Afiwzhgfiéaafi :2.

 

.
t...»
.1; x
1‘
.H.
V :2: \ A
ill; .‘LII
is... ..§....rfi. .5
3" .t......ll!

.v 5.7.531

 

t . n21
..oCh.:i....,.-xxrlai.? o
{ti-.laisxib

3.1".

 

.31..)
.- ’5).

2%.? . i:

 

 

 

THEGlS

.v tat)»-

I,”

This is to certify that the

EXJRNIKA Moi 3:63:20: 611th m g OLA}

A’Yahel-Lion mejml mszpléflXf’S W)

WMnl in la \\r€2entedbyg
Nat/N9 OUJWB

has been accepted towards fulfillment
of the requirements for

Pt. 0 degree in CLWVVD

% may

Major professor

Date /;//fl??

MSU is an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

LIBRARY
Michigan State
University

—4“—‘fi 4 “-—q__

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

moo WW.“

 

EXTENDE
METAL Ci

 

EXTENDED MOLECULAR ARRAYS OF TRANSITION
METAL COMPLEXES WITH POLYNITRILE LIGANDS

By

Xiang Ouyang

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1999

 

EXTEND!
METAL c

 

The mai
Elidostate SUUK
EEterials with
quest in this an
that can beha‘
magnetization i
in may exhib
magnetic propi
We“ Could

temperatures,

 

ABSTRACT

EXTENDED MOLECULAR ARRAYS OF TRANSITION
METAL COMPLEXES WITH POLYNITRILE LIGANDS

by

Xiang Ouyang

The main goal of this project is to design, synthesize and study the
solid-state structures of molecular assemblies and polymeric molecule-based
materials with potentially interesting magnetic or electrical properties. The
quest in this area of research is not just to obtain molecule-based compounds
that can behave as classical solid state magnets, showing spontaneous
magnetization below a certain temperature Tc, but also to unearth materials
that may exhibit completely new physical properties or those in which the
magnetic properties are combined with other properties. Examples of such
systems could be materials showing bistability, tunable magnetic ordering
temperatures, discrete molecules showing magnetic hysteresis
(nanomagnets), or hybrid materials coupling magnetism with conductivity.

In search of new materials, we isolated one-, two- and three-
dimensional polymers with paramagnetic metals joined by organic
acceptors. A main focus has been to synthesize and crystallize paramagnetic
transition metal compounds linked by the stable organic radical
tetracyanoquinodimethanide, T CNQ'. In one case, quadruply bonded metal-
metal building blocks have been introduced into extended networks with
TCNQ and similar molecules. Solid-state structures of l-D polynitrile

structures based on diamagnetic M024+ units have been carried out; these

constitute a \‘3'

 

units such as ti

  
    
   
 

medal attentit
n'siallize and
Inspite of thi
inspired us to t
the CCD area
With the as'ai.
0ppmtunities t
been plagued i
issues will be a

In addit
will“ meta
have been pre;
molecular an,
EChmllufis inc
Wimpy.

fiscussed.

 

constitute a valuable model for other compounds with open shell dimetal
units such as those previously reported for mixed—valence R1125+ compounds.

Two main obstacles were encountered in the project that required
special attention, namely that metal-TCNQ compounds are difficult to
crystallize and that they are notorious for their inherent twinning problems.
In spite of this, their potential applications in molecular based materials
inspired us to undertake crystallographic studies. Fortunately, the advent of
the CCD area detector for single crystal X-ray data collection taken together
with the availability of sophisticated twinning software provided new
opportunities to pursue crystal structure determination. These have long
been plagued by twinning, disorder or small crystal size problems. These
issues will be addressed in detail in the context of modern CCD methods.

In addition to metal-metal bonded precursors and mononuclear
transition metal precursors, polynuclear 3d transition metal halide clusters
have been prepared as potential precursors for the construction of extended
molecular arrays. Characterization of the compounds by solid-state
techniques including single crystal and powder X-ray methods, infrared
spectroscopy, and scanning and transmission electron microscopy are
discussed.

 

To my wife Ying and my son Andrew

 

Firstl u
sise guidance
energeticall y e
hating difficu
costallograph ) |
163m", Robert
Hanhua Zhao,

 

 

many Compou
Juan Modesto
Professor Eug
thank all the p.
Gary, Kenn],
merit and
Many
Chem“? and
immin
25°“ cusual
Whmnm
'W milfvelo
Hie to thank m

D. Ward fOr [h

ahiCEiSaSe

C

ACKNOWLEDGEMENT
First, I would like to thank my advisor-professor Kim R Dunbar. Her

wise guidance and full support were always with me though the years. She
energetically encouraged me to pursue my goal, especially when I was
having difficulties. It was she who led me into my new career —
crystallography. I would also like to thank our “magnetic and conducting
team”, Robert Heintz, Guilio Grandinetti, and especially “Han” (Professor
Hanhua Zhao, Nanjing Normal University, Nanjing, Chaina), who prepared
many compounds for my structural studies. I would also like to thank Dr.
Juan Modesto for his help in fitting magnetic data and to our collaborators
Professor Eugenio Coronado and late Professor Jerry Cowen. I also want to
thank all the past and present Dunbar group members, Staurt, Stacey, Alice,
Gary, Kemal, Calvin, Robert, Matt, Paul, Jen H.& Jen S., Cristian, Liz, Dr.
R. Clerac and Shannon, for their friendship and help.

Many thanks to Michigan Sate University, the Department of
Chemistry and the Center for Fundamental Material Research for financial
support. I would like to thank Dr. Don Ward for many long discussions
about crystallography and twinning. I would like to thank to Dr. Bob
Sparks, for his B—release of the TWINNING 1.0 package that introduced me
to the marvelous world of problematic and twinned structures. I would also
like to thank my committee: Professor G. Baker, Professor J. Geiger and Dr.
D. Ward for their helpful comments, as well as Professor M. Smith for his
advice as a second reader on my defense. Finally appreciation goes to Dr.

C. Campana of Bruker AXS who taught me the operation of the SMART

CCD diffracto
nary helpful d

Last but
their tremendo

hit their patien

 

CCD diffractometer. And to Dr. V. Young of University of Minnesota for
many helpful discussions on the treatment of twinning structures.

Last but certainly not least, I want to thank my wife and my son for
their tremendous support and understanding and my sister and my parents

for their patience and support.

vi

TABLE OF CONTENTS

Page
LIST OF TABLES ..................................................................................... xviii
LIST OF FIGURES .................................................................................... xxiv
LIST OF SYMBOLS AND ABBREVIATIONS ...................................... xxxi
CHAPTER I
INTRODUCTION ........................................................................................... 1
1. History of Molecule-Based Materials (Molecular
Magnets/Molecular Conductors) ............................................................. 2
A. Molecule-based Conductors ................................................................ 2
B. Molecule-based Magnets ..................................................................... 3
2. Interaction of Magnetic Spins .................................................................. 4
A. Orthogonality of Spins ......................................................................... 4
B. Orbital Interactions of Dimetal Complexes with Polynitrile
Ligands ................................................................................................. 6
3. Design of Molecule-based Materials from Inorganic/Organic
Precursors ................................................................................................. 6
A. M(TCNQ) and M(TCNQ); Materials. ................................................. 7
(1) General background ........................................................................ 7
(2) Polymeric metal-TCNQ materials ................................................. 10
B. Metal-Metal Bonded Precursors in Molecule-Based Materials. ....... 12
C. Soluble Transition Metal Coordination Complexes as
Precursors ........................................................................................... 14
4. Handling Crystallographic Problems with a CCD Detector .................. 15
References .................................................................................................. 16
CHAPTER II
SYNTHESES AND CHARACTERIZATION OF EXTENDED
ARRAYS WITH METAL-METAL BONDED COMPLEXES AND
POLYNITRILE ACCEPT OR LIGANDS .................................................... 24

vii

l. Introductit

2. Experimet

A. Starting

B. Synthes

(1) Prep.

(a) 2:1

(b) 2:l
) 2

 

C I

( 1i
(2) Prep:
(

81:1}
be.

 

 

1. Introduction ............................................................................................ 25

2. Experimental Section ............................................................................. 27
A. Starting Materials and Reaction Procedures ..................................... 27
B. Synthesis ............................................................................................ 28

(1) Preparation of Moz(OzCCF3)4(TCNQ) (l). ................................... 28
(a) 2:1 Reaction of M02(02CCF3)4 with TCNQ in THE ................ 28
(b) 2:1 Reaction of Moz(OzCCF3)4 with TCNQ in toluene. ............ 28
(c) 2:1 Reaction of Moz(02CCF3)4 with TCNQ in xylenes ............. 28

(2) Preparation of M02(02CCF3)4(DM-DCNQI) (2). ......................... 29
(a) 1:1 Reaction of M02(OzCCF3)4 with DM-DCNQI in

benzene/CHzClz. ....................................................................... 29
(b) Growth of single crystals by slow diffusion reaction. ............... 29

(3) Preparation of {[M02(OzCCMe3)3(NC)2CC(CN)CONH]
°8CH2C12 }.. (3)_ ............................................................................ 31

(a) Reaction of M02(OzCCMe3)4 with TCNE in CH2C12 at —

78 °C. ........................................................................................ 31

(b) Reaction of M02(02CCMe3)4 with TCNE in CH2C12 at
room temperature. ..................................................................... 31

(c) Single crystal growth of (3) under various low

temperature conditions. ............................................................ 32

(i) Ethylene glycol/C02(s) bath. ................................................... 32

(ii) Freezer storage. ....................................................................... 32
(iii) Low temperature circulating bath .......................................... 32
(iv) Solvent diffusion in a fritted H-tube at ~ -12°C. ................... 33

(4) Preparation of

{[Moz(02CCMe3)3(NC)2CC(CN)CONH]00.5C6116}.. (4). .......... 33

(a) Room temperature reaction. ....................................................... 33

viii

(b) Benzene extraction from the CHzClz reaction product (4). ....... 33

(5) Preparation of [Re2C14(dppm)2]2TCNQ-10THF (5) ...................... 34
(6) Preparation of [Re2C14(dppm)2]2DM-DCNQI-10THF (6). ........... 34
C. X-ray Crystallographic Studies .......................................................... 34
(1) [M02(02CCF3)4]2(TCNQ)'2(m-C8Hto)
0[Moz(OzCCF3)4]-2(m-C8H,0) (1 ). ................................................ 34
(2) M02(02CCF3)4 DM—DCNQI-C6H6 (2). ......................................... 35
(3) {[Moz(02CCMe3)3(NC)2CC(CN)CONH]-8CH2C12 L... (3). .......... 36
(a) Data collected on a Siemens P3N diffractometer. .................... 36
(b) Data collected on a Siemens SMART CCD 1K platform
diffractometer. .......................................................................... 37
(4) {[M02(02CCMe3)3(NC)2CC(CN)CONH]00.5C61-16 }.. (4). ............ 39
(5) Crystallographic study of [Re2C14(dppm)2]2TCNQ°10THF
(5) .................................................................................................. 40
(6) Crystallographic study of [Re2C14(dppm)2]2DM-
DCNQIOIOTHF (6). ...................................................................... 42
3. Results and Discussion .......................................................................... 43
A. Preparation and Characterization ...................................................... 47
(1) Synthesis of the TCNE product of Moz(02CCMe3)4. ................... 47
B. Crystallographic Studies. ................................................................... 50
(l) Moz(02CCF3)4 TCNQ ................................................................... 50
(2) [M02(02CCF3)4(DM-DCN QI)°C6116].. (2) .................................... 53
(3) Structural characterization of transformed TCNE with
M02(OzCCMe3)4 ........................................................................... 61
(a) Data from a P3N instrument partially solved by DIRDIF. ....... 61
(b) Structure solution and refinement of data from a Siemens
SMART CCD. .......................................................................... 63

 

C. Charge
4. Summary

References.

 

(i) [M02(02CCMe3)3(NC)2CC(CN)CONH)]°8CH2C12 (3) ........... 63

(ii) [M02(OgCCMe3)3(NC)2C(CN)CCONH)]00.5C61-16 (4) ........... 66
(4) Structural features of [Re2C14(dppm)2]2TCNQ°10TI-IF(S)
and [Re2C14(dppm)2]2DMDCNQI'10THF (6) .............................. 72
C. Charge Transfer and Infrared Spectral Data. ..................................... 76
4. Summary ................................................................................................ 77
References .................................................................................................. 78
CHAPTER III

PREPARATION, STRUCTURAL CHARACTERIZATION AND
PROPERTIES OF BINARY TRANSITION METAL TCNQ

MATERIALS ................................................................................................ 82
1. Introduction ............................................................................................ 83
2. Experimental Section ............................................................................. 85

A. Synthesis ............................................................................................ 85
(1) Starting materials and reaction procedures. .................................. 85
(2) Bulk syntheses of Cu(TCNQ) phase I (7). .................................... 86

(a) Method A. .................................................................................. 86
(b) Method B. .................................................................................. 86
(3) Bulk synthesis of Cu(TCNQ) phase II (8). ................................... 87
(4) Thin films of Cu(TCNQ) .............................................................. 87
(5) Bulk synthesis of Mn(TCNQ)2(MeOH)2 (9). ................................ 88
(6) Bulk synthesis of Mn(TCNQ)2(H20)2 (10). .................................. 88

B. Single Crystal Growth. ...................................................................... 88
(l) Cu(TCNQ) phase I and phase H (7), (8). ...................................... 88
(2) [Mn(TCNQ)(TCNQ-TCNQ)0,5(MeOH)2]... (9). ............................ 89
(3) [Mn(TCNQ)2(H20)2].. (10). .......................................................... 89
(4) ["Bu4N][TCNQ] (1 1). .................................................................... 89
(5) ["BuaNHTCNQFd (12). ................................................................ 89

(6) [Zn(H20)4('r|l-TCNQ)2°TCNQ-2MeCN].. (13). ............................ 89

(7) [Ni(DMSO)6][TCNQ]3 (14) ........................................................... 89
(8) [Ni(TCNQ)2(HzO)].. (15). ............................................................. 91
C. Physical Measurements ...................................................................... 91
(1) X-ray crystallography. ................................................................... 92
(a) Cu(TCNQ) phase I (7) . ............................................................. 92

(b) Cu(TCNQ) phase H (8) . ............................................................ 92

(c) [Mn(TCNQ)(TCNQ-TCNQ)0,5(MeOH)2]n (9) ........................... 94

(d) [Mn(TCNQ)2(HzO)2].. (10). ....................................................... 96

(e) ["Bu4N][TCNQ] (l 1) .................................................................. 96

(f) ["Bu4N][TCNQF4] (12) ............................................................... 97

(g) {[ZI‘I(H20)4](TIl-TCNQ)2’2TCNQ'2MCCN}.. (13). .................. 98

(h) Structure of [Ni(DMSO)6][TCNQ]3 (14) ................................. 100

(i) Crystal structure of [Ni(TCNQ)2(HzO)].. (15). ........................ 102

(2) Powder X-ray diffraction studies. ............................................... 103
3. Results and Discussion ........................................................................ 105

A. Synthetic Methods and Powder X—ray Characterization of

Cu(TCNQ). ...................................................................................... 105
(1) Bulk synthesis of phase I. ............................................................ 105
(2) Conversion of phase I to phase II Cu(TCNQ) in acetonitrile. 106

B. Syntheses of the Mn(TCNQ)2(solvent)2 phases. ............................. 107
C. Spectroscopic Studies. ..................................................................... 108
( 1) X-ray photoelectron spectroscopy ............................................... 108
(2) Infrared spectroscopic studies. .................................................... 111
(3) SEM studies of Cu(TCNQ). ........................................................ 112
(4) Powder X-ray studies of Cu(TCNQ) bulk phases ....................... 114
(5) Powder X-ray studies of Cu(TCNQ) films. ................................ 119

xi

(a) At room temperature. ............................................................... 119

(b) At elevated temperatures ......................................................... 119
(6) XRD and SEM studies of M(TCNQ); and M(TCNQ)2L ............ 120
(a) Powder patterns of M(TCNQ)2 . .............................................. 120
D. Single Crystal Diffraction Studies. .................................................. 125
(1) Details of collection and refinement for Cu(TCNQ). ................. 125
(a) Cu(TCNQ) phase I. .................................................................. 126
(b) Cu(TCNQ) phase II. ................................................................ 126
(c) Comparison of the Cu(TCNQ) polymorphs. ........................... 127
((1) Further structural analysis of Cu(TCNQ) and a
comparison to Ag(TCNQ). ..................................................... 130
(2) Crystallographic studies of M(TCNQ)2Lx phase (x=solvent)..... 133
(a) [Mn(TCNQ)(TCNQ-TCNQ)0,5(MeOH)2].._ ............................. 133
(b) [Mn(TCNQ)2(HzO)2]., (10). ..................................................... 136
(0) Structures of ["Bu4N][TCNQ] (11) and ["Bu4N][TCNQF4]
(12) salts .................................................................................. 140
(d) { Zn(H20)4(TCNQ)3°2MeCN L... (13). ....................................... 140
(e) [Ni(DMSO)6][TCNQ]3 (14). .................................................... 143
(f) Ni(TCNQ)2(HzO) (15). ............................................................. 155
(3) TCNQ bonding modes in various TCNQ containing
compounds. ................................................................................. 155
(a) Ring interactions and physical properties. ............................... 155
(b) Hydrogen bonding and weak interactions. .............................. 157
(c) Bond length relationship and the degree of charge transfer
in the TCNQ ligand. ............................................................... 157
E. Magnetic Measurements. ................................................................. 157
(1) Cu(TCNQ) phases. ...................................................................... 157

xii

(2) Mn
F. Charge
(1) Cha
G. Conclu

References ..
CREPIER IV
SlRUCIURA
AND TETRn
METAL HAL

l. Introducti
2. Experime
A. Synthe:
(1) Prep
(3) Bu

(b) Sit

(2) Prep1
(a) Bu

 

 

(2) Mn samples .................................................................................. 163

F. Charge-Transport Properties TCNQ. ............................................... 166

(1) Charge-transport properties of Cu(TCNQ). ................................ 166

G. Concluding Remarks. ...................................................................... 169

References ................................................................................................ 171
CHAPTER IV

STRUCTURAL AND MAGNETIC STUDIES OF DINUCLEAR
AND TETRANUCLEAR CLUSTERS OF 3-D TRANSITION

METAL HALIDES ..................................................................................... 178
1. Introduction .......................................................................................... 179
2. Experimental Section. .......................................................................... 179

A. Synthesis. ......................................................................................... 179
(1) Preparation of [Mn5C110(THF)3].. (16). ....................................... 180
(a) Bulk preparation ....................................................................... 180
(b) Single crystal growth ............................................................... 180
(2) Preparation of [MnBr2(THF)2]., (17) ........................................... 180
(a) Bulk preparation ....................................................................... 180
(b) Single crystal growth ............................................................... 180
(3) Preparation of crystals of [MnC12(MeOH)2]...(18) ...................... 181
(4) Preparation of [Fe2C14(bpy)2] from FC4C13(TI'IF)4 (19). .............. 181
(5) Preparation of [FeC12(bpy)]., from Fe4C13(TI-IF)4 (20). ............... 181
(6) Preparation of [ppn]2[CozCl6] (21). ............................................. 182
(7) Preparation of [ppn]2[Mn2C16] (22). ............................................ 182
(8) Preparation of [ppn]2[NiCl4] (23, 24) ......................................... 183
(a) Bulk Preparation. ..................................................................... 183
(b) Single crystals of phase I (23). ................................................ 183
(c) Composite crystals of a triclinic phase (24) and
orthorhombic phase (23). ........................................................ 183

xiii

(9) Preparation of [Mn2(bpym)4(u-F)(tt‘-BF4)(BF4)2(MeCN)

02MeCN]... (25). .......................................................................... 183

(10) [Et4N]Cl°[Fe2C14(MeOH)4(ll-2,2'-bpym)] (26). ........................ 184

(a) Bulk preparation. ...................................................................... 184

(b) Single crystal growth. .............................................................. 184

B. Structural Studies. ............................................................................ 184

(1) Structure of [Mn5C110(TI-IF)8],, (16). ........................................... 184

(2) Structure of [MnBr2(THF)2].., (17). ............................................. 185

(3) [MnC12(MeOH)2].., (18) ............................................................... 186

(4) FezCl4(bpy)2 (19). ........................................................................ 187

(5) [Fe2C14(bpy)].. (20). ..................................................................... 187

(6) Structure of [ppn]2[CozCl6] (21) .................................................. 190

(7) [ppn]2[Mn2C16] (22). .................................................................... 192

(8) Structure of [ppn]2NiCl4 orthorhombic phase (23). .................... 193
(9) Composite phases of [ppn]2NiCl4 with orthorhombic and

triclinic forms (23b, 24). ............................................................. 194

(10) [Et4N]Cl'[Fe2C14(MeOH)4(u-2,2'-bpym)] (26) ......................... 197

3. Discussion ............................................................................................ 199

A. X-ray Structural Results and Discussion. ........................................ 199

(1) Structure of [Mn5C110(THF)3].. (16). ........................................... 199

(2) Structure of [MnBr2(THF)2]., (17) ............................................... 200

(3) Structure of Fe2C14(bpy)2 (l9). .................................................... 206

(4) Structures of [ppn]2[CozC16] (21) and [ppn]2[Mn2C16] (22) ........ 209

(a) [ppn]2[C02C16] (21). ................................................................. 214

(b) [ppn]2[Mn2C16 (22) ................................................................... 214

(c) [Et4N]Cl°[Fe2C14(MeOH)4(p.-2,2'-bpym)] (26) ........................ 215

B. Magnetic Studies .............................................................................. 219

xiv

 

(1) Magnetic properties of [ppn]2[CozC16] and [ppn]2[Mn2C16]. ..... 219

(2) Magnetic fitting of Mn(H) chains of (17). .................................. 224
(3) [Mn5C110(THF)3]..chain of (16) ................................................... 224
(4) Magnetism of the Mn cluster (16) ............................................... 224
4. Conclusion ........................................................................................... 232
References ................................................................................................ 233
CHAPTER V
THE USE OF THE CCD AREA DETECTOR X-RAY
INSTRUMENT TO RESOLVE TWIN NING PROBLEMS ...................... 236
l . Introduction .......................................................................................... 237
A. Background ...................................................................................... 238
(1) A brief history of diffractometers ................................................ 238
(2) Modern area detectors and CCD techniques ............................... 240
B . The Advantages and Disadvantages of Area Detectors ................... 241
C. New Problems in the CCD Era ........................................................ 241
(1) The problem of twinning ............................................................. 241
(a) Definition of twinning .............................................................. 241
D. Manipulation of Data for Twinned Crystals - Why not Grow a
Better Crystal? ................................................................................. 243
(1) Indexing of twinned crystals: ...................................................... 244
(a) Simple algorithm method. ........................................................ 244
(b) Difference vector and related methods. ................................... 245
E. Data Integration and Deconvolution. ............................................... 246
(l) A rotational twin .......................................................................... 246
(2) A nearly perfect twin. .................................................................. 247
F. Structure solution and refinement. ................................................... 247
2- Mastering the CCD Technique for Special Needs ............................... 247

A. Problematic Structures: Seven Case Studies. .................................. 248
(1) Example 1: Large cell, “small molecule” data set: of

{[M02(02CCMe3)3(NC)2CC(CN)CONH] '8CH2C12 }... (3) ....... 248
(2) Example 2: A Pseudo-Merohedral twin: [Mn2(u4-o-TCNQ-
TCNQ)2(u-cis-TCNQ)2(MeOH)2]... (9) ...................................... 249
(3) Example 3: A rotational twin:
[Et4N]Cl-Fe2C14(MeOH)4(bpym)3 (26). ...................................... 250
(a) A data set from a scintillation detector compared to a data
set from CCD detector. ........................................................... 250
(b) Using a rotational twinning matrix in the refinement ............. 251
(4) Examples 4a and 4b: Nearly perfect twins: ................................. 254
(a) Mn2(bpym)2(p.-F)(nl-BF4)(BF4)2(MeCN)°2MeCN (25). ........ 254
(b) Structure of Mn(n'-TCNQ)2(TCNQ)(HOCH2CH20H)3
(28). ......................................................................................... 255
(i) Big box integration. ............................................................... 255
(ii) Small box integration ............................................................ 255
(5) Example 5: Crystal movement during data collection:
Structure of Rh2(pynp)2(OzCCH3)2(BF4)2'C7H3 (27). ................ 258
(6) Example 6: A quadruply twinned crystal:CuTCNQ phase II
(8) ................................................................................................ 261
(7) Example 7: A composite of two polymorphic crystals: Two
morphologies of [ppn]2NiCl4°Me2CO, (23, 24). ........................ 264
B. Discussion ........................................................................................ 265
(1) Considerations before data collection. ........................................ 265
(2) Data collection ............................................................................. 266
(3) Data processing. .......................................................................... 266
(4) Structure solution and refinement. .............................................. 266
xvi

_——rw

iConclusit

References .
APPBV’DICEEl I
lPPEhDIX A I
GENERAL Pi-

]. Data colle
dihacton
A. Crystal
B. Data co
C. Data in
D. Beam i

References .
Diffs-DIX B
llOll POSII

 

 

3. Conclusions ......................................................................................... 269

References ............................................................................................... 270
APPENDICES ............................................................................................ 272
APPENDDC A
GENERAL PROCEDURES FOR DATA COLLECTION ....................... 273

1. Data collection procedure for Siemens CCD platform

diffractometer ...................................................................................... 274

A. Crystal mounting and preliminary data collection .......................... 274

B . Data collection ................................................................................. 274

C . Data integration. .............................................................................. 275

D. Beam inhomogeneity, crystal decay and absorption correction. 275

References ............................................................................................... 27 6
APPENDD( B

ATOM POSITIONS TABLES ................................................................... 277

 

Table 1. St
Ci
[3
lableZ. Cr
Table 3. 51
Table 4. Se

Table 5. Gel

Table 6, Se
Table 7, Co

 

Table 1.

Table 2.

Table 3.
Table 4.
Table 5.

Table 6.
Table 7.

Table 8.
Table 9.
Table 10.
Table 11.

Table 12.
Table 13.

_

 

LIST OF TABLES
page
Structural parameters for {[M02(02CCF3)4]2(p.4-TCNQ)-2m-
CsHtolool [M02(02CCF3)4I'("1'C8Hto)2 loo (1) and
[M02(OZCCF3)4(DM-DCNQI)-C6H6].. (2) .................................. 38
Crystal data and structure refinement for
{[Moz(OgCCMe3)3(NC)2CC(CN)CONH]°8CH2C12}... (3) and

{ [M02(02CCMe3)3(NC)2CC(CN)CONH]00.5C6H.,}..., (4) ........... 41
Structural parameters for [Re2C14(dppm)2]2TCNQ-10THF (5)
and [Re2C14(dppm)2]2DMDCNQI'10THF (6). ........................... 44
Selected bond distances (A) and bond angles (°) for
[Moz(OzCCF3)4(TCNQ)],,, (1). ................................................... 52
Geometric parameters for compounds that exhibit metal DM-
DCNQI o-bonding. ..................................................................... 58
Selected bond distances [A] and bond angles (°) for (2) ............ 59

Comparison of semi-empirical calculations for two 2,3,3-
tricyanoacrylamide anion isomers. ............................................. 62

Comparison of refinement results of (3) in different space

groups .......................................................................................... 63
Structural parameters for a M0211. 11 core coordinated to non-
linear nitriles. .............................................................................. 66
Selected bond distances (A) and bond angles (°) for
{[M02(OzCCMe3)3(NC)2C(CN)CCONH)]°8CH2C12}.., (3). ....... 67
(Cont’d) ....................................................................................... 68

Librational disorder refinement of the CN bond length in (5). 72
Summary of key crystallographic parameters for Cu(TCNQ)

xviii

 

h

pi

Table 14. Ct
[l
[.

Table 15. C11

 

a

Table16. Cr

 

 

 

Table 14.

phase I (7) and phase II (8). ....................................................... 93
Crystal data and structure refinement parameters for
[Mn(TCNQ-TCNQ)”(TCNQ)(MeOH)2].. (9) and
[Mn(TCNQ)2(HZO)2]...(10). ........................................................ 95

Table 15. Crystal data and structure refinement for [”Bu4N][TCNQ] (11)

Table 16.

Table 17.

Table 18.
Table 19.

Table 20.
Table 21.
Table 22.
Table 23.
Table 24.

Table 25.

Table 26.

 

and ["BuM[TCNQF4] (12). ....................................................... 99
Crystal data and structure refinement for {[Zn(H20)4] (n1-
TCNQ)2-2TCNQ-2MeCN}.. (13) and [Ni(DMSO)6](TCNQ)3

(14). ........................................................................................... 101
Crystal data and structure refinement for

[Ni(TCNQ)2(H20)]...(15) .......................................................... 104
IR and XPS data for Cu(TCNQ) and Cu(DM-DCNQI) ............ 113

Key bond distances(A) in the TCNQ unit(s) for
[Mn(TCNQ)(TCNQ-TCNQ)0,5(MeOH)2] .. (9) and
[Mn(TCNQ)2(H20)2] .. (10) according to diagram in Figure

41 ............................................................................................... 137
Selected bond lengths [A] and angles [°] for
[Mn(TCNQ)2(H20)2].. (10). ...................................................... 142
Bond lengths [A] and angles [°] for ["Bu4N][TCNQ] (11). ...... 144
Bond lengths [A] and angles [°] for ["BuaN][TCNQF4] (12).... 145
Stacking interactions for TCNQ'1 units in selected structures. . 158
Hydrogen bonds (with esds except for fixed and riding atoms)

for [Mn(TCNQ)2(H20)2]..,(10). ................................................ 159
Hydrogen bonds (with esds except for fixed and riding atoms)

for [Mn(TCNQ)(TCNQ-TCNQ)0,5(MeOH)2]... (9). .................. 159
Hydrogen bonds (with esds except for fixed and riding atoms)

for [Zn(H20)4(TIl-TCNQ)2-TCNQ°2MeCN]... (13). ................. 159

xix

Table 27.
Table 28.
Table 29.
Table 30.
Table 31.
Table 32.

Table 33.
Table 34.

Table 35.
Table 36.

Table 37.

Table 38.
Table 39.

Table 40.

Crystal data and structure refinement for

[Mn5C110(TI-IF)3]..(16) and [MnBr2(THF)2].. (17). ................... 188
Crystal data and structure refinement for [MnC12(MeOH)2]..

(18). ........................................................................................... 189
Crystal data and structure refinement for Fe2C14(bpy)2 (19)

and [FeC12(bpy)]..(20). .............................................................. 191
Crystal data and structure refinement for [ppn]2[C02C16] (21)

and for [ppn]2[Mn2C16] (22) ...................................................... 195
Crystal data and structure refinement for [ppn]2NiCl4-Me2CO
(23) and (24). ............................................................................ 196
Crystal data and structure refinement for
[EtaN]Cl'Fe/zCl4(bpym)(MeOH)4. (26) ..................................... 198
Bond lengths [A] and angles [°] for [FeC12(bpy)]... (20). ......... 210
Selected bond distances (A) and bond angles (°) for
[ppn]2[CozC16], (21). ................................................................. 216
Selected bond distances (A) and bond angles (°) for
[ppn]2[Mn2C16], (22). ................................................................ 221
Selected bond distances (A) and bond angles (°) for [EuN1c1-
[Fe2C14(MeOH)4(|.t-2,2'-bpym)], (26). ...................................... 226

Rotational transformation of

[EtnN]Cl-Fe4Cl4(MeOH)4(bpym)3 (26). from orientation A] to
orientation A2. ........................................................................... 252
Indexing result of rotationally twinned
[EtaN]C1~Fe2Cl4(MeOH)4(bpym)3 (26). ................................... 253
Rotational transformation of Mn2(bpym)2(p.-F)(nl-BF4)(BF4)2
(MeCN)-2MeCN (25) from orientation A1 to orientation A2...256
Rotational transformation of

h,
A

Table 41. R
R

0

Table 42. C
Table43. R
fr
Table-14. Ct
of

Table 45. At
di

[h
Cs

Table 46. All
dis

Di
lablm. Att
dis

{ll
Table“ Att
dis

A {ll
lilting. A10
dis
libltgo [R
~Ato

 

Mn(TCNQ)2(TCNQ)(HOCH2CHzOH)3 (28) from orientation

A] to orientation A2 ................................................................... 257
Table 41. Rotational transformation of

Rh2(OzCMe)2(pynp)2.(BF4)2°C7H8 (27) from orientation A1 to

orientation A2. ........................................................................... 259
Table 42. Comparison of two TWINDX solutions of (27). ...................... 260
Table 43. Rotational transformation of [ppn]2[NiCl4]°Me2CO (23b, 24)

from orientation A1 to orientation A2. ...................................... 267

Table 44. Comparison of indexing results of the two TWINDX solutions
of ppn]2[NiCl4]'Me2CO (23b, 24). ........................................... 268
Table 45. Atomic coordinates, ( x 104) and equivalent isotropic
displacement parameters (Azx 103) for
[M02(02CCF3)4(TCNQ)’2m'C8Hlo)l~"I M02(02CCF3)4'2m-
C3H10)]... (1). .............................................................................. 278
Table 46. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Mar 103) for [M02(02CCF3)4(DM-
DCNQI)°C6H6]... (2). ................................................................. 282
Table 47. Atomic coordinates (x 104) and equivalent isotropic
displacement parameters (Azx 103) for
{ [M02(OzC(CH3)3)3((NC)2C(CN)CCONH)]~6CH2C12 }.... (3).... 283
Table 48. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) for
{[Moz(OzC(CH3)3)3((NC)2C(CN)CCONH)]-0.5C6H(5}... (4). 289
Table 49. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) for
[Re2C14(dppm)2]2(TCNQ) (5). .................................................. 291
Table 50. Atomic coordinates ( x 104) and equivalent isotropic

xxi

displacement parameters (Azx 103) for

[RezCl4(dppm)2]2(DMDCNQI) (6). .......................................

Table 51. Atomic coordinates ( x 104) and equivalent isotropic

...295

displacement parameters (Azx 103) for Cu(TCNQ) phase I (7).299

Table 52. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) for Cu(TCNQ) phase II
(8). .........................................................................................

Table 53. Atomic coordinates (x 104) and equivalent isotropic
displacement parameters (Azx 103) [Mn(TCNQ-

TCNQ)0,5(TCNQ)(MeOH)2]., (9). ........................................

Table 54. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) [Mn(TCNQ)2(HzO)2]..

(10). .......................................................................................

Table 55. Atomic coordinates ( x 104) and equivalent isotropic

displacement parameters (Azx 103) [Bu4N][TCNQ] (11) .....

Table 56. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) [Bu4N][TCNQF4] (12).
Table 57. Atomic coordinates ( x 104) and equivalent isotropic

displacement parameters (Azx 103) for

[Zn(TCNQ)2(H20)4(TCNQ)2(MeCN)4].... (1 3). .....................

Table 58. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) [Ni(DMSO)6][TCNQ]3

(l4). .......................................................................................

Table 59. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) for [Mn5C110(THF)g]..

....301

....302

....304

....305

....307

....310

....312

(16). ........................................................................................... 314

Table 60. Atomic coordinates ( x 104) and equivalent isotropic

xxii

displacement parameters (Azx 103) for [MnBr2(TI-IF)2]., (17). 316
Table 61. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) [Fe2C14(,u-2,2’-bpy)2] (19).317
Table 62. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) for [FeC12(,u-2,2’-bpy)]..
(20). ........................................................................................... 318
Table 63. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) for
[EWCI- [Fe2Cl4(MeOH)4(,u-2,2’ -bpym)] (26). ....................... 3 l9

xxiii

figure 1. O

 

Figure 3.
Figure 4. Ir.-

ligureS. 0L

figure 2. S"
lr‘

 

figure 13. Til

the

 

Figure 1.

Figure 2.
Figure 3.
Figure 4.
Figure 5.
Figure 6.
Figure 7.

Figure 8.

Figure 9.

LIST OF FIGURES
page
Orbital interactions in (NC)5M-CN-M’(NC)5 dinuclear units:
(a) Orthogonality between tzg and eg magnetic orbitals, e.g. M
= Crm, M’ = Ni". (b) Overlap between two tzg orbitals, e.g. M

= Cr'", M’= v". ............................................................................. 5
Simplified MO diagrams of metal-metal bonded units. ............... 7
Interactions between an axial ligand and dimetal units. ............... 8
Interactions between an equatorial ligand and dimetal unit. ........ 9
Organonitrile acceptors that behave as ligands in their
reduced forms. ............................................................................ 11
Preparation of [M02(OzCCF3)4 DM-DCNQI-C6H6]... (2). ............ 30
Metal-metal bonded orbitals diagrams. ....................................... 45
Interactions from an axial ligand to metal-metal bonded units
such as Mozll‘". ............................................................................ 46

Interactions between equatorial ligands and metal-metal

bonded units such as Rezn'm. ...................................................... 48

Figure 10. Documented reactions involving the decomposition of TCNE. .49
Figure 11. Metallocyclic rings of polynitriles with dimetal units and

idealized angles: (a) Moz(02CCF3)4(TCNQ) (1), (b)
Rh2(OzCCF3)4(TCNE). ............................................................... 54

Figure 12. Thermal ellipsoid representation of (1) showing one M02 unit

bonded with the TCNQ ligand and a second M02 unit bonded
by m-xylene ................................................................................. 55

1Figure 13. Thermal ellipsoid representation of (2) showing a portion of

the chain with the disordered CF; group and benzene

xxiv

 

 

iigure 14.

f
i
Figure 15. iv
Figure 16. 8
it
Figure 17. P
ir
Fcure 18. .\'
“Elite 19. Pl.
bi
“sure 20. E
Figure 21. Ti
figure 22. T)-

 

 

molecule omitted ......................................................................... 56
Figure 14. Plotting diagram of ab plane of (2) showing the stacking

feature of the DM-DCNQI ligand with the benzene molecule

in the lattice ................................................................................. 57
Figure 15. Molybdenum TCNE modeled from the partial P3N structure. .64
Figure 16. Simulation of dimolybdenum TCNE reaction product (3)

indicates an overcrowded hyper iii-TCNE bonding mode. ........ 65
Figure 17. Plot of [M02(02CCMe3)3((NC)2C(CN)CCONH)°8CH2C12] (3)
in space group P6cc . .................................................................. 69
Figure 18. Nitrile bonding modes with a M0211. 11 core. ................................ 70
Figure 19. Pluto representation of (4) clearly shows that the CN groups
bonded to the M02 unit are in an unusual bent mode. ................ 71
Figure 20. Electron density map of nitrile region in (5). ............................. 73
Figure 21. Thermal ellipsoid representation of Re2(dppm)2TCNQ (5) ........ 74
Figure 22. Thermal ellipsoid representation of Re2(dppm)2DMDCNQI
(6). ............................................................................................... 75
Figure 23. Organonitrile acceptors that behave as ligands in their
reduced forms. ............................................................................ 83
Figure 24. Crystal growing setup for two Cu(TCN Q) phases. .................... 90
Fig-Ire 25. XPS data in the Cu 2pm and 2pm regions for phase I (a) and
phase II Cu(TCNQ) (b). ............................................................ 109
Figure.26. XPS data in the Cu 2133/2 and 2pm regions for
[CuI(MeCN)4][BF4] (a) and [Cu"(MeCN)4][BF4]2 (b) .............. 1 10

Figure 27 . SEM micrographs for solution-prepared Cu(TCNQ) crystals. 115
Figure 28. SEM micrographs showing the progressive changes from

- Cu(TCNQ) phase I to phase II. ................................................. 116
Figure 29. SEM of Cu(TCNQ) films grown on copper in MeCN for

XXV

 

1hour at different temperatures: (a) at 60°C, (b) at 80°C. ........ 117
Figure 30. SEM micrographs of Cu(TCNQ) grown on copper in CH3CN:

(a) top and (b) side views of films grown for 1h at 40 °C

(phase I) and (c) top and ((1) side views of films grown for 1 h

at 80 °C (phase II). The reference scale on all micrographs is

10 um. ....................................................................................... 118
Figure 31. XRD powder patterns illustrating the conversion of

Cu(TCNQ) phase I to phase II in CH3CN. ............................... 121
Figure 32. XRD pattern for a Cu(TCNQ) sample grown on a Cu

substrate for 1 h at 80°C in a CH3CN solution of TCNQ ........ 122
Figure 33. XRD pattern of M(TCNQ)2, (M = Mn, Fe, Co, Ni). ................ 123

Figure 34. Mn(TCNQ); small particles in nanometer size could be seen
from (a) SEM (111m scale) and (b) TEM (0.1 pm scale)

micrographs. ............................................................................. 124
Figure 35. Pluto representation of Cu(TCNQ) phase I. ............................. 128
Figure 36. Thermol elipsoid representation of CuTCNQ phase II. ........... 129

Figure 37 . Interpenetrating networks of Cu(TCNQ) phase I features
TCNQ stacking between different networks ( solid bond --
network 1, holo bond - network 2) ........................................... 131
Figure 38. Interpenetrating networks of Cu(TCNQ) phase H. Note that
the TCNQ units between different networks are shifted and
not iii—stacked between different networks (solid bond:
network 1, holo bond: network 2). ........................................... 132
Figure 39. Schematic diagrams of Cu(TCNQ) phase 1 and phase II with
an emphasis on the geometry of the Cu-N connection. ............ 134
Figure 40. Experimental XRD powder pattern for (a) Cu(TCNQ) phase I
and (b) Ag(TCNQ) .................................................................... 135

xxvi

 

Figure 41.
iigurtill

figure 43. 1
Figure 44. 1

Figure 45. ll
ligure 46.

Figure 47. l
I
Fort 48. E

hare 49.1

 

“the so.
Fliure 51.

Flitre 52. p

 

figure 55. E

Figure 41. Various bond lengths in T CNQ compounds (9), (10). ............. 138
Figure 42. Thermal ellipsoid representation of [Mn(TCNQ)(TCNQ-

TCNQ)0,5(MeOH)2].. (9). ........................................................... 139
Figure 43. Thermal ellipsoid representation of Mn(TCNQ)2(H20)4 (10). . 141

Figure 44. Thermal ellipsoid representation of ["Bu4N][TCNQ] (11). ...... 146

Figure 45. Packing diagram of ["Bu4N][TCNQ] (11) along the a axis ...... 147
Figure 46. (a)The dimer radical pair of (TCNQ)2]2'packed in a nearly
eclipsed pattern in (11). (b) The two dimer pairs of (12). ........ 148

Figure 47. Thermal ellipsoid representation of ["BuaN][TCNQF4] (12). .. 149

Figure 48. Packing diagram of [BuaN][TCNQF4] (12) viewed down b
axis shows both the stacking TCNQF4 dimer and the loosely

packed dimer. ............................................................................ 150
Figure 49. Thermal ellipsoid representation of {an 1-
TCNQ)2(H20)4(TCNQ)-2MeCN }... (13°2MeCN). ................... 151
Figure 50. Packing diagram of (13) along the [l 0 1] plane emphasizes
the TCNQ stacking pattern. ...................................................... 152
Figure 51. Thermal ellipsoid representation of [Ni(DMSO)6][TCNQ]3
(14), ........................................................................................... 153
Figure 52. Packing diagram of (14) viewed down a axis showing TCNQ
stacking column. ....................................................................... 154

Figure 53. Superimposed thermal ellipsoid representations of
[Ni(TCNQ)2(H20)]..° (15) depicting the disorder modeling of

(15). ........................................................................................... 156
mm 54. Pluto representation of the [(TCNQ)2]2' unit in (9) and the

F hydrogen bonding scheme. ....................................................... 160
lgure 55. Extended hydrogen bonding networks in
[Mn(TCNQ)2(H20)4].. (10) ....................................................... 161
xxvii

figure 56. it
t
ligure 57. \

 

 

 

 

Figure 56. Hydrogen bonding networks in [anl-
TCNQ)2(H20)4(TCNQ) '2MeCN]... (13). ................................. 162
Figure 57. Variable temperature susceptibility data for Cu(TCNQ) (a) x
versus T for phase I (7); (b) x and p...“ versus T for phase II
(8). ............................................................................................. 164
Figure 58. Variable temperature susceptibility data for Cu(TCNQ) (c)
zero-field cooled and field-cooled x versus T for phase II (8);
(d) hysteresis of phase H(8). ..................................................... 165
Figure 59. Magnetic data for Mn TCNQ compounds (9),(10) ................... 167
Figure 60. Plot of conductivity 0 (S cm") versus temperature for bulk
Cu(TCNQ) phase I and phase 11 measured on pressed pellets. 168
Figure 61. Thermal ellipsoid representation of {Mn5C110(TI-IF)3}..(16). 201
Figure 62. Comparison of different types of Mn(II) sites in (16) with the

metal sites in M4C13(THF)5 clusters. ........................................ 202
Figure 63. Packing view of [Mn5C1w(THF)g].. (16) down the b axis. ....... 203
Figure 64. Thermal ellipsoid representation of a portion of the chain
compound [MnBr2(THF)2].. (17), .............................................. 204
“We 65. Thermal ellipsoid representation of mononuclear
[Mn12(THF)3], ........................................................................... 205
Fig‘n'e 66. Thermal ellipsoid representation of [Fe2C14(bpy)2], (19) .......... 207
Figure 67. Packing view of [Fe2C14(bpy)2](19) down the (1 axis. .............. 208
Figul‘e 68. Thermal ellipsoid representation of [FeC12(bpy)].. (20.) .......... 211
Figure 69. Packing diagram of [FeC12(bpy)].. (20) showing the one-
dimensional chain along c axis. ................................................ 212
Figure 70. Structures of [MC12(bidentateL)]2 ............................................. 213
ngul‘e 71. Packing view of [ppn]2Mn2C16 in the ac plane. ........................ 217
Flgul‘e 72. Thermal ellipsoid representation of (21). ................................. 218

xxviii

Figure 73. Thermal ellipsoid representation of (22). ................................. 222
Figure 74. Packing view of [ppn]2Mn2C16 in the ac plane. ....................... 223
Figure 75. Thermal ellipsoid representation of
[EtaN]Cl-[Fe2C14(MeOH)4(u-2,2’-bpym)] (26) illustrates the
hydrogen bond between the iron dimer units and the chloride
ion. ............................................................................................ 225
Figure 76. Thermal dependence of the susceptibility for the
[ppn]2[CozC16] dimer (filled circles) and the best fit to the
experimental data obtained (solid line). Dashed-dotted lines
represent the perpendicular (upper) and parallel (lower)
susceptibility. The theoretical behavior of a fully isotropic
antiferromagnetic Co(II) dimer is shown as dotted line. .......... 227
Figure 77. Thermal dependence of the product xT for [ppn]2[C02C16]
(filled circles) and the best fit to the experimental data
obtained (solid line). The theoretical behavior of a fully
isotropic antiferromagnetic Co(II) dimer is shown as dotted
line ............................................................................................. 228
Figure 78. Thermal dependence of the susceptibility for [ppn][Mn2C16]
(filled circles) and the best fit to the experimental data
obtained (solid line). The theoretical behavior of a fully

isotropic antiferromagnetic Mn(II) dimer is shown as dotted

line ............................................................................................. 229

Figure 79. Polymeric chain of MnCl2 units in (16). ................................... 230
Figure 80. Fitting with the cluster model of magnetic measurement data

of (16). ...................................................................................... 230

Figure 81. Magnetic data fitting of [MnBr2(THF)2].. (17). ........................ 231

gum 82. The Curie-Weiss behavior of [MnC12(MeOH)2]., (18). Solid

xxix

 

Figure 84. i
l

 

 

line is the Curie-Weiss fit, dot is the llx and diamond dot is

the um plot. ............................................................................... 231
Figure 83. The experimental setup used by Laue, Friedrich and Knipping

to measure X-ray diffraction intensities. This same general

experimental setup is still currently used, although the source

of X rays and the detection system are now much more

sophisticated .............................................................................. 239

Figure 84. Using local symmetry to reduce independently refined

parameters. ................................................................................ 250

CHAPTER I

INTRODUCTION

1. History
Conductor‘
The

to remain

    
  
  
   
  
   
  
  

magnetic
coordinatio
area of ino
argutrlents
metals and
solids, incl
lith magn

illolec

1. History of Molecule-Based Materials (Molecular Magnets/Molecular
Conductors)
The exploration of molecule-based materials is a topic that promises
to remain at the forefront of the search for new conducting, optical and
magnetic materials for years to come.1 The use of soluble transition metal
coordination complexes as precursors to materials is a rapidly expanding
area of inorganic chemistry with many potential applications.2 Topological
arguments that take into consideration the geometrical preferences of the
metals and ligands have allowed chemists to design new molecule-based
solids, including porous,3 magnetic,4 conducting,5 and conducting materials
with magnetic centers.6
A. Molecule-based Conductors
The fast report of electrical conductivity in an organic solid appeared
in 1954,7 namely, a perylene-bromine complex, which has a room-
telnperature conductivity of 0.1 S cm". In 1960, the organic acceptor TCNQ
(TCN Q = 7,7’,8,8’-tetracyanoquinodimethane) was synthesized, a discovery
that opened up a whole field of conducting charge-transfer materials.8 These
discoveries were followed in the 1970s with the synthesis of the organic
d0n0r ’I'I'F(TI‘F = tetrathiafulvalene)9 which led to the first organic metal,
nfinitely [TTF-TCNQ]. Its room-temperature conductivity (500 $2" cm")
inclEases with a decrease in temperature to a value of 6000 82" cm“1 at 60 K
where a metal-insulator transition occurs.10 Since then, great interest has
been devoted to this type of material, and a great number. of new organic
donol‘s and acceptors, as well as their charge-transfer salts, have been

synthesized.
In 1979, Jerome et al. discovered a superconducting state in

 

(ll-1181:).
improverr
symmetric
ones such
acceptor-t
discoverer
reported is
B. llolecu

The
When Wic
diethyldith
ferromagm
based mag
Kihn Who
Chaul of M

 

 

3

(TMTSF)2PF6 at Tc =1.2 K under 10 kbar.11 Since then, significant
improvements in Tc have been obtained in other organic salts based on either
symmetrical molecules such as BEDT-TTF (or “ET”)12 or unsymmetrical
ones such as DMET-TTFl3 or MDT-'I'I‘F.l4 Later, another family of
acceptor-based superconductors of the type M(drnit)2, (M = Ni, Pd) were
discovered, and these continue to be investigated.15 The highest Tc to be
reported is 12.8K in K—(ET)2(CuN(CN)2)Cl.'°

B. Molecule-based Magnets

The origin of molecular magnetism can be traced back to 30 years ago
when Wickman et al., reported that the complex FeCl(L-L)2 (L-L = N,N-
diethyldithiocarbamate) with an S = 3/2 ground state orders
ferromagnetically with a Tc of 2.5K.l7 Since then, the study of molecule-
based magnets has emerged as a new field, with notable breakthroughs by
Kahn who reported the first inorganic-based molecular magnet ferrimagnetic
chain of Mn(II)Cu(Il)18 centers and by Miller et al., who discovered the first
organic based molecular magnet, namely [FeCp*2]”[TCNE]" (TCNE =
tetracyanoethylene).l9 The rapid reports in the ten years since these
developments include high Tc ferromagnets based on the Prussian-blue
motif,20 spin-crossover compounds that undergo abrupt transitions near room
temperature”! compounds that superconduct in the presence of localized
magnetic moments,22 and high spin clusters that mimic the properties of a
single-domain magnet.23 Our approach is to use organocyanide ligands in
combination with inorganic open shell cations as to prepare new materials
with interesting properties.

Inorganic chemists are interested in the fact that organocyanides such
as TCNE and TCNQ can also bond to metals in a 6 fashion to the lone-pair

electrons 0
addition to
acceptorst
also been
species. B
bidentate 0
extended 7
molecule.
inorganicr

electron de

ilnteract

 

A. Ol'lhOgt
To u

baht magi

 

 

4

electrons on the nitrogen atoms, forming covalently linked materials?“25 In

addition to metal-radical charge-transfer complexes with organocyanide 1t-
acceptors, other n—acceptor radicals in the nitronyl nitroxide family”3 have
also been used in establishing communication channels between metal
species. Requirements for the organic ligands are (1) that they exhibit
bidentate or multidentate binding properties and (2) that they possess an
extended rt-network to allow for electron delocalization throughout the
molecule. Cyanide groups on organic molecules are able to bond to
inorganic moieties through their nitrogen lone pair electrons and accept it-

electron density from the metal donor complex lying 1t* orbital.
2. Interaction of Magnetic Spins

A. Orthogonality of Spins

To understand the factors involved in molecular magnet design, some
basic magnetic phenomena must be understood. There are five important
types of magnetic behavior observed in solids, namely diamagnetism,
paramagnetism, ferromagnetism, antiferromagnetism, and ferrimagnetisrn.
The last three properties are characterized by cooperative behavior (or bulk
properties). In ferromagnets, the individual spins are oriented in a parallel
fashion. In both antiferromagnets and ferrimagnets, there are at least two
kinds of spins that are coupled antiparallel to each other. The coupling
constant J, describes the isotropic interaction between two spins S] and 82,
and is defined by the spin Hamiltonian H = -2J(Sl-Sz).27a The energy
separation between the singlet and the triplet states is J. In ferromagnetic
coupling, J > 0, while in antiferromagnetic coupling, J < 0. If the orbitals
containing the unpaired electron(s) are orthogonal to each other,then Hund’s

rule keeps the spins parallel and ferromagnetic coupling between the two

 

 

(b)

Figure l. Orbital interactions in (NC)5M-CN-M’(NC)5 dinuclear units: (a)
Orthogonality between :28 and eg magnetic orbitals, e.g. M = Crm, M’ = Ni".
(b) Overlap between two tzg orbitals, e.g. M = Crm, M’= V".

spins occur
alignment
spins in or
required.28

equal magi

B. Orbital

Com
second and
eonfrgurati
scparations
Open-sheD
ciatllples c
are stabli:
HOMO/Li
Spatial orle

3Design
PMllrsor.

 

 

spins occurs. If there is direct overlap between the orbitals, the antiparallel
alignment will be favored. In order to achieve a ferromagnetic interaction,
spins in orthogonal orbitals in the same spatial region (adjacent sites) is
required.28 F errimagnetism is achieved when the interacting spins are not of

equal magnitude.

B. Orbital Interactions of Dimetal Complexes with Polynitrile Ligands.
Compared to the first row transition metal series (3d metals), the
second and third row transition metals are less commonly found in high spin
configurations with a single metal center due to their larger tzg and eg energy
separations. Additionally, the formation of metal-metal bonds,29 stablizes
open-shell metal ions of the heavier elements. Nevertheless, there are some
examples of M-M compounds with two and three unpaired electrons which
are stablized due to degenerated or nearly degenerated HOMO or
HOMO/LUMO orbitals. The frontier orbitals of the dimetal units and their
spatial orientation with respect to the incoming ligand dictates their bonding

and magnetic interactions (Figure. 2-4).

3. Design of Molecule-based Materials from Inorganic/Organic
Precursors.

In order to better understand the basic principles of spin interactions
between covalently linked dimetal moieties and their bridging n-acceptors,
we have directed our research efforts towards the following goals:

a. Synthesis and characterization of molecular materials based on metal-
metal building blocks and polynitrile ligands.

b. Synthesis and characterization of molecular materials based on inorganic
solvated cations coordinated to polynitrile ligands.

c. Correlation of the structures of metal-TCNQ materials with their

 

 

PTOP‘

an”

Figt

little.
l1) Genet

buttons
”liable ;
eXCellem
COl’ilent s
organqc to
BCin .
allbride g
ml" Studio.
mimetic

 

properties.

 

 

 

 

0*

i it we

ll I #_ as

e c i t' r .
ll ll vll llil till

 

 

 

 

 

 

I;
I
ll;

 

 

 

 

 

 

 

 

 

 

—+

 

 

 

 

Ii ll t It '1 '1 ll Ii 1‘
ll it ll ll 6
'1 ll Ii Ii
Mo,“ “, Rezm' m Rezu‘ “, Ruzm‘ "1 Ru,“ "' Ru,“Il

Figure 2. Simplified MO diagrams of metal-metal bonded units.

A. M(TCNQ) and M(TCNQ); Materials.
( 1) General background

A particularly intriguing approach to the design of molecular solids
that exhibit cooperative behavior is to assemble open-shell metal centers and
organic radicals into inorganic/organic "hybrid" structures.”3 1'32 An
excellent illustration of this strategy is the design of highly conducting
covalent solids with o- as well as th-p’n interactions between the metal and
organic constituents, for example the class of compounds Cu(R,R’-DCNQI)2
(DCNQI = N ,N ’-dicyanoquinonediimine, R, R’ = halide, alkyl group or
alkoxide groups) with direct bonding of the nitriles to the Cu centers.28 X-
ray studies of the Cu(DCNQI); compounds reveal that the structures adopt

polymeric motifs consisting of partially reduced DCNQI molecules that

 

 

Fig]

 

5 m it *
EE—
‘ ‘ non-bonding
dxy dxy

non-bonding

 

Figure 3. Interactions between an axial ligand and dimetal units.

 

alt

dxy 8\§\ bonding
x—xg

n *
dXZ-dYZ dedyz &g\

Figure 4. Interactions between an equatorial ligand and dimetal unit.

‘0 6 *r \"\

   

 

 

lonn one-d
mixed-vale
mainly to
delocalizati
mixing is si
creating a "
(2) Polyme
One
organic radi
ions has a
pr0pelties; t
crystallize it
high electric

 

of donors ar
Properties.
In lig
acc’iPtor mcl
molecules (
tttttltidimenE
comiiOIicnt.
heightenedi
anioustiliilt
Ones
and Organic
lilting Sue

 

10

form one-dimensional colurrms connected to each other by tetrahedral
mixed-valence Cu”2+ cations. Although high conductivity is attributed
mainly to DCNQI-DCNQI stacking, it has been suggested that the
delocalization of metal electrons into the conduction band from dtt-ptt
mixing is significant for removing the degeneracy of the pit bands thereby
creating a "multi-Fermi" surface that stabilizes the metallic state.30

(2) Polymeric metal-TCNQ materials

One of the main issues in low-dimensional salts composed of planar
organic radicals is the mode of stacking. The solid-state arrangement of the
ions has a dramatic effect on the resulting magnetic and charge-transport
properties; for example charge-transfer salts such as [TTF][TCNQ] that
crystallize with segregated stacks of donors and acceptors typically exhibit
high electrical conductivity whereas those that assemble as alternating stacks
of donors and acceptors tend to show interesting optical as well as magnetic
properties.

In light of the promise for the use of tunable organic donor and
acceptor molecules in materials applications, researchers are incorporating
molecules of the types depicted in Figure 4 into complex solids with
multidimensional frameworks that involve a paramagnetic inorganic
component. Interest in hybrid organic/inorganic materials has been
heightened by the recent reports of (BEDT-TTF) salts of paramagnetic metal
anions that are highly conducting and even superconducting.33

One synthetic strategy to obtaining materials that contain both metals
and organic acceptors is to combine metal cations with organic acceptor
anions such as TCNQ" (7,7,8,8-tetracyanoquinodimethane), TCNE"
(tetracyanoethylene) and DCNQI" (N,N’-dicyanoquinonediimine) (Figure 5).

In this manner one can take advantage of the tendency for planar organic

radicals to

imparting n
Among the
metallic pol
and Ag witl
TCAB Not.
of metals

dimensional
of this type
research all
however, sh;
mmhm
ions bridged

 

 

 

 

11

radicals to engage in rt-stacking interactions, while at the same time
imparting new properties based on the presence of metal-based electrons.
Among the interesting materials that have emerged from these studies are
metallic polymers of Cu with DCNQI, electrically bistable materials of Cu
and Ag with TCNQ, and a room temperature bulk ferromagnet of V with
TCNE. Notably, all of these materials are binary compounds composed only
of metals and organic acceptors with no co-ligands to reduce the
dimensionality of the extended arrays. Unfortunately, polymeric materials
of this type are notoriously difficult to crystallize, a fact that has hampered
research efforts aimed at magneto-structural correlation. We have found,
however, that with the proper selection of starting materials and solvents it is
possible to prepare crystalline, neutral framework compounds with metal
ions bridged only by ligands derived from TCNQ".

 

C 3
— , N __
NC CN ”3‘3 NC CN
CH3

TCNE DM-DCNQI TCNQ
(a) (b) (6)

 

 

 

Figure 5. Organonitrile acceptors that behave as ligands in their reduced
forms.

As part of a comprehensive investigation of the magnetic and
conducting properties of M/TCN Q materials we have prepared several
crystalline phases of general formulae M(TCNQ) and M(TCNQ); in an
effort to establish structure/property relationships. During the course of

these studie
water, mm
of metallf
behavior of
Cu(TCNQ)t
crystalline 5
polymorphs
powder X-
Spectroscopy
Spectrosc0py
involving M
Presented.

dill. inflate
but of the
Mid is
pa”magneti

 

  
 
 
 
 
 
  

B-blelalqtl

12

these studies, we became aware of the role of the solvent (e.g., methanol,
water, acetonitrile) in determining the structures, and, therefore, properties
of metal-TCNQ materials. In addition, new insight into the electrical
behavior of Cu(TCNQ) was gained by our discovery that two polymorphs of
Cu(TCNQ) could be accessed by solution routes and readily isolated as pure
crystalline solids. Specifically we report the discovery of two distinct
polymorphs of Cu(TCNQ) and characterization of these two phases by
powder X-ray diffraction, single crystal crystallography, infrared
spectroscopy, conductivity, magnetic susceptibility, and X-ray photoelectron
spectroscopy. The syntheses and characterization of products from reactions
involving Mn(H), Fe(II), Co(II), Ni(II) and Zn(II) ions with and TCNQ'l are
presented. Representative single-crystal X-ray structures, powder X-ray
data, infrared spectral pr0perties and magnetic properties are discussed in
light of the TCNQ packing motif that is adopted in the solid-state. Also
detailed is the solid-state conversion of Mn(TCNQ)2(MeOH)2 from a
paramagnetic crystalline phase to an ordered phase that displays

ferromagnetic behavior.

B. Metal-Metal Bonded Precursors in Molecule-Based Materials.
Clearly, the combination of mononuclear metal complexes and 1t-
organic radicals gives rise to interesting behavior,34 but dinuclear metal
building blocks with open-shell electron configurations based on the
population/depopulation of low lying 1t' and 5/5' orbitals also present
intriguing possibilities for the elaboration of extended arrays?”35 Recently,
there has been a trend towards building larger molecules with building
blocks where interaction between the metal-metal moieties leads to tunable

physical properties.36’37 For example, it has been noted that rhodium(II)

trilluoroace
metal-metal
structures (3
acceptors h:
pitalate di
trilluoroacet
and corwor';
teuapropion
coworkers.‘

Atterr

moieties linl

 

 

13

trifluoroacetate is efficient for propogating spin interactions through its
metal-metal bonded framework.38 A few polymeric and oligomeric crystal
structures of dinuclear metal moieties covalently linked by conjugated 1t-
acceptors have recently been synthesized. An infinite chain of rhodium (II)
pivalate dimers linked by 1,4-benzoquinone and molybdenum (II)
trifluoroacetate units joined by 9,10-anthraquinone were reported by Handa
and coworkers.”40 In addition, an infinite chain consisting of ruthenium
tetrapropionate bridged by phenazine was recently reported by Cotton and
co-workers.41

Attempts to model macromolecular systems of metal dinuclear

moieties linked by n—electron acceptors with oligomeric sub-units have also
been investigated. Eexamples of tetrameric structures discovered in our
laboratories are [Re2C14(dppm)2]2-u-TCNQ, and its analogous DMDCNQI
complexes, synthesized by Stuart Bartley several years ago.“43 These are
the first dinuclear metal complex bridged by an organocyanide n—acceptor.
Finally, a tetrakis(u-6-chloro-2-hydroxypyridinato)diruthenium (II,III)
complex linked by pyrazine, was reported by Cotton and coworkers.“

The wide availability of redox active metal-metal bonded dinuclear
compounds that can be tailored electronically and sterically allows one to
design and modify hybrid macromolecules that contain both inorganic and
Organic moieties. Dinuclear metal carboxylates in alternating frameworks
With organocyanides are targeted in this project for the synthesis of new
materials. In this thesis, syntheses and structural studies of compounds with
IVIoz4+ dimetalcarboxylates coordinated to the polynitriles TCNQ, 2,5-DM-
DCNQI and chemically transformed TCNE are presented. Structural studies
0f Re2C14(dppm)2 with TCNQ and DM-DCNQI are also discussed.

CSoluble

High
their magn
With respet
polynuclear
femc systel
substantial i
interesting}:
magnetic do

A nur
between anl
itinethoxyp
311d fomtati
This serendi
Of Other sta
lower insole
mired in [)1
tin halide
aniii In ti
With 2~2'-bi
mention in

 

mimetic n
chmcten’w
: Mil. and (I
and C0) and

FeAClgflHF
0f Feclzf 2

- t

 

14

C. Soluble Transition Metal Coordination Complexes as Precursors.

High spin polynuclear transition metal compounds are of interest for
their magnetic properties in biological as well as materials chemistry.45
With respect to iron compounds, the magnetic behavior of dinuclear and
polynuclear ferrous systems has not been well investigated compared to
ferric systems. This is due, in part, to the large zero-field splittings and
substantial anisotropies of the magnetic hyperfine interactions, which lend
interesting properties to the compounds but which render it difficult to fit the
magnetic data due to the large number of parameters involved.

A number of years ago, Dunbar and Quillevéré reported a reaction
between anhydrous ferric chloride and the highly basic phosphine tris(2,4,6-
trimethoxyphenyl)phosphine (TMPP) that led to the reduction of FeIn to Fe"
and formation of [Fe2C16]2', a hitherto unknown binary ferrous chloride.46
This serendipitous discovery led us to investigate whether the [Fe2C16]2' unit
or other stable polynuclear species would be readily accessible from the
lower insoluble halides of the type MClz (M = Mn, Fe, Co, Ni). Our main
interest in preparing these salts was to access discrete, soluble forms of high
Spin halide complexes for use as building blocks in larger molecules or
arrays. In this vein, we have investigated reactions of several [F62C16]2- salts
with 2,2'-bipyrimidine, a bis-chelating ligand that has received much
attention in the literature due to its ability to transmit electronic and
magnetic information between metal centers. The syntheses, structural
characterization, and magnetic studies of the dinuclear anions [M2C16]2' (M
= Mn, and Co), polynuclear metal halide clusters of M4C13(THF)6 (M = Fe
and Co) and Mn5C1w(TI-IF)3, and two bipyridine reaction products with the
Fe4C13(THF)5 cluster, Fe2C14(2,2'-bpy)2 (bpy = bipyridine) and the structure
0f FeC12(2,2’-bpy). and the bpym (bpym = bipyrimidine) complex,

 

sordid

4. Handling
Cryst
and biology
to the tech
improved s
Coupled De
has been ca
new stmctu
With the no
is entries
total entries
In Spl
SOlution am
imrOduction
with a CCD
in! AS mt
other PTOble
Processing.
dilicollecm
and Other prt

 

15

[EtaN]Cl-[Fe2C14(MeOH)4(pt-2,2'-bpym)], are presented.

4. Handling Crystallographic Problems with a CCD Detector.

Crystallography has become an essential component of both chemistry
and biology. New structures are emerging at an ever-increasing rate owing
to the technical advances in radiation sources, detectors, computing and
improved structural methods. The introduction of the CCD (Charge
Coupled Device) detector in the last few years has added to the trend of what
has been called a “structural explosion?“ On average, every 40 minutes, a
new structure is deposited in the CSD (Cambridge Structure Data Base)48
with the number published by CSD on their website.49'50 The number of
new entries added to CSD reached 14,799 in 1997 and 15,645 in 1998. The
total entries in the database reached 195,798 by the end of 1998.

In spite of the obvious advantages, problems such as data collection,
solution and refinement for twinned crystals have also emerged with the
introduction of CCD area detector. A typical crystallography lab equipped
with a CCD area detector may collect three to four hundred data sets per
year. As many as 10% of these may be problematic, exhibiting twinning or
other problems,51 which require special handling during data collection and
processing. In this thesis, strategies for handling problematic crystal data,

data collection and processing, as well as solution and refinement of twinned

and other problematic crystals will be presented and discussed.

16

References

(a) Molecular Magnetism: From Molecular Assembles to the Devices
Coronado, E.; Delhaes, P.; Gatteschi, D.; Miller, J .S.Eds.; Kluwer
Academic Publishers: Dordrecht, Netherlands 1996. (b) Molecular
Engineering for Advanced Materials Becher, J .; Schaumburg, K. Eds;
Kluwer Academic Publishers: Dordrecht, Netherlands 1995. (c)
Kaszynski, P.; Frieli, A.C.; Michl, J. J. Am. Chem. Soc. 1992, 114,
601. (d) Mallouk, T.E.; Lee, H. J. Chem. Ed.1990, 67(10), 2704.

(a) Chen, C.-T.; Suslick, K. S. Coord. Chem. Rev., 1993, 128, 293-
322; see also references therein. (b) Archer, R. D.; Lauterbach, A.;
Ochaya, V. O. Polyhedron, 1994, 13, 2043-2048.

(a) Gardner, G. B.; Venkataraman, D.; Moore, J. S.; Lee, S. Nature,
1995, 374, 792-795. (b) Venkataraman, D.; Gardner, G. B.; Lee, S.;
Moore, J. S. J. Am. Chem. Soc., 1995, 117, 11600-11601. (c) Yaghi,
O. M.; Li, H. J. Am. Chem. Soc., 1995, 117, 10401.

(a) Manriquez, J. M.; Yee, G. T.; McLean, S.; Epstein, A. J .; Miller, J.
S. Science 1991, 252, 1415-1417. (b) Tamaki, H.; Zhuang, Z. J.;
Matsumoto, N.; Kida, S.; Koikawa, M.; Achiwa, Hashimoto, Y.;
Okawa, H. J. Am. Chem. Soc., 1992, 114, 6974. (c) Stumpf, H. O.;
Pei, Y.; Kalul, O.; Sletten, J.; Renard, J. P. J. Am. Chem. Soc., 1993,
115, 6738. (d) Inoue, K.; Iwamura, H. J. Am. Chem. Soc., 1994, 116,
3173. (e) Ohba, M.; Maruono, N.; Okawa, H.; Enoki, T.; Latour, J.-
M. J. Am. Chem. Soc., 1994, 116, 11566-11567. (f) Kahn, O. in
Molecular Magnetism: From Molecular Assemblies to the Devices:
NATO ASI Series. Eds., Coronado, E.; Delhaes, Gatteschi, D.; Miller,
J. S. Kluwer Academic, Dordrecht, 1996, 321, pp. 243-288. (g)
Decurtins, S.; Schmalle, H. W.; Schneuwly, P.; Zheng, Li-M.;
Ensling, J.; Hauser, A. Inorg. Chem, 1995, 34, 5501. (h) Miyasaka,
H.; Matsumoto, N .; Okawa, H.; Re, N .; Gallo, E.; Floriani, C. Angew.
Chem. Int. Ed. Engl. 1995, 34, 1446-1448. (i) Ohba, M.; Okawa, H.;
Ito, T.; Ohto, A. J. Chem. Soc. Chem. Commun., 1995, 1545-1546. (j)
Michaut, C.; Ouahab, L.; Bergerat, P.; Kahn, 0.; Bousseksou, A. J.
Am. Chem. Soc., 1996, 118, 3610. (R) de Munno, G. ; Poerio, T.;
Viau, G.; Julve, M.; Lloret, F.; Joumaux, Y.; Riviere, E. Chem.

17

Commun., 1996, 2587.

(a) Lacroix, P.; Kahn, 0.; Gliezes, A.; Valade, L.; Cassoux, P. Nouv.
J. de Chimie, 1985, 643-651. (b) Gross, R.; Kaim, W. Angew. Chem.
Int. Ed. Engl. 1987, 26, 251. (c) Bartley, S. L.; Dunbar, K. R. Angew.
Chem. Int. Ed. Eng. 1991, 30, 448. (d) Ballester, L.; Barral, M.;
Gutierrez, A.; Jiménez-Aparicio, R.; Martinez-Muyo, J.; Perpifian,
M.; Monge, M.; Ruiz-Valero, C. J. Chem. Soc. Chem Commun.
1991, 1396-1397. (e) Humphrey, D.G.; Fallon, G.D.; Murray, K.S. J.
Chem. Soc., Chem. Commun. 1988, 1356. (t) Comelissen, J.P.; van
Diemen, J. H.; Groeneveld, L. R.; Haasnoot, J. G.; Spek, A. L.;
Reedijk, J. Inorg. Chem. 1992, 31, 198-202. (g) Oshio, H.; Ino, E.;
Mogi, I,; Ito, T. Inorg. Chem, 1993, 32, 5697-5703. (h) Ballester, L.;
Barral, M.; Gutierrez, A.; Monge, A.; Perpir'ian, M. F.; Ruiz-Valero,
C.; Sénchez-Pélaez, A. Inorg. Chem. 1994, 33, 2142-2146. (i)
Dunbar, K. R.; Ouyang, X. Mol. Crys. Liq. 0323., 1995, 273, 21-28.
0) Oshio, H.; Ino, E.; Ito, T.; Maeda, Y. Bull. Chem. Soc. Jpn., 1995,
68, 889. (k) Dunbar, K. R. Angew Chem, 1996, 35, 1659. (l)
Decurtins, S.; Dunbar, K. R. ; Gomez-Garcia, C. J.; Mallah, T.;
Raptis, R. G.; Talham, D.; Veciana, J. in Molecular Magnetism: From
Molecular Assemblies to the Devices. (Eds.: E. Coronado, P. Delhaes,
D. Gatteschi, J. S. Miller) NATO ASI Series vol. E321 Kluwer, 1996,
571-582. (m) Dunbar, K. R.; Ouyang, X. Chem. Commun., 1996,
2427. (n) Zhao, H.; Heintz, R. A.; Rogers, R. D.; Dunbar, K. R. J.
Am. Chem. Soc., 1996, 118, 12844. (0) Dunbar, K. R.; Ouyang, X.
Inorg. Chem, 1996, 35, 7188. (p) Azcondo, M. T.; Ballester, L.;
Gutierrez, A.; Perpifian, F.; Amador, U.; Ruiz-Valero, C.; Bellitto, C.
J. Chem. Soc., Dalton Trans. 1996, 3015.

(a) Aumiiller, A.; Erk, P.; Klebe, G.; Hfinig, S.; von Schlitz, J.;
Werner, H. Angew. Chem. Int. Ed. Engl. 1986, 25, 740-741. (b)
Aumttller, A.; Erk, P.; Httnig, S. Mol. Cryst. Liq. Cryst. Inc. Nonlin.
Opt. 1988, 156, 215-221. (c) Erk, P.; Gross, H.-J.; Hilnig, U. L.;
Meixner, H.; Werner, H.-P.; von Schlitz, J. U.; Wolf, H. C. Angew.
Chem Int. Ed. Engl. 1989, 28, 1245-1246. (d) Kato, R.; Kobayashi,
H.; Kobayashi, A. J. Am. Chem Soc., 1989, 111, 5224-5232. (e)
Aumilller, A.; Erk, P.; Hfinig, S.; Hadicke, E.; Peters, K.; von
Schnering, H. G. Chem Ber. 1991, 124, 2001.(f) Sinzger, K.; Hiinig,

10.

11.

12.

13.

14.

15.

16.

18

S.; Jopp, M.; Bauer, D.; Bietsch, W.; von Schlitz, J. U.; Wolf, H. C;
Kremer, R. K.; Metzenthin, T.; Bau, R.; Khan, S. 1.; Lindbaum, A.;
Lengauer, C. L.; Tillmanns, E. J. Am. Chem Soc. 1993, 115, 7696.
(g) Kato, R.; Kobayashi, H.; Kobayashi, A. J. Am. Chem. Soc. 1989,
111, 5224.

(a) Akamatu, H.; Inokuchi, H.; Matsunaga, Y. Nature (London) 1954,
173, 168. (b) Akamatu, H.; Inokuchi, H.; Matsunaga, Y. Bull. Chem
Soc. Jpn. 1956, 29, 213.

Melby, L. R.; Harder, F. J.; Hertler, W. F.; Mahler, W.; Benson, F. E.;
Mochel, W. E. J. Am Chem. Soc. 1962, 84, 3374.

Wudl, F.; Smith, G. M.; Hufnagel, E. J. J. Chem Soc., Chem.
Commun. 1970, 1453.

(a) Ferraris, J.; Cowan, D. 0.; Walatka, V.; Perlstein, J. H. J. Am
Chem Soc. 1973, 95, 498. (b) Coleman, L. B.; Cohen, M. J.;
Sandman, D. J.; Yanagishi, F. G.; Garito, A. F.; Heeger, A. J. Solid
State Commun. 1973, 12, 1125.

(a) Jerome, D.; Mazaud, A.; Ribault, M.; Bechgaard, K. J. Phys. Lett.
1980, 41, L95. (b) Bechgaard, K.; Jacobsen, S.; Mortensen, K;
Pedersen, H. J .; Thorup, N. Solid State Commun. 1980, 33, 1119.
Mizuno, M.; Cava, M. P. J. Org. Chem. 1978, 43, 416.

Kikuchi, K.; Kikuchi, M.; Namiki, T.; Saito, K.; Ikemoto, 1.; Murata,
K.; Ishiguro, T.; Kobayashi, K. Chem. Lett. 1987, 931.

Papavassillou, G. C.; Mousdis, G. A.; Zamboni, J. S.; Terzis, A.;
Hountas, A.; Hilti, B.; Mayer, C. W.; Pfeiffer, J. Synth. Met. 1988, 27,
B379.

Bossard, L.; Ribault, M.; Brousseau, M.; Valade, L.; Cassoux, P. C.
R. Acad. Sci. Paris 1986, Ser. 2, 302, 205.

Williams, J. M.; Kini, A. M.; Wang, H. H.; Carlson, K. D.; Geiser, U.;

 

(*‘J
r9

Mort
Bory
Over
3272

(a) V
and l

 

 

17.

18.

19.

20.

21.

22.

19

Montgomery, L. K.; Pyrka, G. J.; Watkins, D. M.; Kommers, J. M.;
Boryschuk, S. J.; Crouch, A. V. S.; Kwok, W. K.; Schirber, J. E.;
Overmyer, D. L.; Jung, D.; Whangbo, M. H. Inorg. Chem. 1990, 29,
3272.

(a) Wickman, H. H.; Trozzolo, A. M.; Williams, H. J.; Hull, G. W.;
and Merritt, F. R. The Physical Review, 1967, I55, 563. (b) DeFotis,
G. C.; Palacio, F.; O’Connor, C. J.; Bhatia, S. N.; Carling, R. L. J.
Am Chem. Soc., 1977, 99, 8314.

Pei, Y.; Verdaguer, M; Sletten, J; Renard, J. P. J. Am. Chem. Soc.
1986, 108, 7428. ‘

Miller, J. S.; Calabresse, J. C.; Epstein, A. J.; Bigelow, W.; Zhang, J.
H.; Reiff, W. M. J. Chem. Soc., Chem. Commun. 1986, 1026.

(a) Keggin, J. F.; Miles, F. D. Nature 1936, 137, 577. (b) Mayoh, B.;
Day, P. J. C. S. Dalton 1976, 1483. (b) Gadet, V.; Mallah, T.; Castro,
1.; Berdaguer, M. J. Am. Chem. Soc. 1992, 114, 9213. (c) Verdaguer,
M. Science, 1996, 272, 698, and references therein. ((1) Girolami, G.
S.; Entley, W. R.; Treadway, C. R.; Holmes, 8. M. VIth Int’l Confi 0n
Molecule-Based Magnets, Seignosse, France, Sept. 12-17, 1998.
(e)Bushmann, W. E.; Paulson, S. C.; Wynn, C. M.; Girtu, M. A.;
Epstein, A. J.; White, H. S.; Miller, J. S. Chem. Mat, 1998, 10, 1386.
(t) Ferlay, S.; Mallah, T.; Ouahes, R.; Veillet, P,; Verdaguer, M.
Inorg. Chem, 1999, 38, 229.

(a) Miller, J. S.; Epstein, A. J. Chem. Commun., 1998, 1319. (b)
Zhang, J.; Liable-Sands, L. M.; Rheingold, A. L.; Del Sesto, R. E.;
Gordon, D. C.; Burkhart, B. M.; Miller, J. S. Chem Commun., 1998,
1385. (c)Zhang, J .; Ensling, J .; Ksenofontov, V.; Giltlich, P.; Epstein,
A. J .; Miller, J. S. Angew. Chem. Int. Ed., 1998, 37, 637.

(a) Kurmoo, M.; Graham, A. W.; Day, P.; Coles, S. J .; Hursthouse, M.
B.; Caulfield, J. L.; Singleton, J .; Pratt, S. J .; Hayes, W.; Ducasse, L.;
Guionneau, P. J. Am. Chem. Soc. 1995, 117, 12209. (b) Martin, L.;
Turner, S. 8.; Day, P. Chem. Commun. 1997, 15, 1367. (0) Day, P.
Physica Scripta, 1993, T49, 726.

24.

23.

24.

25.

26.

27.

28.

29.

30.

31.

20

(a) Caneschi, A.; Gatteschi, D.; Sessoli, R. J. Am. Chem. Soc., 1991,
113, 5873. (b) Aromi, G.; Aubin, S. M. J.; Bolcar, M. A.; Christou,
G.; Eppley, H. J.; Folting, K.; Hendrickson, D. N.; Huffman, J. C.;
Squire, R. C.; Tsai, H. L.; Wang, S.; Wemple, M. W. Polyhedron,
1998, 17, 3005. (c) Barra, A. L.; Brunel, L. C.; Gatteschi, D.; Pardi,
L.; Sessoli, R. Acct. Chem. Res., 1998, 31, 460.

Summerville, D. A.; Cape, T. W.; Johnson, E. D.; and Basolo, F.
Inorg. Chem 1978, I 7, 3297.

Miller, J. S.; Calabrese, J. C.; Mclean, R. S; Epstein, A. J. Adv. Mat.
1992, 4, 498

(a) Lacroix, P.; Kahn, 0.; Gleizes, A.; Valade, L.; Cassoux, P. New J.
Chem. 1984, 8, 643. (b) Rey Paul Acc. Chem. Res. 1989, 22, 392.

There are different conventions being used in the description of
magnetic interactions. See, for example, (a) Carlin R. L.
Magnetochemistry; Springer-Verlag: Berlin Heidelberg, Chapter 5,
1986. (b) Mattis D. C. The Theory of Magnetism; Springer-Verlag:
New York, 1981; Vol. I. The convention used in this thesis follows
the convention used in reference (a).

(a) Kahn, 0. Structure and Bonding, 1987, 68, 89. (b) Kahn, O.
Angew. Chem, Int. Ed. 1985, 24, 834.

Miller, J. S.; Esptein, A. J. Prog. Inorg. Chem. 1976 V0120 p1.

Manriquez, J. M.; Yee, G. T.; Mclean, S.; Epstein, A. J .: Miller, J. S.
Science 1991, 252, 1415.

(a) Aumiiller, A.; Erk, P.; Klebe, G.; Hiinig, S.; von Schlitz, J. U. and
Werner, H.-P. Angew. Chem. Int. Ed. Engl. 1986, 25, 740. (b)
Aumtiller, A.; Erk P. and Hlinig, S. Mol. Cryst. Liq. Cryst. Inc.
Nonlin. Opt. 1988, 156, 215. (c) Erk, P.; Gross, H.-J.; Httnig,;
Meixner, H.; Werner, H.-P.; von Schlitz, J. U. and Wolf, H. C. Angew.
Chem Int. Ed. Engl. 1989, 28, 1245. (d) Kato, R.; Kobayashi, H.;

S ynt

(a)
(b)
199
199

Vale!

' (M

(b) }
Kmn

32.

33.

34.

35.

36.

37.

21

Kobayashi, A. J. Am. Chem. Soc. 1989, I 11 , 5224.

(a) Bartley, S. L.; Dunbar, K. R. Angew. Chem. Int. Ed. Engl. 1991,
103, 447. (b) Cayton, R. H.; Chisholm, M. H.; Darrington, F. D.
Angew. Chem. Int. Ed Engl., 1990, 29, 1481-1483. (c) Cayton, R. H.;
Chisholm, M. H.; Putilina, E. F.; Folting, K. Polyhedron, 1993, 12,
2627. (d) Stoner, T. C.; Dallinger, R. R; Hopkins, J. Am. Chem. Soc.,
1990, 112, 5651. (e) Cayton, R. H.; Chisholm, M. H.; J. Huffman, C.;
Lobkovsky, E. B. J. Am. Chem. Soc., 1991, 113, 8709. (f) Huckett, S.
C.; Arrington, C. A.; Burns, C. J.; Clark, D. L.; Swanson, B. I. Synt.
Met. 1991, 41-43, 2769. (g) Li, D.; Huckett, S. C.; Frankcom, T.;
Paffett, M. T.; Farr, J. D.; Hawley, M. E.; Gottesfeld, S.; Thompson,
J. D.; Burns, C. J.; Swanson, B. I. in Supramolecular Architecture:
Synthetic Control in Thin Films and Solids ACS Symp. Ser. 499, (Ed:
Bein, T.) American Chemical Society, 1992, Ch. 4, 33. (h) Stoner, T.
C.; Geib, S. J.; Hopkins, M. D. J. Am. Chem. Soc., 1992, 114, 4201.
(i) Dunbar, K. R.; J. Cluster Science 1994, 5, 125. (i) Dunbar, K. R.;
Ouyang, X. Mol. Cryst. Liq. Cryst. 1995, 273, 21.

Triki, S.; Ouahab, L.; Grandjean, D.; Arniell, J .; Garrigou-Lagrange,
C.; Delahes, P.; Fabre, J. M. Synth. Met. 1991, 41 -43, 2589.

(a) Gross, R.; Kaim, W. Angew. Chem. Int. Ed. Engl., 1987, 26, 251.
(b) Gross-Lannert, R.; Kaim, W.; Olbrich-Deussner, B. Inorg. Chem,
1990, 29, 5046. (c) Kaim, W.; Moscherosch, M. Coord. Chem. Rev.,
1994, 129, 157. (d) Ballester, L.; Barral, M.; Gutierrez, A.; Jimenez-
Aparicio, R.; Martinez-Muyo, J.; Perpifian, M.; Monge, M.; Ruiz-
Valero, C. J. Chem. Soc. Chem. Commun., 1991, 1396.

(a) Cotton, F. A. and Norman, J. G. Jr. J. Coord. Chem, 1971, I, 161.
(b) Handa, M.; Yamada, K.; Nakao, T.; Kasuga, K.; Mikuriya, M. and
Kotera, T. Chem. Lett., 1993, 1969. (c) Dunbar, K. R. and Ouyang, X.
Chem Commun., 1996, 2427.

Cayton, R. H.; Chisholm, M. H.; Huffman, J. C.; and Lobkovsky, E.
B. J. Am Chem. Soc. 1991, 113, 8709.

Cayton, R. H.; Chisholm, M. H.; Putilina, E. F.; Flting, K.; Huffman,

 

41.

42.

 

 

38.

39.

40.

41.

42.

43.

45.

46.

22

J. C.; and Moodley, K. G. Inorg. Chem. 1992, 31 , 2928.

Cogne, A.; Grand, A.; Rey, P.; and Subra, R. J. Am. Chem. Soc. 1989,
111, 3230.

Handa, M.; Takata, A; Nkao, T.; Kasuga, K.; Mikuriya, M.; and
Kotera,T. Chem. Letter. 1992, 2085.

Handa, M.; Sono, H.; Kasamatsu, K.; Kasuga, K.; Mikuriya, M.;
Ikenoue, S. Chem. Lett. 1992, 453.

Cotton, F. A.; Kim, Y.; Ren, T. Inorg. Chem. 1992, 31, 2723.

Bartley, S. L.; Dunbar, K. R. Angew. Chem. Int. Ed. Engl. 1991, 30,
448.

Bartley, S. L.; Dunbar, K. R. unpublished results.
Cotton, F. A.; Kim, Y.; Ren, T., Inorg. Chem. 1992, 31, 2608.

(a) Holm, R. H. Acc. Chem. Res. 1977, 10, 427. (b) Holm, R. H.
Chem. Soc. Rev. 1981, 10, 455. (c) Papaefthymiou,V.; Girerd, J .J.;
Moura, I.; Moura, J .J .G.; Mijnck, E. J. Am. Chem. Soc. 1987, 109,
4703. (d) Taft, K. L.; Lippard, S. J. J. Am Chem. Soc. 1990, 112,
9629.(e) Menage, S.; Vincent, J. M.; Lambeaux, C.; Chottard, G.;
Grand, A.; Fontecave, M. Inorg. Chem. 1993, 32, 4766. (f) Taft, K.
L.; Papaefthymiou, G. C.; Lippard, S. J. Science 1993, 259, 1302. (g)
Zang, Y.; Jang, H. G.; Chiou, Y. M.; Hendrich, M. P.; Que, L. Jr.
Inorg. Chim. Acta 1993, 213, 41. (h) Hendrich, M. P.; Day, E. P.;
Wang, C. -P.; Synder, B. S.; Holm, R. H.; Munck, E. Inorg. Chem.
1994, 33, 2848 and references therein. (i) Powell, A. K.; Heath, S. L.
Comments Inorg. Chem. 1994, 15, 255 and references therein. (i)
Goldberg, D. P.; Tesler, J .; Bastos, C. M.; Lippard, S. J. Inorg. Chem,
1995, 34, 3011.

(a) Dunbar, K. R.; Quillevéré, A. Angew. Chem, Int. Ed. Engl. 1992,
31, 1360. (b) Dunbar, K. R.; Quillevéré, A. Polyhedron 1993, 12, 807.
(c) Dunbar, K. R.; Sun, J .-S. Mol. Cryst. Liq. Cryst. 1995, 274, 51.

47.

48.

49.

50.

51.

23

Baker, T. IUCr Newslett. 1997, 5, 1.

(a) CAMBRIDGE DATABASE -Allen, F. H.; Kennard, 0.; Taylor,
R. Acc. Chem. Res. 1983, 16, 655. (b) Allen, F. H. and Kennard, 0.
Chemical Design Automation News, 1993, 8, 31.

“CCDC Annual Report 1997”, Hartley, D and Allen, F. H. May, 1998
http://www.ccdc.cam.ac.uk/news/annrep/AnnRep97.html.

“CCDC Annual Report 1998”, Hartley, D and Allen, F. H. Jun. 23,
1999 http://www.ccdc.cam.ac.ul_<lannrep/report98.html.

Young, V. G. Jr. New Strategies in the Refinement of Rotationally
Twinned Materials, Poster presentation in ACA 1998 Annual
Meeting, St. Louis, MO.

 

CHAPTER II

SYNTHESES AND CHARACTERIZATION OF

EXTENDED ARRAYS WITH METAL-METAL

BONDED COMPLEXES AND POLYNITRILE
ACCEPTOR LIGANDS

24

1. Introdue

The
centers h;
compound
the comml

by simplq

 

 

 

synthesizel
with the N
group on t!
ground sta‘
themagna
bl Cotton
Spins), NC
wasobsen
Ru; Units
diSplays a}
dosed-she
reputed 1
HRuzfithp;~
Willing a

25

1. Introduction

The synthesis and reactivity of compounds containing dinuclear metal
centers has been extensively studied for nearly 30 years.1 Numerous
compounds reported over the years have contributed to the understanding of
the communication pathway between metal-metal bonded M2 units bridged
by simple inorganic atoms or organic ligands. Ru2(02CR)4Cl, first
synthesized by Wilkinson,2 exists as a polymeric zigzag chain compound
with the M-Cl-M angles ranging from 120° to 180°, depending on the R
group on the carboxylate ligand. These R1125+ compounds possess an S = 3/2
ground state based on the (5’215482(1t"‘8"‘)3 electronic configuration. However,
the magnetic study of the Ru2(02C3H7)4Cl compound reported independently
by Cotton3 and by Drago4 shows only paramagnetic behavior (isolated
spins). No interaction between the dimetal units through the chlorine bridge
was observed. A paper by Cotton describes an infinite chain structure where
R112 units are connected by linear cholride bridges. This material actually
displays antiferromagnetic behavior.5 Diruthenium carboxylates bridged by
closed-shell organic conjugated n-species such as phenazine6 were also
reported to exhibit antiferromagnetic coupling, while the compound
{[Ru2(chp)4]2(pyz)}(BF4)2 (where chp = 6-chloro-2-hydroxypy1idinato) with
pyrazine as the 1: bridging unit exhibits no appreciable coupling.7 Studies
involving polynitrile ligands such as DM-DCNQI (2,5-dimethyl-N,N’-
dicyanoquinonediimine) with a series of diruthenium carboxylate complexes
have been reported by Swanson et al.8 When acetate ligands were used, the

magnetic study revealed both antiferromagnetic behavior for T(K) > 150 and
f(".tl‘romagnetic-like behavior for T(K) < 150, while [Ru2(02CR)4]DMDCNQI

(R = Tol) shows only antiferromagnetic coupling. However no structural

idiomatic

Int
coworker
can be oh
diehloroet
order has
021551, wi
reactionb
reported 6
oxidized
(WM
showed m
Organoniu
channel to
11119138611 g
mlfilaction
“’eakened
of M02“
MO3(02CO
”my 1,1

 

Strong C011
axially 0r

characten'z
d“ 10 th:
fittepmrs
SUCCegSfUu

Met;

26

information is available to address the interaction mechanism.

In terms of dimolybdenum redox-active compounds, McCarley and
co-workers have reported9 that M02(02CR)4 (R = C2H5, C(CH3)3, or C6H5)
can be oxidized by 12 in a non-coordinating solvent such as 1,2-
dichloroethane to form 13' salts of the cation [M02(02CR)4]”'. The bond
order has been reduced to B.O. = 3.5 and the electron configuration is
021561, with a single electron residing in the 8 orbital. Evidence of a redox
reaction between TCNE and a quadruply bonded dimolybdenum unit was
reported by Giraudon et al.'°'ll The M02(R2C22H20N4)2 molecule was
oxidized by TCNE to form a diradical species
([M02(R2C22H20N4)2]°+TCNE'-). Research reported by Felthouse et al 12
showed that when M02(02CCF3)4 is used as a bridging unit to connect two
organonitronyl radicals, that the quadruply-bonded unit is not an efficient
channel to propagate spin interactions of the organic radicals TEMPO.
Interestingly, the analogous dirhodium complex exhibits antiferromagnetic
interactions with J = -200 K through its metal-metal bond,13 which should be
weakened by the introduction of axial ligands. There is promise for the use
of M024+ units in this chemistry, however, as Handa has reported that
M02(02CCF3)4 forms an infinite one-dimensional chain with a n-acceptor,
namely 1,4-benzoquinone, as the bridging unit”. Oligomers that exhibit
strong communication between paramagnetic M-M bonded units through
axially or equatorially bound bridging ligands have been structurally
Characterized,” but there is a general lack of X-ray data for polymeric arrays
due to the difficulties in obtaining single crystals. Of the polynitrile
acceptors TCNQ, TCNE and DM-DCNQI,15 only TCNE has been
Sllccessfully incorporated into infinite arrays with a dimetal unit.16

Metal-metal bonded compounds offer a diverse selection of oxidation

state, liga
issues tha ,
the applic

main goal
polymeric
crystallog

compoun

“Ru;(OgC
ferrimagng
Chumu
Structural

   
  
  
  
 
  

contain ml
addilicm, [1
Presented,

2' Eth‘rirr
A' Startlnl

Mo;
Wmmél

 

mfluoloac

pulChaSed
DCNQI w

u- t
lsmjmflhj

WEWWu.

27

state, ligand and metal ion choices for molecule-based materials. The main
issues that must be addressed are the isolation of pure crystalline phases and
the application of various methods to characterize polymeric products. The
main goal of the studies in this chapter is to synthesize pure oligomeric or
polymeric compounds whose structures could be determined by X-ray
crystallography. In this way we hope to model the structures of other related
compound of interests for their magnetic properties, for example
“Ru2(02CR)4(DM-DCNQI)”’7 which has been reported to form
ferrimagnetic chains with DM-DCNQI, e.g., Ruz‘p"5+ (S = 1 or 2), but whose
characterization by X-ray techniques has eluded researchers. In this chapter,
structural studies of M02(02CCF3)4 and Re2Cl4(dppm)2 complexes that
contain the polynitrile ligands DM-DCNQI and TCNQ are described. In
addition, the products of dimolybdenum pivalate reactions with TCNE are

presented.
2. Experimental Section

A. Starting Materials and Reaction Procedures

M02(02CMe)4 and M02(02CCMe3)4 were prepared by the reaction of
Mo(CO)6 with the corresponding carboxylic acid.18 M02(02CCF3)4 was
prepared by the reaction of M02(02CMe)4 with trifluoroacetic acid and
trifluoroacetic anhydride under reflux conditions.19 TCNQ and TCNE were
purchased from Aldrich and purified by sublimation before use. DM-
DCNQI was prepared by the reaction of the 2,5-dimethquuinone with
bis(t1imethylsilyl)carbodiimide in the presence of TiCl4.20 Re2C14(dppm)2
Was prepared by the literature method.21 All solvents were deoxygenated
and freshly distilled before use, and all reactions were carried out in an

al‘gon or N; atmosphere unless otherwise noted.

B. Synthe
(l)Prepa
(a) 2:1 Re

M0]
were loud
THF. Tht
THFwasl
uh3x§
WR spel
Wood
1020(w),5
(ll) 2:1 Re
Mu
Were load?
tuluene. 1
098650]:
had fOl’met
diffractomt
Problem 1
122ng br)
503th), 47

 

28

B. Synthesis
(1) Preparation of M02(02CCF3)4(TCNQ) (1).
(a) 2:1 Reaction of Moz(02CCF3)4 with TCNQ in THF.

Moz(02CCF3)4 (0.095 g, 0.02 mmol) and TCNQ (0.020 g, 0.01 mmol)
were loaded into a Schlenk tube and dissolved in 20 mL of freshly distilled
THF. The yellow-green mixture was stirred for 12 h after which time the
THF was removed by vacuum. The resulting dark green residue was washed
with 3 x 5 mL of diethyl ether. The product was examined by infrared and
NMR spectroscopies. IR (KBr pellet, cm'l): 3435(vs), 2924(m), 2882(m),
2264(m), 2226(w), 1730(s), 1639(m), 1510(w), 1415(w), 1192(vs), 1170(w),
1020(w), 856(w), 825(w), 731(w).

(b) 2:1 Reaction of Moz(02CCF3)4 with TCNQ in toluene.

M02(02CCF3)4 (0.095 g, 0.02 mmol) and TCNQ (0.020 g, 0.01 mmol)
were loaded into a Schlenk tube and dissolved in 20 mL of freshly distilled
toluene. The orange mixture was refluxed for 12 hours, and the resulting
orange solution was stored in the freezer (~ —12 °C) until red needle crystals
had formed. A few crystals was screened on the Nicolet (Siemens) P3N
diffractometer but they were unable to be indexed due to a severe twinning
problem IR (Nujol, cm'l): 2243(w), 2218(m), 1603(s), 1589(vs), 1545(m),
1228(s br), 1190(vs br), 1157(vs br), 858(m), 850(m), 781(m), 731(vs),
503(m), 474(m).

(c) 2:1 Reaction of Moz(02CCF3)4 with TCNQ in xylenes.

Quantities of Moz(02CCF3)4 (0.095 g, 0.02 rmnol) and TCNQ (0.020
g, 0.01 mmol) were loaded into a Schlenk tube along with 20 mL of
degassed xylenes (b.p. 138.2-141.7°C). The red mixture was refluxed for 12

h in an oil bath and slowly cooled to room temperature. IR (Nujol, cm"):

2243(w),
1190(vs
474(w).
hours. (
a dark I
(2) Prepa
(a) 1:1
benzene]
Mo‘
benzene ih
benzene w
undisturbe
"OP of re
wfished w
Pressure
2232(w), j
1234(3), 1
lb) Grow

Sing

29

2243(w), 2220(m), 1603(s), 1589(vs), 1570(m sh), 1545(m), 1228(vs br),
1190(vs br), 1159(vs br), 858(8), 781(m), 733(vs), 518(w sh), 503(m),
474(w). Dark red, rectangular platelet crystals formed over the course of 24
hours. (Reaction in degassed m-xylene led to the immediate precipitation of
a dark red product but no crystals.)

(2) Preparation of M02(02CCF3)4(DM-DCNQI) (2).

(a) 1:1 Reaction of Moz(02CCF3)4 with DM-DCNQI in
benzene/CH2C12.

M02(02CCF3)4 (0.045 g, 0.1 mmol) was dissolved in 10 mL of
benzene in a Schlenk tube and DM-DCNQI (0.018 g, 0.1 mmol) in 15 mL of
benzene was slowly added. The yellow brown mixture was allowed to stand
undisturbed for 24 hours, after which time the solvent was decanted and a
crop of red-brown needle-like crystals was collected. The crystals were
washed with 3 x 5 mL of freshly distilled benzene and dried under reduced
pressure (yield 0.035 g, 56%). IR (Nujol, cm"): 3497(m), 3365(m, br),
2232(w), 1726(w), 1701(w, sh), 1603(s), 1537(s, br), 1410(w), 1354(m, sh),
1234(8), 1194(vs, br), 1167(vs), 858(m), 781(m), 733(s), 476(w).

(b) Growth of single crystals by slow diffusion reaction.

Single crystals of (2) were grown in a sealed tube by a slow diffusion
reaction. A 2 mL CH2C12 solution of M02(02CCF3)4 (0.047 g, 0.01 mmol) in
5 mL of CHZCIZ was transferred to a 1 mm diameter Pyrex tube, and a 2 mL
solution of 2,5-DM-DCNQI (0.018 g, 0.01 mmol in 10 mL of benzene) was
carefully layered on top of the Moz(02CCF3)4 solution via a syringe. The
tube was flame-sealed under a slight vacuum and allowed to stand for one
week. Brown needle crystals were observed to form at the interface of the
light yellow solution. The crystals were harvested for a crystallographic
study.

 

 

«~-

CPS

F igu

 

 

 

30

m
m m om
m m. we 0
m we «.mt Fm.
c m w m almmwd
m w ,e...
M C
. u as
as. m“ m Ht
m cow/Mb m. H m
I flu u WHOM
«a. ,c s
C 3 0 IO 3
are , /o «rm
GI Iva

(:2 N-«v

Me

 

 

Red-brown solid

You) = 2232 cm'1

 

 

Figure 6. Preparation of [M02(OZCCF3)4 DM-DCNQI-Cgl-Ig]... (2).

 

31

(3) Preparation of {[Moz(02CCMe3)3(NC)2CC(CN)CONH]
°8CH2C12}... (3).
(a) Reaction of Moz(02CCMe3)4 with TCNE in CH2C12 at -78 °C.

M02(02CCMC3)4 (0.120 g, 0.02 mmol) was loaded into a 100 mL
Schlenk flask and dissolved in 20 mL of freshly distilled CH2C12. TCNE
(0.026 g, 0.02 mmol) in 30 mL of CH2C12 was added drop-wise through a
cannula to the Schlenk flask which was cooled in an isopropanol/C02(s)
bath (—78 °C) over a period of 1 h. The Schlenk flask was then kept in the
low temperature bath. No visible color change in the dark green solution
was observed after one week. The solution was warmed up to room
temperature, and stirred for 10 h, after which time the solution was stored in
a freezer for three weeks. During this time tiny amounts of red crystalline
product were observed to have formed.

(b) Reaction of Moz(02CCMe3)4 with TCNE in CH2C12 at room
temperature.

M02(02CCMe3)4 (0.120 g, 0.02 mmol) was dissolved in 20 mL of
freshly distilled CHzClz, and TCNE (0.026 g, 0.02 mmol) in 30 mL of
CH2C12 was added drop-wise through a cannula over the period of l h. The
mixture was stirred at r.t. for 12 h, during which time the solution changed
from dark green to dark brown. A small amount of black solid was observed
on the side of the flask. Red hexagonal platelet crystals grew after the
solution was stored at low temperature (-10 to -40°C) for 5 days. The
crystals continued to grow over the period of one month. The crystals are
soft and thin, and the larger ones have the tendency to be twinned. When
removed from the mother liquid, the crystals quickly lose dichloromethane
which leads to rapid loss of crystallinity. IR (Crushed crystals between KBr
plates, em"): 3440(vs, br), 3260(s, sh), 2980(m), 2204(s), 2160(m, sh),

1633(vs.
1020(w),
(c) Singlt
condition
(i)Ethyle
A
glycol/Ch
microcrys
days.
(it) Freezl
A 1
for three
0f~02:
(ill) low
A
Crlogen

 

32

1633(vs, br), 1458(8, br), 1290(w), 1267(m), 1221(8), 1130(w), 1036(w),
1020(w), 1010(w), 895(m), 796(m, sh), 717(8), 696(m, sh), 610(m), 455(w).
(c) Single crystal growth of (3) under various low temperature
conditions.

(i) Ethylene glycol/C02(s) bath.

A reaction flask containing (3) was cooled in an ethylene
glycol/C02(s) bath at -12°C for one week. A finely divided red
microcrystalline product was observed to be present in the solution after 4
days.

(ii) Freezer storage.

A reaction flask containing (3) was stored in the freezer at ~ -12°C
for three to four weeks, during which time small hexagonal platelet crystals
of ~ 0.2 mm diameter and ~ 0.05 mm thickness had formed.

(iii) Low temperature circulating bath.

A reaction solution containing (3) was placed in a programmable
Cryogen II-80 low temperature bath. The temperature was set to circulate
between -10 to —40°C with a gradient of i 1°C per hour. After ~ 1 week,
red hexagonal crystals in the size range of 0.2 — 0.4 mm diameter were
observed. Within two weeks, the crystals were in the m size range. From
time to time, colorless needles were observed at the surface of the solution,
but they melted immediately and dissolved in the CH2C12 solution when the
flask was disturbed or removed from the bath. Attempts were made to
remove the melted crystals by pipette for characterization. IR (liquid drops
between KBr plates, cm"): 2261(w), 2202(w), 1684(vs br), 1606(m, br),
1560(w, sh), 1550(w, sh), 1471(w), 1402(m, br), 1259(vs), 1089(vs, br),
1020(vs, br).

(iv) Sols'

Th
30 ml. 0
diffusion

 

33

(iv) Solvent diffusion in a fritted H-tube at ~ -12°C.

The reaction mixture was transferred to an H-tube via a cannula, and
30 mL of hexane was added to the other side of the H-tube to allow for slow
diffusion crystallization. The H-tube was maintained in an
ethyleneglycol/C02(s) bath (~ -10 to ~ -40°C) for 10 days. A small quantity
of red crystals was obtained by this method.

(4) Preparation of
{[M02(O;CCMe3)3(NC)2CC(CN)CONH]°0.SC6H‘}.. (4).
(a) Room temperature reaction.

Quantities of M02(02CCMe3)4 (0.060 g, 0.1 mmol) and TCNE (0.013
g, 0.1 rrrmol) were dissolved in 10 mL of benzene and stirred for two days at
room temperature. A yellow brown precipitate was collected by filtration
and washed with 3x5 mL of diethyl ether. IR (KBr pellet, em"): 3420(vs,
br), 3330(8, br), 3204(8, br), 2967(m), 2928(m), 2878(vw, sh), 2854(vw, sh
2261(vw), 2216(vw), 2157(w), 1686(vs), 1626(m), 1520(m), 1485(8),
1460(w), 1425(8), 1402(8, 8h), 1381(8, sh), 1365(m, sh), 1228(m), 1196(w,
sh), 1091(vw), 976(8), 734(8, br), 614(w), 488(w, br).

(b) Benzene extraction from the CH2C12 reaction product (4).

The reaction proceeded at room temperature as described in (4). The
dark brown solution was kept in a low temperature bath (~ -20 °C) while
CH2C12 was removed by a dynamic vacuum. Freshly distilled benzene (20
mL) was added to the resulting residue, and the dark green solution was
transferred to a Schlenk flask by a cannula. The solution was stored at room
temperature for one week, during which time, tiny red hexagonal platelet
crystals formed. IR (Nujol, em"): 2216(w), 1684(m, br) 1221(m, br)
1155(m), 1093(w), 1022(w), 970(w), 800(w), 721(m), 638(w), 555(w),
503(m).

34

(5) Preparation of [RezCl4(dppm)2]2TCNQ-10THF (5).
Re2C14(dppm)2 (0.100 g 0.04 mmol) in 3 mL of THF was layered with
a THF solution of TCNQ (0.007 g 0.04 mmol) in a 1 mm Pyrex tube. The
tube was sealed under vacuum and stored at low temperature (~ 2 °C) for
three days during which time small brown blocks had grown. Removal of
the crystals from the THF solvent led to rapid loss of crystallinity.
(6) Preparation of [Re2C14(dppm)2]zDM-DCNQI-10THF (6).
RezCl4(dppm)2 (0.100 g, 0.04 mmol) in THF was layered with a THF
solution of DM-DCNQI (0.007 g, 004an1101) in a 1 mm Pyrex tube. The
tube was sealed and stored at low temperature (~ 2°C) for three days to yield
a small quantity of tiny brown crystals.

C. X-ray Crystallographic Studies

General data collection procedures are described in appendix A.
(1) [M02(02CCF3)4]2(TCNQ)'2(m-Cero)

'[M02(02CCF3)4]'2("1'C8H10) (1)-

Single crystals of (1) used for crystallographic study were grown from
the reaction of M02(02CCF3)4 and TCNQ in xylenes (b.p. 138—141°C). A
rectangular platelet crystal of approximate size 0.73 x 0.55 x 0.14 mm was
secured on the tip of a glass fiber with Dow Corning silicone grease. The
crystal was mounted on a Nicolet (Siemens) P3N diffractometer equipped
with a low temperature device at -108_-t_;1°C. Least-squares refinement using
20 well centered reflections in the range 15° 3 20 5 25° indicated a triclinic
system with a = 11.712(2) A, b = 12.831(3) A, c = 15.882(3) A, a:
73.88(3)°, fl= 73.87(3)°, y: 87.87(3)°, v = 2200.8(8) A3. The data were
collected at -108i1°C by the (1)-20 scan technique in the range 4° $20350°.
The total number of reflections measured were 6908, of which 6536

reflectio
refined
package
the diso
anisotro
occupanfi
oceupantl
molecqu
disordero
residuals
R1: 0.06
and with
(2) Mom
Sir
SOlution I
Motor

appf 0X11}:

 

35

reflections are unique. The structure was solved by direct methods and
refined by full-matrix least squares on I“2 using the Siemens SHEXTL v5.03
package on a Silicon Graphics Workstation. All non-hydrogen atoms except
the disordered -CF3 group and the m-xylene molecules were refined
anisotropically. The disorder in the -CF;; groups was modeled at 50%
occupancy over two conformers for one -CF3 group, and at ~ 33%
occupancy over three conformers for a second —CF3 group. The m-xylene
molecules were modeled at 50% occupancy each for two orientations. All
disordered atoms were refined isotropically. The structure was refined to
residuals of R1 = 0.0478 (wR2 = 0.119) for 5385 reflections with I > 20 and
R]: 0.0616 (wR2= 0.130) for all 6532 reflections against 625 parameters
and with a goodness of fit = 1.232.

(2) Moz(02CCF3)4 DM-DCNQIOC6H6 (2).

Single crystals of (2) were grown by slow diffusion of a benzene
solution of 2,5-DM-DCNQI into a methylene chloride solution of
M02(02CCF3)4 in a 1 mm diameter sealed Pyrex tube. A crystal of
approximate size 0.20 x 0.20 x 0.35 mm was secured on the tip of a glass
fiber with Dow Corning silicone grease and mounted on a Siemens SMART
1K CCD platform diffractometer at a detector distance of 6 cm. The crystal
was bathed in a cold nitrogen stream at -140(2)°C with a Siemens LT2
cooling device. Preliminary examination and final data collection were
carried out with graphite-monochromated Mo K0: radiation (Au = 0.71073
A) powered at 50 KV and 40 mA. An indexing of 45 frames of data with
exposure times of 10 sec/frame indicated a monoclinic crystal system. A
hemisphere of data with 1321 frames was collected with a scan with of 03°
in (o and an exposure time of 30 sec/frame. Final unit cell parameters were

generated from the refinement of the XYZ centroids of the reflections with

indlCat

3C1

36

I > 100(1) from a total of 6044 reflections with a =20.755 (4) A, b = 13.794
(3) A, c = 13.924 (3) A, 13: 130.92 (3) °, and V: 3012.1(11) A3. Symmetry
related and multiply measured reflections were averaged. Crystal decay and
an azimuthal absorption correction were accounted for by the program
Xprep,22 which led to Rim = 0.0415, and transmission factors between 0.789
and 1.0. Systematic absences indicated that the crystal belongs to the
monoclinic space group C2/c (#15), with Rim = 0.0476 and Rsigm, = 0.0349.
The structure was solved by direct methods and refined by full-matrix least
squares on F2 using the Siemens SHEXTL v5.0322 package on a Silicon
Graphics Workstation. A trifluoromethyl group disorder was modeled at
half occupancy for each of the two conformers. All atoms were
anisotropically refined except for the disordered —CF3 group. Hydrogen
atoms were fixed to ideal positions. A summary of crystallographic data for
(2) is listed in Table 1.
(3) {[M02(OzCCMe3)3(NC)2CC(CN)CONH]08CH2C12 }.. (3).
(a) Data collected on a Siemens P3/V diffractometer.

A thin hexagonal platelet of (3) of approximate dimensions 0.5 x 0.5 x
0.1 mm was secured on the tip of a glass fiber with Dow Corning grease.
The crystal was mounted on a Nicolet (Siemens) P3N diffractometer
equipped with a low temperature device at -110(2)°C. Least-squares
refinement using 20 well-centered reflections in the range 15° 5 20 _<_ 25°
indicated a trigonal/hexagonal system with a = b = 27.1010(38) A, c =
22.8610(46) A, a: fl=90°, y: 120°, v = 14,000(8) A3. Axial photographs
confirmed that the Laue symmetry of the crystal is 6/mmm. Data were
collected at -110(2)°C for 3692 reflections of which only 16% had
intensities of I > 40( F). Due to the low number of observed data, only a
partial solution was obtained with the program SHELXS86 and DIRDIF23 of

the TE
reform

(lea
dilirar

secure:
mount
detectt
at -l4t
and fit
K0 ra.
10ml c
indica
bane
30 sec
of the
Was 1
22.80
and 0
pr Ogr;
XPRi
hexag

Straig

37

the TEXSAN24 v5.0 package in the space group P6/mcc (#192). No
refinement was attempted.

(b) Data collected on a Siemens SMART CCD 1K platform
diffractometer.

A crystal of approximate dimensions 0.46 x 0.40 x 0.20 mm was
secured on the tip of a glass fiber with Dow Corning silicone grease and
mounted on a Siemens SMART 1K CCD platform diffractometer at a
detector distance of 6 cm. The crystal was bathed in a cold nitrogen stream
at -140(2)°C with a Siemens LT2 cooling device. Preliminary examination
and final data collection were carried on with graphite-monochromated Mo
K01 radiation (A0, = 0.71073 A) powered at 50 KV and 40 mA. Indexing a
total of 45 frames of preliminary data with exposure time of 10 sec/frame
indicated a monoclinic crystal system. A hemisphere of data with 1321
frames was collected with a scan with of 0.3° in 0) and an exposure time of
30 sec/frame. Final unit cell parameters were generated from the refinement
of the XYZ centroids of 8192 reflections with ”0) 10. The crystal system
was found to be trigonal/hexagonal with a = b = 46.935602) A, c =
22.8009(6) A, [3: 120.0 (3) °, V = 433243301) A3. Beam inhomogeneity
and crystal absorption and decay were corrected for by application of the
program SADABS.25 Data merging was performed with the program
XPREP. Systematic absences indicated that the crystal belonged to the
hexagonal system, but space group determination failed to be
straightforward due to violations. Structure solution proceeded by a trial and
error method, with the best results being achieved in space group P-31c
(#163) with a merohedral twinning matrix (0 -1 0, -1 0 0 ,0 0 -1 ) or
in P6cc (#184) with the merohedral twinning matrix (1 0 0, 0 1 0, 0 0 -1).

Tabb:

Smmo
Tm
(hysui
Space

a,A
uh
(31“
0, deg
fl dc,1
7;deg
V, A3
T. K
Radia

Dsalc,

HOOC
COUC‘t

R] (W
(30F‘

NOle:
gOOdr
refleC

38

Table 1. Structural parameters for {[M02(02CCF3)4]2(u4-TCNQ)02m-
C8H10}°°{ [M02(OZCCF3)4]°(m-C8H10)2}.. (1) and
[M02(02CCF3)4(DM-DCNQI)‘C6Holoo (2).

 

 

Structure (1) (2)
Formula C341'122F 18M03N2012 C24H14F12M02N408
Crystal system Triclinic Monoclinic
Space group P-l C2/c
a, 1 1.712(2) 20.755(4)

b, A 12.8310) 13.7940)

c, A 15.882(3) 13.9240)

a, deg 7388(3) 90
[3, deg . 7387(3) 130.92(3)

7, deg 8787(3) 90

V, A3 2200.8(8) 3012.1(11)
T, K 165(2) 133(2)
Radiation source Mo KOL (A0, = 0.71073 A )
Z 2 4

pcalc, g/cm3 1.932 1.998

n, mm'1 0.978 0.962

F(000) 1248 1768
Collected lunique Data 6908 / 6536 5917 / 2151
Parameters / restraints 625/0 251/72
R1 (wR2) (F2, I > 2 a) 0.0478 (0.119) 0.0695(0.1767)
GoF 1.232 1.098

Scan type 00-20 0)

 

Note: R1 = 2 ||F,| - |Fc|| / >3 |F,|, sz = {z [W(F02 - 18,52] / 2 [w(F°2)2]}m,
goodness-of-fit = S = {[w(F°2 - Fc2)2] / (n - p)}m, n is the total number of
reflections and p is the total parameters refined.

llle st

Subseqt
refinem
disorde
atoms

idealizel
(4) {[Mt
St
benzene
dimensit
with D0
CCD p1
stream
device.
With a g
and 40

 

 

39

The structure solutions were obtained by direct methods in the XS program.
Subsequent atoms were located from difference maps generated during
refinement cycles. Least-squares refinement and t-butyl group and solvent
disorder modeling were performed with the refinement program XL. All
atoms were anisotropically refined, and hydrogen atoms were fixed in
idealized positions.

(4) {[Moz(02CCMe3)3(NC)2CC(CN)CONH]-0.5CJ15}..(4).

Single crystals of (4) were grown from the slow evaporation of a
benzene extraction of (3). A red, hexagonal platelet crystal of approximate
dimensions 0.052 x 0.052 x 0.078 mm was secured on the tip of a glass fiber
with Dow Corning silicone grease and mounted on a Siemens SMART 1K
CCD platform diffractometer. The crystal was bathed in a cold nitrogen
stream at —100(2)°C with the Oxford Cryosystem Cryostream cooling
device. Preliminary examination and final data collection were carried out
with a graphite-monochromated Mo K01 radiation source powered at 50 KV
and 40 mA. Intensity data were collected with 0.3° (1)—scans at a detector
distance of 5 cm. Initial cell parameters and an orientation matrix were
generated from 60 preliminary data frames with a 15 sec exposure time.
The unit cell was found to be monoclinic, C-centered with a = 30.8937(9) A,
b = 11.12530) A, c = 17.9818(6) A, ,6: 105.123(1)°, V = 5966.3408) A3
from the refinement of the XYZ centroids of 2651 strong reflections with I >
100(1) from a total of 17,791 reflections. The intensities were corrected for
beam inhomogeneity, crystal decay and absorption with the program
SADABS which led to Rim: 0.0530, Tm,“ = 0.67, Tmam = 1.0. A total of
17,791 reflections were measured at a maximum 20 = 56.59, of which 6,297
were in the category of I > 20(1), and 7274 were unique. Symmetry related

and multiply measured reflections were averaged with the program Xprep.

Shun
Cflct
aehiev
hefii
smnm

(5) Cr

TCNQ
hm.

and)

40

Systematic absences indicated that the crystal belongs to the space group
C2/c (#15), with Rim = 0.11 and Rsigm, = 0.22. Structure solution was
achieved from direct methods for all non-hydrogen atoms. Amide group and
the fully rotationally t-butyl group disorder were modeled by refining the
sum of the occupancies of the different disordered groups to be unity.
(5) Crystallographic study of [Re2Cl4(dppm)2]2TCNQ-10THF (5).

Single crystals of (5) were grown by diffusion of a THF solution of
TCNQ into a THF solution of Re2C14(dppm)2 at 0°C over the course of three
days. A dark brown crystal of approximate dimension 0.39 x 0.26 x 0.09
mm was secured on the tip of a glass fiber with Dow Corning silicone grease
and mounted on a Bruker (Siemens) SMART 1K CCD platform
diffractometer. The crystal was bathed in a cold nitrogen stream at -
100(2)°C with the Oxford Cryosystem Cryostream cooling device.
Preliminary examination and final data collection were carried on with
graphite-monochromated Mo K01 radiation (A0, = 0.71073 A) powered at 50
KV and 40 mA. Intensity data were collected with 0.3° (1)—scans at a
detector distance of 5 cm.

Initial cell parameters and an orientation matrix were generated from
60 preliminary data frames with 10 sec/frame exposure times. The cell was
found to be triclinic. A full sphere of data was collected with 30 sec/frame
exposure times. The final unit cell obtained from the refinement of the XYZ
centroids of 8,192 reflections with I > 100(1) from a total of 39,411
reflections, is a = 12.18880) A, b = 14.29080) A, c = 23.31250) A, a =
72.8220(10)°, )6: 81.9380(10)°, y: 75.7760(10)° and v = 3752.06(16) A.
The intensities were corrected for beam inhomogeneity, crystal decay and
absorption with the program SADABS which led to R1 = 0.0399, Tm," =
0.78, Tmax = 1.0. A total of 17,791 reflections was measured with the

Table 2.

llM02(02C
{[M02(02C(

E
ldentifrcatio
Empirical ft
Formula we
Temperaturt
Wavelength
Crystal syst
Space grou
Unit cell dii

 

 

41

 

 

Table 2. Crystal data and structure refinement for
{ [M02(02CCMe3)3(NC)2CC(CN)CONH]°8CH2C12}.. (3) and
{ [Moz(02CCMe3)3(NC)2CC(CN)CONH]-0.5C6Hg },., (4).
Identification code (3) (4)
Empirical formula C73H111 C133 M06 N12 021 C50 H66 C14 M04 N3 014
Formula weight 3281.62 1528.67
Temperature 133(2) K 133(2) K
Wavelength 0.71073 A 0.71073 A
Crystal system Hexagonal Monoclinic
Space group P6cc C2/c
Unit cell dimensions a = 46.8862(4) A a = 30.855(4) A
b=46.8862(4) A b= 11.112603) A
c = 22.7568(4) A c = 17.964(2) A
a: 90°. 0: 90°.
[3: 90°. ,6: 105.111(2)°.
y: 120°. y: 90°.
Volume 43324.3(9) A3 5946.503) A3
Z 12 4
Density (calculated) 1.509 Mg/m3 1.708 Mg/m3
Absorption coefficient 1.164 mm'l 1.072 mrn'l
F (000) 19626 3080
Crystal size 0.46 x 0.40 x 0.20 mm3 0.052 x 0.052 x 0.078
mm3
Reflections collected 10711 13892
Refinement method Full-matrix on F2 Full-matrix on F2
Data/ restraints/ 10707/ 1734 / 815 5023 / 32 l 236
parameters
GoF 1.367 1.062
RI (wR2) [Fz, I>20( I )] 0.1762/0.3880 0.1 101/0.21 17

 

Note: R1 = 2 ||F,| - |Fc|| / 2 |F,|, wR2 = {2 [w(F02 - Fc2)2] / 2 [w(F.2)2]}‘°.
goodness-of-fit = s = {[w(F02 — F32] / (n - p)} ”2, w = 1 / [02(F02) + (aP)2 +
bP]

maxim
reflecti

f
with th
was tn
solutio
SHEX'
Hydro;
hydrog
methot
were r
summ;
(0C1:
of DR
COllrse
0.35 x

SillCOH

C1703)

42

maximum 20 = 56.59, of which 26,832 were of the type I > 20(1) and 16,984
reflections were unique.

Symmetry related and multiply measured reflections were averaged
with the program XPREP. Systematic absences confirmed that the crystal
was triclinic P -1 (#2), with Rim = 0.0493 and Rsigm, = 0.0805. Structure
solution was achieved by application of the direct methods program in
SHEXTL v5.04 with subsequent atoms being located from difference maps.
Hydrogen atoms were calculated and fixed during refinement. All non-
hydrogen atoms were refined anisotropically by full matrix least-squares
methods. Disordered solvent molecules situated around the inversion center
were modeled by refining two individual fragments as half occupied. A
summary of refinement details is listed in Table 3.

(6) Crystallographic study of [Re2C14(dppm)2]2DM-DCNQI-lOTHF (6).

Single crystals of (6) were grown by slow diffusion of a THF solution
of DM-DCNQI into a THF solution of Re2Cl4(dppm)2 at -12°C over the
course of a week. A dark brown crystal of approximate dimensions 0.50 x
0.35 x 0.20 mm was secured on the tip of a glass fiber with Dow Corning
silicone grease and mounted on a Siemens SMART 1K CCD platform
diffractometer in a cold nitrogen stream at -100(2) °C with the Oxford
Cryosystem Cryostream cooling device. Preliminary examination and final
data collection were carried out with a graphite-monochromated
MoKa (Au: 0.71073 A) radiation source powered at 50 KV and 40 mA.
Intensity data were collected with 0.3° (1)-scans at a detector distance of 5
cm.

Initial cell parameters and an orientation matrix were generated from
60 data frames with 10 sec/frame exposure times. A full sphere of data was
collected for a total of 2,474 frames and an exposure time of 22 sec/frame.

The fl
c=23
V = 3
7242 .
lhe il
absoq
0.662,
maxin
17,051

with t
triclin
Soluti.
V5.05
differ
fixed

motel

by rt

43

The final unit cell was triclinic with a = 12.12640) A, b = 142531900) A,
c = 23.0826(3) A, a: 74.5780(10)°, ,6: 83.1510(10)°, y: 75.4330(10)° and
V = 3716.42(8) A3 obtained from the refinement of the XYZ centroids of
7242 strong reflections with I > 100(1) from a total of 41,616 reflections.
The intensities were corrected for beam inhomogeneity, crystal decay and
absorption with the program SADABS which led to R1 = 0.0560, Tm,“ =
0.662, Tm = 1.0. A total of 41,123 reflections was measured with the
maximum angle 20 = 56.63, of which 25,260 were of the type I > 20(1) and
17,059 reflections were unique.

Symmetry related and multiply measured reflections were averaged
with the program XPREP. Systematic absences confirmed that the system is
triclinic P -1 (#2), with Rim = 0.071 and Rsigma = 0.12. Initial structure
solution was achieved by application of the direct methods in SHEXTL
V5.04, with subsequent atoms being located from a series of alternating
difference maps. Hydrogen atoms were placed in idealized positions and
fixed during refinement. Full-matrix least-square refinement was carried out
172. All non-hydrogen atoms were refined anisotropically. Disordered THF
molecules which occupy positions around the inversion center were modeled
by refining two fragments at half occupancy each. A summary of

refinement details is printed in Table 3.

3. Results and Discussion

One of the interests in the current work is that metal-metal bonded
compounds with unpaired spins are interesting building blocks for magnetic
materials. The qualitative orbital interactions in metal-metal bonded species
are illustrated in the simplified diagram in Figure 7.

The HOMOs for the different metal-metal bonded units in this study

Table 3.
and [RegCl

 

Structure

 

Crystal sys
Space grou
all

b, A

ch

as degree
A degree

1 degree

V, A3

Z

0931C. g/ Cl
is mm"
F(000)
Radiation .
D313 Colle.
Rim) ‘ I i
PWtEr
R1 (wR2)
00F '1
COHCCITHE
Scan 6
NOte; a R
=ZRFJ‘
Of‘fil 7: S

44

Table 3. Structural parameters for [Re2Cl4(dppm)2]2TCNQ°10THF (5)
and [Re2C14(dppm)2]2DMDCNQI-l0THF (6).

 

 

Structure (5) (6)
Crystal system Triclinic Triclinic
Space group P-1 P-1

a, A 12.18880) 12.12640)
b, A 14.29080) 142531900)
c, A 23.3125(6) 23.0826(3)
a, degree 72.8220(10) 74.5780(10)
[3, degree 81.9380(10) 83.1510(10)
)1 degree 75.7760(10) 75.4330(100
v, A3 3752.06(16) 3716.42(8)
z 2 2

Dcalc, g/cm3 1.546 1.551

)1, mm1 3.501 3.534
F(000) 1750 1740
Radiation Source Mo K01 (ACl = 0.71073 A)
Data collected / unique 3941 1/ 16984 41616/ 17152
R(int.) ' / R(sigma) " 0.0493/0.085 0.0710/0.012
Parameters / constrains 787/100 745/150

R1 (wR2) (re, I > 2 0) ‘ 0.0529 (0.1172) 0.0648 (0.1526)
GoF ‘ 0.985 0.836
Collecting Temperature,°C -100(2) -100(2)

Scan type 0) 0.)

 

Note: ' R,,,, = 2 pr,2 — Fc2(mean)| / 2 [F02], " Rpm = 2 [0(F02) / 2 [F02], ° R1
= 2 ||F,| - |Fc|| / 2 |F,|, wR2 ={2 [w(F02 - F52] / 2 [w(F02)2] }"2, " Goodness-
of-fit = s = { [w(F02 - F52] / (n - p) }"2 , w = 1 / [02(1702) + (aP)2 + bP]

45

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ll ll ll ll 11*
it I +1 or
ii “i H il il 5

4| 1| 1in it it till
H H In IV '1 W H H n
it it it it a

It It u n
M02“. [1’ R e21". 111 R 6211. ll, Ruzlll, I" Ruzll, III R112". 1]

Figure 7. Metal-metal bonded orbitals diagrams.

are 8, 5*, 5*1t* or the W" orbitals (Figure 7). Ligands with unpaired spins
such as TCNQ'I', TCNE'I' or DMDCNQI" may interact with the unpaired
electrons residing on the metal-metal bonded unit. When the HOMO of an
axial ligand is W“, which is the case for nitrile or carbonyl groups, there is no
interaction between the unpaired electron in the ligand with the HOMO of
M025+ and Re?“ unit with unpaired electrons in the 5 or 8* (dxy-based)
orbitals of the metal-metal bond. There will be no direct overlap of the
orthogonal M-L orbitals (Figure 8), therefore, one would expect a
ferromagnetic interaction pathway. When unpaired spins from an axial
ligand residing in 1t* orbitals interact with a metal-metal bonded unit such as
Ru;4+ and RU25+, however, the n* orbital of axial ligands overlaps with the
HOMO (Figure 8), therefore providing a direct antiferromagnetic pathway.
When a ligand with an unpaired spin bonds to the equatorial position
of a metal-meal bonded unit such as R625+, the interaction is quite different.

The 5/6* levels will then have the correct symmetry to overlap with the 1t*

46

/"\ 11*

Non-bonding

7"" A 71*

dxz,dyz dxz,dyz

 

 

Figure 8. Interactions from an axial ligand to metal-metal bonded units
such as M02!" 11.

orbital of t
antiferromr

The
used to bl|

 

compounds
particular I
regard. De
3: 3/2 git
When bridg
anions. \l
TCNE”, T1
or, m),

47

orbital of the ligand, and there should be good direct communication or an
antiferromagnetic interaction (Figure 9).

The tetracarboxylate complexes of Ruzs”, M02“, and R625+ can be
used to build low-dimensional arrays with multidentate ligands. Such
compounds are of potential interest for their magnetic properties. In
particular the diruthenium complexes have attracted high interest in this
regard. Depending on the bridging ligand, the diruthenium complex, with an
S = 3/2 ground state, has been found to exhibit antiferromagnetic coupling
when bridged by closed shell organic bidentate ligands or simple inorganic
anions. When bonded to an S = 1/2 organic n—acceptor radical such as
TCNE", TCNQ", or DMDCNQI", in the case of tetrakispivilatodiruthenium
(II, III), the complex behaves like a ferrimagnet due to local

antiferromagnetic coupling but without cancellation of spins.

A. Preparation and Characterization
(1) Synthesis of the TCNE product of Moz(02CCMe3)4.

TCNE has been used extensively as a building block in conducting
and magnetic materials. One of the most surprising examples of its use is in
the compound V(TCNE)x that shows high Tc ferromagnetic ordering.26 One
of the unfortunate facts of the chemistry of TCNE, however, is that it is
highly reactive toward hydrolysis, condensation and cleavage, especially in
the presence of metal ions that enhance its reactivity.27 In spite of its
drawbacks, it has been noted that TCNE can be incorporated into a 3-D
arrays such as Rh2(02CCF3)4TCNE-C6H6.28

TCNE hydrolysis reactions have been reported in the literature.27 The
common hydrolysis routes involve the release of an HCN molecule when a

base is present, to form a cleavage products as shown in Figure 10. TCNE

48

 

‘6’“! ‘N,

Bonding

' 8‘ \ 1c“
‘ ‘ Bonding
dxy dxy
n*
R “at: Non-bonding

dxz. dyz dxz, dyz

      

   

Figure 9. Interactions between equatorial ligands and metal-metal bonded
units such as Rezn‘m.

 

 

 

N
H N N
N N C \\ //
\ / C C
\ C/ NEG; _ >=—§—
— c C _. GEN
//.> <.. .4 Ne .
N
. 1y
H20,B'
HCN H20,B'
HCN
”I c//N N\\C 0%”
M C c
(4 s.
0
\\\\\~
N
g H20
8
H3CC(O)CH3
N y N
“I III
N C c
//
H C NEC NEC
NEC > /\ GEN \ GEN \ GEN
/c H20 Me Me H
M T \y 0 \~ 0
.. 1.

Figure 10. Documented reactions involving the decomposition of TCNE.

l
I

also form
acetone. 27
polynitrile
relevance 1
route has 0
high pressu
no CN old
believed to

 

tricyanoacr
PIECCdence
unusual rel
COummd

Fror
the llllIOge
“0n. Coon

Energy tha

3' Crystal

(1) Mozlfl
All

Composed

to four ind

50

also form adducts with small molecules such as water, methanol and
acetone. 2" TCNE can also further react with the intermediate to form other
polynitrile anions or act as an olefin in organometallic reactions. Of direct
relevance to the present work is the fact that a very uncommon hydrolysis
route has been reported in which TCNE is hydrolyzed under the condition of
high pressure.27 The product is an addition product of H20 and TCNE with
no CN cleavage. Structure (3) from our reaction with M02(02CBu‘)4 is
believed to be the only example wherein the hydrolysis of TCNE to give the
tricyanoacrylamide anion has occurred under mild conditions. As
precedence for this work, we note that Cotton and Haefner observed the
unusual related hydrolysis reaction of acetonitrile in the presence of a Mop4+
compound and a base.29

From the crystallographic data, one cannot determine the location of
the nitrogen versus the oxygen atom A semi-empirical calculations on the
non- coordinated anion ligand suggest that conformer (1) is of a lower

energy than conformer (2).

B. Crystallographic Studies.
(1) M02(02CCF3)4 TCNQ

An X-ray crystallographic study revealed that compound (1) is
composed, in part, of a chain of 04-TCNQ ligands bound in axial positions
to four independent M02(02CCF3)4 molecules to give a 1-D polymeric motif.
A second type of M02(02CCF3)4 molecule is situated in a cavity with two m-
xylene ligands perpendicularly oriented near the axial positions. The
midpoint of the Mo-Mo bond in the unique M02 unit and the center of the
u‘-TCNQ rings reside on inversion centers. A thermal ellipsoid
representation of (1) is depicted in Figure 12 and selected bond lengths and

l?—

bond anglt

The
2.1131(1 l l
statistically
the MM h!
Mo-N(TClI
distances 1

l.

 

other axial
A, L = I:
L .-. 4,4’_
Stretching

ViCiN) :

51

bond angles are listed in Table 4.

The Mo-Mo bond length of 2.1126(8) A in the polymeric chain and
2.1131(11) A in the molecule of M02(02CCF3)4(m-xylene)2 in (l) are not
statistically different from each other, but they are longer by ~0.023 A than
the M-M bond in M02(02CCF3)4 (2.090(4) A) without axial ligands.30 The
Mo-N(TCNQ) bond length of 2.627(5)A is longer than the corresponding
distances reported for polymeric M02(02CCF3)414 complexes containing
other axial nitrogen donor ligands (Mo-Mo = 2.127(2) A, Mo-N = 2.531(8)
A, L = DM-DCNQI (2); Mo-Mo = 2.128(1) A, Mo-N = 2.557(8) A,
L = 4,4’-bpy).31 These observations taken together with the v(CEN)
stretching frequencies of 2243 and 2220 cm'1 for (1) (neutral TCNQ
v(CEN) = 2230 cm") imply that the interactions between the nitrile and the
Mo-Mo unit in (1) are weak rt-interactions. The TCNQ ligand is also
involved in n—stacking interactions with interstitial m-xylene solvent
molecules, with the shortest distance from the m-xylene to TCNQ being
343(2) A. The presence of these n—interactions helps to stabilize the
structure, and, indeed, crystals will not grow in the absence of xylenes. The
xylenes used in the reaction is a mixture of isomers (b.p. 138.2-141.7 °C),
but only m-xylene is incorporated! Furthermore, neither benzene nor
toluene is suitable as solvents for crystal growth. These collective
observations point to the selective enclathration of m-xylene in the formation
of this rather unusual inclusion compound.

Metal-metal bonded complexes involved in axial coordination to
arenes are rare. In fact only two examples, Cr2(02CCF3)4-CgH¢32 and
Moz(02CP(Bu‘)2)4'2C6Hg,33 have been reported for Group 6 metals.
Calculations performed on Cr2(02CCF3)4-C6H6 by the Xa-SW-SCF method

indicate that the en, orbital of the benzene molecules donate electron

Table 4.

 

M01
M01 1
M01 1
M01 1
M01 1
M01 (
C17 (
C13 1

52

Table 4. Selected bond distances (A) and bond angles (°) for
[M02(02CCF3)4(TCNQ)]«». (1)-

A
M01
M01
M01
M01
M01
M01
C17
C13

M02
C13
N1

B
M02
04
05
01
N1*
0(8)
C18
C14

B

Mol

N1

C13

A-B (A)

2.1126(8)

2.1 18(4)
2.124(4)
2.127(4)
2.630(6)
2.105(4)
1.337(9)
1.431(9)

C
N1"‘
Mol*
C14

Bond Distances

A
M02
M02
M02
M02
M02

N1
C 17
C 14

B
07
02
O3
06
N2

C13
C16
C16

A-B (A)
2.108(4)
2.109(4)
2.110(4)
2.134(4)
2.623(5)
1.148(8)
1.451(8)
1.363(8)

Bond Angles

A-B-C(°)

175.50(13) M01

147.5(5)
176.8(6)

A

C15
N2

M03
M03
M03
M03
M03
N2
C 1 8
C 14

B
M02
N2
C 15

010
O9
011

A-B (A)
2.107(4)
2.108(4)
2.109(4)

M03* 2.113(1)

012
C15

C16*

C15

C

N2
M02
C14

2.122(4)
1.126(8)
1.440(8)
1.453(8)

A-B-C(°)
171.15(12)

163.5(5)
178.5(7)

density to
interactior
with the pl
hme
299(1) 11'
A. These
earlier exd
2.090(4) 1
presence (1
“115'
Units in pr
111(1) is
Adopts w‘)
nitrile t0

 

 

 

53

density to the W“ orbital of the Cr; units.32 As was observed for the
interactions involving benzene, the m-xylene molecules in (1) are oriented
with the planes of the rings perpendicular to the Mo-Mo axis. The average
distance from the least squares plane of the xylene ring to the Mo atom is
2.99(1) A and the shortest Mo-C distance is essentially the same, viz., 3.00
A. These values are much less than the corresponding M-C distances in the
earlier examples.32 Furthermore, a lengthening of the Mo-Mo bond from
2.090(4) A in M02(02CCF3)4 to 2.113101) A in (1) is in accord with the
presence of axial interactions to the arene groups. 32

This study supports the feasibility of using TCNQ to connect dimetal
units in polymeric arrays. The 24-membered metallocyclic ring that forms
in (1) is a consequence of the angle that the TCNQ nitrile functionality
adopts with the M02 units. These angles are expected to vary from one
nitrile to another, thereby giving rise to different metallocyclic ring sizes,
e.g., the structure of Rh2(02CCF3).,TCNE34 involves a 30-membered ring in
a 2-D motif. In the dirhodium compound, the Rh-N—C angle is close to 180°
and four polynitrile ligands are used in the formation of the ring (Figure
11(a)). In compound (1), the two independent Mo-N-C angles of 147.5 °
and 163.5° are closer to 150°, which permits ring closure with only two
ligands (Figure 11(b)). This situation gives rise to a one-dimensional p.4-
polynitrile polymer. In addition, the large cavities created by the packing of
adjacent chains in (1) allow for the inclusion of a separate m-xylene bonded
dimolybdenum complex.
(2) [M02(02CCF3)4(DM'DCNQI)°C6H6]<» (2)

[M02(02CCF3)4(DM—DCNQI)~C6Hg]... (2) crystallizes in the
monoclinic space group C2/c, with the midpoint of the Mo—Mo bond

residing on an inversion center. The structure consists of alternating

54

 

(a) 24 membered octagon ring
or: 150° .0 =120°

   
 
    

_ N— Rh— Rh— N—

(b) 30 rnembered square ring
a=180°, B =120°

N
Figure 11. Metallocyclic rings of polynitriles with dimetal units and
idealized angles: (a) Moz(02CCF3)4(TCN Q) (1),
(b) Rh2(02CCF3)4(TCNE)r

55

6:23-24 3 600:3 3:: ~22 0:88 a
“0:0 95w: 0209 05 55 30:3 :5. 82 0:0 @355 2V mo 5000:8890: 290an Each .3 Earn

(at
_z s. 1, m8 8

., ., s.

to ma .. . .
as . $45 982 mo

06 . (as 8e

   

.v m, V—& ..~©FQ~N_.Q m0 1, ,,
.mz ..ma :2 r c-
t C

It

56

002:8 2:029: 0:35p :5 95%
no.5 3:20:86 05 5MB :35 05 mo 50.89 a @355 A8 no :ouflfivmoao: 20830 38:05. .3 3:3..—

cFZ .mo

 

 

57

.0232 05 E 2:029: 0:35p 2:
53 Ben: 5200-2: 2: do 238.. meadow 2: mazes... S .6 2% as ..e 582% mass: .3 2:3:

 

 

\S‘

 

Uflz

z

.ow—Uchvfi-rb NATVZMVQIEA- unwanvnhh uthm—duflu oznfis Znn~=nva~nhunvnv Endh mwkvvsrd-~ufl~=z “UFHurU-nhaihvmV rift mVNANIfi N.

58

 

 

 

8m 83: €82 $32 $82 o 520:
5 $2”: €33 3:32 Ewen: acme: o .N1 2520929522
8... 830.8: 3:092 2820...: 5%: 9.333 m- «1 520920226
8m 3:2: 8:3: 52:: 8:82 8522 N- .3 520920243:
new 53.: $5.: 53...: 622 6%: 2- ..c N5200.28.30
com 52: 63.2 $3.? $82 $92: 3. ._: 5520:2035
com 53: $3: 289...: 6:22 60%: 2. .Na N520022026
:0: 202v 022v c e a cause 520:2: 6.:
2 \leo a z
a
zIIIUMZ\_2

0:20:36 52400.35 30.: 22:20 005 22:59:00 :8 0.000880; 2.008000 .m 030,—.

Table 6.

M

M01’
M01’

 

59

Table 6. Selected bond distances [A] and bond angles (°) for (2).

BondDistances
A B A-B(A) A B A-B(A) A B A-B(A)
M01 M01’ 2.127(2) M01 02 2.114(6) N1 cs 1.146(12)
M01 04 2.118(6) M01 01’ 2.131(7) C5 N2 1.368(12)
M01 03’ 2.105(6) M01 N1 2.531(8)

Bond Angles

 

A B C A-B-C(°) A B C A-B-C(°)
M01 ’ M01 N1 166.7(2) M01 N1 C5 140.1(6)
M01 ’ M01 01 90.4(2) N1 C5 N2 173.1(9)

M02102

positions
ellipsoid
stacking
be seen ‘
14). Selc
the more
Mo~NsC
throngh
SWCture
DCNQI}
Simone
larger t1
S“ggesti
CsN b0

6O

M02(02CCF3)4 and DM-DCNQI molecules bonded through trans axial
positions of the dimetal unit to give a zig-zag polymeric chain. A Thermal
ellipsoid plot of a segment of the chain is depicted in Figure 13. The
stacking feature of the DM-DCNQI molecule with the benzene molecule can
be seen from the plotting diagram of the ab plane of the molecule (Figure.
14). Selected bond distances and bond angles are listed in Figure 6. One of
the more striking features of the structure is the 140.1(6)° bond angle for the
Mo-NEC bond which is a low value for an sp hybridized N atom interacting
through its lone pair. An inspection of related metric parameters for
structures in which the o—coordinated DM-DCNQI is formulated as [DM-
DCNQI]2' and [DM-DCNQI]32', reveals that the M-Nac angles in these
structures also deviate from 180°, although, with one exception, they are all
larger than 140° (Table 5).36 While the M-NEC bond angle of 140° is
suggestive of an sp2 hybridized N atom, it should be pointed out that the
CEN bond distance of 1.146(12)A, which one would expect to be longer
than for an sp nitrogen, is essentially the same as the corresponding value for
neutral, free DM-DCNQI (1.148(5) A)” The axial Mo-N distances of
2.531(8) A and the Mo-Mo distance of 2.127(2)A are typical of analogous
interactions in other Moz(02CCF3)4 complexes with nonpolymeric N-donors
such as pyridine (Mo-N 2.548(5)A, Mo—Mo 2.129(2)A)38 and 4,4’-
bipyridine (Mo-N 2.530(9)A, Mo-Mo 2.124(1)A)39. The Mo-N distance is
shorter than those structures with polymeric interactions such as
[M02(02CCF3)4(4.4’-bpy)2]~ (2.557(8)A) and [M02(02CCF3)4 (T CNQ)o.s]..
(2.622(5)A) (l). The slightly lengthened CEN single bond (1.368(12)A) in
compound (2) compared to the neutral ligand (1.350(5)A) can be attributed
to electronic effects.

The structural characterization of a one-dimensional polymeric metal

complelo
knowled;

the polyl
stacked (

 

at a sepa
polymeri
(3) Strut
Moz(t
(a) Data
"ll

(only 10'
fragmem
the pr0g,
hexagon

PTOmiSin

61

complex/DM-DCNQI compound has not been previously reported to our
knowledge. The stability of the present structure and the ease with which
the polymer crystallizes from benzene is rationalized on the basis of the n-
stacked columns formed by the DCNQI ligands and the benzene molecules
at a separation of 3.4 A. These additional interactions serve to "stitch" the
polymeric chains together hence stabilizing the structure in two-dimensions
(3) Structural characterization of transformed TCNE with
M02(02CCM€3)4
(a) Data from a P3/V instrument partially solved by DIRDIF.

The data obtained on the Nicolet P3N instrument were very weak
(only 10% of reflections were of the type F > 30). Nevertheless, a structure
fragment was solved (Hexagonal system, space group P6/mcc (#192)) with
the program SHEX886, and DIRDIF, which allowed for the detection of a
hexagonal ring of six Mo-Mo units linked by cis-u—TCNE. This was a very
promising result that encouraged us to keep trying different crystallization
techniques in order to obtain a better data set. From the structure fragment
obtained, it was not possible to know if this was a polymeric structure. In
order to assist in prediction of the structure, a simulation was performed with
the SPARTAN35 program based on the partial structure.

Our main question was whether the product is a polymeric sheet or
just a hexamer. From the simulation in Figures 15 and 16, one can ascertain
that the center of the structure contains a 10 A hole. It is also possible to
conclude that it is too crowded to link four M02 units to one TCNE ligand, in
the manner that four Rh; units were linked by a TCNE ligand.28 It is
impossible then for a polymeric structure to have formed without altering the
Mo; unit or the TCNE ligand. The possibilities that TCNE has decomposed

into another cyanocarbon molecule is not remote, given that many

Table 7

 

OrPrsm

m 0 n.
lWAaUcHWhnwboim
"mt imwmew

n

62

Table 7. Comparison of semi-empirical calculations for two 2,3,3-
tricyanoacrylamide anion isomers.

 

(1) (2)

 

 

 

 

 

 

 

 

 

ll) (2)

Heat of Formation 66.02 64.86
(Kcal/mole)

Amide H to the nearest 2.304 2.405
neighbor, A

N-H bond length, A 0.991 0.993
C-N bond length, A 1.332 1.335
CO bond length, A 1.249 1.247

 

 

63

transformations of the molecule have been reported ( Figure 10).

(b) Structure solution and refinement of data from a Siemens
SMART CCD.

(i) [M02(02CCMe3)3(NC)2CC(CN)CONH)]08CH2C12 (3)

A new batch of crystals of (3) was grown using the low temperature
circulation control bath as described in the experimental section. A crystal
with dimensions of 0.70 x 0.70 x 0.40 was mounted on a Nicolet/Siemens
P3N. It was found that the crystal diffracted strongly, but the intensity
dropped off rapidly at high angles due to extreme solvent disorder. The data
were not solved due to possible twinning as well, although the crystal could
be indexed to a trigonal/hexagonal cell with a z b z 27 A, c z 23 A, a z [3 =
90°, yz120° Vz14000 A3. Another crystal of (3) was taken to Siemens in
Wisconsin in dry ice and the crystal was mounted on a SMART CCD 1K
area detector diffractometer. A pre-data collection routine of 45 frames gave
a hexagonal system, with cell parameters of a z b z 47 A, c z 23 A, dz [3 z
90°, y z 120° V z 46000 A3. This cell is approximately three times the
volume of the cell originally indexed from the data collected on the P3N.
The exact space group was not able to be discerned due to twinning and the
high disorder in solvent molecules. The structure was therefore solved in

several space groups. Among them, solutions from space groups P-3c1

Table 8. Comparison of refinement results of (3) in different space groups.

 

 

Space group P—3c1 (#147) P6cc (#184)
Unique data 20021 (26480) 1748 1 (25328)
R(im,),R(,,m) 0.048(0.091) 0.130(0.104)
RI (wR2) 0.1636(0.375) 0.160(0.363)

GOF 2.12 2.08

 

“We 1

 

 

64

 

Figure 15. Molybdenum TCNE modeled from the partial P3N structure.

65

 

Figure 16. Simulation of dimolybdenum TCNE reaction product (3)
indicates an overcrowded hyper 114-TCNE bonding mode.

(#147) a
refineme
Space gr}
(ii) [M01

In

electrons
axial pol
involved
dangling
This is t
involves
milking i
those Wit
19 and a
in Table

Th
Which is

cGRIPOUn

 

Table 9.

 

66

(#147) and P6cc (#184) produced the best refinements. A comparison of
refinement results is produced in Table 8. A plotting diagram of (3) in P6cc
space group is shown in (Figure 17).

(ii) [M02(02CCMe3)3(NC)2C(CN)CCONH)]00.5C6H(, (4)

In this structure, it can be easily discovered that the lone pair of
electrons on two of the three unperturbed CN groups act as o donors to the
axial positions of another M02 unit. The bond length of the nitrile groups
involved in a weak o-bond with Me2 are 1.141(17) A, 1.138(13) A while the
dangling nitrile group is 1.13(2) A, slightly shorter than the bonding group.
This is easily understood by the fact that the bonding of the CN group
involves a degree of n—backbonding which weakens the CEN bond, thus
making it longer. The Mo-Mo bond length of 2.131(2) A is longer than
those without an axial ligand. A plotting diagram of (4) is depicted in Figure
19 and a comparison of Mo-Mo bond lengths for related structures is listed
in Table 9.

The bonding in this structure involves a highly bent nitrile group
which is unusual. This is the smallest angle within the four dimolybdenum
compounds that we have characterized with polynitrile ligands. Other nitrile

II, II

Table 9. Structural parameters for a Mo; core coordinated to non-linear

 

 

nitriles.

Compd. Mo-N Mo-Mo <Mo-N-C (°) <Mo-Mo-N(°)
distance A distance A

(1) 2.630(6) 2.113(1) 147.5(5),163.5(5) 175.50(13)

(2) 2.531(8) 2.127(2) 140.1(6) 166.6(2)

(3) 254(7) 2.142( 13) 155.8(59) 174.1(16)

(4) 2.559(13) 2.131(2) 122.1( 12) 168.7(3)

 

Ml

Mt

M(

Me

Me

M0

N2

N3

E

67

Table 10. Selected bond distances (A) and bond angles (°) for
{[Moz(02CCMe3)3(NC)2C(CN)CCONH)]°8CH2C12}.. (3).

 

 

Bond Distances
A B A-B (A) A B A-B (A)
M01 M62 214(1) Mol N1 254(7)
Mol N7 199(7) M62 03 203(9)
M01 04 200(10) M62 02 206(6)
M01 01 212(6) M62 05 212(8)
M01 06 216(7) M62 07 216(7)
M62 N6 253(8) 010 C5 1.45(10)
N1 C1 123(11) C1 C2 1.28(14)
N2 C3 1.11(12) C2 C3 1.59(14)
N3 C5 1.32(10) C2 C4 1.53(15)
N4 C6 1.13(14) C4 C5 1.35(13)

 

Table 1

A_
M02
02

03

05
M01

N7

M02

N7

03

Table 11. (Cont’d)

68

 

 

Bond Angles
A B C A-B-C (°) A B C A-B-C (°)
M02 M01 N1 174.1(16) C1 N1 Mol 155.8(59)
02 M02 M01 92.0(18) C2 C1 N1 165.1(93)
03 M02 M01 89.4(27) M03 N4 C6 158.0(90)
05 M02 M01 89.7(22) N4 C6 C4 157.4( 100)
M01 M02 07 91.7(19) N2 C3 C2 173.9( 100)
N7 M01 M02 92.9(20) C1 C2 C3 1 16.8(90)
04 M01 M02 93.0(29) C3 C2 C4 1 l 1.1(84)
M02 M01 06 90. 1 ( 17) C1 C2 C4 129.8(98)
N7 M01 01 174.3(26) C5 C4 C2 1 14.4(89)
04 M01 06 174.1(34) C5 C4 C6 124.6(96)
03 M02 02 177.6(34) N3 C5 C4 124.6(80)
05 M02 07 173.4(30) C4 C5 010 l 17.8(75)

 

69

. 8cm deem 88a 5 6 3on82228022805232628.98% do as .2 95$..—

 

 

 

Mo-

NitriT

70

   

/
4 —— 4 %
Mo--Mo N__C Mo—M0‘—%

Nitrile bonding mode in ( 3) Nitrile bonding mode in (4)

 

 

 

i \

C
up \\
l N\
Mo-i-Mo Mo—-4 Mo

Other nitrile bonding modes with M0211. H cores

Figure 18. Nitrile bonding modes with a M0211. " core.

Fig

71

Q

 

Figure 19. Pluto representation of (4) clearly shows that the CN groups
bonded to the M02 unit are in an unusual bent mode.

72

ligand bonding modes involving axial interactions exhibit a tilt such as the

one shown in Figure 18. The Mo-N distance of 2.53 A is comparable to

other Mo-N bond lengths involving Mo-Mo bond. The structure shows

disorder in the transformed TCNE (amide) group.

(4) Structural features of [RezCl4(dppm)2]2TCNQ-10THF(5) and
[RezCl4(dppm)2]2DMDCNQI°lOTI-IF (6).

Due to librational disorder, the —CN bond length of the nitrile bonded
directly to the Re-Re core cannot accurately be refined. Normal -CN bond
lengths are in the range of 1.10 - 1.15 A. The CN bond length in compound
(5) is 1.05 A. An electron density map clearly shows that the CN bond
should be longer, which we attribute to a phenomenon called librational
disorder.40 This typically occurs when a light atom involved in a multiple
bond which is attached to a heavy atom, or ,in general, when the normal
spherical model for electron density is not correct. Figure 20 shows the
electron density map of the nitrile group involved in the coordination
bonding with the refined position of the -CN group in the absence of a
libration correction. Refinement was carried out first without bond length
constraints, then with soft constraints to match the CN bond length indicated
by the E-map. A comparison of the refined bond length with bond length
constraints, the distance between peaks centers of the —CN group in the
electron density map, and the result of refinement with librational disorder

correction are listed in Table 12.

Table 12. Librational disorder refinement of the CN bond length in (5).

 

 

Refined bond length length from E-map libration correction
1.05 1.105 1.05

 

 

 

 

73

 

 

 

 

 

Figure 20. Electron density map of nitrile region in (5).

74

.Amv OZUHNAEAEBNOMH we 2223:8898 28an 38.65. AN Paw—h

 

75

.Gv

525252822382

 

MO

2093:0852

383:0

_aaeefi

.NN

v.53...—

ex
a 1
ho
on]
me

has

TC.
tem
ind:
C011
are

No

76

C. Charge Transfer and Infrared Spectral Data.

Neutral TCNQ shows only one CN stretching band but TCNQ'l gives
rise to an Alg and a EU mode when interactions are present that lower the
symmetry. Also when the molecule engages in a trans-p. bonding mode, the
D4,, symmetry is broken and there are two modes. The dangling CN group
exhibits a higher frequency, while the o-bonding CN group typically shows
a lower frequency due to backbonding from the metal. It should be noted,
however, that when TCN Q bonds to metal centers with no backbonding and
only o—donation, the stretching frequency of the CN bond shifts to higher
energies due to electron donation from the lone pair of electrons on N which
has antibonding character with respect to the CEN bond. The IR spectrum
of (5) shows that the TCNQ moiety in the rhenium structure is not a typical
TCNQ“) anion radical, but the SQUID measurement shows that, at room
temperature the moment is 3.2 B.M., After subtracting a temperature
independent paramagnetic (TIP) contribution, the moment is about 2.3 B.M.
comparable to the value of 2.47 for two non-interacting S = 1/2 spins. There
are two possibilities for this scenario: (1) a doubly reduced TCNQZ' linking
two one-electron oxidized R625+ units and (2) a one electron reduced
TCNQ'“ anion radical linking mixed valence Reg5+ and Rez4+ species. In the
first case, the unpaired spins in the dirhenium centers would be isolated by
the diamagnetic TCNQZ' bridge. If this is true, then the average oxidation
for the Re) core is +5, this may be reasonable considering the fact that the
dihedral angle of the P—Re—Re-P is about 14° large than the 3° in the Re?“
core found in Re2C15(dppm)2, but smaller than the 30° of a Reg4+ core, such
as Re4(dppm)2C13(CNR)2+. It could be considered as a [Re2](+)-TCNQ(1-)-
[Re2](0) and [Re2](+)-TCNQ(2-)-[Re2](+)equilibrium, in which the

77

TCNQ(l-) bridge provides an AF interaction pathway, while TCNQ(2-) with
a higher LUMO would most likely isolate the two Re;5+ units.

4. Summary

Metal-metal bonded building blocks are capable of being used to build
extended arrays with polynitrile ligands which provide a different
architecture than products with 3d transition metal cations. Such structural
possibilities provide a different approach for building supramolecular
structures as compared to the more strict limitations of 3d transition metals.
In addition, their rich redox chemistry makes them good candidates for the

design of electronic and optical devices for energy storage and conversion.

10.

11.

12.

13.

78

References

Cotton, F. A.; Walton, R. A. Multiple Bonds between Metal Atoms;
2nd ed. 1993; Oxford University Press, New York.

Stephenson, T. A.; Wilkinson G. J. Inorg. Nucl. Chem. 1966, 28, 2285
Cotton, F. A.; Pedersen, E. Inorg. Chem. 1975, 14, 388

Telser, J .; Drago R. S. Inorg. Chem. 1984, 23, 3114

Cotton, F. A.; Kim, Y.; Ren, T. Polyhedron 1993, 12 607

Cotton, F. A.; Kim, Y.; Ren, T. Inorg. Chem. 1992, 31, 2723.

Cotton, F. A.; Kim, Y.; Ren, T., Inorg. Chem. 1992, 31, 2608.

Li, D.-Q.; Huckett, S. C.; Frankcom, T.; Paffett, M. T.; Farr, J. D.;
Hawley, M. E.; Gottesfeld, S.; Thompson, J. D.; Burns, C. J.; and
Swanson, B. I. in "Supramolecular Architecture : Synthetic Comtrol
in Thin Films and Solids" ACS Symp. Ser. 499, Ed., Bein, T.
American Chemistry Society, 1992, Ch. 4, pp33-45

McCarley, R. E.; Templeton, J. L.; Clbum, T. J.; Katovic, V. and
Hoxmeier, R. J. in “Inorganic Compounds with Unusual Properties”.

Adv. Chem.Ser. 1975, 150, 318

Mandon, D.; Giraudon, J. M.; Toupet, L.; Sala-pala, J .; Guerchais, J.
E. J. Am. Chem. Soc. 1987, 109, 3490

Giraudon, J. M.; Sala-Pal, J .; Guerchais, J. E.; and Toupet, L. Inorg.
Chem, 1991, 30, 891

Felthouse, T. R.; Dong, T.-Y.; Hendrickson, D. N.; Shieh, H.-S. and
Thompson, M. J. Am. Chem. Soc 1986, 108, 8201

Cogne A.; Grand A.; Rey P.; and Subra R. J. Am. Chem. Soc. 1989,
111, 3230

 

19
20

21

22.

23.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

79

Handa, M.; Sono, H.; Kasamatsu, K.; Kasuga, K.; Mikuriya, M.;
Ikenoue, S. Chem. Lett. 1992, 453.

TCNQ = 7,7’,8,8’-Tetracyano-quinodimethane, TCNE =
tetracyanoethylene, DCNQI = MN -dicyanoquinonediimine, DM-
DCNQI = 2,5-dimethyl-N,N’ -dicyanoquinone—diimine.

(a) Cotton, F. A.; Kim, Y. J. Am. Chem. Soc. 1993, 115, 8511. (b)
Cotton, F. A.; Kim,Y.; Lu, J. Inorg. Chim. Acta 1994, 221, 1.

(a) Huckett, S. C.; Arrington, C. A.; Burns, C. J.; Clark, D. L.;
Swanson, B. I. Synt. Met. 1991, 41-43, 2769. (b) Li, D.; Huckett, S.
C.; Frankcom, T.; Paffett, M. T.; Farr, J. D.; Hawley, M. E.;
Gottesfeld, S.; Thompson, J. D.; Burns, C. J.; Swanson, B. I. in
Supramolecular Architecture: Synthetic Control in Thin Films and
Solids ACS Symp. Ser. 499, (Ed: Bein, T.) American Chemical
Society, 1992, Ch. 4, 33.

Stephenson, T. A.; Bannister, C.; Wilkinson, G., J. Chem. Soc., 1964,
2538.

Cotton, F. A. and Norman, J. G. J. Coord. Chem, 1971, 1, 161.
Aumfiller, A.; Hiinig, S. Liebigs Ann. Chem. 1986, 142.

Barder, T. J .; Cotton, F. A.; Dunbar, K. R.; Powell, G. L.; Schwotzer,
W. and Walton, R. A. Inorg. Chem, 1985, 24, 2550.

SHELXTL 5.0 suites (XPREP, XS, XL and XP) Structure solution

and refinement package, Siemens Energy & Automation, Inc.
Analytical Instrumentation.

DIRDIF84 - Beurskens, P.T. — Direct Methods for Difference
Structures; Technical Report 1984/1, Crystallography Laboratory,
Toemooiveld, 6525 Ed Nijmegen, Netherlands.

TEXSAN-TEXRAY Structure Analysis Package, Molecular Structure

 

to

2?

32

37.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

80

Corporation, 1985

SADABS — Sheldrick, G. M. — Area Detector Absorption (and other)
Correction Program. 1998. University of Gttingen, Germany.

Manriquez, J. M.; Yee, G. T.; McLean, S.; Epstein, A. J .; Miller, J. S.
Science 1991, 252, 1415

Prince, M. and Homyak, J. Chem. Comm. 1966, I4, 455.

Cotton, F. A. and Kim, Y. J. Am. Chem. Soc. 1993, 115, 8511.
Cotton, F. A. and Haefner S., Personal communication.

Cotton, F. A. and Norman, J. G. Jr., J. Coord. Chem, 1971, I, 161.

Handa, M.; Yamada, K.; Nakao, T.; Kasuga, K.; Mikuriya, M. and
Kotera, T., Chem. Lett., 1993, 1969.

Cotton, F. A.; Feng, X., Kibala, P. and Matusz, J M. Am. Chem. Soc.,
1988, 110, 2807.

Buhro, W. E.; Chisholm, M. H. J.; Martin, D.; Huffman, J. C.;
Folting, K. and Sterib, W. E. J. Am. Chem. Soc., 1989, III, 8149.

Cotton, F, A.; Kim, Y.; Lu, J ., Inorg. Chim. Acta 1994, 221, 1.
SPARTAN V4.0, Wavefunction, Inc

(a) Aumliller, A.; Erk, P.; Klebe, G.; Htinig, S.; von Schlitz, J. U.; and
Werner, H.-P. Angew. Chem. Int. Ed. Engl. 1986, 25, 740. (b) Oshio,
H. Inorg. Chem. 1993, 32, 4123. (c) Kato, R.; Kobayashi, H.;
Kobayashi, A. J. Am. Chem. Soc., 1989, III, 5224. (d) M. A. S.
Aquino, F. L. Lee, E. J. Gabe, C. Bensimon, J. E. Greedan, and R. J.
Crutchley, J. Am. Chem. Soc., 1992, 114, 5130.

Aumfiller, A.; Erk, P.; Hi1nig, S.; Hadicke, K.; Peters, H.; von Schlitz,
J. U. Chem. Ber. 1991, 124, 2001.

38.

39.

40.

81

Cotton, F. A. and Norman, J. G. Jr. J. Coord. Chem, 1971, 1, 161.

Handa, M.; Yamada, K.; Nakao, T.; Kasuga, K.; Mikuriya, M. and
Kotera, T. Chem. Lett., 1993, 1969.

Braga, D. and Koetzle, T. F. Chem. Commun., 1987, 144.

CHAPTER III

PREPARATION, STRUCTURAL
CHARACTERIZATION AND PROPERTIES OF
BINARY TRANSITION METAL TCNQ
MATERIALS

82

83

1. Introduction

The use of soluble transition metal coordination complexes as
precursors to materials is a rapidly expanding area of inorganic chemistry,
rich in structural diversity and potential applications."4 Topological
arguments that take into consideration the geometrical preferences of the
metal and ligand building blocks have allowed chemists to design new
molecular solids, including novel porous,2 magnetic3 and conducting5
materials. Among the compounds in the latter category are charge-transfer
solids composed of metal complexes and the radical forms of the
organonitrile acceptors depicted in Figure 23.4’5 The most well-investigated
of these hybrid inorganic/organic materials are the binary, mixed-valence
compounds Cu(2,5-R2-DCNQI)2(DCNQI = dicyanoquinodiimine) (Figure
23 ).5’6 These metallopolymers exhibit unusually high, anisotropic
conductivities that are attributed, not only to the columns of stacked DCNQI
radicals formed by interpenetrating networks, but to the interaction of the d
orbitals of the metal with the p orbitals of the diimine bridge.

 

 

N N
NC CN
CH3
TCNE DM-DCNQI TCNQ

 

 

(a) (b) (°)

 

Figure 23. Organonitrile acceptors that behave as ligands in their reduced
forms.

84

In considering the remarkable properties exhibited by copper DCNQI
polymers, we were reminded of a class of intriguing metal complexes of 7,7-
8,8-tetracyanoquinodimethane (TCN Q) first reported in 1962. From the
perspective of metal complexation, the main difference between DCNQI and
TCNQ is that TCNQ is capable of behaving as a tetradentate as well as a
bidentate bridging ligand which allows it to form products with both 1:1 and
1:2 metal/TCNQ ratios.“’6‘7 The 1:1 materials of Ag and Cu are the subject
of intense interest due to the fact that crystalline films of these compounds
undergo electric-field induced bi-stable switching under certain conditions.8'
22 In spite of extensive efforts to understand these materials, however, this
phenomenon continues to be the source of debate in the literature. One of the
crucial issues to be resolved is the source of conflicting data regarding the
electrical behavior of Cu(TCNQ). Independent findings of several groups
suggest that the physical appearance of the crystalline films is highly
dependent on film preparation conditions.15 "9’20 22

In light of the high interest in such materials, we find it to be curious
that no X-ray data for Cu(TCNQ) or related binary materials with 3d
elements have been determined. In fact, no full-scale investigation of binary
metal compounds of TCNQ has been undertaken, although materials
formulated as M(TCNQ)2(HZO)2 were recognized as being easily prepared,
tractable solids as early as 1962.6’23 As part of a comprehensive investigation
of the magnetic and conducting properties of M/TCNQ materials and in an
effort to establish structural/property relationships, we have prepared a
variety of crystalline phases of general formulae M(TCNQ) and M(TCNQ)2.
One synthetic strategy to such materials is the combination of metal cations
with organic acceptor anions such as TCNQ" (7,7,8,8-

tetr'acyanoquinodimethane),5 TCNE” (tetracyanoethylene)6 and DCNQI"

85

(N,N’-dicyanoquinonediimine).7 Unfortunately, polymeric materials of this
type are notoriously difficult to crystallize, a fact that has hampered research
efforts aimed at structure/property correlations. We have found, however,
that with the proper selection of starting materials and solvents, it is possible
to prepare crystalline, neutral framework compounds with metal ions
bridged only by ligands derived from TCNQ". During the course of these
studies, we became aware of the role of the solvent (e.g., methanol, water,
acetonitrile) in determining the structures and therefore physical properties
of metal/'1‘ CNQ materials.

In this chapter the syntheses and characterization of TCNQ
compounds with the metal ions Mn(II), Ni(II), Zn(II) and Cu(I) are
discribed. Representative single-crystal X-ray structures, powder X-ray
data, infrared spectral properties and properties are discussed in light of the
TCNQ packing motif that is adopted in the solid-state. The structures of
["BuaNHTCNQ] and ["BuaN][TCNQF4] as well as the salt
[Ni(DMSO)6][TCNQ]3 (all with outersphere TCNQ" anions) are also
presented as a comparison to the compounds in which TCNQ'l acts as a
ligand. Structural studies on Ni(TCNQ)2(HzO), which bears a strong
resemblance to other 3d metal TCN Q hydrate compounds such as
Mn(TCNQ)2(HzO)2, is also discussed.

2. Experimental Section

A. Synthesis
(1) Starting materials and reaction procedures.

All reactions were carried out under an argon atmosphere by the use
of standard Schlenk-line techniques. Methanol was pre-dried over 4 A

molecular sieves and freshly distilled under a nitrogen atmosphere from

86

Mg(OCH3)2. Acetonitrile was dried over 3 A molecular sieves and freshly
distilled under an atmosphere of N2 prior to use. TCNQ was purchased from
TCI Chemical Company. The salt ["Bu.tN][TCNQ]23a was prepared by
metathesis of LiTCNQ with "Buer in boiling water followed by
recrystallization from CH2C12.2‘““b Crystalline LiTCNQ was prepared by
combining TCNQ and 3 equivalents of LiI in hot acetonitrile and allowing
the mixture to cool for 12 hours. MnX2 (X = Cl, Br) and CuI starting
materials were obtained from commercial sources and used without further
purification. [Cu(MeCN)4][BF4],25 [Mn(MeCN)4][BF4]2 and
Ni(MeCN)4(BF4)226 were prepared as described in the literature.

(2) Bulk syntheses of Cu(TCNQ) phase I (7).

(a) Method A.

CuI (0.95 g, 5 mmol) was added to 150 mL of a hot acetonitrile
solution of TCNQ and stirred for 3 minutes.23a The dark blue
microcrystalline product that formed was separated from a brown solution
by filtration, washed with copious amounts of acetonitrile and dried in
vacuo; yield 1.22 g, 91%. Anal. Calcd for C12H4N4Cu: C, 53.83; H, 1.51; N,
20.93. Found: C, 53.46; H, 1.39; N, 20.57. Characteristic IR data (Nujol
mull, KBr plates, cm'l): v(CN), 2199s,br; 2172sh; 5(C-H) 825s.

(b) Method B.

A quantity of ["BuaN][TCNQ] (0.500 g, 1.12 mmol) was dissolved in
50 mL of acetonitrile to give a green solution which was added to a solution
of [Cu(MeCN)4][BF4] (0.352 g, 1.12 mmol) dissolved in 20 mL of
acetonitrile. Upon mixing, a color change to dark blue occurred with the
deposition of a dark blue microcrystalline solid. The solid was collected by
filtration, washed with acetonitrile and dried in vacuo to give 0.265 g (88%
yield) of a dark blue solid whose X—ray pattern is identical to the product

87

obtained in method A.
(3) Bulk synthesis of Cu(TCNQ) phase II (8).

A quantity of acetonitrile (20 mL) was added to 0.300 g of phase I.
The resulting suspension and light green solution was stirred at room
temperature for 4 days after which time the solvent was evaporated to yield a
dark blue solid. If the solution is refluxed for 3 hours, the conversion
process occurs much faster (within 30 minutes). Both methods lead to
virtually quantitative conversion of phase I to phase II as judged by X-ray
powder diffraction and infrared spectroscopy. Anal. Calcd for ClzlrlaNaCu:
C, 53.83; H, 1.51; N, 20.93. Found: C, 53.68; H, 1.33; N, 20.85.
Characteristic IR data (Nujol mull, KBr plates, cm"): v(CN), 2211s; 2172s;
5(C-H) 825s.

(4) Thin films of Cu(TCNQ) .

Cu(TCNQ) thin films were prepared by a method similar to those
described in the literature.8 Rectangular pieces of copper foil were cleaned
mechanically and then chemically with a dilute H2804 solution and washed
with water followed by acetone in an N 2 atmosphere. The dipping
experiments were conducted in an N; purged glove bag in a two-
compartment, fritted reaction vessel that contained a saturated CH3CN
solution of TCNQ. To avoid contaminating the Cu(TCNQ) films with solid
TCNQ, the copper pieces were placed in a glass thirnble with a coarse glass
frit at one end and carefully arranged such that copper surfaces did not
touch. The thimble was lowered into the TCNQ saturated solution and the
reaction vessel was sealed with a ground glass stopper. Upon contact with
the solution, black-purple films were observed to appear on the exposed
copper surfaces. At periodic, timed intervals the reaction vessel was opened

to retrieve one of the copper pieces which was then washed with acetonitrile,

88

dried and placed in a vial under anaerobic conditions for later analysis.
(5) Bulk synthesis of Mn(TCNQ)2(MeOH)2 (9).

A MeOH solution (20 mL) of ["BuaN][TCNQ] (0.447 g, 1.00 mmol)
was slowly added to a MeOH solution (10 mL) of [Mn(MeCN)4][BF4]2
(0.197 g, 0.5 mmol). The resulting reaction solution was stirred for 20 min
to give a blue precipitate that was collected by filtration, washed with
methanol followed by diethyl ether and dried in vacuo. Yield: 0.187 g, 71%.
Analytical data: Mn (x = 2) Calcd for C26H16MnNgOzz C, 59.21; H, 3.06; N,
21.25. Found: C, 59.53; H, 2.41; N, 21.71. Thermogravimetric
measurements indicate that the samples are stable with approximately two
equivalents of coordinated MeOH.

(6) Bulk synthesis of Mn(TCNQ)2(HzO)2 (10).

An aqueous solution (20 mL) of LiTCNQ (0.211 g, 1.0 mmol) was
added to an aqueous solution (20 mL) of MnClz (0.063 g, 0.5 mmol), and
stirred at room temperature for 30 min. The resulting blue precipitate was
collected by filtration, washed with water followed by ethanol and diethyl
ether, and finally dried in vacuo. Yield: 0.242 g, 94%. Elemental analyses:
Calcd for C24H12MnN302: C, 57.73; H, 2.40; N, 22.44. Found: C, 57.76; H,
2.20; N, 22.47.

B. Single Crystal Growth.
(1) Cu(TCNQ) phase I and phase II (7), (8).

Tiny crystals of both Cu(TCNQ) phases were grown by a technique
that involves the use of a three-compartment fritted cell. In the outer two
compartments, small Soxhlet thimbles containing the starting materials CuI
and TCNQ were suspended on a wire and immersed in the solution (Figure

24). The entire cell was placed in a water bath and gently heated to 60 °C

89

for 3-5 days to expedite mixing of the reactants in the center compartment.
Within days, the solution gradually turned green signifying the production of
TCNQ'I, and tiny needles, admixed with platelets, began to form in the
middle compartment. The needles (phase I) were harvested immediately
because they eventually re-dissolve to produce crystals of phase II (the
platelet habit).

(2) [MD(TCNQ)(TCNQ'TCNQ)0,5(MBOH)2]... (9).

Purple needles of [Mn(TCNQ)(TCNQ-TCNQ)0,5(MeOH)2].. (9) were
grown by slow diffusion of a methanolic solution of [Mn(MeCN)4][BF4]2
into a mixture of MeCN/MeOH (2: 1) containing ["BuaN][TCNQ] in a 3 mm
diameter sealed pyrex tube.

(3) [Mn(TCNQ)2(H20)2Lo (10)-

Red-purple monoclinic prisms of [Mn(TCNQ)2(HzO)2].. (10) were
grown by slow diffusion of aqueous solutions of mm; and LiTCNQ in 3
mm diameter sealed pyrex tube.

(4) ["Bu..N][TCNQ] (11)-

Purple prismatic crystals of ["Bu4N][TCNQ] were grown by slow
evaporation of a saturated methanol solution of ["BuaN][TCNQ].
(5) ["BlelTCNQFtl (12).

Blue green platelet-shaped crystals of ["Bu4][TCNQF4] were grown by
slow diffusion of water into a methanol solution of ["Bu4N][TCNQF4].
(6) [Zn(HzO).(n‘-TCNQ).-TCNQ-2MeCN].. (13).

Purple prisms of [Zn(H20),(n‘-TCNQ)2-TCNQ-2MeCN].. were
grown by a slow reaction of Zn metal pellets with a 1:1 acetonitrile/water
solution of TCNQ.

(7) [Ni(DM50ltllTCNle (14)-
Purple platelets of [Ni(DMSO)6][TCNQ]3 were grown by the

90

 

 

 

 

 

 

 

 

2 CuI (s) + TCNQ“ > 2 Cu(TCNQ) + 12
CuI (s) in /© 0 Saturated
a small paper V O 0 solution of
thimble used 55 §§EEEE frit frit TCCNDQ TCNQo in
inaSoxhlet 'llii' \ I 9° MeCN
extractor. @- I <— O <—ooo

 

 

 

 

 

1

Phase 1 \ ~ 0.02 x 0.02 x 0.10 mm
Phase II I ~ 0.01 x 0.10 x 0.10 mm

Figure 24. Crystal growing setup for two Cu(TCN Q) phases.

91

diffusion of acetonitrile into a 1:1 DMSO/HZO solution of Ni(TCNQ)2. The
solid Ni(TCNQ); was prepared from the reaction of [Ni(MeCN)6][BF4]2 with
2 equivalents of ["Bu4N][TCNQ].
(8) [Ni(TCNQ)2(H20)l~ (15)-

Purple prismatic crystals of [Ni(TCNQ)2(HzO)].. were grown by
layering an aqueous solution of [Ni(MeCN)6][BF4]2 with an aqueous
solution of LiTCNQ. A small amount of acetonitrile was used as a buffer

solution to separate the reactant layers.

C. Physical Measurements.

Single crystal X-ray data were obtained on a Siemens
(Bruker) SMART 1K CCD area detector instrument or a Nicolet
P3N diffractometer. Both are equipped with sealed tube Mo
anodes. Powder X-ray data were collected using a Rigaku RU200B
X-ray powder diffractometer with Cu KOL radiation. The
variable temperature magnetic susceptibility data were collected in the
range 5-300 K using a Quantum Design, Model MPMS-5
SQUID magnetometer housed in the Physics & Astronomy Department
at Michigan State University. Infrared spectra were recorded in Nujol
on KBr plates using a Nicolet IR/42 FT-IR spectrometer.
X-ray Photoelectron Spectral (XPS) data were obtained on a PHI
5400 instrument with a Mg source in the Composite Materials and
Structures Center at Michigan State University. Thermogravimetric
analyses were performed between 25-1000 °C on a SHIMADZU,
TGA-SO. Scanning electron microscopy measurements were performed
on a JEOL 6400V instrument with a LaB6 gun housed in the Center
for Electron Optics at Michigan State University.

92

(1) X-ray crystallography.
(a) Cu(TCNQ) phase I (7 ) .

A dark purple needle crystal of dimensions 0.25 x 0.02 x 0.02 mm
was mounted on the end of a glass fiber with silicone grease. A hemisphere
of data was collected at -100(2) °C on a Siemens SMART 1K CCD area
detector diffractometer by Dr. Victor Young at the University of Minnesota.
The frame scan-width was 0.3 in (1), and the exposure time was 60 sec/frame.
No cell information was available before data collection due to the weak
diffraction. Post data collection processing results suggested that the crystal
belongs to the tetragonal 4/m Laue group. The data were integrated with the
cell parameters a = b = 11.266(2) A, c = 3.8878(8) A, a: )6: y: 90°,
V = 493.47(17) A3. After numerous failures to solve the structure in the 4/m
or m crystal symmetry group, the structure was ultimately solved in the
lower symmetry monoclinic space group Pn with the cell parameters
a = 3.8878(8) A, b = 11.266(2) A, c = 11.266(2) A, ,9: 90.00(3)°. The
structural model was refined as disordered and pseudo-merohedrally
twinned. The 3x3 twinning matrix [1 O 0, 0 0 1, 0 -1 0] was used to model
the pseudo 4-fold rotation. A summary of data collection and refinement
results is listed in Table 13.

(b) Cu(TCNQ) phase II (8) .

A tiny, square platelet crystal of dimensions 0.13 x 0.13 x 0.01 mm
was mounted on the tip of a glass fiber with the use of silicone grease. A
full sphere of data was collected at -100:l:(2) °C on a Bruker (Siemens)
SMART 1K CCD area detector diffractometer. Due to the small size of the
crystal and twinning problems, cell parameters were not refined before data
collection. Data were collected in 0.3° (1)-scans with an exposure time of 90

sec/frame (total time 69 hours). Strong reflections (253) were extracted with

93

Table 13. Summary of key crystallographic parameters for Cu(TCNQ)
phase I (7) and phase II (8).

 

 

Corruround Name (7) (8)

Formula CUN4C12H4 CLIN4C12H4

Morphology Needle Square Platelet

Crystal size (mm3) 0.25 x 0.02 x 0.02 0.13 x 0.13 x 0.01

Crystal system Monoclinic Monoclinic

Space group P n P 2/n

a, A 3.8878(8) 5.3337(8)

b, A 11.266(2) 5.3312(8)

c, A 11.266(2) 18.875(3)

a, deg 90 90

[3, deg 90.00(3) 94.036(3)

7, deg 90 90 ,

v, A3 493.5(2) 535.3804)

Z 2 2

T, K 173 173

Radiation " Mo Koe Mo KCX.

pcalc, g/crn3 1.802 1.661

11, mm" 2.188 2.017

Collected! unique data 1995/1 11 1 4456/955

Indexing method SMART TWINDX

Solution and refinement Pseudo merohedral Quadruplet rotational

method twinning and twinning,

disordered, isotropically anisotropically refined.
refined.

R1(wR2)c 0.234(0.485) 0.159(0.335)

goreI 2.576 1.060

 

“Graphite monochromated. b1 > 20(1). °RI = 2||F,|-|F,||/2|F,|; wRZ =
[2[w(F02-Fc2)2]/Z[w(F02)2]]"2. dGoodnees-of Fit (0015) = s = [Z[w(F02-
Fc2)2]/(n-p)]m, n is the total number of reflections and p is the total

parameters refined.

94

the SMART index routine, but this failed to give a reasonable cell. The
indexing solution was obtained by the use of programs in the TWINNING
package written by R. Sparks.27 (The indexing process is discussed in detail
in Chapter V). Four monoclinic cells were indexed with a = 5.3337(8) A, b
= 5.3312(8) A, c = 18.875(3) A, a: 94.036(3)° and V = 535.3804) A3. The
twinning law matrix [0 —1 0, 1 0 0, 0.25 0.25 1] transforms the component A
to the other three components by successive rotations along (0 0 1) axis.
Because the TWHKL module in the TWINNING package only handles two
component twins, a revised version of TWHKL was used to process the four
component twin data. The refinement is not stable because of the severity of
the twinning. The data for solving the structure in the space group P2/n
were integrated with one component only. The final residuals for this
treatment were R1: 0.159 and wR2= 0.345 for 4456 reflections with I >
20(1). A summary of data collection parameters is listed in Table 13.

(C) [Mn(TCNQ)(TCNQ'TCNQ)0.5(MCOH)2]n (9)-

A long needle crystal of approximate size 0.40 x 0.04 x 0.04 mm was
mounted on a glass fiber and secured with silicone grease. A hemisphere of
data was collected on a Siemens SMART CCD area detector instrument at
-140(1)°C by Dr. D. Powell in University of Wisconsin. (This was prior to
the time that Michigan State Univeristy had a CCD instrument). A summary
of data collection parameters for the compound is given in Table 9. The data
were solved and refined in SHELXTL V.5.0 by full-matrix least-squares on
F2. The original cell and systematic absences suggested an orthorhombic C-
centered crystal system, but the structure was ultimately solved as a
33.2/66.8 twin in C2/c (twinning law [1 0 0, 0 1 0, 0 0 —1]). The disordered
MeOH group was modeled in two orientations with occupancies of 0.535
(CZM’) and 0.465 (C2M”). All non-hydrogen atoms were located and

95

Table 14. Crystal data and structure refinement parameters for [Mn(TCNQ-
TCNQ)0.5(TCNQ)(MCOH)21~(9) and [Mn(TCNQ)2(H20)2]...(10).

 

 

Identification code (9) (10)
Empirical formula C26 H16 N3 02 Mn C24 H12 Mn N3 02
Formula weight 527.41 499.36
Temperature 133(2) K 133(2) K
Wavelength 0.71073 A 0.71073 A
Crystal system Monoclinic Monoclinic
Space group C2/c C2/c

Unit cell dimensions

a = 14.4910(6) A
b = 27.4207(10) A
C: 13.1146(5) A
a: 90°

a = 17.8904(6) A
b = 13.7183(4) A
c = 12.6348(4) A
a: 90°

,6 = 90.0049(8)° ,6 = 131.7610(10)°
y = 90° y = 90°
Volume, A3 5211.1(3) 23l3.06(13)
Z 8 4
rcalc, Mg/m3 1.344 1.434
Absorption coefficient, mm" 0.545 0.610
F(000) 2152 1012
Crystal size, mm3 0.40 x 0.04 x 0.04 0.40 x 0.36 x 0.30
Reflections collected 10510 6944
Absorption correction Ill-Scan Multi-scan,
SADABS
Refinement method Full-matrix on F2 Full-matrix on I“2
Data/restraints/parameters 4523 / 0 / 337 2710 / 0/ 184
Goodness-of-fit on F2 1.137 1.048
R indices [I>20 0.0359 (0.0742) 0.0386 (0.0872)

R1 (wR2)

0.395 and -0345 e/A3 0.257 and -0427
e/A3

Largest diff. peak and hole

 

Toots: R1 = 2 pr.) - |Fc|| / 2 |F0|, wRZ = (2 {ME} - F52] / >2 [w(F.2)2]}"2.
goodness-of-fit = s = {[w(F,2 - F52] / (n - p)} “2, w = 1 / [02(F02) + (aP)2 +
bP]

96

refined with anisotropic thermal parameters, and hydrogen atoms were
located from difference Fourier maps and refined with isotropic thermal
parameters. Hydrogen atoms were assigned to ideal positions. The final
refinement gave R]: 0.0361 (wR2= 0.0772) for 4149 reflections with I > 20
and R]: 0.0414 (wR2= 0.0847) for 4532 data and a GOF value of 1.146.

(d) [Mn(TCNQ)2(H20)2lo (10)-

A blue crystal of approximate dimensions 0.40 x 0.36 x 0.30 mm was
mounted on a glass fiber and secured with Dow coming grease. Diffraction
data were collected on a Siemens SMART CCD area detector instrument at
-140(2)°C. Preliminary cell parameters indicated a monoclinic system from
the refinement of 26 reflections taken from 45 frames with 10 second
exposure times. The initial cell parameters were a = 17.839(10) A,
b = 13.692(7) A, c = 12.597(7) A, fl= 131.740(10)°, v = 2296(4) A3. The
final cell parameters were obtained from the refinement of the XYZ
centroids of 3816 reflections with I > 100' from a total of 7,094 integrated
reflections. Systematic absences indicate that the crystal belongs to a
monoclinic system, space group C2/c (#15), with a = 17.8904(6) A,
b = 13.7183(4) A, c = 12.6348(4) A, a: 90°, ,9: 131.7610(10)°, 7: 90°,
V = 23l3.06(13) A3. A summary of data collection parameters is given in
Table 14. The data were solved and refined in SHELXTL V. 5.0 by full-
matrix least-squares on F2. All non-hydrogen atoms were located and
refined with anisotropic thermal parameters, and hydrogen atoms were
located from difference Fourier maps and refined with isotropic thermal
parameters.

(9) [”BueNllTCNQl (11)-
A purple needle crystal of approximate dimensions

0.21 x0.16 x0.10 mm was mounted on a glass fiber and secured with

97

grease. Diffraction data were collected on a Siemens SMART CCD area
detector instrument at —100(2)°C. Preliminary cell parameters were
obtained from the refinement of 63 reflections from 60 frames with
10 second exposure times. Initial unit cell parameters were a = 9.300(2) A,
b = 19.343(5) A, c = 15.385(5) A, p: 100.990(18) °, v = 2716.74(132) A3.
A hemisphere of data was collected at a scan-width of 0.3° in (1) at 30
sec/frame. Final cell parameters and an orientation matrix were obtained
from the refinement of the XYZ centroids of 3296 reflections with I > 100
with a = 9.3028(7) A, b = 19.3520(14) A, c = 15.388001) A,
p = 100.979(1)°, v = 2719.57(59) A3. A summary of data collection
parameters for compound (11) is given in Table 10. The data were solved
and refined in SHELXTL V. 5.04 by full-matrix least-squares on F2. All
non-hydrogen atoms were located and refined with anisotropic thermal
parameters, and hydrogen atoms were fixed in the idealized positions. The
refinement led to RI: 0.0638 (wR2= 0.111) for 3259 reflections with I > 20
and R]: 0.1497 (wR2= 0.1357) for all 6327 data against 298 parameters
with a goodness-of-fit of 0.973. The highest residual peak in the electron
difference map is 0.18 e/A3.
(f) ["BueNllTCNQFcl (12)-

A thin, blue green platelet crystal of approximate dimensions
0.20 x 0.08 x 0.01 mm was mounted on a glass fiber and secured with Dow
Corning grease. Diffraction data were collected on a Siemens SMART CCD
platform diffractometer at —100(2)°C. Preliminary 10 sec/frames indicated a
triclinic system with a = 9.630 A, b = 14.934 A, c = 21.305 A, a: 107.445°,
,6: 93.422°, y: 104.716°, V = 2800 A3. A hemisphere of data was collected
at a scan width of 0.3° in (0. Final cell parameters and an orientation matrix

were obtained from the refinement of XYZ centroids of 2,850 strong

98

reflections with I/ 0‘ > 10 from a total of 15,276 reflections with a =
9.6335(3) A, b = 14.9574(3) A, c = 21.3615(5) A, a = 107.440(1)°,
,6 = 93.436(2)°, = 104.785(1)°, V = 2808.07(17) A3. Absorption was
corrected for with program SADABS which lead to a transmission factor
between 0.56 and 1. Data were further merged and truncated with XPREP
to .85A resolution to reduce the high Rsigma. The structure was solved in
space group P-I by direct method with subsequent atoms being located from
difference maps and refined in SHELXTL V. 5.10 package by full-matrix
least-squares on F2. All non-hydrogen atoms were located and refined
anisotropically, and hydrogen atoms were fixed in idealized positions. The
final refinement was based on 668 parameters and 13369 reflections, of
which 8918 were unique. The refinement led to R1: 0.0621 and wR2=
0.1221 and a goodness-of —fit 1.075 for reflections with I > 20 and R]:
0.1568 (wR2= 0.1582) for all data. The highest peak in the final difference
map was 0.204 e-IA3. A summary of data collection parameters for
compound (12) is given in Table 15.

(g) {[Zn(H20).](n‘-TCNQ),o2TCNQ-2MeCN}.. (13).

Single crystals of (13) were grown from the reaction of Zn pellets
with TCNQ in a 1:1 MeCN/H20 solution. A brown needle crystal of
approximate dimensions 0.40 x 0.10 x 0.08 mm was secured on the tip of a
glass fiber with Dow Corning silicone grease and mounted on a Siemens
SMART 1K CCD platform diffractometer. Preliminary examination
indicated that the unit cell was triclinic with a = 10.7624(4) A,
b = 11.0007(3) A, c = 12.1897(4) A, a = 70.147(1)°,,3 = 88.086(1)°,
y: 89.784(2)°, V = 1356.61(8) A3 reflections with I > 100(1) from 8,554
reflections. The final unit cell was a = 10.7624(4) A, b = 11.0007(3) A,
c = 12.1897(4) A, a = 70.1470(10)°, 73 = 88.0860(10)°, y = 89.784(2)°,

99

Table 15. Crystal data and structure refinement for ["BusN][TCNQ] (11)

 

 

and ["BunN][TCNQF4] (12).

Identification code (11) (12)
Empirical formula C56 H30 N10 C60 H72 F3 N10
Formula weight 893.30 1085.28
Temperature 173(2) K 173(2) K
Wavelength 0.71073 A 0.71073 A
Crystal system Monoclinic Triclinic
Space group P2(1)/n P T

Unit cell dimensions

Volume A3
Z
Density (calculated)

Absorption coefficient mm'1

F(000)

Crystal size, mm3
Reflections collected
Absorption correction
Refinement method

Data / restraints / parameters

Goodness-of-fit on F2

RI (wR2) W, I>20(I)],
Largest diff. peak and hole

a = 9.3028(7) A

b = 19.3520(14) A
c = 15.388001) A
a: 90°

fl: 100.979(2)°
y: 90°

2719.6(3)

2

1.091 Mg/m3
0.065

972

0.21 x 0.16 x 0.10
16240

Multi-scan, SADABS
Full-matrix on re
6327 / 0 / 298
0.973

0.0638 (0.1111)

a = 9.6335(3) A

b = 14.9574(3) A
c = 21.3615(5) A
a: 107.4400(10)°
,6: 93.437(2)°

y: 104.785(2)°
2808.07(12)

2

1.284 Mg/m3
0.096

1148

0.20 x 0.08): 0.01
13369
Multi-scan, SADABS
Full-matrix on I“2
8918 I 0 / 667
0.950

0.0755 (0.1694)

0.178 and 0242 e/A3 0.486 and -0409 e/A3

 

Note: R1 = 2 ||Fo| - |Fc|| / z |F,|, wR2 = {2: [w(F02 - 1°52] /2 [w(F02)2]}m,
goodness-of-fit = s = {[w(F02 - F32] / (n - p)} “2, w = 1 / [o2(F,2) + (aP)2 +

bP]

100

V = 1356.61(8) A3. The intensities were corrected for beam inhomogeneity,
crystal decay and absorption with the program SADABS which led to Rim =
0.0258 and transmission factors Tm,“ = 0.704, Tm = 1.0. A total of 8,554
reflections was measured with the maximum 20 = 56.36°, of which 4230
were in the category I > 20(1) and 5,669 reflections were unique.

Systematic absences indicated that the system belongs to the space
group P—l (#2). Symmetry and multiply measured reflections were then
averaged with SHELXTL to give values of Rim = 0.0303 and Rsig = 0.0887.
The structure was solved by direct methods. All non-hydrogen atoms were
refined anisotropically by a full matrix least-squares refinement on F2.
Hydrogen atoms were fixed in idealized position and were not refined. The
final residuals were RI: 0.0642 (wR2= 0.1339) for 4230 (I > 20(1))
reflections and R]: 0.0991 (wR2= 0.1492) for all 5969 reflections and 408
parameters. A heavy residual of 2.264 e'/A3 was observed to reside at an
inversion center, which was modeled as a disordered Zn atom (3.3%)
bonding to the —CN group trans to the original position in the framework
dominated by the TCNQ stacks. A summary of the data collection and
refinement results is listed in Table 16.

(h) Structure of [Ni(DMSO)6][TCNQ]3 (14).

A square platelet crystal of approximate dimensions
0.52 x 0.52 x 0.18 mm was secured on the tip of a glass fiber with Dow
Corning silicone grease. The crystal was mounted on a Nicolet (Siemens)
P3N diffractometer equipped with a low temperature device at -108(1)°C.
Least-squares refinement using 20 well centered reflections in the range
15° 5 20 _<_ 25° indicated a triclinic system with a = 7.927(2) A, b =
10.057(2) A, c = l7.141(3) A, a: 84.59(3)°, )8: 82.67(3)°, y: 82.26(3)°,
V= 1338.8(5) A3. The data were collected at —108(1)°C by the (1)-20 scan

101

Table 16. Crystal data and structure refinement for {[Zn(HzO)4](’n1-
TCNQ)2-2TCNQ°2MeCN}.. (13) and [Ni(DMSO)6](TCNQ)3 (14).

 

 

Identification code (13) (14)
Empirical formula C23 H18 N10 02 2110.5 C24 H24 N5 N105 03 S3
Formula weight 559.21 570.03
Temperature 173(2) K 165(2) K
Wavelength 0.71073 A 0.71073 A
Crystal system T riclinic Triclinic
Space group P-l P- 1
Unit cell dimensions a = 10.7624(4) A a = 7.9267(16) A
b = 11.0007(3) A b = 10.057(2) A
c=12.1897(4)A c=17.141(3)A
a: 70.1470(10)° a: 84.59(3)°
,6: 88.0860(10)° ,6: 82.67(3)°
y: 89.784(2)° y: 82.26(3)°
Volume A3 1356.61(8) 1338.8(5)
Z 2 2
Density (calculated) 1.369 Mg/m3 1.414 Mg/m3
Absorption coefficient 0.520 mm'1 0.655 mm'1
F(000) 574 592
Crystal size mm3 0.40 x 0.10 x 0.08 0.52 x 0.52 x 0.15
Reflections collected 8554 51 10
Refinement method Full-matrix on I"2 Full—matrix on F2
Data / restraints / 5969 / 6 / 409 4738 / 22 / 358
parameters
Absorption correction Multi-scan, SADABS w-Scan
Goodness-of-fit on F2 0.992 1.088
RI(wR2) [5“, I>20(I)] 0.0642(0.1375) 0.0683(01281)

Largest diff. peak and 2.276 and-1.003 e/A3 0.523 and-1.122 e/A3
hole

 

Note: R1 = E IIFoI - IF.” / 2 IFOI. sz = {2 ME? - Fc2)21/>3 [w(F.2)°]}‘°.
goodness—of-fit = s = {[w(F02 - F,2)2] / (n - p)} "2, w = 1 / [020702) + (aP)2 +
bP].

102

technique in the range 4° $20350°. A total of 5,110 reflections was
measured, of which 4,738 are unique. The structure was solved by direct
methods and refined by full-matrix least squares on F2. All non-hydrogen
atoms except the disordered SMez groups were refined anisotropically. The
disorder in the SMe2 groups was modeled at ~33% occupancy distributed
among three conformers. The structure was refined to residuals of R1 =
0.0683 (wR2= 0.128) for 3048 reflections (I > 20) and a goodness-of-fit =
1.088.

A summary of key data collection and refinement parameters is
provided in Table 16.

(i) Crystal structure of [Ni(TCNQ)2(H20)].. (15).

Single crystals of (15) were grown by layering an acetonitrile solution
of [Ni(MeCN)6][(BF4)2] with LiTCNQ in H20. A small purple
cube crystal of dimensions 0.080 x 0.060 x 0.060 mm was secured on the tip
of a glass fiber with epoxy. The data were collected at -60(1)°C
on a Siemens P4-CCD four circle diffractometer equipped with Mo rotating
anode source by Dr. Jim Britton at McMaster University, in Ontario,
Canada. Preliminary examination indicated that the crystal system belongs
to the tetragonal system, but the crystal diffracted so weakly. A combination
of \ll-(D scan with scan-widths of 0.3° and 60 sec/frames led to 1,750 data
frames with the maximum 20 = 52.96° and a resolution of 0.8A.

The final cell parameters and an orientation matrix were obtained
from the refinement of XYZ centroids of 1290 reflections with I > 200(1)
from a total of 6,634 reflections. Data integration with tetragonal cell
constraints led to a = 12.242002) A, b = 12.242005) A, c = 3.9067(5) A,
a: 90.0000(23)°, ,6: 90.0000(35)°, y: 90.0000(25)°, V = 585480094) A3
and a total of 6,634 reflections of which 2,158 were in the category I > 20(1)

103

and 1,474 were unique. The intensities were corrected for beam
inhomogeneity, crystal decay and absorption with the program SADABS,
which led to Rim: 0.0454, and transmission factors Tm,“ = 0.742, Tm = 1.0.

After many failures to solve the structure in the 4/m or m
crystal symmetry group, the structure was ultimately solved in the
lower symmetry monoclinic space group Pc with the cell parameters
a : 3.9169(5) A, b = 12.294502), c : 12.300505) A, ,6: 90.130(2)°, due
to probable twining. The structural model was refined as disordered
and pseudo-merohedrally twinned. The 3x3 twinning matrix [1 0 0, 0 -1
0, 0 0 -1] was used to model a two-fold rotational twinning. This modeling
led to RI: 0.0829, and wR2= 0.1789 for 1106 (I > 20' (1)) unique
reflections for 151 parameter, and RI: 0.1781, wR2= 0.2386 for all 1184
data and a goodness-of-fit of 1.050. Absolute configuration testing leads
to a Flack constant of 0.00 which indicates the solution is a correct
configuration. The largest difference peak and hole in the residual
difference map is 0.301 and -0.262 e/A3 respectively. A summary of data
collection and refinement results is listed in Table 17.

(2) Powder X-ray diffraction studies.

Powder XRD methods were used to probe the formation
and conversion of the two phases of Cu(TCNQ). XRD was used to probe
the relationship of M(TCNQ)2L (M = Fe, Co, Ni; L = MeOH, H20)
products to the more crystalline Mn phase for which several single
crystal phases were obtained. The binary metal TCNQ phase formulated
as M(TCNQ); (M = Mn, Fe, Co, Ni) were also subjected to powder
XRD studies. All samples were measured from 5 to 50° in 26
with a diffractometer power of 45 kV and 100 mA.

Table 17. Crystal data and structure refinement for

[Ni(TCNQ)2(H20)]...(15).

 

 

Identification code (15)
Empirical formula C12 H6 N4 Ninja O
Formula weight 251.56
Temperature 213(2) K
Wavelength 1.54178 A

Crystal system Monoclinic

Space group Pc

Unit cell dimensions

Volume A3

Z

Density (calculated)
Absorption coefficient mm-l
F(000)

Crystal size, mm3
Reflections collected
Independent reflections
Absorption correction

Max. and min. transmission
Refinement method

Data / restraints / parameters
Goodness-of-fit on F2

Final R indices [I>20( 1)]
Absolute structure parameter
Largest diff. peak and hole

a = 3.9169(5) A
b : 12.294502) A

: 12.3005(15)A
a: 90°
73: 90.130(2)°
y: 90°
592.3502)
2
1.410 Mg/m3
0.857
256
0.08 x 0.06 x 0.06
4874
2264 (Ram, : 0.0619]
Multi-scan, SADABS
1.00 and 0.69
Full-matrix on F2
2264 I46 / 151
1.050
R] : 0.0829, wR2 : 0.1789
0.00
0.301 and -0.262 e/A3

 

Note: R1 = 2 IIFol - IF." / 2 lat. sz = {2 ME? - F521 I >3 muffin”.
goodness-of-fit : s : {[w(F02 - F,2)2] / (n - p)} "2, w : 1 / [02(F02) + (up)2 +

bP]

105

3. Results and Discussion

A. Synthetic Methods and Powder X-ray Characterization of
Cu(TCNQ).
(1) Bulk synthesis of phase 1.

Two methods were used to prepare bulk samples of Cu(TCNQ), (7),
and the products were compared to the well-investigated phase first reported
by Melby et al. in 1962 (Eq 2).2311 In Melby’s method, TCNQ is reduced in
situ by CuI which leads to the immediate formation of Cu(TCNQ) as blue-
black microcrystals and I; as a by-product (Eq 2). The I; is trapped by
excess CuI to form 13', thus avoiding any side-reaction with 12. The second
method, which is new, follows from our interest in the use of acetonitrile
solvated transition metal cations as useful starting materials in non—aqueous

28 In the reaction depicted in Eq. 3, acetonitrile

substitution reactions.
solutions of ["Bu4N][TCNQ] and [Cu(CH3CN)4][(BF4) are combined to give
Cu(TCNQ) as a dark blue precipitate with ["BuaN][BF4] as the by-product
(Eq 2).

3 CuI + 2 TCNQ —> 2 Cu(TCNQ) + Cul3 (Eq. 2)

[Cu(MeCN)4][BF4] + ["Bu4N][TCNQ] —-> Cu(TCNQ) + ["Bu4N] [(BF4]
(EQ- 3)

A variety of characterization tools including X-ray powder diffraction
support the conclusion that the products of the reactions in Eqs. 2 and 3
(hereafter referred to as phase I) are the same material, except that the
product of the CuI reaction is typically more crystalline as judged by the
intensity of the powder patterns and by SEM. We were curious as to
whether a different product could be obtained by the use of the Cu(II)

106

starting material [Cu"(CH3CN)4][(BF4)2]. As depicted in Eq 4 the products
are Cul(TCNQ) and neutral TCNQ and not Cu"(TCNQ)2 as one might have
predicted.

[Cu(MeCN)4][BF4]2 + 2 ["BuaN][TCNQ] —> Cu(TCNQ) + TCNQ +
["BulelBFtl (Ext-4)
This is in accord with results reported for the reaction of CuHSOa with
LiTCNQ in H20 which gives an unstable green powder that is occasionally
written as Cu(TCNQ)2,23°, but which is actually Cu'(TCNQ")(TCNQ°) with
trapped TCNQ0 as established by Raman, XPS and IR spectroscopies.23M
Our results from the reaction performed in CH3CN indicate that
Cu"(TCNQ)2 is also unstable under non-aqueous conditions. An important
difference between the aqueous versus non-aqueous reaction is that the
Cu(TCNQ) material retains the neutral TCNQ molecule as part of the
formula due to the insolubility of neutral TCNQ in H20. TCNQ is soluble in
CH3CN, however, and therefore it remains in solution after the reaction in
Eq. 4. The characteristic yellow color of the acetonitrile filtrate and the
v(CEN) stretch at 2222 cm'1 are convincing evidence that the by-product is
neutral TCNQ.29
(2) Conversion of phase I to phase II Cu(TCNQ) in acetonitrile.

In the course of preparing Cu(TCNQ) by the reactions given in
Equations 2 and 3, we made an unexpected observation. We discovered that
if suspensions of phase I are stirred in CH3CN at room temperature for
several days, or if the reaction solution is refluxed for several hours, a
second phase, (8), is produced in essentially quantitative yields. This was
quite surprising to us as we had observed that phase I can be heated in the

solid-state at 150 °C for two days in air or exposed to air and ambient light

107

for weeks with no change as evidenced by X-ray powder diffraction and IR
spectroscopy. Clearly, acetonitrile is capable of dissolving phase I into its
constituents, i.e., Cu(I) and TCNQ'l ions which then recombine to give a
different phase as judged by physical changes that can be monitoredwma‘22
This conversion produces a form of Cu(TCNQ) (Eq 5) that is stable with
respect to further changes under forcing thermal conditions in the solid-state

or in solution.

Cu(TCNQ) suspended in CH3CN —> Cu(TCNQ) (Eq 5)
phase I phase II

B. Syntheses of the Mn(TCNQ)2(solvent)2 phases.

Reactions of [Mn(MeCN)4][BF4]2 with ["BuaN][BF4] in MeOH or
LiTCNQ in H20 lead to the rapid formation of dark blue-purple crystalline
solids. These materials are robust under ambient conditions upon exposure
to air, moisture and light, with no decomposition being detected after several
months. Elemental analyses for the water compounds fit the empirical
formula M(TCNQ)2(H20)2 rather than M(TCNQ)2(HZO)3 which had been
previously reported?”30 Presumably, handling and storage conditions affect
the quantity of water that these materials retain, which is an important point
because degree of hydration is known to influence the magnetic properties of
these aqueous-derived materials.”

No prior reports of binary metal/TCNQ compounds prepared in
alcohols have appeared in the literature to our knowledge. Analytical data
for the bulk methanol product support the formulation of

Mn(TCNQ)2(MeOH)2 as do the TGA measurements.

108

C. Spectroscopic Studies.
(1) X-ray photoelectron spectroscopy.

The 1:1 ratio of Cu to TCNQ in the material does not necessarily
mean that complete charge transfer between the metal and the acceptor has
occurred, thus it is important to probe the oxidation state of the metal in the
Cu(TCNQ) phases, One method that is useful in this regard is X-ray
Photoelectron Spectroscopy (XPS). In earlier reports, XPS data of
Cu(TCNQ) prepared from CuI revealed characteristic binding energies for
Cu(I)2p1,2 and 2pm for Cu(I),9""d‘31 with no evidence for shoulders or higher
binding energy satellites that can be attributed to Cu(II).32 Likewise, XPS
data for compounds (7) and (8) prepared in our laboratories gave essentially
identical results in accord with Cu(I). Data from this work as well as from
earlier papers are presented in Table 18. The 2133/2 and 2131/2 Signals are
depicted in Figure 25. Both phases exhibit essentially identical binding
energies for the Cu 2p orbitals, with no evidence for shoulders or satellites
due to Cu(II). Likewise, the NIS orbitals appear as a single feature at 398.7
eV in both phases which is indicative of one type of TCNQ. Mixtures of
TCNQ’l and TCNQ can be readily discerned in this region.31 The ease with
which Cu(II) can be observed in bona fide mixed-valence Cul/Cull
compounds and materials is illustrated by a comparison of the data for
chemically prepared Cu(TCNQ) with electrochemically prepared
Cu(DCNQI); materials. For example, the mixed-valence metallic compound
Cu(DM-DCNQI); shows two sets of Cu 2131/2 and 2pm (931.6, 933.6 eV)
signals and a more obvious second set of satellite features at 943 and 962 eV
that are characteristic of paramagnetic Cu(II).33a As further support for the
lack of Cu(II) in our TCNQ samples we undertook XPS (Figure 25 and 26)

109

 

 

 

 

 

 

 

 

 

 

m e e e c e #2:: e e 4. m 4 ‘ 1 1 a ‘ ‘ 4.
3 .. Cu(TCNQ) Phasel ..
. I 293/2 I
1» f
7 1» 1r
.,. eh
i 3 5 ..
e 5 4p .4.
E , ..
3 --
q
2 <-
4
l .
l.
0 3 L ‘r i i ¢ ¢ i i
an as see 355 see 915 so 335
Binding Energy, eV
(2!)
19 L .L J. + 4 e e : e e e e 4 e e : ‘r .L ‘F
9 Cu(TCNQ) PhaseIl ~0-
a 293/2 3'
’ 1:
= ‘ 2p1/2 ..
‘9 5 .C
9 4 1?
1r-
3 «b
{P
2 ‘b
l ..
P
0 .fi‘ . + e : 2+ % e 4 : * 4. e : . : . . .
an ass son :55 so 945 so as sea :25 are
Binding Energy, eV

00

Figure 25. XPS data in the Cu 2133/2 and 2pm regions for phase I (a) and
phase II Cu(TCNQ) (b).

110

 

 

 

 

 

 

 

10 ‘ T 4. f
9 ..
s Cu(MeCN)4(BF4) ..
,. ::
:1 ‘ j. 1
0 .. ..
9. ° 2: .
it , .. ::
, .. 1:
21 it
1* ..
‘12 2:

o .1 + 4: . 4 l J. L e
370 S5 S0 5 fl :6 H :35 fl 3

mam 0V
(a)
1. ° 4‘ ' 1 . 4. 4. i e e : : e + e e e ‘
’ Cu(MeCN)4(BF4)2 i
8 ..
, 1:
4
c-e q»
i ‘ .
s 1
5 . .;
3 I
1 .:
1 ..
J-
':7m m ”75 a a his a':
mom. 0'
(b)

Figure.26. XPS data in the Cu 2pm and 2pm regions for
[Cu‘tMeCNhllBFd (a) and [Cu"(MeCN)4][BF4]2 (b).

111

measurements of the nitrile salts [Cu'(CH,CN),][BF,] and
[Cu"(CH3CN)4][BF4]2 as references. Although they were rather unstable
under the vacuum conditions of the experiment, it was possible to observe
important differences. The Cu(II) salt exhibits four different signals in the
Cu 2p region whereas the Cu(I) analog exhibits only two features. We take
these data as convincing evidence that Cu(II) is not present in our samples to
any appreciable degree. This conclusion is in contrast to data obtained for
electrochemically prepared samples reported by Willett and co-workers
which point to the presence of both Cu(I) and Cu(II) species.33b

(2) Infrared spectroscopic studies.

Infrared spectroscopy in the v(CEN) region is very useful for
assigning charge as well as predicting stacking modes in TCNQ” salts.4‘1’6’34
Unfortunately, analysis of IR data for metal-bound TCNQ ligands is
complicated by the fact that v(CEN) stretches can shift to higher energies if
the TCNQ molecules acts primarily as a o-donor, or to lower energies if
there is significant metal to TCN Q n—backbonding.° A more informative
infrared active mode is the 6(C-H) bend of TCNQ. We have discovered that
this ring bending mode is a fingerprint for the presence of the o—dimer
[TCNQ-TCNQ? versus normal TCNQ'l in the materials. The spectrum for
[Mn(TCNQ)(TCNQ-TCNQ)”(MeOH)2].. contains a feature at ~825 cm“1 in
addition to one at 802 cm”1 which are indications of both [TCNQ]" and
[TCNQ-TCNQJZ' being present as verified in the X-ray structure (see single
crystal section). The spectrum of [Mn(TCNQ)2(HzO)2].. on the other hand
exhibits only one 5(C-H) feature at ~823 cm'l due to the syn-uz-[TCNQI
ligand.4h

Infrared spectra of the two phases of Cu(TCNQ) are quite similar as

would be expected on the basis of their formulations as compounds of

112

TCNQ". No peaks associated with acetonitrile were observed in either

f Compound (7) shows a strong, broad v(CEN) absorption at

phase.4
2199 cm'1 with a shoulder at 2172 cm’1 whereas (8) exhibits two strong,
sharp stretches at 2211 cm‘1 and 2172 cm'l. Perhaps even more indicative of
the similarity of the TCNQ unit in the two phases is the 8(C-H) mode at
825 cm'1 which is very sensitive to changes in oxidation state. These data
are consistent with the presence of TCNQ'l and not TCNQ, TCNQz', or
mixed-valence stacks of TCNQ'l and TCNQ.33 Furthermore, they are in
accord with previously reported IR data for samples of Cu(TCNQ) prepared
by the original CuI method.'°’33‘l Since the existence of a second phase was
unknown until the present work, there are no data with which to compare
phase II.

(3) SEM studies of Cu(TCNQ).

Scanning electron microscopy was used to discern morphologies of
the bulk Cu(TCNQ) samples in this study, and the results were compared to
films of the material. It is evident from the micrographs depicted in Figure
27 (a) and (b) that the two bulk phases are physically quite different. Phase I
is composed of regular needles Whereas phase 11 consists of crystallites With
a platelet morphology. The same two crystallite types have been reported to
appear during the growth of Cu(TCN Q) films from acetonitrile
solutionsw’zoa’22 If one compares these observations to Cu(TCNQ) grown
asfilrns on Cu substrates, it is obvious that the phase with densely packed
needles oriented perpendicular to the Cu surface appears immediately and
survives only during short dipping times at 60 °C (Figure 28 (a)).'9'2°‘
Within 20 minutes, the needles become visibly pitted (Figure 28 (b)) and
after 1 h (Figure 28 (c)), the film consists of regular-shaped blocks. A further

113

.2 as so: so: .

.39 .m 5:08 13:05:0on 05 E @0508 $8553 :
.mmm :0: ..R :0 3:02 Co 0050:: 285:2? OZUH + EU a

 

 

 

 

 

 

mg .mg

om :00 9mm: 6.28 ..... 35mg ”302092920

mm :00 fiwam 6.35 .mémo mcmw among 288 «AOZUCHC

3202:: a 0225

:53 $5 fiwmm 5&3 4&3 mwmw mNSN .2 EN :AOZUHvzu

3:000:

0.53 wE: fiwom 5.35 dd? mmmw mNCN madman H 08:: ...AOZUCRV
22 ”Sam .33 :0

80:80:00 max A>0v .m.m man :80 $-va E 7:8 AZmUv> M:

 

 

 

 

 

 

.QOZUQ-EQV:U 0:0 AOZUCRV :8 02.6 max 0:: w: .3 030,—.

114

illustration of the importance of reaction conditions in dictating the
composition of the films is the comparison of SEM micrographs of films
grown at 80 °C and 60 °C for 1 h (Figure 29). At 80 °C, the film consists of
a single type of crystallite that is identical to bulk phase H as judged by
powder XRD. In contrast, the 60 °C film shows regions of both
polymorphs. At higher resolutions, one can observe from the micrographs in
Figure 30a-d, that the needle morphology is dominant at 40 °C even after
one hour (Figure 30 a,b) and that the platelets are present at 80 °C on Cu foil
under the same conditions as indicated in Figure 30c,d. These results are
compelling evidence for the high degree of variability in films prepared
under different conditions, a fact that helps to explain the lack of
reproducibility in the electrical switching studies reported by various groups.
(4) Powder X-ray studies of Cu(TCNQ) bulk phases.

The most convincing evidence that two distinctly different forms of
Cu(TCNQ) exist can be found in the diffraction studies of microcrystalline
samples prepared by bulk solution routes. A number of crystal structures
containing copper and other ligands in addition to TCNQ have been
reported,4 but there are no published structural data for a binary phase
containing only copper and TCNQ. Powder patterns of phase I prepared at
room temperature from CuI and TCNQ and a suspension of phase I that had
been refluxed in MeCN for three hours to convert it to phase II are depicted
in Figure 31. Powder diffraction patterns from a Rigaku RU2OOB X-ray
powder diffractometer were used for indexing by the programs Dicvol91 and
Treor90. This led to the tetragonal cell a = b = 11.252 A, c = 3.950 A,
V = 500.16 A3 for phase I and the monoclinic cell a = 5.325 A, b = 5.343 A,
c = 18.955 A, 13: 93.570°, V= 538.3 A3 for phase 11.

115

('u'l‘('\() Phase I mag. HNNh

Phase I
(a)

Phase II
(b)

 

Figure 27 . SEM micrographs for solution-prepared Cu(TCNQ) crystals.

116

(a)
Phase I

“101‘ I minulc ul'ilipping

*—| pm

0?)
Phase I
dissolving

.\l'lcr 20 minutes ol'ilipping -| Jun

(C)
Phase II

\l'lcr l lmurol'dipping —| “m

 

Figure 28. SEM micrographs showing the progressive changes from
Cu(TCNQ) phase I to phase II.

 

(b)

.\i'
,1

51») pm

 

Figure 29. SEM of Cu(TCNQ) films grown on copper in MeCN for 1hour
at different temperatures: (a) at 60°C, (b) at 80°C.

118

   

(C)

 

Figure 30. SEM micrographs of Cu(TCNQ) grown on copper in CH3CN:
(a) top and (b) side views of films grown for 1h at 40 °C (phase
I) and (c) top and (d) side views of films grown for 1 h at 80 °C
(phase II). The reference scale on all micrographs is 10 um.

119

(5) Powder X-ray studies of Cu(TCNQ) films.
(a) At room temperature.

It has been reported that Cu(TCNQ) films grown on Cu foil surfaces
at short reaction times are spectroscopically identical to the bulk material

I.10 Since our discovery that a second phase exists,

prepared from Cu
however, it is important to investigate changes in film composition as a
function of reaction time. Consequently, XRD patterns of the Cu(TCNQ)
films were monitored periodically by removing one of the pieces of copper
sheet from the TCNQ solution and collecting an X-ray powder pattern on the
resulting material. Representative data are provided in Figure 31 with the
top and bottom powder pattern being provided as references for pure phase I
and II respectively. The powder pattern collected after 6 hours is dominated
by the features of phase I, but within 46 hours it is evident that the second
phase is being formed at the expense of the first phase; by 76 hours phase H
is the major species (Figure 31). We noted that the color of the films
remained constant during the reaction and that there were no other visible
changes in the film with the exception of the thickness. The progressive
powder X-ray monitoring of Cu film reactions with TCNQ acetonitrile
solution is shown in Figure 31.
(b) At elevated temperatures

Since the Cu(TCNQ) films in devices are required to be at least 5 um
thick, CH3CN solutions of TCNQ are typically heated to ~80 °C to
accelerate the rate of growth on the Cu foil.15’19 Clearly it is important to
know the composition of these films prepared at elevated temperatures.
Accordingly, a piece of Cu foil was dipped as described in the experimental
section at 80 °C in CH3CN and the resulting film was analyzed by X-ray
powder diffraction. Figure 32 shows the resulting XRD pattern obtained on

120

the sample with the underlying Cu substrate. These data strongly support
the conclusion that phase II is the main component of the films used for the
switching experiments. There is a very low concentration of phase I for
films prepared under the conditions cited in the literature for the fabrication
of devices.
(6) XRD and SEM studies of M(TCNQ); and M(TCNQ)2L

Powder patterns of binary metal TCNQ samples indicate that the
M(TCNQ); (M = Mn, Fe, Co, Ni) binary phases are isomorphous. Indexing
results for Mn(TCNQ)2 and N i(TCNQ)2 indicate that both compounds
belong to the tetragonal crystal system. Comparison of the powder patterns
of M(TCNQ)2L (L = MeOH, H20) to a simulated pattern obtained from a
structural model of the single crystal Ni(TCNQ)2(H20) reveal that bulk
prepared samples of Mn(TCNQ)2(MeOH)2 and Mn(TCNQ)2(H20)2 should
exhibit similar structural features, which are different from the known
Mn(TCNQ)2(MeOH)n (n = 2, 4) and the Mn(TCNQ)2(H20)2 phase.
(a) Powder patterns of M(TCNQ); .

The bulk prepared compounds samples of M(TCNQ)2 (M = Mn, Fe,
Co, Ni) from acetonitrile all exhibit essentially the same powder pattern,
which indicates that these products are isostructural. Unfortunately, the
samples suffer from having a small particle size, which hinders their
structural determination by powder techniques. The SEM and TEM
micrographs (Figure 34) of the Mn(TCNQ); sample indicated that it consists
of nano-crystals of approximate size 0.05 x 0.02 x 0.02 um which leads to
severe peak broadening in the powder pattern. Indexing results on the
Mn(TCNQ); compound, the most crystalline sample of the group, indicate a
tetragonal crystal system with a = 12.272(2)A, b = 12.272(2)A c =
8.713(5)A, a = ,8 = y = 90°. V = 1312.12A3 The four strongest peaks

121

 

Cu(TCNQ) phase I

 

 

 

 

 

 

 

j 1
10 20 30 40 50
L l 1 I 't A l l l L A l A ’t s A l - A L L 4 j
Curt-TCNQ In MeCN at
room temp. for 6 h
I
1'0 f 2‘0 ' 3'0 1 4'0 f s
1 l J L L l J; 4 1
Cu + TCNQ in MeCN at
1 room temp. for 46 h
II
I 1 I f T I T 7
10 20 30 40 50
L J; l 4 l I l A _L
II

Cu + TCNQ in MeCN at
room temp. for 76 h

 

 

 

 

 

 

 

Figure 31. XRD powder patterns illustrating the conversion of Cu(TCNQ)
phase I to phase II in CH3CN.

122

 

 

 

 

 

 

T I I I I I I I I I I I I I I I I I I l I I I T r I T I j T I T I T I I I I I
t Cu f
C I
T Cu(TCNQ) film from Cu + TCNQ 1
~ 1
C in MeCN at 80 'C for 1 h 1
r 1
II I
L— T
. .
_ I ' 5
I L l l L L i M i 1 l l l l J_L L L L L l l l l L l l l L l L l l l l l l l J .
5 10 15 20 25 30 35 40 45

28

Figure 32. XRD pattern for a Cu(TCNQ) sample grown on a Cu substrate
for 1 h at 80°C in a CH3CN solution of TCNQ.

123

.32 .00 any :2 M Sc .NAOZUHVE mo 50:3 a .8 «unwr—

'll.~
!

 

 

 

«6282.:

III!"
3 7 9 I R 3 R a. or .

 

 

- v _ - . - . . _ -
----__-—---_-——--_-_—-—._—-—_-—-__-_-

l§§§_

 

 

'1

«620.58

 

 

assigns»

 

 

 

 

 

«620.com

«runs

li§§fiiiii°

124

    

.3 . *I “”3.

Mn(TCNQ)2

 

Mn(TCNQ)2

(b)
Figure 34. Mn(TCNQ); small particles in nanometer size could be seen from
(a) SEM (lum scale) and (b) TEM (0.1 pm scale) micrographs.

125

correspond to d spacings of 6.136, 5.488, 4.339 and 3.404 A respectively.
All four metal TCNQ (Mn, Fe, Co and Ni) have similar pattern.(Figure. 33)
It is important to point out that the strong peak in the powder pattern with a
d pacing of 3.4 A points to well stacked TCNQ columns in the structure.

D. Single Crystal Diffraction Studies.
(1) Details of collection and refinement for Cu(TCNQ).

A tiny needle of Cu(TCNQ) phase I and a very thin platelet of
Cu(TCNQ) phase II were each mounted on the end of a glass fiber and X-
ray data were collected on a Siemens (Bruker) SMART 1K CCD area
detector. In both instances, the crystals were exceedingly small, such that
they tested the limits of the instrument. The weak data problems
notwithstanding, the major features of the structures can readily be
discerned. The results are in accord with the powder patterns, furthermore
they help to explain the observed properties in a qualitative manner. The
refinement of phase I is poor due to a low data-to—parameter ratio and to a
merohedral twinning problem, but phase H actually refines quite well given
the tiny size of the crystal and the fact that it is quadruply twinned.
Numerous independent data sets of these materials have been collected on
CCD diffractometers over the course of nearly two years, and the present
results are the best that we have obtained. Since Cu(TCNQ) is the subject of
much interest and its structural determination has eluded researchers for
decades, these data sets, albeit marginal, are important to report. In keeping
with the lower resolution of these structures, however, we have limited our
discussion to the most important differences in phase I and H; no attempts to

base conclusions on fine points of the structures are made.

126

(a) Cu(TCNQ) Phase I.

The structure was partially solved by direct methods and refined in
SHELXTL V5.10. The presence of twinning rendered the space group
choice very difficult. Solutions were attempted in numerous tetragonal,
orthorhombic and monoclinic space groups, but the only reasonable solution
was obtained in monoclinic Pn. The pseudo-merohedral twinning matrix [1
0 O, O O 1, 0 —1 O] was used in the refinement to represent the pseudo-four-
fold symmetry. The structural solution gave two independent Cu(TCNQ)
orientations in a ratio of ~50% each. A total of 1111 unique data were
refined isotropically on fl to R1 (wR2) = 0.232(0.481) but, not surprisingly,
the lack of data and other complications led to a slightly unstable refinement.
Nevertheless, important features of the relationships within the structure can
be gleaned from this very weak data set. The Cu atoms in the structure are
coordinated to four nitrogen atoms in a highly distorted tetrahedral
environment as evidenced by the N -Cu-N angles of 92° and 142° (Figure 35
and 37). The quinoid rings of the TCNQ units are engaged in interplanar
stacking at a distance of 3.24 A which is less than the van der Waals
distance of 3.4 A for carbon atoms. Adjacent TCNQ stacks are rotated by
90° with respect to each other which is a standard feature in most binary
metal/'1‘ CNQ salts.7‘34 Powder simulations based on the raw reflection data
and on the structure solution are very similar to the experimental powder
pattern for the bulk phase I.

(b) Cu(TCNQ) phase II.

Cu(TCNQ) phase II features a new binary metal/'1‘ CNQ bonding
mode. Other known TCNQ bonding modes including pseudo-tetragonal
stacks with alkali metal TCNQ products, Cu(TCNQ) phase I, and
Ag(TCNQ). The crystal was found to be a quadruplet twin; this was

127

resolved by deconvolution methods, and the data were refined in the space
group P2/n. A thermal ellipsoid diagram of the asymmetric unit is provided
in Figure 36. The N-Cu-N angles around the Cu atom are between 103° to
114.7°, and the average Cu-N bond length is 1.95 A. The metal geometry is
close to tetrahedral, unlike the highly distorted Cu(TCNQ) phase I.
Adjacent TCNQ ligands are parallel to each another with the shortest face-
to-face contact between nearest neighbors in the same network being 6.8 A
(Figure 38). This arrangement of TCNQ'l ions is unprecedented; typically
they are situated around a metal ion as four TCNQ units related by a pseudo-
four fold rotational axis such as that found in Ag(TCNQ) and alkali metal
TCNQ salts.7‘32

(c) Comparison of the Cu(TCNQ) polymorphs.

Schematic views of the two structures are depicted in Figure 39. Both
phases exhibit polymeric motifs based on the repeat pattern of a four-
coordinate Cu ion ligated to nitriles of independent TCNQ molecules. There
are two important differences in the structures, however, that lead to entirely
different spatial arrangements for the TCNQ units in the two structures. The
first difference is the relative orientation of TCNQ moieties around the Cu
atoms. In phase I and in all other structures of 1:1 M(TCNQ) compounds
including that of Ag and alkali metal salts,7’35 neighboring TCNQ molecules
are rotated 90° with respect to one another as illustrated in Figure 37. Phase
II represents a new structural archetype for this type of solid in which
infinite arrays of co-planar TCNQ molecules are oriented in the same
direction, but in two perpendicular planes as depicted in Figure 38. The
second major difference in the two phases is the type of interpenetration that
they exhibit. In phase I, the two independent networks bring the TCNQ

molecules together to give a columnar stack with the closest distance being

128

//// Cull)
N(3)
g NH) 0 0(5) ' cm)
‘ g (3(6)
0 C(1) C(10)
. e
. 9 cm
(3(2) e
c<3> o ‘ 0‘8) C(12)
0(9) .
' NM) 9
M2) 9

Figure 35. Pluto representation of Cu(TCNQ) phase I.

129

 

'\

be
I 3
(a?
«it
6)

Figure 36. Thermol elipsoid representation of CuTCNQ phase II.

130

3.24 A (Figure 37). In sharp contrast, the interpenetration in phase 11 does
not bring the two independent networks together, rather the TCNQ rings are
"slipped" and no n—stacking occurs (Figure 38). The closest distance
between parallel TCNQ units in the same network is 6.8 A.

((1) Further structural analysis of Cu(TCNQ) and a comparison to
Ag(TCNQ).

At this point, it is instructive to return to the powder patterns for the
two Cu(TCNQ) phases and also for the Ag(TCNQ) material that was
structurally characterized by Shields.7 An experimental powder pattern of
Cu(TCNQ) phase I is given in (Figure 40) from which it can be observed
that there are very few intense features in the high angle region. This is an
indication that we are not observing much information about order in the
third dimension. The fact that the sample is composed of very thin platelets
leads to a preferred orientation which limits the amount of observed data.
Since the Cu(TCNQ) phase I exhibits a connectivity reminiscent of
Ag(TCNQ), it is important to comment on the fact that their powder patterns
are very different. Ag(TCNQ) crystallizes in the orthorhombic space group
Pnnm, with a = 6.975(1) A, b = 16.689(1) A, c = 17.4550) A, V = 2031.5
A37 The powder pattern of Ag(TCNQ) simulated from the X-ray data of
Shields shown in (Figure 40) is clearly different from that of Cu(TCNQ)
phase 1. Bulk samples of the Ag(TCNQ) phase (Figure 40) prepared from
the reaction between Ag[BF4] and ["BuaN][TCNQ] in CH3CN were found to
be identical to the sample used for the X-ray simulation. Interestingly, if the
reaction is stopped after only two minutes instead of 3 hours, the higher
symmetry phase in Figure 40 is obtained. This new Ag(TCNQ) phase is
similar to Cu(TCNQ) phase I in the low angle region; this implies a pseudo-

tetragonal cell and a similar connectivity. From its short lifetime, however,

131

 

Figure 37. Interpenetrating networks of Cu(TCNQ) phase I features TCNQ
stacking between different networks ( solid bond -- network 1, holo bond -
network 2).

132

 

Figure 38. Interpenetrating networks of Cu(TCNQ) phase 11. Note that the
TCNQ units between different networks are shifted and not n—stacked

between different networks (solid bond: network 1, holo bond: network 2).

133

it is doubtful that it will be possible to obtain this phase of Ag(TCNQ) as
larger crystals for single crystal analysis. Interestingly, if one simulates the
Ag(TCNQ) powder pattern with the artificially imposed average cell edges
a = b = 17.06 A, the lines collapse to a pattern that resembles both the
Ag(TCNQ) and Cu(TCNQ) phase I structures (Figures 40).

(2) Crystallographic studies of M(TCNQ)2Lx phase (x=solvent).

(a) [Mn(TCNQ)(TCNQ-TCNQ)os(MeOH)2]e.

Crystals of this compound exhibit a 2-D zig-zag network of Mn(II)
ions coordinated to two different types of TCNQ ligands (Figure 41). The
Mn(II) ions are surrounded by four nitrogen and two oxygen atoms from
tetradentate ua-[TCNQ-TCNQF', bidentate cis-[u-TCNQI' and axial MeOH
ligands. The [TCNQ-TCNQ? entity has been documented only three
previous times, and in these instances, it behaves solely as an outer- sphere
anion.3S In the compound [EtP](TCNQ)2 (EtP = N -ethylphenazinium),3 5“ the
o-[(TCNQ-TCNQ]2' dimer contains a long C-C bond of 1.631(5) A that
joins the two rings, whereas in [Pt(bpy)2][TCNQ-TCNQ]35° and
[Cu(DMP)2]2[TCNQ-TCNQ] (DMP = 2,9-dirrtethyl-1,10-phenanthr61ine)35c
the corresponding distances are 165(2) A and 1.630(13) A. A schematic of
the o-dimerized [(TCNQ)2]2' unit is depicted in Figure 41. These C-C bonds
(gl-gl' in Figure 41.) including the one in the current compound of
1.630(13) A are long, but obviously significant, since the C atoms involved
are essentially tetrahedral. The uniformity of the distances in the rings
(summarized in Table 19), and the significant lengthening of the C=C bond
in the planar exocyclic C=C(CEN)2 groups in the [TCNQ-TCNQ? moiety
are indications that there is a main contribution from the aromatic resonance
structure depicted in Figure 41. Distances and angles within (TCNQ-
TCNQ)2' (Table 19) are similar to those found in another

.GOUOOGGOO

2.90 05 mo yoEBw 05 :o £9395 :: 523 = owns: :5 2 82: 62090.0 mo 33: own—:ofim .am 9.5me

~— SSE AOZUERV — 8!: A020h0=0
:0 =0 :0 \ \ao
/ \ J \ le 082
z
z _

1 34
\\
Z
s 3
0
//Z
\\
32
20 is

135

next-eat.)

 

 

 

 

 

~<-—l~wzm-iz

 

 

 

Jest-onto)

 

 

 

-<'-l"‘VJZtTJ—IZ"'

 

 

 

 

 

(b)

Figure 40. Experimental XRD powder pattern for (a) Cu(TCNQ) phase I
and (b) Ag(TCNQ).

136

compound recently crystallized in our laboratories, namely [Mn(TCNQ-
TCNQ)(MeOH)4].., which also contains a o-TCNQ dimer with a C-C bond
joining the TCNQ units of 1.659(10) A. The cis-[tt-TCNQJ' molecules act
as bridging ligands via 1,2-dicyano positions with the two unligated Ni-
groups pointing outward from the edges of the zig-zag layers towards axial
MeOH ligands in adjacent layers to form hydrogen bonds (N3---02A =
2.834(8) A and N4---OlA = 2.846(8) A), Figure 41. The metrical
parameters within the bidentate TCNQ" ligand (Table 19) are in accord with
data reported for other structures that contain [TCNQ]".4‘7 In addition to
participating in hydrogen bonding, the cis-[u—TCNQI ligands are engaged in
n—stacking interactions of 3.295 A that serve to stabilize a densely packed,
interdigitated arrangement of layers (Figure 41).

(b) [Mn(TCNQ)2(H20)2]~(10).

Compound (10) represents a new type of TCNQ bonding mode, i.e.
syn-u-TCNQ, which has not been reported before. A thermal ellipsoid
diagram of the asymmetric unit in [Mn(TCNQ)2(H20)2].. is provided in
Figure 43, and packing diagrams in the ab and be planes are depicted in
Figure 54 (b) and (c) respectively. The manganese ions are in a pseudo-
octahedral environment consisting of four nitrogen atoms from two syn-[p.-
TCNQ]'° ligands and two axial H20 molecules. The Mn-N distance of
2.225(2) A is similar to the corresponding distances observed for the
coordinated TCNQ ligands in [Mn(TCNQ)(TCNQ-TCNQ)”(MeOH)2].. ,
but there are some notable differences (Table 19). The distances within the
TCNQ rings are indicative of the resonance structure in Figure 41c.
[Mn(TCNQ)2(H20)2].. contains alternating single and double bonds in the
quinoid ring of the TCNQ, whereas in the MeOH containing structure, the

bonding in the ring is uniform with charge being localized on the exocyclic

137

Table 19. Key bond distances(A) in the TCNQ unit(s) for

 

 

[Mn(TCNQ)(TCNQ-TCNQ)0,5(MeOH)2] .. (9) and
[Mn(TCNQ)2(H20)2] .. (10) according to diagram in Figure 41 .

(9) (10)
al-bl 1.390(10) a3-b3 1.415(2)
b 1 -c1 1.459(7) b3-c3 1.419(2)
c1 -d1 1.39( 1) c3-d3 1.429(2)
d1 -e1 1.385(8) d3-e3 1.365(2)
el-fl 1.38(2) e3-f3 1.430(2)
f1 -gl 1.528(7) f3-g3 1.419(2)
gl-gl' 1.635(10) g3-h3 1.417(2)
g 1 -h1 1.48(2)
a2-b2 1 .401 (10)
b2-c2 1 .404(8)
c2-d2 1 .43(2)
d2-e2 1 365(8)
e2-f2 1.43( 1)
f2—g2 1.421(7)

1 .407(10)

 

138

 

 

 

 

Mn
\ h Mn
a I
N\\\C lb c dl el f g :5 IN /
_1 l 1 1 1:" C’ ¢N
8
N/ C\\\7 l "1 (a)
C \
Mn/ K? N\
04-[TCNQ-TCNQ] 2’ o-dimer in (9) Mn

Mn

N \\\ 32 d2 62 '12 ’//N/

\
Cs

K? N\

czs—u-[TCNQ] '1n(9) Mn
Mn Mn
\ h N/
N\ a3 d3 63 3 91
M” *‘N

syn-u-[TCNQ] -- in (10)

Figure 41. Various bond lengths in TCNQ compounds (9), (10).

139

5 3Emooseneazufiozobazubeé mo eooaeoaoaoc 286:6 deuce... he 2:5

 

140

carbon atom. The extended structure is a 2-D polymeric layered structure
with extensive intra-layer ( 3.048(2) A) and inter-layer (~3.3 A) n-stacking
of the [TCNQ]" radicals. While the latter value is fairly long to be a
significant interaction, the 3.048 A separation between the [TCNQ]'l ligands
within the layers is a sign of very strongly coupled n—dimers. Typical ranges
for 1t- [TCNQ]22' interactions for selected compounds are provided in Table
19.

(c) Structures of ["BmNHTCNQ] (11) and ["BmNHTCNQFa] (12)

salts.

The structure of (11) is dominated by the packing of the
tetrabutylammonium cation rather than the TCNQ anion radical. Two
TCNQ anion radicals are related by an inversion center, and form a
dimerized divalent anion, which are stacked to a nearly eclipsed pattern
(Figure 46). Refinement of the least-squares plane of the dimer revealed the
inter-planar distance of the dimer is 3.218(2) A, which is shorter than the
van de Waals distance between aromatic rings. The dimerization of the
TCNQ anion radical stabilizes the reactive radical. Important bond lengths
and angles are listed in Table 21.

There are two independent TCNQF4 moieties in the asymmetric positions of
Structure (12). Only one of them is packed in a stacked dimer radical form
With the inter-planar distance of 3.206(5) A. The other pair of TCNQF4
anion are packed together loosely with the inter-planar distance of 3.265(9)
A- The angle between the least-square planes of the two unique dimer pairs
is about 70°. Important bond lengths and bond angles of (12) are listed in
Table 22.

(d) {Zn(H20)4(TCNQ)3-2MeCN},. (13).

The structure of (13) features a Zn(H) cation coordinated to four H20

 

141

 

Figure 43. Thermal ellipsoid representation of Mn(TCNQ)2(H20)4 (10).

Ta

 

 

142

 

 

 

Table 20. Selected bond lengths [A] and angles [°] for
[Mn(TCNQ)2(H20)2].. (10).

A-B A A-B A
Mn(1)-O(1) 2.1481(18) C(7)-C(10) 1.419(2)
Mn(1)-N(2) 2.2237(16) C(2)-C(4) 1.420(3)
Mn(1)-N(4)#2 2.2270(17) C(5)-C(6) 1.369(2)
N(4)-Mn(1)#4 2.2270(17) C(8)-C(9) 1.362(3)
N(1)-C(1) 1.148(3) C(4)-C(5) 1.419(2)
N(2)-C(3) 1.151(2) C(6)-C(7) 1.419(2)
N(4)-C(12) 1.148(2) C(4)-C(9) 1.427(3)
N(3)-C(11) 1.151(2) C(7)-C(8) 1.431(2)
C(10)-C(11) 1.423(2)

C(10)-C(12) 1.412(3)
C(1)-C(2) 1.417(3)

_ C(2)—C(3) 1.419(3)

\ A-B-C ° A-B-C o
O(1)-Mn(1)-N(4)#2 8822(7) O(1)-Mn(1)-N(2) 9070(7)
N(2)-Mn(1)-N(4)#2 8902(6) O(1)-Mn(1)-N(2)#1 8930(7)
O(1)-Mn(1)-N(4)#3 91.78(7) N(1)-C(1)-C(2) 178.2(2)
N(2)-Mn(1)-N(4)#3 9098(6) C(1)-C(2)—C(3) 116.57(17)

\C(3)-N(2)-Mn(1) 161.5905)

 

Symmetry transformations used to generate equivalent atoms:

#1 -x,-y,-z #2 x-1/2,-y-1/2,z-1/2 #3 -x+1/2,y+1/2,-z+1/2

#4 -x+l/2,y-1/2,-z+1/2

143

molecules and two nl-TCNQ molecules in trans-positions. The Zn atom
resides on an inversion center. There is one unique TCNQ molecule and
two half TCNQ fragments in the asymmetric unit. Both TCNQ fragments
are near inversion centers and are stacked in colunms. There are two
acetonitrile molecules in the interstices. These are hydrogen-bonded to the
water molecules. The least-square refinement of the inter-plane distances
between TCNQ molecules is ~3.28 A, (the plane is slightly tilted by ~1.3°).
There is one heavy residual peak in the difference map
(~ 2e'/A3) which is located on an inversion center. This peak was modeled
as a disordered position of the Zn atom with its coordinated water molecules
as a group; the refinement led to a 966/3 .4 ratio over the two positions.
This treatment brought the residual R1 down to 5.85% from 6.42% for 4230
reflections with I > 20 and the heaviest residual peak in the difference map
down from 2.28 to 0.4e'lA3. A thermal ellipsoid representation of
compound (13) is depicted in Figure 49. A packing diagram that shows the
TCNQ stacking pattern is depicted in Figure 50.

(e) [Ni(DMSO)o][TCNQ]3 (14)-

Compound (14) is a salt of the DMSO solvated nickel (II) cation with
TCNQ'1 anion radicals stacked along with neutral TCNQ. The cell volume
is close to that of {Zn(nl-TCNQ)2(H20)4(TCNQ)02MeCN} in which
columns of [TCNQ]32' are dominant the structural framework. The nickel
atom, which resides on an inversion center, is bonded to six DMSO ligands.
There are one and one-half unique TCNQ moieties in the asymmetric unit.
The stacks of TCNQ columns are arranged along the b axis at an inter-planar
distance of ~ 3.32 A, which is essentially the same as the 3.4 A van der Waal
distance for aromatic rings. A PLUTO drawing featuring the TCNQ stacks
is shown in Figure 52. The SMez group on one of the DMSO

Table 21. Bond lengths [A] and angles [°] for ["BuaN][TCNQ] (11).

 

N(1)-C(1)
N(2)-C(3)
N(3)-C(1 1)
N(4)-C(12)
N(5)-C(17)
N(5)-C(13)
N(5)-C(25)
N(5)-C(21)
C(1)-C(2)
C(2)-C(4)
C(2)-C(3)
C(4)-C(9)
C(4)-C(5)
C(5)-C(6)
C(6)-C(7)
C(7)-C(10)
C(7)—C(8)
C(8)—C(9)
C(10)-C(1 1)
C(10)-C(12)
C(13)-C(l4)
C(14)-C(15)

1.155(3)
1.151(3)
1.154(3)
1.154(3)
1.524(3)
1.517(3)
1.525(3)
1.528(3)
1.419(3)
1.418(3)
1.424(3)
1.419(3)
1.419(3)
1.363(3)
1.412(3)
1.420(3)
1.422(3)
1.363(3)
1.415(3)
1.419(3)
1.519(3)
1.524(3)

C(17)-N(5)-C(13)
C(17)-N(5)-C(25)
C(13)-N(5)-C(25)
C(17)-N(5)-C(21)
C(13)-N(5)-C(21)
C(25)-N(5)-C(21)
N(1)-C(1)~C(2)
C(4)-C(2)-C(1)
C(4)-C(2)-C(3)
C(1)-C(2)-C(3)
N(2)-C(3)-C(2)
C(9)-C(4)-C(2)
C(9)-C(4)-C(5)
C(2)-C(4)-C(5)
C(6)-C(5)-C(4)
C(5)-C(6)-C(7)
C(6)-C(7)-C(10)
C(6)-C(7)-C(8)
C(10)-C(7)-C(8)
C(9)-C(8)-C(7)
C(8)-C(9)-C(4)
C(l 1)-C(10)-C(7)

108.53(16)
111.35(16)
108.57(16)
108.65(16)
112.02(l6)
107.75(16)
177.8(3)
122.3(2)
120.9(2)
116.33(19)
179.9(3)
121.43(19)
116.8(2)
121.8(2)
121.4(2)
121.9(2)
121.7(2)
116.93(19)
121.3(2)
121.3(2)
121.7(2)
121.7(2)

 

145

Table 22. Bond lengths [A] and angles [°] for ["BuaN][TCNQF4] (12).

 

Bond length
F(4)-C(8) 1.345(4) C(4)-C(9) 1.409(5)
F(2)-C(6) 1.358(4) C(4)-C(2) 1.416(5)
F(3)-C(9) 1.351(4) C(4)-C(5) 1.418(5)
F(1)-C(5) 1.340(4) C(7)-C(6) 1.407(5)
N(l)-C(1) 1.155(5) C(7)-C(10) 1.416(5)
C(3)-N(2) 1.155(5) F(5)-C(17) 1.340(4)
C(11)-N(3) 1.144(4) F(6)-C(18) 1.346(4)
C(12)-N(4) 1.144(5) F(7)-C(21) 1.354(5)
C(8)-C(9) 1.356(5) F(8)-C(20) 1.358(5)
C(8)-C(7) 1.421(5)
Bond angle

F(4)-C(8)-C(9) 1 18.1(3) N(2)-C(3)—C(2) 176.8(4)
F(4)-C(8)-C(7) 118.3(3) N(1)-C(1)-C(2) 176.8(4)
F(1)-C(5)-C(6) 118.0(3) N(4)-C(12)-C(10) 174.6(4)
F(1)-C(5)-C(4) 118.6(3) C(2)-C(4)—C(5) 123.2(3)
F(3)-C(9)-C(8) 117.3(3) C(6)-C(7)-C(10) 124.0(3)
F(3)-C(9)-C(4) 119.0(3) C(8)-C(9)-C(4) 123.7(3)
F(2)-C(6)-C(7) 118.8(3) C(6)-C(7)-C(8) 112.3(3)
C(9)-C(8)-C(7) 123.6(3) C(10)-C(7)-C(8) 123.7(3)
C(9)-C(4)-C(2) 124.1(3) C(6)-C(5)-C(4) 123.4(3)

C(9)-C(4)-C(5) 1 12.7(3)

 

146

.2 c 2020:2390 :6 somaeoaoaoc 26856 aaoofi .3. 6.5mm.—

 

147

I
L
A
b
& ~ ‘. .
e O
.0 Q
l ,"
5.".
L
O ‘.. ‘1. 0 ‘.
a k 0 ‘a
0 ‘ " "
a ‘ a
. 5
.,

 

(
v
.‘I
O

D
0".
D
O
9
. I
v'.

Figure 45. Packing diagram of ["BuaN][TCNQ] (11) along the a axis.

F i810
ecH1381

 

 

148

 

(b)

Figure 46. (a)The dimer radical pair of (TCNQ)2]2'packed in a nearly
eclipsed pattern in (11). (b) The two dimer pairs of (12).

149

.8 c 5520.: 223m; :o 85:88.? Bowman. 358.: .3. 65mm...

 

:25: coo—om: E862 05 :5 6:5.
£020... mucosa 2: 53 $2: ass e .58 poses a: 5020.: 223a :o 888% mice: .8 95$:

150

 

151

206203 AzuoziozubgoamvgozufiSad 0o 835888 283:. Reece... .3. 88E

 

152

 

 

 

 

 

 

 

 

 

 

 

Figure 50. Packing diagram of (13) along the [l O 1] plane emphasizes the
TCNQ stacking pattern.

 

 

Figu:

 

 

153

.1 N3 '4’ 5
,5. C21 "‘ 025
N6 C24 .
£ 1. C20 n’ (‘7 ,
(.3‘ ‘5 '
N4 C17

 

Figure 51. Thermal ellipsoid representation of [Ni(DMSO)6][TCNQ]3 (14),

154

£828 wad—08m 020,—. wfiaonm mg a Esau 3303 AVS mo 88w“? mat—cam .mm are

 

 

 

 

 

 

 

mc
iso

dra

def.
nicl
this
at01

nicl

the]

(Win.

Cal-[X

155

molecules is disordered over three positions, which were modeled and
isotropically refined to a ratio of 65.7/22.8/11.2. A thermal ellipsoid
drawing of (14) is depicted in Figure 51.

(f) Ni(TCNQ)2(H20) (15).

Compound (15) was solved by a disorder model. A plot of (15) is
depicted in Figure 53. The residual R1 is relatively low, given that only the
nickel atom is refined anisotropically, but there are still major problems with
this model. The elongated ellipsoid of nickel atom indicates that the nickel
atom probably posited at the average position of the real solution, which
nickel atoms could be disordered. The other possibility is that there is a
superlattice problem. We suspect that there is a lack of long range ordering
in the c direction (TCNQ-stacking direction) which is known a stacking
fault. Further studies are needed to address these issues, for example
electron diffraction could be very useful in helping to identify the problem.
(3) TCNQ bonding modes in various TCNQ containing

compounds.
(a) Ring interactions and physical properties.

TCNQ exhibits several types of bonding and stacking modes.36 In
terms of the stacking and connection style of the quinoid structure, TCNQ
radical anions can stack as n-dimers, o-dimers or as n—columns. In terms of
the linkage of the TCNQ ligand to metal centers can be nl-o-bonded (14), o-
trans-u-bonded (5), o-cis-tt-bonded (9), o-syn-u-bonded (10) and o-w-
bonded (7), (8). When the n—stacking distance of the TCNQ rings is shorter
than the 3.4 A van der Waal distance for aromatic rings, the two TCNQ
quinoid rings are considered to be strongly interacting. The o-dimer of
TCNQ is the extreme case of an interaction in which two the quinoid

carbon atoms are actually linked by a weak o-bond. When the n—stacking

156

.G: we 3388
5286 2: massage a: ..zoamvgozobmzH .6 282838 28%; amuse Emaauasm .mm 2:9..—

 

tr:
es;
bll
fin
the

(1)

is ac

SCVe

157

distance is longer than 3.4 A, the interaction between TCNQ rings is weak,
and, although this weak stacking is crucial for holding the structure together,
it is not significant for enhancing magnetic interactions or engendering
electrical conductivity.

(b) Hydrogen bonding and weak interactions.

Hydrogen bonding networks are an important feature of many
supramoleculear architectures. In binary metal TCNQ coordination
polymers crystallized from protonic solvents, hydrogen bonding also plays
an important role in the packing. Figure 55 shows the hydrogen network
found in compound (9) in which hydrogen bonding from the methanol
ligand helps to stabilize the (it-TCNQ)? dimer anion.

(c) Bond length relationship and the degree of charge transfer in the
TCNQ ligand.

The bond lengths in the TCNQ quinoid ring are sensitive to the degree
of charge transfer between the cations and the TCNQ anions. Several
methods have been used in the literature to correlate the degree of charge
transfer to the bond lengths in outer-sphere TCNQ molecules. These are not
especially helpful, however when TCNQ is acting as a ligand, as different
binding modes also contribute to the bond length differences in the quinoid
rings. Table 19 summarizes examples of different kinds of TCNQ ligands in
the Mn/T CN Q MeOH and H20 structures.

E. Magnetic Measurements.
(l) Cu(TCNQ) phases.

The discovery that Cu(TCNQ) exists as two different structural types
is actually not surprising, as numerous TCNQ materials are known to exhibit

several different phases with the same chemical composition.“37 What is

 

11111 «5...... 13,575 c.. 0:2: TCZCL. LCk mtctowfigtm metmkosm. .MiN w~n-wrh

baggage moo—866 wagon: Hobs—H85 98 5.332: *

 

 

 

 

158

 

 

 

 

 

 

 

 

98a. 388x??? 33588:: 358% FHOZUE HZ; mL
auras." asnoe_n-§m *3 4m 21 Aonmfiozubéé
ascousab afimoezvegozofiozub520.55,:

335888 8a .8.” zufixozub:moezxozubaéeé

835858 mmd 3020.5de

2382; N2“ 2028:9338

835388 Nam Eozobuofievsufl

68% 83588:: -.m NAOZUCAEV

Ens—8 335388: 2 .m AOZUCQM

58% Baas 835288: $23 _ozutazamz
£58 magofim 258 mama—HS A<voo§m€ 35::an 3:09:00

 

 

 

 

 

@8325,“ @8028 E 83: .-OZUH H8 $08885 mac—08m. .MN 039—.

0.95

159

Table 24. Hydrogen bonds (with esds except for fixed and riding atoms) for
[Mn(TCNQ)2(H20)2]~.(10)-

 

 

 

D-H H...A D...A <(DHA)
0.81 (3) 209(3) 2.896(2) 169(3) Ol-HlW...N3
O.77(4) 2.06(4) 2.819(3) 170(3) 01 -H2W...N1

 

Table 25. Hydrogen bonds (with esds except for fixed and riding atoms) for
[Mn(TCNQ)(TCNQ-TCNQ)0.5(MeOH)2]oo (9)-

 

 

 

D-H H...A D...A <(DHA)
0.84 1.98 2.822(4) 178.1 01M-
H1M...N16B
0.84 2.04 2.863(4) 166.3 02M-
H2M_a...N 14B

 

Table 26. Hydrogen bonds (with esds except for fixed and riding atoms) for
[Zn(H20)4(n‘-TCNQ)2-TCNQ-2MeCN].. (13).

 

 

D-H H...A D...A <(DHA)
0.84 1.99 2.820(4) 167.6 0 1_a-
H1X_a...N9_a
0.96 2.18 3.057(4) 152.6 Ol__a-
H1 Y_a...N3
0.84 2.02 2.818(4) 159.6 O2_a-
H2X_a...N10__a
0.95 2.06 2.931(4) - 151.2 02_a-

H2Y_a...N8

 

160

680:8 mam—2:5 sumac? 05 28 3v 5 :5 83020.23 05 we 52.35358 GEE .vm 0.53m

    

.. .28 <25...
:3 a <25 8.“. 3::
ac: K
. ..ooiz <me <m8 <m o<<2 . . o
<22 <mz u“ <m8 omez .<<8 .. e e
. <90 . . <8. .95....1 ..
c (550 «<50 I¢<mmo ommwo <<moc Q o I
<3. . e e <8 Immfleo . e e ..
<8 m8 mmome.) - .. .me <<...o . .
o <8. «<5 9 m5... ¢o<mmz o o
9 <3 . . <8 . oez . m8 . 88 e . <<mz
<2 9.2 <82
<32 . . . e
22 .25 . <28

25 .. <28. .

161

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

~
s
‘s

 

 

 

 

 

 

 

 

Figure 55. Extended hydrogen bonding networks in [Mn(TCNQ)2(H20)4]..
(10)

 

162

 

Figure 56. Hydrogen bonding networks in [Zn(nl-TCNQ)2(H20)4(TCNQ) °2MeCN].. (13).

163

important about this observation, however, are the ramifications that it has in
terms of the observed properties of various samples of Cu(TCNQ),
especially conductivity and magnetic coupling. The variable temperature
magnetic susceptibilities of the samples were measured by the SQUID
technique. Plots of x versus T and “eff versus T are given in Figures 57 and
58. Phase I shows behavior typical of a TCNQ radical anion salt with
TCNQ stacking viz., strong coupling of the unpaired spins. Most simple
TCNQ radical anion salts are paramagnetic, but the susceptibilities are only
about 10% of that expected for a system with non-interacting spins.37 The
very low susceptibility of phase I (Figure 57) indicates considerable
magnetic coupling of the unpaired electrons which renders the sample
essentially diamagnetic. This is consistent with early studies on the
magnetism of chemically prepared Cu(TCNQ).38 In phase II, on the other
hand, the TCNQ'l radicals behave like isolated paramagnets in the high
temperature regime; in fact the um of 1.58 B.M. at 320 K is very close to
that of the spin only value for an isolated S = 1/2 system (Figure 57). At low
temperatures the effective magnetic moment for phase II begins to slowly
decrease, indicating the onset of weak antiferromagnetic interactions
between the unpaired spins which at 10 K abruptly changes to an ordering of
the ferromagnetic type. In support of an ordering are the results of the zero-
field cooled and field-cooled experiments (Figure 58c) and the hysteresis
loop observed at 2 K with a coercive field of 500 Gauss (Figure 58d).

(2) Mn samples.

Temperature dependent magnetic susceptibilities (x) for
[Mn(TCNQ)2(HzO)2].. and [Mn(TCNQ)(TCNQ-TCNQ)0_5(MeOH)2].. were
measured from 5 to 300 K in a dc. SQUID magnetometer. Plots of xmlm",
um and llxmolm" versus temperature are provided in Figures 59.

164

 

 

 

 

 

 

 

 

 

 

2103 1 . ,..2.,...-, . ”2 T. .4
.3 ..
"8 ‘0 Cu (TCNQ) phase I i
1.610'3 1
A 1.4103 _:
‘5 1.2 10-3 i:
S 1103 _.
I8, 3104 _:
V 1
>3 610" _
4104 *1
210,. 10000 G 1?
_210—4 L1 I I4 I I I I l I l I I l I I I I l I I4 I l I I I I l I I I
0 so 100 150 200 250 300 350
T (K)
(a)
2.510-2 fiIW.
Cu (TCNQ) phaseI - 800
2102 ‘
9 —————> e.
"a -. - 600 5?
E 1.5 101 _- . __
9 . 6‘
5 ~ . .0. a
‘1 1102 f B
.E . a
. <— V
1 r 200
5103
I SOOG i
01m Ll I L I A4 I l l 4 I :1. I f lirf 0
0 so 100 150 200 250 300
T(K)
(b)

Figure 57. Variable temperature susceptibility data for Cu(TCNQ) (a) x
versus T for phase I (7); (b) x and “eff versus T for phase II (8).

 

 

 

 

 

0.20 . f. r T I 1 . . 1 . r . . , . . I r
- Cu (TCNQ)phaseII‘
" FC 00 1
I.“ 0.15f °° -
B 1- 0 j
a _ . 1
5
0.10 - o ..
E ; -
x ~ °
0.050 P ..
. o d
ZFC "fin" 100G:
. 0000000.... 00000000 0 o O o I}
0.0 I I I41 I I I I 1 I I I I 1 I I I I
0 5 10 15 20
T(K)
(c)
0.03 r' 11111 r11 rfirv I v I I v I l v 1 1 Tfi I l I [j rd
1' .‘
: Cu (TCNQ) phase II , : 3 ‘3
0.02? ..:o .2
. ...o
: ’9‘...
0.01:- / .0. '7
A .'.
an .0
:2 0: .../.° 4
-0.01L .°' ...- ..
C 0.0’... d
-002} .::°' J
53' 2K 1
_003'III1IIILIII1III1III1III1III1II

 

 

 

H (kGauss)

(11)

Figure 58. Variable temperature susceptibility data for Cu(TCNQ) (c) zero-
field cooled and field-cooled x versus T for phase II (8); (d)
hysteresis of phase 11(8).

166

The results of these studies indicate that the Mn compounds exhibit Curie-
Weiss behavior (neg are ~ 5.9 B.M.) with a small contribution from
antiferromagnetic coupling at low temperatures. The conclusion that the
TCNQ radicals do not contribute to the magnetism of these compounds is in
accord with the X-ray structures that reveal the presence of o—bonded
[TCNQ-TCNQ? and/or at stacked [TCNQ]" moieties. Both types of
TCNQ-«TCNQ interactions lead to spin pairing of the radicals. The fact
that the n—stacked [TCNQI' ligands in the structure of [Mn(TCNQ)(TCNQ-
TCNQ)0,5(MeOH)2].. are diamagnetic was verified by a measurement of the
magnetic susceptibility of the isostructural Zn(II) analog which is rigorously

diamagnetic.39

F. Charge-Transport Properties TCNQ.
(l) Charge-transport properties of Cu(TCNQ).

The two structural forms of Cu(TCNQ) were subjected to pressed
pellet conductivity measurements, and it was found that they exhibit quite
different charge-transport properties as illustrated in the plots provided in
Figure 60. Both behave as semiconductors, but (7) has a room temperature
conductivity of 2.5 x 10'1 Scm’1 whereas (8) is nearly insulating with a room
temperature conductivity of only 1.3 x 10'5 Scm'l. The band gaps are 0.137
eV and 0.332 eV for phase I and II respectively. It is interesting to note that
these values are very close to the corresponding electrical properties of the
"unswitched" and "switched" forms of Cu(TCNQ) in the devices. Two
separate studies of the conductivity of Cu(TCNQ) prepared by the method of
Melby et. al., reported values of o (r.t.) = 1.6x10'2 Scm'l and 0(23 °C) =
1x10'2 ohrn'lcm'l and band gaps of 0.11 eV343 and Egap = 0.16eV.40 These
data are similar to what was found for phase I in this study. Clearly they

167

[Mn(TCNQJCNQ)”(TCNQxMeomzloo

 

  
   

 

 

 

 

 

 

 

 

 

 

7_-...,....,....,.....-.r.,....4so
I -‘10
(L5 : .
: -360
r 1
6 - I
t : .00..... 150
0 If C
15.5, 4,140 X
: i
p 130
5.” 1
I £20
4.5L 5
' c = 4.4990, 9 - 41.3714 . ro
: R8039” 3
4"IIIIrIIII1IIIIrLIII1II2I1IIII‘o
0 50 100 150 200 250 300
T(K)
[Mn(TCNQ)2(HzO)2]__
7 _r...,....1<.-.,....,....,.....‘7o
1
6;]..0000000 o o O 0 0 9 9 ’ :
:- «I
1: 1
=5 5: £40
:1. * < z
I : x
4_ 130
I 1
; 120
3 1- I
t c = 4.8269, 9: 4.4375 5 1o
: R.==(L99995 1
2 IIII1IIII1IIIILIIII1IIII1III“n

0 50 100 150 200 250 300

T(K)

Figure 59. Magnetic data for Mn TCNQ compounds (9),(10).

Conductivity (s cm")

168

Conductivity of Cu(TCNQ)

 

 
  

 

 

 

 

 

100 T ' ' I I I HI I I I I I 1 1 1 l l 1T1 1 1 T I 1 1 1 1
10°1
10.2 E a
10'3 E —°—Phase 1 :3

E +Phase 2 2
10" s
10.. 4
10-6 IHIIZIIIIIIIIIIUII1IIII1II.II

0.003 0.005 0.007 0.009

l/Temperature (K'l)

Figure 60. Plot of conductivity 0 (S cm'l) versus temperature for bulk

Cu(TCNQ) phase I and phase 11 measured on pressed pellets.

169

exhibit a much higher conductivity than phase II, which was not recognized

as a distinctly different phase prior to the present investigation.

G. Concluding Remarks.

Nearly twenty years ago, it was reported that the application of an
electric field to a device consisting of a thin film of Cu(TCNQ) (5-20 um)
sandwiched between a Cu and Al electrode induces a switching of the
material from a high to a low resistance state at a critical threshold
potential.8 The device displays a memory effect in that it remains in the low
resistance state (“ON” state) for a short period of time in the absence of an
applied field. There is a controversy surrounding this phenomenon,
however, due to the fact that many researchers have reported an inability to
reproduce this behavior. After reviewing the body of literature on the
subject, we were led to suspect that subtle changes in the Cu(TCNQ) film
preparation could account for the differences in electrical properties of the
devices. The results of our investigations establish the existence of two
polymorphs of Cu[(TCNQ) by X-ray photoelectron spectroscopy, infrared
spectroscopy, powder X-ray diffraction techniques and single crystal X-ray
methods. The chemical composition and electronic characteristics of the
two phases of Cu(TCNQ) are identical, but the structures and properties are
quite different. Powder XRD studies revealed that both phases are present in
thin films of Cu(TCNQ) grown on copper substrates but that even subtle
differences in reaction conditions can lead to variable quantities of the two
phases. This is undoubtedly the cause of the reported inconsistencies in the
properties of Cu(TCNQ) devices.8’17'20 Conductivity data on pressed pellets
of bulk samples of the two phases support the conclusion that the material

responsible for switching in the films is phase II, which is a very poor

170

semiconductor. Phase I is a much better conductor; in fact its conductivity is
basically the same as that of the Cu(TCNQ) films in the switched state. In
our studies, no evidence for switching was observed for the pressed pellets,
but this is not unexpected given that the sample thickness is much greater
than one can achieve with the fihns. What we believe is happening in the
switching phenomenon is that phase H (centrosymmetric, resting state)
converts to another structure (non-centrosyrmnetric, electrically induced
phase) in an electric field. Such a piezoelectrically driven structural changes
would not be absurd given how “soft” these materials are. The next step in
this work will be to verify that a structural change occurs in the switched
state. To do this, one will need a very pure film of phase II for the device,
and a diffractometer devoted to measuring the film before, and during, the
switching. Technical difficulties have prevented us from performing this
experiment to date.

Our investigations of Mn(II) chemistry with TCNQ'l radicals reveal
that a major product in methanol is [Mn(TCNQ)(TCNQ-
TCNQ)0,5(MeOH)2].. with pI-[TCNQ-TCNQF' and cis-u-[TCNQI' bridges.
The latter structure is also exhibited by the Zn(II) derivative and
[Zn(TCNQ)(TCNQ-TCNQ)0_5(MeOH)2]....,37a The X-ray structure of the
layered material [Mn(TCNQ)2(H20)2].. revealed yet another type of TCNQ'l
binding, namely syn—u-[TCNQI' bridges that participate in strong intra-layer
n-stacking (~3.1 A). These results underscore the importance of solvent and
the role of supramolecular interactions (hydrogen-bonding and n-
interactions) in stabilizing a particular phase. The diamagnetic o-[TCNQ-
TCNQ]2' and n—[TCNQhZ’ bridges serve to magnetically isolate the S = 5/2
Mn(II) centers, as evidenced by the Curie-Weiss behavior of the compounds
between 5 and 300 K01...” ~5.9 B.M.).

171

References

(a) Fagan, P. J.; Ward, M. D.; Calabrese, J. C. J. Am. Chem. Soc.,
1989, 111, 1698. (b) Iwamoto, T. in "Inclusion Compounds:
Inorganic and Physical Aspects of Inclusion" Iwamoto, T. Atwood, J.
L.; Davies, J. E. D.; MacNicol, D. D, Eds., Oxford University Press:
Oxford. 1991, 5, Ch. 6, 177. (c) Robson, R.; Abrahms, B. F.; Batten,
S. R.; Gable, R. W.; Hoskins, B. F.; Liu, J. in Supramolecular
Architecture; Bein, T.; Ed.; American Chemical Society: Washington,
DC, 1992, 256. (d) Tannenbaum, R. Chem. Mater. 1994, 6, 550. (e)
Constable, E. C.; Prog. Inorg. Chem, 1994, 42, 67. (f) Lu, J.;
Harrison, W. T. A.; Jacobson, A. J. Angew Chem Int. Ed. Engl. 1995,
34, 2557. (g) Dunbar, K. R., Heintz, K. R. Prog. Inorg. Chem, 1996,
283. (h) Kawata, S.; Kitagawa, S.; Kumagai, H.; Kudo, C.;
Kamesaki, H.; Ishiyama, T.; Suzuki, R.; Kondo, M.; Katada, M.
Inorg. Chem, 1996, 35, 4449. (i) Whiteford, J. A.; Rachlin, E. M.;
Stang, P. J. Angew. Chem. Int. Ed. Engl., 1996, 35, 2524. (i) Sharma,
C V. K.; Zaworotko, M. J. Chem. Commun., 1996, 2655. (k) Hirsch,
K. A., Wilson, S. R.; Moore, J. S. Inorg. Chem, 1997, 36, 2960.

(a) Gardner, G. B.; Venkataraman, D.; Moore, J. S.; Lee, S. Nature,
1995, 374, 792-795. (b) Yaghi, O. M.; Li, G.; Li, H. Nature, 1995,
378, 703. (c) Whiteford, J. A.; Rachlin, E. M.; Stang, P. J. Angew.
Chem Int. Ed. Engl. 1996, 35, 2524. (d) Yaghi, O. M.; Li, H. J. Am.
Chem. Soc., 1995, 117, 10401. (f) Yaghi, O. M.; Li, H. J. Am.
Chem. Soc., 1996, 118, 295. (g) Venkataraman, D.; Gardner, G. B.;
Lee, S.; Moore, J. S. J. Am. Chem. Soc., 1995, 117, 11601. (h) Yaghi,
O. M.; Li, H.; Groy, T. L. J. Am. Chem. Soc., 1996, 118, 9096. (i)
Olenyuk, B.; Whiteford, J. A.; Stang, P. J. J. Am. Chem. Soc., 1996,
118, 8221.

(a) Manriquez, J. M.; Yee, G. T.; McLean, S.; Epstein, A. J .; Miller, J.
S. Science 1991, 252, 1415-1417. (b) Tamaki, H.; Zhuang, Z. J.;
Matsumoto, N .; Kida, S.; Koikawa, M.; Achiwa, Hashimoto, Y.;
Okawa, H. J. Am Chem. Soc., 1992, 114, 6974. (c) Stumpf, H. 0.;
Pei, Y.; Kahn, 0.; Sletten, J .; Renard, J. P. J. Am. Chem. Soc., 1993,
115, 6738. (d) Inoue, K.; Iwamura, H. J. Am. Chem. Soc., 1994, 116,
3173. (e) Ohba, M.; Maruono, N.; Okawa, H.; Enoki, T.; Latour, J .-
M. J. Am. Chem. Soc., 1994, 116, 11566-11567. (f) Kahn, O. in

172

Molecular Magnetism: From Molecular Assemblies to the Devices:
NATO ASI Series. Eds., Coronado, E.; Delhaes, Gatteschi, D.; Miller,
J. S. Kluwer Academic, Dordrecht, 1996, 321, pp. 243-288. (g)
Decurtins, S.; Schmalle, H. W.; Schneuwly, P.; Zheng, Li-M.;
Ensling, J .; Hauser, A. Inorg. Chem, 1995, 34, 5501. (h) Miyasaka,
H.; Matsumoto, N.; Okawa, H.; Re, N.; Gallo, E.; Floriani, C. Angew.
Chem Int. Ed. Engl. 1995, 34, 1446-1448. (i) Ohba, M.; Okawa, H.;
Ito, T.; Ohto, A. J. Chem. Soc. Chem Commun., 1995, 1545-1546. 0)
Michaut, C.; Ouahab, L.; Bergerat, P.; Kahn, O.; Bousseksou, A. J.
Am. Chem. Soc., 1996, 118, 3610. (k) de Munno, G. ; Poerio, T.;
Viau, G.; Julve, M.; Lloret, F.; Joumaux, Y.; Riviere, E. Chem
Commun., 1996, 2587.

(a) Lacroix, P.; Kahn, O.; Gliezes, A.; Valade, L.; Cassoux, P. Nouv.
J. de Chimie, 1985, 643-651. (b) Gross, R.; Kaim, W. Angew. Chem
Int. Ed. Engl. 1987, 26, 251. (c) Bartley, S. L.; Dunbar, K. R. Angew.
Chem Int. Ed. Eng. 1991, 30, 448. (d) Ballester, L.; Barral, M.;
Gutierrez, A.; Jiménez-Aparicio, R.; Martinez-Muyo, J .; Perpifian,
M.; Monge, M.; Ruiz-Valero, C. J. Chem. Soc. Chem. Commun.
1991, 1396-1397. (e) Humphrey, D.G.; Fallon, G.D.; Murray, K.S. J.
Chem. Soc., Chem. Commun. 1988, 1356. (f) Comelissen, J.P.; van
Diemen, J. H.; Groeneveld, L. R.; Haasnoot, J. G.; Spek, A. L.;
Reedijk, J. Inorg. Chem. 1992, 31, 198-202. (g) Oshio, H.; Ino, E.;
Mogi, 1,; Ito, T. Inorg. Chem, 1993, 32, 5697-5703. (h) Ballester, L.;
Barral, M.; Gutierrez, A.; Monge, A.; Perpifian, M. F.; Ruiz-Valero,
C.; Sénchez-Pélaez, A. Inorg. Chem 1994, 33, 2142-2146. (i)
Dunbar, K. R.; Ouyang, X. Molecular Crystals Liquid Crystals, 1995,
273, 21-28. 0) Oshio, H.; Inc, B; Ito, T.; Maeda, Y. Bull. Chem Soc.
Jpn., 1995, 68, 889. (k) Dunbar, K. R. Angew Chem, 1996, 35,
1659. (1) Decurtins, S.; Dunbar, K. R. ; Gomez-Garcia, C. J .; Mallah,
T.; Raptis, R. G.; Talharn, D.; Veciana, J. in Molecular Magnetism:
From Molecular Assemblies to the Devices. (Eds.: E. Coronado, P.
Delhaes, D. Gatteschi, J. S. Miller) NATO ASI Series vol. E321
Kluwer, 1996, 571-582. (m) Dunbar, K. R.; Ouyang, X. Chem
Commun., 1996, 2427. (n) Zhao, H.; Heintz, R. A.; Rogers, R. D.;
Dunbar, K. R. J. Am Chem. Soc., 1996, 118, 12844. (0) Dunbar, K.
R.; Ouyang, X. Inorg. Chem, 1996, 35, 7188. (p) Azcondo, M. T.;
Ballester, L.; Gutierrez, A.; Perpifian, F.; Amador, U.; Ruiz-Valero,
C.; Bellitto, C. J. Chem. Soc., Dalton Trans. 1996, 3015.

10.

11.

173

(a) Aumfiller, A.; Erk, P.; Klebe, G.; Hunig, S.; von Schlitz, J.;
Werner, H. Angew. Chem. Int. Ed. Engl. 1986, 25, 740-741. (b)
Aumfiller, A.; Erk, P.; Hiinig, S. Mol. Cryst. Liq. Cryst. Inc. Nonlin.
Opt. 1988, 156, 215-221. (c) Erk, P.; Gross, H.-J.; Hfinig, U. L.;
Meixner, H.; Werner, H.-P.; von Schiltz, J. U.; Wolf, H. C. Angew.
Chem. Int. Ed. Engl. 1989, 28, 1245-1246. (d) Kato, R.; Kobayashi,
H.; Kobayashi, A. J. Am. Chem. Soc., 1989, 111, 5224-5232. (e)
Aumiiller, A.; Erk, P.; Hfinig, S.; Hadicke, E.; Peters, K.; von
Schnering, H. G. Chem Ber. 1991, 124, 2001.(f) Sinzger, K.; Hunig,
S.; Jopp, M.; Bauer, D.; Bietsch, W.; von Schlitz, J. U.; Wolf, H. C;
Kremer, R. K.; Metzenthin, T.; Bau, R.; Khan, S. 1.; Lindbaum, A.;
Lengauer, C. L.; Tillmanns, E. J. Am. Chem. Soc. 1993, 115, 7696.
(g) Kato, R.; Kobayashi, H.; Kobayashi, A. J. Am. Chem. Soc. 1989,
III, 5224.

An excellent review on the subject of sigma coordination to TCNX
molecules is: Kaim, W.; Moscherosch, M. Coord. Chem. Rev. 1994,
129, 157.

Structure of Ag(TCNQ): Shields, L. J. Chem. Soc., Faraday Trans. 2
1985, 81, 1.

Potember, R. S.; Poehler, T. O.; Cowan, D. O. Appl. Phys. Lett. 1979,
34, 405.

(a) Potember, R. S.; Poehler, T. O.; Hoffman, R. C.; Speck, K.R.;
Benson, R. C. In Molecular Electronic Devices II Carter, F. L., Ed.;
Marcel Dekker: New York. 1987, 91. (b) Potember, R. S.; Poehler, T.
O.; Cowan, D. 0.; Carter, F. L.; Brant, P. I Molecular Electronic
Devices Carter, F. L., Ed.; Marcel Dekker: New York. 1982, 73. (c)
Kamitsos, E. I.; Risen, W. M. Jr. Solid State Commun. 1983, 45, 165.
(d) Potember, R. S.; Poehler, T. O.; Cowan, D. O.; Brant, P.; Carter,
F. L.; Bloch. A. N. Chem. Scripta 1981, 17, 219. (e) Kamitsos, E. I.;
Risen, W. M., Jr. Solid State Commun. 1982, 42, 561.

Kamitsos, E.I.; Risen, W.M. Mol. Cryst. Liq. Cryst. 1986, I34, 31.
(a) Poehler, T. O.; Potember, R. S.; Hoffman, R.; Benson, R. C. Mol.

Cryst. Liq. Cryst. 1984, 107, 91. (b) Potember, R. S.; Poeler, T. O.;
Benson, R. C. Appl. Phys. Lett. 1982, 41, 548.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

174

Kamitsos, E. I.; Risen, W. M. Jr. J. Chem. Phys. 1983, 79, 5808.

Benson, R. C.; Hoffman, R.C.; Potember, R. S.; Bourkoff, E.; Poehler,
T. 0. Appl. Phys. Lett. 1983, 42, 855.

(a) Yamaguchi, S.; Viands, C. A.; Potember, R. S. J. Vac. Sci.
Technol. 1991, 9, 1129. (b) Hoffman, R. C.; Potember, R.S. Applied
Optics 1989, 28(7), 1417. (c) Kamitsos, E.I.; Risen, W. M. Jr. J.
Chem. Phys. 1983, 79, 477.

Potember, R. S.; Poehler, T.O.; Rappa, A.; Cowan, D.O.; Bloch, A. N.
Synth. Metals 1982, 4, 371.

Potember, R. S.; Poehler, T. O.; Rappa, A.; Cowan, D. O.; Bloch, A.
N. J. Am. Chem. Soc. 1980, 102, 3659.

Duan, H.; Mays, M. D.; Cowan, D. O.; Kruger, J. Synth. Metals 1989,
28, C675.

Sato, C.; Wakarnatsu, S.; Tadokoro, K.; Ishii, K. J. Appl. Phys. 1990,
68(12), 6535. In this work, vapor deposition techniques were used to

produce films of Cu on a glass-ceramics substrate which was dipped
into a solution of TCNQ in acetone/CH3CN to yield Cu(TCNQ).

Hoagland, J. J .; Wang, X. D.; Hipps, K. W. Chem. Mater. 1993, 5, 54.

(a) Liu, S. G.; Liu, Y.Q.; Wu, P. J.; Zhu, D. B. Chem. Mater. 1996, 8,
2779. (b) Liu, S.-G; Liu, Y.-Q.; Zhu, D.-B. Thin Solid Films, 1996,
280, 271. (c) Sun, S. Q.; Wu, P. J.; Zhu, D. B. Solid State Commun.,
1996, 99, 237. ((1) Liu, S. G.; Liu, Y.Q.; Wu, P. J .; Zhu, D. B.; Tian,
H.; Chen, K.-C. Thin Solid Films, 1996, 289, 300.

(a) Wakida, S.; Ujihira, Y. Jpn. J. Appl. Physics 1988, 27, 1314. (b)
Hua, Z. Y.; Chen, G. R. Vacuum 1992, 43, 1019.

Duan, H.; Cowan, D. O.; Kruger, J. In Advanced Organic Solid State
Materials Chiang, L. Y; Chaikin, P.M.; Cowan, D. 0. Ed.; Materials
Research Society Symposium Proceedings Vol. 173; Materials
Research Society: Pittsburgh, Pennsylvania, 1990; p.165.

23.

24.

25.

26.

27.

28.

29.

175

(a) Melby, L. R.; Harder, R.J.; Hertler, W. R.; Mahler, W.; Benson, R.
E.; Mochel, W. E. J. Am. Chem. Soc. 1962, 84, 3374. (b) Sano, M.;
Ohta, T.; Akamatsu, H. Bull. Chem. Soc. Jpn. 1968, 41, 2204. (c)
Ikemoto, 1.; Thomas, J. M.; Kuroda, H. Bull. Chem. Soc. Jpn. 1973,
46, 2237. (d) Bozio, R.; Girlando, A.; Pecile, C. J. Chem Soc.
Faraday Trans., 1975, 71, 1237.

(a) Acker, D. S.; Harder, R. J .; Hertler, W. R.; Mahler, W.; Melby, L.
R.; Bensin, R. E.; Mochel, W. E. J. Am. Chem. Soc. 1960, 82, 6408.
(b) Melby, L. R.; Harder, R. J .; Hertler, W. R.; Mahler, W.; Benson,
R. E.; Mochel, W. E. J. Am. Chem. Soc. 1962, 84, 3374. (c)
Torrance, J. B. Acc. Chem. Res. 1979, 12, 79. (d) Endres H. In
Extended Linear Chain Compounds; Miller, J. S. Ed.; Plenum: New
York. 1983, Vol. 3, 263-312. (e) Wudl, F. Acc. Chem. Res., 1984, 17,
227. (f) Jérome, D. Science 1991, 252, 1509. (e) Bryce, M. R. Chem
Soc. Rev. 1991, 20, 355. (g) Williams, J. M.; Schultz, A. J.; Geiser,
U.; Carlson, K. D.; Kini, A. M.; Wang, H. H.; Kwok, W. -K.;
Whangbo, M. -H.; Schirber, J. E. Science 1991, 252, 1501. (h) Ward,
M. D. Electroanal Chem. 1988 I6, 181. (i) Martin, N.; Segura, J. L.;
Seoane, C. J. Mater. Chem. 1997, 7, 1661.

The method of Kubas, G.J. Inorg. Synth. 1979, I9, 90, was used with
the modification that 45-50% HBF4 was used instead of 60-65%
HPF6.

Hathaway, B. J .; Holah, D. G.; Underhill, A. E. J. Chem. Soc. 1962,
2444.

'I'WINDX, TWROT, TWUTIL and TWHKL are components of the
TWINNING program written by Dr. Robert Sparks, private
communication.

(a) Dunbar, K. R. J. Am Chem. Soc. 1988, 110, 8247. (b) Dunbar, K.
R.; Pence, L. E. Inorg. Chem, 1991, 30, 2018. (c) Bartley, S. L.;
Bernstein, S. N.; Dunbar, K. R. Inorg. Chim. Acta 1993, 213, 213-
231. (d) Dunbar, K. R. Comments Inorg. Chem, 1992, 13, 313-357.
(6) Dunbar, K. R. J. Cluster Science, 1994, 5, 125-144.

We note that it is not cost effective to use a Cu(II) starting material to

30.

31.

32.

33.

34.

35.

176

prepare Cu(TCNQ) because it consumes "Bu4N(TCNQ) as a reducing
agent and the product is not as pure as the compound obtained from
either of the other two methods, as judged by X-ray powder
diffraction techniques.

(a) Thompson, R. C.; Gujral, V. K.; Wagner, H. J .; Schwerdtfeger, C.
F. Phys. Stat. Sol. 1979, 53, 181. (b) Chain, E. E.; Kevill, D. N.;
Kimball, C. W.; Weber, L. W. J. Phys. Chem Solids 1976, 37, 817.
(c) Thompson, R. C.; Hoyano, Y.; Schwerdtfeger, C. F. Solid State
Commun. 1977, 23, 633. (d) Kathirgamanathan, P.; Rosseinsky, D. R.
J. C. S. Chem Commun. 1980, 839.

Ikemoto, 1.; Thomas, J. M.; Kuroda, H. Bull. Chem Soc., Jpn , 1973,
46, 2237.

Brant, P.; Fernando, Q. J. Inorg. Nucl. Chem, 1978, 64,45.

(a) Kobayashi, A.; Kato, R.; Kobayashi, H.; Mori, T.; Inokuchi, H.
Solid State Commun., 1987, 64, 45. (b) Yashihiro, Y.; Furukawa, Y.;
Kobayashi, A.; Tasumi, M.; Kato, R.; Kobayashi, H. J. Chem. Phys.,
1994, 100, 2449. (b) Willett, R. D.; Long, G., personal
communication.

(a) Inoue, M.; Inoue, M. B. J. Chem. Soc., Faraday Trans. 2 1985,
81, 539. (b) Inoue, M. B.; Inoue, M.; Fernando, Q.; Nebesny, K. W.
J. Phys. Chem. 1987, 91, 527. (c) Inoue, M.; Inoue, M. B. Inorg.
Chem. 1986, 25, 37. (d) Pukacki, W.; Pawlak, M.; Graja, A.; Lequan,
M.; Lequan, R. M. Inorg. Chem. 1987, 26, 1328. (e) Farges, I. P.;
Brau, A.; Dupuis, P. Solid State Commun. 1985, 54(6), 531. (f)
Chappell, J. S.; Bloch, A. N.; Bryden, A.; Maxfield, M.; Poehler, T.
O.; Cowan, D. O. J. Am. Chem. Soc. 1981, 103, 2442. (g) Van
Duyne, R. P.; Suchanski, M. R.; Lakovits, J. M.; Siedle, A. R.; Parks,
K. D.; Cotton, T. M. J. Am. Chem. Soc., 1979, 101, 2832. (h) Lunelli,
B.; Pecile, C. J. Chem. Phys., 1970, 52, 2375. (i) Bozio, R.; Girlando,
A.; Pecile, C. J. Chem. Far. Trans. 11, 1975, 71, 1237.

(a) Morosin, B.; Plastas, H. J.; Coleman, L. B.; Stewart J. M. Acta
Cryst. 1978, B54, 540. (b) Dong, V.; Endres, H.; Keller, H. J.;
Moroni, W.; Nothe, D. Acta Cryst. Sec. B 1977, B33, 2428. (c)
Hoffmann, S. K.; Corvan, P. J.; Singh, P.; Sethulekshmi, C. N.;

36.

37.

38.

39.

39.

177
Hatfield, W. E. J. Am. Chem. Soc. 1983, 105, 4608.

Kaim, W. and Moscherosch, M. Coord. Chem. Rev. 1994, 129, 157.

(a) Hoekstra, A.; Spoelder, T.; Vos, A. Acta Cryst. 1972, 328, 14. (b)
Murakami, M.; Yoshimura, S. Bull. Chem. Soc. Jpn. 1975, 48, 157.
(c) Konno, M.; Saito, Y. Acta Cryst. 1974, B30, 1294. (d) Konno, M.;
Saito, Y. Acta Cryst. 1975, B31, 2007. (c) Konno, M.; Ishii, T.; Saito,
Y. Acta Cryst. 1977, B33, 763. (d) Endres, H. In Extended Linear
Chain Compounds Vol. 3; Miller, J .S. Ed.; Plenun Press: New York,
1983; PP. 263-317.

(a) Kommandeur, J. In Low-Dimensional Cooperative Phenomena;
Keller, H.J. Ed.; Plenum Press: New York, 1975, 65. (b)
Schwertdfeger, C.F. Solid State Commun. 1977, 23, 621. (c) Siratori,
K.; Kondow, T. J. Phys. Chem. Solids 1978, 39, 225. (d)
Kommandeur, J. In The Physics and Chemistry of Low Dimensional
Solids; Alcacer, L. Ed.; D. Reidel Publishing: Dordrecht, Holland,
1980, 197. (e) Grossel, M. C.; Weston, S. C. Chem. Mater. 1996,
8(5), 977. (f) Grossel, M. C.; Weston, S. C. J. Chem. Soc., Chem.
Commun. 1992, 1510. (g) Hynes, R. C.; Morton, J.R.; Preston, K. F.;
Williams, A. J.; Evans, F.; Grossel, M. C.; Sutcliffe, L.H.; Weston, S.
C. J. Chem. Soc., Faraday Trans. 1991, 87(14), 2229. (h) Iida, Y.
Bull. Chem. Soc. Jpn. 1969, 42, 71.

(a) Zhao, H,; Heintz, R. A.; Ouyang, X.; Dunbar, K. R.; Campana, C.;
Rogers, R. D. Chem. Mater. 1999, 11, 736-746. (b) Dunbar, K. R.;
Cowen, J.; Zhao, H.; Heintz, R. A.; Ouyang, X.; Grandinetti, G.
NATO ASI: Supramolecular Engineering of Synthetic Metallic
Materials: Conductors and Magnets Ed: J. Veciana, Kluwer:
Dordrecht, 1999, in press. (c) O'Kane, S.; Dunbar, K. R. unpublished
results.

Siemons, W. J.; Bierstedt, P. E.; Kepler, R. G. J. Chem. Phy., 1963,
39, 3523.

CHAPTER IV

STRUCTURAL AND MAGNETIC STUDIES OF
DINUCLEAR AND TETRANU CLEAR
CLUSTERS OF 3-D TRANSITION METAL
HALIDES

178

179

1. Introduction

High spin polynuclear transition metal compounds are of interest for
their magnetic applications in biological1 as well as materials chemistry.2 In
particular, 3d transition metal cations are commonly used as spin carriers in
extended arrays with cooperative properties. In spite of the interest in such
chemistry, a building block approach to preparing extended structures of
metal halide magnetic compounds has not been undertaken. In this work,
we report the convenient preparation and X-ray structures of [M2X6]2’ and
M4X3(THF)6 precursors. These 3d transition metal halide dinuclear anions
and tetra nuclear clusters are to be used as building blocks in magnetic
mterials. In this chapter, we report the structures and magnetic properties of
manganese halide clusters with THF, dimetal compounds of the type
[M2C16]2' and products from reactions of Fe4C13(THF)3 and [Fe2C16]2‘ with

nitrogen donor chelates.
2. Experimental Section.

A. Synthesis.

The starting materials MnClz, FeClz, and COO; were purchased from
Strem Chemicals, Inc., and used without further purification. [Et4N]Cl
(tetraethylammonium chloride) was purchased from Sigma Chemical Co.
and [ppn]Cl (ppn = bis(tn'phenylphosphoniumfimminium chloride) was
purchased from Aldrich; all were used as received. Acetone was distilled
over 3 A molecular sieves; diethyl ether, hexanes and THF were distilled
over sodium-potassium/benzophenone, whereas methanol was distilled over
Mg(OMe)2 under a nitrogen atmosphere. Unless otherwise specified, all
reactions were carried out under an argon atmosphere by using standard

Schlenk-line techniques. Due to the extreme moisture sensitivity of these

180

compounds, all glassware was pre-treated with the commercially available
reagent Glassclad.
(1) Preparation of [Mn5Clm('I‘HF)3],o (16).
(a) Bulk preparation

MnClz anhydride (0.500 g, 0.397 mmol) was loaded into a three-
necked Schlenk flask with 40 mL of THF, and was refluxed for 24 hours. A
white precipitate was collected by filtration and washed with 5 mL of THF
followed by 5 mL of diethyl ether and vacuum dried. Yield 0.76g (79%
based on Mn5C110(THF)8). IR (CsI, Nujol, cm") v(Mn-Cl) = 361(br).
(b) Single crystal growth

A saturated solution of (16) in 20 mL of acetone/0.5 mL of methanol
was carefully layered with 2 mL of a buffer solution (acetone/methanol 40:1
v/v) followed by 10 mL of hexanes in a Schlenk tube. The solution was
allowed to stand undisturbed, and after three days a small quantity of needle
crystals was obtained. IR (CsI, Nujol, cm'l) v(Mn-Cl) = 366 (br).
(2) Preparation of [MnBrKI‘HFMw (17).
(a) Bulk preparation

MnBrz anhydride (0.94g) was suspended in 40 mL of THF in a three-
necked Schlenk flask and refluxed for 12 hours. The yellow-brown solution
was reduced in volume to about one-half and layered with 40 mL of
hexanes. The resulting pale pink crystalline product was collected by
filtration and washed with 5 mL of THF followed by 5 mL of diethyl ether
and finally dried in a vacuum; yield: 1.24g (72%).
(b) Single crystal growth

Pale pink needle crystals of (17) suitable for a crystallographic study
were grown by carefully layering 20 mL of a saturated acetone solution of

MnBrz-xTI-IF with 10 mL of hexanes.

181

(3) Preparation of crystals of [MnCl2(MeOH)2],,(l8).

A saturated solution of [Mn5C110(THF)g],, in 20 mL of acetone and 0.5
mL of methanol was carefully layered with a 5 mL mixture of acetone and
methanol (40:1 v/v) followed by 10 mL of hexanes in a Schlenk tube. The
solution was allowed to stand undisturbed, and long needle crystals of
[MnC12(MeOH)2],, were observed to form over a three day period. The
crystals tend to be severely twinned, and they lose interstitial solvent of
crystallization immediately upon removal fiom the mother liquor. Nujol
mull (cm'l), v(Mn-Cl) = 387 (br).

(4) Preparation of [FezCl4(bpy)2] from FC4C13(THF)4 (19).

Stock solutions of 2,2'-bipyridine (10.0 mg, 0.064 mol 5 mL of
acetone) and Fe4Clg(TI-IF)6 (0.400 g, 0.426 mmol 10 mL of acetone) were
prepared. A 1 mL aliquot of the Fe4C18(THF)6 solution was syringed into a
6 mm O.D. Pyrex tube, and carefillly layered with 0.5 mL of acetone
followed by 1 mL of the 2,2’-bipyridine solution. The tube was flame-
sealed under a slight vacuum and allowed to stand undisturbed. After
several days, orange crystals suitable for crystallographic study were
harvested from the interface region of the tube. [R (CsI, Nujol, cm'l): 417,
359, 317, 250. Anal. Calcd.: %C, 42.45; %H, 2.85; %N, 9.90. Found: %C,
41.70; %H, 3.19; %N, 9.17.

(5) Preparation of [FeCI2(bpy)].,o from Fe4Cls(THF)4 (20).

Stock solutions of 2,2'-bipyridine (6.0 mg, 0.038 mmol in 5 mL of
acetone) and Fe4Clg(TI-IF)6 (0.380 g, 0.404 mmol in 15 mL of acetone) were
prepared. A 1 mL aliquot of the Fe4Clg(TI-IF)6 solution was syringed into a
6 mm O.D. Pyrex tube and carefully layered with 1 mL of acetone followed
by 1 mL of the 2,2’-bipyridine solution. The tube was flame-sealed under a
slight vacuum and allowed to stand undisturbed. Afier several days, red

182

needle crystals suitable for crystallographic study were harvested from the
interface region of the tube.
(6) Preparation of [ppn]2[CozCl6] (21).

Anhydrous CoClz (0.260 g, 2 mmol) and [ppn]Cl (1.144 g, 2 mmol)
were dissolved in 60 mL of acetone and stirred at r.t. for 16 hours. A blue
solution and a large quantity of blue solid were present at the end of this
time. The blue product was collected by filtration, and the filtrate was
treated with 40 mL of diethyl ether to yield additional microcrystalline
product; combined yield, 1.02 g (67%). Single crystals were grown by slow
diffusion of diethyl ether into a solution of the compound in acetone. IR
(Nujol, cm") 1705 (ms), 1587 (m), 1261 (br), 1111 (m), 1026 (m), 997 (m),
' 798 (m), 762 (m), 742 (m), 690 (s), 550 (s), 532 (s), 495 (s), 395 (ms), 333
(ms), 306 (ms), 268 (br). Anal. Calcd. for C02C16P4N2C72H50: C, 61.43; H,
4.30; CI, 15.11; N, 1.99. Found: C, 61.46; H, 4.37; N, 1.79.

(7) Preparation of [ppn]2[Mn2Cl6] (22).

Anhydrous MnClz (0.252 g, 2 mmol) and [ppn]Cl (1.145 g, 2 mmol)
were dissolved in 35 mL of acetone and stirred for 16 hours. The resulting
colorless solution was filtered to remove any insoluble particles, and the
filtrate was treated with 40 mL of diethyl ether to give a pale yellow-green
microcrystalline product which was collected, washed with 5 mL of diethyl
ether and dried in vacuo; yield, 1.14 g (97%). Single crystals were obtained
by dissolving [ppn]2[Mn2C16] in 20 mL of acetone and layering with 10 mL
of hexanes in a Schlenk tube. Light yellow, rhombohedral crystals were
harvested after three days. IR (Nujol, cm'l) 1707 (m), 1587 (m), 1265 (br),
1180 (m), 1113 (m, s), 1026 (m), 997 (s), 800 (m, s), 748 (s), 694 (s), 551
(s), 532 (s), 499 (s), 396 (m, s), 330 (s), 287 (m, s), 248 (m, s), 228 (m, 3).
Anal. Calcd. for Mn2C16P4N2C72H60: C, 61.78; H, 4.32; CI, 15.20; N, 2.00.

183

Found: C, 61.65; H, 4.56; N, 1.74.
(8) Preparation of [ppn]2[NiCl4] (23, 24).
(a) Bulk Preparation.

Samples of anhydrous NiClz (0.260 g, 2 mmol) and [ppn]Cl (1.148 g,
2 mmol) were dissolved in 35 mL of acetone and stirred at r.t. for 16 hours.
A small quantity of light blue precipitate was removed by filtration and
discarded, and the filtrate was treated with 40 mL of diethyl ether to produce
a light blue microcrystalline material. The product was collected by
filtration, washed with 3 x 5 mL of diethyl ether and vacuum dried. Yield
0.686 g (47%).

(b) Single crystals of phase I (23).

Blue needle crystals of (23) of an orthorhombic form were obtained
by dissolving 50 mg of the salt in 20 mL of acetone and layering the solution
with 10 mL of diethyl ether.

(c) Composite crystals of a triclinic phase (24) and orthorhombic
phase (23).

Blue platelet crystals of a triclinic phase of [ppn]2[NiCl4] (24) were
grown from layering 10 mL of hexanes on top of 20 mL of an acetone
solution that contained 50 mg of the compound.

(9) Preparation of [Mn2(bpym)4(u-F)(TIl-BF4)(BF4)2(M0CN)
~2MeCN],o (25).

Single crystals of (25) were grown by layering an acetonitrile solution
of [Mn(MeCN)4]2[BF4]2 with a MeCN solution of 2,2'-bypyrimidine. Stock
solutions of 2,2’-bipyridine (10 mg in 5 mL of acetonitrile) and
[Mn(MeCN)4]2[BF4]2 (0.20 g in 10 mL of acetonitrile) were prepared, and 1
mL of the metal containing solution was syringed into a 6 mm O.D. Pyrex
tube and carefully layered with 0.5 mL of acetonitrile followed by 1 mL of

184

the 2,2’-bipyridine stock solution. The tube was flame-sealed under a slight
vacuum and allowed to stand undisturbed. After several days, pale yellow
crystals suitable for crystallographic study were harvested from the interface
region of the tube.
(10) [EttN]C"[FezCMMeOIDAu-LT-bpymfl (26)-
(a) Bulk preparation.

A methanol solution (15 mL) of [Et4N]2[Fe2C16] (26) was added to 15
mL of a methanol solution of 2,2'-bipyrimidine, and the resulting solution
was stirred for 12 h to give a pale red solution. The solution was
concentrated and treated with diethyl ether (20 mL) to give a black solid;
yield: 0.0352 g, 36% based on 2,2'-bpym.
(b) Single crystal growth.

Single crystals of (26) were grown by slow diffusion of a MeOH
solution of [Et4N]2[Fe2Cl5] into a THF solution of 2,2'-bypyrimidine.

B. Structural Studies.

Crystallographic data for (16), (18) and (21) —(26) were collected on a
Bruker (Siemens) SMART 1K CCD platform diffractometer. Data for (l7),
(19) were collected on a Rigaku AFC6S four circle diffractometer. Data for
(20) was collected on a Nicolet P3/V four circle diffractometer. Detailed
procedures for using a Bruker CCD are described in Appendix A.

(1) Structure of [MnsClm(THF)8],, (16).

A colorless, rod-like crystal of approximate dimensions 0.60 x 0.29 x
0.29 mm was mounted on the tip of a glass fiber with Dow Corning silicone
grease. Indexing and refinement of 35 reflections from a total of 60 frames
with an exposure time of 10 sec/frame gave unit cell parameters for a

triclinic unit cell. A hemisphere of data with 1321 frames were collected

185

with a scan width of 0.3° in (0 and an exposure time of 30 sec/frame.
Indexing and refinement of 93 reflections from a total of 180 data frames
was used to generate a precise cell for data integration which led to 7,753
reflections in the range of —12 S h S 7, -14 s k S 14, -16 S l s 15 with a
maximum 20 angle of 56.96°. Of the 5400 unique reflections, a total of
3786 reflections with I > 20(1) and Rim = 0.0303 remained afier data
reduction. The final cell was produced from the refinement of the XY Z
centroids of 5400 reflection with 1/0' > 10. The unit cell parameters are a =
9.4620(2) A, b = 11.13870(2) A, c = 12.4763(2) A, a= 75.2850(10)°, ,6:
76.3050(10)°, y = 79.6870(10)° and V = 1225.72(3) A3. The data were
corrected for beam inhomogeneity, crystal decay and absorption by
SADABS3 which led to a transmission factors ranged fiom 0.83 to 1.00 and
an Rim=0.303.

The positions of the non-hydrogen atoms were located by direct
methods and refined by fiJll-matrix least squares on F2 using the SHELXTL
5.04 package. Two disordered THF molecules were modeled at half
occupancies in two positions. All non-hydrogen atoms were refined
anisotropically except the two disordered THF groups which were refined
isotropically. Hydrogen atoms were placed in idealized positions.
Crystallographic data for (16) are summarized in Table 27.

(2) Structure of [MnBr2(THF)2],, (17).

A pale pink, rod-like crystal of approximate dimensions 0.40 x 0.25 x
0.25 mm was mounted on the tip of a glass fiber with Dow Corning silicone
grease. Indexing and refinement of 20 reflections in the 20 range 25-45 led
to an orthorhombic crystal system with the unit cell parameters a = 7.593(2)
A, b = 18.375(4) A, c = 3.989(1) A, a = ,B= 7= 90°, and V= 556.6(2) A3.
The data were collected in the 26)-co scan mode to 50° in the range of 0 s h S

186

9, 0 S k S 21, 0 S l s 4 with a maximum 20 angle of49.91°. The data were
corrected for Lorentz and polarization effects which led to Rim = 0.115 and
Rsig = 0.0364.

Systematic absences indicated two possible orthorhombic space
groups, namely Pnnm and Pnn2. In the Pnnm space group, the THF
molecule is disordered around the mirror plane. In the space groups Pnn2,
the Mn and Br atoms positions were located by direct methods and refined
by full-matrix least squares on 17"2 using the SHELXTL 5.04 package. Other
non-hydrogen atoms were located from alternating difference Fourier maps
and least squares refinement. A disordered THF group was isotropically
refined in two positions, and the Mn and Br atoms were refined
anisotropically. Hydrogen atoms were fixed in idealized positions.
Crystallographic data for (17) are summarized in Table 27.

(3) [MnClz(MeOH)z],, (18)

A colorless thin needle crystal of dimensions 0.21 x 0.13 x 0.05 mm
was secured on the tip of a glass fiber with Dow Coming silicone grease and
placed in a cold N2(g) stream at 173(2) K. Indexing 8O reflections failed to
produce a solution with default parameters in the auto-indexing routine from
a total of 60 fi’ames with an exposure time of 10 sec/frame. Reindexing with
a reflection percentage fit ratio reduced of 70% (versus 80%) indicated a
monoclinic I-centered crystal system with cell parameters a = 24.14(3)A, b
= 3.69(5)A, c = 22.08(3)A, ,6 = 9727(6), and V= 1950.9(3.6)A3. A hemi-
sphere of data with 1321 frames was collected with a scan width of 0.3 in (o
and exposure times of 20 sec/frame. A further indexing afier data collection
failed to reproduce the monoclinic cell and indicated that crystal might
belong to a primitive triclinic system with a = 3.7288(5) A, b = 15.429(2) A,
c = 17.456(2) A, a = 83.969(2)°, ,8 = 83.875(2)°, y = 82.932(2)°, V =

187

986.64(37) A3. However, data integration with the triclinic cell cannot be
solved. The final cell was eventually indexed by the TWINNING4 package
which indicated the presence of quadruple rotational twins with a = 3.723(2)
A, b = 5.817(3) A, c = 7.695(3) A, a = 94.801(7)°, p = 96.252(8)°, y =
96.072(8)° and V = 163.96(21) A3. The twinning laws for other three
solutions related solution A are [-1 0 0, 0.33 1 0, 0 -O.25 -1], [-1 O O, O -1 O,
0.5 0.25 l] and [1 0 0, -0.33 -1 0, -0.5 0 1] respectively. The data were
integrated by using the orientation matrix of component A. Laue crystal
symmetry constraints were used during data integration, and the cell
parameters and orientation matrix were also refined. Summary of data
collection and refinement are listed in Table 28.
(4) FezCMbPYb (19)-

An orange needle crystal of approximate dimensions 0.20 x 0.20 x
0.30 mm3 was mounted on the tip of a glass fiber with Dow Corning silicone
grease. Indexing and refinement of 20 reflections in the 20 range 25-45°
gave unit cell parameters for a triclinic crystal system, a = 8.205 (2) A, b =
8.436 (2) A, c = 8.749 (2) A, a = 110.25(3)°, ,6 = 105.69(3)°, 7= 93.93 (3)°,
and V = 538.2(2) A3. The data were corrected for Lorentz and polarization
effects. The positions of the non-hydrogen atoms were located by direct
methods and refined by full-matrix least square on F2 using the SHELXTL
5.04 package whereas hydrogen atoms were placed in calculated positions.
Key crystallographic data are summarized in Table 29.
(5) FezClAbpy)... (20)-

A red needle crystal of approximate dimensions 0.35 x 0.26 x 0.26
mm was mounted on the tip of a glass fiber with Dow Corning silicone
grease. Indexing and refinement of 24 reflections in the 20 range 15-30°

gave unit cell parameters for a monoclinic C-centered crystal system

188

Table 27. Crystal data and structure refinement for [Mn5C1w(TI-IF)3]..(16)

 

 

and [MnBr2(TI-IF)2].. (17).
Identification code (16) (17)
Empirical formula C32H64C110MD503 C3H16 BI‘2MI'102
Formula weight 603.02 179.48
Temperature 133(2) K 173(2) K
Wavelength 0.71073 A 0.71073 A
Crystal system Triclinic Orthorhombic
Space group P-l Pnn2
Unit cell dimensions a = 9.4620(2) A a = 7.5930(15) A
b: 11.13870(10) A b= 18.375(4) A
c = 12.4763(2) A c = 3.9890(8) A
a: 75.2850(10)°. (1: fl: 7: 90°
[3 = 76.3050(10)°.
y: 79.6870(10)°.
Volume 1225.72(3) A3 556.5509) A3
Z 1 2
Density (calculated) 1.634 Mg/m3 2.142 Mg/m3
Absorption coefficient 1.840 mm'1 8.332 mm1
F (000) 615 350
Crystal size (mm?) 0.60 x 0.29 x 0.29 0.75 x 0.40 x 0.40
Reflections collected 7753 563
Absorption correction SADABS Scan
Refinement method Full-matrix on F2 Full-matrix on F2
Data/restraints/parameters 5400/ 100 / 242 563/ 1 I42
GOF on F2 0.964 1.113
R] (wR2) [1“2,I>20(I)] 0.0564(O.1446) 0.0506(0.1340)

Largest difference peak 1.024 and -1.407 e'IA3

and hole

Abs. Structure, Extinction

1.566 and 4021 67A3

0.04(7), 0.067(8)

 

R1 = 2: ||F,| - IF,“ / 2 |F,|, wR2 = (2 {ME} - F52] IE [w(F,2)2]}"2, GOF= s
= {lwtrf - 12.52] / (n - 12)} "’3 w = 1 / 16208.2) + <an + bP]

189

Table 28. Crystal data and structure refinement for [MnC12(MeOH)2].. (18).

 

 

Identification code (1 8)
Empirical formula C2H3C12Mn02
Formula weight 189.92
Temperature 143(2) K
Wavelength 0.71073 A
Crystal system Triclinic
Space group P-l
Unit cell dimensions a = 3.7232(16) A
b = 5.817(3) A
c = 7.695(3) A
a: 94.801(7)°.
fl = 96.252(8)°.
y: 96.072(8)°.
Volume 163.96(12) A3
Z 1
Density (calculated) 1.923 Mg/m3
Absorption coefficient 2.726 mm—l
F (000) 95
Crystal size (mm3) 0.10 x 0.20 x 0.5
Indexing method TWINNING
Reflections collected 1025
Absorption correction SADABS

Refinement method

Data / restraints / parameters
GOF on F2

Final R indices [I>2 0(1)]

R indices (all data)

Largest diff. peak and hole

Full-matrix least-squares on F2
720/ 0 / 34

3.703

R1 = 0.1756, wR2 = 0.4677

R1 = 0.1926, wR2 = 0.4721
5.857 and -1075 e/A3

 

R1 = 2 IIFoI - IF." I 2 IFOI. sz = {>3 [W(F02 - 1W] I 2 lwlF.2>’ll"’. 60F: s
= llwlF.2 - F371 / (n - 12)}"2. w = 1 I [02(Fez) + <an + bP]

190

with a = 17.93503) A, b = 9.071(7) A, c = 6.879(5) A, a = 90°,
)6 = 113.l7(10)°, y = 90°, and V = 1028.86(1.33) A3. Unique data of
1346reflections was collected by the (1)-20 technique between 4 5 20 5 55.
The data were corrected for Lorentz and polarization effects. The positions
of the non-hydrogen atoms were located by direct methods and refined by
full-matrix least square on I“2 using the SHELXTL 5.04 package whereas
hydrogen atoms were placed in calculated positions. Key crystallographic
data are summarized in Table 29.

(6) Structure of [ppn]2[CozC16] (21).

Single crystals of [ppn]2[C02Cl6] were grown by dissolving a bulk
sample of [ppn]2[CozC16] in 20 mL of acetone and layering with 10 mL of
hexanes in a Schlenk tube. Light blue crystals were harvested after three
days. A light blue diamond-shaped crystal of approximate dimensions
0.05 x 0.04 x 0.02 mm was mounted on the tip of a glass fiber with Dow
Corning silicone grease and placed in a N2 cool stream at 173(1)K. Indexing
and refinement of 53 reflections from a total of 60 frames with an exposure
time of 10 sec/frame indicated a monoclinic crystal system. A hemisphere
of data with 1321 frames was collected with a scan width of 0.3° in (l) and an
exposure time of 30 sec/frame. Indexing and refinement of 221 reflections
from a total of 200 data frames generated the precise cell for data integration
which led to 40,634 reflections in the range of -28 S h S 18, -17 S k S 17,
-31 S l S 31 with a maximum 20 angle of 56.60°. Final cell parameters were
generated from the refinement of the XYZ centroids of 6220 strong
reflections with I > 10 0(1). The unit cell parameters are a = 21.6564(2) A,
b = 13.1996(2) A, c = 23.85660) A, )3: 102.7312(3)°, and V = 6651.90(12)
A3. The intensities were corrected for beam inhomogeneity and crystal

absorption and decay with SADABS which led to transmission factors

191

Table 29. Crystal data and structure refinement for Fe2C14(bpy)2 (19) and

 

 

[FeCla(bpy)]..(20).
Identification code (19) (20)
Empirical formula C20H16C14F62N4 C10 H8 C12 Fe N2
Formula weight 282.93 282.93
Temperature 193(2) K 173(2) K
Wavelength 0.71073 A 0.71069 A
Crystal system Triclinic Monoclinic
Space group P—l C2/c
Unit cell dimensions a = 8.205(2) A a = 17.935(5) A
b = 8.436(2) A b = 9.071(5) A
c = 8.749(2) A c = 6.879(5) A
a: 110.25(3)° a: 90.000(5)°.
,6: 105.69(3)° ,6: 113.170(5)°
y: 93.93(3)° y: 90.000(5)°
Volume 538.2(2) A3 1028.9(10) A:
z 1 4
Density (calculated) 1.746 Mg/m3 1.827 Mg/m3
Absorption coefficient 1.860 mm'1 1.946 mm'
F(OOO) 284 568
Crystal size (mm3) 0.30 x 0.20 x 0.20 0.35 x 0.26 x 0.26
Reflections collected 2038 1223
Refinement method Full-matrix on F2 Full-matrix on F2
Data/restraints/parameters 1897 / 0 / 168 1 187 / 0 / 70
GOF on F2 1.074 1.060

Final R indices [I>26(I)]

R1 = 0.0314, wR2 =
0.0850

Largest diff. peak and hole 0.504 and -0.919 e'lA3

Extinction coefficient

R1 = 0.0776, wR2 =
0.1943

1.566 and -2.871 e'
/A3

0.008(3)

 

R1 = 2 ||Fo| - |Fc|| / z |F,|, wR2 = {2 [w(F02 - F052] IE [W(F02)

- Fc2)2] / (n - p)} "2, w = 1 I [020702) + 01102 + bP]

211‘”. 00F: s = {[W(F62

192

between 0.89 to 1.00. Of the 15565 unique reflections, a total of 8184
reflections with I > 20(1) and Rim = 0.0332 remained after data reduction.
Equivalent reflections were merged and truncated to 0.85A resolution to
reduce the high Rsigma caused by the small size of the crystal. The structure
was solved and refined with the SHELXTL 5.04 package. The positions of
all non-hydrogen atoms were located by direct methods and refined
anisotropically by full-matrix least squares on F2. Hydrogen atoms were
placed in idealized positions. The final full-matrix refinement was based on
11293 unique reflections (merged and truncated) and 775 parameters led to
R1 = 0.0452 and wR2 = 0.0789 (1 > 2 o). The goodness-of-fit index was
1.019 and the highest peak in the final difference map was 0.470 e'/A3. The
crystallographic data for (21) are summarized in Table 30.

(7) [ppnllenzClt] (22).

Single crystals were obtained by dissolving a bulk sample of
[ppn]2[Mn2Cl6] in 20 mL of acetone and layering with 10 mL of hexanes in a
Schlenk tube. Pale yellow rhombohedral crystals were harvested after three
days. A light yellow rhombohedral crystal of approximate dimensions
0.49 x 0.31 x 0.26 mm was mounted on the tip of a glass fiber with Dow
Corning silicone grease. Indexing and refinement of 274 reflections from a
total of 60 frames with an exposure time of 10 sec/frame gave unit cell
parameters. for a monoclinic unit cell. A hemisphere of data with 1321
frames was collected with a scan width of 0.3° in (0 and an exposure time of
30 sec/frame. Indexing and refinement of 300 reflections from a total of 100
data frames generated the precise cell for data integration which led to
41,252 reflections in the range of -18 S h S 28, -17 S k S 14, -31 SIS 31
with a maximum 20 angle of 56.66°. Final cell parameters were generated

from the refinement of the centroids of 8192 strong reflections with

193

1 > 10 0, are a = 21.70450) A, b = 13.2667(2) A, c = 2400410) A,
,6: 102.549(1)°, and V: 6746.80(29) A3. The reflections were corrected for
beam inhomogeneity, crystal absorption and decay with the program
SADABS which led transmission factors between 0.83 to 1.00. Of the
15,889 unique reflections, a total of 11,647 reflections with I > 20(1) and
Rim = 0.0321 remained after data reduction.

The positions of the non-hydrogen atoms were located by direct
methods and refined by full-matrix least squares on F2 using the SHELXTL
5.04 package. All non-hydrogen atoms were refined anisotropically.
Hydrogen atoms were placed in idealized positions. Crystallographic data
for (22) are summarized in Table 30.

(8) Structure of [ppnhNiCla orthorhombic phase (23).

A blue block-shaped crystal of approximate dimensions
0.52 x 0.31 x 0.18 mm was mounted on the tip of a glass fiber with
Dow Corning silicone grease and placed in a N2 cool stream at 173(1) K. A
hemisphere of data with 1321 frames was collected with a scan width of 0.3°
in (l) and an exposure time of 30 sec/frame. Indexing and refinement of 206
reflections from a total of 200 data frames generated the precise cell for data
integration which led to 21816 reflections in the range of -19 S h S 32,
-16 S k S 16, -14 S l S 14 with a maximum 20 angle of 56.72°. Final cell
parameters were generated from the refinement of the XYZ centroids of
strong reflections with I > 10 0(1). The unit cell parameters are
a = 24.467(2) A, b = 1269470) A, c = 10.8072(7) A, a = e: y: 90°, and
V = 3356.8(4) A3. The intensities were corrected for beam inhomogeneity,
crystal absorption and decay with SADABS which led to transmission
factors between 0.808 to 1.00 and Rim = 0.0452 for a total of 21,816
reflections (8,375 unique). The structure was solved and refined with the

194

SHELXTL 5.04 package. The positions of all non-hydrogen atoms were
located by direct methods and refined anisotropically by full-matrix least
squares on F3. Hydrogen atoms were placed in idealized positions and not
refined. The final full-matrix refinement was based on 4554 unique
reflections and 394 parameters led to R1 = 0.0396 and wR2 = 0.0787
(1 > 2 o). The goodness-of-fit index was 1.024 and the highest peak in the
final difference map was 0.51 e'lA3. The crystallographic data for (23) are
summarized in Table 31.
(9) Composite phases of [ppn]2NiCla with orthorhombic and
triclinic forms (23b, 24).

A hexagonal platelet of crystal (23b, 24) of approximate size 0.36 x
0.29 x 0.17 mm was examined on a SMART 1K CCD Area Detector and
indexed to give the triclinic cell a = 13.62000) A, b = 13.646(7) A,
c = 21.44303) A, a = 104.52(7)°, ,3 = 95.52(5)°, y = 115.35(5)°,
V = 3390.64(388) A3. A hemisphere of data was collected at 30sec/frame.
Without symmetry constraints the data gave a different cell of
a = 14.3861(3) A, b = 14.11480) A, c = 22.6380(5) A, a = 90.000001)°,
,6: 90.000002)°, y: 115.9618(6)°, v: 4132.9205) A3. After reprocessing
the data, one third of the original reflections used for indexing was rejected
during the least-square refinement as the error is bigger than tolerance, and
the cell reverted from the monoclinic setting back to a triclinic cell.

TWINDX4 was then used to index the data, and a triclinic cell was
indexed with default parameters. A second run with the remaining
reflections revealed that the rest of the reflections indexed to an
orthorhombic cell which happened to be the same as the cell for (23). This
is a very rare case in which one can actually index reflections that belong to

different crystal systems. A comparison of the orientation matrices and cell

195

Table 30. Crystal data and structure refinement for [ppn]2[CozC16] (21) and

 

 

for [PpnllenzClo] (22).

Identification code (21) (22)

Empirical formula C72H50C16C02N2P4 C72H60C16MD2N2P4

Formula weight 1407.66 1399.68

Temperature 173(2) K 133(2) K

Wavelength 0.71073 A 0.71073 A

Crystal system Monoclinic Monoclinic

Space group P21/n P21/n

Unit cell dimensions a = 21.6564(2) A a = 21.70450) A
b = 13.1996(2) A b = 13.2667(2) A
c = 23.8566000) A c = 2400410) A
,6: 102.73°. ,6: 102.55°.

Volume 6651.89(12) A3 6746.79(18) A3

Z 4 4

Density (calculated) 1.406 Mg/m3 1.378 Mg/m3

Absorption coefficient 0.880 mm"1 0.750 mm'1

F (000) 2888 2872

Crystal size, mm3 0.050x 0.040 x 0.020 0.49 x 0.31 x 0.26

Reflections collected 11293 41252

Absorption correction Multi-Scan, Multi-scan,
SADABS SADABS

Refinement method Full-matrix on F2 Full-matrix on 1“2

Data / restraints / parameters 1 1293 / 0 / 775 15889 / 0 / 775

GOF on F2 1.020 1.175

Final R indices [1 >20(I)] R1 = 0.0452, wR2 = R1 = 0.0411, wR2 =
0.0788 0.0815

Largest diff. peak and hole 0.470 and -0.386 0.481 and -0.471
e/A3 e/A3

 

R1 = z ||F,| - |Fc|| / 2 |F,|, wR2 = {z [W(F02 - F,2)2] / 2: [w(F,2)2]}"2,
goodness-of-fit = S = {[w(F02 - Fc2)2] / (n - p)} ”2, W = 1 / [02(F02) + (“Pf 4'

bP]

196

Table 31. Crystal data and structure refinement for [ppn]2NiC14-Me2CO

 

 

(23) and (24).
Identification code (23) (24)
Empirical formula C75H63C14N2N10P4 C75H66C12N2NIOP4
Formula weight 1337.70 1264.79
Temperature 133(2) K 293(2) K
Wavelength 0.71073 A 0.71073 A
Crystal system Orthorhombic Triclinic
Space group P21212 P-1
Unit cell dimensions a = 24.467705) A a = 13.661(3) A

Volume

Z

Density (calculated)
Absorption coefficient
F (000)

Crystal size (mm3)
Indexing method
Reflections collected
Refinement method

Data/restraints/parameters

GOF on F2
R1(wR2) [F2, I>20(I)]
Abs. Struct. parameter

Largest diff. peak and 0.51 and -0.33 e.A-3

hole

b = l2.6947(8) A
c = 10.80720) A
a: 90°.

,6: 90°.

y: 90°.
3356.8(4) As

2

1.323 Mg/m3
0.590 mm!

1392

0.45 x 0.30 x 0.28
SMART

21784
Full-matrix on F2
7977 / 0 / 394
1.024
0.0396(0.0787)
0.005(14)

b = 13.7120) A
c = 21.2790) A
a: 104.059(4)°.
,3: 95.699(4)°.
y: 115.919(4)°.
3379.9(12) As

2

1.243 Mg/m3
0.505 mm-1

1320

0.36 x 0.29 x 0.17
TWINNING
20931
Full-matrix on F2
14764 / 15 I772
0.853
0.0867(O.181 1)

0.435 and -0.901 e.A-3

 

R1 = 2 ||F,| - |Fc|| / 2 |F,|, wR2 = (2 [W(F02 - Eff] / 2 [w(F02)2]}"2, GOF =
llwlF.2 - F321 I (n - p)} "2. w = 1 / 1020.2) + <an + bP]

197

parameters with TWROT indicated that the two solutions are rotated 37°
from each another.

Data integration was carried out on the triclinic cell (24) from the
TWINDX result with local and global cell constraints. The data were
partially contaminated from the orthorhombic data. R1 = 0.0888, wR2 =
0.1944 (for I > 20(1)), and R1 = 0.2881, wR2 = 0.2507 (for all data).

Data integration for the orthorhombic cell (23b) with a global
constraint produced a second data set. The reflections are weaker than those
belonging to the triclinic cell. Least-squares refinement on F2 led to
R1 = 0.1355, wR2 = 0.1990 (for I>20(I)), and R1 = 0.3753, wR2 = 0.2746 (for
all data). The crystallographic data for (24) are summarized in Table 31.

(10) [EttN]Cl'lFezCLt(MeOH)4(M-2,2'-bpym)l (26)

A dark brownish-green crystal of dimensions 0.49 x 0.22 x 0.13 mm
was secured on the tip of a glass fiber with Dow Corning silicone
grease and placed in a cold N2(g) stream at 173(2) K. Indexing
and refinement of 39 out of 79 reflections from a total of 60 frames
with an exposure time of 10 sec/frame indicated a monoclinic crystal
system. A full sphere of data With 2474 frames was collected with
a scan width of 0.3 in (l) and exposure times -18< h < 18, -20 < k < 20,
-20 < 1 < 20 with a maximum 20 angle of 56.70°. Re-indexing
the data reflection list with the TWINNING‘ package indicated the
presence of rotational twins which correspond to 54 independent
reflections for component A and 9 independent reflections for component B.
Component B is readily transformed into component A by a 180°
rotation around the (100) axis in direct or reciprocal space. The twinning
law is [l 0 -0.15, 0 -1 0, 0 0 -1]. The data were integrated by using
the orientation matrix of component A. Laue crystal symmetry constraints

198

Table 32. Crystal data and structure refinement for
[Et4N1C1°F62C14(bem)(MCOH)4- (26)

 

 

Identification code (26)
Empirical formula C40H84C110Fe4N 1003
Formula weight 1411.07
Temperature 173(2) K
Wavelength 0.71073 A

Crystal system Monoclinic

Space group P2(1)/c

Unit cell dimensions

Volume

Z

Density (calculated)
Absorption coefficient

F (000)

Crystal size

Indexing method
Reflections collected
Absorption correction
Refinement method

Data / restraints / parameters
GOF on F2

Final R indices [I>20( 1)]
Largest diff. peak and hole

a = 13.5890) A

b = 15.2620) A

c = 15.4230) A

a: 90°.

)3: 94.318(3)°.

7: 90°.

3189.7(8) As

2

1.469 Mg/m3

1.361 mm-1

1464

0.49 x 0.22 x 0.13 mm3
TWINNING

37705

Multiscan (SADABS)
Full-matrix least-squares on F2
7774 / 0 / 325

0.910

R1 = 0.0642, wR2 = 0.1396
0.663 and -O.488 e.A-3

 

R1 = 2 ||F,| - |Fc|| / 2 |1=,|, wR2 = {2 [w(F02 — 12,2)2] / 2 [w(F02)2]}"2, GOF: s
= {[w(F02 - F521 I (n - p)} "2. w = 1 1 1020.2) + <an + bP]

199

were used during data integration, and the cell parameters and orientation
matrix were refined and updated every 100 frames. All final parameters
were generated from the refinement of the XYZ centroids of 6822 strong
reflections with 1 >5 6, which led to a = 13.5890) A, b = 15.2620) A,

= 15.4230) A, B = 94.318(3)°, V = 3189.7(8) A3. The intensities were
corrected for beam inhomogeneity absorption and decay with the program
SADABS3 which led to transmission factors between 0.78 to 1.00. Of the
8060 unique reflections, a total of 3241 reflections with I > 20 (I) and Rim =
0.164 remained after data reduction. The structure was solved and refined in
the SHELXTL 5.10 package. The position of all non-hydrogen atoms were
located by direct methods and refined anisotropically by full-matrix least
squares on F2. Hydrogen atoms were placed in idealized positions. The
final full-matrix refinement based on 4774 unique reflection, of which 3241
with I > 20, and 326 parameters led to R. = 0.064 and wR2 = 0.166 (I > 20).
The goodness-of-fit is 0.910 and the highest peak in the final difference
map is 0.663 e'lA3

3. Discussion

A. X-ray Structural Results and Discussion.
(1) Structure of [Mn5Cllo(THF)3].. (16).

Unlike the related clusters Co4Clg(TI-IF)65 and Fe4Clg(TI-IF)66 that
exist as discrete molecular species, the Mn(II) derivative crystallizes as a
one-dimensional polymeric chain. The structure of [Mn5C110(THF)g].. is best
described as a one-dimensional chain consisted of alternating Mn4Clg(THF)6
and MnC12(THF)2 repeat units linked by chloride bridges. The presence of
the mononuclear "MnC12(THF)2" bridge allows for six coordination for all

the metal ions, as opposed to the Fe and Co clusters in which there are two

200

five-coordinate and two six-coordinate metal centers. The fact that Mn(II)
has a larger radius compared to Co(II) and Fe(II) is likely to contribute to its
tendency to form exclusively six-coordinate structures in the absence of
steric hindrance.

There are three different types of Mn environments in the structure,
namely trans-MnCl4(THF)2, MnC15(THF) and cis-MnC14(THF)2 that exhibit
different numbers of shared chlorides. There are also three types of chloride
ligands in (16). The two terminal chlorides on the five coordinate sites in
the Fe and Co clusters become 112-Cl atoms in [Mn5Clm(THF)3].. due to
bridging interactions with the MnC12(THF)2 unit, while the remaining liq-Cl
and 113-C1 chlorine atoms exhibit similar geometry to the Fe and Co
derivatives. The Mn—(uQ-Cl) distances range from 2.494(1)-2.554(1)A and
the Mn-(ug-Cl) distance are from 2.593(1)-2.602(1) A compared to the
M(uz-Cl) ranges of 2.355(1)-2.488(2) A and 2.285(2)-2.46(1) A for Fe and
Co respectively. In these clusters the M(ltg-Cl) distances range from
2.506(1)-2.737(1) A in Fe and 2.450(2)-2.844(3) A in Co. From these
comparisons, it is obvious that the 113-Cl interactions in (16) are more
symmetrical with a more narrow range of observed bond lengths compared
to the Fe and Co compounds. Schematic diagrams of the Fe and Mn
derivatives are provided in Figure 62. The Mn-Mn separations are longer
(Mnl-Mn2 = 3.7622(7) A, Mnl-MnlA = 3.751000) A, Mn2-Mn3 =
3.6851(5) A) than the distances observed in the structures of the Fe and Co
derivatives (Co-Co range 3.630(2)-3.686(2) A and Fe-Fe range 3.649(2)-
3.751(2) A).

(2) Structure of [MnBr2(THF)2]., (17)
The solid-state structure of (17) is also a one-dimensional polymeric

structure, but in this case there is only one kind of manganese environment,

201

 

Figure 61. Thermal ellipsoid representation of {Mn5C110(THF)g}..(16).

202

CI Cl

THF—Mn3—THF
/ \

THF 03 0'3 0'3
\ / “F’s-V
M1
012’ g \012 02"“?‘012
TH F/ll’h. JZ‘ ‘\\\C' 1”,," ' Mlz"‘\\\\TH F TH F/l’h 'M I 2‘\\\\C| 1”In -M I 2.\\\\\TH F
n n
THF( I ‘cn’ I ‘THF THF’ I ‘cn' | ‘THF
c12\ M'1,012 02‘ le ,012
/ \
Cl3 THF / \\THF

CI: (;3
THF—Mna—THF

Cl Cl

M4C18(TI'IF)69 M = Fe, C0 {Mn5C110(T HFlsln

Figure 62. Comparison of different types of Mn(II) sites in (16) with the
metal sites in M4C13(THF)6 clusters.

203

c.
Q
Q o
. ‘
’§ '
\\
O
I,
\‘
\‘ ’0
t ‘.
O’O
‘ \
1 ' .
e "
.s'
0
e '
I
\\
\\ )I‘

9
\
v . ‘ o
c e_- . ’4 '.'
’1', , ’1" -
" ’1', " 0';-
e o
e ‘ 9
I ‘ e
o e
0 e
‘ ‘

Figure 63. Packing view of [Mn5C110(THF)3].,(16) down the b axis.

204

e e
e 9‘? . o ”,9- O
«- ‘ ‘1 C(3l
‘7’ 0121‘!
, 0141
. ‘= a ' L: D
./. CI‘ll ,/
0 I/ :2: e 9).—>9
011)
Brl1C] Br‘ll]
:- Mom :4
mm 0.1:
5‘ '1
BrllBl r‘l‘lAl
our; 9 o “AQUA
CMA) " C(1Al
° 13g 0 1:34.
‘ § . . g; C2Al
‘-‘" . clam§ .
. 0

Figure 64. Thermal ellipsoid representation of a portion of the chain
compound [MnBr2(TI-IF)2].. (17)_

205

O
C
0(4) gm, (:15)
2") .
e i“
$515.1,
C131 1.4., 1:16)
012m
. 0
1(1) cm
A» om . ( “col
4 an I ( \F} \' I
$0,,"Mnl1i "” ”L, 69’
* 011A] ’ 6‘. 012A)
Il1Al .
O
“‘3 012A)
013A) 45> , e
' ’4. 4016A)
1r ' I 0
01405;! I.
. A“ 015A)
0

Figure 65. Thermal ellipsoid representation of mononuclear [Mn12(THF)3],

206

namely a trans-bis(tht)tetrabromorrlanganate(ll) unit which is the same as
the mononuclear bridging moiety in (16). The fact that there is no evidence
for the formation of the MII4B1'3(TI"IF)6 cluster is probably related to steric
effects due to the much larger radius of Br' as compared to CT. The Mn-Mn
separation in (17) is 3.9890(8) A and the Mn-Br bond length is 2.722(5) A.
As expected, these distances are appreciably longer than those in the Cl
compound. The angles Br-Mn-Br and Mn-Br-Mn are 85.9(2)° and 94.25(3)°
respectively. The oxygen atoms are related by a two-fold axis through the
Mn atom and the O-Mn-O is nearly linear (179.7(15)°).

The steric argument advanced to explain the structural differences in
the MnX2(THF)y (X = Cl, Br) compounds is further supported by an
examination of the iodide derivative.7 As can be seen from the Thermal
ellipsoid diagram in Figure 63, cis-Mn12(THF)3 is a five coordinate trigonal
bipyramidal molecule rather than a cluster or an oligomer. Due to the larger
size of the iodine atoms, the axial oxygen atoms are bent away from the
equatorial positions as evidenced by the angles O-Mn-O is 165.3(3)° and the
I-Mn-I angle is 126.89(5)°. The Mn-I bond length is 2.7244(9) A.

(3) Structure of FezCl4(bpy)2 (19).

Compound (19) crystallizes as a dinuclear compound with a square
pyramidal geometry resulting from the union of two "FeC12(bpy)" units with
two bridging chloride atoms. The Fe atoms are 0.699(1) A above the least-
square plane of the basal plane of the square pyramid. The angle between
the least-squares plane of the bipyridine and the base of the square
pyramidal is 21.54(8)°. The least-square plane refinement results are listed
in Table 5. The bipyridine molecules from different dimer units stack along
the a direction of the unit cell with an average intermolecular contact of
3.393(3) A. A plot of (19) viewed down the a direction of the unit cell is

207

Q

      
 

, C01
1 \ .
0121" 0141
' fl- 0
’ \ . \. \CIS)
ClllAl \ cm I, ‘ .06) ,s\ 017]
..— c1121 ’ _, .
‘ clai
\
Fd‘Felll ”‘2 3, ,1, 0191
—1
011210" ° c1101 '
~ Cll1l

Figure 66. Thermal ellipsoid representation of [Fe2C14(bpy)2], (19)

208

 

 

Figure 67. Packing View of [Fe2C14(bpy)2](19) down the a axis.

209

shown in Figure 66. It is interesting to note that the structure of the
Fe2C14(bpy)2 differs from that of the corresponding manganese chloride
compound of empirical formula “MnC12(bpy)” which crystallizes in
I polymeric chains.8 The dimer feature of (19) versus the one-dimensional
chain of [MnC12(bpy)].., and (20) may be explained by the fact that the
smaller Fe(II) ion does not always prefer six coordination unlike Mn(II). In
the event that more sterically demanding ligands than 2,2'-bipyridine are
used, Mn(II) adopts the dinuclear structure as in the case of Mn2C14(biq)2
(biq = 2,2’-biquinolyl).9 The molecule Mn2C14(biq) (depicted schematically
in Figure 70), possesses a Mn-Mn distance of 3.887(1) A and a Mn-Cl-Mn
angle of 998°, compared to the Fe-Fe distance in (19) of 3.7048(17) A and
an Fe-Cl-Fe angle of 97.35(4)°. Although the two structures are very
similar, the Fe-Fe distance is shorter and the Fe-Cl-Fe angle is more acute.
The bpy ligand deviates only slightly from planarity, while the biq ligand is
bent. It appears from these results that the edge sharing square pyramidal
dimer structures of Fe2C14(bpy)2 and Mn2C14(biCI)2 result from primarily
steric rather than electronic factors. The existence of another similar dimer
metal bipyridyl chloride complex [CuC12(4,4'-Me2bpy)2]2lo also supports
such a hypothesis.

(4) Structures of [ppnlle92Cla] (21) and [ppnllen2Clt] (22)

Compound (21) and compound (22) of general formula [M2C16]2'
crystallize as dinuclear anions consisting of two distorted tetrahedra with
shared chlorine edges. Both structures are similar to that of [ppn]2Fe2C16 in
that the packing is dominated by the relatively bulky [ppn]+ cations. The
two orientations of the [M2C16]2' anions in the asymmetric unit of (21) and
(22) are believed to be the result of a packing effect that forces the chlorine

bridges to adopt more acute angles.

210

Table 33. Bond lengths [A] and angles [°] for [FeC12(bpy)].. (20).

 

A-B (A) A-B (A)
Fe(1)-N(3)#1 2.171(3) N(3)-C(5) 1.341(5)
Fe(1)-N(3) 2.171(3) N(3)-C(4) 1.350(5)
Fe(1)-Cl(2) 2.422002) C(4)-C(6) 1.391(6)
Fe(1)-Cl(2)#1 2.4220(12) C(5)-C(7) 1.396(5)
Fe(1)-Cl(2)#2 2.632(2) C(5)-C(5)#1 1.492(8)
Fe(1)-Cl(2)#3 2.632(2) C(6)-C(8) 1.378(7)
Cl(2)-Fe(1)#3 2.632(2) C(7)-C(8) 1.376(6)
A-B-C (°) A-B-C (°)

 

N(3)#1-Fe(l )-N(3) 74.98(18) Cl(2)#1-Fe(1)-Cl(2)#3 96.62(4)
N(3)#1-Fe(l)-Cl(2) 93.74(10) Cl(2)#2-Fe(1)-Cl(2)#3 l79.92(4)

N(3)-Fe(l)-Cl(2) l62.84(10) Fe(1)-Cl(2)-Fe(l)#3 9656(4)
N(3)#1-Fe(1)-Cl(2)#l 162.84(10) C(5)-N(3)-C(4) 117.7(3)
N(3)-Fe(1)-Cl(2)#1 93.74(10) C(5)-N(3)-Fe(1) 117.0(3)
Cl(2)-Fe(1)-Cl(2)#l 9993(6) C(4)-N(3)-Fe(l) 124.8(3)
N(3)#1-Fe(l)-Cl(2)#2 84.74(9) N(3)-C(4)-C(6) 123.3(4)
N(3)-Fe(1)-Cl(2)#2 95.19(9) N(3)-C(5)-C(7) 122.2(4)
Cl(2)-Fe(1)-Cl(2)#2 96.62(4) N(3)-C(5)-C(5)#1 115.3(2)
Cl(2)#1-Fe(1)-Cl(2)#2 83.44(4) C(7)-C(5)-C(5)#1 122.5(2)
N(3)#l-Fe(1)-Cl(2)#3 95.19(9) C(8)-C(6)-C(4) 118.1(4)
N (3)-Fe(1)-Cl(2)#3 84.74(9) C(8)-C(7)-C(5) 119.2(4)
Cl(2)-Fe(1)-Cl(2)#3 83.44(4) C(7)-C(8)-C(6) 119.5(4)

 

Symmetry transformations used to generate equivalent atoms:
#1 -x,y,-z+1/2

#2 x,-y+1,z-1/2

#3 -x,-y+1,-z+l

211

 

Figure 68. Thermal ellipsoid representation of [FeC12(bpy)],. (20.)

212

 

 

 

 

Figure 69. Packing diagram of [FeC12(bpy)].. (20) showing the one-

dimensional chain along c axis.

213

 

[MnC12(bpy)]n

[FeCIXbPYHZ

 

[MnCltbqua

Figure 70. Structures of [MC12(bidentateL)]2.

214

(a) [ppn12IC92Cltl(21).

The [C02C16]2' moieties can be viewed as an assembly of two edge-
sharing tetrahedral [CoC14]' units related by an inversion center. There are
two independent [C02C16]2' anions in the asymmetric unit. The average Co-
Chm,inal and average Co-Clbfidgm, distances of 2.224(1) A and 2.346(1) A are
comparable to the corresponding distances reported in the literature
(2.212(3)-2.238(6) A for distances of Co-Clmml and 2.329(5)-2.38 (3) A
for Co-Cl.,,.,dg,,,g).24 The C0 """ Co distance of 3.221(1) A is shorter than the
corresponding distances reported in the literature (3.277 (6)-3.366 (3) A).26
The Col-C12-Col* and C12-Co-Cl2* angles of 85.79 and 93.29(3)° are
considerably distorted from an ideal tetrahedral geometry due to the
formation of the four membered ring. Both of the [C02C16]2' units possess
Cl-Co-Cl angles larger than 90°. By comparison, in the [ppn]2Fe2C16
structure, one of the two [Fe2C16]2' anions exhibits an angle less than 90°.
The C0 """ Co distance in (21) is the shortest one of the group, and the Co-Cl-
Co angle is the smallest in our [ppn]2[M2C16] (M = Fe, Co and Mn) series. A
projection diagram viewed in the ac plane in Figure 71 illustrates the
arrangement of the [ppn]+ cations and the [CozCl5]2' anions in the unit cell.
(b) [ppnllenzClt (22)-

The [ppn]2[Mn2C16] structure is similar to the Co analogue, with two
independent dimetal anion units in the unit cell and a Cl-Mn-Cl angle greater
than 90°. The average Mn-Clterminal and average Mn-Clbn-dging distances of
2.309(1) A and 2.446(1)A are comparable to analogous distances reported in
the literature 2.306(2) A for distances of Mn-Clmmiml and 2.440(2) A for
distances for Mn-Clbn-dging.“ The average Mn """ Mn distance of 3.376(1) A is
the same as the distances reported in the literature for [Mn2Cl6]2' anion
(3.376(2) A).Ila The average Mnl-ClZ-Mn1* and C12-Mn-C12* angles of

215

87.25(2)° and 92.75(2)° are considerably distorted from an ideal tetrahedral
geometry due to the formation of the four member ring. A projection
diagram viewed down the ac plane in Figure 74 illustrates the arrangement
of the [ppn]+ and the [Mn2C16]2' units in the unit cell.
(C) [EtctN]CI°IF92C14(MCOH)4(P-'292"bpym)] (26)-

Cleavage of the [Fe2C16]2' core is also observed to occur in the
reaction of [EtaN]2[Fe2C16] with 2,2’-bpym in MeOH. The compound
[EtaN]Cl-Fe2Cl4(MeOH)4(u-2,2'-bpym) consists of two Fe(II) atoms joined
by a bis-chelating 2,2'-bpym ligand that are further bonded to two terminal
chlorides and two methanol ligands. The unit cell also contains an
equivalent of [Et4N]Cl that serves to fill void space as well as to assist in
connecting the dimeric molecules into an infinite chain through hydrogen
bonds between the outer-sphere Cl’ ions and the axial MeOH ligands. The
average hydrogen bonding distance between the outer-sphere Cl' ion and the
axial methanols is 2.315(7) A. There are also weak hydrogen bonding
between the terminal chlorides, which bonded to the metal, and 2,2'- bpym
ring as well as the tetraethylammonium cation at an average distance of
2.805(5) A. The 2,2'-bpym ring is slightly out of the plane containing Fel
and FelA by ~0.015 A. A thermal ellipsoid drawing as well as selected
bond distances and bond angles for [EtaN]Cl-[Fe2C14(MeOH)4(u-2,2'-
bpym)] are presented in Figure 75 and Table 36. The average Fe-Cl bond
distance is 2.383(2) A. Fel-N(bridging) bond distances of 2.224(6) A and
2.239(6) A differ only slightly differs from Fe2-N(bridging) bond distances
at 2.244(6) A and 2.250(6) A. The separation between two intramolecular
Fe(II) atoms is 5.957 A. The closest intermolecular Fe """ Fe contact is 7.964
A. The occurence of rotational twining hindered previous attempts to solve

the structure by point detector data, which was contaminated by twinning

216

Table 34. Selected bond distances (A) and bond angles (°) for
[Ppnlle02Clo], (21)-

 

 

 

 

 

Bond Distances

A B A-B (A) A B A-B (A) A B A-B (A)
Col Col* 3.191(1) Co2 C02* 3.251(1) N1 Pl 1.585(3)
C01 C11 2.224(1) Co2 C14 2.232(1) N1 P2 1.580(3)
C01 C12 2.347(1) Co2 C15 2.362(1) N2 P3 1.585(3)
Col C12* 2.342(1) Co2 C15* 2.334(1) N2 P4 1.578(3)
C01 C13 2.224(1) Co2 C16 2.216(1)

Bond Angles

A B C A-B-C (°) A B C A-B—C (°)
C11 C01 C13 115.94 (4) C14 Co2 C16 115.57 (4)
C11 C01 C12 106.86 (4) C14 Co2 C15 112.34 (4)
C11 Col C12* 114.34 (4) C14 Co2 C15* 113.55 (4)
C13 Col C12 114.81 (2) C16 Co2 C15 109.94 (5)
C13 Col C12* 108.76 (4) C16 Co2 C15* 110.75 (5)
C12 C01 C12* 94.21 (3) C15 Co2 C15* 92.370)
Col C12 C01* 85.790) Co2 C15 C02* 87.63 (3)
P1 N1 P2 137.4 (2) P3 N2 P4 140.6 (2)

 

217

.283 we 05 E Jonazmmamfi «o 32> mew—8m .E: Page."

 

218

.25 we 5323:8298 283:0 Hanoi. .Nb ouswwm

 

m6

<30

219

and led to severe systematic violations and a high R factor. By using CCD
data and the 'I'WINNING4 package, we were able to integrate the data

basically free from twinning contamination (less than 0.1%).

B. Magnetic Studies.

Since one of the goals of this work was to use the dinuclear metal
anion [Fe2C16]2' and the tetranuclear metal cluster Fe4Clg(THF)6 as a source
of the diferrous ion in reactions with polydentate ligands, we undertook a
comprehensive study of the magnetic properties of various salts of this anion
and of analogous anions of Co(II) and high-spin Mn(H). Four complexes
containing the [CozCl6]2' moiety12 and [M2X6]2' salts containing various
metals and halides such as [Fe216]2' and [Mn2x,]2' (x = Br, I Cl), have
appeared in the literature.”'14 No magnetism has been detailed for these
compounds. In this work, magnetic studies of the different salts of [M2C16]2'
with the metal ions M11 = Fe, Co and Mn have been performed. For the Fe
compounds, salts containing four different counterions were studied to check
for the effect of counterion on magnetic interactions.

(1) Magnetic properties of [ppn]2[CozC16] and [ppn]2[Mn2C16].

The magnetic properties of the [ppn]+ salt are depicted in Figures 66,
67 and 68 with plots of x vs. T and xT vs. T. These data suggest the
presence of strong antiferromagnetic exchange interactions between the
Co(II) ions. A rounded maximum in x at 50 K is observed, while xT reveals
a continuous decrease from a value of 4.22 emu'K/mole at 300 K to a value
close to zero at 2 K. In tetrahedral environments, Co(II) is described by
a 4A; term. Therefore, the Hamiltonian in Eq. 7 was expected to be
appropriate to describe its properties,

H. = -2Jstsz + D(3212 + 8.22) (Eq. 7)

220

as it contains an isotropic exchange term supplemented by a zero-field
splitting term to account for the single-ion anisotropy of the spin S=3l2 of
Co(II). In an initial fitting exercise, we attempted to reproduce the magnetic
properties by a fully isotropic Hamiltonian, i.e., by neglecting the single-ion
anisotropy contribution. This simple model completely failed to fit the
magnetic data. In particular, significant differences between theory and
experiment were observed in the region near the maximum in x (dotted line
in Figure 76). In a second step, the ZFS term was taken into account which
introduces a magnetic anisotropy. In a manner similar to the treatment of
the Fe dimers, the model needs to calculate the two components of the
magnetic susceptibility (x ,, and x!) in order to obtain the theoretical curve
for a powder (xvowdir = (7‘4 +27“ Y3). With this additional term the magnetic
behavior is closely reproduced in the whole temperature range from the
following set of parameters: J = - 11.6 cm’l, D = 29.0 cm"1 and g = 2.25
(solid line in Figures. 76 and 77). A small amount of paramagnetic impurity
(0.16 %) was introduced to reproduce the Curie tail observed below 4 K.

In summary, the spin anisotropy is again the dominant contribution
(the ratio AJ/AD is equal to 0.4), although a stronger antiferromagnetic
coupling is present. The large values of D compared to J are a consequence
of the bridging angles M-Cl-M which are not far from the orthogonality
value which favors weak interactions. An additional effect for a distorted
tetrahedral environment is that the ZFS of the ground spin state can mix
Withexcited states through a second-order spin orbit coupling may be quite
significant. The magnetic data for [ppn]2[Mn2C15] are plotted in Figure 78.
A continuous decrease in xT is observed at lower temperatures which is
indicative of dominant antiferromagnetic interactions. In this case the ZFS

is expected to be quite small compared to J, as the ground state of Mn(II) is

221

Table 35. Selected bond distances (A) and bond angles (°) for
[PPnllenzaolt (22)-

 

 

 

 

 

Bond Distances
A B A-B (A) A B A-B (A) A B A-B (A)
Mnl Mnl* 3.344(1) Mn2 Mn2* 3.408(1) N1 P1 1.586(2)
Mnl C11 2.305(1) Mn2 C14 2.305(1) N1 P2 1.583(2)
Mnl C12 2.447(1) Mn2 C15 2.466(1) N2 P3 1.583(2)
Mnl C12* 2.439(1) Mn2 C15* 2.434(1) N2 P4 1.582(2)
Mnl C13 2.310(1) Mn2 C16 2.316(1)

Bond Angles
A B C A-B-C (°) A B C A-B-C (°)
C11 Mnl C13 118.00 (3) C14 Mn2 C16 118.21 (3)
C11 Mnl C12 105.58 (3) C14 Mn2 C15 108.63 (3)
C11 Mnl C12* 113.82 (3) C14 Mn2 C15* 110.04 (3)
C13 Mnl C12 114.490) C16 Mn2 C15 111.520)
C13 Mnl C12* 108.79 (3) C16 Mn2 C15* 113.40 (3)
C12 Mnl C12* 93.63 (2) C15 Mn2 C15* 91.870)
Mnl C12 Mn1* 86.370) Mn2 C15 Mn2* 88.13 (2)
P1 N1 P2 137.3 (2) P3 N2 P4 141.2(1)

 

222

 

ANS e6 cow—3:880»: 28050 38.55.

<20

me 2:2..—

 

223

.053 we 05 E o—Uuqzaafi mo 33> mew—8m .2. guru

 

224

described by a 6A, term. Thus, a fully isotropic Heisenberg Hamiltonian
should be suitable for describing the prOperties. As we can see from the
data, however, this model does not satisfactorily reproduce the experimental
behavior. A model assuming two antiferromagnetic J parameters of
different magnitudes was used to improve this fit. The consideration of a
second exchange coupling in this system may be due to the presence of a
weak antiferromagnetic coupling between Mn dimers. A simple model that
accounts for this effect is that of an alternating J -J ’ chain of spins 5/2. If
these spins are treated as classical, an analytical expression can be obtained.
Using this expression, a good fit of the experimental data was obtained. The
best fit (solid line in Figure 78) corresponds to the following parameter
values: J = -15.6 cm", J’ = — 0.69 cm"1 and g = 2.32. In order to reproduce
the data a paramagnetic impurity has been introduced (3%).

(2) Magnetic fitting of Mn(II) chains of (17).

In order to fit the magnetic properties of these linear chains, the well-
known Fisher expression for classical spins was used.13 The best fit gives[ J
= -0.17 K and g = 2.06. Unfortunately, the data are contaminated with
oxygen just in the region where the xT product starts to decreases which
makes it difficult to obtain a precise fit. Another sample needs to be
measured for this to be a reliable fit.

(3) [Mn5Clm(THF)g]..chain of (16)

With the classical model, fitting parameters were obtained with g
equal to 2.28 and J equals to -0.70 K. A cluster approximation with the
definition of the exchange pathways (as shown in Figure 70-73).

(4) Magnetism of the Mn cluster (16)

In the cluster approximation the best fitting parameters are obtained

when J '/J is small (J '/J = 0.1, J' = J " and J = -0.4K). When compared to the

225

 

Figure 75. Thermal ellipsoid representation of
[EtaN]Cl-[Fe2C14(MeOH)4(u-2,2’-bpym)] (26) illustrates the hydrogen bond
between the iron dimer units and the chloride ion.

226

Table 36. Selected bond distances (A) and bond angles (°) for [EtaN]Cl°

[F62C14(MeOH)4(u-2,2'-bpym)], (26)-

 

 

 

 

Bond Distances
A B A-B (A) A B A-B (A)
Fel 02 2.137(6) Fe2 O4 2.151(6)
Fel Ol 2.168(6) Fe2 O3 2.164(6)
Fel N 1 2.239(6) Fe2 N3 2.244(6)
Fel C11 2.379(2) F02 C13 2.386(2)
Fel C12 2.388(2) Fe2 C14 2.381(2)
Bond Angles
A B C A-B-C (°) A B C A-B-C (°)
N2* Fel N 1 74.6(2) N3 Fe2 N4* 73.8(2)
02 Fel N1 86.8(2) O4 Fe2 N3 85.6(2)
Ol Fel N1 84.7(2) O3 F02 N3 85.7(2)
Ol Fel C11 92.77(18) 03 Fe2 C13 94.36(18)
N1 Fel C11 90.61(18) N3 Fe2 C13 92.95(18)
C11 Fel C12 102.22(8) C14 Fe2 C15 102.43(9)

227

40x10'3 —
35 —J
30 -e

25-

 

X (emu/mol)

20—

15—

 

 

H.0-fl'ul'uflmd'tumo'l-Cfl. .
O a. . .

 

10 l l l I l l

50 100 150 200 250 300
T(K)

C—

Figure 76. Thermal dependence of the susceptibility for the [ppn]2[CozC16]
dimer (filled circles) and the best fit to the experimental data
obtained (solid line). Dashed-dotted lines represent the
perpendicular (upper) and parallel (lower) susceptibility. The
theoretical behavior of a fully isotropic antiferromagnetic Co(H)
dimer is shown as dotted line.

 

228

 

 

 

4_
e,_
E
2.
:3
5.2—
$2
1—
0" r I I I I I
50 100 150 200 250 300

T (K)

Figure 77 . Thermal dependence of the product xT for [ppn]2[CozC16] (filled
circles) and the best fit to the experimental data obtained (solid
line). The theoretical behavior of a fully isotropic
antiferromagnetic Co(II) dimer is shown as dotted line.

229

XT (emu.K/mol)

 

 

 

T" I I I I I I
0 50 100 150 200 250 300

T (K)

Figure 78. Thermal dependence of the susceptibility for [ppn][Mn2C16]
(filled circles) and the best fit to the experimental data obtained
(solid line). The theoretical behavior of a fully isotropic
antiferromagnetic Mn(H) dimer is shown as dotted line.

 

230

WWW. ,VM"\

MD J" MD

KWJ RAJ

Mn

 

Mn

 

 

 

 

 

Figure 79. Polymeric chain of MnClz units in (16).

 

 

 

3O 4 IAAI LALLLA .4 r l r n A n l r n r A l A ‘4 L b

25‘: r

c: I J” ;

O . .-

E 20.. _

9. : ‘ K :

=1 4 u

E . .

3 15: I:

Q ~ .

3 J=04K ;

10 ~ J’IJ =J”/J = 0.1 -

: g=2.28 _
5‘.........,....,..e.,....,....

0 50 100 150 200 250 300

 

 

T (K)

Figure 80. Fitting with the cluster model of magnetic measurement data of
(16).

231

 

5 fivivtfifiTTfth'VrrtliIIIIIV'T

; I
E 1
8 1
E— - - _ 1
a Classrcal Chain S—5/2 ,
J = -0.17 K

g = 2.06 .

4

 

 

 

A l A A A l A A A A

200 250 300

0 50 100 150
T(K)

Figure 81. Magnetic data fitting of [MnBr2(THF)2],. (17).

 

 

 

 

 

 

 

 

 
   

 

 

 

 

 

 

 

 

. 1/zmol (COW) I . gamma) I
I I I I [TI—FI I I I I I l I FYI I Ij IIII I I I I I I I I 6
P 1
1
80 _ ...,. e e e j
, ° - 5.5
O .1
60 F '
E l’ . — 5
5 ”e . S
3 _ 1 3
i. . . 3
: 40 -o 4
. 4 4.5
1
..e I v=M0+Mi'x I .
w I MOI 0.79142 I 4
20 " I M1L0.24732 I _ 4
O
’ [ RI 0.99989 I ‘
d
0 JA l l l L1 1 l I ‘4 I ll ‘ I 1" l I I l l l l I I I A L4 35
o 50 100 150 200 250 300 350

T(K)

Figure 82. The Curie-Weiss behavior of [MnC12(MeOH)2].. (18). Solid line
is the Curie-Weiss fit, dot is the llx and diamond dot is the II,“-
plot.

232

classical model, the data can be interpreted with a simple cluster model
(tetramer plus an isolated Mn(H)). The interaction between isolated Mn and

the tetramer is weaker than the interaction of Mn within the tetramer (J > J’).

4. Conclusion

Metal halide complexes of 3d metals with simple solvent molecules
like THF and MeOH have the potential be used as building block for
molecular arrays with their tendency to form multinuclear clusters, solubility
in common organic solvents, and the presence of unpaired spins. Although
their synthesis and structures may be simple, their solid-state
characterizations is not as easy as one might think due to their tendency to
lose solvent and/or twinning. However, their structural information is
important in understanding the basic magnetic interactions of clusters, which

leads to a better understanding of more complicated molecular arrays.

233

References

(a) Holm, R. H. Acc. Chem. Res. 1977, 10, 427. (b) Holm, R. H.
Chem. Soc. Rev. 1981, 10, 455. (c) Papaefthymiou,V.; Girerd, J .J .;
Moura, I.; Moura, J.J.G.; Miinck, E. J. Am Chem. Soc. 1987, 109,
4703. (d) Taft, K. L.; Lippard, S. J. J. Am. Chem. Soc. 1990, 112,
9629.(e) Ménage, S.; Vincent, J. M.; Lambeaux, C.; Chottard, G.;
Grand, A.; Fontecave, M. Inorg. Chem. 1993, 32, 4766. (f) Taft, K.
L.; Papaefthymiou, G. C.; Lippard, S. J. Science 1993, 259, 1302. (g)
Zang, Y.; Jang, H. G.; Chiou, Y. M.; Hendrich, M. P.; Que, L. Jr.
Inorg. Chim. Acta 1993, 213, 41. (h) Hendrich, M. B; Day, E. P.;
Wang, C. -P.; Synder, B. S.; Holm, R. H.; Mfinck, E. Inorg. Chem
1994, 33, 2848 and references therein. (i) Powell, A. K.; Heath, S. L.
Comments Inorg. Chem. 1994, 15, 255 and references therein. (j)
Goldberg, D. P.; Tesler, J.; Bastos, C. M.; Lippard, S. J. Inorg. Chem,
1995, 34, 3011.

(a) Willett, R. D.; Landee, C. P.; Gaura, R. M., Swank, D. D.;
Groenedijk, H. A.; van Duynevelt, A. J. J. Mag. Mag. Mater. 1980,
15-18, 1055. (b) Groenedijk, H. A.; van Duynevelt, A. J.; Blbte, H.
W. J.; Gaura, R. M., Willett, R. D. Physica 1981, 1063, 47. (c)
Kitajima, N.; Amagai, H.; Tamura, N.; Ito, M.; Moro-oka, Y.;
Heerwegh, K.; Pénicaud, A.; Mathur, R.; Reed, C. A.; Boyd, P. D. W.
Inorg. Chem. 1993, 32, 3583. (d) Taft, K. L.; Delfs, C. D.;
Papaefthymiou, G. C.; Foner, S.; Gatteschi, D.; Lippard, S. J. J. Am.
Chem. Soc. 1994, 116, 823 and references therein. (e) Elrnali, A.;
Elerman, Y.; Svoboda, I.; Fuess, H.; Griesar, K.; Haase, W. Z.
Naturforsch. 1994, 49B, 365. (f) Mathoniere, C.; Carling, S. G.;
Yusheng, D.; Day, P. J. Chem. Soc., Chem. Commun. 1994, 1551. (g)
Klose, A.; Solari, E.; Floriani, C.; Chiese-Villa, A.; Rizzoli, C.; Re, N.
J. Am. Chem. Soc. 1994, 116, 9123. (h) Aromi, G.; Claude, J. P.;
Knapp, M. J.; Huffman, J. C.; Hendrickson, D. N.; Christou, G. J. Am.
Chem. Soc., 1998, 120, 2977.

Sheldrick, G. M., SADABS. Siemens Area Detector Absorption ( and
other) Correction, 1998. University of Gttingen, Germany.

10.

ll.

12.

13.

14.

234

Sparks, R. A. TWINNING (TWINDX, 'I'WUTIL, TWROT and
TWHKL), Program for processing twinned data. Bruker AXS
Instrument, Madison, WI.

Sobota, P.; Olejnik Z.; Utko, J. and Lis, T. Polyhedron, 1993, 12,
613.

(a) Bel’skii, V. K.; Ishchenko, V. M.; Bulychev, B. M.; Protskii, A.
N .; Soloveichik, G. L.; Ellert, O. G.; Seifulina, Z. M.; Rakitin, Y. V.;
Novotortsev, V. M. Inorg. Chim. Acta, 1985, 96, 123. (b) Cotton, F.
A.; Luck, R. L.; Son, K.-A. Inorg. Chim. Acta, 1991, I79, 11.

Stolz, P. and Pohl, S. Z. Natuurforsch., Teil, 8., 1988, 43, 175.

Lubben, M.; Meetsema, A.; Feringa, B. L. Inorg. Chim. Acta, 1995,
230, 169.

Sinn, E. J. C. S. Dalton, 1976, 162.

Gonzalez, 0.; Atria, A. M.; Spodine, E.; Manzur, J .; Garland, M. T.
Acta Cryst. C 1993, 1589.

(a) Pampaloni, G.; Englert, U. Inorg. Chim. Acta 1995, 231, 167. (b)
Brass, C.; Robert, R.; Bachet, B. and Chevalier, R. Acta Crystallogr.,
Sect. B. 1976, 32, 1371. (c) Goodyear, J .; Ali, E. M. and Shutherland,
H. H. Acta Crystallogr., Sect. B. 1978, 34, 2671.

(a) Harrison, W.; Trotter, J. J. Chem. Soc. Dalton Trans. 1973, 61. (b)
Olson, W. L.; Dahl, L. F. Acta Cryst. 1986, C42, 541. (c) Kireeva, O.
K.; Bulychev, B. M.; Streltsova, N. R.; Belsky, V. K.; Dunin, A. G.
Polyhedron, 1992, 11, 1801.

(a) Saak, W.; Haase, D.; Pohl, S. Z. Naturforsch. 1988, 43B, 289. (b)
Ruhlandt-Senge, K.; Miiller, U. Z. Naturforsch. 1992, 47B, 1075.

Pohl, S.; Saak, W.; Stolz, P. Z. Naturforsch. 1988, 43B, 171.

235

15. Khan O. Molecular Magnetism. 1993 258.

CHAPTER V

THE USE OF THE CCD AREA DETECTOR
X-RAY INSTRUMENT TO RESOLVE TWINNING
PROBLEMS

236

237

1. Introduction

Crystallography has become an essential component of both chemical
and biological research.1 New structures are emerging in the literature at an
ever-increasing rate, thanks to technical advances in radiation sources,
detectors, computing and to improved structure-determination methods. The
recent introduction of Charge Coupled Device (CCD) area detectors for
single crystal X-ray analysis has contributed to a recent “structural
explosion”.l As evidence for this, a new structure is deposited in the
Cambridge Structure Data Base (CSD), on average, approximately every
forty minutes, with the total number of crystal structures determined
reaching 190,000 in 1998.2’3 With the new techniques being put to use,
researchers can now obtain crystallographic results much more quickly than
ever before. In the past, it was not uncommon for data to be collected over
the course of weeks, but comparable structures can now be determined in
less than 24 hours, including solution and least-square refinement and
preparation of tables and diagrams for publication. In spite of these
remarkable advances, the practitioner of modern crystallography soon learns
What “seasoned” crystallographers have been warning,2 namely that you
need to have an even deeper understanding of the underlying
crystallographic principles in the CCD era. Diffractometers equipped with
CCD area detectors are being widely used in small molecule crystallography
in place of the conventional serial detectors, and data collections on
microcrystals and weakly diffracting, mosaic crystals have become more
common. A typical crystallography lab equipped with a CCD area detector
may collect three to four hundred data sets a year, of which as many as 10%

are for problem crystals with twinning or other problems. In this chapter,

238

specific strategies for handling problematic crystal data in terms of data
collection and processing, as well as solution and refinement of twinning

and other problematic crystals will be discussed.

A. Background
( 1) A brief history of diffractometers

In the first X-ray diffraction experiment (Figure 83), performed by
Laue, Friedrich and Knipping in 1912’, it was demonstrated that a crystal
lattice is able to diffract X-rays in a manner much like a fine metal grid
diffracts light. The size of the grid corresponds to the wavelength of the
diffracted beam. This result implied to scientists that we may actually be
able to “see” atoms by virtue of their diffraction pattern. The first successful
crystal structure to be solved was carried out on sodium chloride by Bragg.4
It is interesting to point out that the diffraction pattern was recorded on a
photographic film, which is the first type of area detector; in fact, the
experimental setup is essentially the same as the one used today. Today’s
equipment is obviously more elaborate, but the essential components are still
an X-ray source, a crystal bathed in the incident beam, and a detector to
measure the intensity of diffracted X-ray beams. From the early days of
Bragg, various film techniques dominated the X-ray diffractometers until
these gave way to scintillation detectors in the 1960’s. Scintillation
detectors (serial or point detectors) use electronic devices for direct counting
of diffracted photons for more rapid and accurate intensity measurements5 as
compared to the difficult to obtain (and usually less accurate) intensities
measured by film methods. Protein crystallography, which rapidly
developed in the 1980’s, requires the collection of hundreds of thousands of

reflections for a data set, which would take months to collect with a serial

239

diffracted beam
(spot on film)

    
 
 
   

direct beam
\ (absorbed by lead

crystal beamstop)

a

heated filament $6515

photographic film
(detector)

 

X-ray tube
(source)

Figure 83. The experimental setup used by Laue, Friedrich and Knipping to
measure X-ray diffraction intensities. This same general experimental setup
is still currently used, although the source of X rays and the detection system
are now much more sophisticated.

240

detector, thus crystallographers in this field are forced to use film and other
area detector techniques.
(2) Modern area detectors and CCD techniques

The desire to collect large numbers of data sets or to collect from
crystals with very large unit cells has lured crystallographers back to area
detectors that can collect multiple reflections simultaneously when the
detector cuts through the Ewald sphere. Area detectors developed from the
old film technique use electronic devices, such as image plates and multi-
wire area detectors, to record X-ray photons. While these have been very
useful, these area detector techniques usually suffer from slow readout time.

The CCD (Charge Coupled Device) detector was developed in the
mid 1980’s as an electronic camera, and it soon found an application in
crystallography. The CCD camera is an area detector that measures the X-
ray intensity of slices of reciprocal space. The area detector collects
intensities for many reflections simultaneously, resulting in shorter data
collection times as compared to serial diffractometers. The CCD detector
contains a phosphor screen located immediately behind a beryllium window.
The screen converts incoming X-ray photons into optical photons. Because
of the greater energy of X-ray photons compared to optical photons,
hundreds of optical photons are produced for each incident X-ray photon. A
fiber-optics taper transmits the optical photons to the CCD chip, where they
are converted to stored electrons. The detector is very sensitive, permitting
the measurement of useful data from small specimens, like microcrystals
that are much smaller than samples that can be collected with a serial
diffractometer.

Because CCD area detectors collect all data contained in planes of

reciprocal space, it provides the opportunity for studying superstructures,

241

twinning, and crystal defects.

B. The Advantages and Disadvantages of Area Detectors.

One advantage of CCD area detectors is that they can read out data
much faster than other area detectors, thus the collection of thin slices of
data is practical. The area detector collects data from crystal of medium to
large unit cells in a fraction of the time required for serial scintillation
detectors to perform the same operation. On the other hand, (for the average
crystallographer) there is also a disadvantage to the CCD area detector in
that it allows for the collection of all the data which includes reflections
from twinning components, from tiny daughter crystals. These problems are
ordinarily solved by using slits to cut off the background signal or twinning
replications in the serial detector instruments, but this hardware option is no
longer possible for area detectors. In fact most of the “old tricks” used with
scintillation diffractometers cannot be used with area detectors. The
manipulation of the data, then, must rely on powerful new software options,
which is a different strategy than most crystallographers have been

accustomed to using.

C. New Problems in the CCD Era
(1) The problem of twinning
(a) Definition of twinning
Twinning may be classified according to the symmetry relation of the
twin components or to the origin of the twins. As defined in the
International Table of Crystallography Vol C :6
A twin consists of two or more single crystals of the same
species but in different orientations. Twin components are

intergrown in such way that at least some of their lattice

242

directions are parallel. The twinning law describes the

geometrical relation between twin components. It specifies a

symmetry operation, the twin operation that brings one of the

twin components into parallel orientation with the other. The

corresponding symmetry element is called the twin element.

The biggest problem with twinning is that it may not be easy to
distinguish between twinning and disorder.7 Although this is the case,
pursuing the possibility of twinning is advantageous. As George Sheldrick
has suggested, “You only need two more parameters to test if there is
twinning, but you may spend a long time struggling with a disorder model.”8

Below are the equations that describe the structure factor relationship

to reflections in the case of twinning or disorder:

F(h) = 2fi (h)exp(27zithi ) +ngj(h)exp(27tih ‘ xj)+(1-g)2fk(h)exp(2nih7xk)
i j 1:

Eq. 5
F1(h.) = 2f,(h,)exp(2nih,‘x,) Eq. 6
F2(h2) = ifi(h2)exp(2nih2’xi) Eq. 7
HM = {81Ft(ht)]2+(1-g)[F2(h2)]2}"2 Eq. 8

Where the xi’s are the position coordinates and the fi(h)’s are the
scattering factors for the non-disordered atoms, the xj’s are the position
coordinates and the fj(h)’s are the scattering factors for the disordered atoms
in the first position and the xk’s are the position coordinates and the fk(h)’s
are the scattering factors for the disordered atoms in the second position.
The g parameter is the fraction of disordered atoms in the first position and
l-g is the fraction of disordered atoms in the second position (Eq. 5).

In the twinning expression (Eq. 8), g is proportional to the volume of

the first component and l-g is proportional to the volume of the second

243

component. Note how this expression differs from the expression for
disorder. Often when a disordered model does not refine well (good R-
values cannot be obtained), it may be that the sample is twinned instead of
disordered.

Most types of crystallographic twins are merohedral/pseudo-
merohedral or rotational twins. For merohedral twins and pseudo-
merohedral twins there is an exact overlap of the twinning lattice and the
real lattice, thus no special data collection or integration procedures are
needed. For rotational twins, however, the twinning lattice does not exactly
overlap with the original lattice, thus the indexing routine attempts to
include reflections that belong to different twinning components in
determining the unit cell of a single component. In many cases, the program

cannot find a solution.

D. Manipulation of Data for Twinned Crystals - Why not Grow a Better
Crystal?

The progression of crystallography from being the tool of an elite
group of highly trained crystallographers to becoming the tool of the
everyday scientist has led to a new way of viewing the importance of a data
set. Rather than insisting on “perfect” crystals, many modern
crystallographers seek to solve a chemical problem in a practical manner and
in a timely fashion using the available crystals. Structural determination by
single crystal X-ray methods is important to nearly every chemist, and to
inorganic chemists in particular. The Cambridge Structure Data Base (CSD)
has witnessed a tremendous increase in the number of deposited structures
during the past decade. Many of these problems are being pursued because

of the need to address physical properties of new materials, often of

244

marginal crystal quality. Conventional crystallographers are finding
themselves at odds with their training when asked to handle crystals they
consider unacceptable. The conventional wisdom “garbage in, garbage out”
may not apply in the same sense as when structures were being evaluated for
their “crystallographic merit” as opposed to their “chemical information
merit”. The problems of small, twinned, strongly absorbing crystals are
falling into the hands of synthetic chemists who often have more incentive
than professional crystallographers to resolve these issues because the
success of their project depends on knowing the answer.

Before the age of the automated serial diffractometers when film
techniques were still in use, (1950’s and 1960’s), there were many
publications in the literature devoted to theoretical aspects of twinning.
After the advent of scintillation detectors, the number of papers that dealt
with this topic dropped to only an occasional paper because these machines
can’t deal with twinning easily.9 The process has come full circle now with
the widespread use of the CCD, and more researchers are tackling twinning
problems with renewed interest. The reason for this resurgence of interest is
that area detectors collect more data than a point detector, and one cannot
ignore data once it is obtained.

(1) Indexing of twinned crystals:

There are a number of ways one may index reflections from a twinned
crystal. These are summarized in the following sections.
(a) Simple algorithm method.

In the system software for various diffractometers, there is a built-in
routine to index reflections when twinning is suspected. The routine is not
very sophisticated, however, and is essentially a simple approach such as

indexing odd numbered reflections only or even number reflections, or some

245

other arithmetic combination.
(b) Difference vector and related methods.

There is a widely used program called DIRAX10 written by A. J. M.
Duisenberg, which sorts out the projection of reflections into so called T
vectors. Another program is Robert Sparks’ TWINNING11 package which
is distributed with the SMART 1000 CCD platform diffractometer and beta
tested by SMART users before the NT version release. In Sparks’ program,
there are two strategies for indexing. One uses the difference vector method,
and the other uses a full vector method. The SMART system uses difference
vectors for its indexing routine. If twinned data are encountered, many cross
vectors will be generated between different twinning components, which
makes it difficult to generate reasonable cell parameters. By reducing the
“fraction of difference vectors which must fit” number, a temporary cell
usually can be generated, and the XYZ coordinates of reflections calculated.
By loading the file that contains the temporary cell, orientation matrix and
the XYZ coordinates of reflections to TWINDX, one can quickly check if
the reflections belong to a unique domain or to twinned domains. When
running TWINDX, the shortest 100 difference vectors will be generated and
sorted by decreasing number of difference vectors in a group. The program
will ask for the number of difference vectors in a group to be used in the
indexing. Because the first three non-coplanar vectors may not be from the
same crystal, the program generates many combinations of three non-
coplanar vectors from the list of difference vectors. It can be shown that for
at least one such combination, all three difference vectors will come from
reflections from the same twin component. The program then outputs a
summary file which contains the cell parameters, orientation matrices and a

brief summary of fitted reflections with a combined figure of merit and a

246

detailed reflection fitting file. TWUTIL is then used to remove the indexed
reflections for solution A and prepare input file for indexing in the
subsequent run. TWINDX is run again with the new set of reflections.
After removal of reflections that fit solution A, it is easier for the rest of the
reflections to be indexed since they may now fit the same solution. From the
list of possible solutions, the one with the same cell parameters as solution A
is of interest. TWUTIL generates a new input file with only independent
reflections after comparing two indexing results. A further run of TWINDX
with only these reflections is preferred in order to obtain a refined solution
with all independent data. Finally, TWUTIL can be run again to generate a
file with cell parameters, an orientation matrix and the independent
reflection list for a specific solution. Finally, TWROT is run to determine

the rotation angle and the twinning law between a special pair of solutions.

E. Data Integration and Deconvolution.

After obtaining the indexing results from the TWINDX program,
there are several considerations for data integration, depending on the type
of problem encountered.

(1) A rotational twin.

A pair of rotational twins is usually rotated 180° from each other.
There may be some partial overlapping reflections which will affect the
refinement of the cell parameters and crystal orientation. A tight constraint
with integrated cell refinement turned off will generate a desired result. If
one of the components is of a much lower percentage than the other (as
judged by the number of independent reflections contributing to that cell), a
tightly constrained integration with global cell refinement will be

satisfactory.

247

(2) A nearly perfect twin.

In the case of a twin in which one solution only differs from the
other by a small angle, nearly all of the reflections are overlapped.
By using a loosely constrained or unconstrained integration
window to accommodate diffraction spots of both fragments, one has
the best chance for a stable integration. Essentially, the data are being
treated as if they are broad reflections with a high degree of mosaic

spread.

F. Structure solution and refinement.

Structure solution for a rotational twin can be handled in the
same way as normal data, with SHELXS and the corresponding
XS generating satisfactory results. In the case of severely affected
data, SIR92l2 or SIR97 is the program of choice to produce a
reasonable solution by a default procedure. SI-IELXL97 and SHELXTL
can use twinned data in HKLF 5 format and allow the BASF

parameter (twinning volume ratio) to be refined. Other
crystallographic packages such as CRSTAL13 by David Watkin can
also refine structures with twinned data. Merohedrally and

pseudo-merophedrally twinned data can be used directly in the
standard SHELX HKLF 4 format with refinement of the twinning
batch (BASF) parameter.

2. Mastering the CCD Technique for Special Needs

The chemical and structural details of the crystals for the data
sets presented here are discussed in the individual chapters. The
discussions in this chapter are limited to the type of twinning problem

that was evident and to the solution of the problems.

248

A. Problematic Structures: Seven Case Studies.
(1) Example 1: Large cell, “small molecule” data set: of
{[Moz(02CCMe3)3(NC)2CC(CN)CONH] 08CH2C12 }.. (3)

A point detector can only collect one reflection at a time, thus it takes
weeks for data collections from crystals with large unit cells. Unfortunately,
if the crystal is not stable in the X-ray beam, the crystal may not survive.
One such case is a compound prepared at low temperature and with a huge
quantity of incorporated solvent.

Single crystals of (3) were grown from the low temperature
crystallization of a reaction mixture of Moz(OzCBu‘3)4 with TCNE in
dichloromethane. The crystals are red hexagonal platelets of typical size
0.3 x 0.3 x 0.2 mm. Due to twinning and the presence of large channels that
contain numerous disordered solvent molecules, the data are very weak.
Repeated attempts to improve the technique of crystallization were not able
to yield a better crystal.

Data set 1 was collected at Michigan State using a Nicolet P3N
diffractometer controlled by a VAX 3520 workstation. Cell refinement from
14 well-centered reflections gave a hexagonal cell, with a = b = 27.101(4) A,
c = 22.8610) A, a: )3: 90°, y: 1200 and V = 14541.07 A3. T = -100 _4; 1
°C. The cell belongs to the hexagonal system space group P6/mcc or P6cc.
The observed reflections comprised only 14% of the entire data set, and their
average intensity I/0= 4. The structure was partially solved in the space
group of P6/mcc with the program DIRDIF84.”

Data set 2 was collected by Fr. Charles Campana at Siemens (Bruker)
using a SMART CCD 1K diffractometer equipped with a LT2 low
temperature device at —160(1) C. The cell indexed as a hexagonal system,

249

but the cell volume was three times larger than the one obtained using the
point detector instrument, with a = 46.88620) A, b = 46.88620) A,
c = 22.75680) A, a = )3: 90°, y = 120°, V = 43324.3(9) A3. The space
group could not be selected with confidence due to twinning and a large
amount of disorder. The structure was refined in several different space
groups and with different combinations of twinning laws. The differences
between the choice of space group are mainly based on the degree of
disordered methylene chloride.

(2) Example 2: A Pseudo-Merohedral twin: [Mn2(u"-0-TCNQ-

TCNQ)2(tt-cis-TCNQ)2(MeOH)2],, (9)

Crystals of (9) were grown by a layering reaction of
[Mn(MeCN)4](BF4)2 in acetonitrile with LiTCNQ in MeOH. The first
data set was collected using the Rigaku AFC6S instrument at —100 i 1 °C.
The automatic peak search routine indicated an orthorhombic crystal system
with
a = 14.4700) A, b = 27.3730) A, c = 13.1480) A, a = )3 = y: 90,
V = 5207.8(18) A3. There was a problem, however, with the solution.
The two space groups that met the systematic absence check are Cmcm
and Aba2. In these space groups, the TCNQ fragment resides on a
mirror plane that bisects the molecule. When the structure was expanded,
the TCNQ fragment became highly distorted as one side of the quinone ring
is smaller than the other side.

A second data set was collected on a SMART CCD area detector by
Dr. D. Powell at the University of Wisconsin, Madison. The solution turned
out to be a pseudo-merohedral twin in a monoclinic space group with the
beta angle just 0.05 off the perfect 90°. The twinning ratio was found to be
30:70.

250

cs C9

C5 and C9, C6 and C8,

which are related by local
symmetry, are assigned the
same thermal parameters in
the refinement

 

 

C6 C8

Figure 84. Using local symmetry to reduce independently refined
parameters.

The original data were then re-refined as pseudo-merohedrally
twinned data. This led to R1 = 0.0565 for 1078 F0 > 40(F0) and 0.2463 for
all 3069 data and wR2 = 0.2089, GOF = 0.982. Heavy constraints were
introduced during the refinement because of the low data to parameter ratio.
Only one-half of the atoms that form the TCNQ ring were refined
independently (Figure 84). The other half of the atoms in the TCNQ
quinone ling were constrained to have the same atomic displacement
parameters as the refined one by introduction of a mirror plane.

(3) Example 3: A rotational twin:
[EtsN]C1°FezCl4(MeOH)4(bem)3 (26).

(a) A data set from a scintillation detector compared to a data set
from CCD detector.

The first data set was collected on a Rigaku AFC6S diffractometer.

251

The unit cell was generated from the precise refinement of 20 well-centered
reflections in the two theta range from 14 — 30°. A standard quadrant of data
was collected according to convention. Three standard reflections were
periodically measured to monitor the decay of the crystal after every 150
reflections. The structure solution was from DIRDIF/SHELXS86 and
refined against F2 in Texsan V5.0 to give R = 0.112 (wR = 0.129) for 310
variables.

(b) Using a rotational twinning matrix in the refinement

It was noted that there are systematic absence violations to the c glide
plane in the category of h01 (l = Zn). The unit cell parameter has a ,8 angle
equal to 94° (close to 90°), with b and c being nearly equal. This situation
usually leads to rotational twinning. When the twinning matrix
[1 0 0.15, 0 - 1 0, 0 0 —1] was used in the refinement, the twinning ratio was
refined to about 1% contamination from the possible twins. The systematic
absence violation was resolved by this strategy.

A CCD data set was then collected in a monoclinic cell setting. The
unusually high R1 and systematic violation to the glide plane suggested the
possibility of twinning. The data frames were examined, but no visual
splitting of the reflections was noted. The preliminary orientation matrix
was then loaded to the TWINDX program. A pair of solutions from default
runs revealed that the crystal is indeed rotationally twinned. There were 54
out of 71 reflection indexed as independent reflections for twin domain one,
and 9 for twin domain two. There remained 12 completely overlapping
reflections and four diffraction spots that failed to be indexed. The
orientation of domain two was transformed to domain one by the
transformation matrix. The rotation angle is 180° between domain one and

two. The rotation axis is along a* (1 0 0) in reciprocal space or a (1 0 0) in

252

Table 37 . Rotational transformation of [Et4N]Cl-Fe2C14(MeOH)4(bpym)3
(26). from orientation A1 to orientation A2.

 

Orientation Matrix A1 from solution 1 (jalea0.p4p)
Orientation Matrix A; from solution 2 (jaleb0.p4p)

 

Orientation matrix and unit cell parameters for A1
0.05160 0.01662 0.04700
0.04626 0.01563 -0.04526
-0.02556 0.06128 -0.00166
13.5790 15.2931 15.3667 89.612 94.447 90.198

Orientation matrix and unit cell parameters for A2
-0.05247 -0.01658 0.04028
-0.04511 -0.01576 -0.05265
0.02471 -0.06139 0.00366
13.6468 15.2672 15.1025 89.119 94.152 90.827

Transform matrix from orientation 1 to 2 (twinning law)
Ag(inverse) * A1 [transforms hl to h2] =

-l.00375 0.00008 -0.15045

0.01109 -0.99810 0.02493

-0.02190 0.00190 0.98111

Rotation axis in reciprocal space = -0.07630 0.00990 1.00000
Angle of rotation around -0.08 0.01 1.00 = -l79.72050
Rotation axis in direct space = -0.01002 0.00317 1.00000

 

253

Table 38. Indexing result of rotationally twinned
[EttNICI°F62Cl4(MeOH)4(bPYm)3 (26)-

 

Nine of £1103: jaleaa.th / jalebb.th

 

figures of merit/reflection. indexed/reflections used:

0.00155/54/54 0.00339/9/9
unit cells:
13.579 15.293 15.367 13.647 15.103 15.267
0.008 0.006 0.008 0.249 0.275 0.119
89.612 85.553 89.802 3181.43 89.119 89.173 85.848 3137.68
0.040 0.047 0.042 2.92 1.141 1.145 1.425 82.42
Refl. a x L flags 3 x L
1 -5.011 1.001 -6.025 1 - 1 - -4.104 5.987 1.094
2 —0.017 2.987 -3.008 1 + 0 - 0.455 2.914 3.056
3 3.000 5.995 -0.990 1 + 0 - 3.180 0.862 6.042
4 4.000 6.999 1.009 1 + 0 - 3.884 -1.122 7.005
5 —1.992 6.998 -6.976 1 + 0 - -0.930 6.843 7.137
6 -5.008 1.992 -7.017 1 - 1 — -3.952 6.959 2.108
7 0.807 1.984 ~1.074 0 — 1 + 0.991 1.001 2.017
8 4.001 7.997 0.011 1 + 0 - 4.034 -0.145 8.027
9 2.012 7.000 -2.985 1 + 0 - 2.489 2.840 7.084
10 0.990 4.989 -2.992 1 + 0 - 1.464 2.873 5.066
11 0.795 1.871 -1.131 O - 1 + 0.988 1.057 1.905
12 1.993 3.994 -1.002 1 + 0 — 2.171 0.900 4.034
13 -4.608 -1.049 -4.010 0 - 1 + -4.003 4.006 -0.997
14 2.978 2.994 1.975 1 + 0 - 2.711 -2.041 2.973
15 0.991 2.996 -2.014 1 + 0 - 1.318 1.917 3.052
16 0.970 2.054 —0.971 1 + 0 - 1.139 0.896 2.086
17 0.013 7.000 -4.985 1 + 0 - 0.783 4.845 7.112
18 -1.002 4.995 -5.995 1 + 0 - —0.084 5.863 5.124
19 1.981 2.999 -0.025 1 - 1 - 2.012 -0.056 3.016
20 -0.001 3.997 -4.004 1 + 0 - 0.621 3.889 4.089
66 2.009 -4.002 2.016 1 + 0 - 1.732 -2.046 -4.022
67 2.014 -1.979 2.027 1 + O - 1.736 -2.060 —2.003
68 3.558 0.061 3.967 0 - 1 2.994 -4.002 0.002
69 3.010 -2.003 3.023 1 + 0 - 2.586 -3.060 -2.041
70 0.999 —4.995 1.004 1 - 1 - 0.871 —1.029 -4.999
71 -4.000 -6.993 -2.007 1 + 0 — -3.694 2.038 -6.973
72 3.984 -7.002 4.998 1 + 0 — 3.266 -5.009 -7.069
73 1.132 -1.974 1.000 0 - 1 + 1.005 -1.033 —1.982
74 1.130 —2.017 0.918 0 - 1 + 1.016 -0.953 -2.023
75 -2.986 -3.986 —1.989 1 + 0 - -2.679 1.992 -3.961
76 6.004 -1.003 8.005 1 + 0 - 4.841 -8.015 -1.133
77 3.010 -2.996 3.025 1 + 0 - 2.585 -3.060 -3.032
78 -0.011 —8.006 0.984 1 + 0 - -0.140 -0.982 -8.015
79 0.993 -5.859 1.124 0 — 0 - 0.846 —1.145 -5.864

W of indexing result (flags):

0+0+ 0+0- 0+1+ 0+1— 0-0+ 0-0- 0-1+ 0-1— 1+0+ 1+0- 1+1+ 1+1- 1-0+ 1—0- 1-1+ 1-1-
0 0 0 0 0 4 9 0 0 54 0 0 0 0 0 12

net. for tlnga(£itting 0040):

0—0—: Refl not been indexed and not used for indexing;

1+0-: Independent reflections for cell 1;

0-1+: Independent reflections for cell 2;

l+1+: Non-independent reflection used in indexing;

1-1-: Non—independent reflections fitted to both solution, but not been

used in indexing.

254

direct space. The cell parameters are a = 13.558(3) A, b = 15.229(3) A, c =
15.390(3) 11, ,6: 94.30(3)°. The orientation file was subjected to TWINDX
treatment which resulted in two orientation cells with 54 and 6 independent
reflection respectively (see Table 38). Because the second domain was
relatively weak, data were reintegrated with tight cell constraints for
integration and global cell refinement. Cell refinement was set to run on
every 100 frames, with only the strong reflections (1/0 > 20) to avoid
influence from the minor twinning domain. The reprocessed data refined to a
final R] = 0.0633, wR2 = 0.1392 on 2473 (I > 20(1)) against 326 parameters
and R1 = 0.1826, wR2 = 0.1670 (all data) for all 3895 reflections.

(4) Examples 4a and 4b: Nearly perfect twins:

(8) an(bpym)2(H-F)(n 1-BF4)(BF4)2(MeCN)'2MeCN (25).

Crystals of (25) were grown from a layering reaction of
[Mn(MeCN)4][BF4]2 and 2,2’-bypyrimidine in acetonitrile. A crystal of size
0.3 x 0.35 x 0.45 mm was mounted on the tip of a glass fiber with Dow
Corning Silicone grease. Examination of 45 preliminary frames indicated
that the crystal belongs to the triclinic system. A hemisphere of data was
collected with 25 sec/frame. The following cell parameters were indexed
from 45 preliminary frames: a = 12.299(5) A, b = 12.480(4) A, c =12.690(6)
A, a: 100.98(5)°, fl: 116.63(3)°, y: 101.02(4)°, v= 1622.23(98) A3.

The predicted reflection positions matched poorly with the data
frames. Slightly split reflections were found after careful examination of the
frame data. The XYZ coordinates of 38 strong reflections were saved to an
input file and R. Sparks’ TWNNING package was used to deconvolute the
data. Two similar cells with different orientation matrices were indexed
with 17 and 12 reflections separately. The relation of the orientation

matrices is as shown in Table 39. The rotation angle from solution 1 to

255

solution 2 is only 1.5°

Data integration was carried out with triclinic constraints, with only
global cell refinement. The resulting integrated data were processed by
TWHKL to generate the reflection file in HKLF 5 format, which indicated
that there are 2,478 single reflections and 11,340 complewa overlapping
reflections; 6,228 partially overlapped reflections were rejected. Structural
solution and refinement was carried out by using the SHELXTL package.
All non-hydrogen atoms were located from the XS output and refined with
XL on F2. The RI values with and without TWHKL processing are 14.3%
(5272 F0 > 40(Fo) unique reflections) and 18% (6605 unique reflections)
respectively.

(b) Structure of Mn(n1-TCNQ)2(TCNQ)(HOCH2CH20H)3 (28).

For this data set, the TWROT indicates that the two orientation
matrices of the split crystal are separated by only ~1°. Inspection of the
frames indicated that some of the reflection spots are well separated, (about
3—4 frames apart in the z direction) while some frames are not. To better
refine the structure, the following strategy was used:

(i) Big box integration.

In the big box integration, the integration box was set to be very large
so as to include both reflection mates. The box (peak shape) refinement was
turned on. This method treats the total intensity of both crystals as the
intensity of an averaged, but single, mosaic crystal.

(ii) Small box integration.

In the small box integration, several options may be used which lead
to different results.

(1) Refine and update the orientation of the cell every 200 frames, and
refine the box (peak shape) as well.

256

Table 39. Rotational transformation of Mn2(bpym)2(p.-F)('ql-BF4)(BF4)2
(MeCN)°2MeCN (25) from orientation A1 to orientation A2.

 

Orientation Matrix A1 from solution 1 (twin11.p4p )
Orientation Matrix A2 from solution 2 (twin22.p4p )

 

Orientation matrix and unit cell parameters for Al
-0.01138 -0.06826 0.01943
-0.06702 -0.05150 -0.08071
0.05394 -0.03308 -0.04773
12.3459 12.8265 12.3121 116.679 101.227 100.477

Orientation matrix and unit cell parameters for A2
-0.01200 -0.06871 0.01901
-0.06562 -0.05272 -0.08209
0.05557 -0.03091 -0.04578
12.3795 12.8417 12.2807 116.575 100.726 101.494

Transform matrix from orientation 1 to 2 (twinning law)
Ag(inverse) * A1 [transforms hl to h2] =

0.99070 -0.03093 -0.02947

-0.00053 1.00144 0.00079

0.02473 0.00895 1.00619

Rotation axis in reciprocal space = 0.76207 -5.00000 2.15606
Angle of rotation around 0.76 -5.00 2.16 = -1.46653
Rotation axis in direct space = -0.02766 -1.00000 -0.00931

 

257

Table 40. Rotational transformation of
Mn(TCNQ)2(TCNQ)(HOCH2CH20H)3 (28) from orientation A1
to orientation A2.

 

Orientation Matrix A1 from solution 1 (sao6a.p4p)
Orientation Matrix A2 from solution 2 (sao6b.p4p )

 

Orientation matrix and unit cell parameters for A1
0.00456 —0.06331 -0.02571
0.00176 0.03669 -0.06280
0.09500 -0.00100 0.00211
10.5244 13.8093 14.8698 82.128 89.394 87.314

Orientation matrix and unit cell parameters for A2
0.00575 -0.06356 -0.02453
0.00138 0.03578 -0.06284
0.09465 0.00066 0.00242
10.5518 13.8540 14.9611 81.968 89.713 87.879

Transform matrix from orientation 1 to 2 (twinning law)
A2(inverse) * A1 [transforms hl to h2] =

1.00343 -0.01718 -0.00354

0.01751 1.00015 0.01516

0.00401 -0.01480 1.00790

Rotation axis in reciprocal space = 1.25372 0.83509 -3.00000
Angle of rotation around 1.25 0.84 -3.00 = 1.32474
Rotation axis in direct space = 3.86700 2.11682 -5.00000

 

258

(2) Refine only the orientation of the cell with a fixed box size.

(3) Use the orientation matrix from the TWINDX without further peak
shape or orientation matrix refinement. A global cell refinement is
performed when the integration is finished with only those reflections with
U0 > 10.

Because some diffraction spots were not sufficiently well separated,
peak shape refinement led to incorrect cell parameters and the data could not
be used. Data from method (1) were ultimately used to solve the structure.
The refinement led to R1 = 0.102 for 3846 F0 > 40(Fo) and 0.203 for all
9938 data and wR2 = 0.263, GOF = 0.975.

(5) Example 5: Crystal movement during data collection:
Structure of Rh2(pynp)2(OzCCH3)2(BF4)2-C7Hs (27).

Crystals of (27) were grown from the layering reaction
of Rh2(OzCCH3)2(MeCN)6(BF4)2 and a toluene solution of pynp
(pynp = bis(pyridylnaphthyridine)). Preliminary data indicated that the
crystal was monoclinic. The final data, however, failed to reproduce the cell
with the autoindex routine in SMART. After reducing the required
reflection ratio for indexing from 80% (default) to 70%, a temporary cell
was obtained, which enabled the indexing/least-square routine to calculate
the XYZ coordinates of the reflections. SMART was re-run in the debug
mode to produce the XYZ coordinates of the reflections which were
manually edited to remove the temporary indexes. These were saved as a
raw file for use in the TWINNING program. Two solutions had similar cell
parameters but different orientation matrices. TWINDX yielded two cells
with a rotational angle of ~37° between them. Careful inspection of the
indexed reflection list revealed that the first cell is associated with the first
half of the data while the second cell is associated with the second half of the

259

Table 41. Rotational transformation of Rh2(OzCMe)2(pynp)2.(BF4)2-C7H3
(27) from orientation A1 to orientation A2.

 

Orientation Matrix A. from solution 1 (rhl 1r.p4p)
Orientation Matrix A2 from solution 2 (rh22r.p4p)

 

Orientation matrix and unit cell parameters for A1
-0.05238 -0.03188 0.00858
-0.01790 0.02133 0.06063
-0.05031 0.02560 -0.03982

13.4185 21.6846 13.7409 90.000 94.922 90.000

Orientation matrix and unit cell parameters for A;
-0.02631 004084 -0.02498
-0.05100 0.00047 0.04946
-0.04842 0.02170 -0.04809

13.3652 21.6223 13.6780 90.000 94.828 90.000

Transform matrix from orientation 1 to 2 (twinning law)
Ag(inverse) * A1 [transforms hl to h2] =
0.80066 -0.27547 -0.32054
0.48593 0.87310 -0.55440
0.45912 0.13885 0.90063

Rotation axis in reciprocal space = 2.13 -5.00 2.99
Angle of rotation around 2.13 —5.00 2.99 = —37.62720
Rotation axis in direct space = 3.92954 -3.13519 5.00000

 

260

Table 42. Comparison of two TWINDX solutions of (27).

 

Nine of files:rh11.th/ rh22.

tWX

 

figures of merit/reflections indexed/reflections need:
0.00233/55/55

0.00169/77/77

005
007
002
975
986

992
056
992
001
009
987
994
964

.706

524

.258
.063

361
689
699
931

.246

942

.243
.381

.093
.244
.521
.694

unit celle:
Refl!

1 -3.

2 -3.

3 —3.

4 —2.

5 -1.
81 -2.
82 —7.
83 -2.
84 -5.
85 -7.
86 -5.
87 —8.
88 -1.
89 -2
90 -7.
91 -9
92 —2
93 -0.
94 -4.
95 -3.
96 -4.
97 -2
98 -3.
99 -0
100 1
155 4
156 8
157 8
158 7
159 0

.023

FJKJFIKJH

-3

-0

OU‘U‘lfiko

K

.993
.990
.997
.029
.022

.985
-6.
-4.
-6.
-7.
-6.
-8.
.727
.462
—4.

317
003
007
998
041
962

169

.374
.819
.293
.297
.380
-4.
-1.
—3.
.185
.162

829
837
320

.567
.570
.208
.414
.017

-6
-5

-5.
-7.
.041
-6.
-7.
-6.
.715
-6.
.973
.446
.744

-7

-7

-7
-7
-7

-6
10

.007
.011
.002
.005
.010

.990
.853
.008
.998
—1.

4.
-8.
.306
.730

000
021
004

157
154
001
085
942

994

.486
.386
-8.
-8.
-0.

977
033
043

Summary of indexing reeult (llege):
0-0+ 0—0- 0-1+ 0-1— 1+0+

0+0+ 0+0- 0+1+ 0+1-

liege
+ 0 -

HIAIahiH
+ + + +
c>c>c>o

I

I I + + + + + I +
OIaIahah-OIAIAhac>o<3+ac>c>o<3c5c>oz
I

O<DCDC>O<DC3C>O<DCDC>ODHFJFJP‘P‘OO‘
I
I + + + + I + + + I

F‘CDCDCio
I
HIarath
I + + + +

8
-1.955
-1.611
-1.068

1.603
0.967

-5.876
-3.374
-3.149
-4.614
-7.391
-3.885
12.000
—2.970
-3.602
-8.196
-9.837
-2.972
-0.982
—5.998
-4.758
-8.000
-5.032
-6.995
-2.978
-1.100

2.962
1.988
2.998
2.990
0.035

UUlU-lfiw

-5
-1
-7

—3

.206
.284
.870
.864
.571

.826
-O.
—2.
-3.
.049
.846
.295
.008
.240
-1.
-0.
.001
.995
-O.
.304
-4.
-2.
-3.
-0.
.231

620
624
352

989
824

008

983
999
985
991

.986
.005
-2.
-1.
.035

005
014

~1.562
—1.891
0.262
5.698
3.949

-4.466
10.294

2.226

4.320

3.695

7.344
-1.821
-S.988
-4.851
-1.196
-2.804
—6.986
-6.980
-4.975
-5.659
-4.001
-5.045
-4.994
-6.981
-8.661

-10.014
-13.039
-12.042
-11.030

0.056

1+0- 1+1+ 1+1— 1-0+ 1—0— 1-1+ 1-1-

 

0 0 0 0 55 0 0 77 0 0 0 0 0 2
the for f1ege(£itting code):
0-0-: Refl not been indexed and not used for indexing;
1+0— Independent reflections for cell 1;
0-1+- Independent reflections for cell 2;
1+1+ Non-independent reflection used in indexing;
1-1- Non-independent reflections fitted to both solution, but not been
used in indexing.

261

data. This is a clear indication that the crystal moved during data collection.
After examining the data frames with the two orientation matrices, a third
orientation matrix was needed to integrate a hundred data frames obtained
after the time that the crystal moved. In the end, about 100 frames (10% of
the total data) were unusable. This is thought to be related to a large
movement that was ongoing during these frames before the crystal finally
settled into a new position.

Data integration was carried out separately on the two orientation
matrices. The first cell was used for the first part of the data (about one third
of the total frames), and the second cell was used to treat the second half of
the data. A third cell was used to integrate a hundred data frames during the
time the crystal was moving. The final data were corrected for
inhomogeneity, decay and absorption with SADABS and then merged by
XPREP and truncated to a resolution of 0.95 A (20 = 47°) to suppress effects
of solvent and anion disorder.

The structure was solved by SIR97.12 A main structure fragment and
the disordered toluene were refined without problem. The disordered [BF4]'
anions, however, posed a major problem in further refmement. Both [BF4]'
anions are near 2-fold symmetry elements, and were modeled at 50%
occupancy. The final R1 is 8.9% based on 1,399 reflections (1/0 > 4) and
R1 = 14.2% with all 2,341 data refined against 293 parameters.

(6) Example 6: A quadruply twinned crystal:CuTCNQ phase II
(8).

CuTCNQ phase II was grown by a diffusion reaction between CH3CN
solution of CuI and TCNQ in a 3-compartment fiitted cell. Details of the
crystallization technique are described in Chapter 4. The habit of the
crystals is square platelet with a typical size being 0.13 x 0.13 x 0.01 mm.

262

The first data set was collected by Dr. Charles Campana of Siemens
(Bruker) on a SMART 1K platform CCD diffractometer. There were
problems during the indexing, and the final cell was indexed by selectively
removing reflections introduced by possible twinning components. Data
integration was carried out with the program SAINT on the indexed cell
with cell parameters a = 5.3337(8) A, b = 5.3312(8) A, c = 18.875(3) A,
,6: 94.036(3)° and V = 535.3804) A3. Systematic absences indicated that
the space group is P2. Structure solution was obtained from XS and refined
against F2. The MISYMM program in the PLATON software package
indicated that an inversion center exists in the structure model. This result,
however, is contradicted by the systematic absences to the data. Rotational
twinning or multiple twinning was proposed, and a four-fold rotation
twinning law [0 l 0, -1 0 0, 0.25 0.25 1] was used in the refinement to model
the twinning in space group P2/n. The off diagonal terms are the
contributions from the non 90 degree ,8 angle. The R1 dropped from 20% to
16% after refinement. The twinning volume ratio for individual twinning
components are refined as 52%, 8%, 15%, 25%.

A second data set was collected at MSU on a SMART 1K CCD
diffractometer equipped with an Oxford Cryostream LT device. A platelet
crystal of approximate dimensions 0.13 x 0.13 x 0.05 mm was mounted on
the tip of a glass fiber with Dow Corning Silicone grease. The crystal was
also found to be severely twinned. A full sphere data set was collected over
a 72-hour period with an exposure time of 90 sec/frame. Of the total
reflections, 253 reflections with U0 > 20 were selected, but the indexing
routine was unable to produce a reasonable cell. The XYZ coordinates of the
reflections were subjected to the treatment of the TWINDX routine. By
using 100 reflections and fitting ratio of 20%, TWINDX gave three solutions

263

with different orientations. Solution B is rotated 90° from solution A and C
while solution C is rotated 180° from solution A. When the results were
discussed with Dr. V. Young of the University of Minnesota and Dr. R.
Sparks of Siemens, Dr. Young pointed out that there might exist a fourth
solution related to solution B. Dr. Sparks offered to process the data with an
updated version of his TWINDX routine that he redimensioned to handle
this data set, which contained more than two twinned domains for the
normal routine. In this manner, a fourth solution was determined and
confirmed by the SMART indexing routine with the following relation to the
first three.

A., = A2 . A1“. A3

The 238 reflections that were used for the indexing are distributed
among the four twinning domains according to the numbers 99, 100, 88 and
83 respectively. There remained 71 reflections (of maximum intensity I/o <
5) that did not fit any of the four solutions.

The data were integrated separately with four orientation matrices. A
revised version of TWHKL was then used to process the four sets of
integrated data together. Due to the complicated twinning situation and the
exclusion of partial overlapping reflections, however, the combined data
processing did not improve the refinement. After careful examination of the
relations between the four twins, an alternative method was used to generate
a set of HKLF 5 formatted data from a single integrated data set by one of
the following conditions, which are derived from the twinning law. The
orientation matrices of other cells could be transformed from one cell by the
3 x 3 matrix (0 -1 0, 1 0 0 0.25 0.25 1), which rotates 90° successively along
the (0 0 1) axis in reciprocal space. Thus the following relationships related

the other twins to the solution A:

264

h’ = k, k’ = -h, l’ = (h-k)/4 +1;

h” = -h, k” = -k, l” = h/2 +1.

h’” = -k, k’” = h, l’” = (h+k)/4 +1

A simple FOTRAN routine was written to treat the integrated
reflection data from solution A. A new data set with the HKLF 515 format
for overlapping reflection were given different batch numbers according to
their relationship, partial overlapped reflection was removed. The structure
was solved and refined with SHELXTL in the monoclinic space group P2/n
with a final R1 of 15.9% and wR2 of 34.5% (Fo>4 o(F0) ). The volume
ratio of the twinning domains were refined to be 0.35, 0.19, 0.25, and 0.21.
(7) Example 7: A composite of two polymorphic crystals: Two

morphologies of [ppn]2NiCl4-Me2CO, (23, 24).

A hexagonal platelet of crystal (24) was examined on a SMART 1K
CCD Area Detector and indexed to give the triclinic cell a = 13.620(10) A, b
= 13.646(7) A, c = 21.44303) A, a = 104.52(7)°, fl = 95.52(5)°, y =
115.35(5)°, V = 3390(4) A3. A hemisphere of data was collected at 30
sec/frame. Without symmetry constraints the data integration gave a
different cell of a = 14.3861(3) A, b = 14.1148(1) A, c = 22.6380(5) A, a:
90.0000(11)°, 13 = 90.0000(12)°, y = 115.9618(6)°, V = 41329205) A3.
After reprocessing the data, one third of the 465 reflections for indexing was
removed from the least-square refinement as their error in indices are much
bigger than allowed tolerance, and the cell reverted from the monoclinic
setting back to a triclinic cell.

To identify the cause of the problem, TWINDX was then used to
index the data, and a triclinic cell was indexed from 355 out of 464
reflections. A second run with the remaining reflections revealed that the

rest of the reflections indexed to an orthorhombic cell which is the same as

265

(23). This is a very rare case in which one can actually index reflections that
belong to different crystal systems. A comparison of the orientation
matrices and cell parameters with TWROT indicated that the two solutions
are rotated 37° from each another.

Data integration was carried out on the triclinic cell from the
TWINDX result with local and global cell constraints. The data were
partially contaminated from the orthorhombic data. The final statistics were
R1 = 0.0888, wR2 = 0.1944 (for I >20(I)), and R1 = 0.2881, wR2 = 0.2507
(for all data).

Data integration for the orthorhombic cell with a global constraint
produced a second data set. It should be noted that the reflections are
weaker than those belonging to the triclinic cell. Least-squares refinement
on F2 led to R1 = 0.1355, wR2 = 0.1990 (for I >20(I)), and R1 = 0.3753,
wR2 = 0.2746 (for all data).

B. Discussion
(1) Considerations before data collection.

It is important to select the best crystal for a crystallographic
experiment! When problems such as twinning and fragmentation occur due
to crystal defects or stress during growth, some precautions may be taken to
avoid these problems. A quick rotational photograph typically reveals
problems for a crystal of normal size. For microcrystals, a longer exposure
time is needed to examine the quality of the candidate for data collection. In
the event that performing a still photograph is not sufficient, one may collect
all the data at longer exposure times. Sometimes the dilemma becomes
whether to conduct an exhaustive search for the best crystal and versus

collecting data for the “less than perfect” crystal in hand.

266

(2) Data collection.

With a CCD area detector, data may be collected even without
successful indexing of a unit cell. This is especially helpful when the crystal
is very weak. To be able to handle potential twinning problems, a full
sphere of data is preferred.

(3) Data processing.

When using a serial detector to collect data for a problem crystal, one
may vary options such as using a narrow slit. Precise crystal centering can
also be used to reduce the effect of twinning or to prevent the collection of
data on small daughter crystals. When CCD detectors are used, hardware
options are not available. Nevertheless, the software options during data
integration such as the integration box size, and peak-shape profile
refinement are useful resources when facing problem data collection.
Narrowing or widening the integration window is one of the options that can
have major effects on the quality of the final data. In the case of small
angles between the twinning domains, one can widen the integration window
to allow for the chance to treat the nearly perfectly overlapped reflections as
one with broad mosaic spread. When a bigger separation of reflections is
encountered, one can narrow the window to avoid the daughter crystal
reflections altogether so as to be able to treat the data as a small single
crystal. It is still important, however, to select the best crystal and carefully
choose data collection parameters to give one a chance to obtain the best
data.

(4) Structure solution and refinement.

It is common for twinning laws to contain off-diagonal terms when a

monoclinic system exhibits a ,6 angle ~ 90° and the a and b are nearly equal.

A 180° rotation along the (0 0 1) axis will generate a twinning law of

267

Table 43. Rotational transformation of [ppn]2[NiCl4]°Me2CO (23b, 24)
from orientation A1 to orientation A2.

 

Orientation Matrix A1 from solution 1 (40ctra.p4p )
Orientation Matrix A2 from solution 2 (40ctrb.p4p )

 

Orientation matrix and unit cell parameters for A1
0.06467 0.07256 0.00044
-0.02006 -0.02616 -0.04989
-0.04898 0.03680 0.00080

13.6943 13.7336 21.2670 104.039 95.809 115.908

Orientation matrix and unit cell parameters for A2
-0.04288 -0.03938 0.02990
0.07211 0.01337 0.02502
-0.03930 0.06670 0.01278

10.7946 12.7217 24.3727 90.163 90.291 90.246

Transform matrix from orientation 1 to 2 (twinning law)
Ag(inverse) * A1 [transforms hl to h2] =

-0.26483 -0.75304 -0.42303

-0.98377 -0.1 1981 -0.09892

0.48735 1.18885 -0.72199

Rotation axis in reciprocal space = -3.00000 2.97493 1.10602
Angle of rotation around -3.00 2.97 1.11 = 144.57920
Rotation axis in direct space = -2.98588 4.00000 1.18486

 

268

Table 44. Comparison of indexing results of the two TWINDX solutions of
ppn]2[NiCl4]-Me2CO (23b, 24).

 

Nhne of £iles:40tr00.th/40tr01.th

 

figures of merit/reflections indexed/reflections used:

0.00334/355/355 0.00243/94/94
unit cells:
13.694 13.734 21.267 10.795 12.722 24.373
0.009 0.006 0.011 0.007 0.013 0.019
104.039 95.809 115.908 3390.44 90.163 90.291 90.246 3346.94
0.036 0.045 0.039 2.94 0.074 0.058 0.069 4.82
Refl! 8 K L flags 3 x L
1 5.971 -3.987 -3.962 1 + 0 — 3.121 -S.013 1.007
2 -0.973 0.966 11.047 1 + 0 - -5.119 ~0.259 —7.326
3 2.990 -1.000 2.046 1 + 0 - -0.880 -3.032 -1.233
4 6.003 —2.020 5.008 1 + 0 - -2.163 -6.167 -3.116
5 6.017 -2.028 6.002 1 + 0 - -2.582 -6.279 -3.836
456 1.993 -5.789 -2.636 0 - 1 + 4.971 -1.015 -4.031
457 0.517 4.398 -10.481 0 1 + 1.009 -0.007 13.024
458 —0.029 -1.032 -7.935 1 0 - 4.166 0.929 4.464
459 0.868 -2.261 -5.886 0 - 1 + 3.987 -0.009 1.961
460 -0.435 2.451 -8.750 0 - 1 + 1.995 0.992 8.995
461 0.980 -3.033 -5.908 1 + 0 - 4.548 -0.025 1.115
462 0.821 -3.045 -4.458 0 - 1 + 3.986 -0.011 —0.025
463 -0.068 -1.045 -4.931 1 + 0 - 2.914 0.671 2.262
464 —0.077 —1.065 -5.923 0 - 0 - 3.352 0.781 2.950

Gunnery of indexing result (rlegs):
0+0+ 0+0- 0+1+ 0+1- 0-0+ 0—0- 0-1+ 0-1- 1+0+ 1+0- 1+1+ 1+1- 1-0+ 1—0- 1-1+ 1-1-
0 0 0 0 0 14 94 0 0 355 0 0 0 0 0 1

 

the for £1egs(fitting code):

0-0-: Refl not been indexed and not used for indexing;

1+0-: Independent reflections for cell 1;

0-1+: Independent reflections for cell 2;

1+1+: Non-independent reflection used in indexing;

1-1-: Non-independent reflections fitted to both solution, but not been
used in indexing.

269

[-1 0 0, 0 -1 0, .5 0 1] (Example 6). The off-diagonal term can readily be
calculated from the formula 2ccos/3’b.9 In the case where b and c are nearly
equal, the rotation will be around (1 0 0) (Example 3). Structure solution for
a twinned crystal is not as straightforward as for a non-twinned crystal.
Sometimes the Patterson methods can be used to locate the heavy element
position. Structure refinement for twinned crystal data from merohedral
twins and pseudo-merohedral twins (Example 2) could be handled by using
the built-in capacity of a refinement program such as SHELXL978 or
CRYSTAL.l3 Solutions for crystal with rotational twins may be obtained
by using data set integrated from one of the twinned orientation. To achieve
the best results for refinement, reflection data that contain both orientations
should be used. However, unless special programs like UNTWIN16 or
TWHKLS17 are used to handle the partially overlapping reflections, the
refinement residual index (R1) is usually much higher than for an non-

twinned crystal.

3. Conclusions

With the help of a CCD X-ray instrument, graduate students that
depend on single crystal structures can be more productive when
crystallography is routinely needed for their research. Considerable effort,
however, is still required in order to handle problem crystals that are more
frequently encountered when using the high resolution and sensitivity of a
CCD camera. Basic education and training in handling problem crystals,
especially those that suffer from twining, should become a standard teaching
procedure in CCD equipped X-ray facilities.

10.

270
References

Baker T. I UCr Newslett. 1997, 5, 1.
Baker, T. I UCr Newslett. 1999, 7 ,1

Friedrich, W., Knipping, P., and Laue, M. Sitzungsberichte der
mathenatisch-physikalischen Klasse der Kb’niglichen Bayerischen
Akademie der Wissenschafi‘en zu Mu'nchen. 1912, 303-322. English
translation: Stezowski, J. J. In: Structural Crystallography in
Chemistry and Biology. (Ed., Glusker, J. P.) Hutchinson & Ross:
Stroudsburg, PA. 1981, 23.

Bragg, W. L. Proc. Roy. Soc. (London) 1913, A89, 248.

X-ray Structure Determination—A Practical Guide, Stout, G. H. and
Jensen L. H. 2nd Ed. John Wiley & Sons, Inc. USA 1989 173.

Wilson, A. J. C., Ed., International Tables for Crystallography,
Volume C, Kluwer Academic Publishers, Dordrecht, The Netherlands,
1995.

(a) Hoenle, W. and von Schnering, H. G. Z Krist, 1988 184, 301. (b)
SHELXTL V5 Reference Manual. 1997, Ch8. Siemens Energy &
Automation, Inc. Analytical Instrumentation.

SHELX97 [Includes SHELXS97, SHELXL97, CIFI‘AB (and
SHELXA) ] - Programs for Crystal Structure Analysis (Release 97-2).
Sheldrick, G.M., Institiit fur Anorganische Chemie der Universitat,
Tammanstrasse 4, D-3400 Gottingen, Germany, 1998.

Watkin, D. Twinning. Don ’t give up -- yet. Presented at BCA Autumn
Meeting, Bristol, November 12th 1997.
http://darkstarxtlox.ac.uk/refs bris97.html. Online published: Jan 28,
1998, Site Visited: Mar 26, 1998.

(a) Duisenberg, A. J. M. J. Appl. Cryst. 1992, 25, 92. (b) DIRAX,
Duisenberg, A. J. M.Private communication.

11.

12.

13.

14.

15.

16.

17.

271

Sparks R. A. TWINNING ('I'WINDX, TWUTIL, TWROT and
TWHKL), Program for processing twinned data. Bruker AXS
Instrument, Madison, WI.

(a) SIR92 —A program for crystal structure solution. A. J. Appl. Cryst.
1993.26, 343. (b) SIR97, Altomare, A., Cascarano, G., Giacovazzo, C.
and Guagliardi 1997 .

Watkin,D. Crystals32 Release 10 user’s manual.

DIRDIF 84 - Beurskens, P. T.-Direct Methods for Difference
Structures; Technical Report 1984/1, Crystallography Laboratory,
Toemooiveld, 6525 Ed Nijmegen, Netherlands.

Reflection format that contains h, k, 1 F2, 0 and a batch number, the
batch number indicated whether a relation is an independent reflection
or overlapping reflection. Volume ratio of twinning domains could be
refined in least-square refinement.

Young, V. G. Jr. New Strategies in the Refinement of Rotationally
Twinned Materials, Poster presentation in ACA 1998 Meeting St.
Louis, MO.

TWHKL5, Sparks R. A. and Young G. V. Jr. Program for processing
twinned reflection data based on UNTWIN and TWHKL, private
communication.

APPENDICES

272

APPENDIX A

GENERAL PROCEDURES FOR DATA
COLLECTION

273

274

1. Data collection procedure for Siemens CCD platform diffractometer
Data collection procedures on a Siemens CCD platform diffractometer

involved the following steps:

A. Crystal mounting and preliminary data collection.

Crystals were secured on the tip of a short glass fiber typically (1/8 of
an inch) with Dow Corning silicone grease and inserted into a copper pin on
a goniometer. The instrument is a Siemens SMARTl 1K CCD platform
diffractometer with graphite-monochromated Mo Kor radiation source ( AOL =
0.71073 A ) powered at 50 kV and 40 mA. The crystal was bathed in a cold
stream of nitrogen, which serves to solidify the grease, with an Oxford
Cryosystem Cryostream cooling device. Intensity data were collected with
0.3° frames by the (1)-scan technique with a detector-to—crystal distance of 5
cm. Initial cell parameters and an orientation matrix were generated from a
set of preliminary data collection of 10 sec frames, from three detector
positions (detector was positioned at 20 = i 26°, when (I) = 0°, and 20 = -

26° when (I) = 90°, 15 frames were taken for each detector position).

B. Data collection.

Data collection was performed according to the Laue symmetry of the
preliminary data. For crystal systems of monoclinic and higher symmetry,
only a hemisphere of data was collected. In the case of triclinic systems, a
full sphere of data was collected to ensure sufficient redundancy for the
absorption correction (Described in Step three). For a hemisphere data
collection, the detector was successively set to three initial positions ¢ = 0,
90 and 180, with 20 = (D = -28. Frames were measured at 30 sec for all (1)
positions; 606 frames were measured for the first (1) position, and 453 and

230 for the second and third (1) position to ensure sufficient coverage for a

275

resolution of 0.75 A and redundancy above 2 for a hemisphere. An extra 50
frame set at the first detector position was recollected at the end of data
collection check for crystal decay. When collecting a full sphere of data,
another detector position with (b = 270 was added. A total of 606 frames

were collected for all detector positions.

C. Data integration.

Post data collection cell parameters and a second orientation matrix
were obtained by refining reflections from ~ 30-50 frames in six to eight
groups, which were evenly distributed among all data frames. Diffraction
was carefully checked against locations predicted by the new orientation
matrix for signs of mosaic spread, twinning, split crystals or crystals that
moved during data collection. The new orientation matrix was used in data
integration. All frames were then integrated with the new orientation matrix
with appropriate global and local constraints according to the Laue
symmetry. Final cell parameters and the orientation matrix were obtained

from the XYZ centroids of up to 8192 integrated reflections with I > 100(1).

D. Beam inhomogeneity, crystal decay and absorption correction.

The intensities of integrated reflections were corrected for beam
inhomogeneity, crystal decay and absorption with the program SADABS
according to the algorithm described by Blessing,2 which is applied when
high redundancy data such as CCD data are used.

1.

2.

276

References

SMART: Area-Detector Software Package, 1993. Siemens Industrial
Automation, Inc.: Madison, WI.
Blessing J. Acta Cryst. 1995,/151, 33 .

APPENDIX B

ATOM POSITIONS TABLES

277

278

Table 45. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) for [Moz(OzCCF3)4(TCNQ)°2m-
C8H10)]oo"[ M02(02CCF3)4‘2m-C8Hio)]oo (1)-

 

 

): y 2 U(13(1)
M0(1) 9719(1) 8628(1) 7660(1) 27(1)
MO(2) “516(1) 8673(1) 7671(1) 27(1)
MO(3) 5091(1) 5362(1) 5501(1) 28(1)
0(1) 9660(4) 6901(3) 8049(3) 35(1)
0(2) 1 1557(4) 6965(3) 8053(3) 33(1)
0(3) 1 1590(4) 10387(3) 7270(3) 36(1)
0(4) 9692(4) 10344(3) 7258(3) 34(1)
0(5) 9101(3) 8605(3) 9055(3) 32(1)
0(6) 10997(4) 8625(3) 9080(3) 32( 1)
0(7) 12121(4) 8728(4) 6278(3) 34(1)
0(8) 10212(4) 8675(4) 6270(3) 34(1)
0(9) 6930(4) 5101(3) 5217(3) 32( 1)
0( 10) 5404(4) 6913(3) 4555(3) 33(1)
0(11) 3261(4) 5664(3) 5842(3) 32(1)
0(12) 4793(4) 3843(3) 6512(3) 33(1)
N( 1) 17436(5) 8542(5) 7787(4) 44(2)
N(2) 13631(5) 8881(5) 7883(4) 37(1)
C(7) 1 1304(6) 8729(5) 5887(5) 34(2)
C(9) 5373(5) 6965(5) 3761(4) 32( 1)
C( 1 1) 2669(5) 5380(5) 5381(4) 32(2)
C(13) 16493(6) 8763(5) 8122(4) 31(1)
C(14) 15347(5) 9079(5) 8560(4) 26(1)
C(15) 14382(5) 8954(5) 8181(4) 28(1)
C(17) 16138(5) 9646(5) 9644(4) 32(2)
C(18) 15977(5) 101 18(5) 10319(4) 32(2)
C(16) 15168(5) 9519(5) 9272(4) 28(1)
C( l) 10603(6) 6466(5) 8154(4) 32( 1)
C(3) 10664(6) 10837(5) 7 127(4) 32(2)
C(5) 9895(6) 8587(5) 9449(4) 31(1)

279

 

 

Table 45. (cont’d).

x y z U(eq)
c0) 10604(5) 5234(5) 8441(4) 45(2)
F(1A) 1 1520(9) 4882(6) 7861(6) 36(3)
F(2A) 9610(9) 4754(8) 8470(13) 56(3)
F(3A) 1081604) 4867(8) 9255(6) 46(3)
F03) 9489(7) 483201) 892102) 47(5)
F03) 1125507) 488804) 902502) 62(6)
F(3B) 1095700) 481803) 7747(8) 70(6)
F(1C) 994508) 479202) 804205) 27(6)
FOC) 1169800) 4873(22) 8174(23) 82(11)
F(3C) 10177(29) 477909) 9338(6) 7600)
C(4) 10756(5) 12060(5) 6743(4) 45(2)
F(4A) 9853(27) 12462(26) 6405(30) 5503)
F(5A) 11754(25) 12398(33) 6067(23) 94(19)
F(6A) 1075205) 1253208) 739704) 20(9)
F(43) 1171804) 1251301) 684505) 39(5)
303) 9811(13) 1252903) 715005) 55(6)
F(6B) 1089403) 1237203) 5845(6) 47(6)
F(4C) 1133506) 12528(6) 716400) 51(3)
F(5C) 9676(5) 12469(6) 681401) 46(3)
F(6C) 1133503) 123100) 5855(4) 51(3)
C(6) 9484(4) 8472(4) 1040(4) 39(2)
F(7A) 8353(6) 8676(9) 10768(7) 56(3)
F(8A) 10158(8) 9097(7) 107150) 37(3)
F(9A) 9659(8) 7443(5) 10895(6) 46(2)
F(7B) 10311(8) 8847(9) 10760(8) 56(4)
F(83) 9171(9) 7449(5) 10959(6) 55(3)
F(93) 8508(6) 9044(7) 10672(6) 42(2)
C(10) 5506(5) 8076(5) 3077(4) 49(2)
F(10A) 6206(8) 8069(9) 2256(5) 51(3)
F(11A) 6023(9) 8789(7) 3355(6) 55(3)
F(12A) 4457(7) 8430(9) 2987(8) 64(3)
F003) 5557(9) 8894(7) 3433(6) 55(3)

280

 

 

Table 45. (cont’d).

x y 2 U(°‘1)
F(1 13) 4550(7) 8236(8) 2744(7) 55(3)
F023) 6461(8) 8160(10) 2369(6) 60(4)
C(12A) 1371(7) 5673(8) 5495(7) 34(5)
F(l3A) 108300) 6397(8) 5973(7) 42(4)
F(l4A) 682(8) 4772(7) 5902(7) 45(3)
F( 15A) 1 162(9) 6143(8) 4697(6) 50(3)
C023) 1328(7) 5555(8) 5676(7) 33(5)
F033) 924(8) 5894(8) 4941(6) 41(2)
F043) 104200) 6256(8) 6181(7) 41(4)
F053) 697(8) 4646(6) 6206(6) 39(2)
C0) 1 1724(5) 8856(4) 4862(5) 50(2)
F(16A) 0251(9) 9809(6) 4365(7) 66(3)
F(l7A) 107990) 8687(9) 4570(7) 56(3)
F(18A) 125330) 8111(7) 4691(6) 48(2)
F(l6B) 10977(8) 8370(8) 4582(8) 57(3)
F073) 128040) 8500(9) 4587(8) 70(3)
F083) 1064(9) 9931(5) 4450(6) 54(3)
C(19A) 6880(11) 639703) 670401) 31(4)
C(20A) 661904) 7420(13) 621802) 32(5)
C(21A) 542904) 766003) 629102) 26(6)
C(22A) 452804) 686603) 6830(11) 28(5)
C(23A) 477606) 585603) 732202) 35(5)
C(24A) 596404) 5630(14) 724901) 32(4)
C(25A) 813902) 604504) 667502) 46(4)
C(26A) 5280(21) 877805) 570606) 44(6)
C093) 669704) 613603) 693702) 34(4)
C003) 667605) 716505) 637703) 36(5)
C013) 567405) 768805) 619904) 38(7)
C023) 4597( 15) 7099( 12) 66430 1) 29(5)
C033) 458603) 6069(14) 72140 1) 27(4)
C043) 560805) 558304) 737802) 37(5)
C053) 788605) 569307) 701804) 64(5)

281

 

 

Table 45. (cont’d).

x y 2 U(O‘I)
C063) 556605) 880409) 558800) 7400)
C(27A) 556403) 658306) 1007003) 37(6)
C(28A) 488306) 694908) 1080405) 44(7)
C(29A) 366308) 715200) 1094404) 50(7)
C(30A) 308408) 700707) 1032403) 56(6)
C(31A) 371207) 661606) 960205) 52(5)
C(32A) 493409) 641906) 947503) 43(6)
C(33A) 686104) 637506) 997404) 62(5)
C(34A) 293508) 751808) 1174800) 11603)
C073) 549905) 667608) 1026904) 46(7)
C083) 463009) 703109) 1090205) 49(7)
C093) 347607) 710100) 1084405) 51(7)
C(30B) 324107) 682105) 1013102) 41(5)
C(313) 408608) 651305) 946702) 46(5)
C(323) 522806) 641206) 954704) 47(6)
C(333) 677905) 6530(16) 1029304) 61(5)
C(343) 262509) 745309) 1161105) 64(6)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

282

 

 

Table 46. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (A2x 103) for [M02(OzCCF3)4(DM-
DCNQ1)°Cch]e (2)-

x y z U(eq)
M0(l) 433(1) 576(1) 283(1) 19(1)
0(1) 257(4) 4497(5) 559(6) 30(2)
0(2) 4 153(4) -256(5) 4(6) 26(2)
0(3) -338(4) 4164(4) 4562(6) 24(1)
0(4) 4267(4) 51(4) 2174(6) 23(1)
C0) -905(6) 4142(7) 315(8) 21(2)
C0) 4436(6) 4810(6) 437(8) 25(2)
F0) 432502) 2694(12) 24902) 45(6)
F0) 420705) 475200) 156104) 55(7)
F(3) 2240(11) 460101) -398(23) 80(11)
F(l) -980(l7) 2064(24) 164706) 70(9)
F0) 2120(13) 4401(20) 7205) 67(9)
F(3') 465104) 2647(15) 49102) 61(9)
C(3) 4042(6) 217(7) 2366(9) 22(2)
C(4) 4707(6) 4209(7) -3680(9) 28(2)
F(4) 2102(4) 4903(4) -3589(5) 34(1)
F(5) 4332(4) 4609(4) 4076(5) 35(1)
F(6) 2277(3) -583(4) 4545(5) 34(1)
N(l) 4390(5) 2055(6) 4283(7) 24(2)
C(5) 4348(5) 2879(6) 4327(8) 16(2)
N(2) 4403(5) 3861(5) 4480(7) 21(2)
C(6) 205(5) 4389(6) 240(8) 16(2)
C0) 131(6) 3981(7) 218(8) 22(2)
C(8) 833(6) 4567(7) 963(8) 22(2)
C(9) 1705(6) 4179(7) 1948(9) 26(2)
C08) 0 34850 1) 2500 29(3)
C(28) -748(7) 3990(8) 3364(9) 32(2)
C(38) 243(6) 4999(8) 3365(9) 30(2)
C(48) o -5498(12) 2500 39(4)

283

Table 47. Atomic coordinates ( x 104) and equivalent
displacement parameters (Azx 103)
{[M02(02C(CH3)3)3((NC)2C(CN)CCONH)]'6CH2C12}~ (3)-

 

 

x y 2 U(eq)
MO(1) 4045(2) 2387(2) 9291(9) 16(2)
M0(2) 4231(2) 2908(2) 9303(9) 17(2)
M0(3) 2600(2) 936(2) 9152(9) 8(2)
M0(4) 2417(2) 419(2) 9145(9) 16(2)
M0(5) 4647(2) 893(2) 9438(9) 15(2)
M0(6) 4985(2) 708(2) 9391(9) 16(2)
0(1) 4526(14) 2448( 14) 9315(28) 7(13)
0(2) 4708(15) 3007(15) 9410(30) 1 1(14)
0(3) 3762(23) 2821(23) 9229(43) 28(25)
N(7A) 3762(23) 2821(23) 9229(43) 28(25)
0(4) 4100(25) 2379(24) 8420(44) 42(25)
0(5) 4296(21) 2922(20) 8380(37) 29(20)
0(6) 4020( 16) 2376(16) 1024 1 (30) 8(15)
0(7) 4225(17) 2915(17) 10251(32) 16(17)
0(8) 2131(15) 863(15) 9044(30) 13(15)
0(9) 1907(14) 322(14) 9086(28) 6(13)
0( 10) 2879(15) 495(16) 9217(29) 0(16)
N(3A) 2879(15) 495(16) 9217(29) 0(16)
0(11) 2583(17) 944(16) 8254(32) 11(17)
0(12) 2381(16) 410(16) 8291(30) 11(15)
0(13) 2554(15) 931(14) 10110(27) 0(14)
0(14) 2347(15) 398(14) 10092(27) 0(12)
0(15) 5038(14) 1376(14) 9430(28) 3(12)
0(16) 5379(16) 1 193(16) 9368(30) 12(15)
0(17) 4623(15) 233(15) 9406(31) 0(16)
N(9A) 4623(15) 233(15) 9406(31) 0(16)
0( 18) 4642(20) 938(20) 8557(37) 27(20)
0(19) 5006(20) 742(21) 8528(38) 29(21)
0(20) 4674( 18) 921(19) 10400(37) 23(19)

isotropic

for

284

 

 

Table 47 . (cont’ d).
x y z U(eq)

O(2l) 5028(19) 730(18) 10298(36) 18(18)
N(l) 3787(16) 1767(16) 9191(31) 0(14)
N(2) 4315(18) 1214(18) 9408(36) 9(17)
N(3) 3070(15) 1049(16) 9258(29) 6(16)
0( 10A) 3070(15) 1049(16) 9258(29) 6(16)
N(4) 3572(21) 404(21) 9522(40) 23(19)
N(S) 2859(17) 1563(17) 9158(32) 3(15)
N(6) 2339(25) 2110(25) 9156(48) 31(24)
N(7) 3580(17) 2283(17) 9263(32) 13(17)
O(3A) 3580(17) 2283(17) 9263(32) 13(17)
N(8) 3030(23) 2910(23) 9088(45) 26(22)
N(9) 4262(22) 420(22) 9490(42) 38(25)
0( 17A) 4262(22) 420(22) 9490(42) 38(25)
N(lO) 3470(19) -282( 19) 9587(36) 14(17)
N(l 1) 4600(24) -473(24) 9273(43) 26(23)
N(l2) 3552(20) ~1001(l9) 9374(38) 13(18)
C(l) 3753(20) 1495(20) 9297(37) 0(17)
C(2) 3746(27) 1218(26) 9272(50) 23(25)
C(3) 4089(24) 1228(24) 9331(46) 12(22)
C(4) 3457(25) 868(25) 936 l (48) 19(23)
C(5) 3161(20) 826(19) 9195(37) 0(17)
C(6) 3543(27) 615(27) 9349(52) 24(25)
C(7) 2867(29) 1786(29) 9213(54) 26(28)
C(8) 2874(20) 2125(20) 9139(40) 2(17)
C(9) 2591(19) 2106(19) 9115(39) 0(17)
C(10) 3177(25) 2442(25) 9143(50) 16(23)
C(11) 3535(22) 2526(22) 9175(43) 9(19)
C(12) 3125(24) 2703(24) 9077(47) 15(22)
C(13) 4304(34) 169(33) 9427(65) 38(33)
C(14) 4062(20) -l64(20) 9470(38) 2(17)
C(15) 3732(26) -214(25) 9530(48) 18(24)
C(16) 4095(20) 426(20) 9374(39) 0(17)

285

 

 

Table 47 . (cont’d).
x y z U(eq)

C(17) 437801) 44001) 9378(42) 508)
C(18) 376907) 26807) 9373(51) 2004)
C(19) 476509) 2764(32) 9320(59) 34(32)
C(20) 509805) 280806) 9414(32) 1009)
C(21) 536701) 314606) 9562(43) 2304)
C(22) 5132(48) 2739(31) 8787(34) 169035)
C(23) 5130(32) 256601) 9791(54) 61(40)
C(24) 422505) 265305) 811009) 1703)
C(25) 430507) 267406) 7475(30) 69(51)
C(26) 458400) 261 1(22) 7400(53) 36(30)
C(27) 439300) 299406) 7190(39) 1201)
C(28) 400909) 240609) 7176(53) 39(32)
C(29) 416609) 264509) 1052706) 0(17)
C(30) 423809) 265309) 1116508) 3109)
C(31) 425309) 294301) 11465(61) 58(42)
C(32) 398504) 234700) 11459(56) 51(36)
C(33) 456502) 2674(32) 1121903) 83(59)
C(34) 188405) 57509) 9106(38) 0(17)
C(35) 155207) 53109) 9000(33) 35(30)
C(36) 1282(41) 20200) 9194(57) 129(98)
C(37) 1495(32) 58208) 8375(31) 48(35)
C(38) 154307) 790(23) 9366(39) 3208)
C(39) 2510(52) 619(52) 7921(47) 89(62)
C(40) 247102) 67201) 7289(42) 41(33)
C(41) 273808) 631(37) 7020017) 172044)
C(42) 2491(35) 97803) 705206) 73(51)
C(43) 214703) 38504) 7122(69) 62(42)
C(44) 2413(43) 649(45) 1032607) 63(48)
C(45) 233905) 66805) 1095609) 14(21)
C(46) 259800) 95300) 1 1292(53) 4505)
C(47) 227805) 357(19) 1 1247(60) 58(42)
C(48) 2030(17) 68401) 10966(44) 20(23)

286

 

 

Table 47 . (cont’d).

x y z U(eq)
C(49) 535206) 144000) 937501) 4908)
C(50) 563705) 178105) 938109) 1702)
C(51) 554505) 202504) 9169(52) 0(33)
C(52) 590407) 1800(32) 9001(46) 18(42)
C(53) 576208) 1866(40) 999502) 38(57)
C(51) 5570(38) 205609) 9493(95) 0(52)
C(52) 5648(86) 173805) 873505) 186010)
C(53) 5963(45) 1870(57) 9638033) 51036)
C(54) 484904) 87503) 8230(27) 1101)
C(55) 488907) 91807) 758408) 2807)
C(56) 486103) 62609) 7270(43) 2103)
C(57) 522609) 120401) 748601) 81(59)
C(58) 464403) 99804) 7334(51) 3400)
C(59) 489302) 86102) 1061801) 7(20)
C(60) 496703) 94103) 1 1251(33) 3008)
C(61) 465702) 861(46) 11565(97) 148014)
C(62) 5214(38) 129403) 11358009) 167032)
C(63) 509507) 73205) 1149203) 68(49)
Cl(l) 4328(8) 167901) 813802) 6402)
C( 1S) 4715(1 1) 174609) 834203) 3802)
C10) 5018(7) 1969(9) 781708) 48(9)
C10) 2262(8) 1516(9) 806100) 47(9)
C(28) 195102) 161708) 8021(69) 59(50)
CI(4) 1617(8) 1340(10) 759401) 54(10)
0(5) 3382(9) 6470 1) 565907) 59(14)
C(38) 296402) 47202) 5824(81) 68(63)
CI(6) 2718(8) 110(9) 543300) 3200)
C10) 1604(8) 136701) 1064702) 5701)
C(58) 192705) 159800) 10160(45) 43(35)
C100) 2276(9) 156802) 1030604) 6602)
C10 1) 1520(10) 349100) 582902) 60(11)
C(68) 165402) 328803) 535803) 2205)

287

 

 

Table 47. (cont’d).

x y z U(eq)
C102) 188302) 3140(11) 5720(23) 6702)
C103) 476805) 13605) 590109) 9508)
C(78) 5000 o 5518(44) 40(44)
C105) 6939(8) 1967(9) 591108) 45(9)
C(88) 701300) 172408) 542405) 1401)
Cl(l6) 7420(7) 1807(8) 544907) 37(8)
C107) 5954(8) 1059(8) 821409) 5200)
C(98) 634307) 140005) 8335(47) 4406)
C108) 6427(9) 1729(8) 786809) 52(10)
C109) 5920(9) 1068(9) 1040401) 60(11)
C008) 633101) 138506) 1041706) 2104)
C100) 640400) 1681(8) 1095200) 5901)
C103) 323504) 28304) 5597(55) 61(18)
C028) 3340(56) 248(50) 6318(54) 0(61)
C104) 3460(56) -48(52) 6395(83) 10503)
C105) 353903) -586(9) 802906) 8806)
C038) 3843(44) 23604) 7669(53) 10104)
C106) 3973(12) 124(9) 807505) 8204)
C109) 3994(9) 84801) 1077603) 6802)
C058) 368901) 94906) 10915052) 166031)
C100) 3292(9) 61603) 1081005) 80(14)
C103) 349301) 143102) 775703) 7203)
C078) 338507) 169205) 812107) 3401)
c104) 316401) 181601) 767703) 70(13)
C105) 3409(9) 70502) 1266104) 8104)
C088) 302902) 548(38) 1230703) 81(57)
C106) 2704(8) 250(10) 1272401) 6101)
C107) 535800) 22100) 779801) 5801)
C098) 5000 0 821407) 46(50)
C109) 547107) 253406) 641609) 11100)
C008) 543403) 2774(63) 6965002) 80(92)
C100) 5603035) 3062016) 7533078) 644048)

 

 

288
Table 47 . (cont’d).
x y 2 Wall

C101) 4370(9) 575(9) 588800) 6100)
C018) 392504) 377(45) 5797088) 289063)
Cl(42) 37570 1) 4102) 605009) 10608)
Cl(43) 368808) 75109) 7850(34) 4907)
Cl(44) 333701) 18801) 7170(38) 59(21)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

 

289

Table 48. Atomic coordinates ( x 104) and equivalent isotropic

displacement parameters (Azx 103) for
1[M02(02C(CH3)3)3((NC)2C(CN)CCONH)1'O~5C6H6loo (4)-

 

 

X Y 2 1105(1)
Mo(l) 1647(1) 6639(1) 12085(1) 36(1)
Mo(2) 1795(1) 8216(1) 11523(1) 36(1)
0(3) 1135(4) 8313(9) 10846(5) 43(3)
0(2) 968(3) 6671(9) l 1421(5) 38(3)
0(6) 1786(3) 5572(9) 1 1222(5) 40(3)
0(4) 1589(4) 9332(9) 12319(6) 43(3)
0(7) 1924(3) 7187(9) 10623(5) 40(3)
0(5) 1436(3) 7651(9) 12903(5) 35(3)
C( 1) 2590(8) 7349(18) 12659(12) 64(6)
0(1) 2298(5) 6554(14) 12750(6) 74(6)
N( 1) 2445(4) 8207(12) 12156(7) 46(4)
C(2) 3023(7) 6992(19) 13215(12) 37(7)
C(3) 3403(8) 7710(20) 13206(13) 38(7)
N ( 1A) 2298(5) 6554(14) 12750(6) 74(6)
0( 1A) 2445(4) 8207(12) 12156(7) 46(4)
C(2A) 3076(8) 7720(30) 12909( 17) 7(9)
C(3A) 3330(14) 7010(30) 13542( 19) 21(11)
N(4) 4167(5) 7153(17) 14102(9) 101(7)
N(3) 3451(4) 9528(12) 12414(7) 46(4)
N(2) 3078(4) 5193(13) 14l30(7) 49(4)
C(4) 3810(7) 7337(15) 13754( 10) 55(5)
C(5) 3098(6) 6032(16) l3778(10) 46(5)
C(6) 3359(5) 8682(14) 12698(9) 37(4)
C(7) 858(6) 7517(14) 10939(9) 43(5)
C(8) 372(6) 7584(12) 10421(9) 58(5)
C(9) 211(8) 6291(19) 10182(16) 92(11)
C(10) 102(7) 8347(19) 10855(12) 61(8)
C(11) 383(8) 8130(30) 9634(12) 95(11)
C(9A) 68(10) 6830(60) 10790(30) 60(30)

290

 

 

Table 48. (cont’d).

x y z U(eq)
C(10A) 311(13) 8810(30) 1000000) 40(20)
C(1 1A) 354(13) 6950(60) 9655(19) 40(20)
C(12) 1871(6) 605705) 106660) 40(4)
C(13) 1913(6) 531604) 9961(9) 48(5)
C(14) 1413(7) 494707) 953300) 69(6)
C(15) 2190(6) 414906) 1026300) 63(5)
C(16) 2129(5) 602903) 9437(8) 38(4)
C(17) 1433(5) 877704) 12822(8) 32(4)
C(18) 1221(6) 953204) 133420) 44(4)
C(19) 1203(6) 882704) 140800) 50(5)
C(20) 1508(7) 1074000) 1360702) 88(7)
C(21) 729(6) 980105) 1284400) 63(5)
C(lS) 0 5600(50) 2500 190(20)
C(28) 42200) 4770(30) 244605) 13800)
C(38) 383(8) 3540(20) 247103) 99(8)
C(48) 0 2930(30) 2500 9100)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

291

Table 49. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) for [Re2C14(dppm)2]2(TCNQ) (5).

 

 

x y 2 WW)
Re(1) 5380(1) 3931(1) 2800(1) 28(1)
Re(2) 6786(1) 4190(1) 3253(1) 28(1)
P( l) 3936(2) 5481(1) 2880(1) 34( 1)
Cl(l) 3883(2) 3605(1) 2237(1) 39(1)
Cl(3) 6391(2) 3934(1) 4307(1) 41(1)
Cl(2) 4316(1) 3201(1) 3699(1) 36( 1)
P(3) 6487(2) 2288(1) 2639(1) 34(1)
P(2) 5973(2) 5956(1) 3214(1) 32( 1)
CI(4) 8237(2) 4573(1) 2487(1) 41(1)
P(4) 7965(2) 2491(1) 3480(1) 34(1)
N( 1) 5934(5) 4583(4) 1934(3) 35(1)
N(2) 7666(8) 6542(7) 431(4) 94(3)
C( 1) 6123(6) 4899(6) 1484(4) 41(2)
C(2) 6268(7) 5356(6) 829(3) 51(2)
C(3) 7053(8) 6008(7) 619(3) 61(2)
C(4) 5629(7) 5178(5) 407(3) 45(2)
C(5) 4861(7) 4511(5) 61 1(3) 48(2)
C(6) 4262(7) 4346(6) 213(3) 53(2)
C(7) 4438(6) 6002(5) 3397(3) 36(2)
C(8) 7964(6) 2074(5) 2807(3) 39(2)
C(9) 2514(6) 5348(5) 3214(3) 41(2)
C(10) 2187(7) 5298(6) 3817(4) 56(2)
C(l l) 1097(8) 5181(7) 4067(4) 69(3)
C(12) 324(8) 51 18(7) 3703(5) 69(3)
C(13) 629(7) 5173(6) 3111(5) 68(3)
C(14) 1720(7) 5281(6) 2872(4) 57(2)
C(15) 3626(6) 6485(5) 2186(3) 39(2)
C(16) 3354(7) 7501(5) 2189(4) 56(2)
C(17) 3047(9) 8238(6) 1659(4) 73(3)
C(18) 3037( 10) 7997(7) 1 134(4) 88(3)

292

 

 

Table 49. (cont’d).

X y 2 U(eq)
C(19) 3295(8) 6998(6) 1130(4) 64(2)
C(20) 3598(6) 6250(5) 1654(3) 45(2)
C(21) 6077(6) 7050(5) 2570(3) 38(2)
C(22) 5665(7) 8014(5) 2656(3) 51(2)
C(23) 5742(8) 8846(6) 2188(4) 64(3)
C(24) 6214(9) 8761(6) 1634(4) 71(3)
C(25) 6614(8) 7818(6) 1530(3) 63(3)
C(26) 6549(7) 6948(5) 2005(3) 45(2)
C(27) 6423(6) 6331(5) 3817(3) 33(2)
C(28) 5730(7) 6446(5) 4322(3) 45(2)
C(29) 6131(8) 6699(6) 4780(3) 56(2)
C(30) 7223(8) 6833(6) 4718(3) 53(2)
C01 ) 7922(7) 6711(6) 4215(3) 50(2)
C(32) 7520(6) 6466(5) 3762(3) 42(2)
C(33) 6634(7) 2210(5) 1858(3) 42(2)
C(34) 7429(8) 2651(7) 1437(3) 63(2)
C(35) 752200) 2578(9) 843(4) 89(4)
C(36) 682902) 2074(9) 680(4) 97(4)
C(37) 604502) 1676(6) 1084(4) 88(4)
C(38) 5932(9) 1744(6) 1670(3) 63(3)
C(39) 5977(7) 1161(5) 3068(3) 37(2)
C(40) 6741(7) 240(5) 3273(3) 48(2)
C(41) 6315(9) -607(6) 3599(4) 64(3)
C(42) 517100) -549(6) 3719(4) 67(3)
C(43) 4416(8) 350(6) 3514(4) 63(2)
C(44) 4820(7) 1211(5) 3191(3) 47(2)
C(45) 7656(6) 1442(4) 4115(3) 35(2)
C(46) 6566(6) 1461(5) 4387(3) 38(2)
C(47) 6308(7) 619(5) 4829(3) 48(2)
C(48) 7182(8) 225(6) 5005(4) 62(2)
C(49) 8262(8) 243(6) 4740(4) 68(3)
C(50) 8515(7) 586(6) 4298(3) 54(2)

293

 

 

Table 49. (cont’d).

x y 2 U(eq)
C(51) 9440(6) 2492(5) 3572(4) 44(2)
C(52) 103180) 2425(6) 3138(4) 59(2)
C(53) 1 1404(8) 2460(7) 3246(6) 82(3)
C(54) 1 1588(8) 2592(8) 3782(6) 88(4)
C(55) 107300) 2647(7) 4209(5) 83(3)
C(56) 9620(7) 2626(6) 4114(4) 57(2)
0(1) 2962(6) 8171(5) 3618(3) 87(2)
C(57) 238500) 7928(8) 4197(4) 102(4)
C(58) 1319(9) 8616(9) 4191(5) 104(4)
C(59) 143200) 9473(8) 3660(5) 99(4)
C(60) 230400) 9036(8) 3257(4) 89(3)
0(2) 892206) 3660(20) 421(7) 25701)
C(61) 9740(20) 3120(14) 26102) 25304)
C(62) 980109) 381309) 4334(7) 19801)
C(63) 9380(20) 481305) 4277(9) 20101)
C(64) 8950(30) 467608) ~686(l3) 330(20)
0(3) 1239000) 2290(18) 1331(11) 33803)
C(65) 1264209) 129507) 174902) 23703)
C(66) 1 1640(20) 98402) 1800(12) 24704)
C(67) 1072904) 186809) 167500) 189(8)
C(68) 1123000) 2480(30) 118903) 460(30)
0(4) 9417(8) 9522(8) 2622(5) 124(3)
C(69) 907407) 97390 1) 2041(7) 156(7)
C(70) 938608) 8850(17) 1855(8) 179(8)
C(71) 9310(20) 8090(13) 240300) 21802)
C(72) 944905) 847303) 2860(8) 150(6)
0(5) 6090(20) 35306) 109(8) 129(8)
C(73) 6530(30) 35000) 2305) 16107)
C(74) 5570(50) -860(40) 44000) 320(50)
C(75) 4560(30) 410(30) 3000) 250(40)
C(76) 4980(30) 750(30) -9705) 14508)
0(6) 1056000) 39000) 24000) 400(60)

294

 

 

Table 49. (cont’d).

x y z U(eq)
C(77) 9530(40) 230(40) -9000) 26000)
C(78) 9860(50) 1060(40) 2000) 430(60)
C(79) 1105000) 760(50) 110(30) 310(40)
C(80) 1147000) 40(40) 20000) 240(30)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

295

Table 50. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) for [RezCl4(dppm)2]2(DMDCNQI) (6).

 

 

x y z U(eq)
Re(l) 320(1) 3863(1) 2812(1) 31(1)
Re(2) 1712(1) 4137(1) 3286(1) 30(1)
C10) 4187(2) 3509(2) 2233(1) 46(1)
C10) 246(2) 3164(2) 3690(1) 40(1)
C10) 3171(2) 4479(2) 2548(1) 45(1)
C10) 1286(2) 3950(2) 4333(1) 43(1)
P(4) 4138(2) 5421(2) 2883(1) 38(1)
P(3) 896(2) 5906(2) 3221(1) 35(1)
P(2) 2903(2) 2452(2) 3531(1) 39(1)
P(l) 1458(2) 2229(2) 2650(1) 41(1)
N(l) 858(6) 4484(5) 1971(3) 39(2)
N(2) 1162(7) 5209(6) 911(3) 51(2)
C0) 2936(8) 201 1(7) 2841(4) 45(2)
C(2) -631(7) 5982(6) 3391(4) 36(2)
C(3) 965(8) 4794(7) 1477(5) 43(2)
C(4) 584(9) 5099(7) 456(4) 54(3)
C(5) 48300) 4449(8) 577(4) 55(3)
C(6) -776(9) 4362(8) 132(4) 51(3)
C(7) 458802) 367700) 253(5) 94(5)
C(8) 943(9) 1095(7) 3058(4) 48(2)
C(9) 48700) 1152(8) 3176(4) 56(3)
C(10) 38302) 308(9) 3482(5) 71(3)
C(11) 167(15) 358(9) 3681(5) 83(4)
C(12) 130203) -644(8) 3574(6) 78(4)
C(13) 171201) 188(7) 3262(5) 62(3)
C(14) 163500) 2148(7) 1870(5) 58(3)
C(15) 990(12) 1649(9) 1657(5) 82(4)
C(16) 110107) 161400) 1068(7) 113(6)
C(17) 187607) 2060(12) 678(6) 112(6)
C(18) 250903) 25760 1) 872(6) 99(5)

296

 

 

Table 50. (cont’d).

x y z U(eq)
C(19) 238400) 2602(9) 1462(5) 71(3)
C(20) 4342(8) 2458(7) 3655(5) 49(3)
C(21) 5253(9) 2364(8) 3239(5) 64(3)
C(22) 6346(9) 241 1(9) 3363(6) 73(4)
C(23) 647900) 2598(9) 3896(7) 77(4)
C(24) 55980 1) 2699(9) 4320(6) 73(4)
C(25) 4503(8) 2640(7) 4194(5) 53(3)
C(26) 2612(8) 1431(7) 4151(4) 40(2)
C(27) 1511(8) 1462(7) 4410(4) 41(2)
C(28) 1271(9) 640(8) 4833(4) 53(3)
C(29) 213300) 495(7) 5016(5) 60(3)
C(30) 322201) 240(8) 4769(5) 79(4)
C(31) 3462(9) 568(7) 4343(5) 59(3)
C(32) 1374(8) 6335(6) 3802(4) 37(2)
C(33) 715(8) 6470(7) 4321(4) 45(2)
C(34) 1134(9) 6756(8) 4761(5) 60(3)
C(35) 222400) 6880(8) 4696(5) 61(3)
C(36) 2899(9) 6734(7) 4196(5) 57(3)
C(37) 2470(8) 6449(7) 3744(4) 44(2)
C(38) 998(8) 6924(6) 2557(4) 38(2)
C(39) 1370(8) 6761(7) 2005(4) 49(3)
C(40) 14060 1) 7563(9) 1508(5) 77(4)
C(41) 106302) 8538(9) 1572(5) 83(4)
C(42) 68202) 8690(9) 2125(6) 85(4)
C(43) 65300) 7908(7) 2622(5) 62(3)
C(44) 4473(8) 6393(7) 2195(4) 42(2)
C(45) 4472(8) 6139(8) 1653(4) 54(3)
C(46) - 1756( 10) 6868(8) 1 128(5) 70(3)
C(47) 20560 1) 7840(8) 1 136(5) 74(4)
C(48) 2016(12) 81 14(9) 1667(6) 88(4)
C(49) 4730(9) 7396(8) 2189(4) 58(3)
C(50) 2554(8) 5311(7) 3218(5) 49(3)

297

 

 

Table 50. (cont’d).

x y z U(eq)
C(51) 3357(9) 5232(8) 2871(6) 67(3)
C(52) 4427(11) 5120(9) 3117(8) 85(4)
C(53) 474201) 5080(9) 3704(7) 79(4)
C(54) 3931(1 1) 5164(9) 4069(6) 83(4)
C(55) -2851(9) 5264(8) 3817(5) 66(3)
0(1) 2162(9) 8149(7) 3599(5) 115(4)
C(56) 279904) 901202) 3245(7) 122(6)
C(57) 367804) 9470(11) 3650(7) 138(7)
C(58) 377302) 862303) 4165(6) 124(6)
C(59) 267605) 795302) 4181(6) 135(7)
0(2) 436302) 344(12) 2588(7) 165(6)
C(60) 3960(20) 35105) 2009(9) 18200)
C(61) 4320(20) 421509) 175700) 23704)
C(62) 4270(20) 2021(15) 22880 1) 21502)
C(63) 4390(20) 467506) 2790(9) 19501)
0(3) 384000) 5860(40) 880(20) 790(40)
C(69) 3110(50) 6460(20) 53408) 540(30)
C(70) 398000) 5860(40) 700(20) 680(50)
C(71) 4290(30) 5170(30) 125805) 450(30)
C(72) 3160(50) 4860(30) 102000) 540(40)
0(4) 272000) 2080(30) 161409) 640(30)
C(73) 380000) 2450(20) 135709) 560(40)
C(74) 439000) 1700(30) 170805) 33108)
C(75) 342000) 930(30) 201408) 520(40)
C(76) 246000) 1060(30) 159607) 430(30)
0(5) 1250(30) 80(30) 50(20) 240(20)
C(77) 60(40) 48000) 90(20) 14108)
C(78) 440(40) 370(50) 190(30) 220(30)
C(79) 530(50) 417000) 80(30) 300(40)
C(80) 1380(40) 380(40) -260(20) 210(30)
0(6) 3640(60) 1114000) 210(30) 700(60)
C(82) 343000) 1014000) 110(20) 360(30)

298

 

 

Table 50. (cont’d).

x y 2 U(O‘l)
C(83) 4390(80) 9810(40) 20(40) 400(60)
C(84) 330000) 1072000) 2000) 310(30)
C(85) 480000) 1151000) 30(30) 560(60)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

299

 

 

Table 51. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) for Cu(TCNQ) phase I (7).

x y z U(eq)
Cu(1) 7270(60) 2525(6) 342808) 64(3)
N(l) 5690000) 3850(30) 2650(30) 20(8)
C(8) 3810(40) 6030(20) 1910(20) 29(7)
C(5) 4150(50) 8950(20) 4940(20) 20(8)
C(9) 1860(30) 6720(30) 267707) 24(8)
C(10) 490(40) 6340(20) 379907) 18(9)
C(ll) 447000) 7100(30) 450907) 28(10)
C(12) 2200(40) 8290(30) 4150(30) 39(9)
C(13) 32000) 8670(20) 302000) 3601)
C(14) 1140(60) 7910(30) 2310(20) 1900)
C( 1) 4320(1 10) 4780(30) 2320(40) 12(8)
N(2) 4310(1 10) 8320(30) 7100(30) 20(9)
C(2) -5560( 150) 8550(80) 61 10(30) 22(1 1)
N(4) -4260(130) 1114000) 4100(40) 4303)
C(4) 3380000) 1019000) 4550(30) 17(9)
N(3) 4010000) 6650(40) 40000) 25(9)
C(3) 5640000) 6290(40) 760(30) 19(8)
Cu(lA) 2320(70) 2520(6) 343509) 88(4)
N(3A) 5680(120) 6580(40) 47000) 43(9)
C(8A) 5550(40) 5950(20) 1910(30) 32(7)
C(SA) 1320000) 8880(20) 4950(30) 20(7)
C(9A) 7410(30) 6660(20) 2645(19) 60(9)
C(lOA) 8630(40) 6280(20) 378509) 4900)
C(l 1A) 1053000) 7030(30) 4515(18) 7(8)
C(12A) 1133000) 8220(30) 4160(30) 41(8)
C(13A) 1011000) 8610(20) 3030(30) 30(10)
C(14A) 8210(60) 7860(30) 2290(20) 34(8)
N(2A) 13910000) 8310(40) 6980(40) 32(10)
C(2A) 15290060) 8540000) 602000) 3303)
C(3A) 4120000) 6350(40) 750(30) 24(8)

300

 

 

Table 51. (cont’d).

x y 2 U(GQ)
N(4A) 1423000) 1 1030(20) 4130(30) 0(6)
C(4A) 1446000) 1007000) 4620(30) 0(7)
N(lA) 3520(80) 3980(30) 2660(30) 27(7)
C(lA) 5730(90) 4690(30) 2330(30) 23(8)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

Table 52. Atomic coordinates

301

( x 10‘) and equivalent
displacement parameters (Azx 103) for Cu(TCNQ) phase II (8).

 

 

x y z U(eq)
Cu(1) -2500 l 1732(5) 7500 24(1)
N( 1) -348(23) 9759(19) 6916(5) 21(3)
C(l) 1292(26) 8940(24) 6636(6) 20(3)
N(2) 5431(22) 4041(20) 6886(6) 24(3)
C(2) 4421(27) 5743(26) 6604(7) 28(4)
C(3) 3215(26) 7839(23) 6257(6) 20(3)
C(4) 4204(24) 8933(23) 5628(6) 20(3)
C(5) 2925(27) l 1070(26) 5308(6) 30(3)
C(6) 3828(24) 12080(26) 4700(6) 27(3)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

isotropic

302

Table 53. Atomic coordinates ( x 104) and equivalent isotropic

 

 

displacement parameters (Azx 103) [Mn(TCNQ-
TCNQ)0,5(TCNQ)(MeOH)2].. (9).

x y z U(eq)
Mn(l) 5000 5514(1) 7500 16(1)
O(lM) 4838(2) 5532(1) 9150(2) 29(1)
C(lM) 4020(3) 5497(2) 9746(3) 40(1)
N(lA) 6092(2) 4965(1) 7533(2) 22(1)
C(2A) 6750(2) 4739(1) 7688(2) 17(1)
N(3A) 9000(2) 5028(1) 7898(2) 26(1)
C(4A) 8371(2) 4768(1) 7891(2) 21(1)
C(SA) 7569(2) 4478(1) 7882(2) 17(1)
C(6A) 7565(2) 3966(1) 8180(2) 16(1)
C(7A) 6738(2) 3701(1) 8247(2) 20(1)
C(8A) 6729(2) 3226(1) 8598(2) 22(1)
C(9A) 7547(3) 2996(1) 8885(2) 21(1)
C(lOA) 8381(2) 3246(1) 8774(2) 22(1)
C0 1A) 8391(2) 3726(1) 8429(2) 20(1)
C(12A) 7531(3) 2485(1) 9371(2) 22(1)
C(13A) 6708(3) 2207(1) 9029(3) 25(1)
N(14A) 6036(2) 2016(1) 8792(2) 36(1)
C(15A) 8384(2) 2209(1) 9130(3) 25(1)
N(16A) 9072(2) 2016(1) 8970(2) 34(1)
Mn(2) 10000 5591(1) 7500 18(1)
O(2M) 106200) 5631(1) 9023(2) 31(1)
C(2M’) 1 1564(6) 5552(4) 9295(7) 40(3)
C(2M") “330(8) 5358(5) 9492(8) 40(3)
N(lB) 6080(2) 6081(1) 7788(2) 24(1)
C(23) 6692(2) 6327(1) 8014(2) 19(1)
N(3B) 8994(2) 6146(1) 8073(2) 23(1)
C(43) 8327(2) 6367(1) 8176(2) 18(1)
C(53) 7469(2) 6616(1) 8272(2) 17(1)
C(63) 7405(2) 7122(1) 8529(2) 17(1)

303

 

 

Table 53. (cont’d).

x y 2 WW)
C(73) 6532(2) 7365(1) 8549(2) 18(1)
C(83) 6468(2) 7846(1) 8795(2) 19(1)
C(93) 7272(2) 8126(1) 9044(2) 19(1)
C003) 8140(2) 7884(1) 9014(2) 19(1)
C013) 8207(2) 7398(1) 8769(2) 19(1)
C023) 7208(2) 8631(1) 9309(2) 20(1)
C033) 6350(2) 8875(1) 9363(2) 22(1)
N(14B) 5629(2) 9058(1) 9416(2) 33(1)
C053) 8018(2) 8910(1) 9488(2) 22(1)
N063) 8699(2) 9125(1) 9628(2) 31(1)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

304

Table 54. Atomic coordinates ( x 10") and equivalent isotropic
displacement parameters (Azx 103) [Mn(TCNQ)2(HzO)2].. (10).

 

 

x y z U(eq)
Mn(l) o o 0 25(1)
N(l) 3647(2) 2886(1) 1856(2) 45(1)
C0) 3442(1) 2116(2) 1944(2) 29(1)
N(2) 1348(1) 758(1) 643(2) 34(1)
C(2) 3199(1) 1153(1) 2032(2) 24(1)
N(3) 7949(1) -l476(1 ) 6179(2) 37(1)
C(3) 2170(1) 951(1) 1239(2) 27(1)
N(4) 5617(1) 3597(1) 4980(2) 34(1)
C(4) 3925(1) 399(1) 2790(2) 22(1)
C(5) 4964(1) 61 1(1) 3699(2) 22(1)
C(6) 5667(1) 4 16(1) 4393(2) 22(1)
C(7) 5388(1) 4109(1) 4242(2) 21(1)
C(8) 4341(1) 4322(1) 3338(2) 24(1)
C(9) 3643(1) 398(1) 2641(2) 24(1)
C(10) 6120(1) 4858(1) 4954(2) 24(1)
C(11) 7137(1) -1662(1) 5649(2) 26(1)
C(12) 5858(1) 2828(1) 4958(2) 25(1)
0(1) 251(2) 174(1) 2200(2) 47(1)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

305

Table 55. Atomic coordinates ( x 10") and equivalent isotropic
displacement parameters (Azx 103) [Bu4N][TCNQ] (11).

 

 

x y z U(eq)
N( 1) 2488(2) —785( 1) 3067(1) 47(1)
N(2) 5106(2) -1088(1) 1026(1) 45(1)
N(3) -2346(2) 2280(1) - 104(1) 49(1)
N(4) 373(2) 1956(1) 2110(1) 42(1)
N(S) -1246(2) 1782(1) -5151(1) 25(1)
C( l) 2646(2) -664( l) 2357(2) 30(1)
C(2) 2888(2) -504( l) 1496(1) 26(1)
C(3) 41 14(2) -828(l) 1236(1) 30(1)
C(4) 2078(2) 15(1) 962(1) 23(1)
C(5) 869(2) 350(1) 1216(1) 28( 1)
C(6) 102(2) 852(1) 702(1) 29(1)
C(7) 478(2) 1066(1) - 103(1) 24(1)
C(8) 1707(2) 739(1) -352(1) 26(1)
C(9) 2468(2) 234(1) 158(1) 26( 1)
C( 10) -322(2) 1589(1) -637( 1) 28(1)
C(l l) - 1452(2) 1963(1) -352(l) 33(1)
C(12) 40(2) 1788(1) - 1455(2) 29(1)
C(13) -2423(2) 1229(1) -5286(1) 28(1)
C(14) -3979(2) 1464(1) -5644(1) 32(1)
C(15) -5021(2) 850(1) -5710(2) 38(1)
C(16) -6593(2) 1034(1) -6109(2) 45(1)
C(17) - 1533(2) 2272(1) -4427( 1) 27(1)
C(18) -386(2) 2827(1) ~4157( 1) 32(1)
C(19) -856(2) 3322(1) -3495( 1) 34(1)
C(20) 193(2) 3927(1) -3301(2) 46(1)
C(21) - 1243(2) 2200(1) -5993( l) 28( 1)
C(22) -1 181(2) 1787(1) -6826( 1) 39(1)
C(23) -1 109(2) 2277(1) -7598(2) 44(1)
C(24) 424(3) 2520(1) -7630(2) 62(1)

306

 

 

Table 55. (cont’d).

x y z U(eq)
C(25) 244(2) 1433(1) 4895(1) 28(1)
C(26) 497(2) 1023(1) 4038(1) 36(1)
C(27) 2065(2) 774(1) 3815(2) 47(1)
C(28) 2371(3) 368(1) 2954(2) 74(1)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

307

Table 56. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) [Bu4N][TCNQF4] (12).

 

 

x y z U(eq)
F0) 39(2) 7356(1) 505(1) 44(1)
F0) 9(3) 4718(2) 4384(1) 49(1)
F(3) 2288(3) 6229(2) 808(1) 50(1)
F(l) 2195(3) 3597(2) 4078(1) 52(1)
N(9) 5531(3) 107810) 1809(1) 30(1)
C(8) 380(4) 6421(3) 106(2) 32(1)
C(4) 2378(4) 4842(3) -118(2) 31(1)
C(7) 96(4) 61 12(3) 461(2) 29(1)
N(lO) 2792(3) 5975(2) 3268(2) 32(1)
C(5) 4702(4) 4531(3) 383(2) 32(1)
C(9) 4740(4) 5833(3) 258(2) 32(1)
C(12) 1963(5) 7733(3) 244(2) 37(1)
C(ll) 1886(5) 6428(3) 4226(2) 37(1)
C(6) 354(4) 5123(3) 3370) 33(1)
C(29) 4372(4) 104550) 2209(2) 32(1)
C(3) 4255(5) 4531(3) 614(2) 41(1)
C(10) 1302(4) 6727(3) 322(2) 32(1)
C(38) 3976(4) 115130) 1194(2) 37(1)
N(4) 2555(4) 8543(3) 19(2) 53(1)
C(25) 6455(4) l 1808(2) 2200(2) 33(1)
N(l) 4947(4) 2456(3) -685(2) 58(1)
C(2) -3564(4) 4215(3) 46(2) 34(1)
C(l) 4300(5) 3245(3) 368(2) 42(1)
C(34) 7792(4) 103550) 1363(2) 34(1)
C(37) 4857(4) 107970) 1150(2) 34(1)
C(41) 2916(4) 6162(3) 4001(2) 34(1)
C(30) 3271(4) 9470(3) 1891(2) 36(1)
C(33) 6435(4) 100620) 1668(2) 33(1)
C(51) -6489(4) 4048(3) 2377(2) 43(1)
N(2) 4854(4) 4742(3) 1064(2) 58(1)

308

 

 

Table 56. (cont’d).

x y z U(eq)
C(49) 3955(4) 5053(3) 2834(2) 35(1)
C(35) 8615(4) 9587(3) 1286(2) 41(1)
C(45) 4340(4) 5831(3) 3114(2) 38(1)
C(31) 2290(5) 9257(3) 2389(2) 46(1)
C(46) 20(4) 6666(3) 3491(2) 42(1)
C(26) 7347(4) 1 1934(3) 2849(2) 41( 1)
C(39) 3263(5) l 1396(3) 512(2) 43(1)
C(50) 3524(4) 5045(3) 2831(2) 41(1)
N(3) 2401(4) 6260(2) 4701(2) 53(1)
C(53) 2990(4) 6874(3) 3123(2) 37(1)
C(32) 1202(5) 8268(3) 2172(2) 59(1)
C(42) 2537(5) 5427(3) 4291(2) 43(1)
C(54) 2918(5) 6875(3) 2414(2) 45(1)
C(36) 9954(4) 9839(3) 959(2) 48(1)
C(40) 2410(5) 121340) 534(2) 57(1)
C(27) 8275(5) 129920) 3142(2) 52(1)
C(47) 1368(4) 6403(3) 3348(2) 54(1)
C(43) 2877(5) 5627(3) 4998(2) 52(1)
C(55) 3154(6) 7804(4) 2334(3) 64(1)
C(28) 9075(6) 132230) 3828(2) 75(2)
C(44) 2676(5) 4841(3) 5287(2) 69(2)
C(52) -808l(5) 3979(3) 2313(2) 70(2)
C(48) 2694(5) 7217(4) 3748(3) 78(2)
C(56) 4625(7) 7905(4) 2455(3) 103(2)
F(5) 191(2) 200(2) 3450(1) 49(1)
F(6) 4(2) -9270) 4183(1) 50(1)
F0) 4741(3) 2196(2) 4710(1) 78(1)
F(8) 4538(3) 1079(2) 5456(2) 94(1)
C(23) 933(5) 4410(3) 5183(2) 40(1)
C(18) 1 192(4) 460(3) 4323(2) 39(1)
C(16) 2498(4) 1244(3) 4015(2) 37(1)

309

 

 

Table 56. (cont’d).

x y z U(eq)
C(17) 1294(4) 422(3) 3935(2) 38(1)
C(19) 2273(5) 3(3) 4862(2) 42(1)
C(15) 3890(5) 2608(3) 3654(2) 46(1)
N(7) 36(4) 2070(3) 5142(2) 57(1)
C(14) 2642(4) 1832(3) 3608(2) 36(1)
C(13) 1585(5) 1672(3) 3057(2) 44(1)
C(21) 3550(5) 1415(3) 4565(2) 52(1)
C(22) 2161(5) 395(3) 5273(2) 46(1)
N(5) 820(5) 1595(3) 2601(2) 68(1)
N(6) 4863(5) 3245(3) 3656(2) 71(1)
C(24) 3243(6) 429(4) 5806(3) 68(2)
C(20) 3445(5) 835(3) 4949(2) 56(1)
N(8) 4089(5) 345(4) 6246(3) 99(2)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

310

Table 57. Atomic coordinates ( x 104) and equivalent isotropic
displacement parameters (Azx 103) for
[Zn(TCNQ)2(H20)4(TCNQ)2(MeCNMoo (13).

 

 

x y z U(eq)
Zn( 1) 0 0 0 21(1)
0( 1) 1564(2) -47 1(2) 1059(2) 30(1)
0(2) -5 36(2) - 1935(2) 718(2) 28( 1)
C(1) -l6l8(3) 114(3) 2260(2) 19(1)
C(2) -237 1(3) -91(3) 3280(2) 18(1)
C(3) -3033(3) 1008(3) 3358(2) 22(1)
C(4) -2462(3) -1326(3) 4169(2) 18(1)
C(5) - 1754(3) -2401(3) 4090(3) 19(1)
C(6) -l838(3) 3588(3) 4951(3) 19(1)
C(7) -2638(3) -379l(3) 5961(2) 20(1)
C(8) -3342(3) —2711(3) 6038(2) 20(1)
C(9) -3253(3) 4526(3) 5185(2) 19(1)
C(10) -2746(3) -5008(3) 6856(3) 21(1)
C(11) -2055(3) -61 16(3) 6833(3) 25(1)
C(12) -3587(3) -5214(3) 7831(3) 25(1)
C(13) -462(3) -6466(3) 3296(3) 27(1)
C(14) 210(3) -757 1(3) 3227(3) 22(1)
C(15) 1015(3) -7360(3) 2211(3) 26(1)
C(16) 105(3) -8771(3) 4096(3) 21(1)
C(17) -688(3) -8962(3) 5129(3) 23(1)
C(18) -794(3) -10121(3) 5984(3) 22(1)
C(19) 4553(3) - 1487(3) 3220(3) 25(1)
C(20) 5260(3) -2589(3) 3201(2) 19(1)
C(21) 6125(3) —2410(3) 2234(3) 22(1)
C(22) 5133(3) -3788(3) 4090(2) 18(1)
C(23) 5844(3) 4894(3) 4054(3) 19(1)
C(24) 571 1(3) -6056(3) 4924(2) 19(1)
C(25) 4486(3) - 1528(4) -410(3) 38(1)
C(26) 5454(4) -2065(4) -961(3) 44(1)

311

 

 

Table 57. (cont’d).

x y 2 mm)
C(27) 1780(4) 4082(4) 391(4) 41(1)
C(28) 2693(4) 4693(4) 3290) 46(1)
N(l) -9910) 259(2) 1437(2) 25(1)
N(2) 3554(3) 1909(3) 3413(2) 34(1)
N(3) 4537(3) 2035(3) 6832(3) 37(1)
N(4) 4293(3) 3353(3) 8599(2) 35(1)
N(5) 4000(3) 3584(3) 3349(3) 40(1)
N(6) 1666(3) -7l78(3) 1401(2) 37(1)
N(7) 3978(3) 318(3) 3255(3) 39(1)
N(8) 6838(3) -2287(3) 1481(2) 30(1)
N(9) 3743(3) 4107(4) 30(3) 61(1)
N(lO) 1070(4) 3592(4) 26(4) 66(1)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

312

 

 

Table 58. Atomic coordinates ( x 104) and equivalent
displacement parameters (Azx 103) [Ni(DMSO)5][TCNQ]3 (14).

x y 2 U(OQ)
Ni(l) 5000 o 5000 22(1)
8(2) 8271(2) -1765(2) 5409(1) 30(1)
8(3) 6483(2) 1202(2) 3272(1) 24(1)
0(5) 5038(5) 619(4) 3828(2) 26(1)
0(6) 4753(5) 4924(4) 4706(2) 28(1)
0(7) 7626(5) 469(4) 4924(2) 32(1)
C(8) 6362(9) 2886(6) 3498(4) 34(2)
C(17) 8471(7) 4319(5) 4350(3) 19(1)
C(14) 102210) 3391(5) 3(3) 20(1)
C(19) 111380) 3660(6) 2650) 23(1)
C(11) 102300) 2608(6) 1400(4) 25(1)
N(3) 9190(8) 5233(6) 3399(3) 47(2)
C(20) 7602(7) 4736(6) 2020(3) 24(1)
N(l) 9547(7) 2323(6) 2025(3) 40(1)
N(2) 143940) 2435(6) 567(3) 41(1)
C(16) 7557(7) 4043(5) 388(3) 22(1)
N(4) 4330(7) 4953(6) 4953(3) 42(2)
C(12) 110870) 2913(5) 648(3) 21( 1)
C(18) 103100) 4118(6) 4406(3) 22(1)
C(21) 8493(8) 5011(6) 2779(4) 31(2)
C(15) 8376(7) 361 1(5) 55(3) 20(1)
8(4) 5791(4) 2255(3) 3939(2) 38(1)
C(9) 646707) 4005( 10) 41 14(8) 72(4)
C(10) 428408) 237905) 3281(8) 66(4)
80) 455201) 225300) 3870(4) 26(3)
C(10) 6640(20) 252000) 335606) 34(7)
C(9) 4160(30) 396206) 398906) 35(8)
S(4") 4830(20) 301000) 413202) 48(8)
C(10") 6840(40) 328000) 3540(30) 60(20)
C(9") 3660(60) 2090(50) 340000) 2404)

isotropic

313

 

 

Table 58. (cont’ (1).

x y z U(eq)
C(27) 8200(8) 218(6) 68(4) 27( 1)
C(24) 8405(8) 969(6) - 1341(4) 30(2)
C(22) 5785(8) 4864(6) -1974(4) 30(1)
C(13) 12913(8) 2658(6) 596(3) 26(1)
C(26) 9177(8) 493(6) -677(4) 27(1)
C(25) 9340(9) 1263(7) -2098(4) 36(2)
C(28) 8971(8) -268(6) 7 15(4) 30(2)
C(23) 6557(9) 1 163(6) - 1330(4) 35(2)
N (6) 5102(8) 1309(6) - 1345(4) 51(2)
N (5) 10029(9) 1502(7) -2709(4) 58(2)
C(7) 5602(9) 1491(7) 2363(3) 38(2)
C(6) 9084(8) -1230(7) 6228(4) 39(2)
C(5) 10249(8) 2371(8) 4849(4) 45(2)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

Table 59.

Atomic coordinates

314

( x 10") and equivalent
displacement parameters (Azx 103) for [Mn5C110(TI-IF)3].. (16).

 

 

x y Z ”(39)
Mn( 1) 4059(1) 3554(1) 5529(1) 19(1)
Mn(2) 3574(1) 5864(1) 7365(1) 20(1)
Mn(3) 5000 5000 10000 22(1)
Cl(l) 1993(1) 4803(1) 6581(1) 23(1)
Cl(2) 5774(1) 4621(1) 6246(1) 20(1)
Cl(3) 3152(1) 4276(1) 9219(1) 26(1)
Cl(4) 5270(1) 6722(1) 8199(1) 26( 1)
Cl(5) 3844(1) 7614(1) 5576(1) 24(1)
0(1) 4116(3) 1933(2) 7015(2) 28(1)
C(1) 2880(4) 1632(4) 7907(3) 30(1)
C(2) 3530(5) 813(4) 8882(3) 36(1)
C(3) 4944(5) 147(3) 8322(3) 35( 1)
C(4) 5243(5) 880(4) 7110(3) 45(1)
0(2) 2547(8) 2663(6) 4982(8) 22(2)
C(5) 1210(11) 3257(8) 4609(11) 31(3)
C(6) 372( 10) 2227(9) 4602(13) 40(3)
C(7) 1605(9) 1235(8) 4298(1 1) 34(3)
C(8) 2885(9) 1465(7) 4675(1 1) 26(2)
0(2A) 2361(9) 2603(7) 5213(9) 21(2)
C(SA) 1014(12) 3210(8) 4838(12) 23(3)
C(6A) 225( 10) 21 16(9) 4964(13) 31(3)
C(7A) 1369(12) 1046(1 1) 4763(16) 47(3)
C(8A) 2680(1 1) 1342(8) 5046(13) 30(3)
0(4) 3221(3) 6242(2) 10843(2) 30( 1)
C(13) 3440(4) 7421(3) ”023(4) 34(1)
C(14) 1928(5) 7972(4) 1 1532(5) 52(1)
C(15) 1227(5) 6836(5) 12223(4) 50(1)
C(16) 1844(4) 5867(4) 1 1528(4) 38( 1)
0( 3) 1535(9) 71 1 1(6) 7875( 10) 22(2)
C(9) 1643(12) 8367( 10) 7937(17) 43(4)

isotropic

315

 

 

Table 59. (cont’d).

x y 2 U199)
C(10) 11401) 8942(9) 816904) 35(3)
C(11) 30702) 791400) 868306) 43(4)
C(12) 9501) 6766(9) 834304) 33(3)
O(3A) 1682(7) 7057(5) 8115(8) 22(2)
C(9A) 1733(8) 8275(7) 825200) 28(2)
C(10A) 18700) 8881(8) 8510(13) 49(3)
C(l 1A) 337(9) 7815(8) 9079(1 1) 39(3)
C(12A) 221(8) 6722(7) 8597(9) 27(2)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

316

Table 60. Atomic coordinates ( x 104) and equivalent isotIOpic
displacement parameters (Azx 103) for [MnBr2(TI-IF)2].. (17).

 

 

x y 2 Wm)
Br(1) 6908(1) 4372(1) 3533(4) 16(1)
Mn( 1) 5000 5000 8545(17) 15(1)
0(1) 3207(8) 4058(4) 8560(70) 20(2)
C( 1) 1290(15) 4082(6) 8900(60) 24(3)
C(2) 614(15) 3370(6) 9020(50) 22(3)
C(3) 2144(13) 2833(6) 8190(50) 19(3)
C(4) 3731(16) 3336(6) 9140(40) 23(3)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

Table 61. Atomic coordinates

317

( x 10") and equivalent
displacement parameters (Azx 103) [Fe2C14(/t-2,2’-bpy)2] (l9).

 

 

x y 2 U(GQ)
Fe(l) 1856(1) 1758(1) 741(1) 19(1)
Cl(2) 531(1) -984(1) -1707(1) 25(1)
Cl(l) 4308(1) 1384(1) 2446(1) 30(1)
N( 1) 2522(3) 2867(3) -932(3) 21( 1)
N(2) 1710(3) 4438(3) 1867(3) 20(1)
C( 1) 2982(4) 1991(4) -2310(4) 27( 1)
C(2) 35 14(4) 2762(4) -3292(4) 30( 1)
C(3) 3578(4) 4517(4) -2845(4) 29(1)
C(4) 3107(4) 5444(4) - 1425(4) 25( 1)
C(5) 2592(3) 4590(3) 493(3) 20( 1)
C(6) 2125(3) 5462(3) 1089(3) 20( 1)
C(7) 2145(4) 7222(4) 1759(4) 27( 1)
C(8) 1776(4) 7947(4) 3278(4) 30( 1)
C(9) 1392(4) 6901(4) 4102(4) 29(1)
C( 10) 1350(4) 5157(4) 3345(4) 25( 1)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

isotropic

Table 62. Atomic coordinates

318

( x 104) and equivalent
displacement parameters (Azx 103) for [FeC12(,u-2,2’-bpy)].. (20).

isotropic

 

 

x y z U(eq)
Fe(l) o 4142(1) 2500 16(1)
Cl(2) 3510) 5860(1) 3390(2) 18(1)
N(3) 735(2) 2243(4) 2484(5) 16(1)
C(4) 151 1(2) 2308(4) 2661(7) 22(1)
C(5) 425(2) 909(4) 2563(6) 16(1)
C(6) 1999(3) 1069(5) 2938(7) 25(1)
C(7) 871(3) 385(4) 2770(6) 22(1)
C(8) 1666(3) 292(5) 2985(7) 27(1)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

 

 

 

319

Table 63. Atomic coordinates ( x 10") and equivalent isotropic
displacement parameters (Azx 103) for [Et4N]Cl-[Fe2C14(MeOH)4(,u-2,2’-
bpym)] (26).

 

 

x y z U(eq)
Fe( 1) 4897(1) 8535(1) 1266(1) 22(1)
Fe(2) -3 10( 1) 8727(1) -1465( 1) 23(1)
Cl(l) 5009(1) 8624(1) 2810(1) 31(1)
Cl(2) 4618(1) 7007(1) 1022(1) 36(1)
Cl(3) -511(1) 8991(1) -2992(1) 36(1)
CI(4) -464(l) 7176(1) -1384(1) 37(1)
Cl(S) 2482(1) 10010(l) -371(l) 30(1)
0(1) 3341(3) 8852(3) 1161(3) 41(1)
0(2) 6457(3) 8416(3) 1196(3) 35(1)
0(3) - 1835(3) 9002(3) -1252(3) 47( 1)
0(4) 1271(3) 8614(3) - 1440(3) 39(1)
N(l) 5113(3) 9983(3) 1170(3) 19(1)
N(2) 5156(3) l 1 177(3) 152(3) 19(1)
N(3) -38(3) 10147(3) -1156(3) 21(1)
N(4) 176(3) lll67(3) 6(3) 15(1)
N(5) 2527(4) 9882(3) 4678(3) 30(1)
C( 1) 5258(4) 10573(4) 181 1(4) 25(2)
C(2) 5348(5) l 1452(4) 1667(4) 25(2)
C(3) 5307(4) l 1719(4) 817(4) 23(2)
C(4) 5073(4) 10323(4) 358(4) 17(1)
C(5) 2567(5) 8505(5) 1623(5) 55(2) .
C(6) 7086(5) 7849(4) 1732(5) 46(2)
C(7) 61(4) 10805(4) - 173 1(4) 24(2)
C(8) 216(4) 1 1642(4) - 1448(4) 28(2)
C(9) 262(4) 1 1816(4) -578(4) 25(2)
C( 10) 35(4) 10359(4) -318(4) 19(1)
C(l l) -2704(5) 8859(5) - 1779(5) 51(2)
C(12) 1837(5) 8018(4) - 1916(5) 46(2)
C(13) -2182(5) 8944(4) 4844(4) 36(2)

 

320

 

 

Table 63. (cont’d).

x y 2 Wall
C(14) 2942(6) 8242(5) 4629(5) 53(2)
C(15) 4663(6) 104540) 4948(5) 53(2)
C(16) 48470) 114270) 4828(6) 87(3)
C(17) 2864(5) 100200) 3714(4) 34(2)
C(18) 2104(6) 9853(5) 3099(4) 45(2)
C(19) 3415(6) 100800) 5181(5) 53(2)
C(20) 3237(8) 9961(6) 6156(5) 89(4)

 

U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

IIIIIIIIIIIIIIIIIIIIIIIIIIIIII

111111(Igjallljijjl111121111