°C and 180°C for the green and purple materials respectively. 5.2.2.4 M A G N E T I C PROPERTIES A N D PROPOSED STRUCTURAL FEATURES Variable temperature magnetic measurements were recorded for powder samples of the purple material on a SQUID magnetometer. The powder magnetic susceptibility and magnetic moment data are listed in Appendix II, Table 11-10. The magnetic susceptibilities and moments have been calculated assuming the purple material has the same parent mass (1206 gmol"1) as the green complex and assuming that two copper(II) ions are present per molecule. In other words, the molar magnetic susceptibility was obtained by multiplying the gram magnetic susceptibility by 603 gmol"1. If indeed there are new O H groups coordinated to some of the metal centers in the purple material, the parent mass would obviously be different, as would the distribution of the various oxidation states of the copper ions present. The calculation has been performed this way to aid in comparisons between the magnetic behaviour of the purple and green complexes, and the results should not be mistaken for accurate values. The resulting variable temperature magnetic susceptibility and moment data are presented graphically in Figure 5.5. 209 0 50 100 150 200 250 300 Temperature (K) Figure 5.5. Variable temperature magnetic susceptibility and magnetic moment plots for the purple decomposition product of [Cu3(3,5-F6diMepz)5]. The values have been calculated per mole of Cu(II), assuming that two thirds of the copper ions are in the 2+ oxidation state, the molecular weight is 1206 gmol"1. Even though the magnitude of the values determined for the magnetic susceptibilities and moments are likely erroneous, the trends observed are real. The magnetic moment decreases with temperature from 300 K to 110 K. As the temperature is decreased further, there is a very evident upturn in the magnitude of the magnetic moment. The moment continues to increase as the temperature is reduced until a maximum value is reached at 60 210 K. The magnitude of the magnetic moment then falls again with decreasing temperature to the lowest temperature studied of 2 K. This unusual behaviour may be the result of two competing exchange pathways in the material. That the observed decrease in magnetic moment as temperature is reduced is consistent with strong antiferromagnetic coupling in the system. The increase in the magnetic moment as the temperature is decreased from 110 K to 60 K may result from a ferromagnetic exchange pathway also present in the material. At temperatures below 60 K the antiferromagnetic component may begin to dominate again. If, in the purple material, there are OH groups bridging rings of [Cu3(3,5-F 6diMepz) 5], two exchange pathways would not be unexpected. It is already known that the triple pyrazolate bridge between the two Cu(II) centers in the green complex provides an efficient pathway for antiferromagnetic exchange. The presence of additional O H bridges between rings may therefore provide a new pathway for magnetic coupling. Another possibility is the presence of uncompensated spin. If the purple complex consists of three copper(II) ions linked by pyrazolate and/or hydroxide bridges it would not be possible for all of the spins to align antiparallel. In the extreme of complete antiferromagnetic coupling, the spin on one of the three copper(II) centers would remain uncoupled. However, if the molar magnetic susceptibilities are calculated assuming there are three copper(II) ions present per trimetallic ring, the resulting magnetic moment at the lowest temperatures studied is only half that expected for the presence of a single unpaired electron. 211 As was mentioned above, the magnetic data were treated in an identical fashion to the magnetic data of the [Cu3(3,5-F6diMepz)5] complex to aid in comparisons between the two systems. As can be seen by comparing Figures 5.5 and 5.6, the magnitude of the magnetic susceptibility is radically different for these two materials. The presence of even a few percent of the purple material in samples of [Cu3(3,5-F6diMepz)5] would dramatically affect the magnetic data of the latter. It is possible that the ferromagnetic impurity observed in "pure" samples of the green complex were actually small amounts of the purple decomposition product described here. The physical properties of the purple material are consistent with a polymeric or a large oligometallic structure. The compound is insoluble in solvents in which it does not decompose and is involatile below 100°C. Infrared spectroscopy results indicate the presence of O H groups in this material. These groups may bridge individual rings of Cu3(3,5-F6diMepz)5 to form an oligomeric complex or a 1-D, 2-D or 3-D polymeric system. 5.3 S U M M A R Y A N D CONCLUSIONS Attempts to prepare the binary copper(II) 3,5-bis(trifluoromethyl)pyrazolate polymer, [Cu(3,5-F6diMepz)2]x, by the reaction of copper metal shot with 3,5-212 bis(trifluoromethyl)pyrazole under an atmosphere of dioxygen at ~100°C led to the formation of a dark green material of composition Cu3(3,5-F6diMepz)5. Infrared studies established that this material was devoid of any residual N - H functionalities and mass spectral analysis showed the parent ion, P + , cluster as the highest m/e signals in the spectrum, with strong signals arising from the ions resulting from the consecutive loss of two 3,5-F6diMepz ligands P - L + and P-2L + . Single crystal X-ray diffraction studies showed that the complex consists of a triangular arrangement of copper ions linked by five bridging 3,5-F6diMepz ligands, resulting in two Cu(II) ions and one unique Cu(I) ion. Magnetically the complex can be described as a copper(II) dimer with four bridging groups (three 3,5-F 6diMepz bridges and the Cu(I)(3,5-F6diMepz)2 acting as the fourth bridge) linking the two paramagnetic centers. The calculated value of the magnetic exchange coupling constant, J, is -235 cm"1, consistent with several other pyrazolate bridged copper(II) systems. Prolonged exposure to moisture results in the formation of a purple compound. The elemental composition of this purple complex is quite similar to the green parent. However the purple material has significantly different physical and magnetic properties and may be polymeric in nature. 213 Chapter 6 AN 1R O N (11) IMID A Z O L E IMIDAZOLATO CO MPOUND 6.1 INTRODUCTION Bridging pyrazolate and 3,4,5-substituted pyrazolates have been shown, both in the current work and by others, to provide an efficient pathway for magnetic exchange in several polymeric and oligometallic transition metal complexes. The 1,3-diazolate ligand, imidazolate (imid), when acting as a bridging ligand, has also been shown to promote magnetic exchange interactions (263-275). However, in the case of imidazolate complexes, the steric effects imposed by the C H groups in the 2-position makes the double bridging motif, which leads to 1-D chain polymers in the pyrazolate complexes, unlikely. The possibility of forming extended structures of higher dimensionality with the imidazolate ligand makes structural and magnetic studies on these systems potentially very interesting. There have been a number of relevant studies reported in the literature on metal imidazolate complexes. In 1960, Jarvis and Wells (276) reported the 3-D structure for [Cu(imid)2]x. In 1964 the syntheses of binary metal imidazolates of nickel, copper, silver and zinc were reported by Bauman and Wang (277). Also in 1964 Brown and Aftergut 214 described the preparation of bis(imidazolates) of copper, cobalt and zinc (278). Interestingly the [Cu(imid)2]x complex isolated by Bauman and Wang was described as reddish-purple while the material prepared by Brown and Aftergut was blue. Inoue et al. (243) described the preparation of a third form (green) of [Cu(imid)2]x in 1965 and reported on the temperature dependence of the magnetic susceptibilities of the three forms (green, brown and blue) in the following year (263). In 1967 the structure of the zinc imidazolate [Zn(imid)2]x (279) was reported and the structure of the cobalt analogue was published in 1975 (280). Lundberg reported in 1972 on the structure of an imidazolato-bridged copper(II)-imidazole chloride complex [Cu(imid)(imidH)2Cl]x (281). The material was determined to be comprised of chains of copper(II) ions linked by single imidazolate bridges and has subsequently been shown to exhibit strong antiferromagnetic exchange (264). In 1974 a single imidazolate bridged polymer of an iron(III) hemin was shown to possess unusual magnetic properties (265). More recently, Chaudhuri et al. reported the structure and magnetic properties of an imidazolato-bridged trinuclear copper(II) complex (266). The copper(II) ions in this compound form a triangular array and antiferromagnetic coupling via the bridging ligands leads to uncompensated spin and so-called "spin frustration" behaviour. An electrochemical synthesis and magnetic studies of a number of neutral transition metal imidazolates have also been recently reported (94). In addition to the metal-imidazolate polymeric materials, several oligometallic complexes in which imidazolate acts as a bridging ligand have been reported. Both homobimetallic and heterobimetallic imidazolate-bridged complexes have been described (268, 282-285) and 215 magnetic studies on these systems (283, 285) again reveal the imidazolate linkage to provide an efficient pathway for antiferromagnetic exchange. A series of 4(5)-substituted imidazolate-bridged bimetallic and trimetallic complexes have been thoroughly investigated by Matsumoto and others (267, 286-292). Again the results show the imidazolate linkage to promote magnetic exchange between paramagnetic metal centers. In 1968, Seel and Sperber reported the synthesis, but no further characterization, of an iron(II) complex of composition Fe(imid)2-0.5(imidH) (75). This complex was most likely the same material as the one described here and determined to be [Fe3(imid)6(imidH)2]x. 6 2 RESULTS A N D DISCUSSION 6.2.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPERTIES Details of the synthesis of [Fe3(imid)6(imidH)2]x are described in Chapter 9, Section 9.2.6. The success of the reactions between nickelocene and manganocene with pyrazoles to generate polymeric complexes, along with the previously reported preparation of an iron(II) material from the reaction of ferrocene with imidazole, were the primary factors leading towards the synthetic approach used here. Ferrocene and imidazole were combined and sealed under vacuum in a Carius tube and the mixture was heated (150 °C) well above the melting temperature of imidazole (90°C) for several days. Over this period, the bright 216 orange colour of the initial solution of ferrocene in molten imidazole was discharged and pale yellow crystals deposited from the solution. The product and excess imidazole were allowed to cool to room temperature and the Carius tube was opened under a dinitrogen atmosphere. The excess imidazole was extracted with dry and dioxygen-free acetonitrile solvent and the compound was isolated as pale yellow crystals. As single crystals, [Fe3(imid)6(imidH)2]x appeared to be fairly air-stable and could be handled briefly under normal atmospheric conditions without any obvious physical changes. However, powdered samples would quickly turn brown upon exposure to dioxygen and moisture. The material was therefore treated under inert conditions whenever possible. The complex was insoluble in all common organic solvents and water. Attempts to dissolve the complex in wet solvents or water generally resulted in the apparent decomposition of the compound into a brown insoluble material. The compound is involatile and does not melt up to the temperature at which thermal decomposition begins (200°C). T G A and DSC plots are shown in Figure 6.1. T G A results suggest the complex decomposes in a multistep process as temperature is ramped slowly upwards. Approximately 7% of the materials initial weight was lost between 200°C and 270°C. The next 11% was lost between 270°C and 290°C. A further 4% of the complexes initial weight was lost between 217 Figure 6.1. Thermal decomposition of [Fe3(imid)6(imidH)2]x as studied by (a) T G A and (b) DSC. 290°C and 325°C, followed by a loss of roughly 43% of the initial weight between 325°C and 560°C. No additional loss of weight was observed up to the maximum temperature studied of 800°C. DSC measurements were consistent with the TGA results. At the onset of thermal decomposition a large endothermic peak is observed centered at 280°C. Two 218 additional endothermic peaks at 405°C and 480°C are also observed. It is possible that the large endothermic event at 280°C, and the corresponding weight loss of-18% is a result of the loss of the two neutral imidazole groups in the [Fe3(imid)6(irnidH)2] repeat unit. The weight percent due to the two neutral imidazoles is -19%. 6.2.2 X - R A Y DIFFRACTION STUDIES Single crystal X-ray diffraction studies performed by S.J. Rettig reveal [Fe3(imid)6(imidH)2]x to be comprised of linear chains of tetrahedral iron centers cross-linked by octahedral iron ions resulting in a 3-dimensional structure. Crystallographic data, atomic coordinates, bond lengths and bond angles appear in Appendix I, Tables I-10 through 1-13. The structure of the repeat unit in the polymer is shown in Figure 6.2 and the unit cell is shown in Figure 6.3. All of the iron centers are bridged to four other iron centers via imidazolate ions, the two remaining (trans) coordination positions of the octahedral centers are occupied by neutral imidazole molecules. The chains of tetrahedral iron ions are oriented along the c-axis giving sheets of tetrahedral iron chains in the (100) plane. Each chain within a sheet is linked via the octahedral irons to six different tetrahedral iron chains, two in each of the adjacent parallel sheets and two neighbouring chains within its own sheet. The Fe-N bond distances around the octahedral Fe(l) are significantly longer at 2.183 - 2.253 A than those 219 Figure 6.2. The molecular structure of the repeat unit of [Fe3(imid)6(imidH)2]x with atom numbering scheme and 33% probability thermal ellipsoids for all non-hydrogen atoms. around the tetrahedral Fe(2) at 2.020 - 2.053 A. The angles N-Fe-N around octahedral Fe(l) are all close to 90° ranging from 88.18 - 91.82°; those around the distorted tetrahedral Fe(2) range widely from 99.3 - 128.0°. 220 Figure 6.3. Stereoview of the unit cell of [Fe3(imid)6(imidH)2]x. For clarity all hydrogen atoms, carbons 4 and 5 of imid and all atoms except the coordinating nitrogen of imidH have been removed. The literature contains few references to structures of divalent transition metal imidazolate polymers but the structures of [Cu(imid)2]x (276), [Zn(imid)2]x (279), and [Co(imid)2]x (280) have been reported. The [Cu(imid)2]x structure, like the present structure, displays the metal ions in two distinct environments, in this case square planar and distorted tetrahedral. The [Zn(imid)2]x and [Co(imid)2]x structures have the metal(II) ions in a distorted tetrahedral environment. A common feature in the above three structures is the formation of M4(imid)4 rings which are cross-linked in the 3-D network with 221 imidazolate bridges between the M(II) ions. In the present structure again we see large ring structures, in this case Fe6(imid)6 units, crosslinked into a 3-D network. This basic preference for larger metal ligand ring systems over the simple M 2 L 2 rings of the metal(II) pyrazolate structures is readily understood on steric grounds. The adjacent donor N atoms of the pyrazolate ligands allows the formation of planar or boat-like M-(N-N) 2 -M six membered rings in the extended [M(4-Hpz) 2] x chain polymers. With the imidazolate bridge the donor N atoms are now separated by a C atom and M 2 L 2 rings in the planar or boat-like conformations required for a linear chain polymeric structure are sterically impossible. This M 2 L 2 arrangement, however, was suggested recently, together with even more unlikely triply bridged M 2 L 3 moieties as possible structural motifs in binary metal imidazolate systems (94). Similar structural patterns have been documented with discrete molecular species in that [Me2Ga(4-Hpz)]2 is dimeric with the Ga-(N-N) 2-Ga six membered ring as the central feature (293); whereas, [Et2Ga(2-Meimid)]4 is tetrameric with the central feature being a Ga4(2-Meimid)4 puckered ring system (294). 6.2.3 SPECTROSCOPIC STUDIES 6.2.3.1 INFRARED SPECTROSCOPY Band frequencies and relative intensities are listed in Appendix III, Table III-11. As has been mentioned in most of the previous chapters, infrared spectroscopy was primarily used to determine the presence or absence of any neutral azole. This was accomplished by 222 looking for a VNH band in the spectrum of the product in question. In the case of [Fe(imid)6(imidH)2]x there was a sharp and intense band in this region at 3380 cm"1 indicating that there is neutral imidazole present and that it is not involved, to any significant degree, in any hydrogen bonding. Interestingly this N - H stretch is also visible in the electronic spectrum discussed below. 6.2.3.2 ELECTRONIC SPECTROSCOPY The UV-Vis-NIR spectrum of the complex is shown in Figure 6.4 and displays two d-d absorption bands centered at 1265 nm and 1960 nm. These we assign to the 5T 2 g—» 5E g transition for high spin octahedral Fe(II) and to the 5 E-> 5 T 2 transition for tetrahedral Fe(II) respectively. One interesting aspect of the spectrum is the presence of the sharp and intense v N -H IR band at 2950 nm (3390 cm"1). 223 O •-cs & o CO T T 500 1000 1500 2000 Wavelength (nm) 2500 3000 Figure 6.4. UV-Vis-NIR spectrum for [Fe3(imid)6(imidH)2]x. (The sharp peak at 2950 nm results from the V N - H IR band (-3390 cm"1) and the sharp peaks at about 2500 nm arise from the mulling agent, Nujol). 6.2.4 M A G N E T I C BEHAVIOUR Magnetic susceptibilities were measured at 500 G and 10 000 G from 2 to 300 K using a SQUID magnetometer. Magnetization studies were done at applied fields ranging from 0 G to 55 000 G at temperatures of 2 K , 4.8 K, 13 K, 20 K, 50 K , 100 K, and 300 K. Hysteresis magnetization data were obtained by oscillating the applied magnetic field 224 between 55 000 G and - 55 000 G at 4.8 K. The powder magnetic susceptibilities and magnetic moments versus temperature data for the two applied fields (three runs in total), along with the hysteresis magnetization data and magnetization data at several different temperatures are tabulated in Appendix II, Tables 11-11,11-13 and 11-14 respectively. The presence of linear chains of imidazolate-bridged tetrahedral iron ions in the 3-D network of the compound gives rise to the possibility of magnetic exchange between the paramagnetic metal centers in these chains. These chains of tetrahedral iron centers are cross-linked by imidazolate-octahedral iron-imidazolate bridges which leads to an additional potential pathway for magnetic exchange. Magnetic susceptibility and %T versus T data on powdered samples of Fe3(imid)6(imidH)2]x in an applied field of 500 Gauss are shown in Figure 6.5. The results show an average magnetic moment which decreases from 5.48 to 3.74 M-B (xT = 3.76 to xT = 1.75 cn^Kmol"1) with temperature decreasing from 300 K to 17 K. There is a sharp upturn in the average moment to a maximum value of 14.3 ( i B (xT =25.7 cm^Kmol"1; Figure 6.5) at 11 K. Magnetization versus applied field behaviour is illustrated in Figure 6.6. 225 l H ooooo 50 ~i— 50 100 150 200 Temperature (K) 100 150 200 Temperature (K) 250 300 250 300 Figure 6.5. Plot of the magnetic susceptibility (O) and xT ( • ) versus temperature at 500 G for a powdered sample of [Fe3(imid)6(imidH)2]x-Above 17 K the magnetization varies linearly with applied field to the maximum field strength studied of 55 000 G. At temperatures below 17 K the magnetization no longer varies linearly with the applied field and a net magnetization of the compound is observed at zero field. The highest magnetization reached at 2 K and 55 000 G is 7280 cm JGmol"\ 226 8000 6000 o E a E o 4000 o C3 _ N a BO 2 2000 R g 3^ o • O A • • 6 o A O A O • O 10000 20000 30000 Field (G) 40000 50000 • 60000 Figure 6.6. Plot of magnetization versus applied magnetic field for [Fe3(imid)6(imidH)2]x at 300 K (red O), 100 K ( • ) , 50 K (blue A), 20 K (V), 13 K (O), 4.8 K (green O), and 2 K (yellow A). significantly below the theoretical saturation magnetization of 22 300 cm3Gmol"1 (295). Cycling the field between 55 000 G and -55 000 G at 4.8 K generates a hysteresis loop, shown in Figure 6.7, with a remnant magnetization of 2500 cm 3Gmor 1 and a coercive field of 200 G. 227 Figure 6.7. Field dependence of [Fe3(imid)6(imidH)2]x at 4.8 K. In attempting to explain the magnetic behaviour of this compound the results obtained at temperatures above 17 K are considered first. In the absence of magnetic exchange effects the tetrahedral iron centers are expected to make a temperature independent contribution to the magnetic moment while the octahedral centers, by virtue of the orbitally degenerate 5 T 2 g ground state, are expected to make a temperature dependent 228 contribution (296). The observed temperature dependent moment behaviour above 17 K may then be accounted for by single-ion effects associated with the octahedral metal centers combined, possibly, with antiferromagnetic exchange interactions in the lattice. A potentially very complicated modeling problem may be simplified by utilizing information available from the X-ray determined structure. The presence of unpaired electron density in both e (g ) and t2(g> type metal orbitals in both T d and Oh iron(II) centers suggests that orbital symmetry effects may not be important in determining whether exchange involving one metal type is stronger or weaker than that involving the other type. Metal to bridging-ligand bond strengths may, however, be important. The structure reveals that the Fe-N (bridge) bonds involving tetrahedral iron centers are significantly shorter than those involving the octahedral centers suggesting that exchange interactions involving the tetrahedral iron should dominate. Accordingly, the magnetic susceptibility data above 17 K may be modeled as arising from separate contributions from the different metal centers: XFXe P=(2/3)X^I+(l/3)xr [6-1] Values of %l* ( the susceptibility per mole of octahedral iron(II)) at different temperatures were interpolated from the theoretical data given by Figgis et al. (296). The theoretical values are given as functions of three parameters: (i) X, the spin-orbit coupling constant; (ii) 229 k, the orbital reduction factor and (iii) v, the axial distortion parameter. Two sets of parameters were chosen to model X F J ^ n both, k was fixed at 1 (no orbital reduction assumed) and A, was fixed at the free ion value for iron(II) of -100 cm"1. In one set, v was fixed at 0 (assumes no tetragonal distortion) and in the other v was fixed at +10. A positive value of v is dictated by the X-ray structure which shows an axial elongation. Negative v parameters would apply to axial compression. The range of 0 to +10 represents the total range of distortion for which calculations were done in reference (296). From the measured susceptibilities (% Fe P) and these theoretical susceptibilities (x F e X values of %Te were then calculated from equation [1] and these data were plotted against temperature as shown in Figure 6.8. The molar magnetic susceptibilities and magnetic moments corresponding to the tetrahedral iron(II) ions are listed in Appendix II, Table 11-12. The results indicate antiferromagnetic interactions between tetrahedral iron centers. This interaction was analyzed by employing the model for linear chains of antiferromagnetically coupled S = 2 metal ions developed by Weng (202). Hiller et al. generated equation [2], which reproduces Weng's numerical results (203), and the equation A £ V ( 2 + 71.938x2) Xchai„ (i+ io.482x +955.56x3)AT L J 230 Figure 6.8. Plots of x*l versus temperature at 500 G. Magnetic contribution fromxFe subtracted out as described in text for v = 0 ( • ) and v = +10 (O). The lines are calculated from theory as described in the text. has been used to fit the experimental data. In equation [2] x = \ J\/kT and J is the exchange coupling constant. The best fit value of J and g with v set at 0 were -2.33(2) and 2.282(6) cm"1 (F = 0.020) while the best fit values of J and g with v set at +10 were -2.54(2) and 2.334(6) cm"1 (F = 0.021) respectively. F has been described earlier (Equation [3.9]) and 231 provides a measure of the goodness of fit. As seen from the F values and by visual inspection of the data compared to theory (Figure 6.7), the quality of the fits to theory is not affected significantly by the magnitude of the distortion parameters chosen for the octahedral metal centers and the g and J parameters obtained for the antiferromagnetically coupled centers are also relatively insensitive to the varying degrees of distortion chosen for the octahedral centers. To put the magnitude of this antiferromagnetic coupling in perspective the results obtained here may be compared with those reported earlier for the closely related [CuCl(imid)(imidH)2]x, the structure of which consists of chains of copper ions bridged by imidazolate ligands (281). The magnetic susceptibilities of the compound were fit to a linear chain model yielding J = -84 cm"1 (264). The model employed in reference (264) used the Heisenberg spin Hamiltonian in the form: H = JSi*S2; whereas, the model employed in the present work has H = -2JSi'S2- Hence " J " = -42 cm"1 should be used for the copper(II) system in making comparisons with the current study. Moreover, to compare complexes with different total spin, S (as described earlier in this thesis), it is appropriate to compare values of \4JS2\ rather than J (204)). Employing J = -2.3 cm"1 for [Fe3(imid)6(imidH)2]x, \4JS2\ = 37 cm"1 for this compound compared to 42 cm"1 for the imidazolate bridged copper(II) polymer. Interestingly, the value of \4JS2\ for the previously studied imidazolate bridged triangular complex, calculated as above for [CuCl(imid)(imidH)2]x (266), is 38 cm"1, in very close agreement with the other values. The 232 magnitude of the magnetic exchange coupling can also be compared with previously reported dimetallic Fe(II) and Mn(II) irnidazolate-bridged metalloporphyrins (273). The value of |4V,S2| for the iron(II) complex, [Bu4N][Fe2(|i-imid)(Ar,#-Bis(5-(o-phenyl)-10,15,20-triphenylporphyrin)urea)], is 37 cm"1 while an estimate of the value of - J for the Mn(II) analogue (-J= 2 cm"1) generates a \4JS2\ value of 50 cm"1. Clearly the magnitude of the antiferromagnetic coupling in the extended chains of [Fe3(imid)6(imidH)2]x is of the order expected for imidazolate bridged metal ions. The above analysis provides strong evidence for significant antiferromagnetic coupling between tetrahedral metal centers in [Fe3(imid)6(imidH)2]x. The magnetic phase transition at 17 K is consistent with long-range ordering at low temperatures. The observation of net magnetization at zero applied field below the transition temperature, combined with the fact that the maximum magnetization obtained falls considerably short of the theoretical saturation value, provides support for the conclusion that this is an example of canted spin antiferromagnetic coupling leading to weak ferromagnetism at low temperatures (297). Support for a canted spin structure also comes from the X-ray determined molecular structure which shows a systematic alternation of the relative orientation of the iron(II) chromophores along the imidazolate bridged chains. As a measure of this we calculated a dihedral angle of 43° between the vectors bisecting the N -Fe-N intrachain angles on adjacent tetrahedral iron atoms along the chain. 233 A previously reported canted spin iron(II) complex is [Fe(4-imidazoleacetate)2]x-2CH3OH (34). The compound behaves magnetically as a 2-D antiferromagnetic system in which 3-D long-range ordering gives a net magnetic moment below 15 K. Magnetization experiments at 4.2 K on this compound revealed a coercive field of 6200 G and a remnant magnetization of 1200 cm3Gmol"1. In this system the antiferromagnetic coupling occurs within sheets of octahedral iron(II) centers and hydrogen bonding interactions between the sheets are presumed to promote the observed 3-D weak ferromagnetism. In contrast, in the system described in the present work the antiferromagnetic coupling seems to occur in chains of tetrahedrally coupled iron(II) centers and coordinative links between these chains involving octahedral iron(II) centers promote 3-D order and weak ferromagnetism. More recently, the biimidazolate complex, [Fe(biimidazolate)]x-CH3OH-0.5H2O (35), has been reported as another example of a canted spin iron(II) complex. Magnetic studies revealed a magnetic phase transition at 25 K and magnetization data collected at 4.2 K again revealed a hysteresis loop with a reported coercive field of 2000 G and a remnant magnetization of 950 cm3Gmol"1. Unfortunately the structure of this compound has not been determined by single crystal X-ray diffraction. Moreover, it should be noted that the structure proposed by the authors for the complex does not correlate with the empirical formula. 234 6.2.4.1. PROTOCOL FOR OBTAINING MAGNETIC D A T A FOR M O L E C U L A R M A G N E T S As was shown in the previous section, [Fe3(imid)6(imidH)2]x can become a permanent molecular magnet below 17 K. This magnetic behaviour can lead to misleading results if the utmost care is not taken during the collection of the magnetic data. Problems arise if the remnant magnetization present after the collection of one set of magnetic data is not removed before another set of magnetic data is collected. If not removed, the remnant magnetization will still be present during any subsequent magnetic data collection and any new magnetic signal detected by the magnetometer may be greatly influenced or dominated by this persisting magnetization. The need to remove this residual magnetization is best illustrated by considering the hysteresis loop (Figure 6.6) generated by oscillating the applied magnetic field between positive and negative values. At 4.8 K and zero applied field the material can have a magnetization of anywhere between -2500 cm 3Gmor 1 and +2500 cm 3Gmor 1, depending on the sample conditions prior to the measurement. The only convenient way to remove this magnetization is to heat' the sample above the temperature at which the material becomes a magnet, set the applied field to zero by oscillating the field between positive and negative values (decreasing the magnitude of the field in each oscillation), and finally cooling the sample to the target temperature in zero applied field. This procedure should be used in all cases in which residual magnetization is observed. 235 6.3 S U M M A R Y A N D CONCLUSIONS The reaction of ferrocene with molten imidazole at 150°C yields pale yellow crystals of composition [Fe3(imid)6(imidH)2]x. The structure of this material has been determined by single crystal X-ray diffraction studies and is shown to consist of chains of iron(II) centers cross-linked via octahedral iron(II) ions to generate a 3-D array. Each iron center is bridged to four other metal centers via the imidazolate ions with the two remaining (trans) coordination positions of the octahedral centers being occupied by neutral imidazole molecules. Magnetic susceptibilities on powdered samples have been measured over the temperature range 2 to 300 K, and at fields ranging from 0 to 55 000 G. The compound exhibits antiferromagnetic coupling along chains of tetrahedrally coordinated iron centers. The magnitude of the magnetic exchange coupling constant, J, has been estimated by modeling the data above 17 K, and is approximately -2.4 cm"1. This value is in very good agreement with other reported coupling constants for imidazolate bridged transition metal complexes. A canted spin structure leads to weak ferromagnetism at temperatures below a magnetic phase transition temperature of 17 K. Upon cycling the applied magnetic field between 55 000 G and -55 000 G at 4.8 K, a hysteresis loop is obtained with a remnant magnetization of 2500 cm3Gmol"1 and a coercive field of 200 G. 236 This compound is particularly interesting in that long-range magnetic ordering is observed at low temperatures. Only two complexes prepared in this work, [Fe3(imid)6(imidH)2]x and [Cu(trz)2]x (discussed in the next chapter), exhibit long-range magnetic ordering in the temperature range studied (i.e. above 2 K). It should also be noted that reports on coordination compounds that exhibit long-range magnetic ordering are rare and, of these reports, only a handful have magnetic ordering that is attributable to a canted structure. 237 Chapter 7 BIS (1 , 2 , 4 - T R IA Z O L A T O ) C O P P E R (11) [Cu (trz) 2] x 7.1 INTRODUCTION In this chapter, a triazolate complex of copper(II) is examined. This material, [Cu(trz)2]x, has been prepared by others in the past (241, 243) and some characterization and magnetic studies have been reported (246). However, these previous reports did not discuss the magnetic behaviour of this complex at any temperature below 77 K, and the results to be discussed here show that this compound possesses interesting magnetic properties at these low temperatures. In addition, a different synthetic pathway from the one used previously is described and some structural proposals are presented. 7.2 RESULTS A N D DISCUSSION 7.2.1 SYNTHESIS, PHYSICAL A N D T H E R M A L PROPERTIES Details of the synthesis are summarized in Chapter 9, section 9.2.5.1. As mentioned earlier [Cu(trz)2]x has been previously prepared by others (241, 243). In those reports the material was obtained by first adding an aqueous solution of 1,2,4-triazole (2 moles) to an ammoniacal solution of copper(II) sulfate (1 mole). The resulting blue precipitate was dissolved in 6 M ammonia solution and the desired product (a purple powder) was obtained 238 by allowing the ammonia solution to slowly evaporate. In the current work a material of the same formulation was isolated by employing the molten ligand/metal polymerization method developed by Ehlert (85) and described earlier in the thesis. Attempts to produce the product in a form suitable for single crystal X-ray diffraction studies were unsuccessful. These attempts generally involved modifications to the procedure designed to reduce the rate of reaction. One such attempt involved lowering the temperature of the reaction (below that of the melting temperature (120°C) of 1,2,4-triazole) by including small amounts of solvent (benzene and xylenes) in the reaction mixture. Another approach involved a reduction in the amount of dioxygen gas present by reacting the copper shot with the triazole under normal atmospheric conditions, rather than under a pure O2 atmosphere. Although most of these attempts did produce the desired product, no macroscopic single crystals were obtained. Finally, and again in an attempt to produce large single crystals, the previously published synthetic approach (described above) was employed, but only powder products were obtained. As is the case with most of the polymeric materials discussed thus far, [Cu(trz)2]x is an insoluble, involatile and thermally robust solid. The complex is stable both in air and in contact with wet (with water) solvents. [Cu(trz)2]x is not soluble in any common organic solvent or water, but does dissolve with decomposition in concentrated mineral acids. Evidence for the involatility of the compound and its thermal stability comes from T G A and DSC analyses. The T G A and DSC plots are shown in Figure 7.1. 239 Figure 7.1. T G A and DSC plots for [Cu(trz)2]x. As can be seen there is no loss in mass due to thermal decomposition or sublimation at any temperature below 300°C. The TGA data suggests that two steps occur in the thermal decomposition of [Cu(trz)2]x. The first event involves a rapid loss of nearly 40% of the initial mass of the sample between 320°C and 380°C. The second event is more gradual with roughly 30% of the initial mass being lost between 380°C and 560°C. As is seen in some of the nickel species described earlier, an apparent weight gain is observed between temperatures of 560°C and 700°C. The weight then remains constant as the temperature is increased up from 700°C to the maximum temperature studied of 800°C. The gain in weight at high temperatures is not understood, but may be the result of the formation of 240 metal oxides. As was the case with some of the previously described nickel polymers, the observed minimum weight percent, in this case -30% at 560°C, corresponds to the weight of metal present in the compound. In addition, the final observed weight percent (-40%) is consistent with a final decomposition product copper(II) oxide, CuO. DSC results indicate the presence of two separate endothermic events occurring at roughly 355°C and 380°C respectively. Interestingly there is what appears to be a very sharp exothermic event occurring almost in the center of the second endothermic peak. The first endothermic peak coincides with the temperature region spanning the first step observed in the T G A results. The second endothermic peak observed in the DSC data (and the exothermic event within this peak) occurs at the temperature of the onset of the second step in the T G A plot. The DSC data obtained do not show clearly any event associated with the apparent weight gain seen in the T G A results. This is partly due to the temperature limitations of the DSC equipment. The absence of a well defined event occurring in the DSC may also be a result of the different environments the material is in during the DSC and T G A data collection processes. In the T G A experiment, the material is placed on a platinum pan that is attached to an analytical balance. The experiment is said to be performed under a dinitrogen atmosphere as N 2 gas is continuously introduced into the sample chamber. However, there is little chance that all of the dioxygen gas is removed by this process as a very large "exhaust pipe" is open to normal atmospheric conditions. This differs significantly from the DSC sample environment in which the material is placed in a crimped pan. This pan provides additional protection to the sample from any dioxygen gas that may be present in 241 the sample chamber. Like the T G A process, the sample chamber used in the DSC experiments is flushed with dinitrogen. However, the DSC system utilizes an exhaust pipe with a much smaller diameter, resulting in less chance of diffusion of dioxygen gas back into the sample area. It is therefore possible that some or all of the thermal decomposition analyzed by T G A is different from the thermal decomposition analyzed by DSC. 7.2.2 SPECTROSCOPIC STUDIES 7.2.2.1 INFRARED SPECTROSCOPY Band frequencies and their relative intensities for [Cu(trz)2]x are tabulated in Appendix III, Table 111-12. N - H stretching and bending bands are absent from the spectrum obtained indicating that all of the triazole present is in its deprotonated form. The spectrum is very similar to that obtained for [Mn(trz)2]x (see Appendix III, Tables III-8 and III-12). 7.2.2.2 ELECTRONIC SPECTROSCOPY The UV-Vis-NIR spectrum of [Cu(trz)2]x is shown in Figure 7.2. The spectrum consists of two absorptions, one at 575 nm (17 400 cm"1) and the other at 280 nm (35 700 cm"1). Copper(II) complexes with CuN x (x = 4-6) commonly yield electronic spectra with 242 I 1 1 1 1 1 1 0 500 1000 1500 2000 2500 3000 Wavelength (nm) Figure 7.2. UV-Vis-NIR spectrum of a concentrated mull (a) and a dilute mull(b) of[Cu(trz)2]x. only one main band resulting from d—>d transitions. Hathaway (257), generated a correlation between the energy of the center of gravity of this main band to the stereochemistry of the chromophore, CUN4.6, present. These correlations serve as a useful guide in determining the coordination about the copper ions in such materials. Conveniently, the band maximum of the d—»d transition observed for [Cu(trz)2]x, 17 400 cm"1, lies in an unambiguous region in Hathaway's correlation diagram. This region is consistent with a tetragonal octahedral CuN 6 chromophore. According to Hathaway, the lower energy limit for a square planar CuN 4 chromophore geometry is 18 000 cm", the upper limit for a compressed tetrahedral CuN 4 arrangement lies at 16 000 cm"1, and the upper limit for a square pyramidal CuN 5 chromophore geometry is 17 000 cm"1. This result is therefore consistent with six-coordinate copper(II) ions in [Cu(trz)2]x. 243 7.2.3 PROPOSED STRUCTURE A N D MAGNETIC B E H A V I O U R Magnetic susceptibilities were measured at 500 G and 10 000 G from 2 to 300 K using a SQUID magnetometer. Magnetization studies were done at applied fields ranging from 0 G to 55 000 G at temperatures of 2 K, 4.8 K, 13 K, 25 K, 50 K, 100 K, and 300 K. Hysteresis magnetization data were obtained by oscillating the applied magnetic field between 55 000 G and - 55 000 G at 4.8 K. The powder magnetic susceptibilities and magnetic moments versus temperature data for the two applied fields, along with the hysteresis magnetization data and magnetization data at several different temperatures are tabulated in Appendix II, Tables 11-15,11-17 and 11-16 respectively. Magnetic susceptibility and magnetic moment versus temperature data on powdered samples of [Cu(trz)2]x in an applied field of 500 Gauss are shown in Figure 7.3. The results indicate that the magnetic moment of [Cu(trz)2]x decreases from a room temperature (300 K) value of 1.59 B . M . to 0.95 B . M . at 35 K. The moment then increases dramatically as temperature is reduced more, reaching a maximum value of 3.92 B . M . at 22 K. The magnetic moment subsequently begins to decrease again, as the temperature is lowered further, with a measured value of 1.33 B . M . at the lowest temperature studied of 2 K. This sudden change in the magnetic behaviour is also seen in 244 Figure 7.3. Magnetic susceptibility and magnetic moment versus temperature data at 500 Gauss for [Cu(trz)2]x. the magnetic susceptibility versus temperature plot. As the temperature is lowered from 35 K to 22 K the magnetic susceptibility value increases from 0.0037 cu^mol"1 to 0.088 cm3mol"1. As the temperature is decreased further, the magnitude of the magnetic 245 350 300 A 250 A 200 A 150 100 A 50 9 v o ? —I*-BV A • O A • • O 0 O 3^ o y o 0 o • • 10000 20000 30000 Field (G) 40000 50000 60000 Figure 7.4. Plot of magnetization versus applied magnetic field for [Cu(trz)2]x at 300 K (red O), 100 K (green • ) , 50 K (A), 25 K (V), 13 K (O), 4.8 K (green O), and 2 K (yellow • ) . susceptibility begins to level off, reaching an apparent maximum value of 0.11 cm3mol"1 at all temperatures below 7 K. The magnetic phase transition at 30 K prompted a magnetization study of the compound at several different temperatures. The results of these studies are shown as magnetization versus applied field plots in Figure 7.4. 246 At all temperatures above 30 K the magnetization versus applied field data behave linearly (up to the maximum applied field of 55 000 Gauss) with no magnetization observable at zero applied field. The data collected at 25 K , 13 K, 4.8 K and 2 K are not linear over the entire applied field range, and upon extrapolating the magnetization data to zero applied field, there is an observable intercept in the magnetization axis. It should be noted that the data shown in Figure 7.4 do not include any measurements at applied fields below 500 G. Even at the lowest temperatures examined, the material exhibits close to zero magnetization at zero applied field, provided that the compound was cooled to the target temperature in the absence of an applied external field (zero-field cooling). However, at temperatures below 30 K, a zero-field cooled sample will obtain a net magnetization upon the application of a small magnetic field. This magnetization remains even after the external magnetic field is removed. In order to destroy this net magnetization the temperature must be increased above 30 K (and the external field removed) or an opposing magnetic field (coercive field) of sufficient strength must be applied. To illustrate this better hysteresis magnetization studies were done at 4.8 K and the resulting data are shown in Figure 7.5. 247 -150 -I- , , , 1 , , , -20000 -15000 -10000 -5000 0 5000 10000 15000 20000 Applied magnetic field (G) Figure 7.5. Field dependence of [Cu(trz)2]x at 4.8 K. It is apparent from this magnetic data that [Cu(trz)2]x, like [Fe(imid)6(imidH)2]x described in the previous chapter, exhibits long-range magnetic ordering at low temperatures. This behaviour is consistent with a canted-spin, 3-dimensional structure. Unfortunately, without definitive structural information, the nature of this canting can in no way be quantified. However, a 3-dimensional structure for this material is not surprising. 248 The presence of three potentially donating nitrogen atoms on the triazolate rings generates many structural possibilities. One such possibility is shown in Figure 7.6. M M " M i j y j i ' i i i i j y j i i n " i i i j ^ j i i M ' N - N ^ N - N X < ? ^ N - N K ^ N > N M M M M M Figure 7.6. Possible 3-dimensional structure of [Cu(trz)2]x in which each copper(II) ion is six coordinate. 249 In the above representation, the structure is comprised of linear chains of copper(II) ions similar to the linear chain pyrazolate polymers described earlier in this thesis. These chains are cross linked to other chains through bonds between the unique nitrogen atoms on the triazolate rings and the copper(II) centers. The result is a 3-dimensional material in which the copper ions are octahedrally coordinated, and each nitrogen atom on the triazolate rings is coordinated to a metal center. This proposed structure is consistent with the electronic spectroscopy data in which the d—»d transition observed (17 400 cm"1) is in agreement with copper(II) ions in an octahedral CuN 6 chromophore geometry (257). However, such a structure does not account for the observed canted-spin magnetic behaviour. In addition, : H nmr measurements made previously by others (298) suggest that there are two different proton environments in this material. To explain this the authors proposed a different structure for [Cu(trz)2]x shown in Figure 7.7. It is not difficult to see that this proposed chain structure can be cross-linked to generate a higher dimensional material by introducing bonds between the available nitrogen atoms in the triazolate rings and copper ions in other chains. 250 Figure 7.7. Structure of [Cu(trz)2]x proposed by Inoue and Kubo (298) to account for two different proton environments in this material. Without any additional structural information it is obviously not possible to correlate the magnetic data to the structure. It is useful, however, to compare the magnitude of some of the magnetic properties with those of other, previously reported materials. Specifically, the temperature at which the long range magnetic ordering is observed (Tc), the residual magnetization at zero applied field and the coercive field measured for the complex are of interest. Values for these parameters for several materials are tabulated in Table 7.1. 251 Table 7.1. Selected magnetic parameters for canted-spin magnetic materials. Material Magnetic phase transition temperature, T c (K) Coercive Field (G) Remnant Magnetization (cm3Gmol"1) Reference [Cu(trz)2]x 30 2700a 53a This work [Fe(imid)6(imidH)2]x 18 200a 2500a This work [Fe(4-imidacetate)2]x-2CH3OH 15 6200b 1200b (34) [Fe(biimid)] xCH 3OH 1/2H 20 25 2000b 950b (35) a measured at 4.8 K. b measured at 4.2 K. The magnitude of the remnant magnetization at zero field is a measure of the strength of the magnet. By comparing the remnant magnetization for the materials listed in Table 7.1, it can be seen that the copper triazolate complex is a relatively weak magnet. The size of this magnetization depends on a number of factors. Firstly, the magnetization is proportional to the magnetic susceptibility of the material, which in turn depends on the total spin of the ions present. The ions responsible for the paramagnetism in the systems being compared here are copper(II), with a total spin of/2, and high-spin iron(II), with a total spin of 2. It is therefore expected that with everything else being equal, the iron materials will have a larger magnetization than the copper system. Another important factor in comparing remnant magnetizations in canted-spin materials is the extent of the canting of the magnetic orbitals. To clarify this point two possibilities for a 2-D system are represented in Figure 7.8. For 252 systems in which the magnetic orbitals align in nearly an antiparallel manner (a), the net magnetic moment is relatively small and thus a small remnant magnetization is expected. In materials in which the canting is severe (b), a significantly larger magnetic moment results and therefore a larger remnant magnetization is expected. /\ l\ !\ / \ / \ y \ i\ !\ i\ y \ /\ y \ J\ !\ J\ /\ /\ /\ (b) Small net moment (a) Large net moment Figure 7.8. Two potential alignments for canting of magnetic orbitals leading to (a) a small net magnetization and (b) a large net magnetization. The magnitude of the coercive field in these systems offers information about the type of magnet these materials generate. A large coercive field signifies that the remnant magnetization is difficult to remove. In other words, in order to destroy the magnetization of the material, a significantly large and opposing applied magnetic field is required. Such a magnet is termed a hard magnet. Conversely, a soft magnet is one in which the remnant magnetization is relatively easy to destroy, and is identified by a small coercive field. [Cu(trz)2]x appears to be a relatively hard magnet at 4.8 K, requiring an opposing field of 2700 cir^Gmol"1 in order to destroy the remnant magnetization. 253 The remaining magnetic property compared in Table 7.1 is the magnetic phase transition temperature. This can be described as the highest temperature at which the long range magnetic ordering is observed and the temperature below which the material begins to behave as a permanent magnet. [Cu(trz)2]x has one of the highest T c values reported, to date, for all "canted-spin" transition metal complexes. 7.3 S U M M A R Y A N D CONCLUSIONS Reactions between copper metal shot and 1,2,4-triazole in the presence of dioxygen result in the formation of a purple powder of formulation [Cu(trz)2]x. The structure of this complex has not been determined, but its physical, thermal and magnetic properties are consistent with a polymeric material. Extensive magnetic studies on this compound reveals the presence of antiferromagnetic exchange coupling between copper(II) centers. Moreover, at temperatures below 30 K, long range magnetic ordering, possibly due to a canted-spin structure, occurs. This 3-D magnetic ordering results in a material which is a weak ferromagnet. Hysteresis magnetization studies at 4.8 K reveal a remnant magnetization of 53 cm3Gmor1 and a coercive field of 2700 G. The magnetic transition temperature of 30 K is among the highest reported for a molecular compound. One complex exhibiting a higher Tc (35.8 K) is chromium(II) methylphosphonate (39). 254 Chapter 8 GENERAL SUMMARY AND SUGGESTION FOR FUTURE WORK 8.1 G E N E R A L S U M M A R Y In this study, several polymeric and oligometallic transition metal azolate complexes have been prepared and characterized. In particular, the magnetic properties of these materials were investigated. A series of six dimetallic nickel(II) complexes, [CpNi(4-X-3,5-diMepz)]2 (X = H , C H 3 , CI, Br, N 0 2 ) and [CpNi(3,5-F6diMepz)]2, has been prepared (The X = H derivative had been prepared previously, but not characterized structurally). These materials consist of two nickel(II) centers doubly bridged by substituted pyrazolate ligands and end-capped with Cp groups. The complexes were isolated as single crystals and single crystal X-ray diffraction studies were performed on three representative compounds. The compounds have been shown to be isostructural, with the central Ni(N-N) 2 Ni six membered ring existing in a boat-like conformation in the solid state. Magnetic studies reveal that the compounds are diamagnetic with a small percentage of structural paramagnetic impurity. A series of four trimetallic nickel(II) complexes, [CpNi(4-X-3,5-diMepz)2]2Ni (X = H , C H 3 , CI, Br) has also been prepared and characterized. The complexes are similar to the dimetallic complexes in that double substituted pyrazolate bridges link the nickel(II) centers 255 and the short chain is end-capped with Cp groups. Like the dimetallic nickel(II) species, these trimetallic complexes were obtained in a form suitable for single crystal diffraction studies. Two representative structures have been solved, revealing a square planar central nickel ion linked to two other nickel(II) ions via a pair of double pyrazolate bridges. In the solid state both of the Ni(N-N) 2 Ni rings in each complex are present in a boat-like conformation. Magnetic studies reveal that the complexes exhibit the same magnetic behaviour as the dimetallic systems. This is expected as the outer two nickel(II) ions are in an identical environment to the nickel(II) ions in the dimetallic complexes, while the central nickel(II) chromophore geometry is square planar and thus diamagnetic. Along with the oligometallic nickel(II) complexes isolated, a series of seven polymeric nickel(II) azolate compounds, [Ni(4-Xpz) 2] x (X = H , CI), [Ni(indz)2]x and [Ni(4-X-3,5-diMepz) 2] x (X = H , C H 3 , CI, Br), has been obtained. Attempts to prepared these compounds in a form suitable for X-ray diffraction studies were unsuccessful. However, there is considerable evidence that these materials consist of infinite chains of Ni(II) ions doubly bridged by the substituted pyrazolate ions. The series of seven polymers is divisible into two categories; one in which the nickel complexes are paramagnetic, and the other in which the polymers are diamagnetic. The three diamagnetic complexes, [Ni(4-Xpz) 2] x (X = H , CI) and [Ni(indz)2]x, are thought to contain square planar N i N 4 chromophores while the paramagnetic materials, [Ni(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI, Br), are thought to contain pseudo-tetrahedral N1N4 chromophores. Steric interactions between methyl 256 substituents on the 4-X-3,5-diMepz derivatives are thought to be responsible for the tetrahedral geometry of the N i N 4 chromophore geometry in these compounds. Variable temperature magnetic studies on the paramagnetic series indicate the presence of antiferromagnetic exchange (J values ranging from -14 cm"1 to -16 cm"1 (|4J5 2 | = 64 cm"1 to 76 cm"1)) between nickel(II) ions in the chains. The magnitude of the magnetic exchange observed in these polymers is comparable to the exchange observed in similar previously studied systems of copper(II) and cobalt(II). A series of six manganese(II) substituted pyrazolate compounds, [Mn(4-Xpz)2(4-XpzH)] x (X = CI, Br) and [Mn(4-X-3,5-diMepz)2]x (X = H , C H 3 , CI, Br), has been prepared and characterized. Indirect evidence suggests that these materials, like the nickel(II) complexes, are infinite chains of Mn(II) ions doubly bridged by the substituted pyrazolate ions. In the case of the two 4-Xpz derivatives, an additional neutral 4-XpzH ligand is thought to coordinate to'the manganese ions giving rise to high spin, five-coordinate Mn(II) centers in the chains. For the most part the manganese polymers are very sensitive to normal atmospheric conditions. However, the [Mn(4-Clpz)2(4-ClpzH)]x complex is relatively stable, perhaps due to the presence of the additional coordinated neutral 4-ClpzH. Variable temperature magnetic studies on these materials reveal antiferromagnetic exchange interactions between the Mn(II) ions in the chains. The magnitude of this magnetic exchange for the [Mn(4-X-3,5-diMepz)2]x complexes (/value range: -1.2 cm"1 to -2.1 cm"1 (|4t/S'2j = 30 cm"1 to 52 cm"1)) is only slightly less than those observed in the nickel(II) 257 analogues. The magnitude of the magnetic exchange interaction is lower for the five coordinate Mn(II) chains, most likely due to the presence of the non-bridging ligand on each manganese ion. An attempt to produce a copper(II) polymer, [Cu(3,5-F6diMepz)2]x, resulted, instead, in the formation of a mixed valence (two Cu(II) and one Cu(I)) green material of formulation Cu3(3,5-F6diMepz)5. This complex was isolated as single crystals and an X-ray study revealed that the compound consists of a triangular arrangement of the copper ions linked by the substituted pyrazolate ligands. The complex can be described as a copper(II) dimer in which the two Cu(II) ions are linked by a triple pyrazolate bridge and fourth bridge involving the remaining two pyrazolate ligands and the copper(I) ion. Magnetic studies reveal very strong antiferromagnetic coupling between the two copper(II) ions. Exposure of this green material to dioxygen and water results in the formation of an insoluble purple powder. Indirect evidence suggests the presence of bridging O H groups in the purple material which has radically different magnetic properties from the parent green complex; variable temperature magnetic measurements show the magnetic moment to have an interesting temperature dependence. As the temperature is reduced from room temperature to 110 K the magnetic moment decreases. As the temperature is decreased further (to 60 K) the magnetic moment increases, and finally, as the temperature is reduced even more (to a minimum, temperature studied of 2 K) the magnetic moment decreases again. This behaviour suggests the presence of uncompensated spins. 258 A polymeric iron compound of formulation [Fe3(imid)6(imidH)2]x was prepared and characterized. In this case single crystals suitable for X-ray diffraction were obtained and the solid state structure has been solved. The complex consists of both octahedral and tetrahedral iron(II) ions singly bridged by imidazolate ions (two neutral coordinated imidazole ligands occupy the two remaining sites on the octahedral iron centers). The compounds is best described as consisting of a series of infinite chains of tetrahedral iron(II) centers singly bridged by imidazolate groups. Each of these chains is cross-linked via two octahedral iron(II) centers to a total of six other chains of tetrahedral Fe(II) ions, generating a complicated three dimensional structure. Variable temperature magnetic studies reveal the presence of antiferromagnetic exchange, predominantly along the tetrahedral chains (the value of J was determined to be between -2.3 cm"1 and -2.5 cm"1). At temperatures below 18 K, long-range magnetic ordering is observed and the material becomes a weak ferromagnet. A hysteresis loop (coercive field = 200 G, remnant magnetization = 2500 cm3Gmol"1) was obtained. A canted-spin structure has been used to explain this observed magnetic behaviour. Two triazolate complexes of formulation [M(trz)2]x ( M = Mn, Cu) have been prepared. Neither compound has been obtained in a form suitable for single crystal X-ray diffraction studies. Variable temperature magnetic measurements reveal that both of these complexes exhibit antiferromagnetic exchange. However, in the copper system, long-range magnetic ordering is observed at temperatures below 30 K and, like the iron imidazolate 259 complex, the material becomes a weak ferromagnet (coercive field = 2700 G, remnant magnetization = 53 cn^Gmol"1) at these temperatures. This long range magnetic ordering is likely a result of a canted spin structure. As anticipated, imidazolate and triazolate ligands are capable of bridging transition metal centers to generate 3-dimensional polymeric complexes. The two low-temperature molecular magnets prepared in this work illustrate that desirable long-range magnetic ordering is achievable in these 3-D materials. The hysteresis behaviour observed for these compounds may lead to applications for this class of material if the long-range magnetic ordering becomes achievable at more convenient temperatures. The potential applications of such materials include molecular switches and data storage devices. Continued attempts to prepare three dimensional complexes incorporating paramagnetic centers will undoubtedly lead to the discovery of stronger molecular magnets and higher long-range ordering temperatures. However, at this point it is still not possible to rationally design a system that will have these desirable properties. Nonetheless it is feasible to strategically select ligands that tend to generate 3-D structures, and thus increase the likelihood of achieving these goals. Finally, metallocenes have been shown to be an excellent source of metal(II) ions for the preparation of transition metal polymers. 260 8.2 SUGGESTIONS FOR FUTURE WORK The complexes and preparation methods investigated in this work present many possibilities for future studies. Firstly, in regard to the nickel(II) and manganese(II) polymers, continued attempts to grow single crystals large enough for X-ray diffraction studies could be carried out. In addition, dimetallic and trimetallic manganese complexes analogous to the nickel materials could be targeted. Secondly, considering the interesting magnetic properties exhibited by both the purple decomposition product of Cu3(3,5-FediMepz)5 and the polymer of formulation [Cu(trz)2]x, further attempts at obtaining single crystals of these materials is definitely worthwhile. In regards to the 3-D iron polymer, attempts to remove thermally the coordinated neutral imidazole may generate a new, related compound with interesting magnetic properties. In addition, other derivatives of the imidazole ligand should be targeted in an attempt to extend the range of systems studied. 261 Chapter 9 EXPERIMENTAL 9.1 INTRODUCTION The details of the experimental procedures for the syntheses of the isolated compounds are described here. The apparatus and instruments used for physical measurements are also described. 9.2 SYNTHESES The majority of the chemicals used in this work were of reagent grade and used without additional purification. Several of the compounds discussed are sensitive to dioxygen and/or water vapour and were handled only in the absence of these gases. This was typically done by utilizing Schlenk techniques and/or by handling the compounds in a Vacuum Atmospheres Corporation Model H E 43-2 Dri-Lab glovebox under a dinitrogen atmosphere. All solvents used with these air-sensitive compounds were free of water and deoxygenated. The solvents were dried using the following procedures: THF was refluxed with sodium metal and benzophenone and distilled under a dinitrogen atmosphere; benzene was refluxed with potassium metal and distilled under dinitrogen; acetonitrile was refluxed 262 separately with both phosphorus pentoxide and calcium hydride and then distilled under dinitrogen; hexanes were refluxed with sodium metal and benzophenone and distilled under dinitrogen; and diethyl ether was refluxed with calcium hydride and distilled under a dinitrogen atmosphere. Other solvents including deuterated N M R solvents were dried using molecular sieves and deoxygenated by freeze-pump-thaw cycles on a vacuum line. 9.2.1 A Z O L E DERIVATIVES Pyrazole (4-HpzH), 3,5-dimethylpyrazole (4-H-3,5-diMepzH), indazole (indzH), imidazole (imidH), and 1,2,4-triazole (trzH) were obtained commercially (Aldrich). The remaining substituted azole ligands described were prepared by way of literature preparations or modifications of these literature methods. 9.2.1.1 4-CHLOROPYRAZOLE, (4-ClpzH) 4-Chloropyrazole was prepared using the method of Hiittel et al. (299). Pyrazole (13.6 g, 0.200 mmol) was dissolved in glacial acetic acid. Aqueous NaOCl (617 mL of 0.324 M , 0.200 mol) was added, with stirring, to the resulting solution. The mixture was stirred overnight at room temperature and then neutralized with Na2C03. The 4-ClpzH was extracted from this neutralized solution with 4x300mL portions of CH2CI2. The extracts were subsequently combined and the solvent, was removed under vacuum. The solid residue was dissolved in dilute aqueous NaOH. The 4-ClpzH was then extracted with 3x75 263 mL portions of CH2CI2. These extracts were combined, the solvent was removed and the compound dried under vacuum yielding 18.1 g (88%) of 4-chloropyrazole as a white powder. Mp 75-76°C (lit. (299) 76-77°C). Anal, calcd. for C 3 H 3 C1N 2 : C 35.1, H 3.0, N 27.3; found C 35.3, H 2.9, N 27.4. N M R spectroscopy (*H in CDC13): 5 7.57 (s, 2, CH), 11.13 (br s, 1,NH). 9.2.1.2 4 -BROMOPYRAZOLE, (4-BrpzH) Pyrazole (13.6 g, 0.200 mol) was dissolved in 50 mL of boiling water. Bromine (32.0 g, 0.200 mol) was added dropwise with stirring until the yellow colour of Br 2 persisted. The mixture was refluxed for 1 hr and subsequently allowed to cool. The solution was then made slightly basic by the addition of concentrated NaOH (aq). This led to the precipitation of a white solid which was isolated by vacuum filtration, washed with small portions of cold water and dried in air. Any NaBr present was removed by dissolving the solid obtained in a minimum amount of diethyl ether and passing the solution through a fine ground glass filter. The diethyl ether was then removed under vacuum yielding 22.3 g of 4-bromopyrazole (76%) as a white powder, mp 92-94°C (lit. (300) 96-97°C). Anal. calcd. for C 3 H 3 B r N 2 : C 24.5, H 2.1, N 19.1; found: C 24.1, H 2.2, N 18.9. N M R spectroscopy ( XH in CDC13): 5 7.63 (s, 2, CH), 12.1 (br s, 1, H). 264 9.2.1.3 4-CHLORO-3,5-DIMETHYLPYRAZOLE, (4-Cl-3,5-diMepzH) 4-Chloro-3,5-dimethylpyrazole was prepared in a method similar to that used in the synthesis of 4-chloropyrazole described earlier. 3,5-Dimethylpyrazole (20.7 g, 0.215 mol) was dissolved in 200 mL of water to which 34 mL of glacial acetic acid had been added. To this mixture NaOCl (575 mL of 0.375 M , 0.215 mol) was added and the mixture was stirred continuously at room temperature for 12 hours. During this time a white precipitate formed. The solution was neutralized with Na2C0 3 and subsequently made slightly basic by the addition of a few drops of concentrated aqueous ammonia. The solid product was isolated by suction filtration, washed with several small portions of ice cold water and dried under vacuum for 12 hours. The product was then purified by vacuum sublimation (130°C) yielding 24 g (86%) of 4-chloro-3,5-dimethylpyrazole as a white powder, mp 114°C (lit. (301) 112.5-113.5°C). Anal, calcd. for C 5 H 7 C1N 2 : C 46.0, H 5.4, N 21.4; found: C 46.2, H 5.3, N 21.1. N M R spectroscopy (TT in CDC13): 5 2.25 (s, 6, CH 3 ) , 9.70 (br s, 1, NH). 9.2.1.4 4-BROMO-3,5-DIMETHYLPYRAZOLE, (4-Br-3,5-diMepzH) 3,5-Dimethylpyrazole (40 g, 0.42 mol) was dissolved in 1.2 L of warm water. Bromine (23 mL, 0.42 mol) was added dropwise, with stirring, to the solution resulting in a reddish brown oil forming below the aqueous layer. The mixture was neutralized with 20 mL of concentrated aqueous NaOH resulting in the precipitation of an off-white solid. The solid was collected by vacuum filtration and the remaining solution was cooled to 5°C for 265 12 hours. Additional solid precipitated out of solution and was isolated by vacuum filtration. The two solid fractions were combined and recrystallized from toluene (75 mL), washed with several portions of ice cold petroleum ether and dried under vacuum for 12 hours. The product was further purified by vacuum sublimation (140°C ) yielding 63 g (89%) of 4-bromo-3,5-dimethylpyrazole as a white powder, mp 116-117°C (lit. (302) 118°C). Anal, calcd. for C 5 H 7 B r N 2 : C 34.3, H 4.0, N 16.0; found: C 34.5, H , 4.2, N 16.1. N M R spectroscopy in CDC13): 5 2.20 (s, 6, CH 3 ) , 10.8 (br s, 1, NH). 9.2.1.5 3,4,5-TRIMETHYLPYRAZOLE, (4-CH3-3,5-diMepzH) 3,4,5-Trimethylpyrazole was prepared using the method of Chambers et al. (303) Sodium metal pieces (29.9 g, 1.30 mol) were added slowly to 400 mL of anhydrous methanol under a dinitrogen atmosphere. Upon completion of this reaction, pentane-2,4-dione (129 mL, 1.25 mol) was added to the solution and the mixture was allowed to cool. To this solution, iodomethane (86 mL, 1.39 mol) was added dropwise, with stirring, over a 1 hour period. The methanol was removed from the mixture by fractional distillation (bp 60°C at 540 rnmHg). The product was isolated by distillation (bp 161 °C at 540 mmHg) and dried at room temperature over H2SO4 for 24 hours yielding 98.0 g (68.7%) of 3-methylpentane-2,4-dione as a pale yellow liquid. This liquid was added dropwise, with stirring, to a solution of hydrazine (35%, 75 mL, 0.86 mol), 2 mL of glacial acetic acid, and 150 mL of water (15°C). The mixture was cooled to 5°C for 3 hours resulting in the 266 precipitation of a white solid. The white solid was collected by vacuum filtration, washed with ice cold water and dried under vacuum for 12 hours yielding 86 g (92%) of 3,4,5-trimethylpyrazole, mp 131°C (lit. (304) 137-138°C). Anal, calcd. for C 6 H i 0 H 2 : C 65.4, H 9.2, N 25.5; found: C 65.3, H 9.4, N 25.7. N M R spectroscopy (*H in CDC13): 5 1.90 (s, 3, CH 3 ) , 2.15 (s, 6, CH 3 ) , 9.7 (br s, 1, NH). 9.2.1.6 4-NITRO-3,5-DIMETHYLPYRAZOLE, (4-N02-3,5-diMepzH) 4-Nitro-3,5-dimethylpyrazole was prepared using the method of Morgan and Ackerman (302). 3,5-dimethylpyrazole (15.0 g, 0.156 mol) was dissolved in 30 mL of concentrated H 2 S 0 4 at 0°C. Concentrated H N 0 3 (18 mL) was then added slowly, with stirring, followed by the addition of another 60 mL of concentrated H 2 S 0 4 . The solution was left at room temperature for 12 hours and was heated to 100°C for an additional 5 hours. The solution was allowed to cool to room temperature and then poured onto 300 mL of ice. The mixture was neutralized by dropwise addition of concentrated NaOH (aq) resulting in the precipitation of a white solid. This solid was collected by suction filtration and washed with several small portions of ice cold water. The product was recrystallized from CH 2 C1 2 (120 mL) and dried under vacuum overnight yielding 22 g (95%) of 4-nitro-3,5-dimethylpyrazole as a white solid, mp 123°C (lit. (302) 126°C). Anal, calcd. for C 5 H 7 N 3 0 2 : C 42.6, H 5.0, N 29.8; found: C 42.5, H 5.0, N 29.8. N M R spectroscopy (*H in CDC13): 8 2.60 (s, 6, CH 3 ) , 10.50 (br s, 1, NH). 267 .9.2.1.7 3,5-BIS (TRJFLUOROMETFFYL)P Y R A Z O L E , (3,5-F6diMepzH) 3,5-bis(trifluoromethyl)pyrazole was prepared by the method described by Trofimenko (305). l,5-bis(trifluoro)-2,4-pentadione (25 g, 0.12 mol), hydrazine (13.3 g, 0.15 mol) and 1 drop of acetic acid were dissolved in 175 mL of ethanol. The mixture was refluxed for 18 hours and the ethanol was removed by flash evaporation. The resulting brown oil was placed in a small distillation apparatus and refluxed (150°C oil bath) overnight. During this time the volume of the oil decreased and the remaining brown liquid solidified upon cooling to room temperature. The desired product was separated and from the brown residue by vacuum sublimation yielding 6.8 g (28%) of a white crystalline solid, mp 80-81°C (lit. (305) 84°C). Anal, calcd. for C 5 H 2 N 2 F 6 : C 29.2, H 1.0, N 13.6; found: C 29.1, H 1.2, N 13.7. 9.2.2 M E T A L L O C E N E S Biscyclopentadienyliron (ferrocene) was obtained commercially (Strem Chemicals). The other metallocenes used in this work were synthesized using literature preparations or modifications thereof. Cyclopentadiene was freshly prepared by thermal de-dimerization, or cracking, of commercial dicyclopentadiene (Aldrich). The syntheses were carried out under a dry dinitrogen atmosphere. 268 9.2.2.1 BISCYCLOPENTADffiNYLNICKEL(II) (NICKELOCENE), (Ni(Cp)2) Nickelocene was prepared by the method described by Cordes (306). Nickel powder (29.4 g, 0.5 mol) and 1,2-dimethoxyethane (500 mL) were combined under dinitrogen in a three-necked 1 liter flask. Bromine (27.3 mL, 0.5 mol) was added dropwise, with stirring, forming a yellow etherate of nickel(II) bromide in an exothermic reaction. The solution was left to cool and the solvent was removed under vacuum revealing a yellow solid. The flask was refilled with dinitrogen, and diethylamine (400 mL) was added resulting in a blue solution. The mixture was treated dropwise with freshly prepared cyclopentadiene (98 mL, 1.2 mol) at which point the solution turned green. This green solution was left to stir at room temperature overnight. The solvent was removed under vacuum leaving a mixture of the desired product and diethylammonium bromide. This solid mixture was transferred under dinitrogen to thimbles of a large Soxhlet extraction apparatus. The biscyclopentadienylnickel(II) was continuously extracted with boiling hexane until no additional green material remained in the thimble (12 - 18 hours). The hexane solution was allowed to cool to room temperature and the crystallized portion of the product was collected by filtration. Additional nickelocene was isolated by the removal of the solvent under vacuum. Both portions of the solid green product were then transferred into a large sublimation apparatus and purification was performed by vacuum sublimation (100°C) yielding 45 g (48%) of nickelocene as large green crystals, mp 173°C (lit. (306) 173-174°C). 269 9.2.2.2 B I S C Y C L O P E N T A D I E N Y L M A N G A N E S E (MANGANOCENE) , (Mn(Cp)2) Manganocene was prepared using the method described by Wilkinson et al. (307). Manganese powder (27.5 g, 0.5 mol) and 1,2-dimethoxyethane were mixed in a three-necked 2 liter flask fitted with a motor stirrer and a dinitrogen inlet. The system was first purged with dry dinitrogen gas and then bromine (27.3 mL, 0.5 mol) was added dropwise forming a white suspension in the 1,2-dimethoxyethane. In a separate 1 liter three-necked flask fitted with a motor stirrer, a dinitrogen-inlet reflux condenser, and a pressure-equalized dropping funnel, a solution of 1 molar sodium cyclopentadienide in 1,2-dimethoxyethane was prepared. This solution was produced by first granulating 23 g (1 mol) of sodium metal in dry and dioxygen-free xylene (300 mL). To accomplish this the mixture was stirred vigorously as the solvent was refluxed. The xylene was subsequently decanted and 500 mL of 1,2-dimethoxyethane was added forming a suspension of sodium in the solvent. This suspension was treated dropwise with enough freshly prepared cyclopentadiene (123 mL, 1.5 mol) to dissolve all of the sodium, resulting in the formation of a violet solution of sodium cyclopentadienide. This sodium cyclopentadienide suspension was added slowly to the manganese bromide suspension with a rapid stream of dinitrogen passing through both flasks. The resulting mixture was refluxed for 4 hours, left to cool and then the solvent was removed under vacuum. Dinitrogen gas was admitted to the flask and the residue was transferred 270 into a large sublimation apparatus. The biscyclopentadienylmanganese (-45% yield) was purified by vacuum sublimation at 150°C and subsequently stored in a dinitrogen filled Schlenk tube. 9.2.3 NICKEL(II) P Y R A Z O L A T E S Unless otherwise stated, the source of nickel(II) in the syntheses described in this section was nickelocene which had been freshly vacuum-sublimed prior to use. 9.2.3.1 POLY-BIS(|a-PYRAZOLATO-N,N')NICKEL(II), ([Ni(4-Hpz)2]x) This reaction is a variation of the method described by Blake et al. (82). Nickelocene (0.25 g, 1.3 mmol) was mixed with an excess of pyrazole (2.32 g, 34 mmol) in a round bottom flask under a dinitrogen atmosphere. The flask was fitted with a condenser and the mixture was heated (110°C) with stirring, causing the pyrazole to melt. The initial green colour of the solution faded as the reaction proceeded and after 48 hours a yellow to orange precipitate had formed. The mixture was allowed to cool and solidify. The excess pyrazole was dissolved by the addition of several portions of THF. The remaining yellow solid was isolated by vacuum filtration and washed with additional THF and hexanes. The product was dried under vacuum overnight yielding 0.21 g (83%) of [Ni(4-Hpz)2]x as a yellow to orange powder. Anal, calcd. for N i C 6 H 6 N 4 : C 37.4, H 3.1, N 29.0; found: C 37.4, H 3.3, N 28.9. 271 9.2.3.2 P0LY-BIS( |^ -4-CHL0R0PYRAZ0LAT0-N,N' )NICKEL(II ) , ([Ni(4-Clpz)2]x) Nickelocene (0.25 g, 1.3 mmol) was mixed with an excess of 4-chloropyrazole (1.4 g, 14 mmol) in a Carius tube. The tube was flame-sealed under vacuum and the mixture was heated in an oven at 140°C. Upon melting of the 4-ClpzH, the nickelocene dissolved and a green solution was produced. Upon continued heating the green colour slowly disappeared and a yellow solid precipitated out of the solution. After 48 hours the mixture was allowed to cool and solidify. The tube was cracked open, the contents were removed and the soluble portions were dissolved in THF. The remaining yellow solid was isolated by vacuum filtration and was washed with additional THF and hexanes. The product was dried under vacuum overnight yielding 0.32 g (94%) of Ni(4-Clpz) 2] x as a yellow to orange powder. Anal, calcd. for N i C 6 H 4 N 4 C l 2 : C 27.5, H 1.5, N 21.4; found: C 27.8, H 1.4, N 21.2. 9.2.3.3 POLY-BIS(|a-3,5-DIMETHYLPYRAZOLATO-N,N')NICKEL(II) , ([Ni(4-H-3,5-diMepz)2]x) Nickelocene (0.27 g, 1.4 mmol) was mixed with 3,5-dimethylpyrazole (2.32 g, 24 mmol) and the reactants were placed in a Carius tube. The tube was flame-sealed under vacuum and the reaction vessel was put in an oven at 150°C. Initially the colour of the molten 3,5-dimethylpyrazole solution, in which nickelocene was dissolved, was green. After about one hour the green colour had been replaced by a deep red colour. The 272 reactants were left undisturbed at this elevated temperature for 48 hours after which no colour remained in the molten phase, but a purple solid had formed inside the tube. The mixture was allowed to cool and solidify, and then the Carius tube was cracked open under a dinitrogen atmosphere. THF was added to dissolve any soluble portion and to extract the desired product from the vessel. The extractions were passed through a glass filter and the solid was washed with additional portions of THF and hexanes. The product was subsequently dried under vacuum overnight yielding 0.311 g (94.3%) of [Ni(4-H-3,5-diMepz) 2] x as a purple powder. Anal, calcd. for N i C i 0 H i 4 N 4 : C 48.2, H 5.7, N 22.5; found: C 48.4, H 5.8, N 22.7. 9.2.3.4 POLY-BIS(^-4-CHLORO-3,5-DIMETHYLPYRAZOLATO-N,N')NICKEL(II) , ([Ni(4-Cl-3,5-diMepz)2]x) 4-Chloro-3,5-dimethylpyrazole (2.6 g, 20 mmol) was combined with nickelocene (0.41 g, 2.2 mmol) in a Carius tube. The tube was flame-sealed under vacuum and placed in an oven at 150°C. Upon melting of the 4-Cl-3,5-diMepzH a green solution was observed. After approximately 1 hour the solution had undergone a colour change from green to deep red. This red colour then slowly dissipated over two days during which a brown precipitate formed. After 48 hours no colour remained in the liquid phase. The mixture was then allowed to cool and solidify, and the tube was cracked open under a dry dinitrogen atmosphere. The contents of the tube were slurried with THF which both removed the desired product from the reaction vessel and dissolved the excess 4-Cl-3,5-273 diMepzH. The THF slurry was filtered and the solid portion washed with several portions of both THF and hexane to remove any remaining soluble portions. The product was subsequently dried under vacuum overnight yielding 0.54 g (80%) of [Ni(4-Cl-3,5-diMepz) 2] x as a light brown powder. Anal, calcd. for CioHi4N 4NiCl 2: C 37.8, H 3.8, N 17.6; found: C 38.1, H 3.8, N 17.5. 9.2.3.5 POLY-BIS(|x-4-BROMO-3,5-DIMETHYLPYRAZOLATO-N,N')NICKEL(II) , ([Ni(4-Br-3,5-diMepz)2]x) Nickelocene (0.3 g, 1.6 mmol) and 4-bromo-3,5-dimethylpyrazole (2.4 g, 14 mmol) were intimately mixed and placed in a Carius tube. The tube was sealed under vacuum in the usual fashion and the reaction vessel was placed in an oven (150°C) for 48 hours. The initial green colour due to the nickelocene in.molten 4-Br-3,5-diMepzH turned red in less than 1 hour., This colour faded slowly over the reaction time as a brown solid precipitated out of the melt. The mixture was cooled, the Carius tube was opened under a dry dinitrogen atmosphere, and the reaction mixture was washed with copious amounts of THF. The resulting slurry was filtered and the isolated solid washed with additional portions of THF and hexanes. The product was dried under vacuum for 12 hours yielding 0.55 g (86%) of [Ni(4-Br-3,5-diMepz)2]x as a light brown powder. Anal, calcd. for Ci 0 Hi 4 N4Br 2 Ni: C 29.5, H 3.0, N 13.8; found: C 29.9, H 3.0, N 14.0. 274 9.2.3.6 POLY-BIS(ix-3,4,5-TRIMETHYLPYRAZOLATO-N,N')NICKEL(II), ([Ni(4-CH3-3,5-diMepz)2]x) Nickelocene (0.5 g, 2.7 mmol) and 3,4,5-trimethylpyrazole (2.9 g, 27 mmol) were mixed and placed in a Carius tube. The vessel was flame-sealed under vacuum and placed in the oven at 150°C. Again the initial colour of solution upon the melting of the substituted pyrazole was green. The green colour persisted for less than 1 hour at which time the solution became red. After 48 hours the molten 3,4,5-trimethylpyrazole was pale yellow in colour and a purple precipitate had formed. The reaction vessel was cooled to room temperature and opened under a dinitrogen atmosphere. The mixture was extracted from the reaction vessel using THF and the extracts were passed through a filter to remove the soluble 3,4,5-triMepzH. The solid product was washed with additional amounts of THF and hexanes, and dried under vacuum for 15 hours. 0.7 g of [Ni(4-CH3-3,5-diMepz)2]x (97%) was prepared as a purple powder. Anal, calcd. for Ci 2Hi8N 4Ni: C 52.0, H 6.6, N 20.2; found: C 52.3, H 6.4, N 20.1. 9.2.3.7 POLY-BIS(n-iNDAZOLATO-N,N')NICKEL(II), ([Ni(indz)2]x) Ni(Cp) 2 (0.5 g, 2.7 mmol) and indazole (3.1 g, 26.5 mmol) were mixed with xylene (50 mL) in a 100 mL round bottom flask under a dinitrogen atmosphere. The flask was fitted with a condenser and the solution refluxed (140°C) for 72 hours. Initially the solution was green, but turned orange after ~2 hours of heating. During the reaction an orange solid 275 precipitated out of the solution. After the 72 hours the reaction mixture was allowed to cool and the solid portion was isolated by filtration. The product was washed with THF and hexanes, and dried under vacuum for 10 hours yielding 0.72 g (93%) of [Ni(indz)2]x as an orange solid. Anal, calcd. for Ci4H 1 0 N 4 Ni: C 57.4, H 3.4, N 19.1; found: C 57.1, H 3.6, N 18.9. 9.2.3.8 Di(r)-CYCLOPENTADIENYL-|a-3,5-DEVffiTHYLP Y R A Z O L A T O - N , N ' -NICKEL(II)), ([CpNi(4-H-3,5-diMepz)]2) The dimeric compound [CpNi(4-H-3,5-diMepz)]2 was prepared by a variation of the method of Blake et al. (82). Nickelocene (1.9 g, 0.010 mol) and 3,5-dimethylpyrazole (0.96 g, 0.010 mol) were dissolved in dry and dioxygen-ffee benzene under a dinitrogen atmosphere. The solution was stirred continuously at room temperature for 6 hours, with a colour change from green to red taking place after about 30 minutes. The benzene was removed by flash evaporation. The red solid residue was dried under vacuum and any remaining nickelocene or 4-H-3,5-diMepzH was removed by sublimation. The product was then redissolved in a minimum of dry benzene and single red crystals suitable for X-ray diffraction were obtained by slow evaporation of this benzene solution in a dinitrogen atmosphere. Yield: 1.6 g (42%) of [CpNi(4-H-3,5-diMepz)]2. Anal, calcd. for C 2 oH 2 4 N 4 Ni 2 : C 54.9, H 5.5, N 12.8; found: C 55.1, H 5.5, N 12.7. N M R spectroscopy ( XH in C 6 D 6 ) : 5 2.20 (s, 12, CH 3 ) , 5.20 (s, 10, C 5 H 5 ) , 5.55 (s, 2, CH(pz)). 276 9.2.3.9 Di(ri-CYCLOPENTADIENYL -n-4-CHLORO-3,5-DIMETHYLPYRAZOLATO-N,N' -NICKEL(II ) ) , ([CpNi(4-Cl-3,5-diMepz)]2) Nickelocene (0.25 g, 1.3 mmol) was mixed with 4-Cl-3,5-diMepzH (0.17 g, 1.3 mmol) and the mixture was dissolved in 100 mL of benzene. The solution was stirred for 12 hours with a colour change from green to red taking place after about 30 minutes. The benzene was removed by flash evaporation revealing a red solid residue. The solid was dried, and any remaining volatile reactants (Ni(Cp)2 and 4-Cl-3,5-diMepzH) were removed by exposing the sample to a dynamic vacuum overnight. Yield: 0.20 (59%) of [CpNi(4-Cl-3,5-diMepz)]2 as a red powder. Anal, calcd. for C 2 oH 2 2N 4 Cl 2 Ni 2 : C 47.4, H 4.4, N 11.0; found: C 47.8, H 4.3, N 11.0. N M R spectroscopy (TT in C 6 H 6 ) : 5 2.10 (s, 12, CH 3 ) , 5.00 (s, 10, C 5 H 5 ) . 9.2.3.10 Di(r | -CYCLOPENTADIENYL-u-4-BROMO-3,5-DIMETHYLPYRAZOLATO-N,N' -NICKEL(II ) ) , ([CpNi(4-Br-3,5-diMepz)]2) Nickelocene (0.5 g, 2.7 mmol) was combined with 4-bromo-3,5-dimethylpyrazole (0.47 g, 2.7 mmol) and the mixture was dissolved in benzene (100 mL). The resulting green solution turned red within 30 minutes and was left to stir for an additional 12 hours. The solvent was removed by flash evaporation and the excess volatile starting materials were removed by sublimation under vacuum overnight. The involatile red powder left over from the sublimation was the target compound [CpNi(4-Br-3,5-diMepz)]2. Anal, calcd. for 277 C 2 oH2 2N4Br 2 Ni2: C 40.3, H 3.7, N 9.4; found: C 40.7, H 3.7, N 9.4. N M R spectroscopy (lU in C 6 H 6 ) : 5 2.10 (s, 12, CH 3 ) , 4.95 (s, 10, C 5 H 5 ) . 9.2.3.11 Di(Ti-CYCLOPENTADffiNYL-|a-3,4,5-TRIMETHYLPYRAZOLATO-N,N'-NICKEL(II)), ([CpNi(4-CH3-3,5-diMepz)]2) Nickelocene (0.5 g, 2.7 mmol) and 4-CH 3-3,5-diMepzH (0.30 g, 2.7 mmol) were dissolved in 100 mL of benzene under a dinitrogen atmosphere. The mixture was stirred at room temperature for 18 hours and a colour change from green to red took place after approximately 30 minutes. The benzene was removed by flash evaporation and the residue was exposed to a dynamic vacuum for 12 hours to remove any of the remaining volatile starting materials. The red solid remaining was the desired product, [CpNi(4-CH3-3,5-diMepz)]2 (0.33 g, 52%). Anal, calcd. for C 2 2 H 2 8 N 4 N i 2 : C 56.7, H 6.1, N 12.0; found: C 57.1, H 6.3, 11.8. N M R spectroscopy ( 'H in CeHe): 5 1.62 (s, 6, CH 3 ) , 2.15 (s, 12, CH 3 ) , 5.22 (s, 10, C 5 H 5 ) . 9.2.3.12 Di( r i -CYCLOPENTADIENYL- | i -4-NITRO-3,5-DIMETHYLPYRAZOLATO-N,N'-NICKEL(II)), ([CpNi(4-N02-3,5-diMepz)]2) Nickelocene (0.5 g, 2.7 mmol) and 4-N0 2-3,5-diMepzH (0.37 g, 2.7 mmol) were dissolved in benzene (100 mL). The solution was stirred at room temperature for 12 hours under a dinitrogen atmosphere. 30 minutes into the reaction the colour had turned from the 278 initial green to a red-brown colour. The desired product was obtained by first removing the solvent by flash evaporation and subsequently removing the remaining volatile starting materials by vacuum sublimation. [CpNi(4-N02-3,5-diMepz)]2 was then collected as a brown powder (0.35 g, 49%). Anal, calcd. for C 2 oH 2 2 N 6 04Ni 2 : C 45.5, H 4.2, N 15.9; found: C 45.7, H 4.2, N 16.0. N M R spectroscopy ( 'H in C 6 H 6 ) : 5 2.20 (s, 12, CH 3 ) , 5.20 (s, 10, C 5 H 5 ) . Single crystals suitable for X-ray diffraction were obtained by recrystallizing from benzene. About 20 mg of the brown powder was dissolved in a minimum of benzene in a dinitrogen atmosphere. The resulting red to brown solution was placed in a 20 mL Erlenmeyer flask and the solvent was allowed to evaporate slowly. Over the period of one week the solvent completely evaporated revealing a mixture of brown powder and red irregular single crystals. 9.2.3.13 Di(r | -CYCLOPENTADIENYL-u-3,5-BIS(TRffLOUROMETHYL)PYRAZOLATO-N,N ' -NICKEL(II)), ([CpNi(3,5 -F6diMepz)]2) In a 100 mL round bottom flask nickelocene (0.50 g, 2.7 mmol) was mixed with 3,5-bis(trifluoromethyl)pyrazole (0.44 g, 2.2 mmol). Under a dinitrogen atmosphere, the mixture was dissolved in 75 mL of benzene and the solution was stirred for 18 hours at room temperature. After approximately 30 minutes the colour of the solution changed from 279 green to red. The product was isolated by allowing the benzene to evaporate and then purified by removal of the excess nickelocene under vacuum. Upon complete sublimation of the excess Ni(Cp)2, the [CpNi(3,5-F6diMepz)]2 remained as very thin red crystal plates. Anal, calcd. for C 2 oH 1 2 F 1 2 N 4 Ni 2 : C 36.75, H 1.85, N 8.6; found: C 37.1, H 2.0, N 8.6. N M R spectroscopy ( L H in CeHe): 5 3.37(s, 10, C 5 H 5 ) , 6.67 (s, 2, CH(pz)). Single crystals suitable for X-ray diffraction were obtained by recrystallization from benzene. A saturated solution of [CpNi(3,5-F6diMepz)]2 was prepared by dissolving approximately 30 mg of the red solid in a minimum of the solvent under dry dinitrogen conditions. The solution was then placed in a small vial which was fitted with a plastic cap. Several small holes were punctured into the cap to allow the benzene to evaporate very slowly. Approximately 2-3 weeks later the solvent had completely evaporated and the product had formed as red needle crystals. 9.2.3.14 [CpNi(4-Cl-3,5-diMepz)2]2Ni Nickelocene (0.33 g, 1.7 mmol) was mixed under a dinitrogen atmosphere with 4-Cl-3,5-diMepzH (2.3 g, 17 mmol) and placed in a 100 mL round bottom flask fitted with a condenser. The mixture was heated, with stirring, to 140°C causing the 4-Cl-3,5-diMepzH to melt and the nickelocene to dissolve in this molten ligand. The initial green colour of the solution turned red after about 30 minutes and the mixture was allowed to cool and solidify after 3 hours. The residue was washed with several portions of THF to remove the excess 280 4-Cl-3,5-diMepzH and the suspension was passed through a filter to collect the red solid. The filtercake was then washed with hexanes and dried under vacuum for 12 hours yielding 0.17 g (36%) of [CpNi(4-Cl-3,5-diMepz)2]2Ni as a red powder. Anal, calcd. for C30H34N8M3CI4: C 43.7, H 4.2, N 13.6; found: C 43.7, H 4.2, N 13.9. N M R spectroscopy (TT in CeHe): 5 2.15 (s, 12, CH 3 ) , 2.18 (s, 12, CH 3 ) , 5.65 (s, 10, C 5 H 5 ) . Single crystals suitable for X-ray diffraction were obtained for this product by recrystallization in benzene. A few mg of the red solid was dissolved in a minimum amount of benzene and placed in a sealed glass vessel under vacuum. The solution was left undisturbed for 8 months allowing the crystals to form. The tube was then cracked open under dinitrogen and the crystals carefully isolated and dried under vacuum. 9.2.3.15 [CpM(4-H-3,5-diMepz)2]2Ni Nickelocene (0.5 g, 2.7 mmol) and 3,5-dimethylpyrazole (2.3 g, 24 mmol) were dissolved in 150 mL of hexane under dinitrogen. The green solution was heated (60°C), without stirring, for 24 hours. After about 15 minutes of heating the solution turned red, and orange-red crystals began to precipitate out of solution within 3 hours. The solution was allowed to cool and the hexane was decanted. The orange-red plate crystals (0.55 g, 30%) were washed with small portions of hexane and dried under vacuum for 12 hours. Anal, calcd. for C3oH38N8Ni3: C 52.5, H 5.5, N 16.3; found: C 52.6, H 5.5, N 16.2. N M R 281 spectroscopy ( XH in C 6 H 6 ) : 5 2.15 (s, 12, CH 3 ) , 2.20 (s, 12, CH 3 ) , 5.40 (s, 4, CH(pz)), 5.63 (s, 10, C 5 H 5 ) . 9.2.3.16 [CpNi(4-Br-3,5-diMepz)2]2Ni Nickelocene (0.5 g, 2.7 mmol) was mixed with 4-bromo-3,5-dimethylpyrazole (1.88 g, 10.8 mmol) under a dinitrogen atmosphere. The mixture was dissolved in benzene (75 mL) and heated (60°C) for 18 hours without stirring. The solution turned red after 15 minutes of heating and some red solid precipitated out of solution after a few hours. The solution was allowed to cool and about 40 mL of the benzene was removed by flash evaporation causing more red solid to come out of solution. The solid was isolated by suction filtration and was then washed with small portions of hexane. The product was dried under vacuum for 12 hours yielding 0.40 g (44%) of [CpNi(4-Br-3,5-diMepz)2]2Ni as tiny red crystals. Anal, calcd. for C 3oH 3 4N8Ni 3Br 4: C 35.9, H 3.4, N 11.1; found: C 36.0, H 3.4, N 11.3. N M R spectroscopy (TT in C 6 H 6 ) : 5 1.95 (s, 12, CH 3 ) , 2.1 (s, 12, CH 3 ) , 5.15 (s, 10, C 5 H 5 ) . 9.2.3.17 [CpNi(4-CH3-3,5-diMepz)2]2Ni Nickelocene (0.5 g, 2.7 mmol) and 3,4,5-trimethylpyrazole (1.5 g, 13.5 mmol) were combined and dissolved in 100 mL of benzene. The solution was initially green but turned red after approximately 15 minutes. The solution was heated under dinitrogen to 60°C for 282 18 hours, with a red precipitate forming after about 3 hours. The solution was subsequently allowed to cool and about 50 percent of the solvent was removed by flash evaporation. The remaining slurry was suction filtered and the red solid was washed with small portions of hexanes. The product was dried under vacuum for 12 hours yielding 0.22 g (33%) of [CpNi(4-CH 3-3,5-diMepz)2]2Ni as red microcrystals. Anal, calcd. for C34H46N8Ni3: C 55.0, H 6.2, N 15.1; found: C 55.2, H 6.3, N 15.0. N M R spectroscopy ( XH in C 6 H 6 ) : 6 1.55 (s, 12, CH 3 ) , 2.05 (s, 12, CH 3 ) , 2.23 (s, 12, CH 3 ) , 5.53 (s, 10, C 5 H 5 ) . 9.2.4 MANGANESE(II) AZOLATES In all of the manganese pyrazolates prepared, manganocene was used as the source of manganese(II). In each instance the manganocene was sublimed immediately prior to use to ensure its purity. Due to the reactivity of both the manganocene itself and the manganese containing products, the reactions and purifications were carried out in either a dry dinitrogen atmosphere or under vacuum. 9.2.4.1 POLY-BIS( |a-3,5-DIMETHYLPYRAZOLATO-N,N')MANGANESE(II) , ([Mn(4-H-3,5-diMepz)2]x) Manganocene (1.1 g, 5.9 mmol) and 3,5-dimethylpyrazole (5.19 g, 54 mmol) were intimately mixed and placed in a Carius tube. The tube was flame-sealed under vacuum and placed in an oven (150°C). The 4-H-3,5-diMepzH melted after a few minutes resulting in a 283 mixture of molten ligand and solid manganocene. It was not possible to determine the extent of the reaction by visual means as both the manganocene and desired product appear as beige solids in the colourless and clear molten 4-H-3,5-diMepzH. The mixture was therefore heated for 2 weeks to ensure complete reaction. Following this the tube was allowed to cool, and the excess ligand solidify, prior to the tube being opened under a dinitrogen atmosphere. The beige solid product was isolated by dissolving the excess 3,5-dimethylpyrazole in THF, benzene and hexanes. The slurry was vacuum filtered and the [Mn(4-H-3,5-diMepz)2]x collected on the filter. The product was dried, and any remaining volatile starting materials removed, under vacuum for 48 hours yielding 0.74 g (52%) of [Mn(4-H-3,5-diMepz)2]x as a beige powder. Anal, calcd. for Ci 0Hi 4N4Mn: C 49.0, H 5.8, N 22.9; found: C 49.0, H 5.7, N 23.0. 9.2.4.2 POLY-BIS( |a-4-CHLORO-3,5-DIMETHYLPYRAZOLATO-N,N')MANGANESE(II) , ([Mn(4-Cl-3,5-diMepz)2]x) Manganocene (0.25 g, 1.3 mmol) was mixed intimately with 4-chloropyrazole (1.4 g, 13 mmol) and placed in a Carius tube. The tube was flame-sealed under vacuum and subsequently heated to 160°C for two weeks. The 4-Cl-3,5-diMepzH melted after a few minutes forming a mixture of solid manganocene in the liquid pyrazole. The reaction was terminated by allowing the tube to cool and the pyrazole to solidify. The vessel was cracked open under a dinitrogen atmosphere and the mixture washed with THF, benzene and hexanes to dissolve the excess 4-Cl-3,5-diMepzH. The solutions were filtered to 284 collect the solid product. Any remaining volatile starting materials were removed by exposing the sample to a dynamic vacuum for 48 hours. Yield: 0.21 g (51%) of [Mn(4-Cl-3,5-diMepz)2]x as a white powder. Anal, calcd. for CioHi2N4MnCl2: C 38.2, H 3.9, N 17.8; found: C 38.5, H 3.8, N 17.7. 9.2.4.3 POLY-BIS( |^-4-BROMO-3,5-DIMETHYLPYRAZOLATO-N,N')MANGANESE(II) , ([Mn(4-Br-3,5-diMepz)2]x) Manganocene (0.25 g, 1.3 mmol) was mixed with 4-Br-3,5-dimethylpyrazole (2.5 g, 13 mmol) and, under a dinitrogen atmosphere, the mixture was transferred to a Carius tube. The tube was flame-sealed under vacuum prior to being placed in an oven (160°C) for a two week period. After a few minutes of heating the pyrazole melted giving a molten ligand and solid manganocene mixture. Following the two weeks of heating the tube was removed from the oven and allowed to cool. The tube was cracked open under a dinitrogen atmosphere and the solid mixture extracted by washing with benzene, THF and hexanes. The addition of these solvents dissolved the excess 4-Br-3,5-diMepzH and enabled the desired insoluble product to be collected by filtration. Any additional remaining volatile components (manganocene or 4-Br-3,5-diMepzH) were removed by exposing the solid to a dynamic vacuum for 48 hours. This procedure yielded 0.28 g (53%) of [Mn(4-Br-3,5-diMepz) 2] x as a white powder. Anal, calcd. for C i 0 H i 2 N 3 r 2 M n : C 29.8, H 3.0, N 13.9; found: C 30.0, H 3.0, N 14.0. 285 9.2.4.4 POLY-BIS(p-3,4,5-TRIMETJJn.PYRAZOLATO-N,N')]VLANGAI^SE(II), ([Mn(4-CH3-3,5-diMepz)2]x) Manganocene (0.25 g, 1.3mmol) was intimately mixed with an excess of 3,4,5-trimethylpyrazole (1.49 g, 13 mmol) and placed in a Carius tube under a dinitrogen atmosphere. The tube was flame-sealed under vacuum and was heated to 160°C in an oven. The 4-CH 3-3,5-diMepzH melted a few minutes later giving a liquid/solid mixture of the pyrazole and manganocene respectively. To ensure complete reaction, the tube was left at this temperature for two weeks. The mixture was allowed to cool and solidify, and the tube was cracked open under a dinitrogen atmosphere. The residue was slurried out of the tube by washing with benzene, THF and hexanes. The washings were suction-filtered and the filtercake was rinsed with additional quantities of these solvents. Any remaining volatile starting material were removed by exposing the sample to a dynamic vacuum for 24 hours. Yield 0.21 g (59%) of [Mn(4-CH3-3,5-diMepz)2]x as a beige powder. Anal, calcd. for Ci 2 H 1 8 N4Mn: C52.8, H 6.6, N 20.5; found: C 52.7, H 6.6, N 20.3. 9.2.4.5 POLY-BIS(p.-4-CHLOROPYRAZOLATO-N,N')(4-CHLOROPYRAZOLE)MANGANESE(II) , ([Mn(4-Clpz)2(4-ClpzH)]x) A mixture of manganocene (0.25 g, 1.3 mmol) and 4-chloropyrazole (1.38 g, 13 mmol) was placed in a Carius tube under a dinitrogen atmosphere. The tube was flame-sealed under vacuum and was heated to 160°C by placing the tube in an oven. Upon 286 heating the 4-ClpzH melted resulting in a molten ligand and solid manganocene mixture. The heat was applied for a two week period after which the Carius tube was allowed to cool and the contents solidify. The contents were removed from the tube by first cracking open the tube in a dinitrogen atmosphere and then slurrying the mixture out with benzene, THF and hexanes. The resulting solutions were vacuum filtered to collect the insoluble solid product and this solid was washed with additional portions of benzene and hexanes. The filtercake was then exposed to a vacuum for 24 hours to remove any remaining volatile starting materials. Yield: 0.22 g (66%) of [Mn(4-Clpz)2(4-ClpzH)]x as a white powder. Anal, calcd. for C9H7N6CI3M11: C 30.0, H 2.0, N 23.3; found: C 29.5, H 2.0, N 23.0. 9.2.4.6 POLY-BIS(|i-4-BROMOPYPvAZOLATO-N,N')(4-BROMOPYRAZOLE)MANGANESE(II), ([Mn(4-Brpz)2(4-BrpzH)]x) Manganocene (0.25 g, 1.3 mmol) and 4-bromopyrazole (1.9 g, 13 mmol) were mixed and transferred into a Carius tube under a dinitrogen atmosphere. The reaction vessel was evacuated, flame-sealed and placed in a 160°C oven. At this temperature the 4-BrpzH melted giving a molten ligand and solid manganocene mixture. The elevated temperature was continued for two weeks after which the mixture was allowed to cool and solidify. The tube was opened under a dinitrogen gas atmosphere. The contents of the vessel were slurried with benzene, THF and hexanes, and these solutions were vacuum filtered. The filtercake was exposed to a dynamic vacuum for 48 hours to remove any remaining manganocene or other volatiles present yielding 0.25 g (55%) of [Mn(4-287 Brpz)2(4-BrpzH)]x as a white powder. Anal, calcd. for CgHTNeBrjMn: C 21.9, H 1.4, N 17.1; found: C 21.6, H 1.2, N 16.8. 9.2.4.7 POLY-BIS(n-TRIAZOLATO)MANGANESE(II), ([Mn(trz)2]x) Manganocene (0.20 g, 1.1 mmol) and triazole (0.75 g, 10.9 mmol) were mixed together and placed in a Carius tube in a dinitrogen atmosphere. Xylene (20 mL) was added to the mixture and the vessel was flame-sealed under vacuum. Some of the triazole dissolved in the xylene. However, at room temperature, the Mn(Cp) 2 appeared to remain as a solid. The tube was placed in an oven at 140°C and after a few hours the solid manganocene had dissolved or reacted resulting in a pale orange solution. The reaction was heated for three days at 140°C at which time the solution was pale yellow in colour and a light beige solid product had formed. The temperature was increased to 180°C and the reaction allowed to continue for an additional 48 hours. The vessel was subsequently cooled and opened under a dinitrogen gas atmosphere, and the mixture filtered to isolate the solid portion. This solid was washed with several portions of acetonitrile to remove any remaining triazole. The solid was then exposed to a dynamic vacuum for 48 hours to remove any remaining manganocene or solvent yielding 0.15 g (71%) of [Mn(trz)2]x as a white powder. Anal, calcd. for C ^ N g M n : C 25.2, H 2.1, N 44.0; found: C 25.3, H 2.1, N 44.1. 288 9.2.5 COPPER(II) AZOLATES In the copper compounds described herein, copper metal beads were used as the metal source. These beads were cleaned immediately prior to use by washing in 12 M HC1, followed by a thorough water wash and finally an acetone rinse. 9.2.5.1 POLY-BIS(u-TRIAZOLATOCOPPER(II), ([Cu(trz)2]x) Approximately 1 gram of copper metal beads (3-5 mm in diameter), which had been cleaned in HC1 (12 M), water and acetone prior to use, was placed in a 20 mL round bottomed flask. 1,2,4-Triazole (3 g, 0.15 mmol) was added and, to the flask, a reflux condenser was connected. The reaction mixture has heated under an dioxygen atmosphere to 140°C causing the triazole to melt. This molten ligand and solid copper mixture was maintained at this elevated temperature, without stirring, for 5 days. During that time a bright purple solid could be seen forming a coating on the copper beads. The reaction mixture was allowed to cool and solidify, and the unreacted triazole was removed by vacuum sublimation. The product was then physically separated from the copper beads as a bright purple powder. Anal, calcd. for C ^ N e C u : C 24.1, H 2.0, N 42.1; found: C 24.4, H 2.0, N 41.7. 289 9.2.5.2 A M I X E D V A L E N C E COPPER(II)/COPPER(I) COMPOUND, (Cu3(3,5-F 6diMepz) 5) Clean copper shot (1.954 g, 30.75 mmol) was mixed with 3,5-F6diMepzH (1.00 g, 4.90 mmol) and heated, without stirring, under an atmosphere of dioxygen at 95-105°C for 4 days (melting point of 3,5-F6diMepzH, 79-81°C). Within an hour the molten 3,5-F 6diMepzH had become green. At the cessation of the reaction, green crystals had formed together with a small amount of white material. The product mixture was then heated at 60°C under vacuum to remove excess 3,5-F6diMepzH and finally the dark green compound was isolated by physical separation from the copper shot. Approximately 0.2 g of Cu3(3,5-F 6diMepz) 5 was collected (71% yield based on amount of copper reacted). Anal, calcd. for C 2 5H 5 Cu 3 F 3 oN 1 0 : C 24.9, H 0.4, N 11.6; found: C 24.7, H 0.5, N 11.7. 9.2.6 A N IRON(II) IMIDAZOLATE COMPOUND, ([Fe3(imid)6(imidH)2]x) Ferrocene (0.25 g, 1.3 mmol) was mixed intimately with imidazole (3 g, 44 mmol) and transferred into a dry Carius tube. The tube was flame-sealed under vacuum and placed in an oven set at 160°C. The imidazole melted almost immediately giving an orange solution of ferrocene dissolved in the molten ligand. Over the course of two days the intensity of the orange colour faded as small pale-yellow crystals formed inside the tube. The reaction mixture was removed from the heat after 72 hours, at which time almost all of 290 the orange colour in the solution had dissipated and a large amount of the crystals had formed. Upon cooling the remaining imidazole solidified. The tube was opened under a dinitrogen atmosphere and the contents extracted by washing with benzene. The slurry was suction-filtered to separate the insoluble crystals from the soluble excess imidazole and the crystals were then further washed with large amounts of benzene and hexanes. The [Fe3(imid)6(imidH)2]x was isolated as pale yellow crystals. Anal, calcd. for: C24H26Ni6Fe: C 40.9, H 3.4, N 31.8; found: C 41.3, H 3.5, N 31.3 9.3 PHYSICAL METHODS 9.3.1 M A G N E T I C M E A S U R E M E N T S A N D P A R A M E T E R CALCULATIONS Room temperature magnetic susceptibilities were measured on a Johnson-Matthey magnetic susceptibility balance. This balance employs the Gouy method and in all cases the measurements were corrected for the signals corresponding to the sample holder and the contributions from the diamagnetic atoms in the sample. The diamagnetic contributions were calculated from Pascal's constants (259). Sample preparation involved packing powdered samples in a narrow Pyrex tube to a height of ~lcm. This method was primarily used to determine if a sample was diamagnetic or paramagnetic. Al l paramagnetic samples were subsequently subjected to more rigorous analysis on a SQUID magnetometer. Variable temperature (2 K - 300 K) magnetic susceptibilities were measured using a Quantum Design M P M S SQUID Magnetometer. This magnetometer is capable of measuring samples in applied fields from 0 to 55 000 Gauss and between temperatures of 291 1.7 to 350 K . Two thermometers are used to monitor the temperature. From 1.7 to 40 K a germanium resistance thermometer is employed, and at temperatures above 40 K a platinum resistance thermometer is used. Temperature calibrations are performed using an external platinum resistance thermometer and temperature accuracy within 0.1% is achieved. Magnetic susceptibility signals were calibrated using ultrapure nickel and bismuth standards and are accurate to within 1%. The sample holder is constructed from P V C plastic and is designed to go undetected by the magnetometer. A more detailed description of this particular sample holder, along with its dimensions and instructions for building new ones, is given in M . K . Ehlert's PhD thesis (193). Background corrections were made by subtracting the signal of the sample holder from the data at all temperatures studied. Table 9.1. Diamagnetic corrections of ligands and metal ions. Species Xdia (x 106 ci^mor 1 ) 4-Hpz -38 4-Clpz -55 4-Brpz -66 4-H-3,5-diMepz -62 4-CH3-3,5-diMepz -74 4-Br-3,5-diMepz -90 4-Cl-3,5-diMepz -79 3,5-F6diMepz -82 imid -38 trz -34 N i 2 + -11 M n 2 + -14 C u 2 + -11 Fe 2 + -13 292 A list of the diamagnetic corrections for the ligands and metal ions considered in this work is given in Table 9.1. The magnetic exchange parameters were calculated using the non-linear regression function in SigmaPlot 3.2 for Windows 95. The theoretical models for magnetic exchange interactions are entered into the computer as mathematical equations with assigned variables. The computer calculates the appropriate parameters (J, g, P, TIP etc.) by systematically minimizing the overall difference between the experimental data and the resulting calculated values. The calculated parameters have, associated with them, a standard error value (Standard errors have been given in parentheses for the magnetic parameters quoted in this thesis). These error values should be put into context. The standard error values represent the amount by which a given parameter can be changed without affecting the quality of the fit. These values do not necessarily represent the quality of the fit itself, nor the experimental error. The quality of the fit is indicated by the F-value, which is calculated separately (discussed previously). The significance of these error values is illustrated in Figure 9.1. 293 Temperature ^> Figure 9.1. Arbitrary plot illustrating the significance of the standard error values of the calculated magnetic parameters. The circles represent experimental data, the dashed line represents the theoretical values generated by the computed magnetic parameters, and the two solid lines represent other equally "good" fits generated with the extreme parameter values (parameters ± standard error). The dashed line represents the theoretical fit, which is generated using the calculated parameters. If, in combination, other values of the calculated parameters are used, one or more equally good fits may be obtained. The upper solid line may, for example, represent the fit generated by using the calculated parameters plus the standard error values on those parameters, while the lower fit may represent the fit obtained by employing the minimum accepted values of the calculated parameters. In one case the high temperature fit is good, 294 but the low temperature data fit is poor. In the other scenario the lower temperature fit is good, but not the high temperature fit. Therefore, a standard error on a calculated parameter indicates the range of that parameter that can be used to generate different, yet equally "good", fits. 9.3.2 SINGLE C R Y S T A L X - R A Y DIFFRACTION The single crystal X-ray diffraction study of [Cu3(3,5-F6diMepz)5] was performed by Dr. V. G. Young of the X-Ray Crystallographic Laboratory at the University of Minnesota, using a Siemens SMART Platform CCD system (Mo Ka radiation). Al l other single crystal X-ray diffraction studies reported in this thesis were performed by Dr. Steven J. Rettig of this department on a Rigaku AFC6S single crystal diffractometer (graphite monochromated, Mo Ka or Cu Ka radiation). 9.3.3 POWDER X - R A Y DIFFRACTION Powder diffractograms were recorded on a Rigaku Rotaflex RU-200BH rotating anode, powder X-ray diffractometer (graphite monochromated Cu Ka radiation). Samples were prepared by applying an octane slurry of a compound onto a glass plate and allowing the solvent to evaporate. 295 9.3.4 E L E M E N T A L ANALYSIS Elemental analyses were performed by Mr. P. Borda of this department. Carbon hydrogen and nitrogen percentages were determined using a Carlo Erba Model 1106 elemental analyzer or a Fisons (Erba) Instruments E A 1108 C H N - 0 Elemental Analyzer. Elemental analyses are considered to have an absolute accuracy within ±0.3%. 9.3.5 INFRARED SPECTROSCOPY Infrared spectra (400-4000 cm"1) were obtained using a Bomem MB-100 spectrometer. Samples were prepared as K B r pellets and band energies are accurate to within ±4 cm"1. 9.3.6 ELECTRONIC SPECTROSCOPY Electronic spectra (200 - 3000 nm) were obtained using a Varian Instruments Cary 5 UV-Vis-NIR Spectrophotometer. Samples were prepared as Nujol mulls pressed between quartz plates. A reference cell containing Nujol was used to minimize unwanted background signals . 296 9.3.7 N M R SPECTROSCOPY Nuclear magnetic resonance spectra were recorded using a Bruker AC-200 FT-N M R Spectrometer. The nmr solvents were used as internal standards for calibration of the observed chemical shifts. 9.3.8 T G A Thermogravimetric analyses were performed using a T A Instruments T G A 51 Thermogravimetric Analyzer. Powdered samples (5 -15 mg) were heated in a dinitrogen atmosphere at a rate of 10°C per minute to a maximum temperature of 800°C. 9.3.8 DSC Differential scanning calorimetry data were collected on a T A Instruments 91 OS Differential Scanning Calorimeter. Accurately weighed powdered samples ( 2 - 8 mg) were placed in disposable aluminum pans. A heating rate of 10°C per minute was used to a maximum temperature of 600°C. Heat flow calibration was achieved using accurately weighed samples of indium. 297 9.3.9 S E M Scanning electron microscopy was performed using a Hitachi S-570 Scanning Electron Microscope. Samples were prepared by applying a thin layer of a compound, slurried in ethanol, to a stainless steel stub. The solvent was allowed to evaporate and the sample was coated with gold and platinum. 9.3.10 M A S S SPECTROMETRY Mass spectra were recorded by the mass spectrometry service in this department. Low resolution electron impact ionization spectra were obtained using a Kratos MS50 mass spectrometer. 298 R E F E R E N C E S 1. O. Kahn, J. Krober, and C. Jay. Adv. Mater. 4, 718 (1992). 2. F. L . Carter, (ed.), "Molecular Electronic Devices." Marcel Decker. New York. 1987. 3. O. Kahn and J. P. Launay. Chemtronics 3, 140 (1988). 4. J. Zarembowitch and O. Kahn. New. J. Chem. 15, 181 (1991). 5. O. Kahn. Molecular magnetism. V C H Publishers. New York. 1993. 6. R. L . Carlin. Magnetochemistry. Springer-Verlag. Berlin. 1986. 7. C. J. O'Connor. Prog. Inorg. Chem. 29, 203 (1982). 8. F. E. Mabbs and D. J. Machin. Magnetism and transition metal chemistry. Chapman and Hall. London. 1973. 9. A. Earnshaw. Introduction to magnetochemistry. Academic Press. New York. 1968. 10. A. Earnshaw. Introduction to magnetochemistry. Academic Press. New York. 1968. p. 10 11. V. H . Crawford, H . W. Richardson, J. R. Wasson, D. J. Hodgson, and W. E. Hatfield. Inorg. Chem. 15, 2107 (1976). 12. L . J. De-Jong and A. R. Miedema. Adv. Physics 23, 1 (1974). 13. W. E. Hatfield, W. E. Estes, W. E. Marsh, W. M . Pickens, L . W. Ter-Haar, and R. R. Weller. Extended linear chain compounds. Edited by J. S. Miller. Vol. 3, p. 43. Plenum Press. New York. 1983. 14. R. D. Willet, R. M . Gaura, and C. P. Landee. Extended linear chain compounds. Edited by J. S. Miller. Vol. 3, p. 143. Plenum Press. New York. 1983. 15. C P . Landee. Organic and inorganic low-dimensional crystalline materials. Plenum Press. London. 1987. p. 75 299 16. J. S. Miller, J. C. Calabrese, A. J. Epstein, R. W. Bigelow, J. H . Zhang, and W. R. Reiff. J. Chem. Soc., Chem. Commun., 1026 (1986). 17. J. S. Miller and A. J. Epstein. Phys. Rev. Lett. 58, 2695 (1987). 18. J. S. Miller, J. C. Calabrese, H . Rommelmann, S. R. Chittipeddi, J. H . Zhang, W. R. Reiff, and A. J. Epstein. J. Am. Chem. Soc. 109, 769 (1987). 19. J. S. Miller, A. J. Epstein, and W. M . Reiff. Science 240, 40 (1988). 20. J. S. Miller, A. J. Epstein, and W. M . Reiff. Acc. Chem. Res. 21, 114 (1988). 21. J. S. Miller, A. J. Epstein, and W. M . Reiff. Chem. Rev. 88, 201 (1988). 22. W. E. Broderick, J. A. Thompson, E. P. Day, and B. M . Hoffman. Science 249, 401 (1990). 23. G. T. Yee, J. M . Manriquez, D. A. Dixon, R. S. McLean, D. M . Groski, R. B. Flippen, K. S. Narayan, A. J. Epstein, and J.'S. Miller. Adv. Mat. 3, 309 (1991). 24. J. S. Miller, J. C. Calabrese, R. S. McLean, and A. J. Epstein. Adv. Mater. 4, 499 (1992). 25. P. G. Lacroix, R. Clement, K. Nakatani, J. Zyss, and I. Ledox. Science 263, 658 (1994). 26. P. Rabu, S. Rouba, V. Laget, C. Hornick, and M . Drillon. J. Chem. Soc, Chem. Commun. , 1107(1996). 27. J. M . Manriquez, G. T. Yee, R. S. McLean, A. J. Epstein, and J. S. Miller. Science 252, 1415 (1991). 28. O. Kahn, Y . Pei, M . Verdaguer, J. P. Renard, and J. Sletten. J. Am. Chem. Soc. 110, 782 (1988). 29. A. Caneschi, D. Gatteschi, P. Rey, and R. Sessoli. Inorg. Chem. 27, 1756 (1988). 30. A. Caneschi, D. Gatteschi, J. P. Renard, P. Rey, and R. Sessoli. Inorg. Chem. 28, 2940 (1989). 31. V. Baron, B. Gillon, A. Cousson, C. Mathoniere, O. Kahn, A. Grand, L . Ohrstrom, B. Delley, M . Bonnet, and J. Boucherle. J. Am. Chem. Soc. 119, 3500 (1997). 300 32. C. Mathoniere, C. J. Nuttall, S. G. Carling, and P. Day. Inorg. Chem. 35, 1201 (1996). 33. N . Achiewa, T. Matsuyama, and T. Yoshinari. Phase Transitions 28, 79 (1990). 34. M . A. Martinez-Lorente, V. Petrouleas, R. Poinsot, J.-P. Tuchagues, J. M . Savariault, and M . Drillon. Inorg. Chem. 30, 3587 (1991). 35. M A . Martinez-Lorente, F. Dahan, Y . Sanakis, V. Petrouleas, A. Bousseksou, and J.-P. Tuchagues. Inorg. Chem. 34, 5346 (1995). 36. S. Flandrois, N . B. Chanh, R. Duplessix, T. Maris, and P. Negrier. Phys. Stat. Sol. (a) 149, 697 (1995). 37. K. Miyagawa, A. Kawamoto, Y . Nakazawa, and K. Kanoda. Phys. Rev. Lett. 75, 1174 (1995). 38. D. C. Dender, D. Davidovic, D. H . Reich, and C. Broholm. Phys. Rev. B53, 2583 (1996). 39. C. Bellitto, F. Federici, and S. A. Ibrahim. J. Chem. Soc, Chem. Commun. , 759 (1996). 40. P. Day. J. Chem. Soc, Dalton Trans., 701 (1997). 41. M . Gerloch. Magnetism and ligand field analysis. Cambridge University Press. Cambridge. 1983. 42. M , F. Ladd and R. A. Palmer. Structure determination by X-ray crystallography. Plenum Press. New York. 1988. 43. L . V. Azaroff and M . J. Buerger. The powder method in X-ray crystallography. McGraw-Hill. New York. 1958. 44. K. Nakamoto. Infrared spectra of inorganic and coordination compounds. Wiley-Interscience. New York. 1969. 45. Y . Tanabe and S. Sugano. J. Phys. Soc. Jpn. 9, 753 (1954). 46. A. P. B. Lever. Inorganic electronic spectroscopy. Elsevier. Amsterdam. 1984. 47. J. W. Akitt. Nmr and chemistry. Chapman and Hall. London. 1983. 301 48. J. K. M . Sanders, E. C. Constable, and B. K. Hunter. Modern nmr spectroscopy: a workbook of chemical problems. Oxford. Toronto. 1989. 49. T. S. M a and R. C. Rittner. Modern organic elemental analysis. Marcel Decker Inc. New York. 1979. 50. A. R. West. Solid state chemistry and its applications. John Wiley & Sons. New York. 1984. 51. H . H . Willard, L . L . Merrit, J. A. Dean, and F. A. Settle-Jr. Instrumental methods of analysis. Wadsworth Publishing Co. Belmont, California. 1981. 52. S. Trofimenko. Chem. Rev, 72, 497 (1972). 53. S. Trofimenko. Prog. Inorg. Chem. 34, 115 (1986). 54. G. La-Monica and G. A. Ardizzoia. Prog. Inorg. Chem. 46, 151 (1997). 55. P. J. Steele. Coord. Chem. Rev. 106, 227 (1990). 56. E. Buchner. Ber. 22, 842(1889). 57. N . Masciocchi, M . Moret, P. Cairati, A. Sironi, G. A. Ardizzoia, and G. La-Monica. J. Am. Chem. Soc. 116, 7668 (1994). 58. E. Buchner and M . Fritsch. Ann. 273, 256 (1893). 59. J. G. Vos and W. L . Groeneveld. Transition Met. Chem. 4, 137 (1979). 60. H . Okkersen, W. L . Groeneveld, and J. Reedijk. Revl. Trav. Chim. Pays-Bas. 92, 945 (1973). 61. N . M . Arakelyan, S. A. Papyan, and S. I. Isabekyan. Arm. Khim. Zh. 35, 202 (1982). 62. C. B. Singh, S. Satpathy, and B. Sahoo. J. Inorg. Nucl. Chem. 33, 1313 (1971). 63. M . K. Ehlert, S. J. Rettig, A. Storr, R. C. Thompson, and J. Trotter. Can. J. Chem. 68, 1444 (1990). 64. F. Bonati, G. Minghetti, and G. Banditelli. J. Chem. Soc, Chem. Commun. , 88 (1974). 302 65. G. Banditelli, A. L . Bandini, F. Bonati, R. G. Goel, and G. Minghetti. Gazz. Chim. Ital. 112, 539 (1982). 66. B. Boivu, F. Bonati, and G. Banditelli. Inorg. Chim. Acta 87, 25 (1984). 67. R. G. Raptis and J. P. Fackler-Jr. Inorg. Chem. 27, 4179 (1988). 68. H . H . Murray, R. G. Raptis, and J. P. Fackler-Jr. Inorg. Chem. 27, 26 (1988). 69. M . K. Ehlert, S. J. Rettig, A. Storr, R. C. Thompson, and J. Trotter. Can. J. Chem. 70, 2161 (1992). 70. J. A. Smith, B. Wehrle, F. Aguilar-Parilla, H . H . Limbach, C. Foces-Foces, F. H. Cano, J. Elguero, A. Baldy, M . Pierrot, M . M . T. Khurshid, and J. B. Larcombe-McDougall. J. Am. Chem. Soc. I l l , 7304 (1989). 71. A. L . Llamez-Siaz, C. Foces-Foces, F. H . Cano, P. Jimenez, J. Layner, W. Meutermans, J. Elguero, H . H . Limbach, and F. Aguilar-Parilla. Acta. Crystallogr. B50, 746 (1994). 72. E. J. Fischer. Wiss. Vereffentlich Siemens Konzern 4, 178 (1925). 73. O. Heim. Ind. Eng. Chem., Anal. Ed. 2, 38 (1930). 74. R. T. Pflaum and L. H . Dieter. Proc. Iowa Acad. Sci. 64, 235 (1957). 75. F. Seel and V. Sperber. Angew Chem. Int. Ed. 7, 70 (1968). 76. M . J. Bagley, D. Nicholls, and B. A. Warburton. J. Chem. Soc. (A), 2694 (1970). 77. C. B. Singh, S. Satpathy, and B. Sahoo. J. Inorg. Nucl. Chem. 35, 3947 (1973). 78. J. G. Vos and W. L . Groeneveld. Inorg. Chim. Acta 24, 123 (1977). 79. J. G. Vos and W. L . Groeneveld. Inorg. Chim. Acta 27, 173 (1978). 80. J. G. Vos and W. L . Groeneveld. Inorg. Chim. Acta 26, 71 (1978). 81. J. G. Vos and W. L . Groeneveld. Inorg. Chim. Acta 29, 117 (1978). 82. A. B. Blake, D. F. Ewing, J. E. Hamlin, and J. M . Lockyer. J. Chem. Soc, Dalton Trans. , 1987 (1977). 303 83. J. Strahle, F. Kubel, W. Hiller, and R. Dantona. Mol. Cryst. Liq. Cryst. 81, 265 (1982). 84. F. F. Seelig. Mol. Cryst. Liq. Cryst. 81, 285 (1982). 85. M . K. Ehlert, S. J. Rettig, A. Storr, R. C. Thompson, and J. Trotter. Can. J. Chem. 67, 1970 (1989). 86. M . K . Ehlert, S. J. Rettig, A. Storr, R. C. Thompson, and J. Trotter. Can. J. Chem. 69, 432 (1991). 87. M . K. Ehlert, A. Storr, and R. C. Thompson. Can J. Chem. 70, 1121 (1992). 88. M . K. Ehlert, A. Storr, and R. C. Thompson. Can. J. Chem. 71, 1412 (1993). 89. S. Emori and K. Sadakata. Bull. Chem. Soc. Jpn. 67, 1743 (1994). 90. M . K. Ehlert, S. J. Rettig, A. Storr, R. C. Thompson, and J. Trotter. Can. J. Chem. 68, 1494 (1990). 91. W. Burger and J. Strahle. Z. Anorg. Allg. Chem. 539, 27 (1986). 92. W. Burger and J. Strahle. Z. Anorg. Allg. Chem. 529, 111 (1984). 93. G. A. Ardizzoia, G. La-Monica, S. Cenini, M . Moret, and N . Masciocchi. J. Chem. Soc, Dalton Trans. , 1351 (1996). 94. A. M . Vecchio-Sadus. Transition Met. Chem. 20, 46 (1995). 95. W. L . Driesen, F. Paap, and J. Reedijk. Reel. Trav. Chim. Pays-Bas. 114, 317 (1995). 96. G. A. Ardizzoia, S. Cenini, G. La-Monica, N . Masciocchi, and M . Moret. Inorg. Chem. 33, 1458 (1994). 97. G. A. Ardizzoia and G. La-Monica. Inorg. Chem. 31, 2706 (1992). 98. G. A. Ardizzoia, G. La-Monica, M . A. Angaroni, and F. Cariati. Inorg. Chim. Acta 158, 159 (1989). 99. G. A. Ardizzoia, M . A. Angaroni, G. La-Monica, N . Masciocchi, and M . Moret. J. Chem. Soc, Dalton Trans., 2277 (1990). 304 100. B . F. Fieselmann and G. D. Stucky. Inorg. Chem. 17, 2074 (1978). 101. J. A. Bailey, S. L . Grundy, and S. R. Stobart. Organometallics 9, 536 (1990). 102. L . A. Oro, D. Carmona, M . P. Lamata, C. Foces-Foces, and F. H . Cano. Inorg. Chim. Acta 97, 19 (1985). 103. G. Lopez, G. Garcia, J. Ruiz, G. Sanchez, J. Garcia, and C. Vicente. J. Chem. Soc. Chem. Commun., 1045 (1989). 104. G: Lopez, G. Garcia, G. Sanchez, J. Garcia, J. Ruiz, J. A. Hermoso, A. Vegas, and M . Martinez-Ripoll. Inorg. Chem. 31, 1518 (1992). 105. G. Lopez, J. Ruiz, G. Garcia, J. M . Marti, G. Sanchez, and J. Garcia. J. Organomet. Chem. 412, 435 (1991). 106. G. Lopez, G. Garcia, G. Sanchez, M . D. Santana, J. Ruiz, and J. Garcia. Inorg. Chim. Acta. 188, 195 (1991). 107. G. Lopez, J. Ruiz, G. Garcia, C. Vicente, J. Casabo, E. Molins, and C. Miravitlles. Inorg. Chem. 30, 2065 (1991). 108. G. Lopez, J. Ruiz, G. Garcia, C. Vicente, J. M . Marti, J. A . Hermoso, A. Vegas, and M . Martinez-Ripoll. J. Chem. Soc, Dalton Trans. , 53 (1992). 109. G. Lopez, J. Ruiz, G. Garcia, C. Vicente, V. Rodriguez, G. Sanchez, J. A. Hermoso, and M . Martinez-Ripoll. J. Chem. Soc, Dalton Trans., 1681 (1992). 110. V. K. Jain, S. Kannan, and E. R. T. Tiekink. J. Chem. Soc, Dalton Trans. , 3625 (1993). 111. K. S. Chong, S. J. Rettig, A. Storr, and J. Trotter. Can. J. Chem. 59, 996 (1981). 112. K. S. Chong, S. J. Rettig, A. Storr, and J. Trotter. Can. J. Chem. 57, 3090 (1979). 113. M . K. Ehlert, S. J. Rettig, A. Storr, R. C. Thompson, and J. Trotter. Can. J. Chem. 71, 1425 (1993). 114. R. Uson, L . A. Oro, M.A.Ciriano, M . T. Pinillos, A. Tiripicchio, and M . Tiripicchio-Camellini. J. Organomet. Chem. 205, 247 (1981). 115. I. Amer, H . Schumann, V. Ravindar, U . Baidossi, N . Sorem, and J. Blum. J. Mol . Cat. 85, 163 (1993). 305 116. H . Schumann, H. Hemling, V. Ravindar, Y . Backieh, and J. Blum. J. Organomet. Chem. 469, 213 (1994). 117. L . A. Oro, D. Carmona, J. Reyes, C. Foces-Foces, and F. H . Cano. J. Chem. Soc, Dalton Trans. , 31 (1986). 118. S. Trofimenko. Inorg. Chem. 10, 1372 (1971). 119. K. A. Beveridge, G. W. Bushnell, S. R. Stobart, J. L . Atwood, and M . J. Zaworotko. Organometallics 2, 1447 (1983). 120. F. Barcelo, P. Lahuerta, M . A. Ubeda, C. Foces-Foces, F. H . Cano, and M . Martinez-Ripoll. J. Chem. Soc. Chem. Commun. , 43 (1985). 121. K . A. Beveridge, G. W. Bushnell, K. R. Dixon, D. T. Eadie, S. R. Stobart, J. L . Atwood, and M . J. Zaworotko. J. Am. Chem. Soc 104, 920 (1982). 122. J. Powell, A. Kuksis, S. C. Nyburg, and W. W. Ng. Inorg. Chim. Acta 64, L211 (1982). 123. J. L . Atwood, K. A. Beveridge, G. W. Bushnell, K. R. Dixon, D. T. Eadie, S. R. Stobart, and M . J. Zaworotko. Inorg. Chem. 23, 4050 (1984). 124. D. G. Harrison and S. R. Stobart. J. Chem. Soc. Chem. Commun. 23, 285 (1986). 125. D. O. K . Fjeldsted, S. R. Stobart, and M . J. Zaworotko. J. Am. Chem. Soc. 107, 8258 (1985). 126. L . A. Oro, D. Carmona, M . P. Puebla, M . P. Lamata, C. Foces-Foces, and F. H. Cano. Inorg. Chim. Acta 112, L l l (1986). 127. J. A. Cabeza, C. Lanzaduri, L. A. Oro, A. Tiripicchio, and M . Tiripicchio-Camellini. J. Organomet. Chem. 322, C16 (1987). 128. J. A. Cabeza, C. Lanzaduri, L . A. Oro, D. Belletti, A. Tiripicchio, and M . Tiripicchio-Camellini. J. Chem. Soc, Dalton Trans. , 1093 (1989). 129. J. Camona, F. J. Lahoz, L . A. Oro, J. Reyes, and M . P. Lamata. J. Chem. Soc, Dalton Trans. , 3551 (1990). 130. D. Carmona, F. J. Lahoz, L. A. Oro, M . P. Lamata, and S. Buzarra. Organometallics 10, 3123 (1991). 306 131. M . P. Garcia, A. M . Lopez, M . A. Esteruelas, F. J. Lahoz, and L . A. Oro. J. Chem. Soc, Dalton Trans. , 3465 (1990). 132. D. Carmona, J. Ferrer, R. Atenico, F. J. Lahoz, L . A. Oro, and M . P. Lamata. Organometallics 14, 2057 (1995). 133. M . P. Garcia, A. Portilla, L. A. Oro, C. Foces-Foces, and F. H . Cano. J. Organomet. Chem. 322, 111 (1987). 134. M . Maekawa, M . Munakata, T. Kuroda, and Y. Nozaka. Inorg. Chim. Acta 208, 243 (1993). 135. K. S. Chong, S. J. Rettig, A. Storr, and J. Thompson. Can. J. Chem. 57, 3099 (1979). 136. S. Trofimenko. J. Am. Chem. Soc. 91, 5410 (1969). 137. D. Carmona, M . P. Lamata, J. Ferrer, J. Modrego, M . Perales, F. J. Lahoz, R. Atencio, and L. A. Oro. J. Chem. Soc. Chem. Commun., 575 (1994). 138. A. R. Barron, G. Wilkinson, M . Motevalli, and H . B. Hursthouse. Polyhedron 4, 1131 (1985). 139. V. K . Jain and S. Kannan. Polyhedron 11, 27 (1992). 140. T. N . Doman, D. E. Williams, J. F. Banks, R. M . Buchanan, H . R. Chang, R. J. Webb, and D. N . Hendrickson. Inorg. Chem. 29, 1058 (1990). 141. W. Mazurek, B. J. Kennedy, K. S. Murray, M . J. O'Connor, J. R. Rodgers, M . R. Snow, A. G. Wedd, and P. R. Zwack. Inorg. Chem. 24, 3258 (1985). 142. Y . Nishida and S. Kida. Inorg. Chem. 27, 447 (1988). 143. D. Ajo, A. Bencini, and F. Mani. Inorg. Chem. 27, 2437 (1988). 144. K . Matsumoto, S. Ooi, W. Mori, and Y . Nakao. Bull. Chem. Soc. Jpn. 60, 4477 (1987). 145. P. Iliopoulos, K. S. Murray, R. Robson, J. Wilson, and G. A. Williams. J. Chem. Soc, Dalton Trans. , 1585 (1987). 146. M . G. B. Drew, P. C. Yates, F. S. Esho, J. Trocha-Grimshaw, A. Lavery, K. P. McKillop, S. M . Nelson, and J. Nelson. J. Chem. Soc, Dalton Trans., 2995 (1988). 307 147. P. Iliopoulos, G. S. Fallon, and K. S. Murray. J. Chem. Soc., Dalton Trans. , 1823 (1988) . 148. C. J. McKenzie and R. Robson. Inorg. Chem. 26, 3615 (1987). 149. P. Cheng, D. Liao, S. Yan, Z. Jiang, and G. Wang. Transition Met. Chem. 21, 515 (1996). 150. J. R. Shapley, D. E. Samkoff, C. Bueno, and M . R. Churchill. Inorg. Chem. 21, 634 (1982). 151. D. Carmona, M . P. Lamata, M . Esteban, F. J. Lahoz, L. A. Oro, M . C. Apreda, C. Foces-Foces, andF. H. Cano. J. Chem. Soc, Dalton Trans. , 159 (1994). 152. V. Ashworth, D. C. Liles, and E. Singleton. Inorg. Chim. Acta 98, L65 (1985). 153. L . A. Oro, D. Carmona, J. Reyes, C. Foces-Foces, and F. H . Cano. Inorg. Chim. Acta 112, 35 (1986). 154. C. Claver, P. Kalck, M . Ridmy, A. Thores, L. A. Oro, M . T. Pinillos, M . C. Apreda, F. H . Cano, and C. Foces-Foces. J. Chem. Soc, Dalton Trans. , 1523 (1988). 155. M . A. Ciriano, M . A. Tena, and L . A. Oro. J. Chem. Soc, Dalton Trans. , 2123 (1992). 156. M . T. Pinillos, A. Elduque, J. A. Lopez, F. J. Lahoz, L. A. Oro, and B. E. Mann. J. Chem. Soc, Dalton Trans., 2389 (1992). 157. D. Carmona, L . A. Oro, M . P.. Lamata, M . P. Puebla, J. Ruiz, and P. M . Maitlis. J. Chem. Soc, Dalton Trans., 639 (1987). 158. M . T. Pinillos, A. Elduque, and L. A. Oro. Inorg. Chim. Acta 178, 179 (1990). 159. R. D. Brost and S. R. Stobart. J. Chem. Soc. Chem. Commun. , 498 (1989). 160. M . T. Pinillos, A. Elduque, L . A. Oro, F. J. Lahoz, F. Bonati, A. Tiripicchio, and M . Tiripicchio-Camellini. J. Chem. Soc, Dalton Trans. , 989 (1990). 161. F. Neumann, H . Stoeckli-Evans, and G. Suss-Fik. J. Organomet. Chem. 379, 151 (1989) . 162. T. V. Ashworth, D. C. Liles, and E. Singleton. Inorg. Chim. Acta 98, L65 (1985). 308 163. M . O. Albers, S. F. A. Crosby, D. C. Liles, D. J. Robinson, A. Shaver, and E. Singleton. Organometallics 6, 2014 (1986). 164. R. Atencio, C. Bohanna, M . A. Esteruelas, F. J. Lahoz, and L . A. Oro. J. Chem. Soc, Dalton Trans. , 2171 (1995). 165. M . I. Bruce, M . G. Humphrey, M . R. Snow, E. R. T. Tiekink, and R. C. Wallis. J. Organomet. Chem. 314, 311 (1986). 166. L . A. Oro, D. Carmona, M . P. Garcia, F. J. Lahoz, J. Reyes, C. Foces-Foces, and F. H . Cano. J. Organomet. Chem. 296, C43 (1985). 167. M . T. Pinillos, A. Elduque, E. Berkovich, and L. A. Oro. J. Organomet. Chem. 509, 89 (1996). 168. M . T. Pinillos, A. Elduque, E. Martin, N . Navarro, F. J. Lahoz, J. A. Lopez, and L. A. Oro. Inorg. Chem. 34, 111 (1995). 169. M . T. Pinillos, C. Tejel, L . A. Oro, M . C. Apreda, C. Foces-Foces, and F. H. Cano. J. Chem. Soc, Dalton Trans. , 1989 (1989). 170. M . T. Pinillos, A. Elduque, E. Martin, N . Navarro, L . A. Oro, A. Tiripicchio, and F. Ugozzoli. Inorg. Chem. 34, 3105 (1995). 171. J. Manzur, A. M . Garcia, M . T. Garland, V. Acuna, O. Gonzalez, O. Pena, A. M . Atria, and E. Spodine. Polyhedron 15, 821 (1996). 172. K. Sakai, Y. Yamada, T. Tsubomura, M . Yabuki, and M . Yamaguchi. Inorg. Chem. 35, 542 (1996). 173. G. A. Ardizzoia, M . A. Angaroni, G. La-Monica, F. Cariati, S. Cenini, M . Moret, and M . Masciocchi. Inorg. Chem. 30, 4347 (1991). 174. M . K. Ehlert, S. J. Rettig, A. Storr, R. C. Thompson, and J. Trotter. Inorg. Chem. 32, 5176 (1993). 175. J. C. Bayon, P. Esteban, G. Net, P. G. Rasmussen, and D. F. Bergstrom. Inorg. Chem. 30, 4771 (1991). 176. T. Kamiusuki, H . Okawa, E. Kitaura, M . Koikawa, N . Matsumoto, S. Kida, and H . Oshio. J. Chem. Soc, Dalton Trans., 2077 (1989). 309 177. T. Kamiusuki, H . Okawa, N . Matsumoto, and S. Kida. J. Chem. Soc., Dalton Trans. , 195 (1990). 178. T. Kamiusuki, H . Okawa, E. Kitaura, K. Inoue, and S. Kida. Inorg. Chim. Acta. 179, 139 (1991). 179. M . Itoh, K. Motoda, K. Shindo, T. Kamiusuki, H. Sakiyama, N . Matsumoto, and H . Okawa. J. Chem. Soc, Dalton Trans., 3635 (1995). 180. B. Mernari, F. Abraham, M . Lagrenee, M . Drillon, and P. Legoll. J. Chem. Soc, Dalton Trans. , 1707 (1993). 181. K. Shindo, Y . Mori, K. Motoda, H . Sakiyama, and N . M . H . Okawa. Inorg. Chem. 31, 4987 (1992). 182. J. Casabo, J. Pons, K. S. Siddiqi, F. Teixidor, E. Molins, and C. Miravitlles. J. Chem. Soc, Dalton Trans. , 1401 (1989). 183. M . Munakata, L . P. Wu, T. Yamamoto, T. Kuroda-Sowa, M . Maekawa, and S. Kitagawa. J. Chem. Soc, Dalton Trans., 4099 (1995). 184. M . Kumar, V. J. Arau, and P. Navarro. Tetrahedron Lett. 34, 3459 (1993). 185. C. Acerete, J. M . Bueno, L. Campayo, P. Navarro, M . I. Rodriguez-Franco, and A. Samat. Tetrahedron Lett. 50, 4765 (1995). 186. T. G. Schenk, J. M . Dowens, C. R. C. Milne, P. B. Mackenzie, H . Boucher, J. Whelan, and B. Bosnich. Inorg. Chem. 24, 2334 (1985). 187. J. Pons, X . Lopez, E. Benet, and J. Casabo. Polyhedron 9, 2839 (1990). 188. J. Pons, F. J. Sanchez, A. Labarta, J. Casabo, F. Teixidor, and A. Caubet. Inorg. Chim. Acta 208, 167 (1993). 189. H . Zhang, D. Fu, F. Ji, G. Wang, K. Yu, and T. Yao. J. Chem. Soc, Dalton Trans. , 3799 (1996). 190. V. P. Hanot, T. D. Robert, J. Kolnaar, J. G. Haasnoot, J. Reedijk, H . Kooijman, and A. L . Spek. J. Chem. Soc, Dalton Trans. , 4275 (1996). 191. H. V. R. Dias, W. Jin, H. J. Kim, and H . L . Lu. Inorg. Chem. 35, 2317 (1996). 192. H . V. R. Dias and W. Jin. Inorg Chem. 35, 3687 (1996). 310 193. M . K. Ehlert. PhD Thesis. University of British Columbia, 1992. 194. M . K. Ehlert, A. Storr, R. C. Thompson, F. W. B. Einstein, and R. J. Batchelor. Can. J. Chem. 71, 331 (1993). 195. J. Reedijk, J. C. A. Windhorst, N . H . M . V. Ham, and W. L. Groeneveld. Rec. Trav. Chim. 90, 234 (1971). 196. G. Wilkinson, P. L . Pauson, and F. A. Cotton. J. Am. Chem. Soc. 76, 1970 (1954). 197. E. R. Lippincott and R. D. Nelson. Spectrochim. Acta. 10, 307 (1955). 198. D. Nicholls. Comprehensive inorganic chemistry. Edited by J. C. Bailar, H. J. Emeleus, R. Nyholm, and A. F. Trotman-Dickenson. Vol. 3. Pergamon Press. New York. 1973. 199. M . K. Ehlert. Unpublished results (1989). 200. A. Zecchina, L. Cerruti, S. Coluccia, and E. Borello. J. Chem. Soc. B, 1363 (1967). 201. G. Banditelli, A. L . Bandini, G. Minghetti, and F. Bonati. Can. J. Chem. 59, 1241 (1981). 202. C. H . Weng. PhD Thesis. Carnegie-Mellon University, 1968. 203. W. Hiller, J. Strahle, A. Datz, M . Hanack, W. F. Hatfield, and P. Gutlich. J. Am. Chem. Soc. 106, 329 (1984). 204. S. L . Lambert and D. N . Hendrickson. Inorg. Chem. 18, 2683 (1979). 205. F. A. Cotton and G. Wilkinson. Advanced Inorganic Chemistry. John Wiley and Sons. New York. 1980. p. 741 206. J. Delaunay and R. P. Hugel. Inorg. Chem. 25, 3957 (1986). 207. G. Ciani, M . Manaddero, and M . Sansoni. J. Inorg. Nucl. Chem. 34, 1760 (1972). 208. F. L . Phillips, F. M . Shreeve, and A. C. Skapski. Acta. Cryst. B32, 687 (1976). 209. O. Kahn. Molecular magnetism. VCHPublishers. New York. 1993. p. 241 210. L . R. Groeneveld, G. Vos, G. C. Verschoor, and J. Reedijk. J. Chem. Soc, Chem. Commun. , 620 (1982). 311 211. J. A. J. Jarvis. Acta. Cryst. 15, 964 (1962). 212. K. H . Sugiyarto and H. A. Goodwin. Aust. J. Chem. 47, 263 (1994). 213. V. P. Sinditskii, V. I. Sokol, A. E. Fogel'zang, M . D. Dutov, V. V. Serushkin, M . A. Porai-Koshits, and B. S. Svetlov. Russ. J. Inorg. Chem. 32, 1149 (1987). 214. F. J. Rietmeijer, J. G. Haasnoot, A. J. Den-Hartog, and J. Reedijk. Inorg. Chim. Acta 113, 147 (1986). 215. G. A. Van-Albada, R. A. G. De-Graaff, J. G. Haasnoot, and J. Reedijk. Inorg. Chem. 23, 1404 (1984). 216. F. J. Rietmeijer, G. A. Van-Albada, R. A. G. De-Graaff, J. G. Haasnoot, and J. Reedijk. Inorg. Chem. 24, 3597 (1985). 217. L . Antolini, A. C. Fabretti, D. Gatteschi, A. Giusti, and R. Sessoli. Inorg. Chem. 30, 4858 (1991). 218. W. Vreugdenhil, J. G. Haasnoot, and J. Reedijk. Inorg. Chim. Acta 167 (1990). 219. A. Michalowicz, J. Moscovici, B. Ducourant, D. Cracco, and O. Kahn. Chem. Mater. 7, 1833 (1995). 220. O. Kahn and E. Codjovi. Phil. Trans. R. Soc. Lond. A , 359 (1996). 221. G. Vos, R. A. G. De-Graaff, J. G. Haasnoot, A. M . Van-Der-Kraan, P. De-Vaal, and J. Reedijk. Inorg. Chem. 23, 2905 (1984). 222. J. Krober, E. Codjovi, O. Kahn, F. Groliere, and C. Jay. J. Am. Chem. Soc. 115, 9810(1993). 223. G. Vos, R. A. Le-Febre, R. A. G. De-Graaff, J. G. Haasnoot, and J. Reedijk. J. Am. Chem. Soc. 105, 1682 (1983). 224. C. W. Reiman and M . Zocchi. Chem. Commun. , 272 (1968). 225. A. L. Spek and G. Vos. Acta. Cryst. C39, 990 (1983). 226. L . Antolini, A. C. Fabretti, D. Gatteschi, A. Giusti, and R. Sessoli. Inorg. Chem. 29, 143 (1990). 312 227. L . R. Groeneveld, R. A. Le-Febre, R. A. G. De-Graaff, J. G. Haasnoot, G. Vos, and J. Reedijk. Inorg. Chim. Acta 102, 69 (1985). 228. P. J. Van-Koningsbruggen, J. G. Haasnoot, W. Vreugdenhil, J. Reedijk, and O. Kahn. Inorg. Chim. Acta 239, 5 (1995). 229. G. Vos, J. G. Haasnoot, G. C. Verschoor, J. Reedijk, and P. E. L . Schaminee. Inorg. Chim. Acta 105, 31 (1985). 230. W. Vreugdenhil, J. G. Haasnoot, M . F. J. Schoondergang, and J. Reedijk. Inorg. Chim. Acta 130 (1987). 231. G. Vos, J. G. Haasnoot, G. C. Verschoor, and J. Reedijk. Inorg. Chim. Acta. 102, 187 (1985). 232. W. Vreugdenhil, M . F. J. Schoondergang, J. G. Haasnoot, and J. Reedijk. Z. Naturforsch 42b, 791 (1987). 233. G. Vos, A. J. De-Kok, and G. C. Verschoor. Z. Naturforsch 36b, 809 (1981). 234. D. W. Engelfriet, G. C. Verschoor, and S. Goiter. Acta. Crystallogr., Sect B. 35, 2927 (1979). 235. D. W. Engelfriet, W. D. Brinker, G. C. Verschoor, and S. Goiter. Acta. Crystallogr., SectB. 35, 2922 (1979). 236. L. A. Oro, M . T. Pinillos, and C. Tejel. J. Chem. Soc, Dalton Trans., 2193 (1986). 237. J. Krober, I. Bkouche-Waksman, C. Pascard, M . Thomann, and O. Kahn. Inorg. Chim. Acta 230, 159 (1995). 238. L . A. Oro, M . T. Pinillos, C. Tejel, C. Foces-Foces, and F. H . Cano. J. Chem. Soc, Chem. Commun. , 1687 (1984). 239. L . A. Oro, M . T. Pinillos, C. Tejel, C. Foces-Foces, and F. H . Cano. J. Chem. Soc, Dalton Trans. , 1087 (1986). 240. L . A. Oro, M . T. Pinillos, C. Tejel, C. Foces-Foces, and F. H . Cano. J. Chem. Soc, Dalton Trans. ,2193 (1986). 241. A. D. Paolini and E. Goria. Gazz. Chim. Ital. 63, 1048 (1932). 242. E. Sanero. Period. Miner., Roma. 7, 171 (1936). 313 243. M . Inoue, M . Kishita, and M . Kubo. Inorg. Chem. 4, 626 (1965). 244. M . Inoue, S. Emori, and M . Kubo. Inorg. Chem. 7, 1427 (1968). 245. S. Emori, M . Inoue, and M . Kubo. Bull. Chem. Soc. Jpn. 45, 2259 (1972). 246. M . Inoue and M . Kubo. J. Coord. Chem. 6, 157 (1977). 247. S. Zaydoun and M . S. Idrissi. Polyhedron 14, 1477 (1995). 248. L . G. Groeneveld, G. Vos, G. C. Verschoor, and J. Reedijk. J. Chem. Soc, Chem. Commun. , 620 (1982). 249. R. Prins, P. J. M . W. L . Birker, J. G. Haasnoot, G. C. Verschoor, and J. Reedijk. Inorg. Chem. 24, 4128 (1985). 250. P. M . Slangen, P. J. Van-Koningsbruggen, J. G. Haasnoot, J. Jansen, S. Gorter, J. Reedijk, H . Kooijman, W. J. J. Smeets, and A. L . Spek. Inorg. Chim. Acta 212, 289 (1993). 251. P. M . Slangen, P. J. Van-Koningsbruggen, K. Goubitz, J. G. Haasnoot, and J. Reedijk. Inorg. Chem. 33, 1121 (1994). 252. P. J. Van-Koningsburggen, J. W. Van-Hal, E. Muller, R. A. G. De-Graaff, J. G. Haasnoot, and J. Reedijk. J. Chem. Soc, Dalton Trans. , 1371 (1993). 253. W. Vreugdenhil, J. H . Van-Diemen, R. A. G. De-Graaff, J. G. Haasnoot, J. Reedijk, A. M . Van-der-Kraan, O. Kahn, and J. Zarembowitch. Polyhedron 9, 2971 (1990). 254. W. W. Porterfield. Inorganic chemistry. A unified approach. Addison. Reading, Mass. 1984. p. 168 255. M . Angaroni, G. A. Ardizzoia, T. Beringhelli, G. La-Monica, D. Gatteschi, N . Masciocchi, and M . Moret. J. Chem. Soc, Dalton Trans. , 3305 (1990). 256. F. B. Hulsbergen, R. W. M . t. Hoedt, G. C. Verschoor, J. Reedijk, and A. L. Spek. J. Chem. Soc, Dalton Trans., 539 (1983). 257. B. J. Hathaway. J. Chem. Soc, Dalton Trans. , 1196 (1972). 258. B. Bleaney and K. D. Bowers. Proc. R. Soc. London, Ser. A 226, 95 (1952). 314 259. E. Konig. Landolt-Bornstein numerical data and functional relationships in science and technology. New series. Edited by K. H . Hellwege and A. M . Hellwege. Vol. 8. Springer-Verlag. Berlin. 1966. 260. J. Pons, X . Lopez, J. Casabo, F. Teixidor, A. C. Ruis, and C. Miravitlles. Inorg. Chim. Acta 195, 61 (1992). 261. B . Mernari, F. Abraham, M . Legrenee, M . Drillon, and P. Legoll. J. Chem. Soc, Dalton Trans. , 1707(1993). 262. L . H . Jones. J. Chem. Phys. 22, 217 (1954). 263. M . Inoue, M . Kishita, and M . Kubo. Bull. Chem. Soc. Jpn. 39, 1352 (1966). 264. A. Bencini, C. Benelli, D. Gatteschi, and C. Zanchini. Inorg. Chem. 25, 398 (1986). 265. I. A. Cohen and D. Osterfeld. ACS Symp. Ser. 5 (Extended Interact. Met. Ions Transition Met. Complexes Symp.) , 221 (1974). 266. P. Chaudhuri, I. Karpenstein, M . Winter, C. Butzlaff, E. Bil l , A. X . Trautwein, U . Florke, andH.-J. Haupt. J. Chem. Soc, Chem. Commun. , 321 (1992). 267. N . Matsumoto, J. Kanesaka, A. Ohyoshi, M . Nakamuru, S. Kohata, and H . Okawa. Bull. Chem. Soc. Jpn. 60, 3056 (1987). 268. J. P. Costes, F. Dahan, and J. P. Laurent. Inorg. Chem. 30, 1887 (1991). 269. G. Kolks, S. J. Lippard, J. V. Waszczak, and H . R. Lilienthal. J. Am. Chem. Soc. 104, 717 (1982). 270. C. Benelli, R. K. Bunting, D. Gatteschi, and C. Zanchini. Inorg. Chem. 23, 3074 (1984). 271. J. C. Dewan and S. J. Lippard. Inorg. Chem. 19, 2079 (1980). 272. M . G. B . Drew, M . McCann, and S. M . Nelson. J. Chem. Soc, Dalton Trans. (1981). 273. J. T. Landrum, D. Grimmett, K. J. Haller, W. R. Scheidt, and C. A. Reed. A. Am. Chem. Soc. 103 (1981). 274. J. T. Landrum, C. A. Reed, K. Hatano, and W. R. Scheidt. J. Am. Chem. Soc. 100, 3232 (1978). 315 275. K. Nonoyama, W. Mori, and M . Nonoyama. Polyhedron 13, 891 (1994). 276. J. A. J. Jarvis and A. F. Wells. Acta. Cryst. 13, 1027 (1960). 277. J. E . Bauman-Jr. and J. C. Wang. Inorg. Chem. 3, 368 (1964). 278. G. P. Brown and S. Aftergut. J. Polymer Sci. A2, 1839 (1964). 279. H . C. Freeman. Advan. Protein Chem. 22, 257 (1967). 280. M . Strum, F. Brandl, D. Engel, and W. Hoppe. Acta. Cryst. B31, 2369 (1975). 281. B. K. S. Lundberg. Acta. Chem. Scand. 26, 3902 (1972). 282. J. P. Costes, J. F. Serra, F. Dahan, and J. P. Laurent. Inorg. Chem. 25, 2790 (1986). 283. J. P. Costes and J. P. Laurent. Inorg. Chim. Acta 134, 245 (1987). 284. J. P. Costes and M . I. Fernandez-Garcia. Inorg. Chim. Acta 173, 247 (1990). 285. M . Zong-Wan, H. Qin-Wei, T. Wen-Xia, L . Shi-Xianung, W. Ze-Min, and H. Jin-Ling. Polyhedron 15, 321 (1995). 286. N . Matsumoto, S. Yamashita, and A. Ohyoshi. J. Chem. Soc, Dalton Trans. , 1943 (1988). 287. N . Matsumoto, Y . Maeda, H . Okawa, T. Akui, T. Kawaji, A. Ohyoshi, M . Kodera, H . Oshio, and S. Kida. J. Chem. Soc, Dalton Trans., 943 (1989). 288. N . Matsumoto, H. Okawa, S. Kida, T. Kawaji, and A. Ohyoshi. Bull. Chem. Soc. Jpn. 62, 3812 (1989). 289. N . Matsumoto, H. Takami, K. Inoue, M . Koikawa, Y . Maeda, H . Okawa, and S. Kida. Chem. Lett. , 1393 (1991). 290. N . Matsumoto, K. Inoue, N . Ohba, H . Okawa, and S. Kida. Bull. Chem. Soc. Jpn. 65, 2283 (1992). 291. N . Matsumoto, T. Nozaki, U . Ushio, K. Motoda, M . Ohba, G. Mago, and H . Okawa. J. Chem. Soc, Dalton Trans. , 2157 (1993). 292. K. Inoue, N . Matsumoto, and H . Okawa. Chem. Lett., 1433 (1993). 293. D. F. Rendle, A. Storr, and J. Trotter. Can. J. Chem. 53 (1975). 316 294. K. R. Breakell, D. F. Rendle, A. Storr, and J. Trotter. J. Chem. Soc, Dalton Trans., 1584(1975). 295. R. L . Carlin. Magnetochemistry. Springer-Verlag. Berlin. 1986. p. 7 296. B. N . Figgis, J. Lewis, F. E. Mabbs, and G. A. Webb. J. Chem. Soc. (A), 442 (1967). 297. O. Kahn. Molecular magnetism. V C H Publishers, Inc. New York. 1993. p. 322 298. M . Inoue and M . Kubo. Inorg. Chem. 5, 70 (1966). 299. R. Huttel, O. Schafer, and G. Welzel. Ann. 598, 186 (1956). 300. R. J. Brooklyn and I. L . Finar. J. Chem. Soc. (C) , 1515 (1969). 301. K. J. Von-Auwers and K. Bahr. J. Prakt. Chem. 116, 65 (1927). 302. G. T. Morgan and I. Ackerman. J. Chem. Soc. 123, 1308 (1923). 303. D. Chambers, W. A. Denny, J. S. Buckleton, and G. R. Clark. J. Org. Chem. 50, 4736 (1985). 304. L. Knorr. Justus Liebigs Annln. Chem. 279, 188 (1894). 305. S. Trofimenko. J. Am. Chem. Soc. 89, 3165 (1967). 306. J. F. Cordes. Chem. Ber. 95, 3084 (1962). 307. G. Wilkinson, F. A. Cotton, and J. M . Birmingham. J. Inorg. Nucl. Chem. 2, 95 (1956). 317 A P P E N D I X I S I N G L E C R Y S T A L X - R A Y D I F F R A C T I O N D A T A Table 1-1. Crystallographic data for [CpNi(4-H-3,5-diMepz)]2, [CpNi(4-N0 2-3,5-diMepz)]2 and [CpNi(3,5-F 6diMepz)] 2 a Formula C 2 o H 2 4 N 4 N i 2 C 2 0 H 2 2 N 6 N i 2 O 4 C 2 o H i 2 F 1 2 N 4 N i 2 fw 431.38 527.83 653.72 Crystal system Orthorhombic Orthorhombic Orthorhombic Space group Atrial P 2 i 2 i 2 i Pnma a, A 15.204(1) 17.145(3) 18.7134(8) b, A 17.967(2) 17.738(1) 15.619(2) c , A 6.9786(9) 7.006(1) 7.7263(9) a , ° 90 90 90 p , ° 90 90 90 Y,° 90 90 90 V, A 3 1906.4(6) 2130.8(5) 2258.2(6) z 4 4 4 Pcalc, g/cm3 1.525 1.645 1.923 F(000) 912 1088 1296 Radiation Mo Mo Cu \i, cm"1 19.86 18.08 31.87 Crystal size, mm 0.22 x 0.35 x 0.40 0.13 x 0.25 x 0.25 0.15 x 0.15 x 0.45 Transmission factors 0.76-1.00 0.80-1.00 0.93-1.00 Scan type ©-29 ro-29 co-26 Scan range, deg in co 1.42 + 0.35 tan 6 1.00+ 0.35 tan e A 1.00 + 0.20 tan u Scan speed, deg/min 32 (up to 9 scans) 16 (up to 9 scans) 16 (up to 9 scans) Data collected +h, +k, +1 +h, +k, +/ +h, +k, +1 26max, d e § 65 60 155 318 Crystal decay, % negligible 2.5 2.4 Total reflections 1963 3530 2681 Total unique reflections 1963 3530 2681 Rmerge — — Reflections with I>3G(F2) 1157 1894 1642 No. of variables 117 289 179 R 0.035 0.032 0.058 Rw 0.021 0.026 0.053 gof 2.24 1.43 1.32 Max A/a (final cycle) 0.007 0.002 0.003 Residual density e /A 3 -0.66 to 0.58 -0.26 to 0.28 -0.42 to 1.36 a Temperature 294 K, Rigaku AFC6S diffractometer, Mo Ka radiation (X = 0.71069 A ) or CuKa radiation (k = 1.54178 A ) , graphite monochromator, takeoff angle 6.0°, aperture 6.0 x 6.0 mm at a distance of 285 mm from the crystal, stationary background counts at each end of the scan (scan/background time ratio 2:1), a^F2) = [^(C + 4B)]/Lp2 (S - scan rate, C = scan count, B = normalized background count), function minimized Z^( |F 0 | - |Fc | ) 2 where w = 1 for [CpNi(3,5-F6diMepz)]2 and w= 4F02/G2(F02) for other two complexes, R = E | |Fo | - |F C | | /Z |F 0 | , Rw = (ZH\F0\-\Fc\)2/Zw\F0\2y/2, and gof = [Z W ( |F 0 | - |F C | ) 2 / ( / I I -W)]^. Values given for R, Rw, and gof are based on those reflections with / > 3G(F2). Table 1-2. Crystallographic data for [CpNi(4-H-3,5-diMepz)2]2Ni and [CpNi(4-Cl-3,5-diMepz) 2] 2Ni. a Formula C 3 oH3 8 N 8 Ni 3 C 3oH34Cl 4N 8Ni 3 fw 686.78 824.56 Crystal system Triclinic Monoclinic Space group PI P2\la a, A 10.078(2) 10.2201(7) b, A 16.134(2) 16.174(1) c , A 9.992(1) 10.7207(9) a , ° 91.586(10) 90 111.050(9) 108.193(5) y,° 86.95(1) 90 319 V, A3 1513.9(4) 1683.6(2) z 2 2 Pcalc, g/cm3 1.506 1.626 F(000) 716 844 Radiation Mo Cu \x, cm"1 18.81 51.48 Crystal size, mm 0.15 x 0.15 x 0.45 0.10 x 0.25 x 0.30 Transmission factors 0.78-1.00 0.59-1.00 Scan type Q>-26 ro-29 Scan range, deg in co 1.47+ 0.35 tan G 0.85 + 0.20 tan 9 Scan speed, deg/min 32 (up to 9 scans) 16 (up to 9 scans) Data collected +h, ±k, ±1 +h, +k, ±1 26max, deg 60 155 Crystal decay, % negligible negligible Total reflections 9291 3859 Total unique reflections 8823 3674 Rmerge 0.029 0.018 Reflections with / > 3a(F2) 4499 2113 No. of variables 374 206 R 0.032 0.033 Rw 0.028 0.032 gof 1.61 1.77 Max A/rj (final cycle) 0.0008 0.0008 Residual density e/A3 -0.28 to 0.25 -0.35 to 0.27 a Temperature 294 K, Rigaku AFC6S diffractometer, Mo radiation (k = 0.71069 A) or CuKa radiation (k = 1.54178 A), graphite monochromator, takeoff angle 6.0°, aperture 6.0 x 6.0 mm at a distance of 285 mm from the crystal, stationary background counts at each end of the scan (scan/background time ratio 2:1), a^F2) = [5^(C + 45)]/Lp 2 (S = scan rate, C = scan count, B = normalized background count), function minimized Ivf( |F 0 | - |F c | ) 2 where w = 4F02/o2(F02) for the complexes, R = L | |Fo | - |F c | | /E |F 0 | , Rw = (2w(|F 0|-\Fc\)Vi:w\F0\2y/2, and gof = [Sw(|F0|-|F c|)2/(/w-n)]1 / 2. Values given fori?, Rw, and gof are based on those reflections with / > 3a(F 2). 320 Table 1-3. Final atomic coordinates (fractional) and Beq (10 3 A 2 ) for [CpNi(4-X-3,5-diMepz)]2 (X = H , N0 2 ) , [CpNi(3,5-F6diMepz)]2 ) and [CpNi(4-X-3,5-diMepz)2]2Ni (X = H , CI).* Atom X y z Beq [CpNi(4-H-3,5-diMepz)]2 Ni(l ) 3/4 0.31565(4) 0 2.17(2) Ni(2) 3/4 0.13981(4) 0 2.34(2) N( l ) 0.6601(2) 0.2651(2) 0.1386(6) 2.26(8) N(2) 0.6600(2) 0.1889(2) 0.1339(6) 2.16(8) C(l) 0.5942(3) 0.2878(2) 0.2530(8) 2.24(10) C(2) 0.5518(3) 0.2252(3) 0.3276(8) 2.94(9) C(3) 0.5950(3) 0.1649(3) 0.2500(9) 2.7(1) C(4) 0.5762(4) 0.3683(3) 0.2896(9) 3.5(1) C(5) 0.5796(4)' 0.0834(3) 0.2808(9) 3.8(1) C(6) 3/4 0.4293(4) -0.1047(10) 5.0(2) C(7) 0.6773(4) 0.3896(4) -0.161(1) 5.7(2) C(8) 0.7073(5) 0.3332(4) -0.295(1) 6.6(2) C(9) 3/4 0.0272(4) -0.0960(10) 3.6(2) C(10) 0.6783(4) 0.0656(3) -0.172(1) 4.3(1) C ( l l ) 0.7048(5) 0.1230(3) -0.2929(9) 5.6(2) [CpNi(4-N02-3,5-diMepz)] 2 Ni( l ) 0.09968(4) 0.37237(3) 0.20355(10) 2.52(1) Ni(2) -0.07488(3) 0.36276(4) 0.06292(10) 2.35(1) 0(1) -0.0054(3) 0.6528(2) 0.5594(7) 4.5(1) 0(2) -0.1280(2) 0.6276(2) 0.5260(7) 5.6(1) 0(3) 0.0256(3) 0.0844(2) 0.5690(7) 5.1(1) 0(4) -0.1000(3) 0.1018(2) 0.5481(7) 5.8(1) N( l ) 0.0279(2) 0.4486(2) 0.2881(8) 2.31(10) N(2) -0.0492(2) 0.4410(2) 0.2335(7) 2.3(1) N(3) 0.0378(2) 0.2898(2) 0.2903(8) 2.35(10) N(4) -0.0405(2) 0.2895(2) 0.2397(7) 2.3(1) N(5) -0.0584(3) 0.6120(3) 0.5032(7) 3.2(1) N(6) -0.0325(3) 0.1216(3) 0.5172(7) 3.9(1) C(l) 0.0353(3) 0.5106(3) 0.3949(8) 2.3(1) 321 C(2) -0.0388(3) 0.5436(3) 0.4087(8) * 2.5(1) C(3) -0.0907(3) 0.4973(3) 0.3085(8) 2.5(1) C(4) 0.1103(3) 0.5327(3) 0.4833(9) 3.7(1) C(5) -0.1769(3) 0.5010(3) 0.282(1) 4.2(2) C(6) 0.0511(3) 0.2280(3) 0.3979(8) 2.5(1) C(7) -0.0195(3) 0.1906(3) 0.4163(9) 2.7(1) C(8) -0.0763(3) 0.2310(3) 0.3216(8) 2.7(1) C(9) 0.1300(3) 0.2130(3) 0.4845(10) 4.2(2) C(10) -0.1629(3) 0.2211(3) 0.306(1) 4.2(2) C ( l l ) 0.1544(6) 0.4096(9) -0.055(2) 9.7(4) C(12) 0.1898(6) 0.4383(4) 0.101(2) 7.3(3) C(13) 0.2229(4) 0.3819(8) 0.198(1) 7.1(2) C(14) 0.1998(7) 0.3185(6) 0.118(2) 9.8(4) C(15) 0.1599(6) 0.3385(9) -0.046(2) 10.2(5) C(16) -0.0514(4) 0.3999(4) -0.2261(9) 3.9(2) C(17) -0.1272(4) 0.4207(3) -0.162(1) 3.9(2) C(18) -0.1709(3) 0.3551(3) -0.1283(9) 3.9(1) C(19) -0.1188(4) 0.2957(3) -0.1523(10) 3.8(2) C(20) -0.0471(4) 0.3228(3) -0.2214(10) 3.7(2) [CpNi(3,5-F6diMepz)]2 Ni(l) 0.37708(7) 1/4 0.5094(2) 3.53(3) Ni(2) 0.21014(7) 1/4 0.4879(2) 3.15(3) F(l) 0.4726(2) 0.3426(3) 0.2089(7) 7.5(1) F(2) 0.4641(2) 0.4552(4) 0.3535(7) 8.7(1) F(3) 0.4436(2) 0.4573(4) 0.0854(7) 8.7(1) F(4) 0.1287(2) 0.3436(3) 0.1597(6) 6.3(1) F(5) 0.1305(2) 0.4546(3) 0.3130(6) 8.0(1) F(6) 0.1632(2) 0.4604(3) 0.0510(6) 8.0(1) N( l ) 0.3333(2) 0.3374(3) 0.3704(5) 3.16(10) N(2) 0.2603(2) 0.3377(3) 0.3606(5) . 2.94(9) C(l) 0.3565(3) - 0.3957(4) 0.2546(7) 3.4(1) C(2) 0.3004(3) 0.4333(4) 0.1707(7) 3.9(1) C(3) 0.2417(3) 0.3960(3) 0.2397(7) 3.3(1) C(4) 0.4346(3) 0.4117(5) 0.2277(9) 4.5(1) C(5) 0.1657(3) 0.4119(4) 0.1945(8) 4.1(1) C(6) 0.4820(6) 1/4 0.608(2) 8.2(4) C(7) 0.4442(6) 0.3209(6) 0.668(1) 8.2(3) C(8) 0.3885(5) 0.2940(6) 0.774(1) 8.5(3) C(9) 0.1029(6) 1/4 0.565(1) 6.8(3) 322 C(10) 0.1381(7) 0.3190(5) 0.632(1) 9.9(3) C ( l l ) 0.1889(6) 0.291(1) 0.744(1) 15.4(7) [CpNi(4-H-3,5 -diMepz) 2] 2Ni Ni( l ) 0 0 0 2.16(1) Ni(2) 0.26578(4) 0.11724(3) 0.16918(4) 3.145(10) Ni(3) 1/2 1/2 0 2.30(1) Ni(4) 0.74189(4) 0.37670(3) 0.22317(4) 3.408(10) N( l ) 0.0215(2) 0.0459(1) 0.1838(2) 2.38(5) N(2) 0.1360(2) 0.0932(1) 0.2576(2) 2.69(6) N(3) 0.0077(2) 0.1078(1) -0.0692(2) 2.53(5) N(4) 0.1170(3) 0.1589(1) 0.0063(2) 2.82(6) N(5) 0.4734(3) 0.4608(1) 0.1653(2) 2.75(6) N(6) 0.5764(3) 0.4108(1) 0.2615(2) 3.08(6) N(7) 0.5277(2) 0.3883(1) -0.0527(2) 2.55(5) N(8) 0.6321(2) 0.3373(1) 0.0390(2) 2.66(5) C(l) -0.0564(3) 0.0393(2) 0.2679(3) 2.54(6) C(2) 0.0069(3) 0.0819(2) 0.3944(3) 3.14(7) C(3) 0.1273(3) 0.1144(2) 0.3854(3) 2.90(7) C(4) -0.1888(3) -0.0080(2) 0.2231(3) 3.50(8) C(5) 0.2339(4) 0.1666(2) 0.4936(3) 4.51(9) C(6) -0.0802(3) 0.1496(2) -0.1839(3) 3.11(7) C(7) -0.0281(4) 0.2277(2) -0.1822(4) 4.10(9) C(8) 0.0948(4) 0.2312(2) -0.0635(4) 3.73(8) C(9) -0.2126(3) 0.1147(2) -0.2889(3) 4.12(8) C(10) 0.1924(5) 0.3011(2) -0.0097(4) 5.9(1) C ( l l ) 0.4525(4) 0.0516(3) 0.2968(5) 5.5(1) C(12) 0.4762(4) 0.1372(3) 0.3292(4) 6.0(1) C(13) 0.4668(4) 0.1758(3) 0.2053(5) 5.7(1) C(14) 0.4335(3) 0.1141(3) 0.0942(4) 4.9(1) C(15) 0.4354(3) 0.0365(3) 0.1544(5) 4.9(1) C(16) 0.3657(3) 0.4748(2) 0.2128(3) 3.35(8) C(17) 0.3989(4) 0.4334(2) 0.3414(4) 4.31(9) C(18) 0.5299(4) 0.3946(2) 0.3686(3) 3.98(9) C(19) 0.2354(4) 0.5265(2) 0.1338(4) 4.34(9) C(20) 0.6175(5) 0.3427(2) 0.4952(4) 5.9(1) C(21) 0.4634(3) 0.3455(2) -0.1740(3) 2.84(7) C(22) 0.5265(4) 0.2667(2) -0.1600(3) 3.51(8) C(23) 0.6315(3) 0.2633(2) -0.0267(3) 3.00(7) C(24) 0.3420(4) 0.3820(2) -0.2957(3) 4.17(9) 323 C(25) 0.7320(4) 0.1929(2) 0.0443(4) 4.16(9) C(26) 0.9218(5) 0.3403(4) 0.4138(5) 7.4(1) C(27) 0.8955(4) 0.4287(3) 0.3986(5) 6.6(1) C(28) 0.9230(4) 0.4527(3) 0.2784(5) 5.5(1) C(29) 0.9460(4) 0.3807(3) 0.2106(5) 6.0(1) C(30) 0.9518(4) 0.3123(3) 0.2975(6) 6.9(1) [CpNi(4-Cl-3,5-diMepz)2]2Ni Ni( l ) 1/2 1/2 1/2 2.91(2) Ni(2) 0.26032(6) 0.38608(4) 0.29523(6) 3.84(1) Cl(l) 0.5322(1) 0.18318(6) 0.7342(1) 5.98(3) Cl(2) 0.6983(1) 0.42486(9) 0.0772(1) 6.89(3) N( l ) 0.4747(3) 0.3878(2) 0.5384(2) 3.08(6) N(2) 0.3718(3) 0.3406(2) 0.4561(3) 3.30(6) N(3) 0.5162(3) 0.4658(2) 0.3359(2) 3.17(6) N(4) 0.4180(3) 0.4176(2) 0.2500(3) 3.36(6) C(l) 0.5422(3) 0.3418(2) 0.6429(3) 3.32(8) C(2) 0.4818(3) 0.2651(2) 0.6274(3) 3.63(8) C(3) 0.3758(4) 0.2651(2) 0.5097(4) 3.62(8) C(4) 0.6644(4) 0.3718(2) 0.7503(4) 4.42(9) C(5) 0.2797(4) 0.1971(2) 0.4465(4) 5.2(1) C(6) 0.6213(3) 0.4768(2) 0.2885(3) 3.33(8) C(7) 0.5903(4) 0.4356(2) 0.1712(3) 3.98(9) C(8) 0.4636(4) 0.3995(2) 0.1479(3) 3.92(9) C(9) 0.7464(4) 0.5262(3) 0.3555(4) 4.68(10) C(10) 0.3820(4) 0.3495(3) 0.0323(4) 5.7(1) C ( l l ) 0.0845(4) 0.4616(3) 0.2735(6) 6.0(1) C(12) 0.0671(4) 0.3858(4) 0.3233(5) 6.1(1) C(13) 0.0615(4) 0.3243(3) 0.2297(6) 6.6(1) C(14) 0.0832(5) 0.3616(4) 0.1239(5) 6.9(1) C(15) 0.1040(4) 0.4486(4) 0.1529(5) 6.9(1) * Beq = (m)n2ZZ(Uijaj*aj*(aj!ij)). 324 Table 1-4. Bond lengths (A) with estimated standard deviations in parentheses for [CpNi(4-X-3,5-diMepz)]2 (X = H, N0 2 ) , [CpNi(3,5-F6diMepz)]2, and [CpNi(4-X-3,5-diMepz) 2] 2Ni (X = H , Cl) . a Bond Length Bond Length [CpNi(4-H-3,5-diMepz)]2 Ni( l ) -- N ( l ) 1.905(4) N i ( l ) --C(6) 2.169(8) Ni ( l ) --C(7) 2.063(6) N i ( l ) --C(8) 2.183(7) Ni ( l ) --Cp( l ) 1.78 Ni(2)--N(2) 1.877(4) Ni(2)--C(9) 2.131(7) Ni(2)--C(10) 2.101(6) Ni(2)-- C ( l l ) 2.177(6) Ni(2)--Cp(2) 1.78 N ( l ) - -N(2) 1.370(5) N ( l ) - -C(l) 1.344(6) N(2) - -C(3) 1.348(7) C ( l ) - -C(2) 1.396(7) C ( l ) - -C(4) 1.494(6) C(2) - -C(3) 1.378(7) C(3) - -C(5) 1.498(7) C(6) - -C(7) 1.373(8) C(7) - -C(8) 1.45(1) C(8) - -C(8)* 1.30(2) C(9) - -C(10) 1.396(7) C(10> - C ( l l ) 1.390(9) C ( l l > - C ( l l ) * 1.37(1) [CpNi(4-N0 2 -3,5-diMepz)]2 Ni( l ) -- N ( l ) 1.922(4) Ni ( l ) --N(3) 1.908(4) Ni ( l ) -- C ( l l ) 2.141(9) Ni ( l ) --€(12) 2.066(7) Ni ( l ) --C(13) 2.119(6) Ni ( l ) --C(14) 2.054(7) Ni ( l ) --C(15) 2.12(1) Ni ( l ) -- C p ( l ) 1.77 Ni(2)--N(2) 1.884(4) Ni(2)--N(4) 1.889(4) Ni(2)--C(16) 2.167(6) Ni(2)--C(17) 2.084(6) Ni(2)--C(18) 2.127(5) Ni(2)--C(19) 2.063(6) Ni(2)--C(20) 2.167(7) ' Ni(2)--Cp(2) 1.76 0(1)- -N(5) 1.226(5) 0(2)- -N(5) 1.236(5) 0(3)- -N(6) 1.247(6) 0(4)- -N(6) 1.229(6) N ( l ) - -N(2) • 1.383(5) N ( l ) - -C( l ) ' 1.336(6) N(2)- -C(3) 1.333(6) N(3)- -N(4) 1.389(6) N(3)- -C(6) 1.350(6) N(4)- -C(8) 1.335(6) N(5)- -C(2) 1.423(7) N(6)- -C(7) 1.432(7) C ( l ) - -C(2) 1.401(7) C ( l ) - -C(4) 1.481(7) C(2)- -C(3) 1.400(7) C(3)- -C(5) 1.491(7) C(6)- -C(7) 1.387(7) C(6)- -C(9) 1.506(8) 325 C(7) - -C(8) 1.379(8) C(8)- -C(10) 1.499(8) C(ll)--C(12) 1.35(1) C(ll)--C(15) 1.27(1) C(12)--C(13) 1.34(1) C(13)--C(14) 1.32(1) C(14)--C(15) 1.38(2) C(16> -C(17) 1.423(8) C(16> -C(20) 1.370(7) C(17> -C(18) 1.404(7) C(18)--C(19) 1.391(7) C(19> -C(20) 1.406(9) [CpNi(3,5-F6diMepz)]2 Ni( l ) -- N ( l ) 1.921(4) N i ( l ) --C(6) 2.11(1) Ni ( l ) --C(7) 2.077(7) N i ( l ) --C(8) 2.167(7) Ni ( l ) --Cp( l ) 1.76 Ni(2)--N(2) 1.930(4) Ni(2)--C(9) 2.09(1) Ni(2)--C(10) 2.055(8) Ni(2)-- C ( l l ) 2.119(8) Ni(2)--Cp(2) 1.75 F ( l ) - -C(4) 1.301(7) F ( 2 ) - -C(4) 1.308(7) F ( 3 ) --C(4) 1.320(7) F ( 4 ) --C(5) 1.300(7) F ( 5 ) --C(5) 1.311(7) F ( 6 ) --C(5) 1.344(7) N ( l ) - -N(2) 1.368(5) N ( l ) - -c ( i ) 1.348(7) N(2) - -C(3) 1.350(6) C ( l ) - -C(2) 1.367(7) C ( l ) - -C(4) 1.497(7) C(2)- -C(3) 1.352(7) C(3)- -C(5) 1.486(7) C(6) - -C(7) 1.39(1) C(7)--C(8) 1.39(1) C(8) - -C(8)' 1.37(2) C(9)- -C(10) 1.37(1) C(10> - C ( l l ) 1.36(2) C ( l l > -any 1.27(4) [CpNi(4-H-3,5 -diMepz) 2] 2Ni Ni ( l ) -- N ( l ) 1.904(2) Ni ( l ) --N(3) 1.905(2) Ni(2)--N(2) 1.882(2) Ni(2)--N(4) 1.884(2) Ni(2)-- C ( l l ) 2.105(4) Ni(2)--C(12) 2.183(4) Ni(2)--C(13) 2.187(3) Ni(2)--C(14) 2.078(3) Ni(2)--C(15) 2.138(4) Ni(2)--Cp(l) 1.77 Ni(3)--N(5) 1.896(2) Ni(3)--N(7) 1.899(2) Ni(4)--N(6) 1.892(3) Ni(4)--N(8) 1.887(2) Ni(4)--C(26) 2.174(4) Ni(4)--C(27) 2.069(4) Ni(4)--C(28) 2.148(4) Ni(4)--C(29) 2.110(4) Ni(4)--C(30) 2.188(4) Ni(4)--Cp(2) 1.77 N ( l ) - -N(2) 1.382(3) N ( l ) - -C( l ) 1.349(3) N(2)- -C(3) 1.345(4) N(3)- -N(4) 1.388(3) N(3)- -C(6) 1.342(3) N(4)- -C(8) 1.343(4) N(5)- -N(6) 1.375(3) N(5)- -C(16) 1.340(4) 326 N ( 6 ) - -C(18) 1.349(4) N ( 7 ) - -N(8) 1.371(3) N ( 7 ) - -C(21) 1.342(3) N ( 8 ) - -C(23) 1.346(3) C ( l ) - -C(2) 1.377(4) C ( l ) - -C(4) 1.491(4) C ( 2 ) - -C(3) 1.381(4) C ( 3 ) - -C(5) 1.498(4) C ( 6 ) - -C(7) 1.389(4) C ( 6 ) - -C(9) 1.496(4) C ( 7 ) - -C(8) 1.375(4) C ( 8 ) - -C(10) 1.494(5) C ( l l ) - -C(12) 1.424(6) C ( l l ) - -C(15) 1.386(5) C(12)--C(13) 1.371(6) C(13> -C(14) 1.429(5) C(14> -C(15) 1.400(5) C(16)--C(17) 1.387(4) C(16)--C(19) 1.489(4) C(17)--C(18) 1.368(5) C(18)--C(20) 1.503(4) C(21)--C(22) 1.379(4) C(21)--C(24) 1.488(4) C(22)--C(23) 1.372(4) C(23)--C(25) 1.491(4) C(26)--C(27) 1.439(6) C(26)--C(30) 1.361(6) C(27)--C(28) 1.394(6) C(28)--C(29) 1.380(6) C(29)--C(30) 1.408(6) [CpNi(4-Cl-3,5-diMepz)2]2Ni N i ( l ) -- N ( l ) 1.897(3) N i ( l ) --N(3) 1.900(3) Ni(2)--N(2) 1.894(3) Ni(2)--N(4) 1.892(3) Ni(2)-- C p 1.77 Ni(2)-- C ( l l ) 2.124(4) Ni(2)--C(12) 2.088(4) Ni(2)--C(13) 2.175(4) Ni(2)--C(14) 2.175(4) Ni(2)--C(15) 2.094(4^ C l ( l ) - -C(2) 1.722(3) Cl(2)--C(7) 1.721(4^ N ( l ) - -N(2) 1.373(3) N ( l ) - -C(l) 1.342(4) N(2) - -C(3) 1.344(4) N(3) - -N(4) 1.373(3) N(3) - -C(6) 1.336(4) N(4) - -C(8) 1.350(4) C ( l ) - -C(2) 1.372(4) C ( l ) - -C(4) 1.491(4) C(2) - -C(3) 1.382(5) C(3) - -C(5) 1.490(5) C(6) - -C(7) 1.370(4) C(6) - -C(9) 1.488(5) C(7) - -C(8) 1.371(5) C(8) - -C(10) 1.497(5) C ( l l > -C(12) 1.370(6) C ( l l > -C(15) 1.384(7) C(12> -C(13) 1.402(6) C(13> -C(14) 1.362(7) C(14> -C(15) 1.443(7) a Cp refers to the unweighted centroid of the appropriate cyclopentadienyl ring. Symmetry operations: (*) 3/2-x,y, z (') x, \/2-y, z 327 Table 1-5. Bond angles (deg) with estimated standard deviations in parentheses for [CpNi(4-X-3,5-diMepz)]2 (X = H , N0 2 ) , [CpNi(3,5-F6diMepz)]2, and [CpNi(4-X-3,5-diMepz) 2] 2Ni (X = H , Cl) . a Bonds Angle(deg) Bonds Angle(deg) [CpNi(4-H-3,5-diMepz)]2 N ( l ) - -Ni ( l ) -- N ( l ) 91.7(2) N ( l ) - -Ni ( l ) --Cp( l ) 133.8 N(2) - -Ni(2)--N(2) 93.6(2) N(2)- -Ni(2)--Cp(2) 132.8 Ni ( l ) - - N ( l ) - -N(2) 117.8(3) Ni( l ) -- N ( l ) - - C ( l ) 133.8(3) N(2) - - N ( l ) - -C( l ) 108.4(4) Ni(2)--N(2)-- N ( l ) 118.7(3) Ni(2)--N(2)--C(3) 133.1(4) N ( l ) - -N(2)--C(3) 107.9(4) N ( l ) - - C ( l ) - -C(2) 108.8(4) N ( l ) - - C ( l ) - -C(4) 122.1(5) C(2) - - C ( l ) - -C(4) 129.1(5) C ( l ) - -C(2) - -C(3) 105.5(4) N(2) - -C(3) - -C(2) 109.5(4) N(2)- -C(3) - -C(5) 120.9(5) C(2) - -C(3) - -C(5) 129.6(5) C(7)--C(6) - -C(7)* 107.2(8) C(6) - -C(7) - -C(8) 107.1(5) C(7)--C(8) - -C(8)* 108.3(4) C(10> -C(9) - -C(10)* 102.7(7) C(9)--C(10> - C ( H ) 111.8(6) C(10> - C ( l l ) — C ( l l ) * 106.9(4) [CpNi(4-N0 2 -3,5-diMepz)]2 N ( l ) - -Ni ( l ) --N(3) 94.9(2) N ( l ) - -Ni ( l ) -- C p ( l ) 131.9 N(3)- -Ni( l ) -- C p ( l ) 132.6 N(2)- -Ni(2)--N(4) 91.0(2) N(2)- -Ni(2)--Cp(2) 134.7 N(4)- -Ni(2)--Cp(2) 133.5 Ni ( l ) -- N ( l ) - -N(2) 117.3(3) Ni( l ) -- N ( l ) - - C ( l ) 133.7(4) N(2)- - N ( l ) - -C( l ) 109.0(4) Ni(2)--N(2)-- N ( l ) 118.0(3) Ni(2)--N(2)--C(3) 132.6(4) N ( l ) - -N(2)--C(3) 109.2(4) Ni ( l ) --N(3)--N(4) 117.4(3) Ni( l ) --N(3)--C(6) 135.0(4) N(4)- -N(3)--C(6) 107.6(4) Ni(2)--N(4)--N(3) 117.8(3) Ni(2)--N(4)--C(8) 132.3(4) N(3)--N(4)--C(8) 109.8(4) 0(1)- -N(5)--0(2) 122.8(5) o(i)--N(5)--C(2) 118.6(5) 0(2)- -N(5)--C(2) 118.6(5) 0(3)- -N(6)--0(4) 123.3(5) 0(3)- -N(6)--C(7) 118.2(5) 0(4)- -N(6)--C(7) 118.4(5) N ( l ) - - C ( l ) - -C(2) 107.2(5) N ( l ) - - C ( l ) - -C(4) 122.3(5) C(2)-- C ( l ) - -C(4) 130.4(5) N(5)--C(2) - -C( l ) 127.1(5) N(5)- -C(2) - -C(3) 125.7(5) C ( l ) - -C(2) - -C(3) 107.2(4) N(2)- -C(3) - -C(2) 107.3(4) N(2)--C(3) - -C(5) 120.8(5) C(2)--C(3) - -C(5) 131.8(5) N(3)--C(6) - -C(7) 107.0(5) 328 N(3)—C(6)—C(9) 121.4(5) N(6)—C(7)—C(6) 126.1 (5) C(6)—C(7)—C(8) 108.9(5) N(4)—C(8)—C(10) 121.0(5) C(12)—C(ll)—C(15) 107(1) C(12)—C(13)—C(14) 107.1(9) C(ll)—C(15)—C(14) 109(1) C(16)—C(17)—C(18) 109.0(5) C(18)—C(19)—C(20) 110.1(5) C(7)—C(6)—C(9) 131.5(5) N(6)—C(7)—C(8) 125.0(5) N(4)—C(8)—C(7) 106.6(5) C(7)—C(8)—C(10) 132.4(5) C(ll)—C(12)—C(13) 108.8(9) C(13)—C(14)—C(15) 106.4(9) C(17)—C(16)—C(20) 107.5(6) C(17)—C(18)—C(19) 105.3(5) C(16)—C(20)—C(19) 107.6(6) [CpNi(3,5-F6diMepz)]2 N ( l ) - -Ni( l ) -- N ( i y 90.6(3) N ( l ) - -Ni ( l ) -- C p ( l ) 134.6 N(2) - -Ni(2)--N(2) ' 90.4(2) N(2)- -Ni(2)--Cp(2) 134.6 Ni ( l ) -- N ( l ) - -N(2) 117.3(3) Ni( l ) -- N ( l ) - - C ( l ) 135.5(4) N(2)- - N ( l ) - -C( l ) 106.4(4) Ni(2)--N(2)- - N ( l ) 117.1(3) Ni(2)--N(2)- -C(3) 134.9(3) N ( l ) - -N(2)- -C(3) 107.3(4) N ( l ) - - C ( l ) - -C(2) 111.0(5) N ( l ) - - C ( l ) - -C(4) 121.3(5) C(2)- - C ( l ) - -C(4) 127.7(5) C ( l ) - -C(2) - -C(3) 104.6(5) N(2)- -C(3)- -C(2) 110.7(5) N(2)--C(3) - -C(5) 121.5(5) C(2)--C(3)- -C(5) 127.8(5) F ( l ) - -C(4) - -F(2) 106.5(6) F ( l ) - -C(4)- -F(3) 106.5(6) F ( l ) - -C(4) - -C(l) 114.2(5) F (2 ) - -C(4)- -F(3) 106.5(6) F(2) - -C(4) - -C(l) 113.3(5) F (3 ) - -C(4)- -C(l) 109.3(5) F(4) - -C(5) - -F(5) 107.1(5) F (4 ) - -C(5)- -F(6) 105.9(5) F(4)- -C(5) - -C(3) 115.0(5) F (5 ) - -C(5)- -F(6) 105.8(5) F(5)- -C(5) - -C(3) 113.7(5) F (6 ) - -C(5)- -C(3) 108.7(5) C(7)--C(6) --cay 105(1) C(6)--C(7)- -C(8) 109.6(10) C(7)--C(8) - -C(8)' 107.7(6) C(10> -C(9> —C(10)' 104(1) C(9)--C(10> - C ( l l ) 108(1) C(10> — C ( l l ) — C ( l l ) ' 109.0(9) [CpNi(4-H-3,5-diMepz)2]2Ni N ( l ) - -Ni( l ) -- N ( i y 180.0 N ( l ) - -Ni ( l ) --N(3) 90.22(10) N ( l ) - -Ni( l ) --N(3) ' 89.78(10) N(3)--Ni( l ) --N(3) ' 180.0 N(2)- -Ni(2)--N(4) 91.23(10) N(2)--Ni(2> - C p ( l ) 136.0 N(4)--Ni(2> - C p ( l ) 132.8 N(5)--Ni(3> -N(5)" 180.0 N(5)--Ni(3> -N(7) 88.89(10) N(5)--Ni(3)--N(7)" 91.11(10) N(7)--Ni(3> -N(7)" 180.0 N(6)--Ni(4)--N(8) 91.6(1) N(6)--Ni(4)--Cp(2) 135.7 N(8)--Ni(4)--Cp(2) 132.6 Ni( l ) -- N ( l > -N(2 ) 120.9(2) Ni(l)-- N ( l > - C ( l ) 131.5(2) 329 N(2)—N(l)—C(l) Ni(2)—N(2)—C(3) Ni(l)—N(3)—N(4) N(4)—N(3)—C(6) Ni(2)—N(4)—C(8) Ni(3)—N(5)—N(6) N(6)—N(5)—C(16) Ni(4)—N(6)—C(18) Ni(3)—N(7)—N(8) N(8)—N(7)—C(21) Ni(4)—N(8)—C(23) N ( l ) - C ( l ) - C ( 2 ) C ( 2 ) - C ( l ) - C ( 4 ) N(2)—C(3)—C(2) C(2 ) -C(3 ) -C(5 ) N ( 3 ) - C ( 6 ) - C ( 9 ) C(6 ) -C(7 ) -C(8 ) N(4)—C(8)—C(10) C(12)—C(ll)—C(15) C(12)—C(13)—C(14) C(ll)—C(15)—C(14) N(5)—C(16)—C(19) C(16)—C(17)—C(18) N(6)—C(18)—C(20) N(7)—C(21)—C(22) C(22)—C(21)—C(24) N(8)—C(23)—C(22) C(22)—C(23)—C(25) C(26)—C(27)—C(28) C(28)—C(29)—C(30) N(l)—Ni(l )—N(l)* N(l)—Ni(l)—N(3)* N(2)—Ni(2)—N(4) N(4)—Ni(2)—Cp N i ( l ) - N ( l ) - C ( l ) Ni(2)—N(2)—N(l) N ( l ) - N ( 2 ) - C ( 3 ) Ni(l)—N(3)—C(6) 107.6(2) Ni(2)-- N ( 2 ) - N ( l ) 119.0(2) 132.8(2) N ( l ) - -N(2)-C(3) 108.1(2) 120.3(2) N i ( l ) - -N(3) -C(6) 131.5(2) 108.1(2) Ni(2)--N(4)—N(3) 119.3(2) 132.7(2) N ( 3 ) - -N(4)-C(8) 108.0(2) 120.6(2) Ni(3)--N(5)-C(16) 130.6(2) 108.7(2) Ni(4)--N(6)—N(5) 119.4(2) 133.3(2) N ( 5 ) - -N(6)—C(18) 107.3(3) 120.5(2) Ni(3)--N(7)—C(21) 131.3(2) 108.2(2) Ni(4)--N(8)—N(7) 119.7(2) 132.2(2) N ( 7 ) - -N(8)—C(23) 108.1(2) 109.1(3) N ( l ) - - C ( l ) - C ( 4 ) 122.6(3) 128.3(3) C ( l ) - -C(2)-C(3) 106.3(3) 108.8(3) N(2 ) - -C(3)-C(5) 122.8(3) 128.3(3) N(3 ) - -C(6)-C(7) 108.3(3) 123.3(3) C(7 ) - -C(6)-C(9) 128.4(3) 106.8(3) N(4 ) - _ C (8)-C(7) 108.8(3) 121.9(3) C(7 ) - -C(8)—C(10) 129.3(3) 109.4(4) C ( l l ) --C(12)—C(13) 107.8(4) 107.2(4) C(13)--C(14)—C(15) 109.0(4) 106.2(4) N(5) - -C(16)—C(17) 108.3(3) 122.9(3) C(17)--C(16)—C(19) 128.8(3) 106.5(3) N(6) - -C(18)—C(17) 109.2(3) 122.0(4) C(17> -C(18)—C(20) 128.8(3) 108.4(3) N(7) - -C(21)—C(24) 122.2(3) 129.3(3) C(21)--C(22)—C(23) 106.8(3) 108.5(3) N(8) - -C(23)—C(25) 122.0(3) 129.5(3) C(27)--C(26)—C(30) 106.9(4) 108.2(4) C(27> -C(28)—C(29) 106.7(4) 109.2(4) C(26)--C(30)—C(29) 108.4(4) [CpNi(4-Cl-3,5-diMepz)2]2Ni 180.0 N ( l ) - -Ni(l)—N(3) 88.8(1) 91.2(1) N(3) - -Ni(l)—N(3)* 180.0 91.1(1) N(2) - -Ni(2)—Cp 133.8 135.0 Ni ( l ) -- N ( l ) - N ( 2 ) 121.4(2) 129.9(2) N(2) - - N ( l ) - C ( l ) 108.6(3) 119.4(2) Ni(2)--N(2 ) -C(3 ) 132.2(2) 108.4(3) Ni ( l ) --N(3)—N(4) 121.9(2) 128.8(2) N(4) - -N(3)-C(6) 109.1(3) 330 Ni(2)--N(4)—N(3) 119.0(2) Ni(2)--N(4) -C(8) 133.4(3) N(3) - -N(4)-C(8) 107.6(3) N ( l ) - - C ( l ) - C ( 2 ) 107.8(3) N ( l ) - _C(1)_C(4) 123.6(3) C(2 ) - -C( l ) -C(4 ) 128.6(3) C l ( l ) -- C ( 2 ) - C ( l ) 125.8(3) C l ( l ) --C(2) -C(3) 126.4(3) C ( l ) - -C(2)-C(3) 107.8(3) N(2) - -C(3)-C(2) 107.5(3) N(2) - _ C (3) -C(5) 123.3(3) C(2 ) - -C(3)-C(5) 129.2(3) N(3) - -C(6)-C(7) 107.5(3) N(3) - _ C (6)-C(9) 124.0(3) C(7) - -C(6)-C(9) 128.5(3) Cl(2)-_ C (7 ) -C(6 ) 125.4(3) Cl(2)--C(7) -C(8) 126.4(3) C(6 ) - -C(7)-C(8) 108.0(3) N(4) - _ C (8) -C(7) 107.8(3) N(4) - -C(8)—C(10) 123.0(4) C(7) - -C(8)—C(10) 129.2(4) C(12> —C(ll)—C(15) 107.6(5) C ( l l > -C(12)—C(13) 109.7(5) C(12)--C(13)—C(14) 107.5(4) C(13> -C(14)—C(15) 107.7(5) C( l l ) --C(15)—C(14) 107.1(5) a Cp refers to the unweighted centroid of the appropriate cyclopentadienyl ring. Symmetry operations: (*) 3/2-x,y, z for [CpNi(3,5-diMepz)]2 and 1-x, \-y, 1-z for [CpNi(4-Cl-3,5-diMepz) 2] 2Ni (') x, l/2-y, z for [CpNi(3,5-F6dmpz)]2 and -x, -y, -z for [CpNi(3,5-diMepz) 2] 2Ni (") 1-x, l-y, -z. 331 Table 1-6. Crystallographic data, data collection, and solution and refinement for Cu 3(F 6dmpz) 5. a Formula C25H5CU3F30N10 fw 1206.01 Color, habit green, block Crystal size, mm 0.30x0.20x0.11 Crystal system triclinic Space group PI a, A 9.0557(2) b, A 9.6164(2) c,k 11.8874(3) a, deg 105.406(1) P\ deg 112.317(1) y, deg 90.662(1) V, A 3 915.88(4) Z 1 p c aic g/cm3 2.187 Absorption coefficient, mm"1 1.919 F(000) 582 Diffractometer Siemens SMART Platform CCD Radiation Mo-Ka Wavelength X = 0.71073 A Temperature, K 173 (2) 6 range for data collection, ° 1.94 to 25.09 Index ranges - 1 0 < / » < 9 , - l l < * £ l l , 0 < / < 1 4 Total reflections 5493 Unique reflections 3181 (R i n t = 0.0282) System used SHELXTL-V5.0 Solution Direct methods Refinement method Full-matrix least-squares on F 2 Weighting scheme w = [a 2(F 0 2) + (AP) 2+ (BP)]"1, where P = (F 0 2 + 2Fc2)/3, A = 0.0982, and B = 0.6479 Absorption corrections SADABS (Sheldrick, 1996) Transmission factors 0.675 - 1.000 Data / restraints / parameters 3181 /740/547 332 R indices (I > 2a (I) = 2591) R indices (all data) R l =0.0538, wR2 = 0.1477 R l =0.0654, wR2 = 0.1570 Goodness-of-fit on F z 1.084 Largest diff. Peak and hole, eA' 3 0.768 and -0.743 a Some equations of interest: R i n t = £ | F 2 - ( F O 2 )|/2>2|, R l = S|k|-|Fc||/SlFo|» wR2 = [ £ [ w ( F 2 _ F 2) 2]/s[w(F 2) 2]] , / 2 where w = q / a 2 ( F 2 ) + (a*P) 2 + b*P, GooF = S = [2:[w(F 2-F 2) 2]/(n-p)] , / 2 333 Table 1-7. Atomic coordinates [ x 104] and equivalent isotopic displacement parameters [ A x 103] for Cu3(F6dmpz)5.* Atom X y z U(eq) Cu(l) 4864(1) 6397(1) 6192(1) 29(1) Cu(2) 4169(1) 4927(1) 3473(1) 29(1) Cu(3) 6123(1) 3267(1) 5538(1) 32(1) N ( H ) 2661(8) 4135(8) 4094(6) 30(1) N(12) 2968(7) 4801(7) 5343(5) 28(1) F ( l l ) -622(20) 1458(22) 1930(20) 62(4) F(12) 792(21) 2922(17) 1527(17) 62(4) F(13) 1733(23) 1095(18) 2115(19) 72(4) C ( l l ) 833(22) 2146(21) 2300(13) 48(1) C(12) 1461(24) 3043(30) 3662(13) 36(3) C(13) 911(43) 3060(61) 4585(21) 41(2) C(14) 1979(26) 4044(32) 5663(14) 41(3) C(15) 2073(25) 4428(25) 6981(15) 46(2) F(14) 883(22) 3753(23) 7082(20) 90(6) F(15) 2138(32) 5854(20) 7490(22) 54(4) F(16) 3406(25) 4057(31) 7789(19) 69(5) N(21) 6434(7) 5979(7) 4461(6) 30(1) N(22) 6762(8) 6606(8) 5722(6) 32(1) F(21) 9407(19) 5830(22) 3100(19) 66(4) F(22) 6881(27) 5904(29) 2138(21) 69(5) F(23) 7741(33) 4031(18) 2691(22) 48(3) C(21) 7943(26) 5471(24) 3058(16) 46(2) C(22) 7836(26) 6087(32) 4304(14) 41(3) C(23) 8988(41) 6962(62) 5404(21) 41(2) C(24) 8348(24) 7114(30) 6289(13) 36(3) C(25) 9107(22) 7886(22) 7683(14) 48(1) F(24) 10396(22) 8764(24) 7960(22) 89(7) F(25) 8092(20) 8615(19) 8106(16) 63(4) F(26) 9584(22) 6949(21) 8353(19) 64(4) N(31) 3326(9) 6803(7) 3770(8) 32(2) N(32) 3672(8) 7501(7) 5029(6) 30(1) F(31) 3082(9) 6990(12) 1301(6) 107(3) F(32) 1024(10) 5891(8) 1163(6) 91(2) 334 F(33) 1059(18) 8093(12) 1175(8) 135(6; C(31) 1879(21) 7166(17) 1680(12) 47(5; C(32) 2473(9) 7637(8) 3093(6) 33(2; C(33) 2287(10) 8886(8) 3889(7) 41(2; C(34) 3045(10) 8734(8) 5068(7) 36(2; C(35) 3274(19) 9773(15) 6322(13) 56(2; F(34) 2438(29) 10815(19) 6240(13) 220(6; F(35) 4743(13) 10211(15) 7087(9) 204(8; F(36) 2724(11) 9187(10) 6995(7) 110(3 N(41) 5701(7) 2411(6) 3814(6) 32(1 N(42) 4871(8) 3036(7) 2898(6) 35(1 F(41) 8292(8) 845(8) 4918(8) 98(3 F(42) 6179(8) -554(10) 4323(10) 122(4 F(43) 7554(27) -754(18) 3243(12) 220(6 C(41) 6980(18) 173(14) ' 3923(14) 56(2 C(42) 6030(10) 1123(9) 3201(8) 42(2 C(43) 5483(10) 958(8) 1926(7) 42(2 C(44) 4768(17) 2159(14) 1790(9) 37(3 C(45) 3852(33) 2561(21) 598(13) 51(6 F(44) 3375(34) 3868(17) 864(9) 59(2 F(45) 2609(53) 1631(33) -235(33) 73(6 F(46) 4818(62) 2724(59) 5(45) 67(3 N(51) 6010(7) 5576(7) 7571(5) 32(1 N(52) 6493(7) 4231(6) 7234(5) 30(1 F(51) 7286(53) 8290(33) 10374(33) 73(6 F(52) 5033(62) 7195(59) 10048(45) 67(3 F(53) 5298(12) 8103(9) 8682(7) 57(2 C(51) 6014(30) 7366(19) 9492(12) 37(3 C(52) 6454(68) 5981(34) 8842(17) 59(2 C(53) 7238(9) 4941(8) 9328(7) 39(2 C(54) 7238(9) 3897(7) 8327(7) 34(2 C(55) 7915(20) 2489(16) 8290(13) 43(5 F(54) 7716(10) 1897(8) 9074(8) 80(2 F(55) 7320(12) 1529(8) 7165(6) 113(3 F(56) 9495(9) 2676(10) 8641(10) 93(3 * U(eq) is defined as one third of the trace of the orthogonalized Uy tensor. The site occupancy factor (SOF) is 0.5 for all atoms. 335 Table 1-8. Bond lengths (A) for Cu(II)2Cu(I)(F6dmpz)5, with estimated standard deviations in parentheses. Bond Length Bond Length Cu(l)-N(32) 1.969(6) Cu(l)-N(51) 1.950(6) Cu(l)-N(12) 2.030(6) Cu(l)-N(22) 2.022(6) Cu(2)-N(42) 1.968(6) Cu(2)-N(31) 1.965(6) Cu(2)-N(21) 2.026(6) Cu(2)-N(ll) 2.007(6) Cu(3)-N(41) 1.877(6) Cu(3)-N(52) 1.877(6) N(ll)-N(12) 1.367(8) N(ll)-C(12) 1.35(2) F ( l l ) - C ( l l ) 1.32(2) N(12)-C(14) 1.37(2) F(13)-C(ll) 1.33(2) F(12)-C(ll) 1.32(2) C(12)-C(13) 1.363(14) C(ll)-C(12) 1.495(12) C(14)-C(15) 1.480(13) C(13)-C(14) 1.366(14) C(15)-F(15) 1.33(2) C(15)-F(14) 1.31(2) N(21)-C(22) 1.36(2) C(15)-F(16) 1.35(2) N(22)-C(24) 1.35(2) N(21)-N(22) 1.370(8) F(22)-C(21) 1.32(2) F(21)-C(21) 1.35(2) C(21)-C(22) 1.483(12) F(23)-C(21) 1.32(2) C(23)-C(24) 1.360(14) C(22)-C(23) 1.366(14) C(25)-F(24) 1.31(2) C(24)-C(25) 1.496(12) C(25)-F(25) 1.33(2) C(25)-F(26) 1.33(2) N(31)-N(32) 1.376(9) N(31)-C(32) 1.331(10) F(31)-C(31) 1.32(2) N(32)-C(34) 1.317(9) F(33)-C(31) 1.291(14) F(32)-C(31) 1.30(2) C(32)-C(33) 1.374(10) C(31)-C(32) 1.490(13) C(34)-C(35) 1.494(14) C(33)-C(34) 1.356(10) C(35)-F(35) 1.28(2) C(35)-F(34) 1.26(2) N(41)-C(42) 1.359(9) C(35)-F(36) 1.32(2) N(42)-C(44) 1.330(11) N(41)-N(42) 1.360(9) F(42)-C(41) 1.29(2) F(41)-C(41) 1.32(2) C(41)-C(42) 1.478(13) F(43)-C(41) 1.290(14) C(43)-C(44) 1.347(12) C(42)-C(43) 1.367(10) C(45)-F(45) 1.31(2) C(44)-C(45) 1.500(14) C(45)-F(46) 1.35(2) C(45)-F(44) 1.331(14) N(51)-N(52) 1.377(8) N(51)-C(52) 1.35(2) 336 F(51)-C(51) 1.33(2) F(53)-C(51) 1.322(14) C(52)-C(53) 1.36(2) C(54)-C(55) 1.490(13) C(55)-F(55) 1.31(2) C(55)-F(56) 1.32(2) N(52)-C(54) 1.345(9) F(52)-C(51) 1.32(2) C(51)-C(52) 1.493(14) C(53)-C(54) 1.337(10) C(55)-F(54) 1.28(2) 337 Table 1-9. Bond angles (deg) for Cu(II)2Cu(I)(F6dmpz)5, with estimated standard deviations in parentheses. Bond Angle (deg) Bond Angle (deg) N(51)-Cu(l)-N(22) 94.8(3) N(51)-Cu(l)-N(32) 170.1(3) N(51)-Cu(l)-N(12) 94.3(3) N(32)-Cu(l)-N(22) 89.8(3) N(22)-Cu(l)-N(12) 128.3(2) N(32)-Cu(l)-N(12) 89.6(3) N(31)-Cu(2)-N(ll) 89.1(3) N(31)-Cu(2)-N(42) 171.2(3) N(31)-Cu(2)-N(21) 89.3(3) N(42)-Cu(2)-N(ll) 94.7(3) N(ll)-Cu(2)-N(21) 129.7(2) N(42)-Cu(2)-N(21) 94.3(3) C(12)-N(ll)-N(12) 107.6(7) N(52)-Cu(3)-N(41) 176.2(3) N(12)-N(ll)-Cu(2) 113.8(4) C(12)-N(ll)-Cu(2) 138.4(7) C(14)-N(12)-Cu(l) 139.4(7) C(14)-N(12)-N(ll) 107.0(7) F(12)-C(ll)-F(l l) 109.0(14) N(ll)-N(12)-Cu(l) 113.1(4) F(ll)-C(ll)-F(13) 104.7(14) F(12)-C(ll)-F(13) 106.9(13) F(l l)-C(l l)-C(12) 111.5(13) F(12)-C(ll)-C(12) 112.7(14) N(ll)-C(12)-C(13) 109.7(11) F(13)-C(ll)-C(12) 112(2) C(13)-C(12)-C(ll) 130.5(13) N(l l ) -C(12)-C(l l ) 119.4(12) C(13)-C(14)-N(12) 108.8(12) C(12)-C(13)-C(14) 105.6(10) N(12)-C(14)-C(15) 121.0(12) C(13)-C(14)-C(15) 129.9(13) F(14)-C(15)-F(16) 104(2) F(14)-C(15)-F(15) 108.6(14) F(14)-C(15)-C(14) 113.0(14) F(15)-C(15)-F(16) 104.4(14) F(16)-C(15)-C(14) 112.9(14) F(15)-C(15)-C(14) 113(2) C(22)-N(21)-Cu(2) 138.8(7) C(22)-N(21)-N(22) 107.6(7) C(24)-N(22)-N(21) 106.4(7) N(22)-N(21)-Cu(2) 112.7(4) N(21)-N(22)-Cu(l) 114.0(4) C(24)-N(22)-Cu(l) 139.5(7) F(22)-C(21)-F(21) 107.2(14) F(22)-C(21)-F(23) 107.5(14) F(22)-C(21)-C(22) 113(2) F(23)-C(21)-F(21) 104(2) F(21)-C(21)-C(22) 112.0(13) F(23)-C(21)-C(22) 112.4(14) N(21)-C(22)-C(21) 122.4(12) N(21)-C(22)-C(23) 109.0(12) C(24)-C(23)-C(22) 105.1(10) C(23)-C(22)-C(21) 128.2(13) N(22)-C(24)-C(25) 120.3(12) N(22)-C(24)-C(23) 110.3(12) F(24)-C(25)-F(26) 107(2) C(23)-C(24)-C(25) 128.9(12) F(26)-C(25)-F(25) 105.6(13) F(24)-C(25)-F(25) 110(2) F(26)-C(25)-C(24) 111(2) F(24)-C(25)-C(24) 110.9(14) C(32)-N(31)-N(32) 107.3(6) F(25)-C(25)-C(24) 112.6(13) 338 N(32)-N(31)-Cu(2) 114.3(6) C(34)-N(32)-Cu(l) 139.5(5) F(33)-C(31)-F(32) 109.2(13) F(32)-C(31)-F(31) 103.3(12) F(32)-C(31)-C(32) 112.7(11) N(31)-C(32)-C(33) 110.0(6) C(33)-C(32)-C(31) 129.4(9) N(32)-C(34)-C(33) 111.6(6) C(33)-C(34)-C(35) 128.6(8) F(34)-C(35)-F(36) 99.7(13) F(34)-C(35)-C(34) 114.1(12) F(36)-C(35)-C(34) 112.8(10) C(42)-N(41)-Cu(3) 131.4(5) C(44)-N(42)-N(41) 107.5(7) N(41)-N(42)-Cu(2) 116.5(5) F(43)-C(41)-F(41) 102.1(13) F(43)-C(41)-C(42) 111.0(12) F(41)-C(41)-C(42) 114.9(10) N(41)-C(42)-C(41) 120.7(8) C(44)-C(43)-C(42) 103.9(7) N(42)-C(44)-C(45) 118.8(9) F(45)-C(45)-F(44) 109(2) F(44)-C(45)-F(46) 103(2) F(44)-C(45)-C(44) 111.1(10) C(52)-N(51)-N(52) 107.5(7) N(52)-N(51)-Cu(l) 116.6(4) C(54)-N(52)-Cu(3) 132.7(5) F(53)-C(51)-F(52) 107(2) F(52)-C(51)-F(51) 106.8(14) F(52)-C(51)-C(52) 115(2) N(51)-C(52)-C(53) 109.6(9) C(53)-C(52)-C(51) 129.9(13) C(53)-C(54)-N(52) 111.7(6) N(52)-C(54)-C(55) 119.2(8) F(54)-C(55)-F(56) 104.4(10) F(54)-C(55)-C(54) 112.5(10) F(56)-C(55)-C(54) 111.3(13) C(32)-N(31)-Cu(2) 138.4(6) C(34)-N(32)-N(31) 106.9(6) N(31)-N(32)-Cu(l) 113.7(5) F(33)-C(31)-F(31) 107.0(14) F(33)-C(31)-C(32) 112.5(11) F(31)-C(31)-C(32) 111.5(11) N(31)-C(32)-C(31) 120.6(8) C(34)-C(33)-C(32) 104.2(6) N(32)-C(34)-C(35) 119.7(8) F(34)-C(35)-F(35) 112(2) F(35)-C(35)-F(36) 101.8(12) F(35)-C(35)-C(34) 114.8(11) C(42)-N(41)-N(42) 106.1(6) N(42)-N(41)-Cu(3) 122.4(4) C(44)-N(42)-Cu(2) 136.0(6) F(43)-C(41)-F(42) 107.3(13) F(42)-C(41)-F(41) 106.6(11) F(42)-C(41)-C(42) 114.0(11) N(41)-C(42)-C(43) 110.5(6) C(43)-C(42)-C(41) 128.6(8) N(42)-C(44)-C(43) 111.9(8) C(43)-C(44)-C(45) 129.2(9) F(45)-C(45)-F(46) 107.0(14) F(45)-C(45)-C(44) 115(2) F(46)-C(45)-C(44) 111(2) C(52)-N(51)-Cu(l) 135.9(7) C(54)-N(52)-N(51) 105.7(5) N(51)-N(52)-Cu(3) 121.6(4) F(53)-C(51)-F(51) 104(2) F(53)-C(51)-C(52) 111.1(11) F(51)-C(51)-C(52) 113(2) N(51)-C(52)-C(51) 120.4(12) C(54)-C(53)-C(52) 105.4(9) C(53)-C(54)-C(55) 129.1(8) F(54)-C(55)-F(55) 107.5(13) F(55)-C(55)-F(56) 106.8(11) F(55)-C(55)-C(54) 113.7(10) 339 Table I-10. Crystallographic data for [Fe3(imid)6(imidH)2]x. Formula C24H26Fe3N16 fvv 323.20 Crystal system Monoclinic Space group P2xlc a, A 10.5912(9) b, A 12.958(2) c , A 10.617(1) P,° 92.696(9) F, A3 1455.5(3) z 2 (formula units) Pca/c> g/cm3 1.611 F(000) 720 Radiation (X, A) Mo (0.71069) cm"1 15.21 Crystal size, mm 0.20x0.25 x0.45 Transmission factors 0.94-1.00 Scan type (0-26 Scan range, deg in co 1.15+ 0.35 tan 9 Scan speed, deg/min 16 (up to 8 rescans) Data collected +h, +k, ±1 20max> deg 60 Crystal decay, % negligible Total reflections 4627 Total unique reflections 4413 R-merge 0.029 Reflections with I > 3G(F2) 2260 No. of variables 196 R 0.033 Rw 0.031 gof 1.53 Max A/a (final cycle) 0.0003 Residual density e/A3 -0.31 to 0.29 a Temperature 294 K, Rigaku AFC6S diffractometer, graphite monochromator, takeoff angle 6.0°, aperture 6.0 x 6.0 mm at a distance of 285 mm from the crystal, stationary 340 background counts at each end of the scan (scan/background time ratio 2:1), ^(F2) = [^(C + 4B)]/Lp 2 (S = scan rate, C = scan count, B = normalized background count), function minimized I W ( | F 0 | - | F C | ) 2 where w = 4F 0 2 /c a (F 0 2 ) , R = E | |F 0 | - |F C | | /Z |F 0 | , Rw = (2w((F0|-|F c | ) 2 /Ew|F 0 | 2 ) 1 / 2 , and gof = [Iw(|F 0 | - |F c | ) 2 / (m-«)] 1 / 2 . Values given fori?, Rw, and gof are based on those reflections with I > 3G(7). I > 3G(F2). 341 Table I-11. Final atomic coordinates (fractional) and Beq (A 2) for [Fe3(imid)6(imidH)2]x. * Atom X y z Fe(l) 1/2 0 1/2 2.18(1) Fe(2) -0.01220(4) 0.13074(4) 0.23981(4) 2.306(9) N( l ) 0.3561(2) 0.0357(2) 0.3525(2) 2.55(6) N(2) 0.1716(2) 0.0900(2) 0.2620(2) 2.61(6) N(3) 0.6504(2) 0.0011(2) 0.3643(2) 2.46(6) N(4) 0.8325(2) 0.0441(2) 0.2746(2) 2.60(6) N(5) 0.5317(2) 0.1704(2) 0.5291(2) 2.71(6) N(6) 0.6341(3) 0.3122(3) 0.5793(3) 4.57(9) N(7) -0.0347(2) 0.2462(2) 0.3682(2) 2.75(6) N(8) -0.0392(2) 0.3298(2) 0.5532(2) 2.56(6) C(l) 0.2327(3) 0.0501(3) 0.3639(3) 2.78(7) C(2) 0.2660(3) 0.1025(3) 0.1775(3) 2.92(8) C(3) 0.3760(3) 0.0685(3) 0.2325(3) 2.96(8) C(4) 0.7384(3) 0.0725(3) 0.3499(3) 2.50(7) C(5) 0.7992(3) -0.0545(3) 0.2389(3) 3.57(9) C(6) 0.6894(3) -0.0798(3) 0.2936(3) 3.36(8) C(7) 0.6162(3) 0.2137(3) 0.6079(3) 3.50(9) C(8) 0.5579(4) 0.3347(3) 0.4759(4) 4.4(1) C(9) 0.4948(4) 0.2480(3) 0.4477(3) 3.93(9) C(10) -0.0061(3) 0.2448(3) 0.4925(3) 2.96(7) C ( l l ) -0.0944(3) 0.3913(2) 0.4610(3) 2.72(7) C(12) -0.0915(3) 0.3405(3) 0.3495(3) 2.75(7) *Beq = (8/3)7t2EI/7//a/*a/*(a/-.a7). 342 Table 1-12. Bond lengths (A) with estimated standard deviations in parentheses for [Fe3(imid)6(imidH)2]x. * Bond Length Bond Length Fe(l)—N(l) 2.183(2) Fe(l)--N(3) 2.197(2) Fe(l)—N(5) 2.253(3) Fe(2)--N(2) 2.020(2) Fe(2)—N(4)a 2.039(3) Fe(2)--N(7) 2.046(3) Fe(2)—N(8)b 2.053(2) N ( l ) - -C(l) 1.331(4) N ( l ) - C ( 3 ) 1.369(4) N(2 ) --C(l) 1.338(4) N(2)-C(2) 1.385(4) N(3 ) --C(4) 1.327(4) N(3)-C(6) 1.365(4) N(4 ) --C(4) 1.357(4) N(4)-C(5) 1.375(4) N(5 ) --C(7) 1.320(4) N(5)-C(9) 1.371(4) N(6 ) --C(7) 1.328(4) N(6)-C(8) 1.363(4) N(7 ) --C(10) 1.340(4) N(7)—C(12) 1.373(4) N(8 ) --C(10) 1.332(4) N ( 8 ) - C ( l l ) 1.371(4) C(2 ) --C(3) 1.352(4) C(5)-C(6) 1.363(4) C(8 ) --C(9) 1.333(5) C(ll)—C(12) 1.355(4) * Superscripts refer to symmetry operations: (a) -l+x,y, z (b)x, l/2-y, -1/2+z. 343 Table 1-13. Bond angles (deg) with estimated standard deviations in parentheses for [Fe3(imid)6(imidH)2]x. * Bonds Angle(deg) Bonds Angle(deg) N ( l ) - -Fe(l)-- N ( l ) c 180.0 N ( l ) - -Fe(l)--N(3) 91.82(9) N ( l ) - -Fe(l)--N(3) c 88.18(9) N ( l ) - -Fe(l)--N(5) 89.15(9) N ( l ) - -Fe(l)--N(5) c 90.85(9) N ( 3 ) - -Fe(l)--N(3) c 180.0 N ( 3 ) - -Fe(l)--N(5) 88.59(10) N ( 3 ) - -Fe(l)--N(5) c 91.41(10) N ( 5 ) - -Fe(l)--N(5) c 180.0 N ( 2 ) - -Fe(2)--N(4) a 128.0(1) N ( 2 ) - -Fe(2)--N(7) 104.8(1) N ( 2 ) - -Fe(2)--N(8) b 105.5(1) N(4) b--Fe(2> - N ( 7 ) 99.3(1) N(4) b--Fe(2> —N(8) b 103.58(10) N ( 7 ) - -Fe(2)--N(8) b 116.5(1) Fe(l)-- N ( l ) - - C ( l ) 128.3(2) Fe(l)-- N ( l ) - -C(3) 126.9(2) C ( l ) - - N ( l ) - -C(3) 103.6(2) Fe(2)--N(2) -- C ( l ) 128.5(2) Fe(2)--N(2)--C(2) 128.0(2) C ( l ) - -N(2) --C(2) 103.4(2) Fe(l)--N(3)--C(4) 128.0(2) Fe(l)--N(3) --C(6) 127.1(2) C(4 ) --N(3)--C(6) 103.8(2) Fe(2)d - N ( 4 > - C ( 4 ) 125.4(2) Fe(2)d - N ( 4 ) - C ( 5 ) 131.3(2) C(4 ) --N(4) --C(5) 103.2(3) Fe(l)--N(5)--C(7) 126.5(2) Fe(l)--N(5) --C(9) 126.6(2) C(7 ) --N(5)--C(9) 104.6(3) C(7 ) --N(6) --C(8) 107.7(3) Fe(2)--N(7)--C(10) 128.2(2) Fe(2)--N(7) --C(12) 128.2(2) C(10> -N(7> -C(12) 103.4(3) Fe(2)e - N ( 8 > -C(10) 130.1(2) Fe(2)e -N(8> - C ( l l ) 125.4(2) C(10)--N(8) -- C ( l l ) 104.4(3) N ( l ) - - C ( l ) - -N(2) 114.9(3) N(2) - -C(2 ) --C(3) 108.4(3) N ( l ) - -C(3) --C(2) 109.6(3) N(3) - -C(4 ) --N(4) 114.8(3) N(4) - -C(5) --C(6) 108.6(3) N(3) - -C(6 ) --C(5) 109.7(3) N(5) - -C(7) --N(6) 111.3(3) N(6) - -C(8 ) --C(9) 105.8(3) N(5) - -C(9) --C(8) 110.6(3) N(7) --C(10)--N(8) 114.4(3) N(8) - - C ( l l ) --C(12) 108.4(3) N(7) - -C(12)-- C ( l l ) 109.5(3) "Superscripts refer to symmetry operations: (a) -l+x, y, z (b) x, l/2-y, -1/2+z (c) 1-x, -y, -z (d) l+x,y,z(e)x, I/2-3/, 1/2+z. 344 A P P E N D I X II M A G N E T I C D A T A All susceptibility measurements were made in a SQUID magnetometer at 10 000 G unless otherwise indicated. \ies values calculated from p,eff = 2.828^/XM^ Table II-1. Magnetic susceptibility data for the [Ni(4-X-3,5-diMepz)2]x complexes. [Ni(4-H-3,5-diMepz)2]x [Ni(4-Br-3,5-diMepz)2]x T X M M e^ff T X M M-eff (K) (10'6 c m W 1 ) (UB) (K) (10"6 c m W 1 ) (p.B) 2.00 11100 0.421 2.00 8870 0.377 3.00 8470 0.451 3.00 7770 0.432 4.00 6990 0.473 4.00 7160 0.479 5.00 6150 0.496 5.00 6750 0.520 6.00 5720 0.524 6.00 6550 0.561 7.99 5420 0.589 7.99 6360 0.638 9.99 5450 0.660 9.99 6370 0.713 12.00 5580 0.731 11.99 6440 0.786 15.00 584.0 0.837 14.99 6640 0.892 20.00 6270 1.001 20.00 7030 1.060 25.01 6650 1.153 25.01 7410 1.217 30.02 6960 1.292 30.02 7720 1.361 39.99 7400 1.538 39.99 8120 1.612 50.00 7620 1.746 50.00 8280 1.820 60.00 7670 1.918 60.00 8250 1.990 70.00 7620 2.065 70.00 8090 2.128 80.01 7480 2.188 80.00 7860 2.243 89.99 7290 2.291 89.99 7620 2.342 94.98 7180 2.335 94.98 7480 2.384 99.99 7080 2.379 99.98 7350 2.424 104.99 6970 2.419 104.99 7220 2.462 345 109.98 6870 2.458 109.98 7090 2.497 115.00 6770 2.495 115.01 6960 2.530 120.00 6650 2.526 120.00 6830 2.560 125.01 6550 2.559 125.01 6710 2.590 130.01 6440 2.588 130.01 6580 2.616 135.01 6330 2.614 135.01 6460 2.641 140.00 6230 2.641 140.01 6340 2.664 145.01 6120 2.664 145.01 6230 2.688 150.00 6020 2.687 150.00 6110 2.707 160.00 5820 2.729 160.00 5900 2.748 170.00 5640 2.769 170.00 5700 2.784 180.00 5470 2.806 180.00 5510 2.816 190.00 5310 2.841 190.00 5330 2.846 200.00 5140 2.867 200.00 5160 2.873 210.00 5000 2.898 210.00 5000 2.898 220.00 4860 2.924 220.00 4850 2.921 230.00 4730 2.950 230.00 4720 2.947 240.01 4600 2.971 240.01 4590 2.968 250.01 4490 2.996 250.00 4470 2.990 260.01 4370 3.014 260.01 4350 3.008 270.01 4260 3.033 270.01 4240 3.026 280.01 4160 3.052 280.01 4130 3.041 290.00 4060 3.069 290.01 4030 3.057 300.02 3980 3.090 300.01 3930 3.071 [Ni(4-Cl-3,5-diMepz)2]x [Ni(4-CH3-3,5-diMepz)2]x T X M M-eff (K) (lO"6 c m W 1 ) (n,B) T X M Heff (K) (lO" 6 cm'mol' 1) (UB) 2.00 3.00 4.00 5.00 6.00 8.00 6640 5799 5472 5266 5194 5220 0.326 0.373 0.418 0.459 0.499 0.578 2.00 3.00 4.00 5.00 6.00 6.99 8273 7009 6519 6370 6332 6374 0.364 0.410 0.457 0:505 0.551 0.597 346 9.99 5356 0.654 11.99 5557 0.730 15.00 5879 0.840 20.00 6436 1.015 25.02 6924 1.177 30.02 7311 1.325 39.99 7843 1.584 50.00 8090 1.799 60.00 8126 1.975 70.00 8027 2.120 80.00 7861 2.243 89.99 7640 2.345 94.98 7530 2.392 99.99 7408 2.434 104.98 7298 2.475 109.98 7166 2.511 115.00 7044 2.545 120.00 6922 2.577 125.01 6800 2.607 130.01 6679 2.635 135.01 6558 2.661 140.01 6447 2.687 145.01 6337 2.711 150.00 6226 2.733 160.00 6005 2.772 170.00 5795 2.807 180.00 5607 2.841 190.00 5431 2.873 200.00 5265 2.902 210.00 5122 2.933 220.01 4978 2.960 230.01 4845 2.985 240.01 4713 3.008 250.01 4592 3.030 260.01 4470 3.049 270.01 4360 3.068 280.01 4249 3.085 290.01 4150 3.103 299.98 4040 3.113 7.99 6465 0.643 9.00 6563 0.687 9.99 6678 0.730 10.99 6786 0.772 11.99 6889 0.813 12.99 6982 0.852 15.00 7135 0.925 17.00 7235 0.992 20.00 7342 1.084 22.01 7396 1.141 25.00 7475 1.223 30.01 7628 1.353 35.01 7787 1.477 39.99 7925 1.592 45.00 8011 1.698 50.00 8061 1.795 60.00 8056 1.966 70.00 7953 2.110 80.00 7782 2.231 89.99 7579 2.335 99.98 7359 2.426 109.99 7131 2.505 120.01 6904 2.574 130.02 6681 2.636 140.00 6460 2.689 149.99 6254 2.739 160.00 6049 2.782 170.00 5856 2.822 180.00 5670 2.857 190.00 5487 2.887 200.00 5313 2.915 210.00 5150 2.941 220.01 4992 2.964 230.00 4839 2.983 240.01 4695 3.002 250.01 4559 3.019 260.01 4429 3.035 270.01 4309 3.050 280.01 4199 3.067 290.01 4097 3.083 299.99 3882 3.052 347 Table II-2. Magnetic susceptibility data for the [CpNi(4-X-3,5-diMepz)]2 complexes [CpNi(4-H-3,5-diMepz)]2 T X M Heff (K) (10"6 cm'mol"1) 2.00 348.5 0.075 3.00 318.3 0.087 4.00 300.3 0.098 5.00 275.3 0.105 6.00 259.7 0.112 7.99 235.7 0.123 9.99 218.6 0.132 12.00 206.1 0.141 15.00 192.7 0.152 20.00 181.2 0.170 25.02 172.4 0.186 30.01 164.8 0.199 39.99 152.6 0.221 50.00 148.0 0.243 60.00 140.0 0.259 70.00 129.0 0.269 80.00 123.0 0.281 89.99 120.1 0.294 94.98 118.7 0.300 99.99 118.1 0.307 104.99 117.2 0.314 109.98 117.3 0.321 115.00 116.4 0.327 120.00 116.4 0.334 125.01 114.9 0.339 130.01 114.7 0.345 135.00 114.1 0.351 140.01 114.1 0.357 145.01 113.5 0.363 150.00 112.9 0.368 [CpNi(4-N02-3,5 diMepz)]2 T ( K ) X M (10"6 cn^mor1) M-eff (UB) 2.00 827.6 0.115 3.00 755.1 0.135 4.00 692.1 0.149 5.00 650.0 0.161 6.01 608.6 0.171 7.00 573.4 0.179 7.99 545.0 0.187 9.00 519.0 0.193 9.99 497.9 0.199 10.99 477.1 0:205 11.99 458.6 0.210 12.99 442.2 0.214 15.00 414.7 0.223 17.00 390.1 0.230 20.00 362.8 0.241 22.01 349.0 0.248 25.00 329.1 0.257 30.01 308.6 0.272 35.00 292.5 0.286 39.99 279.7 0.299 45.00 268.8 0.311 50.00 261.9 0.324 60.00 250.9 0.347 70.00 234.0 0.362 80.01 222.6 0.377 89.99 217.2 0.395 99.98 213.0 0.413 109.99 209.3 0.429 120.01 206.9 0.446 130.02 204.4 0.461 348 160.00 111.4 0.378 140.00 202.8 0.477 170.00 110.7 0.388 150.00 200.7 0.491 180.00 109.3 0.397 160.00 199.3 0.505 190.00 108.3 0.406 170.00 198.2 0.519 200.00 106.2 0.412 180.00 197.1 0.533 210.01 106.0 0.422 190.00 195.7 0.545 220.01 114.1 0.448 200.00 196.2 0.560 230.01 112.8 0.456 210.01 195.3 0.573 240.01 112.8 0.465 220.01 194.3 0.585 250.01 112.0 0.473 230.01 193.2 0.596 260.01 111.1 0.481 240.01 193.2 0.609 270.01 111.0 0.490 250.01 192.6 0.620 280.01 109.8 0.496 260.01 191.4 0.631 290.00 108.9 0.503 270.01 191.3 0.643 299.97 107.9 0.509 280.01 192.0 0.656 290.01 192.2 0.668 299.99 191.2 0.677 Table II-3. Magnetic susceptibility and magnetic moment data for the [CpNi(4-X-3,5-diMepz) 2]2Ni complexes. [CpNi(4-H-3,5-diMepz)2]2Ni [CpNi(4-Cl-3,5-diMepz)2]2Ni T X M Heff T X M Heff (K) (lO"6 c m W 1 ) (|iB) . (K) (10"6 c m W 1 ) ( u ^ 299.99 218.0 0.723 299.97 271.7 0.807 290.01 219.9 0.714 290.01 277.5 0.802 280.01 220.8 0.703 280.01 282.3 0.795 270.01 222.5 0.693 270.01 287.2 0.788 260.01 224.5 0.683 260.01 292.9 0.780 250.01 227.4 0.674 250.01 299.6 0.774 240.01 230.0 0.664 240.00 306.1 0.766 230.01 232.2 0.654 230.01 311.7 0.757 220.01 235.2 0.643 220.01 319.3 0.750 210.01 238.4 0.633 210.00 326.8 0.741 349 200.00 242.6 0.623 190.00 247.0 0.613 180.00 251,8 0.602 170.00 257.4 0.592 160.00 263.5 0.581 150.00 270.8 0.570 140.00 279.0 0.559 130.02 288.4 0.548 120.01 299.6 0.536 109.99 313.0 0.525 99.98 330.0 0.514 89.99 350.7 0.502 80.01 378.0 0.492 70.00 418.3 0.484 60.00 492.7 0.486 50.00 586.0 0.484 45.00 632.9 0.477 39.99 693.6 0.471 35.00 768.3 0.464 30.01 865.9 0.456 25.00 1003 0.448 22.01 1120 0.444 .20.00 1214 0.441 17.00 1384 0.434 15.00 1515 0.426 12.99 1660 0.415 11.99 1740 0.408 10.99 1827 0.401 9.99 1925 •. 0.392 9.00 2035 0.383 7.99 2167 0.372 7.00 2321 0.360 6.01 2508 0.347 5.00 2746 0.331 4.00 3041 0.312 3.00 3417 0.286 2.00 3863 0.249 200.00 334.1 0.731 190.00 343.3 0.722 180.00 352.7 0.713 170.00 363.2 0.703 160.00 374.5 0.692 150.00 386.5 0.681 140.01 399.7 0.669 130.01 413.4 0.656 120.01 428.9 0.642 109.99 445.1 0.626 99.98 464.6 0.609 90.00 485.6 0.591 80.00 512.3 0.573 70.00 550.8 0.555 60.00 616.3 0.544 50.00 691.5 0.526 45.00 729.3 0.512 39.99 751.9 0.490 35.01 776.3 0.466 30.01 803.4 0.439 25.00 834.0 0.408 22.01 864.2 0.390 20.00 885.9 0.376 17.00 937.3 0.357 15.00 986.5 0.344 12.99 1050 0.330 11.99 1090 0.323 10.99 1137 0.316 9.99 1193 0.309 8.99 1258 0.301 7.99 1336 0.292 7.00 1430 0.283 6.00 1546 0.272 5.00 1690 0.260 4.00 1860 0.244 3.00 2098 0.224 2.00 2397 0.196 350 Table II-4. Magnetic susceptibility data for the [Mn(4-X-3,5-diMepz)2]x complexes [Mn(4-H-3,5-diMepz)2]x [Mn(4-CH3-3,5-diMepz)2]x T X M HrfF T X M M-dr ( K ) ( c m W 1 ) (ji B) ( K ) ( c m W 1 ) (UB) 2.00 0.08426 1.161 2.00 0.04715 0.868 3.00 0.08143 1.398 3.00 0.04482 1.037 4.00 0.08021 1.602 4.00 0.04364 1.182 5.00 0.07995 1.788 5.00 0.04312 1.313 6.00 0.08025 1.962 6.00 0.04300 1.436 7.00 0.08079 2.127 7.00 0.04309 1.553 7.99 0.08140 2.281 7.99 0.04333 1.664 8.99 0.08197 2.428 9.00 0.04365 1.772 9.99 0.08246 2.567 9.99 0.04399 1.875 9.99 0.08219 2.563 10.99 0.04436 1.975 10.99 0.08252 2.693 11.99 0.04475 2.071 11.99 0.08271 2.816 12.99 0.04512 2.165 12.99 0.08276 2.932 15.00 0.04581 2.344 13.99 0.08268 3.041 17.00 0.04640 2^512 14.99 0.08271 3.149 20.00 0.04700 2.742 14.99 0.08247 3.144 22.02 0.04725 2.885 16.00 0.08217 3.243 25.01 0.04732 3.077 18.00 0.08128 3.421 30.02 0.04686 3.354 20.00 0.08031 3.584 35.01 0.04583 3.582 20.00 0.08011 3.580 39.99 0.04442 3.769 25.01 0.07658 3.914 44.99 0.04294 3.931 30.02 0.07242 4.170 50.00 0.04147 4.072 35.01 0.06825 4.372 60.00 0.03860 4.304 39.99 0.06422 4.532 70.00 0.03590 4.483 45.00 0.06048 4.665 80.00 0.03342 4.624 50.00 0.05704 4.776 90.00 0.03120 4.739 60.00 0.05106 4.950 99.98 0.02923 4.834 70.00 0.04611 5.081 110.00 0.02746 4.915 80.00 0.04197 5.182 120.01 0.02587 4.983 89.99 0.03845 5.261 130.01 0.02443 5.040 351 99.99 0.03547 5.326 109.99 0.03291 5.381 120.01 0.03069 5.427 130.01 0.02872 5.465 140.00 0.02699 5.497 150.00 0.02545 5.526 160.00 0.02409 5.552 170.00 0.02286 5.575 180.00 0.02175 5.595 190.00 0.02074 5.614 200.00 0.01982 5.631 210.01 0.01900 5.649 220.00 0.01823 5.663 230.01 0.01752 5.678 240.00 0.01687 5.690 250.01 0.01626 5.702 260.01 0.01569 5.712 270.01 0.01516 5.722 280.01 0.01468 5.734 290.01 0.01423 5.745 299.99 0.01377 5.748 [Mn(4-Cl-3,5-diMepz)2]x T (K) X M (crr^mor1) Heff (UB) 2.00 0.2205 1.878 3.00 0.1863 2.114 4.00 0.1576 2.245 5.01 0.1351 2.327 6.00 0.1205 2.404 7.00 0.1099 2.481 8.00 0.1024 2.560 8.99 0.09708 2.642 9.99 0.09270 2.721 140.00 0.02314 5.090 150.00 0.02197 5.134 160.00 0.02091 5.173 170.00 0.01995 5.208 180.00 0.01907 5.240 190.00 0.01826 5.267 200.00 0.01750 5.291 210.00 0.01683 5.316 220.00 0.01619 5.337 230.00 0.01561 5.359 240.01 0.01506 5.377 250.01 0.01455 5.394 260.01 0.01407 5.409 270.01 0.01363 5.425 280.01 0.01321 5.439 290.01 0.01282 5:454 300.01 0.01246 5.468 [Mn(4-Br-3,5-diMepz)2]x T X M U-eff (K) (cm3mor1) (M-B) 2.00 • 0.04542 0.852 3.00 0.04663 1.058 4.00 0.04807 1.240 5.00 0.04958 1.408 6.00 0.05101 1.564 7.00 0.05234 1.712 8.00 0.05361 1.852 9.00 0.05479 1.986 9.99 0.05591 2.113 352 10.99 0.08955 2.806 11.99 0.08681 2.885 12.99 0.08462 2.965 15.00 0.08133 3.124 17.00 0.07873 3.272 20.00 0.07572 3.480 22.01 0.07394 3.608 25.00 0.07161 3.784 30.01 0.06778 4.033 35.01 0.06409 4.236 39.99 0.06094 4.415 45.00 0.05765 4.555 50.00 0.05464 4.674 60.00 0.04931 4.864 70.00 0.04479 5.008 80.01 0.04096 5.120 90.00 0.03781 5.217 99.99 0.03494 5.286 110.00 0.03261 5.356 120.01 0.03056 5.416 130.01 0.02851 5.444 140.01 0.02687 5.485 150.00 0.02536 5.516 160.00 0.02413 5.557 170.00 0.02304 5.596 180.00 0.02194 5.620 190.00 0.02098 5.647 200.00 0.02016 5.679 210.00 0.01934 5.699 220.01 0.01866 5.730 230.00 0.01797 5.750 240.01 0.01729 5.761 250.01 0.01674 5.786 260.01 0.01619 5.803 270.01 0.01565 5.814 280.01 0.01524 5.842 290.01 0.01469 5.837 300.00 0.01428 5.853 10.99 0.05701 2.238 11.99 0.05789 2.356 12.99 0.05864 2.468 14.99 0.05984 2.678 17.00 0.06058 2.870 20.01 0.06100 3.124 22.01 0.06087 3:273 25.00 0.06026 3.471 30.01 0.05843 3.745 35.01 0.05613 3.964 39.99 0.05363 4.142 45.00 0.05111 4.289 50.00 0.04860 4.408 60.00 0.04406 4.598 70.00 0.04021 4.744 80.01 0.03695 4.863 90.00 0.03414 4.957 99.98 0.03178 5.041 109.99 0.02973 5.114 120.01 0.02791 5.176 130.01 0.02632 5.231 140.01 0.02489 5.279 150.00 0.02362 5.323 160.00 0.02249 5.365 170.00 0.02146 5.402 180.00 0.02053 5.437 190.00 0.01968 5.468 200.00 0.01891 5.499 210.00 0.01821 5.530 220.00 0.01756 5.558 230.01 0.01697 5.587 240.01 0.01641 5.612 250.01 0.01590 5.638 260.01 0.01542 5.662 270.01 0.01496 5.684 280.01 0.01453 5.703 290.01 0.01413 5.725 300.00 0.01375 5.743 353 Table II-5. Magnetic susceptibility data for the [Mn(4-Xpz) 2(4-XpzH)] x complexes. [Mn(4-Clpz)2(4-ClpzH)] : T (K) X M (cm 3mol 1) M-eff (MB) 2.00 0.2255 1.899 3.00 0.2389 2.394 4.00 0.2446 2.797 5.00 0.2447 3.129 6.00 0.2408 3.399 6.99 0.2347 3.623 7.99 0.2276 3.814 8.99 0.2200 3.978 9.99 0.2123 4.119 10.99 0.2045 4.240 11.99 0.1970 4.347 12.99 0.1898 4.441 15.00 0.1767 4.605 17.00 0.1648 4.734 20.00 0.1494 4.889 22.01 0.1404 4.972 25.01 0.1286 5.073 30.02 0.1129 5.207 35.01 0.1005 5.307 39.99 0.09061 5.383 45.00 0.08229 5.442 50.00 0.07534 5.489 60.00 0.06445 5.561 70.00 0.05630 5.614 80.01 0.05002 5.657 90.00 0.04492 5.686 99.98 0.04090 5.719 109.99 0.03747 5.741 120.01 0.03459 5.762 130.01 0.03209 5.777 140.01 0.02996 5.792 [Mn(4-Brpz)2(4-BrpzH)]x T (K) X M (cm3mof1) Meff (MB)' 2.00 0.22461 1.895 3.00 0.23777 2.388 4.00 0.24343 2.791 5.00 0.24349 3.120 6.00 0.23945 3.390 7.00 0.23337 3.614 8.00 0.22613 3.804 8.99 0.21846 3.963 9.99 0.21073 4.103 10.99 0.20299 4.224 11.99 0.19557 4.331 12.99 0.18838 4.424 15.00 0.17517 4.584 17.01 0.16333 4.714 20.00 0.14792 4.864 22.01 0.13901 4.947 25.00 0.12737 5.046 30.00 0.11173 5.177 35.01 0.09955 5.280 39.99 0.08969 5.356 45.00 0.08144 5.414 50.00 0.07455 5.460 60.00 0.06373 5.530 70.00 0.05565 5.581 80.00 0.04947 5.626 90.00 0.04446 5.657 99.98 0.04035 5.680 109.99 0.03701 5.706 120.01 0.03414 5.725 130.01 0.03167 5.738 140.00 0.02955 5.752 354 150.00 0.02808 5.804 150.00 0.02769 5.763 160.00 0.02641 5.813 160.00 0.02605 5.774 170.00 0.02494 5.823 170.00 0.02461 5.784 180.00 0.02362 5.831 180.00 0.02329 5.791 190.00 0.02245 5.840 190.01 0.02216 5.803 200.00 0.02138 5.847 200.00 0.02113 5.813 210.00 0.02043 5.858 210.00 0.02018 5.822 220.01 0.01955 5.864 220.00 0.01931 5.829 230.00 0.01873 5.870 230.00 0.01852 5.836 240.01 0.01798 5.875 240.01 0.01779 5.843 250.01 0.01730 5.881 250.01 0.01711 5.848 260.01 0.01665 5.884 260.01 0.01648 5.854 270.01 0.01606 5.889 270.01 0.01590 5.859 280.01 0.01551 5.893 280.01 0.01536 5.864 290.01 0.01494 5.886 290.01 0.01486 5.871 299.99 0.01444 5.886 299.99 0.01438 5.875 Table II-6. Magnetic susceptibility data for [Mn(trz)2]. [Mn(trz)2]x 1 XM P-eff (K) (crn'mof1) (UB) 2.00 0.1873 1.731 3.00 0.1804 2.080 4.00 0.1725 2.349 5.01 0.1661 2.580 6.00 0.1605 2.775 7.00 0.1554 2.949 8.00 0.1506 3.104 8.99 0.1460 3.240 9.99 0.1417 3.364 10.99 0.1375 3.477 11.99 0.1337 3.580 12.99 0.1300 3.675 355 15.00 0.1232 3.844 17.00 0.1171 3.990 20.00 0.1089 4.174 22.01 0.1041 4.280 25.00 0.09757 4.417 30.01 0.08839 4.606 35.00 0.08079 4.756 39.99 0.07423 4.873 45.00 0.06871 4.973 50.00 0.06392 5.056 60.00 0.05619 5.192 70.00 0.05009 5.296 80.01 0.04517 5.376 89.99 0.04113 5.440 99.98 0.03778 5.496 109.99 0.03493 5.543 120.01 0.03246 5.582 130.01 0.03031 5.614 140.00 0.02847 5.646 150.00 0.02683 5.673 160.00 0.02533 5.694 170.00 0.02402 5.714 180.00 0.02283 5.733 190.00 0.02177 5.751 200.00 0.02079 5.766 210.01 0.01992 5.785 220.01 0.01910 5.797 230.01 0.01835 5.809 240.01 0.01765 5.821 250.01 0.01701 5.832 260.01 0.01641 5.842 270.01 0.01586 5.852 280.01 0.01535 5.863 290.01 0.01487 5.873 300.01 0.01443 5.885 356 Table II-7. Magnetic susceptibility data for [Cu3(F6diMepz)5]: runs #1 and #2 at 10 0 0 0 G . (Data are per mole of Cu(II)). [Cu3(F6diMepz)5], run#l T X M M-eff ( K ) (10" 6 cn^mol"1) (MB) 2 . 0 0 5 2 0 . 9 0 . 0 9 1 3 . 0 0 3 6 7 . 1 0 . 0 9 4 4 . 0 0 3 3 7 . 4 0 . 1 0 4 5.01 2 8 6 . 1 0 . 1 0 7 6 . 0 0 2 4 7 . 6 0 . 1 0 9 7 . 0 0 2 1 7 . 3 0 . 1 1 0 7 . 9 9 1 9 4 . 0 0.111 8 . 9 9 175.7 0 . 1 1 2 9 . 9 9 161.3 0 . 1 1 4 1 0 . 9 9 149.5 0 . 1 1 5 1 1 . 9 9 139.7 0 . 1 1 6 1 3 . 0 0 131.8 0 . 1 1 7 1 5 . 0 0 113.6 0 . 1 1 7 17.01 101.3 0 . 1 1 7 2 0 . 0 2 9 2 . 0 2 0.121 2 2 . 0 2 8 5 . 2 8 0 . 1 2 2 2 5 . 0 1 7 7 . 3 5 0 . 1 2 4 3 0 . 0 2 6 8 . 7 3 0 . 1 2 8 3 5 . 0 2 5 9 . 9 7 0 . 1 3 0 3 9 . 9 9 5 4 . 8 2 0 . 1 3 2 4 5 . 0 0 5 2 . 1 5 0 . 1 3 7 5 0 . 0 0 5 0 . 1 9 0 . 1 4 2 6 0 . 0 0 4 7 . 7 4 0.151 6 9 . 9 9 4 4 . 9 7 0 . 1 5 9 7 9 . 9 9 4 4 . 9 7 0 . 1 7 0 8 9 . 9 9 4 8 . 7 5 0 . 1 8 7 9 9 . 9 9 5 7 . 3 8 0 . 2 1 4 [Cu3(F6diMepz)5], run #2 T X M (K) ( 1 0 - 6 cnr'mol"1) (MB) 2 . 0 0 4 4 4 . 4 0 . 0 8 4 3 . 0 0 3 0 3 . 5 0 . 0 8 5 3 . 9 9 2 9 5 . 2 0 . 0 9 7 5 . 0 0 2 5 4 . 4 0 . 1 0 1 6 . 0 0 2 1 9 . 3 0 . 1 0 3 7 . 0 0 1 9 3 . 0 0.104 8 . 0 0 173.2 0 . 1 0 5 8 . 9 9 158.2 0 . 1 0 7 9 . 9 9 147.6 0 . 1 0 9 1 0 . 9 9 136.3 0 . 1 0 9 1 1 . 9 9 1 2 3 . 0 0 . 1 0 9 1 2 . 9 9 114.2 0 . 1 0 9 1 5 . 0 0 100.5 0 . 1 1 0 1 7 . 0 0 9 3 . 2 3 0.113 2 0 . 0 0 8 3 . 5 8 0 . 1 1 6 2 2 . 0 1 7 7 . 7 4 0 . 1 1 7 2 5 . 0 0 7 0 . 9 7 0 . 1 1 9 3 0 . 0 1 6 2 . 5 2 0 . 1 2 2 3 5 . 0 0 5 5 . 8 7 0 . 1 2 5 3 9 . 9 9 5 1 . 0 8 0 . 1 2 8 4 5 . 0 0 4 9 . 3 3 0 . 1 3 3 5 0 . 0 0 4 8 . 2 5 0 . 1 3 9 6 0 . 0 0 4 6 . 4 6 0 . 1 4 9 7 0 . 0 0 4 4 . 9 3 0 . 1 5 9 8 0 . 0 0 4 6 . 3 5 0 . 1 7 2 9 0 . 0 0 4 8 . 3 7 0 . 1 8 7 9 9 . 9 8 5 9 . 0 9 0 . 2 1 7 3 5 7 109.98 68.03 0.245 119.99 84.51 0.285 130.00 103.5 0.328 140.01 126.2 0.376 150.00 152.1 0.427 159.99 179.2 0.479 170.00 205.4 0.528 180.00 234.9 0.582 190.00 262.1 0.631 200.00 286.1 0.676 210.00 315.8 0.728 220.00 339.0 0.772 230.00 360.7 0.815 240.00 381.7 0.856 250.00 402.2 0.897 260.01 417.7 0.932 270.01 433.1 0.967 280.01 446.4 1.000 290.01 458.0 1.031 300.00 466.3 1.058 109.99 71.35 0.251 120.01 85.55 0.287 130.01 104.2 0.329 140.00 127.2 0.377 150.00 151.5 0.426 160.00 180.4 0.480 170.00 208.0 0.532 180.00 236.3 0.583 190.00 262.9 0.632 200.00 290.1 0.681 210.00 316.1 0.729 220.00 339.9 0.773 230.01 361.5 0.815 240.01 381.7 0.856 250.00 403.0 0.898 260.01 419.5 0.934 270.01 433.7 0.968 280.01 446.4 1.000 290.01 459.6 1.033 299.99 469.4 1.061 358 Table II-8. Magnetic susceptibility data for [Cu3(F6diMepz)5]: run #3 at 20 000G. [Cu3(F6diMepz)5], run #3 T (K) X M (10"6 cnr'mol"1) (MB) 2.00 493.1 0.089 3.00 393.3 0.097 4.00 314.4 0.100 5.00 271.2 0.104 6.00 235.8 0.106 7.00 209.8 0.108 8.00 188.0 0.110 8.99 170.6 0.111 9.99 157.2 0.112 11.00 143.8 0.112 11.99 136.2 0.114 12.99 128.1 0.115 14.99 114.8 0.117 17.00 103.3 0.119 20.00 92.49 0.122 22.01 85.51 0.123 25.00 77.73 0.125 30.01 68.62 0.128 35.00 61.76 0.132 39.99 56.81 0.135 45.00 52.62 0.138 50.00 50.76 • 0.142 60.00 48.49 0.153 70.00 46.16 0.161 80.01 46.90 0.173 89.99 51.40 0.192 99.98 59.76 0.219 359 110.00 68.91 0.246 120.01 85.51 0.286 130.01 104.4 0.329 140.01 127.7 0.378 150.00 149.7 0.424 160.00 180.5 0.481 170.00 211.2 0.536 180.00 239.4 0.587 190.00 267.8 0.638 200.00 293.7 0.685 210.00 318.6 0.731 220.01 343.1 0.777 230.01 365.3 0.820 240.01 385.3 0.860 250.01 406.7 0.902 260.01 421.4 0.936 270.01 435.3 0.970 280.01 450.7 1.005 290.01 462.4 1.036 299.99 472.6 1.065 Table II-9. Magnetization data for [Cu3(F6diMepz)5] Applied field Magnetization (10"6 E M U ) (Gauss) 2 K « 50 K 100 K 300 K 0 2.504 0.4872 -0.4709 -1.328 50 6.561 2.472 2.082 4.433 100 10.705 4.700 4.636 9.056 250 23.483 12.50 12.63 23.58 500 46.011 24.91 25.51 45.59 750 31.74 33.28 61.16 1000 76.936 36.07 37.27 73.12 2500 35.80 36.14 122.1 5000 225.5 10.87 6.385 179.6 7500 -34.04 -29.80 229.9 360 10000 347.4 -63.04 -66.60 280.7 20000 463.7 -208.9 -210.9 475.9 30000 448.9 -348.3 -355.2 671.2 40000 367.0 -489.7 -500.3 867.8 50000 276.3 -628.2 -643.6 1058 55000 192.9 -689.4 -711.3 1153 Table 11-10. Magnetic susceptibility data for the purple complex from [Cu3(F 6diMepz) 5]. (Xg converted to % M using a molecular weight of 606.3 gmol"1) T X M M-eff ( K ) (lO - 6 ci^mor 1) (MB) 2.00 22350 0.598 3.00 18690 0.670 4.00 15700 0.709 5.01 13450 0.734 6.00 11820 0.753 7.00 10580 0.770 7.99 9598 0.783 9.00 8800 0.796 10.00 8150 0.807 10.99 7605 0.818 12.00 7128 0.827 12.99 6737 0.837 15.00 6089 0.855 17.00 5578 0.871 20.00 5009 0.895 22.01 4722 0.912 25.00 4361 0.934 30.01 3942 0.973 35.01 3627 1.008 39.99 3377 1.039 45.00 3184 1.070 50.00 3010 1.097 60.00 2565 1.109 70.00 2092 1.082 361 80.01 1785 1.069 89.99 1531 1.050 99.98 1326 1.030 109.99 1173 1.016 120.01 1078 1.017 130.01 1025 1.033 140.00 996.1 1.056 150.00 972.4 1.080 160.00 954.0 1.105 170.00 932.8 1.126 180.00 917.4 1.149 190.00 903.1 1.171 200.01 889.2 1.193 210.01 860.9 1.202 220.01 860.7 1.231 230.01 849.2 1.250 240.01 842.8 1.272 250.01 834.5 1.292 260.01 821.1 1.307 270.01 819.6 1.330 280.01 806.7 1.344 290.01 801.5 1.363 299.99 790.7 1.377 Table 11-11. Magnetic susceptibility data for [Fe3(imid)6(irnidH)2]: runs #1, #2 and #3 Run # 1: 10 000 G, data are per mole of Fe Run #2: 10 000 G, data are per mole of Fe T X M Meff T X M | ! e f f (K) ( c m W 1 ) (UB) (K) (crn'mof1) (M-B) 2.00 0.2150 1.854 2.00 0.2150 1.854 3.00 0.2145 2.269 3.00 0.2152 2.272 4.00 0.2130 2.610 4.00 0.2146 2.620 5.00 0.2098 2.897 4.99 0.2129 2.915 362 6.00 0.2062 3.146 7.00 0.2014 3.358 7.99 0.1957 3.536 8.99 0.1895 3.691 9.99 0.1831 3.825 10.99 0.1762 3.936 11.99 0.1681 4.014 12.99 0.1578 4.048 15.00 0.1242 3.859 17.00 0.09733 3.638 20.00 0.09008 3.796 22.01 0.08714 3.917 25.00 0.08314 4.077 30.01 0.07690 4.296 35.00 0.07100 4.458 39.99 0.06561 4.581 45.00 0.06085 4.680 50.00 0.05664 4.759 60.00 0.04944 4.871 70.00 0.04376 4.950 80.00 0.03915 5.005 89.99 0.03537 5.045 99.98 0.03228 5.081 110.00 0.02970 5.112 120.01 0.02742 5.130 130.02 0.02548 5.148 140.00 0.02382 5.165 150.00 0.02234 5.177 160.00 0.02105 5.190 170.00 0.01986 5.196 180.00 0.01881 5.204 190.00 0.01785 5.208 200.00 0.01701 5.215 210.00 0.01628 5.229 220.01 0.01560 5.239 230.01 0.01497 5.248 240.01 0.01439 5.255 250.01 0.01386 5.264 260.01 0.01336 5.271 270.01 0.01290. 5.277 280.01 0.01247 5.285 6.00 0.2100 3.175 7.00 0.2061 3.397 8.00 0.2013 3.588 9.00 0.1961 3.757 10.00 0.1903 3.902 11.00 0.1841 4.024 12.02 0.1769 4.124 13.03 0.1676 4.179 15.03 0.1294 3.944 17.03 0.1040 3.763 20.03 0.09330 3.866 22.03 0.08882 3.956 25.02 0.08317 4.079 30.02 0.07564 4.261 35.01 0.06930 4.405 40.00 0.06399 4.525 45.00 0.05942 4.624 50.00 0.05537 4.705 60.00 0.04907 4.852 70.00 0.04373 4.948 80.00 0.03945 5.024 89.99 0.03598 5.088 99.99 0.03303 5.139 110.01 0.03044 5.175 119.98 0.02824 5.205 130.00 0.02634 5.233 140.01 0.02466 5.255 150.00 0.02317 5.272 160.00 0.02184 5.287 170.00 0.02067 5.301 180.00 0.01960 5.312 190.00 0.01864 5.322 200.00 0.01780 5.336 210.00 0.01701 5.344 220.00 0.01625 5.348 230.00 0.01562 5.360 240.00 0.01501 5.367 250.00 0.01442 5.369 259.99 0.01388 5.372 270.00 0.01338 5.376 280.00 0.01293 5.381 363 290.01 0.01208 5.293 290.00 0.01252 5.388 299.98 0.01170 5.298 300.00 0.01212 5.393 Run #3: 500 G, data are per mole of Fe T M-eff X M T ( K ) (cm3morx) (MB) (cm3Kmol 299.98 0.01253 5.483 3.759 290.01 0.01295 5.480 3.754 280.02 0.01337 5.471 3.743 270.01 0.01380 5.458 3.725 260.00 0.01424 5.442 3.702 250.00 0.01472 5.425 3.679 239.99 0.01526 5.412 3.663 229.98 0.01585 5.399 3.644 219.97 0.01649 5.386 3.627 209.96 0.01718 5.371 3.607 200.01 0.01793 5.355 3.586 190.02 0.01875 5.338 3.563 180.03 0.01967 5.321 3.540 170.04 0.02066 5.301 3.513 160.04 0.02179 5.281 3.487 150.06 0.02302 5.256 3.454 139.99 0.02442 5.229 3.419 130.00 0.02600 5.199 3.380 120.02 0.02781 5.167 3.338 110.00 0.02988 5.127 3.287 99.96 0.03229 5.080 3.227 89.95 0.03515 5.028 3.162 79.96 0.03858 4.967 3.085 70.02 0.04278 4.894 2.995 60.00 0.04800 4.800 2.880 50.00 0.05464 4.674 2.732 45.00 0.05869 4.596 2.641 40.03 0.06331 . 4.502 ' 2.534 364 35.01 0.06858 4.382 2.401 30.02 0.07481 4.238 2.246 25.01 0.08224 4.056 2.057 22.01 0.08761 3.927 1.928 20.01 0.09195. 3.836 1.840 19.00 0.09463 3.792 1.798 18.00 0.09799 3.756 1.764 17.00 0.1028 3.739 1.748 16.00 0.1119 3.784 1.791 15.00 0.1520 4.271 2.280 14.00 1.153 11.36 16.14 12.99 1.766 13.55 22.94 11.99 2.093 14.17 25.10 10.99 2.338 14.33 25.69 9.99 2.538 14.24 25.36 9.00 2.709 13.97 24.39 7.99 2.869 13.54 22.92 6.99 3.008 12.97 21.03 6.00 3.125 12.25 18.75 5.00 3.216 11.34 16.08 4.00 3.275 10.24 13.10 3.00 3.303 8.902 9.908 2.00 3.309 7.275 6.618 Table 11-12. Magnetic data (500 G) for [Fe3(imid)6(imidH)2]: treated as tetrahedral Fe(II) chains with the paramagnetism arising from the octahedral high spin Fe(II) ions subtracted from the data. Octahedral high spin iron susceptibility values used are described in the text. Heff values are the calculated [ies per mole of tetrahedral Fe(II). 1 X M X M M-eff M-eff (K) (crn'mof1) (cm'mol 1) (UB) (UB) v = 0* v = 10* v = 0* v = 10* 18.00 0.05451 0.05220 2.801 2.741 19.00 0.05372 0.05138 2.857 2.794 20.01 0.05376 0.05145 < 2.933 2.869 22.01 0.05423 0.05190 3.090 3.023 365 25.01 0.05482 0.05261 3.311 3.244 30.02 0.05449 0.05252 3.617 3.551 35.01 0.05286 0.05198 3.847 3.815 40.03 0.05076 0.04933 4.031 3.974 45.00 0.04823 0.04710 4.166 4.117 50.00 0.04566 0.04496 4.273 4.240 60.00 0.04092 0.04059 4.431 4.413 70.02 0.03686 0.03691 4.543 4.547 79.96 0.03356 0.03383 4.632 4.651 89.95 0.03081 0.03126 4.708 4.742 99.96 0.02848 0.02905 4.771 4.819 110.00 0.02652 0.02719 4.830 4.891 120.02 0.02480 0.02555 4.879 4.953 130.00 0.02331 0.02410 4.923 5.006 139.99 0.02204 0.02282 4.967 5.055 150.06 0.02089 0.02166 5.007 5.099 160.04 0.01991 0.02065 5.048 5.141 170.04 0.01897 0.01968 5.079 5.174 180.03 0.01816 0.01884 5.113 5.208 190.02 0.01739 0.01804 5.141 5.237 200.01 0.01671 0.01733 5.170 5.266 209.96 0.01608 0.01668 5.197 5.293 219.97 0.01551 0.01608 5.224 5.319 229.98 0.01497 0.01552 5.247 5.342 239.99 0.01448 0.01500 5.272 5.366 250.00 0.01403 0.01452 5.296 5.388 260.00 0.01364 0.01411 5.326 5.417 270.01 0.01329 0.01373 5.357 5.446 280.02 0.01293 0.01336 5.381 5.469 290.01 0.01257 0.01297 5.399 5.485 299.98 0.01219 0.01258 5.408 5.495 * parameter used in calculating the magnetic susceptibility for the magnetically dilute octahedral Fe(II) ions. 366 Table 11-13. Hysteresis magnetization data for [Fe3(imid)6(imidH)2] (data are per mole Fe) Applied field X M Magnetization (G) ( c m W 1 ) (cm'Gmol-1) 0 -288.5 50 -2.861 -143.1 100 2.512 251.2 250 15.59 3898 500 12.56 6282 1000 6.927 6927 5000 1.765 8827 10000 1.061 10610 20000 0.6784 13570 40000 0.4513 18050 50000 0.3982 19910 55000 0.3777 20770 50000 0.3978 19890 40000 0.4526 18110 20000 - 0.6915 13830 10000 1.108 11080 5000 1.904 9521 2500 3.459 8648 1000 8.131 8131 500 15.87 7937 300 26.21 7864 200 39.10 7820 150 51.95 7792 100 77.48 7748 75 102.5 7690 50 152.7 7637 40 188.8 7553 30 248.3 7448 20 364.4 7289 10 697.5 6975 0 6601 -10 -622.7 6227 367 -20 -291.3 5827 -30 -178.3 5348 -40 -122.3 4891 -50 -87.88 4394 -75 -35.05 2629 -100 -14.32 1432 -150 1.212 -182 -200 6.897 -1379 -300 9.635 -2890 -500 7.378 -3689 -1000 5.620 -5620 -2500 3.208 -8019 -5000 1.841 -9206 -10000 1.093 -10930 -20000 0.6929 -13860 -40000 0.4571 -18290 -50000 0.4027 -20140 -55000 0.3814 -20980 -50000 0.4023 -20120 -40000 0.4580 -18320 -20000 0.6959 -13920 -10000 1.114 -11140 -5000 1.910 -9549 -2500 3.476 -8690 -1000 8.154 -8154 -500 15.94 -7969 -300 26.25 -7874 -200 39.14 -7829 -150 51.97 -7795 -100 77.57 -7757 -75 102.9 -7718 -50 152.6 -7629 -40 188.6 -7544 -30 247.9 -7436 -20 364.7 -7293 -10 700.3 -7003 0 -6570 10 -615.8 -6158 368 20 -287.6 -5752 30 -178.1 -5342 40. -121.9 -4877 50 -88.73 -4437 75 -38.73 -2905 100 -15.63 -1563 150 -0.4681 -70.22 200 8.906 1781 300 8.273 2482 500 7.581 3791 1000 6.094 6094 2500 3.262 8155 5000 1.851 9253 10000 1.098 10980 20000 0.6047 15120 40000 0.4274 19230 50000 0.4031 20160 55000 0.3814 20980 Table 11-14. Magnetization data for [Fe3(imid)6(imidH)2] (data are per mole Fe) Applied field Magnetization (cm3Gmof1) (G) 2 K 4.8 K 13 K 20 K 50 K 100 K 300 K 500 2769 2681 1380 48.46 23.53 15.98 6.348 1000 2851 2766 1461 96.76 46.07 31.96 12.54 1500 2918 2832 1532 145.0 68.59 47.92 18.71 2000 2979 2895 1598 193.6 91.03 63.75 24.82 2500 3038 2955 1662 242.0 113.2 79.71 30.89 5000 3327 3248 1968 484.4 224.2 159.2 61.22 7500 3604 3529 2263 726.3 335.7 238.9 91.62 10000 3870 3796 2551 966.9 447.9 319.0 122.3 15000 4368 4301 3112 1447 671.6 478.3 183.3 20000 4827 4764 3647 1923 893.5 637.8 244.2 25000 5250 5195 4158 2395 1115 796.2 305.3 369 30000 5644 5595 4646 2855 1336 955.9 366.4 35000 6012 5960 5106 3315 1554 1113 427.5 40000 6359 6311 5536 3779 1559 1272 487.8 45000 6684 6639 5955 4349 1751 1429 547.8 50000 6988 6952 6346 4862 1940 1586 607.5 55000 7283 7260 6711 5294 2124 1744 666.2 Table 11-15. 500G, 10 000G susceptibility and moment data for [Cu(trz)2]x. 500 G data Temperature Magnetic Magnetic (K) Susceptibility Moment (cn/mol 1) (B.M.) 2.00 0.1110 1.333 3.00 0.1108 1.631 3.00 0.1109 1.632 4.00 0.1107 1.881 4.00 0.1109 1.883 5.00 0.1107 2.104 5.00 0.1106 2.103 6.00 0.1103 2.301 6.00 0.1103 2.301 7.00 0.1100 2.479 7.00 0.1099 2.479 8.00 0.1094 2.644 8.00 0.1094 2.644 9.00 0.1088 2.799 9.00 0.1088 2.798 10.00 0.1080 2.938 10.00 0.1081 2.938 11.00 0.1072 3.070 11.00 0.1072 3.070 12.00 0.1062 3.192 10 000 G data Temperature Magnetic Magnetic (K) Susceptibility Moment (cn/mol 1) (B.M.) 2.00 0.009623 0.392 3.00 0.009585 0.480 4.00 0.009551 0.553 5.00 0.009523 0.617 6.00 0.009507 0.675 7.00 0.009501 0.729 8.00 0.009498 0.780 9.00 0.009499 0.826 10.00 0.009506 0.872 11.00 0.009511 0.914 12.00 0.009512 0.955 13.00 0.009512 0.994 15.00 0.009484 1.067 17.00 0.009426 1.132 20.00 0.009234 1.215 22.00 0.009018 1.260 25.00 0.008423 1.298 30.00 0.004514 1.041 35.00 0.003630 1.008 40.00 0.003461 1.052 370 12.00 0.1062 3.192 13.00 0.1052 3.306 13.00 0.1051 3.305 14.00 0.1040 3.412 14.00 0.1040 3.412 15.00 0.1026 3.509 15.00 0.1025 3.507 16.00 0.1012 3.599 16.00 0.1010 3.596 17.00 0.09949 3.678 17.00 0.09933 3.675 18.00 0.09770 3.750 18.00 0.09747 3.746 19.00 0.09560 3.811 19.00 0.09541 3.808 20.00 0.09330 3.864 20.00 0.09308 3.859 22.00 0.08753 3.925 22.00 0.08731 3.920 25.00 0.07453 3.861 25.00 0.07452 3.861 30.00 0.005536 1.153 30.00 0.006050 1.205 35.00 0.003273 0.957 35.00 0.003278 0.958 40.00 0.003097 0.996 40.00 0.003103 0.997 45.00 0.002970 1.034 45.00 0.002973 1.034 50.00 0.002865 1.070 60.00 0.002679 1.134 60.00 0.002692 1.137 70.00 0.002520 1.188 70.00 0.002530 1.190 80.00 0.002376 1.233 80.00 0.002385 1.235 90.00 0.002248 1.272 90.00 0.002256 1.274 45.00 0.003340 1.096 50.00 0.003233 1.137 60.00 0.003048 1.209 70.00 0.002880 1.270 80.00 0.002737 1.323 90.00 0.002610 1.371 100.00 0.002497 1.413 110.00 0.002395 1.451 120.00 0.002302 1.486 130.00 0.002214 1.517 140.00 0.002134 1.546 150.00 0.002061 1.572 160.00 0.001993 1.597 170.00 0.001930 1.620 180.00 0.001872 1.641 190.00 0.001818 1.662 200.00 0.001767 1.681 210.00 0.001721 1.700 220.00 0.001676 1.717 230.00 0.001634 1.734 240.00 0.001594 1.749 250.00 0.001556 1.764 260.00 0.001520 1.778 270.00 0.001486 1.791 280.00 0.001454 1.805 290.00 0.001422 1.816 300.00 0.001388 1.825 371 100.00 0.002134 1.306 100.00 0.002140 1.308 110.00 0.002030 1.336 120.00 0.001934 1.362 130.00 0.001846 1.385 140.00 0.001764 1.405 150.00 0.001692 1.425 160.00 0.001625 1.442 170.00 0.001562 1.457 180.00 0.001501 1.470 190.00 0.001446 1.482 200.00 0.001396 1.494 210.00 0.001349 1.505 220.00 0.001306 1.516 230.00 0.001264 1.525 240.00 0.001226 1.534 250.00 0.001191 1.543 260.00 0.001166 1.557 270.00 0.001140 1.569 280.00 0.001111 1.577 290.00 0.001082 1.584 300.00 0.001057 1.592 300.00 0.001082 1.611 Table 11-16. Magnetization data for [Cu(trz)2]x. Applied field Magnetization (cm3Gmor1) (G) 2K 4.8K 13K 25K 50K 100K 300K 500 55.65 55.50 52.64 37.15 1.435 1.071 0.541 1000 57.82 57.65 55.02 39.93 2.855 2.130 1.061 1500 59.96 59.75 57.32 42.62 4.263 3.181 1.567 2000 62.02 61.81 59.59 45.26 5.671 4.226 2.065 2500 64.10 63.84 61.81 47.84 7.080 5.274 2.566 372 5000 73.93 73.53 72.47 60.35 14.08 10.48 5.040 7500 83.49 83.01 82.77 71.83 21.08 15.71 7.510 10000 93.31 92.66 93.11 82.56 28.16 20.95 9.992 15000 115.0 114.0 114.1 102.3 . 42.25 31.44 14.93 20000 139.0 137.8 135.4 120.4 56.30 41.95 19.86 25000 162.5 160.6 155.8 137.6 70.41 52.48 24.84 30000 184.8 182.8 175.4 154.6 84.52 62.93 29.77 35000 206.7 204.3 194.9 171.3 98.65 73.41 34.74 40000 228.1 225.7 214.2 187.8 112.8 83.89 39.68 45000 249.2 246.7 233.5 204.4 126.8 94.34 44.59 50000 270.1 267.5 252.9 221.0 140.8 104.7 49.50 55000 290.7 288.2 272.1 237.4 154.9 115.1 54.39 Table 11-17. Hysteresis magnetization data for [Cu(trz)2]x at 4.8 K. Applied Field Magnetic Susceptibility Magnetization (G) (x lO 3 cn^tnol'1) (cm 3Gmol 1) 0 -52.59 10 -5252 -52.52 20 -2613 -52.26 50 -1033 -51.64 100 -511.7 -51.17 250 -201.2 -50.29 500 -94.88 -47.44 750 -55.75 -41.81 1000 -34.36 -34.36 2000 -2.832 -5.66 3000 5.052 15.16 4000 7.608 30.43 5000 8.569 42.84 7500 9.058 67.94 10000 8.660 86.60 20000 6.874 137.5 30000 6.092 182.8 40000 5.643 225.7 50000 5.349 267.5 55000 5.236 288.0 50000 5.344 267.2 40000 5.642 225.7 30000 6.097 182.9 20000 6.900 138.0 10000 9.325 93.25 7500 11.15 83.64 5000 14.81 74.06 4000 17.54 70.16 3000 22.06 66.17 2000 31.03 62.06 1000 57.84 57.84 750 75.60 56.70 500 111.2 55.58 400 137.7 55.06 300 182.4 54.71 200 271.1 54.22 100 537.3 53.73 75 714.8 53.61 50 1069 53.46 40 1336 53.44 30 1779 53.37 20 2664 53.28 10 5314 53.14 0 52.89 -10 -5252 52.52 -20 -2616 52.32 -30 -1742 52.25 -40 -1304 52.18 -50 -1041 52.05 -75 -692.1 51.91 -100 -517.4 51.74 -200 -256.1 51.23 -300 -168.6 50.58 -400 -124.1 49.64 -500 -96.65 48.32 374 -750 -58.20 43.65 -1000 -37.88 37.88 -2000 -7.050 14.10 -3000 1.957 -5.870 -4000 5.605 -22.42 -5000 7.334 -36.67 -7500 8.701 -65.26 -10000 8.562 -85.62 -20000 6.873 -137.5 -30000 6.088 -182.6 -40000 5.647 -225.9 -50000 5.353 -267.6 -55000 5.237 -288.0 -50000 5.347 -267.4 -40000 5.644 -225.8 -30000 6.099 -183.0 -20000 6.906 -138.1 -10000 9.331 -93.31 -7500 11.15 -83.63 -5000 14.81 -74.05 -4000 17.54 -70.14 -3000 22.05 -66.16 -2000 31.03 -62.07 -1000 57.82 -57.82 -750 75.59 -56.69 . -500 111.2 -55.60 -400 137.8 -55.13 -300 182.4 -54.71 -200 270.9 -54.18 -100 537.5 -53.75 -75 715.0 -53.62 -50 1070 -53.52 -40 1336 -53.44 -30 1779 -53.38 -20 2665 -53.31 -10 5312 -53.12 0 -52.85 10 -5247 -52.47 375 20 -2616 -52.33 30 -1741 -52.22 40 -1302 -52.08 50 -1040 -51.98 75 -691.6 -51.87 100 -517.4 -51.74 200 -256.1 -51.23 300 -168.5 -50.56 400 -124.0 -49.59 500 -96.54 -48.27 750 -58.19 -43.64 1000 -37.84 -37.84 2000 -7.060 -14.12 3000 1.999 5.998 4000 5.646 22.58 5000 7.340 36.70 7500 8.710 65.33 10000 8.560 85.60 20000 6.876 137.5 30000 6.093 182.8 40000 5.644 225.7 50000 5.351 267.6 55000 5.234 287.9 376 A P P E N D I X III I N F R A R E D D A T A Table III-1. Band frequencies (cm"1) and relative intensities for the [CpNi(4-X-3,5-diMepz)]2 (X = H , C H 3 , CI, Br, N 0 2 ) and [CpNi(3,5-F6diMepz)]2 dimetallic compounds. [CpNi(4-H-3,5-diMepz)]2 3097m, 2949m br, 2909m br, 1525m, 1415s, 1398sh, 1359m, 1152w, 1120vw, 1050w, 1019m, 976w, 897w, 834m, 795s, 770s, 653w, 591w, 413m [CpNi(4-CH3-3,5-diMepz)]2 3095w, 2918s br, 1857m, 1514m, 1429m br, 1415w, 1375m, 1351s, 1219m, 1160w, l l l l v w , 1040w, 1021m, 975m, 950vw, 896m, 834m, 783s, 727m, 65 lw, 593w [CpNi(4-Cl-3,5-diMepz)]2 3097w, 2942m br, 1519m, 1420m, 1399m, 1380m, 1362m, 1162m, 1116s, 1039w, 1020m, 979w, 897w, 833m, 786s, 476m [CpNi(4-Br-3,5-diMepz)]2 3096w, 2940w sh, 2919m br, 2856w sh, 1513m, 1421m, 1398m, 1377m, 1355m, 1147m, 1104s, 1095w sh, 1039w, 1020m, 979w, 897w, 833m, 785s, 476m [CpNi(4-N02-3,5-diMepz)]2 3096m, 2940w, 2919w, 1546m, 1471m, 1415m, 1371m, 1353s, 1179m, 1094w, 1038w, 996w, 898vw, 834m, 793m, 773w [CpNi(3,5-F6diMepz)]2 1637vw, 1537m br, 1499m, 1441w, 1399vw, 1376m, 1263s, 1214m, 1169s, 1134s br, 1072w sh, 1026s, 1003w, 989w, 967vw, 916vw, 833w, 820m, 793m br, 770m br, 747vw, 737w, 716w, 680vw * Abbreviations: b, broad; m, medium; s, strong; sh, shoulder; w, weak; vw, very weak. 377 Table III-2. Band frequencies (cm"1) and relative intensities for the [CpNi(4-X-3,5-diMepz) 2]2Ni (X = H , C H 3 , CI, Br) trimetallic compounds. [CpNi(4-H-3,5-diMepz)2]2Ni 3095vw, 2942sh, 2924m br, 2856w sh, 1732m, 1575w, 1527s, 1416s, 1357m, 1332w, 1310w, 1142m, 1039m, 981w, 896vw, 835w, 787s, 770w sh, 754m, 653w, 591w, 413w [CpNi(4-CH 3-3,5-diMepz) 2] 2Ni 3095w, 2919m br, 2858m sh, 1730m, 1620w, 1517m, 1431m, 1415w, 1357m, 1219m, 1140m, 1040w, 1021m, 985m, 898m, 833m, 785s, 727w, 653w, 593w [CpNi(4-Cl-3,5-diMepz)2]2Ni 3096w, 2942sh, 2920m br, 2856sh, 173Is, 163 lw, 1522m, 1435m, 1420m, 1401vw, 1383m, 1357m, 1245w, 1225w, 1162s, 1119s, 1039w, 1019w, 977w, 899vw, 834m, 789s, 770w sh [CpNi(4-Br-3,5-diMepz)2]2Ni 3096w, 2942sh, 2919m br, 2856sh, 1732s, 1629w, 1523m, 1435m, 1422m, 1382m, 1356m, 1219w, 1163s, 1120s, 1041w, 1020w, 996w, 899w, 834m, 790s, 771w sh * Abbreviations: b, broad; m, medium; s, strong; sh, shoulder; w, weak; vw, very weak. Table III-3. Band frequencies (cm"1) and relative intensities for the [Ni(4-Xpz) 2] x (X = H, CI) polymeric compounds. [Ni(4-Hpz)2]x 1717vw, 1488m, 1424m, 1389s, 1283m, 1183 s, 1067s, 869m, 744s, 676w, 626m, 452m, 415vw [Ni(4-Clpz)2]x 1631m, 1517m, 1396s, 1304s, 1219s, 1157s, 1066s, 973s, 824s, 697w sh, 672m, 615s, 467m Abbreviations: m; medium, s, strong; vw, very weak; w, weak 378 Table III-4. Band frequencies (cm"1) and relative intensities for [Ni(indz)2]x. [Ni(indz)2]x 1620s, 1504m, 1379s, 1319m, 1247w, 1209s, 1148m, 1090m, 1015w, 946w, 911m, 809m, 752s, 697w, 655m, 569w, 435m Abbreviations: m, medium; s, strong; vw, very weak; w, weak Table III-5. Band frequencies (cm"1) and relative intensities for the [Ni(4-X-3,5-diMepz) 2] x (X = H , C H 3 , CI, Br) polymeric compounds. [Ni(4-H-3,5-diMepz)2]x 2942sh, 2924m br, 2856w sh, 1730vw, 1576w, 1526s, 1418s br, 1357vw, 1330m, 1137m br, 1084w, 1039m, 979vw, 934w, 772m br, 657w, 587w, 450w [Ni(4-CH3-3,5-diMepz)2]x 2919s br, 2860m sh, 1512s, 1424s, 1374m, 1317s, 1214s, 1134m, 1025w, 987m, 950w, 770m, 721m, 668w, 571w, 503w [Ni(4-Cl-3,5-diMepz)2]x 2924m br, 2856sh, 1576w, 1516s, 1478w, 1423s, 1377m, 1323s, 1145s, 1130s, 1119s, 1039m, 1009vw, 987w, 941vw, 770w sh, 676vw, 66lw, 575w, 515m [Ni(4-Br-3,5-diMepz)2]x 2925m br, 2855sh, 1697w, 1561m, 1508w, 1457w, 1413w, 1376w, 1318w, 1131w, 1098w, 1055s, 1035s, 1011s, 986s, 771m, 674m, 659m, 608m. 509m, 498m, 453w Abbreviations: b, broad; m, medium; s, strong; sh, shoulder; w, weak; vw, very weak. 379 Table III-6. Band frequencies (cm"1) and relative intensities for the [Mn(4-Xpz)2(4-XpzH)] x (X = CI, Br) polymeric compounds. [Mn(4-Clpz)2(4-ClpzH)]x 3347s, 3148m, 2925m, 285 lw, 1720vw, 1648m, 1629m, 1557s, 1511s, 1464m, 1405m sh, 1388s, 1345s, 1309w, 1267m, 1197m, 1168m, 1124s, 1114m, 1088w, 1046s, 1006m, 974s, 968s, 952s, 929w, 861s, 826m, 797w, 749m, 684s, 645w, 604m sh, 592s, 563w sh, 496w [Mn(4-Brpz)2(4-BrpzH)]x 3337m, 3145m br, 3122w sh, 2925w, 2852w, 1552s, 1505m br, 1462m, 1405w sh, 1382s, 1338m, 1310w, 1295m, 1265w, 1235vw, 1197w sh, 1168m, 1125w, 1115m, 1088w, 1041s, 1010m, 975w, 954s, 970w sh, 860w sh, 838m, 799m, 747m, 675m, 650w, 605m, 595w sh, 495w Abbreviations: m, medium; s, strong; vw, very weak; w, weak Table III-7. Band frequencies (cm"1) and relative intensities for the [Mn(4-X-3,5-diMepz) 2] x (X = H , C H 3 , CI, Br) polymeric compounds. [Mn(4-H-3,5-diMepz)2]x 3201w, 3116w, 2923s br, 2870m sh, 1577m, 1524s, 1417s br, 1318s, 1147w, 1113m, 1074m, 1035s, 978w, 775s, 748m sh, 609m [Mn(4-CH3-3,5-diMepz)2]x 3200w, 3116w, 2922s br, 2865m sh, 1573s, 1524s, 1394m, 1317s, 1138m, 1077w, 1025w, 987m, 770s, 725m, 668w, 61 l w [Mn(4-Cl-3,5-diMepz)2]x 3204m, 3128m, 2925m br, 2880m, 1585m, 1512s, 1478m, 1421m br, 1384s, 1313s, 1229s, 1043m, lOOOw, 776m, 725w sh, 604m br [Mn(4-Br-3,5-diMepz)2]x 3200m, 3116m, 2926m br, 2882m br, 1575w, 1506s, 1474w, 1415s br, 1384w, 1311s, 1102s, 1038m, 997w, 774m, 608m br, 441m Abbreviations: m, medium; s, strong; w, weak 380 Table III-8. Band frequencies (cm"1) and relative intensities for [Mn(trz)2]x. [Mn(trz)2]x 1737w, 1500s, 1493s, 1483m, 1402w, 1269s, 1254w, 1193w, 1153s, 1069s, 1046w, 1020w, 991s, 963s, 963w, 912w, 902w, 875m, 863w, 697w, 671s Abbreviations: m, medium; s, strong; w, weak Table III-9. Band frequencies (cm 1) and relative intensities for [Cu3(3,5-F6diMepz)5]. [Cu3(3,5-F6diMepz)5] 3163w, 1640w, 1557m, 1543m, 1505m, 1337m, 1264s, 1231s, 1141sbr, 1034s, 1019m, 992w, 824s, 761 m, 73 7m, 716m, 618w, 558w, 543w, 477w, 442w, 419w, 362w, 307w Abbreviations: m, medium; s, strong; w, weak Table III-10. Band frequencies (cm 1) and relative intensities for the purple product from [Cu3(3,5-F6diMepz)5]. Purple decomposition product 3675m, 1651w, 1557m, 1542m, 1505m, 1384w, 1362w, 1257s, 1237m, 1139sbr, 1071w, 1025s, 985m, 826m, 759m, 736m, 718m, 557m, 545m, 468m, 427w, 413w, 389w, 380w, 369w, 358w, 348w, 341w, 326m, 317m Abbreviations: m, medium; s, strong; w, weak 381 Table III-l 1. Band frequencies (cm"1) and relative intensities for [Fe3(imid)6(imidH)2]x. [Fe3(irmd)6(imidH)2]x 3380s br, 3130w, 1677w, 1618m br, 1532m, 1493wsh, 1462s, 1432w, 1373m, 1322w, 1307m, 1260w, 1159m, 1135w, 1060w sh, 1083s, 1060m sh, 978w,950w sh, 939m, 909w, 83 5w sh, 830m, 774m, 746m, 700w, 673m, 646m, 593 m * Abbreviations: m, medium; s, strong; vw, very weak; w, weak Table III-12. Band frequencies (cm"1) and relative intensities for [Cu2(trz)2]x. [Cu2(trz)2]x '. 174 lw, 1504s, 1486s, 1399w, 1295m, 1274m, 1251m, 1193w, 1156s, 1090s, 1078s, 1032w, 963w, 878m, 865m, 671s Abbreviations: m, medium; s, strong; vw, very weak; w, weak 382 APPENDIX IV MASS SPECTRA Figure TV-1. Mass spectrum of [Cu3(3,5-F6diMepz)5] . « t c : s 2 S S 8 11 111 11 111 1111111111 • ' • ' ' ' ' 1 1 1 s • " " " " s • 383 Figure IV-2. Mass spectrum of the purple product from [Cu3(3,5-F6diMepz)5] 384 A P P E N D I X V P O W D E R X - R A Y D I F F R A C T I O N D A T A Table V - l . d-spacings (A) and relative intensities (%) (in parentheses) for [Ni(4-Hpz)2]x. [Ni(4-Hpz)2]x 3.028(15), 3.244(7), 4.231(9), 4.576(29), 6.027(72), 8.347(56), 8.846(100) Table V-2. d-spacings (A ) and relative intensities (%) (in parentheses) for [Ni(4-X-3,5-diMepz)2]x. [Ni(4-H-3,5-diMepz)2]x 4.647(87), 4.271(100), 3.590(13), 2.764(7), 2.731(9), 2.282(24), 2.151(8), 1.778(8), 1.692(5), 1.616(7), 1.566(9) [Ni(4-CH3-3,5-diMepz)2]x 5.156(100), 3.739(9), 3.059(3), 2.607(11), 1.965(18), 1.818(3), 1.728(4), 1.664(5), 1.578(5), 1.515(3), 1.401(6) [Ni(4-Cl-3,5-diMepz)2]x 5.097(100), 4.530(3), 2.550(5), 2.344(4), 1.722(18), 1.698(2) [Ni(4-Br-3,5-diMepz)2]x 5.216(100), 4.55(14), 3.834(24), 3.079(24), 2.867(12), 2.417(70), 2.321(38), 1.998(39), 1.759(68), 1.728(76), 1.700(28), 1.600(23), 1.467(19), 1.379(36) Table V-3. d-spacings (A ) and relative intensities (%) (in parentheses) for [Mn(4-Clpz)2(4-ClpzH)]x. [Mn(4-Clpz)2(4-ClpzH)]x 5.406(100), 4.696(6), 4.251(81), 3.427(10), 3.315(12), 2.867(13), 2.739(12), 2.699(12), 2.287(28), 2.217(5), 2.181(8), 2.023(3), 2.019(9), 1.828(3), 1.746(4), 1.675(5) 385 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share