19—-

«.«nnu‘fi

”HMOOIM U! I

11??

.. e... i . .1... ._
was“ . . .mh. . .25.

5 3

in?

c .v f.‘4

.n..§.r..w,
141s:

 

 

a“.

4&3
Rani? .

.
.25};
...r1s..

.
..

.z 5..
'9 . 2.... .gvtél -
:r xiii...

N

{In

 

$4.. .
.0525;
b :

 

, 1:5
.I a}?
(.1; :1...

2...; 1....

on . 11.
I)!» Iv. .15.

THESIS

.%
2%0

This is to certify that the

dissertation entitled

DESIGN, SYNTHESIS, AND PROPERTIES
OF A NEW CLASS OF BUILDING BLOCKS
FOR ASSEMBLY OF MOLECULE-BASED MAGNETS

presented by

ANDRZEJ WLADYSLAN MI SIOLEK

has been accepted towards fulfillment
of the requirements for

PH.D. degmin CHEMISTRY

 

 

/ Major profess?’

Date June 27. 2000

MS U is an Affirmative Action/Equal Opportunity Institution 0- 1 2771

LIBRARY
Michigan State
University

. o -

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

woo WWW-p.14

 

DESIGN, SYNTHESIS, AND PROPERTIES OF A NEW CLASS OF BUILDING

BLOCKS FOR ASSEMBLY OF MOLECULE-BASED MAGNETS

By

Andrzej Wladyslaw Misioiek

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Chemistry

2000

ABSTRACT

DESIGN, SYNTHESIS, AND PROPERTIES OF A NEW CLASS OF BUILDING

BLOCKS FOR ASSEMBLY OF MOLECULE-BASED MAGNETS

By

Andrzej Wladyslaw Misiolek

In this work, mono and dianions of the tetrone l were proposed as discrete

building blocks for assembly of molecular magnets.

   

These anionic subunits are designed to form extended multidimensional structures
upon complexation with metal cations. This particular framework also was chosen so that
dianion 12‘ (or its derivatives) would possess a paramagnetic triplet as a ground state.

High quality ab-im'tio CASSCF(14, 12)/6-31+G* calculations showed that indeed
the singlet-triplet gap of metal complexes of 12' is small and its sign and magnitude

tongly depends on structural modifications. For example a 17.3 kcal/mol preference for a

singlet is predicted for isolated 12' while the triplet state is preferred by 4.3 kcal/mol for

the symmetrically coordinated complex l(AlF2)2.

- o o -
F x x F
TA!» @ AL?
F“ ‘o o' "F' o o
1(AlF2)2 2

Synthetic procedures leading to a family of neutral aryl substituted tetrones (2)
has been developed and electrochemical studies have shown that they may be reversibly
reduced in two successive one-electron processes to mono- and dianions.

These mono- and dianions have been generated on a preparative scale by
reduction of the parent tetrones with alkali metals under anaerobic conditions. Solutions
of anions were characterized by Vis-NIR, NMR and EPR spectroscopy. EPR of the anion
radicals 2' reveals the high symmetry of the spin delocalisation. Spectroscopic studies of
dianions were hampered by low solubility and aggregation phenomena and no definite
answer concerning their ground spin states could be obtained.

Several attempts were made to grow single crystals of 2' and 22' in order to probe
the self-assembling properties of those species and to obtain definite answers concerning
the ground state of the dianion. These attempts turned out to be successful only for alkali
metal salts of the monoanion (t-Bu)g(MeO)42'. Its KJr salt crystallized as l-D chains with
the expected bridging and chelating mode of coordination. The closely related Na+ salt,

recently crystallized by Robert Gentner, shows a similar l-D chain structure.

Copyright by
Andrzej Wladyslaw Misiolek

2000

ACKNOWLEDGMENTS

I would like to thank my mentor James (Ned) Jackson for encouragement and
guidance I have received from him while pursuing that project and his struggles with my
quite specific way of communicating in English. I believe that he is not only a very
creative chemist and teacher but also a gentleman. I deeply appreciate the help from all
the other members of the Jackson’s group who formed its distinctive atmosphere I
enjoyed for years. Most of all one from Dalila Kovacs the person I could always count on
in the most diffith moments, and during our hiking tn'ps, the most beautiful ones. I
would also like to mention my labmate Eric Mayer who always inspired me with his truly
Viking zeal for doing things which for others seemed to be impossible.

Many of the performed work would not be possible without technical support
from Professor Jim Dye who also served as a member of my guidance committee. Special
thanks to him and to Andrew Ichimura who thought me secrets of EPR and high-vacuum
techniques. I would also like to thank Professors Michael Rathke, and John McCracken
for serving as the other members of my guidance committee.

During the years spent in East Lansing, United States became my second
homeland. I would like to thank all the people I met here for generosity and friendliness I

received. I am going to keep them in my memories forever.

TABLE OF CONTENTS

LIST OF TABLES ........................................................................... ix
LIST OF FIGURES .......................................................................... xiii
LIST OF ABBREVIATIONS .............................................................. XX
INTRODUCTION ............................................................................ 1
BIBLIOGRAPHY ...................................................... 6
CHAPTER 1. BACKGROUND ........................................................... 7
1.1. Dianion 12' vs. II conjugated diradicals ................................ 7
12- Self-assembly of 1' and 12' into magnetic structures ................. 28
BIBLIOGRAPHY ...................................................... 32

CHAPTER 2. COMPUTATIONAL STUDIES OF SINGLET-TRIPLET GAPS

IN DERIVATIVES OF 12‘ ................................................ 36

2. 1 . Introduction ................................................................ 36
2.2. Results ..................................................................... 41
2.3. Discussion ............................................................... 50
2.4. Computational details .................................................... 55
BIBLIOGRAPHY ...................................................... 57
CHAPTER 3. SYNTHESIS OF THE PRECURSORS OF l'AND 12' ............... 58
3.1 . Introduction ................................................................ 58
3.2. Results ..................................................................... 61
3.3. Discussion ................................................................. 70
3 .4. Experimental Section ..................................................... 72

vi

BIBLIOGRAPHY ...................................................... 78

CHAPTER4. GENERATION AND EPR/ENDOR STUDY OF THE

DERIVATIVES or 1' AND 12' .......................................... 76

4. 1 . Introduction ................................................................ 79

4.2. Results ..................................................................... 84

4.3. Discussion ................................................................. 96
4.4. Experimental ............................................................... 101
BIBLIOGRAPHY ...................................................... 104
CHAPTER 5. NMR ........................................................................... 105
5. 1 . Introduction ................................................................ 105
5 .2. Background ................................................................ 1 06
5.3. Results ..................................................................... 107
5.4. Discussion ................................................................. l 1 1
5.5. Experimental ............................................................... l l 3
BIBLIOGRAPHY ........................................................ 1 16
CHAPTER 6. CYCLIC VOLTAMMETRY .............................................. 117
6.1 . Introduction ................................................................ 1 17
6.2. Results ..................................................................... 1 17
6.3. Discussion ................................................................. 1 19
6.4. Experimental ............................................................... 121
BIBLIOGRAPHY ...................................................... 123
CHAPTER 7. X-RAY DIFFRACTION STUDIES ................................... 124
7. 1 Introduction ............................................................... 124

vii

7.2 Results and discussion ................................................... 124

7.3 Experimental ............................................................ 1 36
BIBLIOGRAPHY ....................................................... 137

CHAPTER 8. ELECTRONIC SPECTROSCOPY ..................................... 138
8.1. Introduction ................................................................ 13 8

8.2 Results ................................................................... 138

8.3 Discussion ............................................................... 139

8.4 Experimental ............................................................ 143
BIBLIOGRAPHY ....................................................... 144

CHAPTER 9. SUMARY AND CONCLUSIONS .................................... 145
APPENDIX A .............................................................................. 152
APPENDIX B .............................................................................. 164

viii

TABLE

1.1.

2.1.

2.2.

2.3.

2.4.

2.5

4.1.

4.2.

LIST OF TABLES

Values of zero field splitting parameter D and S-T gaps for some
cyclopentadiene cations ......................................................... 19
Total energies of the investigated complexes in their lowest singlet

('Ag) and triplet (3 B 3,.) electronic states at all levels of the performed
calculations ....................................................................... 42-44
Occupancies of orbitals A and B in the TCSCF singlet wavefunctions

and values of AE(B-A) in the ROHF calculations ............................. 45
Occupancies of the active orbitals of investigated complexes in the

singlet 'Ag CAS(14,12) and CAS(2,3) wavefunctions ..................... 46
Occupancies of the active orbitals of investigated complexes in

the triplet 3B3“ CAS(14,12) wavefunctions .................................. 47
Selected bond lenghts (in A) of 12' and its complexes optimized

at the CAS(14,12)/6-31G* level ............................................... 48
Values of the hyperfine constants for monoanions of 2 and its

derivatives ......................................................................... 85
Values of the hyperfine constants of species present in the solution of

dianions ........................................................................... 91

ix

6.1.

7.1.

7.2.

7.3.

7.4.

8.1.

A1.

A2.

Bl.

B2.

B3.

Cyclic voltammograms of the 1 mM solutions of the tetrones with
0.1M of (n-Bu)4N(PF6) as supporting electrolyte. Swept rate 0.1V/s,
glassy carbon working electrode, platinum wire counterelectrode,

silver wire quasireference calibrated to Fesz/Fesz+ couple (0.46V). 118

The most important crystallographic data for 2, (NMez)420(MeZSO)3,
(t-Bu)g(MeO)42K-(C4HgO)4 and (t-Bu)3(MeO)42Na°(C5H1002):;. . . . . 125
Selected bond distances for tetrones and their monoanions ................ 126
Selected dihedrals for tetrones and their monoanions ...................... 127
Angles of the mean ring planes vs. the planes of the methylene

centers of tetrones and their monoanions ..................................... 128
Absorption maxima of tetrones and their mono- and dianions ............ 139
Geometry of the unique atoms of 12' and its complexes optimized at

indicated level of geometry (6-31G* basis set) .............................. 152-159
CAS(2,2) Frequencies (cm") of vibrations of 12’ and its complexes in 160-163

singlet (TCSCF) and triplet (ROHF) state. Starred values are

translations and rotations of molecules ....................................
Crystal data for 2 ................................................................ 164
Atomic coordinates (x 104) and equivalent isotropic displacement 165
parameters (A2 x 103) for 2. U(eq) is defined as one third of the trace 0

the orthogonalized Uij tensor ..................................................

Bond lengths [A] and angles [deg] for 2 ..................................... 166-167

B4.

B5.

B6.

B7.

B8.

B9.

B10.

B11.

B12.

B13.

B14.

Anisotropic displacement parameters (A2 x 103) for 2.

The anisotropic displacement factor exponent takes the form:
-21t2[h2a*2U..+...+2hka*b*U12] ............................................... 168
Hydrogen coordinates (x 104) and isotropic displacement parameters

(A2 x 103) for 2 .................................................................. 168
Crystal data for (NMe2)42-((CH3)ZSO)3 .................................... 169
Atomic coordinates (x 104) and equivalent isotropic displacement

parameter (A2 x 103) for (NMe2)42-((CH3)ZSO)3. U(eq) is defined as

one third of the trace of the orthogonalized Uij tensor ..................... 170-171
Bond lengths [A] and angles [deg] for (NMe2)42-((CH3)ZSO)3 ......... 172-175
Anisotropic displacement parameters (A2 x 103) for

(N Me2)42-((CH3)2SO)3. The anisotropic displacement factor exponent

takes the form: —21t2[hza*2U1 1+. . .+2hka*b*U12] ........................... 176-177
Hydrogen coordinates (x 104) and isotropic displacement parameters

(A2 x 103) for (NMe2)42-((CH3)ZSO)3 ......................................... 178-179
Crystal data for (t-Bu)g(MeO)42K'(C4HgO)4 .............................. 180
Atomic coordinates (x 104) and equivalent isotropic displacement

parameters (A2 x 103). U(eq) is defined as one third of the trace of the
orthogonalized Uij tensor ...................................................... 181-183
Bond lengths [A] and angles [deg] for (t-Bu)3(MeO)42K-(C4HgO)4. .. 184-190
Anisotropic displacement parameters (A2 x 103) for

(t-Bu)3(MeO)42K-(C4HgO)4. The anisotropic displacement factor

exponent takes the form: —21t2[h2a*2U1 1+. . .+2hka*b*U.2] ............... 191-193

xi

B15. Hydrogen coordinates (x 104) and isotropic displacement parameters

(A2 x 103) for (t-Bu)g(MeO)42K-(C4H30)4 ................................... 194-196

xii

FIGURE

1.1.

1.2.

1.3.

1.4.

1.5.

1.6.

1.7.

1.8.

1.9.

1.10

1.11.

LIST OF FIGURES

Generation of a,oc,a’ ,a’ -tetraphenyl-m-benzoquinonodimethane. . .. 9
Some non-Kekiile hydrocarbons ............................................... 9
Possible spin states of a biradical (a) Triplet (b) Singlet .................. 12
Generation of TMM by photolysis of pyrazoline 5 ........................ 12

Generation of TME by photolysis of
dimethylenecyclopentanone ................................................... 1 3

Generation of MBQDM from dicarbene .................................... 13

An example of non-Kekule hydrocarbons with (a) disjoint and
(b) non-disjoint NBDMOs ...................................................... 14

Examples of odd and even alternant non-Kekule hydrocarbons. . . . . . 16

Molecular orbital diagram for It orbitals of 4c antiaromatic ions
(a) cyclopropene anion (b) cyclopentadiene cation and
(c) benzene dication .......................................................... 13

Generation of the cations of cyclopentadienes ............................. 18

Perturbation of the molecular degeneracy caused by Jahn-Teller

distortion on the example of the cyclopentadiene cation .................. 19

xiii

1.12.

1.13.

1.14.

1.15.

2.1.

2.2.

3.1.

3.2.

3.3.

3.4.

3.5.

3.6.

Dianion 12' as a derivative of DMCB (15) .................................. 24

Molecular orbital diagram for benzene, rhodazonoic acid dianion

(152') and 12' ..................................................................... 26
Hypothetical 2-D honeycomb magnetic structures formed from

anions of 1 and metal cations (X) with spin relay centers (ax)

symmetry related to high-spin semiquinone complexes (a'x) ............ 29
Oxalate bridged ferromagnetic honeycomb structures with bulky
counterions filling void space ................................................. 30

All possible determinants in CAS(2, 2) active space ....................... 39

S-T gaps calculated for a) 12' complexed by different metal cations b)

lLiz with different substituents R, vs. AE(B-A) ............................... 49
Two hypothetical modes of generation of 12' ............................... 58
Three main methods of synthesiszing quinone methides ................ 59
Attempt to prepare 2 from 3H2 and benzophenone ......................... 61

Preparation of 1,2,4,5-tetrahydroxybenzene from
dihydroxybenzoquinone ....................................................... 61

Preparation of dichlorodiphenylmethane .................................... 62

Reaction between tetahydroxybenzene and

dichlorodiphenylmethane ...................................................... 63

xiv

3.7.

3.8.

3.9.

3.10.

3.11.

3.12.

3.13.

3.14.

3.15.

4.1.

4.2.

4.3.

4.4.

Preparation of 2 by a substitution-dehydrogenation sequence ............ 63

Dehydrogenation of 2H, to 2H2 and 2 ....................................... 64
Prepared phenyl substituted derivatives of 2 ................................ 66
Preparation of benzhydrol 9 ................................................... 66

Preparation of 10 and its attempted condensation with

dihydroxyquinone .............................................................. 67
Possible preparation of 1 by dehydration of tetrol 11 ...................... 67
Preparation of diol 13 ........................................................... 69
Reduction of 13 to 14 instead of its attempted deprotection to 15 ...... 69
Oxidative deprotection of 13 with CAN ..................................... 69
A typical solid state spectrum of triplet diradical ........................ 83

ENDOR of 2'[K@c(2.2.2)]+/T HF recorded at 246K

(a1=3.23; a2=1.43 MHz) ....................................................... 86
A) EPR of 2'[K@c(2.2.2)]+/T HF recorded at 246K.

B) Simulation with parameters obtained from ENDOR ................... 86
ENDOR of (MeO)42'[K@c(2.2.2)]+fTHF recorded at 215K,

(a.=3.24; a2=l .31 MHz) ......................................................... 87

XV

4.5.

4.6.

4.7.

4.8.

4.9.

4.10.

4.11.

4.12.

4.13.

4.14.

4.15.

A) EPR of (MeO)42‘K@c[2.2.2]+/THF recorded at 215 K.
B) Simulation with parameters obtained from ENDOR ................
ENDOR of (N Me2)42'[K@c(2.2.2)]+/THF recorded at 210 K,

(a,=3.26; a2=1.08) ............................................................

EPR of (NMez)42'[K@c(2.2.2)]+/THF recorded at 243 K. . . . . . . . .

ENDOR of (t-Bu)3(MeO)42'[K@c(2.2.2)]+/THF recorded at

197, 223 and 263K, ala=5.4 MHz; a|b=1.5 MHz (197K),

a.=3.39 (263 K) .................................................................

EPR of (t-Bu)g(MeO)42'K@c[2.2.2]+/T HF recorded at 180 K ...........

A) EPR of (t-Bu)g(MeO)42'K@c(2.2.2)+/THF recorded at 263 K B)

Simulation with parameters obtained from ENDOR ......................

ENDOR of 22'[K@c(2.2.2)*]2/THE recorded at 240 K

(a1=7.87 MHz, a2=3.22 MHz) .................................................

EPR of 22‘[K@c(2.2.2)]*2/rHF recorded at 247 K .........................

ENDOR of (t-Bu)g(MeO)422‘[K@c(2.2.2)+]2/THF recorded

at 263 K, (a1=8.30 MHz) .......................................................

A) EPR of (t-Bu)g(MeO)422'(K@c[2.2.2]+)2/THF recorded at 290K
B) Simulation with parameters obtained from ENDOR .................
Progress (from A to C) of reduction of 2 with Cs mirror monitored

by EPR. Spectra collected at 4.2 K (A), 3.9 K (B), 4.2 K (C) .........

xvi

87

88

88

89

90

90

92

92

93

93

95

4.16.

4.17.

4.18.

4.19.

4.20.

5.1.

5.2.

5.3.

5.4.

5.5.

Possible reasons for splitting of the ENDOR signal of the meta proton

of the (t-Bu)g(MeO)42° ......................................................... 96
Possible species responsible for EPR signal found in the solution of
dianions ........................................................................ 98

Proposed concerted (a) and stepwise (b) paths of fragmentation

of 22' and its derivatives ........................................................ 99

Two possible modes of aggregation of 2Cs .............................. 100
A typical H-cell used for preparation of EPR samples .................. 102
Reductions investigated by H1 NMR ........................................ 105

'H NMR of the mixture of (t-Bu)g(MeO)42 and c(2.2.2)

(THF-d8, RT) a) Before and b) After a very short reduction time

A) o-hydrogens of the phenyl group, B) metoxy protons,

C) t-Bu protons, D) c(2.2.2) peaks ........................................... 108
Progress of the reduction of (t-Bu)3(MeO)42 observed in the 1-2 ppm
range, THF and. A) t-Butyl protons may be observed.

B) probably belongs to small diamagnetic impurity ....................... 109
Aromatic region of the 1H NMR spectrum of the diamagnetic

products obtained by decomposition of the solutions of a) 22’,

b) (MeO)422' and c) (MezN)422' ............................................. 110
Proposed species responsible for NMR signals presented on

figure 5.4 ........................................................................ 113

xvii

5.6.

6.1.

7.1.

7.2.

7.3.

7.4.

7.5.

8.1.

8.2.

8.3.

Instrument used for preparation of the NMR samples .................. 115

Cyclic voltammogram of (t-Bu)g(MeO)42 in THF ...................... 118

Symbols of bonds and dihedrals used in tables 7.2 and 7.3 ............ 125

ORTEP drawing (50% probability) of 2. Hydrogen atoms are not

shown ............................................................................ 130
ORTEP drawing (50% probability) of (N Me2)42-(Me280)3. Hydroge
atoms and DMSO molecules are not shown ................................ 131
ORTEP drawing (50% probability) of chains formed in the solid state

by radical anion salt (t-Bu)3(MeO)42K-(THF)4 with one unique

molecule of complex. Hydrogen atoms, THF molecules (excluding
coordinated oxygen atoms) t-butyl and metoxy fragments, are not

shown .......................................................................... 133
ORTEP drawing (50% probability) of chains formed in the solid state

by radical anion salt (t-Bu)3(MeO)42Na-(DME)4 with two unique
molecules of complex. Hydrogen atoms, DME molecules (excluding

coordinated oxygen atoms), t-butyl and metoxy fragments are not

shown ............................................................................ 134
UV-Vis absorption spectra of neutral tetrones (acetonitrile) ............. 140
Vis-NIR spectra of the THF solutions of monoanions .................... 141
Vis-NIR spectra of dianions ................................................... 142

xviii

9- 1- Photochemical generation of 2(SnBU3)2 ....................................................... 139

xix

AcOH-
c(2.2.2)-

CAN-

CASSCF (CA8)-

DDQ-
DDQHT
DHQ-
DMF-
ENDOR-
EtzO-
EtOH-
MeCN-
MeOH-
ROHF-
TCSCF-

THF-

LIST OF ABBREVIATIONS

Acetic Acid (CH3COOH)

4,13 ,16,21,24,-hexaoxa-1,10-diazabicyclo-[8.8.8]hexacosane
Ceric Ammonium Nitrate ((N 1-14)2Ce(N03)6)
Complete Active Space Self-Consistent Field
2,3-Dichoro-5,6-dicyano-l ,4-benzoquinone
2,3-Dichoro-5,6-dicyano-l ,4-catechole
2,5-Dihydroxy-1 ,4-benzoquinone
N,N-Dimethylformamide (HCOON(CH3)2)
Electron-Nuclear Double Resonance

Diethyl ether (C2H5)ZO

Ethanol (C2H50H)

Acetonitryle (CH3CN)

Methanol (CH3OH)

Restricted Open-Shell Hartree-Fock
Two-Configuration Self-Consisted Field

Tetrahydrofuran (C4H30)

XX

INTRODUCTION

In this work mono and dianions of tetrone 1 are proposed as new building blocks

for assembly of molecular magnets.

 

R R R R
O O O E O
O O O l O
R R R R
1 1 ' 12'“

Molecular magnets are materials where the magnetically active elements are
organic molecules instead or in addition to traditional inorganic components like
transition metal or lanthanide ions and atoms.1 Since synthetic chemists have much better
control over organic than inorganic molecules, the main advantage of such materials
would be the feasibility of their rational design and ‘fine tuning’ of their properties.
Properties unknown for traditional magnets such as optical transparency, chirality,
solubility in organic solvents, etc. may also lead to new applications in future
technologies.

A purely organic and hybrid inorganic-organic (preferred in our group) approach
is used to synthesize these materials. In the first one, only organic molecules (mainly
stable organic radicals) are assembled into the magnet. In the hybrid approach the organic
components are used together with metal cations. The cations aid their organization into

desired extended magnetic solid state structures and additionally play the role of robust

1

spin carriers or relays. So far, neither of these approaches have produced any molecular

magnet promising practical application. Among the reasons for this failure are the

shortcomings of the organic components available to date. The design and synthesis of 1

is an attempt to change that situation and the following discussion will rationalize that

choice.

Organic molecules used as building blocks for assembly of molecular magnetic

materials must meet a few crucial conditions. First of all two basic requirements

concerning their abilities to form solid state magnetic networks must be met:

They should possess strong and predictable spin communicating abilities. Magnetism

is a cooperative phenomenon and organic spacers must possess ability to strongly
‘couple’ separate paramagnetic centers of the magnetic solid (usually unpaired
electrons of the metal cations or of the other spacers). A high magnitude of coupling

is required in order to obtain ‘room temperature’ magnets

They should h_ave higtwrobabilitv of self-assembly into extended structures in which
r_n_agnetic interaction_s can be propagated along multidimensional extended networks.
Two coupled paramagnetic centers do not make a magnet. In order to produce a
domain of magnetic material extended cooperative magnetic interactions among
thousands of paramagnetic components in the solid state is required. But even an
infinite chain of high-spin coupled paramagnetic centers is not enough.
Multidimensional interactions (2D sheets or better 3D nets) are required in order to

observe a ferromagnetic phase transition.

The first requirement should be easily met by 1' (or 12'). Due to its chelating and
bidentate structure, paramagnetic anions of 1 should have high chance to bind two
metallic centers at the same time. The unpaired electron of such a spacing ligand 1’ would
have close contact with the unpaired electrons concentrated on the metal cations or their
coordination sphere. Such a close contact should assure particularly strong spin-spin
interactions and coupling of considerable strength.

Fulfillment of the second requirement, so far, is the most difficult for all
compounds investigated as a discrete elements of Molecular Magnets. Although some
paramagnetic spacers with relatively good coupling abilities has been prepared, their self-
organizing properties were lefi too much to chance. Consequently solids obtained from
them do not form robust magnetic materials because the structures obtained do not
possess required multidimensional topology of intermolecular magnetic interactions. We
believe that contrary to most other such molecules, anions of 1 have a high probability of
forming extended magnetic networks with metal cations. Some evidence for this
assumption will be presented in Chapter 1.

In addition to the above two pivotal ‘magnetic’ requirements, anions of l have
other interesting properties, which should increase the chances of their successfill self-

assembly into molecular magnets.

o 1 may be paramagnetic on multiple stages of reduction. As a result of careful design
12' has a high chance of having paramagnetic triplet as its ground state (see Chapter

2). This dianion diradical could be thus used as a paramagnetic coupler in

combination with or in place of the monoanion monoradical. This notion offers
flexibility in terms of charge balancing with counteranions and consequently,
increases the chance of achieving the desired structures. A higher spin density
diradical could be favorable for forming magnets based on a fully ferromagnetic
coupling network while the monoradical might be preferred for construction of
ferrimagnets. Fine tuning of the triplet-singlet gap of 1 on the molecular level should
also enable control over material properties and lead to magnets with ‘custom
designed’ properties.

Both 1' and 12’ are anionic. In many of the reported solids discrete organic
paramagnetic components are neutral and solid state intermolecular interactions
between them rely on relatively weak Van der Waals or second order electrostatic
forces. The opposite charges of the anions of l and the metal cations should
strengthen intermolecular interactions, bringing them closer together and facilitate
their magnetic communication. It would also allow synthesis of a hybrid organic-
inorganic material without disruption of the magnetic networks by insulating
diamagnetic counterions.

Electronic and packing properties of 1 should be easy to modify by different
substituents R. Since one of the most desirable properties of molecular magnets is the
possibility of fine-tuning of their properties on the molecular level, the electronic
properties of the basic unit should be easily modifiable. Direct conjugation of R with

the 1! system should provide that possibility for the electronic properties of 1.

The later parts of this thesis will provide some background information
concerning the possibility of the triplet ground state for 12' and the chances of 1' and 12'
to self-assemble into hoped for magnetic structures, followed by presentation and

discussion of results obtained during the course of the study.

BIBLIOGRAPHY

(1) (a) Caneschi, A.; Gatteschi, D.; Sessoli, R.; Rey, P. Accounts Chem. Res. 1989,
22, 392. (b) Gatteschi, D.; Kahn, 0.; Miller, J. S.; Palacio, F. Adv. Mater. 1991, 3,
161. (c) Gatteschi, D.; Sessoli, R. J. Magn. Magn. Mater. 1992, 104, 2092. (d)
Gatteschi, D. Adv. Mater. 1994, 6, 635. (e) Kollmar, C.; Kahn, 0. Accounts
Chem. Res. 1993, 26, 259. (f) Kahn, 0. Nature 1995, 378, 667. (g) Miller, J. S.;
Epstein, A. J. Angew. Chem. -Int. Edit. Eng]. 1994, 33, 385. (h) Miller, J. S.;
Epstein, A. J. In Materials Chemistry, 1995; Vol. 245; pp 161-188. (i) Iwamura,
H. Pure Appl. Chem. 1987, 59, 1595. (j) Rajca, A. Chem. Rev. 1994, 94, 871.

Chapter 1

BACKGROUND

1.1 Dianion 12' vs. 7! conjugated diradicals

The possibility that derivatives of dianion 12' may have triplet ground states is
intriguing both from theoretical and practical points of view. Triplet ground state organic
molecules are very uncommon and despite great theoretical and experimental effort
invested into the study of such compounds there are only a few classes of them known to
date. Triplet state derivatives of 12' would not only be interesting examples of a novel
class of such molecules, but thanks to their potential for self-assembly via interaction
with metal cations, they would be of special interest for material science. The reason for
the rarity of triplets is not only their diradical character and consequent high reactivity.
The frontier orbitals of such molecules must also fulfill some specific conditions. Only
when their HOMO-LUMO gap is small or zero and the two orbitals overlap significantly
in space may a triplet ground state be expected. Fulfillment of these conditions leads to
Hunds rule for atoms and ions or some simple molecules (e.g. 02) which are probably the
best known examples of high-spin species.

Dianion 12' could be classified as an organic diradical with both unpaired
electrons belonging to a single planar 1t conjugated system. Other rare examples of
similar high-spin organic molecules are Non-Kekule hydrocarbons where HOMO-LUMO
degeneracy is based on their topology and 4m: electron annulenes where it is based on

the symmetry of the molecule. Both groups of diradicals will be briefly presented and
7

their relation to 12' will be used to rationalize the hope for triplet as a ground state of that
molecule.

Non-Kekule hydrocarbons. In 1858 Friedrich August Kekulé put forward the idea
that carbon atoms are quadrivalent and link together to form ‘backbones’ of organic
compounds.1 The proposed rules were so successful in assigning structures of the already
known and newly discovered compounds that with time it became unimaginable that
substances that violate Kekule’s principles might exist. However Gomberg’s discovery of
triphenylmethyl, a free radical with trivalent carbon, proved such violations to be
possible.215 years later, an even more fascinating molecule was prepared. Following the
methodology used by Gomberg, Schlenk and Brauns3 allowed a solution of
dichloroxylene 2 to react with powdered copper under a C02 atmosphere (figure 1.1).

Abstraction of chlorine atom from 2 generated a yellow monoradical 3' that reacted
further with another equivalent of copper forming a blue colored product 4".
Remarkably, 4" retained the radical character of 3' despite the fact that both of its radical

centers belonged to a single conjugated system. Although such results might be

intuitively explained considering the fact that no Kekule structure with all valences paired
could be drawn for 4”, no rigorous explanation based on fundamental physical principles
could be supplied at that time. Better insight into the character of this and similar

hydrocarbons had to wait for refinement of the theoretical methods of quantum

chemistry.

CI Ph Ph C. Ph Ph CI Ph Ph
—-p. ' _.. O 0
Ph Ph Cu Ph Ph Cu Ph Ph

2 3. 4..

Figure 1.1. Generation of a,or,or’,a’-tetraphenyl-m-benzoquinonodimethane

'U' A. H I):

MBQDM TMM TME TMB

Figure 1.2. Some non-Kekiile hydrocarbons

In 1950 Longuet-Higgins proved that, in the non-Kekule hydrocarbons (as he
called hydrocarbons for which no closed shell structure could be drawn despite the fact
that they possessed even number of free valences) that do not contain 4n-membered
rings, the number of atoms that cannot be assigned to 1: bonds is equal to both the number
of it nonbonding degenerate molecular orbitals (N BDMO’s) and the number of electrons
that must to be accommodated in them.4 Some simple examples of such hydrocarbons are
m-benzoquinodimethane (MBQDM), trimethylenemethane (TMM),
tetramethyleneethane (TME), and 1,2,4,5-tetramethylene-benzene (TMB). Thus each of
the molecules on figure 1.2 has two it NBDMO, which contain a total of two electrons. In
order to avoid coulombic repulsion, each of those electrons tends to occupy a different
NBDMO, which explains the diradical character of the molecule.

A full description of the electronic states of such diradicals requires specification
of their spin states. It is possible for the spins of the frontier electrons to be parallel or
anti parallel to each other producing a molecule with spin number S=0 (singlet) or 8:1
(triplet) (figure 1.3). Longuet-Higgins came up with the idea that Hund’s rule, which so
successfully explained the spin states of atoms and atomic ions could be
straightforwardly extended to molecules, so that each of the diradicals in figure 1.2
should have a triplet as their ground state.

The introduction of Electron Pararnagnetic Resonance Spectroscopy (EPR) and
matrix isolation techniques offered experimental verification of those speculations,
generally supporting Longuet-Higgins’ hypothesis. In 1966 Dowd generated the simplest
non-Kekule hydrocarbon, TMM, by photolysis of pyrazoline 5 in a frozen glass of

perfluorinated solvents (Figure 1.4).5 The presence of TMM in its triplet state was

10

confirmed by ESR spectroscopy. The broad spectrum obtained, a general feature of triplet
diradicals, could be fitted to a Hamiltonian with zero field splitting parameters‘ being in a
fair agreement with those predicted for TMM by M0 calculations. Using a similar
technique, in 1970, Dowd prepared and recorded the triplet spectrum of TME (Figure
1.5)6’7 while in 1983 the spectrum of unsubstituted MBQDM (generated from the
dicarbene) was reported by Platz’s group (Figure 1.6).8 The intensity of the EPR triplet
signals vs. temperature of all those diradicals obeyed the Curie law which meant that they
had either a strong preference for the triplet state or energetic near degeneracy of triplet
and singlet states (within ca. 10 cal/mol) with the second possibility being obviously
much less likely. Such ideal Curie behavior of obtained diradicals seemed to verify the

applicability of Hund’s rule for all non-Kekule hydrocarbons.

 

.Zero field parameter D relates to the average distance between unpaired electrons while B relates to

deviation from the two fold symmetry of the molecule (see Chapter 4.1).

11

—i_ —1— —1— +

NBDMO NBDMO NBDMO NBDMO
:0 =1
a) b)

Figure 1.3. Possible spin states of a biradical (a) Triplet (b) Singlet

Figure 1.4. Generation of TMM by photolysis of pyrazoline 5.

12

Figure 1.5. Generation of TME by photolysis of dimethylenecyclopentanone

Figure 1.6. Generation of MBQDM from dicarbene.

13

This conclusion however was challenged by W. T. Borden. Inspired by his own
failure to observe the triplet of Dar, cyclooctatetraene (COT), Borden realized the
necessity of considering the effects of electron-electron repulsion more carefully. He
classified non-Kekule hydrocarbons into two groups: with disjoint (like TME, TMB) and
non-disjoint NBDMOs ( like MBQDM, TMM).9’lo The physical basis for Hund’s rule is
that, when two electrons have the same spin, the Pauli exclusion principle does not allow
the electrons to occupy the same regions of space simultaneously, even if they are in
different molecular orbitals. Since electrons of opposite spins are not prevented by the
Pauli principle from appearing in the same regions of space, they usually have a much

higher mutual Coulombic repulsion energy than two electrons of the same spin.

3% H? 6336 OJ»

TME TMM

(a) (b)

Figure 1.7. An example of non-Kekule hydrocarbons with (a) disjont and (b) non-disjoint
NBDMOs

However, as can be shown on the example of TME (figure 1.7), the two singly
occupied molecular orbitals can be chosen so that they are disjoint, i.e., have no atoms in
common. In both the lowest singlet and the lowest triplet states then, one electron
occupies each one of those NBDMOs. Because these MOS are disjoint, regardless of
whether the electrons in them have the same or opposite spin, there is no probability that

both electrons will simultaneously occupy the same AOs. Thus to a first approximation,

14

the lowest singlet and the lowest triplet state of TME have exactly the same energy.
Moreover electron correlation between the electrons in the NBMOs and those that remain
in the bonding MOs of the molecular fragment usually drops the energy of the singlet
well below that of the triplet. Although the diradical character of such hydrocarbons is
retained, there is little to favor the triplet anymore, and indeed, a modest singlet

11,12

preference is expected. Later high quality ab-initio calculations along with gas phase

photoelectron SpCCtI‘OSCOpyB’M

confirmed the strong triplet preference of both TMM
(~15kcal/mol) and MBQDM (~10 kcal/mol), while CISD(Q) calculations on planar
TME gave 1.6 kcal/mol preference for singlet.” The preference for the singlet ground

state was also established by photoelectron spectroscopy.16
:3 .

These results for TME seemed to be in disagreement with the previously observed Curie
behavior of the intensity of its EPR signal. This paradox prompted Matsuda and Iwamura
to reinvestigate the magnetic behavior of this system with the use of a SQUID
magnetometer, a much more precise instument than the EPR spectrometer. Indeed, study
of the magnetization of conforrnationally fixed 6 indicated a miniscule singlet rather than
a strong triplet preference.l7

To distinguish quickly between disjoint and non-disjoint non-Kekule
hydrocarbons the concept of odd and even alternate hydrocarbons may be applied. Each

of the carbon atoms from the extended 1t system of non-Kekule hydrocarbons may be

marked (starred) or remain unmarked (unstarred) (figure 1.8) such that starred carbons

15

adjoin only with unstarred and vice-versa. It has been proven18 that when the number of
starred (N *) and unstarred (N) carbons are equal (N *-N=O), each of the two NBMOs of
the hydrocarbon can be localized on a separate set of carbons so they can remain disjoint.
However, if there is an excess of one type of carbons over another, all NBMOs are
nondisjoint. As a consequence, such an entity follows Hund’s rule and the spin number of

the molecule is equal to (N *-N)/2.

a: * i * t * * a
U A H EC
* * * i * * ‘R
S=(N*-N)/2=1 S=(N*-N)l2=0

Figure 1.8. Examples of odd and even alternant non-Kekule hydrocarbons.

Determination of the singlet as a ground state of the another even alternant non-
Kekule hydrocarbon TMB (figure 1.2) seems to confirm the viability of Borden’s
hypothesis. '9'”

4m: electron (‘antiarmaticfi annulenes. Besides the non-Kekule systems, another
group of fully conjugated hydrocarbons possessing degenerate HOMO and LUMO are
annulenes with 4n1t electrons. Such hydrocarbons are particulary unstable, in contrast to

the highly stabilized aromatic 4n+2 systems, and are called ‘antiaromatic’. As explained

earlier degeneracy of frontier orbitals does not always lead to a triplet as a ground state.

16

 

 

 

 

 

 

CB COT

Neutral 4n1t annulenes square cyclobutadiene (CB) and planar 1,3,5,7-cyclooctatetraene
(COT) are not subject to Hund’s rule because of the disjoint character of their half-
occupied orbitals.10 The 4m: rule of degeneracy can be extended, however to charng
conjugated cyclic systems (figure 1.9). For example, besides cyclobutadiene other
systems with 4 electrons which could potentially have a triplet ground state are the anion
of cyclopropene (a), the cation of cyclopentadiene (b) and the dication of benzene (c).
Frontier orbitals of charged 4n1t annulenes are unable to localize their unpaired electrons
on separate sets of carbon atoms to become disjoint so their ground states are thus
expected to be triplets, barring significant molecular distortion. Such charged
antiaromatic systems have attracted interest of chemists for a long time and experiments

probing their electronic structure date back to the early sixties.

17

 

(a) (b) (C)

Figure 1.9. Molecular orbital diagram for it orbitals of 4e antiaromatic ions

(a) cyclopropene anion (b) cyclopentadiene cation and (c) benzene dication.

R OH R
R R BF3 R @ R
——. ‘
-60C
R R

CPR;

Figure 1.10. Generation of the cations of cyclopentadienes.

18

Figure 1.11. Perturbation of the molecular degeneracy caused by Jahn-Teller distortion

on the example of the cyclopentadiene cation.

 

 

R D (cm'l) Ems) (kcal/mol)
R H 0.1865 ~0
R R
CI 0.1495 ~0
R R
Ph 0.1050 0.35-1.15

p-MeO-Ph Not rep 0.05

 

Table 1.1 Values of zero field splitting parameter D and S-T gaps for some

cyclopentadiene cations.

l9

The simplest member of this group, the cyclopropenyl anion, is known only as a
transient reactive intermediate and has not been experimentally investigated in detail.

Study of the cyclopentadienyl cation system turned out to be more successfirl. Its
derivatives were prepared by reaction of cyclopentadienols or cyclopentadienyl halides
with Lewis acids below —60°C (Figure 1.10).” The cations so prepared are extremely
reactive and decompose when their solutions are warmed up. EPR spectroscopy of frozen
solutions detected triplet diradical signals for some of them, but only a couple of them
had linear Curie plots that could indicate triplet ground states. Evidently their preference
for the Triplet State is intrinsically low. D values and relative energies of the triplet and
singlet states for some investigated cations are presented in table 1.1. As can be expected
with higher delocalization of the frontier orbitals, spin-spin interactions of the frontier
electrons become weaker (as indicated by decreasing D values). Higher delocalization of
interacting electrons also leads to smaller preferences for the triplet state, as expected
considering the Colmnbic basis of Hund’s rule.

A very important factor determining the magnitude of the S-T gap is the
symmetry of the molecule. Not all systems with 4M electrons have the potential to have
triplet ground states. Only if they have C 3 or greater symmetry does one expect
degeneracies in some orbitals with half occupancy of a degenerate orbital pair. However
even a molecule which could have this symmetry, need not adopt it if distortions are
readily available. Instead the possible degeneracy of the orbitals can be lifted by Jahn-
Teller distortion, which stabilizes the singlet state relative to the triplet since the paired

electrons of the singlet state can both go into the orbital whose energy has been lowered

by distortion (Figure 1.11).23

20

The freedom of rotation of phenyl rings makes molecules such as Cp(p-MeO-
Pb); especially prone to this type of behavior and may be another reason besides
delocalization for the failure to observe triplet signal in some cases. Due to the highly
reactive nature of cyclopentadiene cations, efforts to use them for synthesis of molecular
magnets has been essentially futile, and no direct information about the structures of
Cpr is available.

Derivatives of benzene dications displayed sufficient thermal stability that
crystals of their charge transfer complexes could be grown. Although addition of a
second positive charge might be expected to make the molecule more reactive, this
drawback is offset by the synthetic accessibility of systems with more electron donating
substituents than are possible for cyclopentadienyl system. The ground state of 72+ in the
solid state is a singlet by ~4.8 kcal/mol.24 Examination of the crystal structures reveals
that this dication undergoes Jahn-Teller distortion to adopt D 2;, rather than higher D6
symmetry optimal for the triplet. Bond lengths of 72+ suggest a description of its structure
as a union of two cyanine dye fragments. EPR signals of 72+ in solution and solid state
vary, in solution the exited triplet state molecule is able to minimize its energy by coming
back to its high symmetry state with an E value equal to 0. Such interconversion is
impossible in the crystalline solid, where molecule is immobilized in the lattice and the
resulting EPR spectrum is that of a D21, species. This phenomenon indicates that the
thermal equilibria between both spin states observed in solution is complicated by

equilibria between many possible conformers of the molecules.

21

OMe

RIN

NIR

12

1
1

 

22

Dication 82+ has a much more delocalized 7! system than 72+. EPR studies in this
system established the linearity of the Curie plot (meaning either near degeneracy of
singlet and triplet or strong preference for the triplet state). The less flexible framework
presumably helps to prevent distortion.23 A definite answer, however, will be difficult to
reach until a crystal structure of a salt of dication 82+can be determined. Unfortunately,
this goal remains elusive due to the low stability of 82+. Similar EPR behavior was
displayed by the even less stable (although easier to prepare) dication 92+.

Besides being oxidized to 41: electrons, benzene derivatives may be also reduced
to 81: electron systems. Several series of such dianion-diradicals were studied in the
sixties and some of them (102', 112', 122') produced triplet signals.”27 Values of the
observed zero-field splitting parameters were solvent and counter anion dependent
pointing to the great importance of ion pairing phenomena. The most informative studies
have been done recently on the silyl stabilized dianion 132'.28 Crystals of its lithimn salt
suitable for x-ray diffraction could be grown and their structure determined. Like 72+, 132'
undergoes Jahn-Teller distortion leading to D2}, symmetry in the dianion despite the C3
mode of substitution. An especially interesting case is coronene 14 for which the triplet

states of both dianion 142' and dication l42+ have been observed. 2930

Classification of 12'. The dianion 12' can be simply classified neither as non-

 

Kekule hydrocarbon nor as antiaromatic annulene. In a rather straightforward way,
however, it can be related to both groups of diradicals. Dianion 12' can be looked at as a
formal derivative of the well-characterized non-Kekule diradical, dimethylene-

cyclobutane-1,3-diyl (DMCD) 15.73"32

23

   

O

 

a (lb3g) b (2b1u) A (21723) B (3b1u)

Figure 1.12. Dianion 12' as a derivative of DMCB (15).

24

Exchange of methine carbon atoms with semidione radical anion (Figure 1.12)
produces 12' with a pair of frontier orbitals A (2b 28) and B (3121.) related to the frontier
orbitals of 15 (a (lb 38) and b (2b 1“». Preliminary semiempirical calculations have shown
that this exchange should not excessively perturb the degeneracy of the frontier orbitals.
It can be rationalized by the relative closeness of the SOMO energies of the exchanged
fragments (methine carbons and semidione units). Since the new orbitals should retain
degenerate and non-disjoint character of a and b, 12' (if stable) can have a triplet as
ground state. In a more systematic manner the It orbitals of 12’ can be derived from those
of benzene. In order to do this, it is convenient to consider also the orbitals of the related
dianion of rhodizonic acid (162'). Figure 1.13 presents energies of molecular orbitals
calculated for benzene, 162' and 12' calculated with use of the Hiickel method. Each of the
it orbitals of the benzene can be assigned to two orbitals of 162'. Both of them will have
the same nodal planes as the parent benzene orbital and one of them in addition includes
a radial node going around the molecule in-between the carbon and oxygen atoms rings.
It can be considered as an effect of constructive and destructive interactions of the p A0
of internal carbon and external oxygen rings. Since the lowest orbital with a radial node
has higher energy then the highest one without it, two distinct, six orbital subsets are
formed, preserving the orbital pattern of benzene. Out of the two degenerate SOMOs of
12‘, the one not having coefficients over external carbon atoms (A) remains relatively
unperturbed compared to the related orbital in 162', while the energy of the delocalized
over carbon atoms B gets higher. The Magnitude of that shift has value close to the

HOMO/LUMO gap in 162' and A and B orbitals in 12' are almost degenerate.

25

 

O 0
C O

 

 

 

1%}

 

—H—
929 8 8 6.2.6
—H— —H— 03930 .‘o‘. 0o g
0 ‘0 O .9
—H-— "HT —u—
888 ~e
'""' 03°20 -"" o t“-
. ‘ . 00000
O O O
—H—

 

 

 

 

. . O
.8 8.
3b/:(B)

0.8.0
0'8'0

2193,;

0.3.0
00 O 00

1b).,

1"

 

20.,

_ +2.0

_ +3.0

 

Figure 1.13. Molecular orbital diagram for benzene, rhodazonoic acid dianion (162') and

12'

26

These interesting preliminary results encouraged us to pursue a more detailed

theoretical study which will be described in chapter 2.

27

1.2. Self-assembly of 1' and 12' into magnetic structures

Proposed anions 1" and 12' are structurally akin to 172' and the abilities of such
derivatives of dihydroxyquinone dianion and metal cations to self-assemble into extended

networks is well documented.

R
O 0 O
O O' O
R
172' 18

Structures with various dimensionalities, including infinite chains,33 7‘ honeycomb
layers42 or even diamond-like 3D nets43 have been reported for salts of 172' or its
derivatives. Replacing 172' with paramagnetic structural equivalents should not destroy
its self-organizing properties and would hopefirlly replicate such motifs. In addition to
ensuring the required multidimensional extended topology of interactions, the expected
honeycomb motifs like X (figure 1.14) should posses high symmetry of the coordination
sphere (ax) of the metal cation. This fact gives hope that predictable intermolecular
coupling in such networks can also be fulfilled. Strong evidence supporting that hope is
the fact that radical anions 18' coordinated around cations of some main group metals
(Ga3+, Al”) form complexes a'x with almost identical to ax symmetry and are high-spin
coupled (figure 1.14). The J coupling between paramagnetic ligands equal to 6.2 cm'I
(17.7 cal/mol) and 8.6 cm'l (24.6 cal/mol) for Ga”, A13+respectively.44 In, related to X,
A12n13n or Ga2nl3n honeycomb structures all unpaired spins would be supplied by organic

anions with diamagnetic cations playing inert, structural role.

28

’ O
3 ”'0’2. 0.... .50 :. 3X

--0

Figure 1.14. Hypothetical 2-D honeycomb magnetic structures formed from anions of 1
and metal cations (X) with spin relay centers (ax) symmetry related to high-

spin semiquinone complexes (1150.

29

Figure 1.15. Oxalate bridged ferromagnetic honeycomb structures with bulky counterions

filling void space.

30

The strength of the ferromagnetic interactions in such networks may be at best
moderate and in order to increase the couplings paramagnetic cations may be used in
place of the above diamagnetic main group cations.

A strong magnetic couplings between transition metal cations and various
semiquinone ligands coordinated to them has been reported.”46 Strong interactions of
both high- and low-spin nature are possible, depending on the type of metal cation and
geometry of the coordination. As expected antiferromagnetic interactions are generally
more common then the more desired ferromagnetic ones. The former type of interactions
may still be utilized however for construction of magnets based on the ferrimagnetic
principle. Many ferrimagnets with honeycomb magnetic structures have been indeed
prepared. A net magnetic moment in those structures is achieved by uneven canceling of
different low-spin coupled transition metal cations linked trough oxalate anions (figure
1.15)."7'50 Void space of such structures is filled by bulky counterions which also help to
balance the charge of the solid. Although the highest phase transition temperatures of
those solids were only in the range of 40K, use of paramagnetic (like 1') instead of
diamagnetic linkers offers the possibility of significantly strenghtening the coupling and

hence the ferrimagnetic transition temperatures.

31

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

BIBLIOGRAPHY

Kekule, A. Annalen 1858, 106, 129.

Gomberg, M. Ber. Deutsch. Chem. Ges. 1900, 33, 3150.

Schlenck, W.; Brauns, M. Ber. Deutsch. Chem. Ges. 1915, 48, 661.

Longuet-Higgins, H. C. J. Chem. Phys. 1950, I8, 265; 275; 283.

Dowd, P. J. Am. Chem. Soc. 1966, 88, 2587.

Dowd, P. J. Am. Chem. Soc. 1970, 92, 1066.

Dowd, P.; Chang, W.; Paik, Y. H. J. Am. Chem. Soc. 1986, 108, 7416.

Wright, B. B.; Platz, M. S. J. Am. Chem. Soc. 1983, 105, 628.

Borden, W. T.; Davidson, E. R. J. Am. Chem. Soc. 1977, 99, 4587.

Borden, W. T.; Iwamura, H.; Berson, J. A. Accounts Chem. Res. 1994, 27, 109.

Cramer, C. J .; Smith, B. A. J. Phys. Chem. 1996, 100, 9664.

Fort, R. C.; Getty, S. J.; Hrovat, D. A.; Lahti, P. M.; Borden, W. T. J. Am. Chem.
Soc. 1992, 114, 7549.

Wenthold, P. G.; Hu, J .; Squires, R. R. J. Am. Chem. Soc. 1994, 116, 6961.

Wenthold, P. G.; Kim, J. B.; Lineberger, W. C. J. Am. Chem. Soc. 1997, 119,
1354.

32

(15)

(16)

(17)

(18)

(19)

(20)

(21)

(22)

(23)

(24)

(25)

(26)

Nachtigall, P.; Jordan, K. D. J. Am. Chem. Soc. 1992, 114, 4743.

Clifford, E. P.; Wenthold, P. G.; Lineberger, W. C.; Ellison, G. B.; Wang, C. X.;
Grabowski, J. J.; Vila, F.; Jordan, K. D. J. Chem. Soc-Perkin Trans. 2 1998,
1015.

Matsuda, K.; Iwamura, H. J. Am. Chem. Soc. 1997, 119, 7412.
Ovchinnikov, A. A. Theor. Chim. Acta 1978, 47, 297.

Reynolds, J. H.; Berson, J. A.; Scaiano, J. C.; Berinstain, A. B. J. Am. Chem. Soc.
1992, 114, 5866.

Reynolds, J. H.; Berson, J. A.; Kumashiro, K. K.; Duchamp, J. C.; Zilm, K. W.;
Rubello, A.; Vogel, P. J. Am. Chem. Soc. 1992, 114, 763.

Reynolds, J. H.; Berson, J. A.; Kumashiro, K. K.; Duchamp, J. C.; Zilm, K. W.;
Scaiano, J. C.; Berinstain, A. B.; Rubello, A.; Vogel, P. J. Am. Chem. Soc. 1993,
115, 8073.

Breslow, R.; Chang, H. W.; Hill, R.; Wasserman, E. J. Am. Chem. Soc. 1967, 89,
11 12.

Breslow, R. Pure App]. Chem. 1982, 54, 927.

Miller, J. 8.; Dixon, D. A.; Calabrese, J. C.; Vazquez, C.; Krusic, P. J .; Ward, M.
D.; Wasserman, E.; Harlow, R. L. J. Am. Chem. Soc. 1990, 112, 381.

van Willigen, H.; van Broekhoven, J. A. M.; de Boer, E. Mo]. Phys. 1967, 12,
533.

van Broekhoven, J. A. M.; van Willigen, H.; de Boer, E. M0]. Phys. 1968, 15,
101.

33

(27)

(28)

(29)

(30)

(31)

(32)

(33)

(34)

(35)

(36)

(37)

(38)

(39)

(40)

Broekhoven, J. A. M.; van Willigen, H.; de Boer, E. Mo]. Phys. 1971, 20, 993-
997.

Ebata, K.; Setaka, W.; Inoue, T.; Kabuto, C.; Kira, M.; Sakurai, H. J. Am. Chem.
Soc. 1998, 120, 1335.

Glasbeek, M.; van Voorst, J. D. W.; Hoijtink, G. J. J. Chem. Phys. 1966, 45,
1852.

Krusic, P. J.; Wasserman, E. J. Am. Chem. Soc. 1991, 113, 2322.

Snyder, G. J.; Dougherty, D. A. J. Am. Chem. Soc. 1989, 111, 3927.

Du, P.; Hrovat, D. A.; Borden, W. T. J. Am. Chem. Soc. 1989, 111, 3773.

Kawata, S.; Kitagawa, S.; Kondo, M.; Katada, M. Synth. Met. 1995, 71, 1917.

Kawata, S.; Kitagawa, S.; Kumagai, H.; Kudo, C.; Karnesaki, H.; Ishiyama, T.;
Suzuki, R.; Kondo, M.; Katada, M. Inorg. Chem. 1996, 35, 4449.

Robl, C.; Weiss, A. Z.Naturforsch. (B) 1986, 41, 1337.

Robl, C. Mater. Res. Bull. 1987, 22, 1483.

Robl, C. Z. Naturforsch. (B) 1987, 42, 972.

Robl, C.; Weiss, A. Mater. Res. Bull. 1987, 22, 497.

Robl, C.; Kuhs, W. F. J. Solid State Chem. 1989, 79, 46.

Cueto, S.; Straumann, H. P.; Rys, P.; Petter, W.; Gramlich, V.; Rys, F. S. Acta
Crystallogr. Sect. C-Cryst. Struct. Commun. 1992, 48, 458.

34

(41)

(42)

(43)

(44)

(45)

(46)

(47)

(43)

(49)

(50)

Decurtins, S.; Schmalle, H. W.; Zheng, L. M.; Ensling, J. Inorg. Chim. Acta 1996,
244, 165.

Weiss, A.; Riegler, E.; Robl, C. Z.Naturforsch. (B) 1986, 41, 1501.

Abraharns, B. F.; Coleiro, J .; Hoskins, B. F .; Robson, R. Chem. Commun. 1996,
603.

Lange, C. W.; Conklin, B. J.; Pierpont, C. G. Inorg. Chem. 1994, 33, 1276.

Pierpont, C. G.; Lange, C. W. In Progress in Inorganic Chemistry, 1994; Vol. 41;
pp 331-442.

Dei, A.; Gatteschi, D. Inorg. Chim. Acta 1992, 200, 813.

Tamaki, H.; Zhong, Z. J .; Matsumoto, N.; Kida, S.; Koikawa, M.; Achiwa, N.;
Hashimoto, Y.; Okawa, H. J. Am. Chem. Soc. 1992, 114, 6974.

Okawa, H.; Matsumoto, N.; Tamaki, H.; Ohba, M. 1993, 232, 617.

Ohba, M.; Tamaki, H.; Matsumoto, N.; Okawa, H. Inorg. Chem. 1993, 32, 538.

Decurtins, S.; Schmalle, H. W.; Oswald, H. R.; Linden, A.; Ensling, J .; Gutlich,
P.; Hauser, A. Inorg. Chim. Acta 1994, 216, 65.

35

Chapter 2

COMPUTATIONAL STUDY OF SINGLET-TRIPLET GAPS IN DERIVATIVES OF

12'

2.1 . Introduction

Preliminary considerations concerning the electronic structure of 12' were based
on very simple Hiickel theory. In order to look at this topic in a more quantitative way
extensive ab—initio computations had to follow, both on the parent 12' and on derivatives
of this dianion. The most straightforward modifications of 12' are (a) replacement of
hydrogen atoms with other substituents and (b) coordination of the semidione units to
different metal cations. An understanding of the relationship between electronic
properties and structural modifications of 12' should enable the design of derivatives,
which are most appropriate for assembly of molecular magnets. One of the most desirable
properties of such derivatives from the ‘magnetic’ point of View would be a strong
preference for a triplet ground state. The majority of our theoretical studies therefore
focused on correlation of structural modifications with the magnitude and sign of AE(S-T).
Since the value of AE(S-T) is directly linked to the degree of degeneracy of the frontier
orbitals (assuming that their spatial overlap remains relatively unperturbed), an attempt
was made to correlate these two factors.

Calculations were planned to be performed first at the CAS(2, 2) level of theory
also known as TCSCF (Two Configuration Self-Consistent Field), for singlet states and
ROHF (Restricted Open-shell Hartree-Fock) for triplets. To gain insight into influence of

correlation between 1: electrons they were also repeated at the CAS(14, 12) (fourteen

36

electrons in twelve orbitals) level. The set of calculated species included ‘naked’ dianion
12' and its complexes 1M2 with metal cations of increasing charge (M=Li+, Bey, A13 +). In
the latter two systems, fluoride anions were added to balance the charge and keep the
overall complexes neutral. The lithium complex was chosen to assess the influence of
substituents such as fluorine (F41Liz) and hydroxy groups ((OH)41L12). CAS(2, 2) and

CAS(6, 6) calculations on model DMCB were also performed.

.0 0.
Fl-B:e2' © :Bezi-F‘
‘o o’
lLiz 1(BCF)2
H H
I \
F F o o
F' ,0 I 0. .0 I 0.
:‘Aq’a: 21.34 *0: Q 11* +1.1 @ 11*
F' ‘o l o’ ‘o I o'
F F o o
\ I
H H
1(AIF2)2 F41Liz (OH)41Li2
DMCB

37

Theoretical Background. Hartree-Fock (HF), the most basic computational ab-
initio method determines the best one-determinantal trial wave function of the molecule
kI”=’(I)o

where:

(91(1) (91(2) (111(0)

(132(1) (192(2) (02(0)

(P0411040

(911(1) (Pn(2) (Pom)

 

 

and (pn symbolizes spin-orbital.

The HF solution neglects electron correlation effects (EC). Those effects,
although accounting for only ~1% of the total energy of a molecule may play an
important role in some chemical problems (especially those involving bond breaking
processes or diradicaloid species). To describe properly the exact wave function one

should include also determinants of states excited with respect to (130,.
W=M¢0+2 aid),

Inclusion of all possible excited states would fully account for correlation of all electrons
and the exactness of the found solution would be limited only by the size of the basis set
used. The computational cost of determining such a firnction is, however, very high, even

for moderate-size molecular systems. To shorten the time of calculation the Complete

38

Active Space Self-Consistent Field (CASSCF or CAS) method is often employed. This
method takes into account all possible excitations only in the selected (chemically
relevant) orbital subspace. CASSCF calculations optimize both It, and the form of the
orbitals used to define both (Do and the selected excited configurations. The CASSCF
method proved to be reliable for prediction of sign and magnitude of the singlet-triplet
gaps of many of non-Kekfile hydrocarbons, and therefore was selected for use in our

calculations.

 

 

 

 

 

 

4-1- —1— 4—
— +— —-1-

1:1

_ .1.
.1... 4—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

CDO (D1 (D2 (D3 (D4 (D5

Figure 2.1. All possible determinants in CAS(2,2) active space

In the simplest two orbital scheme for two active electrons (CAS(2, 2)) six
possible determinants may be written (figure 2.1). The first two determinants, (Do and (1)1,
represent a singlet state while (D; and (D3 form triplet configurations with corresponding
values of 1 and —1 , respectively, for M5, the electron spin quantum number. Linear
combinations of (1)4 and (1)5 lead to one singlet and one triplet (Ms=0) configuration. The
only difference between the three triplet configurations is the alignment of their magnetic
moments in the presence of an external magnetic field, so calculation of only one is
required to find out its electronic energy. To describe the singlet state one should use
three singlet configurations. However, those three configurations are not unique. In

particular, calculations with only (Do and (I); (which is equivalent to the General Valence

39

Bond method) give exactly the same energy as is obtained from calculations with the
three configurations.

This two electron, two-orbital level of theory, sometimes considered to be the
minimal acceptable, provides relatively consistent results at modest computational cost.l
It can not, however, be expected to accurately predict the exact energies of the S-T gaps.
Correlation with other electrons of the molecule may substantially change the magnitude
or even the sign of the calculated values. To account for correlation at least among the
most important electrons for conjugated diradicals, an active space including all it
electrons and all basic it orbitals is required. Although such a level of calculations
describes energy differences rather accurately, further improvement may be achieved by
inclusion of corrections for ‘dynamic’ correlation of the ‘active’ and inactive (in the case
of It diradicals, 0 bonds) electrons. The most common method of calculating this
correlation is to use second-order Moller-Plesset (MP2) perturbation theory for the CAS
wavefunction (CASPT2N method).2 Although the resulting corrections are minimal, they
are necessary for correct prediction of the orderings of closely spaced states (for example

exited states for analysis of electronic spectra).

40

2.2. Results

Table 2.1 summarizes the total energies of the lowest lying singlet and triplet
states for all the species. Reported energies are for geometries optimized in Dzh symmetry
at the indicated computational level, except for the CAS(14,12)/6-31+G* energies which
were computed using the CAS(14,12)/6-31G* geometries. In most cases calculations
found lAg and 3B3u to be the lowest lying singlet and triplet electronic states respectively.
The only exception was the naked dianion 12' whose 3B2u (|. . .2b 3g23b ,u'3b 381)) state was
found to be 0.2 kcal/mol below 3B3u (l. . .2b 3323b1u'2b281». To keep all the comparisons
consistent, however, all orbital populations and geometry analyses for 12' and congeners
(Tables 3 and 4) were based on the 3B3u state. Table 2.1 also contains values of S-T gaps
(AE(S-T)) expressed in kcal/mol along with the ZPE corrections calculated at the
CAS(2,2)/6-31G* level. Figure 5 graphically depicts the (AE(S-T)) gaps as a function of
AE(B-A), the difference between the one-electron energies of the singly occupied frontier
orbitals B and A in the ROHF triplets. The AE(B-A) values and occupancies of A and B in
the TCSCF (CAS(2,2)) singlet wavefunctions are presented in Table 2.2.

Table 2.3 and 2.4 lists the occupancies of the It (active) orbitals in the CAS(14,12)
wavefunctions for the lAg and 3B3u states, together with the CAS(2,3) results for lAg.
Table 2.5 contains selected bondlengths; cartesian coordinates of the unique atoms and

vibrational frequencies in all calculated complexes may be found in the appendix A.

41

Table 2.1. Total energies of the investigated complexes in their lowest singlet (lAg) and
triplet (3 B 3“) electronic states at all levels of the performed calculations.

 

 

 

 

Level of theory IAg 3 B 3., AE(S-T)
/6-3 1G*(d) Hartree Hartree Kcal/mol
12'
CAS(2,2) -604 .55602 -604 .54691 -5.7
ZPECASQQ) 0 . 10740 0 . 10835 -0.6
CAS(2,3)/(2,2) -604 .56193 -604 .54691 -9.4
CAS(14,12) -604 .67936 -604 .65936 -12.6
-604 .65955* -12.4
CAS(14,12) -604 .73282 -604 .70526 -17.3
/6-31+G*(d)
lLiz
CAS(2,2) -619 .72305 -619 .72589 1.8
ZPEcasm) 0 .l 1678 0 .1 1699 -0.1
CAS(2,3)/(2,2) -6l9 .72569 -619 .72589 0.1
CAS(14,12) -619 .82437 -619 .82360 -0.5
CAS(14,12) —619.84374 -619 .84055 -2.0
/6-31+G*(d)
*33211

42

Table 2.1 (cont). Total energies of the investigated complexes in their singlet ('Ag) and
tripet (3 B 3“) electronic states at all levels of the performed calculations.

 

 

 

 

Level of theory 1Ag 3 B 3.. AE(3-1~)
/6-3 1G*(d) Hartree Hartree Kcal/mol
1(BeF)2
CAS(2,2) -833 .19055 -833 .19756 4.4
ZPECASQJ) O . 12690 O . 12694 0.0
CAS(2,3)/(2,2) -833 .19243 -833 .19756 3.2
CAS(14,12) -833 .28259 -833 .28859 3.8
CAS(14,12) -833 .30793 -833 .31300 3.2
/6-31+G*(d)
1(A1F2)2
CAS(2,2) -1486.7613l -1486 .76908 4.9
ZPECASQJ) 0.12879 0.12877 0.0
CAS(2,3)/(2,2) -1486 .76282 -l486 .76908 3.9
CAS(14,12) -l486 .85099 -1486 .85871 4.8
CAS(14,12) -1486 .88976 -l486 .89663 4.3
/6-31+G*(d)
F41Li2

CAS(2,2) -1015.11315 -1015.11899 3.7
ZPEcasag) 0 .08402 0 .08419 -0.1
CAS(2,3)/(2,2) -1015 .11443 -1015 .11899 2.9
CAS(14,12) -1015.20809 -1015.21185 2.4
CAS(14,12) -1015.2396l -1015.24l96 1.5

/6-31+G*(d)

 

43

Table 2.1 (cont). Total energies of the investigated complexes in their singlet (1A8) and
tripet (3 B 3..) electronic states at all levels of the performed calculations.

 

 

 

Level of theory 'Ag 3B3u AE(s-n
/6-31G*(d) Hartree Hartree Kcal/mol
(OH)41Liz
CAS(2,2) -9l9.10912 -9l9.11529 3.9
ZPECASQJ) 0.13544 0.13521 0.1
CAS(2,3)/(2,2) -919 .10971 -919 .11529 3.5
CAS(14,12) -919.l9l30 -919.19899 4.8
CAS(14,12) -919 .22139 -919 .22854 4.5
/6-31+G*(d)
DMCB

CAS(2,2) -230 .538275 -230 .54967 7.2
ZPECASQJ) 0 .09902 0 .09846 0.4
CAS(6,6) -230 .587662 -230 .61921 19.8
CAS(6,6) -230 .59436 -230 .62581 17.7
/6-31G*(d)

 

44

Table 2.2. Occupancies of orbitals A and B in the TCSCF singlet wavefunctions and
values of AE(B-A) in the ROHF calculations

 

AE(B-A) OCCUP.

ROHF TCSCF

 

 

B A

12' 0.0453 1.59 0.41
1Li2 0.0209 1.34 0.66
1(BeF)2 0.0091 1.19 0.81
1(A1F2)2 0.0041 1.11 0.89
I=.,1Li2 0.0043 1.08 0.92

(OH)41Li2 -0.0112 0.78 1.22

DMCB 0.0338 1.53 0.47

 

45

Table 2.3. Occupancies of the active orbitals of investigated complexes in the singlet lAg
CAS(14,12) and CAS(2,3) wavefunctions.

 

 

 

[Ag

Orbital 12' lLiz 1(BeF)2 1(A1F2)2 F41Li2 (OH)41Li2
1(1),.) 1.98 1.98 1.99 1.99 1.98 1.99
II(bzg) 1.96 1.98 1.98 1.99 1.98 1.98
111(b3g) 1.97 1.97 1.98 1.99 1.98 1.98
IV (a,) 1.95 1.97 1.98 1.99 1.97 1.98
V (but) 1.92 1.91 1.91 1.92 1.92 1.93
VI(b3g) 1.94 1.92 1.92 1.92 1.93 1.94
VII B (b,,,) 1.66 1.41 1.25‘ 1.17‘ 1.14‘ 0.82‘

1.68 1.56 1.44 1.26 1.37 1.20
VIIIA(b2g) 0.28 0.58 0.74‘ 0.82 0.86‘ 1.18"

0.17 0.40 0.53 0.73 0.62 0.80
Ix (b3g) 0.06 0.02 0.01‘ 0.01 0.01‘ 0.00"

0.29 0.17 0.15 0.12 0.12 0.09
x (bru) 0.05 0.03 0.03 0.02 0.03 0.03
x1 (a..) 0.05 0.07 0.07 0.06 0.06 0.06
XII (63,) 0.03 0.03 0.02 0.01 0.03 0.02
*CAS(2,3)

46

Table 2.4. Occupancies of the active orbitals of investigated complexes in the triplet 3B3“
CAS(14,12) wavefunctions.

3Bju

 

Orbital 12' 1Li2 1(BeF)2 1(A1F2)2 F41Li2 (OH)4lLi2

 

1(1),.) 1.98 1.99 1.99 1.99 1.99 1.99
110228) 1.98 1.99 1.99 1.99 1.99 1.99
111(b3g) 1.96 1.98 1.98 1.98 1.98 1.98
IV (11,.) 1.96 1.98 1.98 1.98 1.98 1.98
V(b1u) 1.91 1.92 1.92 1.93 1.93 1.94
VI(b3g) 1.91 1.90 1.90 1.90 1.93 1.93
VIIB(b/a) 1.02 1.01 1.00 1.00 1.00 1.00
VIIIA(bzg)1.01 1.01 1.01 1.00 1.01 1.01
IX(b3g) 0.11 0.11 0.11 0.11 0.08 0.08
X(b/u) 0.04 0.03 0.03 0.03 0.03 0.03
x1(a..) 0.07 0.07 0.07 0.07 0.06 0.06

XII (b3g) 0.03 0.02 0.02 0.02 0.02 0.02

 

47

Table 2.5. Selected bond lengths (in A) of 12' and its complexes optimized at the

CAS(14,12)/6316* level.

 

 

 

3(3B3u) am.) M3133.) b(‘Ag) c683.) c(‘Aa 46133..) d(‘Ag)
12’ 1.436 1.487 1.501 1.457 1.356 1.383 1.252 1.240
1Li2 1.431 1.451 1.473 1.461 1.357 1.357 1.263 1.257
1(BeF)2 1.423 1.437 1.457 1.458 1.358 1.351 1.269 1.264
1(AlF2)2 1.423 1.433 1.457 1.461 1.358 1.349 1.271 1.266
F41Li2 1.439 1.441 1.471 1.478 1.366 1.356 1.263 1.261
(OH)41Li2 1.433 1.441 1.474 1.474 1.351 1.345 1.259 1.256

 

48

 

 

 

 

 

 

 

 

10 '
1(A1F2)2
5 - ‘ 1(BeF)2
-0.02 0.06
-5 . AE(B-A)
(hartree)
-10 «
-1 -
5 AE(S-T) “an
(kcal/mol)
-20
b)
10 '
(011)41L12
-0.02 0 02 0.04 0.06
-5 - ' Alias-A)
(hartree)
-10 '
-1 "
5 AE(s-T)
(kcal/mol)
-20 '

 

Figure 2.2. S-T gap calculated for a) 12' complexed by different metal cations b) 1Li2
with different substituents R, vs. AE(B-A) at
(—o—) CAS(2,2)/6-31G*(d)
(+) CAS(14,12)/6-31G*(d),
(«m-om) CAS(14,12)/6-31+G(d) level of the theory.
( -------- a.--) Zero point corrections at CAS(2,2) level.

49

2.3. Discussion

Series (a): Effect of metal ifion coordipation. Electrostatic interactions of
electrons with metal cations should lower the energies of all occupied MOs. Orbitals
concentrated close to the cations should, however, be affected more strongly then those
further away. With substantial density on the methylene carbons, orbital B (3b/u) does not
interact with the cations as strongly as orbital A (2b 28), which is distributed only on the
semidione units (fig 1). These disparate interactions lead to a decrease in the gap between
A and B (AE(B.A)) as the cation charge increases. The more charged the cations the
stronger an effect is observed; AE(A-B) drops from 0.0453 through 0.0209 and 0.0091, to
0.0041 Hartree for 12', 1Li2, 1(BeF)2 and 1(A1F2)2 respectively. The values of AE(A-B) are
obtained at the single configuration ROHF level of theory, so their physical relevance to
the CAS results is approximate. Nonetheless, a correlation between AE(A-B) values and S-
T gaps is expected, and indeed, an increasing preference for the triplet state tracks the
shrinking interorbital gap. At the CAS(2,2) level of theory the ground state of 12' is a
singlet lying 5.7 kcal/mol below the triplet; for 1L12 the singlet already lies above the
triplet by 1.8 kcal/mol, while for 1(BeF)2 and 1(A1F2)2 this value increases to 4.4 and 4.9
kcal/mol, respectively. Extrapolation of the curve obtained leads to a predicted triplet
' preference of about 5 kcal/mol for AE(A-B)=0. At the current level of theory, this value is
probably close to the maximum achievable by tuning of the relative energies of the

orbitals.

Series (b): Effect of substitution. Unlike B, orbital A has no coefficients on the

exocyclic methylene carbons in 12'. Replacement of the hydrogen atoms with other

50

substituents should thus affect the energy of B much more than that of A. Since orbital B
lies lower than A, the desired decrease of the A-B gap should be achieved by substitution
of 12' with electron donating groups. Figure 5b shows the effect of such structural
modifications. Fluorine, a weak rt donor, already has a substantial effect, lowering AE(B-A)
from 0.0209 for H-substituted 1Liz to 0.0043 for F41Li2. The effect of the more strongly
donating hydroxy groups is powerful enough to reverse the HOMO-LUMO orbital order
(AE(A-B)= -0.0112). Although in both cases a strong preference for the triplet follows, the
AE(5-T)~AE(A-B) correlation can not be directly related to that obtained in series (a). For
example, complexes 1(A1F2)2 and F41Liz have almost the same values for AE(A-B) (0.0041
vs 0.0043 Hartree) but somewhat different AE(S-T) values (4.9 vs 3.7 kcal/mol). Inspection
of the MO coefficients confirms that replacement of hydrogens with heavy atoms causes
more extensive delocalization of orbital B and lowers its spatial overlap with A. This in
turn weakens the electrostatic interactions between the frontier orbitals and slightly
lowers the triplet preference. The HOMO-LUMO gap for the model DMCB (0.0338
hartree) is larger than for lLiz. Nonetheless, its triplet preference calculated at the
CAS(2,2) level of theory is the highest of all the investigated species (7.2 kcal/mol). In
this case the electrostatic interactions between the frontier electrons must by magnified

by the small size and resulting strong spatial overlap of the two SOMOs.

Correlation effects. Examination of results obtained at the CAS(14,12) level of
theory shows that the correlation energy corrections for the calculated species depend on
their structures. The more charged are the cations coordinated to 12', the lower is the

magnitude of (ECASQQYECAgualm) for both singlet and triplet states. Apparently, with

51

increasing charge of the coordinated cation, the bonding electrons of the 11: system
localize more on the carbonyl oxygens while the antibonding frontier electrons localize
on the carbons, weakening their mutual interactions. Coordination also increases the gap
between bonding and antibonding orbitals leading to lower polarizability (i.e. higher
excitation energies) in the It system. Stabilization of the singlet drops faster than for the
triplet. Thus, the improvement in correlation description favors singlet versus triplet for
12', the first member of series (a), by 6.9 kcal/mol but for the last member, 1(AlF2)2, by
only 0.1 kcal/mol. A similar tendency is observed for the substituent effects in series (b)
(2.3 kcal for 1Li2 vs. -0.9 for (OH)4lLi2). One of the factors contributing to this
covariation must be the fact that leveling of the HOMO-LUMO gap results in singlets
having more diradical and less ionic character, but it has no analogous effect on triplets.
Since correlation generally favors more ionic wavefunctions,2 the relative stabilization by
correlation of the singlet vs. the triplet by correlation thus decreases.

Especially impressive is the relative stabilization of the triplet state in the case of
the model DMCB, where the triplet preference rises from 7.2 for CAS(2,2) to 19.8
kcal/mol for CAS(6,6). Additional inclusion of all fi-SD and o-S, n-SD CI from the CAS
multireference wavefunction were reported to yield preferences of 20.2 and 18.2
kcal/mol, respectively.3 The authors attributed this large correction to a higher degree of
delocalization of the frontier electrons in the triplet compared to the singlet and
consequently their stronger interaction with other electrons. The second value reflects the
correlation of TI with o electrons, which usually provides extra stabilization for singlets.
Unfortunately, the species described herein were too large for such studies via CASPTZN

or MR SD CI methods with the software available to us.

52

Examination of the partial occupancies of the active orbitals in the CAS(14,12)
wavefirnctions of both singlet and triplet states (tables 2.3 and 2.4), shows high
occupation of 3b 38, especially for 12'. For this dianion, a triplet state of different
symmetry, 3 B 2,, (|. . .2b 3g23b1u'3b 381)), is slightly lower in energy than the 3B3u
(|. . .2b3gz3b1u12bzg'» state. In this doubly charged situation, the low lying 3b3g orbital
could be electrostatically preferred over A (2bzg) due to the more disjoint relationship of
B and 3b 3g, than of B and A. The less complete CAS(2,3) (B, A, 3b3g and 2 frontier
electrons) calculations, however, show rather low populations of 3b 38 indicating that in
the hill 11: space, it is mostly excitations from orbitals other than A and B that fill this
orbital.

To more accurately describe the anionic character of the organic core 12' of the
species studied here, single point CAS(14,12) calculations with the 6-31+G* basis set
were performed on the geometries optimized at the CAS(14,12)/6-31G‘1‘ level. Addition
of the diffuse "+" sets of s and p functions on heavy atoms generally favors singlets. The
corrections, however, are not dramatic (see table 1) and as expected, they decrease with

tighter coordination of the metal ions.

Molecular Geometry. Potential degeneracy in non-disjoint frontier orbitals does
not necessarily assure that a molecule will have a triplet ground state. The degeneracy
may be lifted by Jahn-Teller distortion. For example, lowering of the symmetry from
threefold or higher to pseudo D2,, stabilizes singlet states of 4M electron ‘antiaromatic’

annulenes. L4 In the present systems evidence for such distortions was sought at the

CAS(2,2) level by vibrational analysis of the Dzh optimized structures.

53

All frequency values for series (a) (12', lLiz, 1(BeF)2 and 1(AlF 2);) turned out to
be real in both singlet and triplet states. Values for some modes of bending out of plane
were however very low, showing that the planar geometry is a rather shallow minimum.
This finding is unsurprising, in light of the nonplanar geometries we have found for the
neutral tetraketone, both computationally and experimentally (with Ph substituents).

In series b, for triplet state F41Li2 all the frequencies are real, while there is one
(16 cm") imaginary mode for the singlet state. For (OH)41Li2, there are three imaginary
frequencies for each state. Two of them in each state (426, 425 cm'l (s) and 458, 457 cm'l
(1)) belong to combinations of out of plane torsions of the hydrogens of the OH groups.
The third ones (18 cm'l(s) and 23 cm"(t)) similar like the imaginary frequency of singlet
state F41Li2, represent boat-like out of plane bending of the molecular frame. Presumably
these distortions arise from the steric interactions introduced when H was replaced by the
larger F and OH groups.

While OH torsions may easily be attributed to the artificially high D 2;, symmetry
chosen for the molecules, bending of the whole molecule must reflect an intrinsic
characteristic. Does it indicate a tendency to Jahn-Teller distortions or just high flexibility
of the molecular frame? Optimization of singlet F41Li2 in the boat-like C 2,, geometry
indicated by imaginary vibration led to energy being less then 15 cal/mol lower then in
the D 2;, one. At the same time occupation numbers of the frontier orbitals change only
slightly (from 0.92(A), 1.08(B) to 0.91(A), 1.09(B)), suggesting that the distortion that
may be caused more by geometry difference rather then an energetic splitting of the
singlet frontier orbitals. Whatever is nature of this distortion the magnitude is negligible

in comparison with the value of the S-T gap.

54

To probe further the flexibility of the molecular frame of lLiz, single point
CAS(14,12) calculations on the triplet at the optimal geometry of the singlet (and vice
versa) were performed. The energy of the triplet at the singlet geometry was only 1.6
kcal/mol higher then at its own optimum, but for the less forgiving singlet state molecule
The analogous increase was 2.8 kcal/mol.

Based on the bond lengths of both singlet and triplet state complexes (table 2.5), it
appears that the molecular frame is best described as two semidione units connected
through vinylidene bridges. Shortening of bond a and elongation of d with increasing
charge of the coordinated cation may be understood as arising from increasing
recruitment of the unpaired electron density to the semidione fragment, wherein the
SOMOs are C-C bonding and C-0 antibonding. Besides the previously discussed
deviation from planarity of F41Li2, some bond length differences were found between
singlet and triplet geometries. For singlet structures bonds at and b are longer while c and
d are shorter than for the corresponding triplets. Those variations can be understood in

terms of the calculated shifts in electron occupancy from orbital B to A.

2.4 Computational details

All calculations with the 6-31G"‘5 basis set were performed using the
GamessD986 program running on a SGI/CRAY Origin 2000 computer. Geometries of all
complexes in both electronic states were fully optimized in D 2;, symmetry at each level of
calculation. The lowest energy orbitals with the symmetries listed in table 2.3 and 2.4

were selected as the active space. Due to the very low electron occupation of the

55

LUMO+1 in the triplet CAS(2,3), calculations were automatically terminated by the
program. Since GamessD98 can not calculate analytical derivatives at the MCSCF level,
vibrational frequencies were calculated only at the CAS(2,2) level with the use of the
GVB method for singlet and ROHF for triplet (these methods gave the same single point
energies as CAS(2,2)). The symmetry adapted CAS(14,12) active space generates 70 840
configurations for the triplet state and 42 756 for the singlet state. Multiconfigurational
Gamess calculations that were attempted with the 6-31+G* basis set failed to converge.
They were, however, successfully completed by use of the Molpro package798 after
verification that this software yielded identical results to Gamess D98 for the 6-31G*

basis set.

56

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

BIBLIOGRAPHY

Borden, W. T. in Diradicals; Borden, W. T., Ed.; Wiley: New York, 1982.

Borden, W. T.; Davidson, E. R. Accounts Chem. Res. 1996, 29, 67.

Du, P.; Hrovat, D. A.; Borden, W. T. J. Am. Chem. Soc. 1989, 111, 3773.

Breslow, R. Pure App]. Chem. 1982, 54, 927.

Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213.

Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.; Gordon, M. S.;
Jensen, J. H.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.; Su, S. J .; Windus, T. L.;
Dupuis, M.; Montgomery, J. A. J. Comput. Chem. 1993, I4, 1347.

Werner, H. J .; Knowles, P. J. J. Chem. Phys. 1985, 82, 5053.

Knowles, P. J .; Werner, H. J. Chem. Phys. Lett. 1985, 115, 259.

57

Chapter 3

SYNTHESIS OF THE PRECURSORS OF 1' AND 12'

3.1 . Introduction

Before starting synthetic work, the crucial question ‘What to synthesize’ had to be
answered. The main goal of the experimental studies is to investigate properties of the
derivatives of 12'. Two hypothetical modes of generation of dianion 12' might be

envisioned. One is reduction of the tetrone l, the other deprotonation of the ‘acid ‘ 1H2

(Figure 3.1).
R R R R
o 0 2e- 0 0 _2H, 0 OH
_..._.... *__
o o o a o o o
R R R R R R
1 12 1H2

Figure 3.1. Two hypothetical modes of generation of 12'

Since reduction of 1 gives access cleanly to both radical anion and dianion, this tetrone
was given higher priority then the acid as the main target of the synthetic work. The other
factor which also had to be determined before beginning the synthesis was selection of
appropriate substituents on the methylene group. Both anion, dianion, and their
precursors Should possess reasonable stability so they Can be handled and investigated

with relative ease.

58

P1120012

CSzlAlCla

a) Direct condensation of phenols and derivatives of benzophenones

Ar Ar
Ar 2CHOH
—>
R R H +
H R R
H

b) Dehydrogenation of diarylmethyl phenols

ng Ar OHAr
Ar2CO
R R + '
ng H leO R R
H

c) Dehydration of triarylmethyl alcohols

Ar Ar
-2[H1
——9
R R
Ar Ar
-H 20
—.
R R

Figure 3.2. Three main methods of synthesizing quinone methides

59

A very common method of stabilizing radical intermediates is to sterically protect then
with phenyl rings. Indeed the first known radicals and diradicals were stabilized that
way.1 Phenyl rings are however not only bulky spectators. Their conjugated 1t systems
may transfer electronic perturbations induced by various substituents and electronically
tune the tetrone unit. Many synthons necessary for synthesis of such ring substituted
compounds are readily available. Phenyl rings also have beneficial effects on the
solubility of compounds in non-polar solvents. Considering all those practical factors, a
Specific precursor 2 (1, R=Ph) and 2H2 (1H2, R=Ph) and their derivatives were selected
as initial synthetic targets.

Preparation of 2 has no precedent. There are, however, some records describing
preparation of related fuchsones, diphenyl substituted derivatives of quinone methides.2
Such compounds were investigated as early as the end of the XIX century and attracted
some attention due to their potential application as dyes. Preparation of fuchsones can be
classified into three main methods (Figure 3.2): (a) Condensation of phenols 'with
benzophenone derivatives; (b) alkylation followed by dehydrogenation and (c)
dehydration of triarylmethyl alcohols. In this work all three methods were tried in an

attempt to synthesize 2.

60

3.2. Results

Since the first strategy is the most straightforward, investigations were started
with the reaction between dihydroxybenzoquinone and benzophenone.
No reaction however was observed either in alkaline (3Na2 as a substrate) nor acidic

conditions (H+, TiCl4, BF3-THF as catalysts).

P Ph P P"
H0 0 -2 H20 0 0
Ph Ph
3H2 2

Figure 3.3. Attempt to prepare 2 from 3H2 and benzophenone

Apparently, the highly conjugated carbonyl group of the benzophenone was not active
enough toward weakly nucleophilic 32'(pK2=5.2). In order to increase the nucleophilicity
of the tetraoxalane unit, the more electron rich tetrahydroxybenzene (3H4) was prepared

by reduction of 3H2 (Figure 3.4).3

O OH sn/HClaq HO OH
HO O HO OH
3H2 3H4

Figure 3.4. Preparation of 1,2,4,5-tetrahydroxy benzene from dihydroxybenzoquinone

61

Replacement of 3H2 by 3114 did not bring any breakthrough and unreacted benzophenone
was recovered from reaction mixtures under all conditions investigated. The more
reactive benzophenone equivalent dichlorodiphenylmethane was then prepared from

benzene and tetrachloromethane following the procedure of Gomberg (Figure 3.5).4

CI Cl
"" O
CCI4/AIC13

Figure 3.5. Preparation of dichloro-diphenylmethane

Although 3H2 reacted with diphenyldichloromethane, the reaction did not produce 2.
Basic catalysts like N33 or pyridine (neat or in DMF or DMSO) led to tars or solids that
were insoluble in common solvents. Acidic conditions (heating in HZSOalAcOH, AcOH,
CF3COOH or just a neat mixture of reagents) led to evolution of HCl and formation of
benzophenone as a main product. It has been concluded that even if 2 was formed as an
intermediate, it was too delicate to survive such strenuous conditions and reacted firrther
to form the thermodynamically more stable products. Apparently milder conditions were
required for preparation of 2. Reactions between 3114 and dichlorodiphenylmethane were
also investigated. When both compounds were heated neat a in 1:2 stoichiometry, HCl
evolved and a yellowish solid identified as 5 was produced (Figure 3.6). Given the above

failures, the one-step strategy for the synthesis of 2 had to be discarded. A new

62

preparation was attempted by the substitution-dehydrogenation sequence presented on

HO OH thccu Q o o):
HODEOH “”01 :(0:(5>:0 0

Figure 3.6. Reaction between tetrahydroxybenzene and dichlorodiphenylmethane

H
P Ph P Ph
Ph 6 Ph
0 OH I 0 0“ -4[H] o 0
II -—> —*
Ho 0 -H* Ho 0 o 0
Ph Ph
H Ph Ph
2114

2

Figure 3.7.

Figure 3.7. Preparation of 2 by a substitution-dehydrogenation sequence

The attempt to prepare 2H4 turned out to be successful. When 3H2 and benzhydrol where
refluxed in hot acetic acid with a few drops of H2804, orange needles of a new compound
separated afier cooling. Spectroscopic analysis determined its structure to be 2H4. After
some optimization this reaction turned out to give a high yield (96%) of product that was
convenient to isolate and purify. Dehydrogenation turned out to be a much more difficult
task. Out of the many oxidizing agents which were tried, only high-potential quinones
(DDQ and o-chloranil) proved to be effective. Metal salts (Pb(OAc)4, K3Fe(CN)6,

Ce(NH4)2(NO3)6 and oxides (AgZO, activated MnOz, PbOz) did not react or formed

63

insoluble and unreactive salts of 2114. Boiling of 2H4 with DDQ in toluene led to a
mixture in which 2 and 2H2 (Figure 3.8) could be identified with other side products. The
more expensive and hard to handle o-Chloranil was less effective, while p-Chloranil did

not react at all.

P Ph P Ph P Ph
0 OH 0 OH (3 o
DDQ +
—>
Ho 0 o o o 0
Ph Ph Ph Ph Ph Ph
2Ha 2H2 2

Figure 3.8. Dehydrogenation of 2114 to 2H2 and 2

Separation and purification of products was an unexpectedly difficult task. In addition to
the fact that 2 is a minor product of the reaction, most of the compounds present in the
mixture could be donors or acceptors (or both!) in donor-acceptor charge-transfer
complexes. This situation led to a complicated mixture of such complexes from which
separation of pure components could not be achieved by such standard techniques as
chromatography. Additionally compound 2 proved too delicate to be sublimed or handled
with protic or electron donating solvents. In order to separate the mixture special care had
to be taken that no DDQ (a particularly effective acceptor) remained in solution by
driving the reaction to completion. Precipitation from dichloromethane solution of the

reaction mixture with ether and recrystallization from hot acetonitrile finally achieved

64

isolation of 2. The structure of the isolated compound was confirmed by X-ray
crystallography (Chapter 8)

A method to isolate 2H2 was also developed. After reaction of DDQ with excess
of 2H4, 2H2 was isolated from remaining 2H4 by extraction with Acetic acid and then
purified by recrystallization from cyclohexane.

The Simplicity of this two-step synthetic procedure, once developed, encouraged
us to apply it for preparation of phenyl ring substituted derivatives of the parent tetrone 2
(figure 3.9). Since ab-initio calculations indicated that electrondonating substituents
should stabilize the triplet state of the dianion diradical relative to the singlet state, we
focused on such derivatives. Condensation of commercially available 4,4’-
dimethoxybenzhydrol and 4,4’-bis(dimethylamino)benzhydrol with 1H2 led to
(MeO)42Ha and (MezN)42H4 under significantly milder conditions than required for 2H4

Dehydrogenations were also much more facile for the substituted compounds.
The shorter reaction times and the higher stability of the substituted tetrones minimized
side reactions, increasing their yields and simplifying work-up procedures. A new
problem, however, appeared with the new compounds. Donation of electron density from
the phenyl substituents to the tetrone moieties increased the polarity of the compounds
and consequently lowered their solubility in organic solvents. To overcome this problem
a tetrone with eight t-butyl groups attached to the phenyl rings was prepared. The
benzhydrol 9 required for this synthesis was prepared from readily available 4,4’-

Methylenebis(2,6-di-t-butyl-phenol) (6) following figure 3.10.

65

R NMez OMe OMe
o o

R R (MezN)42 (MeO)42 (t-Bu)g(MeO)42

Figure 3.9. Prepared phenyl substituted derivatives of 2.

HO OH NaH/Mel CH3O OCH: CAN
0 0 —~ —’
THF AcOH

6 7
CH30 OCH3 NaBH, CHaO OCHa
————- O 0
MeOH
0 OH
8 9

Figure 3.10. Preparation of benzhydrol 9

66

Cl CIC
Cl 0'
CHC|3IAIC|3 O O
D —* ’ 5
CI Cl c1 0' CI

10

Figure 3.1 1. Preparation of 10 and its attempted condensation with dihydroxy-

quinone.

R CIR R
0 OH _2H20 o o
———>
Ho 0 o o
R
R 0']: R
11

Figure 3.12. Possible Preparation of 1 by dehydration of tetrol 11

67

This strategy proved to be successful and (t-Bu)3(MeO)42 was the most soluble
out of all the tetrones prepared. It dissolved well even in such a relatively nonpolar
solvent as toluene. In addition the bulky t-butyl groups are expected to stabilize the
radical against covalent bond-forrning processes by providing steric hindrance.

Replacement of the hydrogens by chlorine atoms on the phenyl ring in the ortho
and para positions has been reported to be a very effective protection strategy for
triphenylmethyl radicals.5 Such chlorinated radicals are even air stable. To prepare
analogous tetrone condensation of 3H2 with benzhydrochloride 10 was attempted (figure
3.11). This promising looking reaction failed despite the fact that the benzhydylic
chlorineof 10 exchanged for an acetoxy group in the AcOH solvent.

An alternative method for preparation of additional tetrones based on dehydration
of tetrol 11 was also attempted (figure 3.12). The protected adamantane substituted diol
13 was prepared following figure 3.13. Intermediate 13 was designed to be deprotected to
15, then oxidized to 11 and then dehydrated. Deprotection however turned out not to

work as planned. Product analysis showed 14 instead of 15 as a main product (Figure

3.14).

68

 
 

”1:?“ 91° 021::0 0: o .0

HO OH TsOH

Toluene! DMF fl”

Figure 3.13. Preparation of diol l3

 

Figure 3.14. Reduction of 13 to 14 instead of its attempted deprotection to 15

0 o o
<:><° °0 ._.. 06’ o .L...
O 0 0 l O ? o O
OH "
16 17

Figure 3.15. Oxidative deprotection of 13 with CAN

69

Protonation of the hydroxy] group probably leads to formation of a very stable benzylic
carbocation, which eventually abstracts hydride from the solvent (EtOH). Attempts at
oxygenative deprotection were also stunning. After stirring of 13 with an acetonitrile
solution of CAN an orange product precipitated which could be identified as 16 (Figure
3.15). What is interesting is that the oxidation state of molecule did not change. This
means that Ce“ cations were apparently acting as a catalyst, not an electron acceptor.
The other explanation may be that oxidation took place at the dioxy substituted carbon of
the cyclohexyl group instead of at the 7: system. Further stirring of 16 with the
CAN/MeCN solution produced a mixture of products. Although MS showed a peak
corresponding to the molecular mass of 17, peaks with higher mass were also present,

some of them with odd molecular weights, suggesting the presence of nitrogen.

3.3. Discussion

This work has shown that derivatives of tetrone 2 can be prepared in a simple
two-step procedure from readily available substrates. The simplicity of these reactions
allows preparation of multigram quantities of target molecules in a matter of days (or
even hours). The rate of the first step (electrophilic substitution) clearly correlates with
the higher stability of the intermediate carbocation, pointing to its concentration rather
than reactivity as a limitation of the reaction. The other reason my be that lower acid
concentration used in later synthesis prevents protonation of the dihydroxyquinone. The
condensation step did, however, show some limitations. The lack of reaction between
3H2 and chlorinated benzhydrol 10 suggests that electrophilic attack of

diarylmethylcarbenium ions on 2,5-dihydroxy-1,4-benzoquinone is rather sensitive to the

70

steric hindrance. This fact impedes preparation of tetrones with ortho-substituted phenyl
rings, a substantial disadvantage since the resulting mono and dianions would be
potentially oxygen stable. As a means to overcome this problem, the use of 18 instead of
10 may be examined. Due to the electron donating methoxy groups (in contrast to the
electron withdrawing chlorines) the substitution reaction should be accelerated. In
addition the methoxy groups, with their smaller Size than chlorine, should also hinder the

reaction less, but still be big enough to protect final anion radicals from oxygen.

\ /
O O
OH
O O
/ 18 \

The rate of dehydrogenation in the second step of the scheme developed also
positively correlates with the electron-donating abilities of the substituents. This is not
surprising since electron-donating substituents stabilize both transient cation and
electron-deficient tetrone ring of the final product.6 Thus to prepare 2, 24h reflux is
required and the yields are generally poor, while the methoxy substituted derivatives
require only a couple of hours of reflux; and reaction for (MezN)42 is almost
instantaneous.

DDQ appears to be the only effective dehydrogenating agent in this reaction. Its
biggest advantage besides high oxidizing power is its ability to dissolve in mild nonprotic
organic solvents. Compounds 2H2 and 2 are unstable in the presence of nucleophilic

solvents which excludes many water-soluble metal-ion based oxidizers.

71

The main disadvantage of the developed two step sequence is the limitation to
substituents R that do not rearrange in the vicinity of the carbocationic center during the
first step of electrophilic substitution. The synthetic scheme developed for preparation of
13 was an attempt to overcome that problem. Unfortunately, the oxidative deprotection,
which would complete the scheme, probably did not succeed because of the high
reactivity of the final product. Its is very likely that the alkyl substituted tetrone, lacking
the stabilization provided by phenyl groups, is too electrophilic to survive in the presence
of the ammonium cations introduced with CAN. Conceivably, Ce+4 triflate may be

effective as an alternative to that more traditional reagent.

3.4. Experimental Section

General Methods. Melting points were determined on a Thomas-Hoover
apparatus and are uncorrected. 1H and '3 C { 1H} spectra were obtained at 300 and 75.5
MHz, respectively on Varian GEMINI 300 spectrometer. The 1H NMR shifts are
referenced to residual lH resonances in the deuterated solvents: CDCl3 (8 7.24); DMSO-
dg (6 2.49). The 13 C shifts are referenced to those of the deuterated solvents: CDCl:, (5
77.0); DMSO-d8 (8 39.5).

Toluene, Benzene and THF were dried by refluxing over sodium/benzophenone
under nitrogen; acetonitrile by refluxing over CaHz under nitrogen. The rest of the

reagents and solvents were used as supplied (Aldrich). Ketone 9 was prepared following

ref 7, compound 12 following ref 3.

72

Synthesis of 42_H‘1: 2g of 2,5-dihydroxy-l,4-quinone (14.3 mmol), 5.5 g of
benzhydrol (29.9 mmol) and 0.5 ml of conc. H2804 were refluxed for 1b in 20ml of
AcOH and then left at room temperature to cool. The crystalline mass obtained was
crushed with a spatula and vacuum-filtered through a glass frit. The collected yellow
crystals were washed with a small amount of AcOH and then thoroughly with water.
Crystals were dried overnight at ca. 90 °C in air giving 6.5 g (96.5%) of orange powder.
The purity of the resulting 2H4 proved to be sufficient for the next step. It could, however, -
be recrystallized from 20 ml of AcOH giving 6g (89% yield) of the product: mp 190-192
°C, 'H NMR (CDC13, 300 MHz): 8:569 (s; 1 H), 7.273 (m; 10 H), 7.95 (s; 1 H), ”c
NMR(CDC13, 75 MHz): 5:456, 117.3, 126.75, 128.4, 129.0, 140.6 (Broadening caused

by concerted proton shift makes it difficult to detect Signals of C=O and C-OH carbons).

Synthesis of 2_H;. 2H4 (6g, 12.7 mmol), DDQ (1 g, 4.4 mmol) and benzene (30 ml)
were gently refluxed with magnetic stirring for 18h under a nitrogen atmosphere. The
benzene was evaporated, 75 ml of CHCl3 were added to the flask, and after a few minutes
of stirring at room temperature, the solution was suction filtered. The separated solid of
2,3-dichloro-5,6-dicyano-1,4-hydroquinone was washed with more CHCl3 and the
combined chloroform solutions were evaporated to dryness. After addition of 25ml of
Acetic acid to the flask, the solution was refluxed until the entire solid dissolved. It was
then left to cool. The precipitated 2114 (which can be reused) was filtered off and washed
with more acetic acid. The combined acetic acid solutions were concentrated on vacuum
to a dark red oil, cold water was added, and mixture stirred until the oil solidified. The

resulting orange-brown solid of crude 2H2 was filtered off, washed with water and dried.

73

The dry solid was then crushed, refluxed with 800 ml of cylohexane for about 25 min and
the solution was cooled, filtered and concentrated to about 80 ml of volume. The
precipitated orange crystals of 2H2 (1.2g, 58%yield based on DDQ) were filtered off,
dried, and once again recrystallized from cyclohexane; mp 115-116 °C (dec.), 1H NMR
(CDCI3, 300 MHz): 5.93 (s; 1 H), 7.1-7.6 (m; 20 H), 8.0-8.2 (broad; 1H), l3C NMR
(CDC13, 75.5 MHz): 46.3, 126.8, 128.0, 128.4, 129.1, 131.4, 132.1, 132.6, 140.3, 140.9,

157.4, 177.2, 179.4, 180.3.

Synthesis 011 10g of2Ha (21.2 mmol), 10g of DDQ (44.1 mmol) and 70 ml of
dry benzene were gently refluxed with stirring for 24h under a nitrogen atmosphere, the
solvent was evaporated, and the solid washed with ca. 200 ml of EtzO and air dried. The
resulting red powder was dissolved in 200 ml of CHCl3 and filtered. The separated solid
DDQHZ was washed with more CHC13 and combined chloroform solutions were
concentrated to about 25 ml. When 300 ml of 320 was poured into the concentrate, after
a while red crystals of crude 2 separated. The product was purified by recrystallization
from hot MeCN (Yield 2-10%); mp 288-290 °C (dec.), lH NMR(CDC13, 300 MHz): 7.2-
7.6 (m; 20 H), 13C NMR(CDC13, 75.5 MHz): 128.1, 132.0, 132.5, 140.0, 180.1, 186.6.

Analysis: calc. C82%, H 4.3%, found C 81.04%, H 4.55%.

Synthesis of (MeOIaZ_Ha, To the well stirred refluxing solution of 2g of 3H2 (14.3
mmol) in 100 ml of AcOH containing a few drops of cone. H2804 was added 7.0g (28.7
mmol) of 4,4'-dimethoxybenzhydrol. Solution first became red and after a while yellow

product precipitated. The suspension obtained was refluxed for about 15 min more,

74

cooled to room temperature and filtered. The collected powder was washed first with
AcOH, then water, and dried giving 8g (13.5 mmol 94.4%) of crude product which was
recrystallized from dioxane; mp=263-4°C, 'H NMR (DMSO-d6, 300 MHz): 8:371 (s; 6
H), 5.46 (s; 1 H), 6.82 (d; J=8.7 Hz, 4 H), 7.07 (d; J=8.7 Hz, 4 H), 11 (broad; 2 H), 13C

NMR (DMSO-d6, 75.5 MHz): 8:434, 55.0, 113.5, 118.0, 129.6, 133.9, 157.5.

Synthesis of (MeOJa; 1.3 g of (MeO)42H4 (2.2 mmol) and 1 g of DDQ (4.4
mmol) were refluxed in 50 ml of toluene for 3h. The solvent was evaporated and the solid
extracted in a soxhlet extractor with 100ml of CHzClz. The resulting dark purple solution
of tetrone was diluted with 100ml of Et20 and, left aside for ca. lb and filtered giving 0.9
g (70%) of product: mp>300; lH NMR(CDC13, 300 MHz): 8:3.87 (s; 3 H), 6.88 (d; J=10
Hz, 2 H), 7.28 (d; J=10 Hz, 2 H), Analysis: calc. C73.5%, H 4.8%, found C 72.52%, H

5.06%.

Synthesis of (MezthL 2g of 3H2 (14.3 mmol) and 7.8 g (28.9 mmol) of 4,4'-
Bis(dimethylamino)-benzhydrol was refluxed with stirring in 100ml of absolute EtOH for
15 min, the solution was cooled and the purple (zwitteronic) product of spectroscopic
purity (8.7g, 13.5 mmol, 94.3%) filtered off, washed with ethanol and dried: mp=169-170
°C (dec.), 1H NMR (CDC13, 300 MHz): 5:2.90 (s; 12 H), 4-4.5 (broad; 1 H), 5.49 (s; l
H), 6.64 (d; J=9 Hz, 4 H), 7.10 (d; J=9 Hz, 4 H), '3 C NMR (CDC13, 75.5 MHz): 15:40.7,

43.8, 112.5, 117.7, 129.6, 129.75, 148.9, 168.4.

75

Synthesis of (MezNy; To the well stirred solution of 1 g (1.55 mmol) of
(MezN)42H4 dissolved in 100 m1 of CH2C12 was added slowly 0.7g (3.1 mmol) of DDQ
dissolved in 100 ml of CHC13. During addition the solution turned dark blue. It was left
stirring for ca. 20 min more and the solvent was removed by evaporation. Separation of
the product from DDQH2 was achieved on a glass frit by slow dissolving of the
protonated product with conc. HCl until the acid no longer was yellow. The resulting
acidic solution was neutralized first with NaOH and then with sodium bicarbonate until
solution reached pH 7-8. The precipitated blue suspension of product was suction filtered,
washed with water, dried, and recrystallized from hot DMSO giving 0.9 g (67%) of
(MezN)42-((CH3)ZSO)3: mp>300 (dec.), 1H NMR (DMSO-d6, 300 MHz): 5:3.13 (s; 12
H), 6.76 (d; J= 9 Hz, 4 H), 7.20 (d; J= 9 Hz, 4 H). Analysis: calc C 63.1% H 6.7% N

6.4%, found C 62.43%, H 6.94% N 6.61%.

Synthesis of benzhydrol 9. To the well stirred suspension of 30 g (71.7 mmol) of
benzophenone 8 in 250 m1 of MeOH, was slowly added 7g of NaBH4. During the process
the mixture warmed up, become homogenous and started to froth. When the reaction
slowed down it was refluxed for ca. 30 min more. After cooling, 250 m1 of water was
added and the milky solution was extracted with three 100 m1 portions of CH2C12. The
combined extracts were dried over Na2SO4 and evaporated to dryness. The waxy residue
was recrystallized from a small volume of hexanes giving 26 g of 9: mp=115-116.5, H '
NMR (CDC13, 300 MHz): 8:1.38 (s; 36 H), 2.2 (s broad; 1 H), 3.66 (s; 6 H), 5.72 (s; 1
H), 7.22 (s; 4 H), C'3 NMR (CDC13, 75.5 MHz): 8:320, 35.8, 64.2, 76.7, 125.1, 137.6,

143.4, 158.8.

76

Synthesis of (t-Bulgmggflflfo the well stirred refluxing suspension of 1.4g of
3H2 (10 mmol) in 100 m1 of AcOH containing a few drops of conc. H2804 was added
10.0g (21.3 mmol) of benzhydrol 9. The solution was refluxed for about 15 min, 40 ml of
water was added, and the mixture was cooled to room temperature and filtered. The
collected yellow crystals were washed first with water, then MeOH, and dried giving 10 g
(9.7 mmol, 97 %) of crude product which was recrystallized from AcOH: mp=232-3°C,
Hl NMR (CDC13, 300 MHz): 5:1.36 (s; 72 H), 3.68 (s; 12 H), 5.51 (s; 2 H), 7.11 (s; 8
H), 8 (s broad; 2 H), C13 NMR (CDC13, 75 MHz): 8=32.1, 35.7, 45.2, 64.1, 117.9, 127.2,

134. 8, 142.9, 157.9.

Synthesis of (t-BuMflQkfi, 4.6g of (t-Bu)g(MeO)42H4 (4.425 mmol), 2g of
DDQ (8.8 mmol) and 50 ml of dry toluene were gently refluxed with stirring for 5h. The
hot solution was filtered and the filtrate evaporated to dryness. The resulting dark red
solid was triturated with two 50ml portions of hexane giving the crude product which
could be recrystallized from toluene/hexane giving 2.6g (56%) of tetrone: mp>300 (dec.),
H' NMR (CDC13, 300 MHz): 6:1.35 (s; 36 H), 3.71 (s; 6 H), 7.15 (s; 4 H), C13 NMR
(CDC13, 75 MHz): 8=31.9, 35.8, 64.6, 129.8, 133.3, 134.5, 143.1, 165.0, 178.4, 184.2.

Analysis: calc. C 78.7%, H 8.9%, found C 77.77%, H 9.58%.

77

(1)

(2)

(3)

(4)

(5)

(6)

(7)

BIBLIOGRAPHY

Gomberg, M. Ber. Deutsch. Chem. Ges. 1900, 33, 3150.

Grunanger, P. Methoden Der Organischen Chemie; Vol. 7/3b Chinone; Houben-
Weyl: Stuttgart Thieme, 1979; pp 395. (b) Wagner, H.-U.; Gompper, R.
Chemistry of fimctional groups; The chemistry of the quinonoid compounds;
Interscience: London ; New York, 1974; pp 1145.

Orlemans, E. O. M.; Lammerink, B. H. M.; Vanveggel, F.; Verboom, W.;
Harkema, S.; Reinhoudt, D. N. J. Org. Chem. 1988, 53, 2278.

Gomberg, M.; Jickling, R. L. J. Am. Chem. Soc. 1915, 37, 2575.

Ballester, M. Accounts Chem. Res. 1985, 18, 380.

Becker, H.-D. Chemistry of fimctional groups.; The chemistry of the quinonoid
compounds; Interscience: London ; New York, 1974; pp 335.

Anderson, K. K.; Shultz, D. A.; Dougherty, D. A. J. Org. Chem. 1997, 62, 7575.

78

Chapter 4
GENERATION AND EPR/ENDOR STUDY OF THE DERIVATIVES OF 1- AND 12'

4.1 Introduction

The main aim of conducting EPR/ENDOR studies of radical monoanions was to
gain basic information concerning their structure (especially the symmetry of the spin
delocalization) and their stability. The related study of the dianion focused also on
determination of its ground state (singlet vs. triplet). The required samples of anions and
dianions were prepared by reduction of the parent tetrones with alkali metals (Scheme
3.1). EPR/ENDOR spectroscopic characterization started with investigation of ‘naked’
anions (non-coordinated to metal cations). Generation of such species is possible when
the metal cation is trapped inside the cage of a tightly coordinating organic ligand such as
a crown ether or cryptand. Use of the cryptand c(2.2.2) as an additive in potassium mirror
reductions not only traps potassium cations but also facilitates the whole dissolution
process. The dissolved macrocycle, upon contact with the potassium surface, forms a
solution of potassium potasside [K@c(2.2.2)]+K', a strong and effective reducing agent,1

which then may reduce substances that are themselves insoluble in the reaction solvent.

("‘13 mm
L. .1.) [.3 $1.13
K/NJ) RU)

c(2.2.2) K@c(2.2.2)+

79

Although ab-initio calculations (Chapter 2) showed tight coordination should increase the
chances of the triplet being the ground state of the dianions, the strategy of using naked
anions allows the acquisition of information about them free of complication caused by
ion pairing. Aggregation phenomena can often obscure spectroscopic results by allowing
intermolecular spin-spin interactions and decreasing the solubility of the investigated
species. Studies of the reduction products without macrocyclic co-ligands were also

performed.

Theoreticafliackground. EPR (electron paramagnetic resonance) and the closely
related ENDOR (electron-nuclear-double-resonance) spectroscopies are the most direct
spectroscopic methods for characterization of paramagnetic organic molecules.2 In
particular they can map delocalization of the unpaired electron(s) over the atoms of the
molecule and in the case of polyradicals, provide information about the strength of the
magnetic spin-spin interactions of the unpaired electrons.

Electron Paramagnetic Resonance spectra of monoradicals (S=1/2, doublets) are
usually split into many lines. This splitting of the resonances, known as hyperfine
structure, is caused by interactions of the unpaired ‘resonating’ electron with the
magnetic moments of the nuclei of the molecule (usually H', N”, F19, C”, P31 for organic
molecules). The strenght of these electron nuclear spin interactions is quantified via the
hyperfine coupling constant (a) and may be obtained directly from the EPR spectra by
measuring the separation between the lines of the spectrum. The value of a depends
linearly on the probability of finding the unpaired spin of the electron on the interacting

nuclei (Fermi contact |‘Y(0)|2) and may be expressed as:

80

a=(2/3)uogeuBYI‘P(0)lz
where:
110- magnetic permeability of vacuum, ge-Lande's factor, tip-Bohr magneton y-
gyromagnetic ratio of the nucleus.
For a group of structurally related atoms or molecular units a linear correlation
exists between the Fermi contact and the spin population localized on that unit. One such

correlation is the McConnell relationzz'3

aH=Qpn
where the spin population localized on an sp2 hybridized carbon (pa) is related to the
hyperfine splitting at hydrogen (an) directly attached to it, via the constant Q. This
relation is often used for estimating spin distribution over 1t conjugated systems.

Other important information besides the strength of the hyperfine coupling is its
multiplicity (number of lines signal is split into). For proton hyperfine splitting, the
variety most commonly encountered, the multiplicity of the splitting relates to the
number of interacting nuclei n, as n+1 (generally interpretation of the splitting of the
electron EPR signal is similar to that for NMR spectra). In the case when the electron
interacts with a large number of nuclei (highly delocalized radicals) interpretation of the
spectrum by inspection is difficult and computer simulation of the spectra are needed in
the analysis. If this approach fails, ENDOR spectroscopy must be used. In the ENDOR
experiment, the intensity of the EPR signal of the radical is measured as a function of the
frequency of electromagnetic irradiation in the NMR region while the electron resonance
frequency is kept constant. When such irradiation reaches a resonant value for nuclei

‘coupled’ to the unpaired electron (usually protons) the intensity of the measured EPR

81

signal increases. Each coupled nucleus will resonate with two frequency values,
corresponding to its splitting by the unpaired electron’s spin-Le: depending whether spin
of the unpaired electron is parallel or antiparallel to the field of the instrument. The
difference between those two resonances (usually measured in MHz) linearly relates to
the value of the hyperfine coupling constant. The main advantage of the method is that
one can measure very small hyperfine splittings under conditions where the hyperfine
structure of the EPR spectrum contains so many lines that they can overlap.

In principle EPR spectra of triplet diradicals in solution do not differ significantly
from those of monoradicals. The most notable difference is a broadening of the hyperfine
pattern caused by the diradical molecule’s spin-spin interactions between the unpaired
electrons, averaged by tumbling. A large difference may be observed in powder or glass
solid matrices. In this case tumbling of the molecules is impossible and the anisotropic
spin-spin interactions of the two unpaired electrons of the molecule do not average. As a
consequence the external magnetic field felt by each electron is strongly perturbed by the
magnetic moment of the other unpaired electron. The degree of such perturbation
depends on the position of the molecule vs. the external magnetic field. Since the
distribution of diradicals is random, a very broad (usually hundreds of Gauss) EPR
spectrum may be observed.4 Figure 4.1 displays a typical example of such a triplet
spectrum. The D value reflects the magnetic interaction between the unpaired spins and is
proportional to r'3 where r is an average distance between unpaired electrons. The E

value, in turn, relates to the deviation from threefold or higher symmetry of the diradical.

82

 

 

 

 

 

 

 

 

 

 

20

Figure 4.1. A typical solid state spectrum of triplet diradical

Another characteristic feature of the triplet state diradical is the presence of the
so-called half-field transition. This is a weak signal (for organic diradicals in ca. 1600 G,
g=4) indicating formally forbidden AS=2 (from spin S=-1 to S=1) transition. In the case
of doublet radicals, neither a half-field transition nor anisotropic spin-spin interactions
may be observed and frozen solutions produce only a sharp peak in the regular field (ca.
3200 G, but depends on the instrument used) intensity. Of course singlet state molecule

will not produce any EPR signal at all.

83

4.2 Results

Monoanions. All para substituted monoanions, in the present work could be
handled at room temperature without noticeable decomposition. Under these conditions,
however, 2' decayed slowly with a half-life of ca. 1h, so it was handled below —20 C. All
prepared monoanions were air sensitive. Monoanions in solution produced strong EPR
signals with g values characteristic for organic radicals. Values of hyperfine coupling
constants of the phenyl rings were determined for all anion radicals by ENDOR and are
presented in table 4.1. The ENDOR spectra are also reproduced in figures 4.(2,4,6,8).
Experimental EPR spectra of those anion radical could be successfully simulated (figures
4.(3,5,10) by assigning a1 values to all 8 ortho (and in the case of 2', 4 para) protons and
a; to all 8 meta protons of the four phenyl groups. Simulation of the signal of (MezN)42'
(figure 4.7) was not attempted, due to the low resolution of the experimental EPR
spectrum and the inability to find values of the coupling constants for the -NMe2 protons
by ENDOR. The monoanion t-Bu3(MeO)42' showed strong temperature dependence of its
ENDOR and EPR spectra. At low temperature (197 K), its ENDOR spectrum (fig 4.8)
showed two couplings caused by its ortho hydrogens (5.4 and 1.5 MHz). Raising the
temperature to 223 K decreased those signals intensities with simultaneous appearance of
a new averaged single value of 3.39 MHz. At 263 K only the latter one may be observed.

The general TRIPLE experiment (nuclear-nuclear-electron resonance) has
determined that for all of the investigated monoanions the spin population on the ortho
and para hydrogens has the opposite from that that on the meta sites. Assuming that each
phenyl ring has three carbons with negative spin density (two orto and 1 para) and three

with positive (two meta and 1 ipso) and assuming the spin population on the ipso carbon

84

is equal to the one on meta, population of unpaired electron residing outside the tetrone

ring may be estimated with use of the McConnell’s relation (Q=-70MHz) as:
4*(3*a2-3*a1)/-70=0.3

where: 4-numer of phenyl rings, 3-number of carbon atoms, with the same spin density,

on each phenyl ring.

Table 4.1. Values of the hyperfine constants for monoanions of 2 and its derivatives

 

 

Radical a1 (ortho) a2 (meta) a3 (para)
anion (MHz) (MHz) (MHz)
2' 3.23 1.43 3.23

(MeO)42' 3.24 1.31 0.36
(NMe2)42' 3.26 1.08 -

(t-Bu)3(MeO)42' 3.39 - -

 

85

1 y} ' ; 1MHz

 

‘ 1' ~H V I. ’1 I | 9,
1 l 1 , . 4. 1'11"."3rl“ { U“ ,‘ . 1 l.“ “H- .111 ‘1“ . r r
‘31". 1 ”1311‘!" .‘lttl ’1 . ”11"" 1 1 :1“: ‘1 nl "31"."! 1 1 3'» .A
i f l ‘. ‘

Figure 4.2. ENDOR of 2'[K@c(2.2.2)]+/T HF recorded at 246K (a1=3.23; a2=1.43 MHz)

Figure 4.3. A) EPR of 2'[K@c(2.2.2)]+/THF recorded at 246K.

B) Simulation with parameters obtained from ENDOR.

86

 

.. 1MHz

Figure 4.4. ENDOR of (MeO)42'[K@c(2.2.2)]+/THF recorded at 215K, (a1=3.24; a2=1.31

MHz).

 

Figure 4.5. A) EPR of (MeO)42'K@c[2.2.2]+/T HF recorded at 215 K.

B) Simulation with parameters obtained from ENDOR.

87

 

1.?! . . |
' 'i

"l “1 I r, at") 1
1.. . 1 1 ‘ J ‘
111“. .11! ‘ 11“ l l l
l ‘ 1: .

ll '1 “a J 1 ’1.

n | ‘ ‘

1 "M ' h 111‘ i If,

“1111 ‘ 1" lt H. I” A“ 1‘ 111111 1 1.“? I 1 11111;} . ll .
1‘” 1. 1| ‘ ‘1" l' ‘ .1111, ’1’11 ~12 4“,“; ”ii-‘2‘"
' 1 11.“) H 1 1 1'1. ‘1 1.1.1131“! l 711111 1“! 111' ‘11 l'

‘1" 1 ‘ ' l 1
I" 1" t j. 11”. 1 1 I til-11*
I! l,‘ 1“ 1' .‘ ‘ ‘ 5
~11 11 1‘ 1 '1 1‘ 511
l 1 fit ‘ F 11 ‘1
1‘" 1. 1 1'
.1"
‘1

F1gure 4 6 ENDOR of (NMe2)42'[K@c(2.2.2)]+/THF recorded at 210 K (a1—3 26
82:1..08)

F1gure 4 7. EPR of (N Me2)42'[K@c(2.2.2)]+/T HF recorded at 243 K

88

11111

.1 11‘. " ,I ‘ 111

1 1 1“ 111 pl 1III 111
1111'11'111‘1'11'1111111" 1111111 111 1'1," 111
1 ‘ .

1 11 11‘ 11 #1“ .III 11 1, 1 :1
1111111. 1111111111 11, III“ 1 I III

1 I) I 1 1, I
111111 1 W111 1 1111;131:111 ‘11 11"":1
1 V!“ . I
197 K ‘ ll I_1Iih1
[‘1 11
1);” I Il'1' 1 115111
1II
1111 '1 11.1 .111 111 I1I 1.11111 II11'1II
1 1 I1IIIIIII|J I111 11III1I‘II1I111IKI1I 11 11IV11I‘1I111kl IIIIIlI; LI .IlhlIll‘I‘ 1 1,11 '1‘ '1.I II I I I
11111 1111111 111111111 1 11 1 111131111 1‘ 111 l 1 1 1 11.1111
1 1'" 11111” "1 11 111 1111.11
11111 11” ‘ I 11:11, I 11 1
223 K I 111 '11 1111.111 1
1 .1.
'1
, 111
1111 III.‘ 1 III
[J 11 I; I 11111111111111,“ 11‘ 1.11 W11
1 11‘ 511111
‘ 1 1 1~. ".1111 11 ‘11'1-1'11"1‘12111‘131V4111‘”1.1/")1" 1'1 11 II
”9'1qu'11,.1'51‘111'131w’11' 11.1) .0 hfl' 1 1 ”if”! 1 m1 . 1 I 'III
1 1"1 1 1 111111:
1 1'1 ', ——
263 K 1" 1")
' 111

F1gure 4 8 ENDOR of (t-Bu)g(MeO)42'[K@c(2.2.2)]+/1'HF recorded at 197 223 and

263K, a1,=5.4 MHz; a,.,=1.5 MHz (197K), a.=3.39 (263 K)

89

Figure 4.9. EPR of (t-Bu)3(MeO)42'K@c[2.2.2]+/T HF recorded at 180 K

Figure 4.10. A) EPR of (t-Bu)g(MeO)42'K@c(2.2.2)+/T HF recorded at 263 K

B) Simulation with parameters obtained from ENDOR.

9O

Table 4.2. Values of the hyperfine constants of species present in the solution of dianions

 

 

Dianion a1 (ortho) a2 (meta) a3 (para)
(MHz) (MHz) (MHz)
22' 7.87 3.22 7.87
(t-Bu)g(MeO)422' 8.30

 

Dianions. The main goal of EPR studies of the dianions was to observe signals of
their triplet states. Frozen solutions of 22' did not show any broad signals characteristic
for typical triplet diradicals nor did they show a half-field transition peak. Only narrow
monoradical peaks could be observed. Analysis of the solution spectrum showed that the
species responsible for that signal is not simply the monoanion. A clean signal of a new
radical was obtained for the sample of 22' (fig. 4.12). The ENDOR spectrum was also
obtained (fig. 4.11), showing two values of proton hyperfine splitting, approximately
twice as big as those of the monoanion. Attempted simulation of EPR spectrum with all
the protons of the tetrone molecule did not produced satisfying results. Halving the
number of interacting protons, however allowed however a simulation much closer to
the experiment (although still not perfect and thus not reproduced). The signal produced
by paramagnetic species in the solution of (t-Bu)g(MeO)422' was much easier to

interpret.

91

I I M”) ,IIIH

I vii I if; ”H WNW” #111. ? IIH

[H N 'IEI I I :l‘ ; ili’i‘li ”HI in} a“ «m H U Ih
rI igwir ”1"“ I“ i“ WITH” W" i 4h ”It! f ”Jim Iii

I I
I

III” “I 1MHz

 

Figure 4.11. ENDOR of 22'[K@c(2.2.2)+]2/T HF recorded at 240 K (a1=7.87 MHz,

a2=3.22 MHz)

106

 

Figure 4.12. EPR of 22'[K@c(2.2.2)]+2/THF recorded at 247 K.

92

 

'III
II I
III I 1MHZ
I It I.
II‘I I II
1“” F‘III ‘I‘
I'l I' III I I II
I. I I I I II I II I' III I III III“
I I .l’ (I I I.“ III I “I. (III'IIIIII I II II I I ’I'LII II‘II “III :III “(III ‘ I.“ II .1 I'IIII’
III-I‘I'IIII I "III" III " III II I" III “I ‘ “I" ' I'I } I I' I'
III ~‘ I II.
I (I I‘ I I I I: III
I'I, III I I {I
III II III! I
II II J

Figure 4.13. ENDOR of (t-Bu)g(MeO)422'[K@c(2.2.2)+]2/THF recorded at 263 K,

(a1=8.30 MHz)

Figure 4.14. A) EPR of (t-Bu)3(MeO)422'(K@c[2.2.2]+)2/THF recorded at 290 K

B) Simulation with parameters obtained from ENDOR.

93

Lack of the meta and para protons simplified the spectrum and ENDOR (fig. 4.13) has
shown only one hyperfine coupling which could be assigned to the ortho protons. EPR

simulation unambiguously assigned it to 4 rather than 8 protons (fig. 4.14)

Aggregates of 2' and Cs+._Reduction of tetrones with alkali metals in the absence
of macrocycles was hampered by their low solubility in THF. Out of the three tetrones
initially synthesized (2,(Me0)42 and (MezN)42) only 2 dissolved in THF to the extent that
reduction could be carried out. But even if this precursor could be dissolved, the
potassium mirror quickly covered with a film of insoluble reduction products, which
prevented further reaction. A cesium mirror appeared to form more soluble products of
reduction than potassium. When solution of 2 contacted this mirror, color of the solution
darkened and an EPR signal of 2' could be detected. The only difference between the
solution spectra of 2Cs and 2'[K@c(2.2.2)]+ was the much broader linewidth of the first
one. A large change, however, could be observed for the powder spectra. While frozen
2[K@c(2.2.2)] produced only the sharp peak of the monoradical, the spectrum of 2Cs had
the structure characteristic for the triplet state species. Figure 4.15 shows the signal of
2C5 as a function of reduction progress. A clear change of the spectrum’s appearance was
observed during the reduction. At first, the spectrum seemed to consist almost entirely of
a pure triplet signal with a D value of ca. 90G. Then as the reduction progresses, a
stronger and stronger doublet spectrum overlapping with the triplet one could be
observed. Unfortunately, even with Cesium, after a while the reaction stops and no

sample of dianion can be prepared.

94

 

'(\
. h
~ III. \. - .
33"" ’I'I""'v“."‘“r'rvf‘” I I l I .1
I: I I
A * I
I ‘
. I .4 .- ‘
I f“, ‘1
W1
. -I~ I _ 1
4‘1 r"'
I
I
I
B \ i
,1/1 I

Figure 4.15. Progress (from A to C) of reduction of 2 with Cs mirror monitored by EPR.

Spectra collected at 4.2 K (A), 3.9 K (B), 4.2 K (C).

95

4.3 Discussion
‘Naked’ monoanions. The experimental spectra of monoanion radicals may be

interpreted rather unambiguously. The data obtained agrees with the simplest most
symmetric model of hyperfine interactions, pointing to high degree of symmetry (on the
EPR timescale) of spin delocalization over the whole molecule. Temperature dependent
study of (t-Bu)g(MeO)42' shows however that this symmetry can be broken. Separation of
the ortho protons into two nonequivalent groups at low temperature may be explained by
substituent induced inhibition of rotation and freezing of the phenyl groups into
conjugated and non-conjugated groups (fig. 4.16a), or localization of the electron on one
methylene site (fig. 4.16b).

Para-substituted monoanions are much more stable, what points to this location as
being the most reactive of the whole molecule, as it is in the simple triphenylmethyl

radical.

 

Figure 4.16. Possible reasons for splitting of the ENDOR signal of the meta protons of

the (t-Bu)3(MeO)42'.

96

‘Naked’ Dianions. No signal of the triplet state dianion could be observed in the
whole range of investigated temperatures for ‘naked’ dianions. What was surprising was
the observation of a new ‘monoradical’. Various types of species may be assigned to it
(Figure 4.17). Products of dimerization (a) at first seem to be the most appealing
explanation. Further analysis however shows certain drawbacks to that hypothesis. First,
22' could dimerize in one of three ways: ‘head to head’, ‘head to tail’, or ‘tail to tail’. As
in the case of the triphenylmethyl radical, tail to head coupling should prevail. ENDOR
shows however signals of only two types of protons, consistent with ‘head to head’
coupling only. The other hard-to-explain fact is the presence of the related signal in the
case of (t-Bu)g(MeO)422'. Due to the effective steric protection this dianion should not
dimerize easily. Formation of the dimer peroxide (b) would require the presence of
oxygen. Although it can not be absolutely ruled out, the greatest care has been taken to
eliminate such possibility. The diradical peroxide would also be expected to be unstable.
A weak oxygen-oxygen bond might break easily and the free oxygen is valence couple
with the radical part of the molecule producing diamagnetic compounds like epoxides.
Other modes of formation of a new radical may be protonation of monoanion or
abstraction of the hydrogen from the solvent (THF) by a photo-exited dianion (c). The
first case is very unlikely since the monoanion should not be a very strong base and the
solvents were carefully dried before the use as well as being continually dried by the

presence of the metal mirror.

97

 

a) Dimers of dianions b) Peroxides

Ar H Ar Ar H Ar Ar ' r
o o o 0' 0 O
- '0 0
Ar Ar Ar ' Ar Ar . Ar
c) Products of protonation or hydrogen d) Localized diradicals

abstraction

Ar .Ar

e) Products of fragmentation

Figure 4.17. Possible species responsible for EPR signal found in the solution of dianions

98

In addition in both cases some of the spin density should be delocalized onto the newly
acquired proton which in turn should be observable in EPR and ENDOR spectra. The
signal may also be attributed to a localized diradical (d). It is however difficult to imagine
why the conjugated dihydroxy-quinone anion spacer should isolate the radical centers.
The most plausible explanations seems to be fragmentation of the dianion.
Concerted or stepwise processes (figure 4.18) may be imagined as a way of getting there.
Radical anions such as (e) would easily explain both the high-spin density and the spin

delocalisation on only two of the phenyl rings, as well as the presence of only one type of

species.
Ar Ar Ar
2\ Ar , Ar
'0 O '0 . Ar
a d I I 2 H
'0 O -ZCO -
(J Ar ,
Ar Ar fl:
Ar . Ar Ar 0 Ar
- ,0 1.0
b - C 2 \9§\0
L 0
Ar . Ar Ar ° Ar

Figure 4.18. Proposed concerted (a) and stepwise (b) paths of fragmentation of 22' and its

derivatives.

99

Ion aggregates. The low solubility of the complexed anion shows that 2 and its
derivatives may have the desired ability to form extended structures in the solid state.
Similar conclusions may be deduced from observation of the powder EPR spectra. A
triplet signal detected in frozen solutions of the 2Cs points to strong intermolecular spin-
spin interactions, probably through the metal cation. This proposed origin for the signal is
supported by the fact that in the presence of the cryptand, only a sharp signal of the
monoanion doublet may be seen and that similar signals were recorded for analogous
complexes of quinones.5 It is interesting is that while the reaction proceeds the signal of
the triplet disappears and the doublet signal seems to be more and more intense. This
observation seems to contrast with the expectation that a higher concentration of anions
should lead to a higher degree of aggregation. Problem this may be easily solved if we
assume that predominating later signal is quartet or higher multiplet rather than doublet.
Out of two modes of aggregation which may be expected (Figure 4.19). Solid state

crystallographic study (Chapter 7) seems to support structure b).

O
O R
R R
R
R Q R R R R R R
0.. E ,0 O 0., ..O 0. ".0 O
‘ch.’ O .505. 50%.
R 0'" § '-.0 o o o o o o o
o R R R R R R
R R R
R
O a b
o R ) )

Figure 4.19. Two possible modes of aggregation of 2Cs

100

4.4. Experimental

Purification and Handling of THF. Anhydrous THF (Aldrich) was placed in a
Shlenk bottle with a few drops of NaKz alloy, degassed through a few of freeze-pump-
thaw cycles and left to sit until occasional shaking turned its color to blue. Solvent was
then freeze-pump-thawed until vacuum over the frozen solid reached 1"‘10'S torr and high
vaccum distilled to the storage bottle. Just before the experiment the needed quantity of
the solvent was distilled to a bottle containing a few drops of NaKz, shaken until blue,
and distilled back to the H-Cell.

Preparation of the EPR/ENDOR samples of mono and dianions. In a typical
experiment tetrone (ca. 10mg) and stoichiometric amount (1:1) of cryptand were placed
in chamber A of the H-cell (figure 4.20), evacuated on a high-vacuum line, and placed in
a Helium filled glove box. A small piece of potassium (ca. 5mg) was placed in chamber
B. Then the H-cell was sealed, put back on the vacuum line, and evacuated until the
vacuum reached ~5*10-6'torr. In order to minimize the possibility of an oxygen leak, the
pyrex glass tube leading to chamber B was sealed off just below the swagelock joint.
Potassium mirror was then prepared in chamber B by heating the potassium with a small
gas flame. Chamber A was immersed in liquid nitrogen and THF (ca. 30ml) distilled into
it. After thawing of the solvent, the H-cell was placed in an isopropanol/dry ice bath (—
40-50°C) and the solution was sloshed between chambers A and B through the glass fi'it.
The progress of the reduction was monitored by Vis-NIR spectroscopy in the opticall cell
attached to the sidearm. Control over the degree of reduction was facilitated by the fact

that after depletion of the cryptand, reduction considerably slowed down.

101

 

  
 
   

 

 

 

I I
F usher-Porter JOInt @ Teflon
Stopcock
Sidearm . 7
FE
EPR Tube 5» T
Glass Frit

Optical Cell
Chamber B
Chamber A

Figure 4.20. A typical H-cell used for preparation of EPR samples.

102

Sealing

Temporary
&/ Swagelok

A small amount of solution was then transferred to the quartz EPR tube and the tube
sealed off while the sample was kept in liquid nitrogen.

Spectroscopy. EPR and ENDOR spectra were recorded on a Bruker ESP300E
spectrometer equipped with a DICE ENDOR unit. Typical conditions for EPR were:
modulation frequency 100kHz, modulation amplitude 0.1 gauss, power 0.1mW, gain 105 .
Typical conditions for ENDOR were: modulation frequency 12.5 kHz, microwave power

20-32 mW, prower 200-300 W, modulation depth 100kHz, gain 105.

103

(1)

(2)

(3)

(4)

(5)

BIBLIOGRAPHY

(a) Jedlinski, Z. Accounts Chem. Res. 1998, 31, 55. (b) Dye, J. L.; Jackson, J. E.;
Cauliez, P. In New Aspects of Organic Chemistry, Organic Synthesis for
Materials and Life Sciences; Yoshida, Z., Ohshiro, Y., Eds; VCH Publishers:
New York, 1992.

(a) Carrington, A.; McLachlan, A. D. Introduction to magnetic resonance with
applications to chemistry and chemical physics; Harper & Row: New York, 1967.
(b) Atherton, N. M. Principles of electron spin resonance; Ellis Horwood : PTR
Prentice Hall: New York, 1993.

McConnell, H. M.; Chesnut, D. B. J. Chem. Phys. 1958, 28, 107.

(a) Dougherty, D. A. In Kinetics and Spectroscopy of Carbenes and Biradicals,
Platz. M. S., Eds.; Plenum Press: New York, 1990, pp. 117-143. (b) Berson, J. A.
In The Chemistry of Quinonoid Compounds, Vol II, Patai, S.; Rappoport, Z.,
Eds.;Wiley: New York, 1988, pp. 455-536. (c) Berson, J. A. In Biradicals,
Borden, W. T., Eds.;Wiley: New York, 1982, pp. 151-194.

Brown, M. A.; McGarvey, B. R.; Ozarowski, A.; Tuck, D. G. .1. Am. Chem. Soc.
1996, 118, 9691.

104

Chapter 5

NMR

5.1 Introduction

Since EPR studies did not indicate the presence of triplet state tetrone dianions, a
complementary NMR study was performed to gain some information about the structures
of the species generated by reduction of tetrones. NMR signals can not be observed for
most paramagnetic organic molecules. The reason is that the unpaired spin of the radicals
~ delocalized over the molecule interacts with the NMR active nuclei leading to their fast
relaxation and consequently extreme broadening. In most cases the broadening is so
strong that no signal may be observed at all. If dianions of the investigated tetrones could

be observed by NMR, it would mean that their ground state is singlet and give additional

insight into their structure.

0 e- E 0 e- 0
O O 0'
Q
Ar r Ar r
O 0 e' O E 0 e-
O O O l 0
Ar Ar Ar Ar

 

Figure 5.]. Reductions investigated by H1 NMR

105

In these studies tetrone/c(2.2.2) mixtures in dg-THF were reduced with potassium mirrors
and the progress of reduction was monitored by the NMR spectroscopy. In addition to
derivatives of 2 reduction of model quinone l was also investigated under the same
conditions. (Figure. 5. 1)

Singlet state species with low HOMO-LUMO gaps possess a substantial
admixture of the low-lying Triplet State (through a spin-orbit coupling mechanism). This
in turn leads to local perturbation of the magnetic field, which, in analogy to the
diamagnetic ring current of aromatic compounds, is called a ‘paramagnetic ring current’.
The effect of such a perturbation is ‘paratropic’, yielding an upfield shift of the exocyclic
ring protons.

Extensive study performed on ‘antiaromatic’ dianions of fused polycyclic

hydrocarbons has shown a correlation between the HOMO—LUMO gap and the degree of

the paratropic shift. 1'4 The lower the HOMO-LUMO gap the higher the contribution of
the triplet and consequently the higher the shift. In the case of very low HOMO-LUMO
gaps, in addition to the paratropic chemical shift, paramagnetic broadening of the NMR
lines can also be observed. In some cases spectra are highly temperature dependent.
Raising the temperature leads to a higher triplet admixture and higher pararnagnetism of
the molecule. This in turn leads to broadening of the peaks relative to the low temperature

case .

106

5.2. Results

Reduction of guinone Q. After very short contact of the c(2.2.2)-quinone solution
with a potassium mirror, its yellow color turned green. At the same time the signals of all
the quinone protons disappeared, including even the protons of the t-Butyl groups. The
only observable signals left were those of the solvent and broadened lines belonging to
the cryptand. Further reduction did not dramatically change the spectrum besides
broadening of the cryptand lines. During that process the color of the solution gradually
changed from green to blue, which is presumably the color of the ‘pure’ semiquinone.
Continuation of the reduction process bleached that color, resulting finally in a faintly
yellowish almost transparent solution. At that stage the sample became diamagnetic and
sharp lines of diamagnetic products could be observed by proton NMR. Further reduction
led to generation of the transient blue color of [K+@c(2.2.2)]K' which at room
temperature decomposed afier a while.

Reduction of the tetrone (t-BU)§(M€O)5L Brief contact of the dark red solution of
(t-Bu)gMeO42 with the potassium mirror led to quick disappearance of the signals of the
phenyl and methoxy protons. The signal of the tert-butyl protons remained, although it
became somewhat broader (figure 5.2). Further reduction broadened the t-Butyl line more
and more (figure 5.3) and at the same time the center of the peak slowly moved
downfield from its initial shift of 1.36 ppm (A1) to ca. 1.5 ppm (A3). Upon longer contact
with the mirror, the behavior of the t-Butyl peak reversed. The line became sharper and
moved back to the position of ca. 1.4 ppm (A4). At the same time the color of the solution

changed from purple to green. Further reduction of the green solution did not change the

107

position of the peak. The signal, however, became broader (A5) again while the solution

became red.

a) C\

 

 

b)

 

 

 

 

 

 

C
\
l I 'l""l l I"'l""l"l"‘
9 8 7 6 5 \0 Y3 J2 1 pp
D THF

Figure 5.2. 1H NMR of the mixture of (t-Bu)3(MeO)42 and c(2.2.2) (THF-d8, RT)

a) Before and b) After a very short reduction time
A) o-hydrogens of the phenyl group, B) metoxy protons, C) t-Bu protons,

D) c(2.2.2) peaks.
108

A1

THF

 

 

 

A2

 

 

 

 

IITIIIIIIIT]ITIIIITI—rijIiITrTI—T‘lfirrlIIIIIIIIIII]

1.9 1.. 1.7 1.6 1.5 1.4 1.3 1.2 1.1 m

Figure 5.3. Progress of the reduction of (t-Bu)g(MeO)42 observed in the 1-2 ppm
range, THF and. A) t-Butyl protons may be observed. B) probably belongs
to small diamagnetic impurity.

109

A1 I'.
I! !' A3
I ll
ll ‘ I
I: . 3' I" I B C
‘l I, I \T l_
a r/ k“ Rx I I
N l
M A1 MW“ WIWWQ
IA:
I I
II c
B I
I' II. I
I I I I
b) M I I I. h. “l‘ x I
I IIIWIIMI W “IN. II“MH\MW
A1 I A2
I c
II II .
I; I}. I
, B I
: I I
I . I

I
l

’ I

r .
C) I I I I

III J K“... II I‘
“I _ I "MWMWWJ W'v‘mw" ' ““00,er New“ My»

8.0 7.5 7.0 6.5 6.0 5.5 ppm

Figure 5.4. Aromatic region of the 'H NMR spectrum of the diamagnetic products
obtained by decomposition of the solutions of a) 22', b) (MeO)422' and c)
(MezNhZZ'.

110

Reductions of the tetrones ZLIMergz and (NMe_2)32_. Unlike (t-Bu)3(MeO)42, 2 does not
possess t-Butyl groups. The signal of 2 thus disappears completely at a very early stage of
the reduction experiment and does not reappear (at low temperature) at any other stage of
the reduction. However when the solution of the dianion 22’(as judged by its green color)
is warmed up to room temperature, a relatively clean spectrum of a new diamagnetic
compound appears in the aromatic region (Figure 5.4). Similar behavior may be observed
for the other two tetrones (MeO)42 and (N Me2)42 with the main difference being that no
signal of the starting material may be observed since these compounds are insoluble in
THF. In both cases, the final room temperature formation of a diamagnetic product may
be seen (Figure 5.4). In all three cases, attempts at firrther reduction of the dianion led to
precipitation of the solid.

No ’3 C NMR signal could be observed for any of the tetrones at any stage of the

reduction besides the neutral precursor.

5.3. Discussion

Even short contact of the THF soluble quinone and tetrones leads to dramatic

changes in the NMR spectra of those compounds. Presumably electron transfer between

reduced and unreduced molecules

Q + 0': Q'+Q

111

is fast on the NMR time-scale, leading to observation of the averaged signal of tetrone
and its radical anion. Even a minimal concentration of the unpaired electrons leads to
intensive broadening of the lines and finally disappearance of the 1H NMR signals. In the
case of (t-Bu)g(MeO)42, the signal of the saturated t-Butyl groups, with their weak
couplings to the paramagnetic n-regions, are still observable even for a pure anion
radical.

When both monoanion and dianion are present in the solution, an analogous
process must happen since no separate signal of the dianion can be observed. Unlike the
behavior of the model quinone no sharp signal of the dianion is observed for any tetrone.
Only the broad signal of the t-Butyl groups can be seen for (t-Bu)3MeO422'. The
broadness and the position of this signal indicate that (t-Bu)3MeO422' (or immediate
products formed from it) are ‘less paramagnetic’ then the monoanion, which seems to
decrease the likelihood of it being a triplet under those conditions. The best explanations
of the tetrone dianions at that stage is that they are probably strongly paratropic
compounds with a low HOMO-LUMO gap and consequently high admixture of the
triplet into the ground singlet state even at low temperatures.

The other important issue is the further fate of the dianions. Their slow
decomposition leads to the formation of diamagnetic compounds that are cleary related to
each other. Analysis of the multiplicity of the aromatic NMR signals allows for
generalizations like: (a) There is only one major product (b) The product has one type of
phenyl rings only. The relative positions of the signals of the ortho- and meta- protons
prompts the conclusion that the phenyl rings are conjugated to the 112 system of the

product.

112

Ar Ar Ar Ar

 

 

 

 

Figure 5.5. Proposed species responsible for NMR signals presented on figure 5.4

Having in mind the radical intermediates proposed in chapter 4, one of its diamagnetic
dimers may be responsible for the observed NMR signals (figure 5.5). Besides phenyl
ring signals, additionally two sharp peaks (B and C) may be observed in figure 5.4. The
only source of the non-aromatic protons in the solution may be C(2.2.2) (if the presence
of water is completely excluded). Those peaks may belong to the products of its

decomposition.

5.4. Experimental

NMR reduction experiments were performed in high precision pyrex NMR tubes
attached to the high vacuum line through a swagelock connection, Teflon stopcock and
Fisher-Porter joint (figure 5.6). Tetrone (ca. 5mg) and C(2.2.2) cryptand (more then 2:1
stoichiometric excess with respect to tetrone) were placed in the tube, the tube was
evacuated and lefi to a pressure of ca.10'5 torr, the stopcock was closed, and the whole
device was introduced into a helium-filled dry box. Inside the box, the NMR tube was

detached and a small quantity of potassium metal was smeared with a pipet inside the

113

upper part of the tube (in the place where the tube was supposed to be sealed off later
with a torch). The tube was then reattached to the stopcock and put back on vacuum line.
After 10'5 torr was reached again, the NMR tube was immersed into liquid nitrogen and
ca. 0.5-0.75 ml of previously dried and degassed THF-d8 was distilled into it. While still
in liquid nitrogen and under vacuum, the tubewas sealed-off with the torch in such a way
that at the same time the potassium formed a mirror in the upper part. After slow thawing
of the THF, the tube was stored and manipulated vertically so no contact of solution with
the mirror was allowed prior to spectroscopic studies

Between NMR measurements, the solution in the tube was briefly allowed to
contact with the mirror while remaining in the isopropanol-dry ice bath (below —60°C). It

was then quickly dried off with a cloth and inserted into the pre-cooled probe of the NMR

instrument

114

  
 

Teflon stopcock

}— Fisher-Porter joint

Swagelok connection

 

 

 

 

 

Potassium

 

NMR Tube

 

 

Figure 5.6. Instrument used for preparation of the NMR samples.

115

(1)

(2)

(3)

(4)

BIBLIOGRAPHY

Minsky, A.; Meyer, A. Y.; Rabinovitz, M. Tetrahedron 1985, 41, 785.

Rabinovitz, M.; Cohen, Y. Tetrahedron 1988, 44, 6957.

Rabinovitz, M. Top. Curr. Chem. 1988, 146, 99.

Rabinovitz, M.; Cohen, Y. Adv. Chem. Ser. 1988, 53.

116

Chapter 6

CYCLIC VOLTAMMETRY

6.1 Introduction

In order to get detailed insight into the process of formation of mono- and
dianions of tetrones 2, their redox behavior was investigated by means of cyclic
voltammetry (CV). This technique allows determination of reduction potentials (E°) of
the investigated tetrones and provides some information concerning the chemical
behavior of produced from them reduced species.l Our main interest was focused on
verification of whether mono- and dianions are indeed stable products of reduction. The
study was performed in non-polar THF (for 2 and (t-Bu)g(MeO)42) and polar DMF (for 2,

(MeO)42 and (MezN)42).

6.2 Results

Figure 6.1 presents a voltammogram of (t-Bu)3(MeO)42/T HF as a typical
example; the rest of the tetrones produced similar patterns. Four reduction waves (E1°,
E2°, E3°, E4°) could be observed for these species. The first two waves appeared to be
reversible, with corresponding oxidations waves E1’° and E2’°, while the third and fourth

were irreversible. The values of the measured reduction potentials (versus ferrocene/

ferrocenium couple) are collected in the table 6.1.

117

Table 6.1. Cyclic voltammograms of the 1 mM solutions of the tetrones with 0.1M of (n-
Bu)4N(PF6) as supporting electrolyte. Swept rate 0.1V/s, glassy carbon

working electrode, platinum wire counterelectrode, silver wire quasireference

calibrated to F esz/F esz+ couple (0.46V).

 

131° 132° 133° 134° AEI,2° AE2,3° AE3,4°

 

 

THF
2 -0.39 -l .10 - -2.01 0.71 - -
(t-Bu)3(MeO)42 -0.57 -1.22 -2.58 -2.41 0.65 0.36 0.83
DMF
2 -0.28 -0.91 -1.23 -1.82 0.63 0.32 0.59
(MeO)42 -0.39 -0.96 -1.29 -1.85 0.57 0.33 0.51
(MezN)42 -0.78 -1.27 -1.51 -2.03 0.49 0.24 0.52

 

 

 

 

 

 

Figure 6.]. Cyclic voltammogram of (t—Bu)g(MeO)42 in THF

118

6.3 Discussion

The first wave apparently corresponds to the reduction of the tetrone to its
monoanion. These species are stable reduction products, as has been firmly established
by both EPR spectroscopy and the X-ray diffraction study.

A clear correlation between the strength of the electron-donating abilities of the
phenyl substituents and the reduction potential may be established. As may be expected,
substituents with stronger donating abilities make the tetrone more difficult to reduce. For
example the difference between E1° for 2 and (MezN)42 (in DMF) is 0.50 V. The lower
reduction potential E1° of 2 in DMF than in THF (-0.28 vs. -0.39V) may be rationalized
by its better solvating abilities for the charged anion. Oxidizing abilities of the most
electron-deficient 2 are better then for simple p and o-benzoquinones (-0.43 V and —0.47
V for benzoquinone in DMF and 3,5-Di(tert-butyl)-o-quinone in THF respectively)" 3
measured in similar conditions but due to electron donation from the conjugated phenyl
ring are lower then for some other systems containing four conjugated carbonyl groups.2
Apparently tetrone I meant to be strong electron acceptor will require electron-
withdrawing substituents R.

The second wave E2° most likely corresponds to the reduction of the monoanion
to the dianion. Supporting this assigment are (a) the fact that the values of in (peak
current) of E2° is similar to that for E1° (so it is likely to be a one electron process) and
(b) the presence of the oxidation wave E2’°, corresponding to E2°. This well-behaved
electrochemistry points to the reversibility of the reduction with regeneration of the

monoanion. If the dianion produced by one-electron reduction of monoanion underwent

prompt decomposition, regeneration of the monoanion would be impossible.

119

The separation between the first and the second reduction waves is rather high (in
the range of ~0.5 V and more). This shift points to the strong interactions between the
two frontier electrons in the dianion and rules out their localization on separate parts of
the molecule. Such a high separation between the first and second reduction potentials
precludes disproportionation of the monoanion to the dianion and neutral tetrone. It is
significant that the gap between the first two waves decreases with increasing electron-
donating abilities of the phenyl ring substituents (0.63, 0.57, 0.49 V for 2, (MeO)42,
(MezN)42 in DMF respectively). At first, one might expect that more electron-rich phenyl
rings would confine the frontier electrons to the tetrone unit and consequently strengthen
their mutual electrostatic interactions. Electron donation may, however (see chapter 2)
decrease electron-electron repulsion of frontier electrons by formation of the more and
more open-shell singlets.

The origins of the third and fourth waves cannot be assigned easily. The third one
is separated from the second by about 0.3V and is usually of much lower intensity (which
varies from tetrone to tetrone). For 2 it is especially small and can be observed only as a
shoulder. A physical phenomenon such as ion pairing or film deposition on the electrode,
may be responsible for this peak. Another explanation may be protonation of the dianion
by traces of water (or by the ethanol used for recrystallization of the supporting
electrolyte). The protonated dianion with its negative charge may then be quickly be
reduced, producing this extra third wave. The fourth wave has a much higher intensity
than the second one and two electrons are probably transferred (which would lead to the
formation of the tetraanion). This wave (like the third one) in all cases appears to be

irreversible. Apparent irreversibility must be caused by slow rate of the oxidation rather

120

than chemical decomposition of the formed species since dianion and monoanion are
recovered. An additional small oxidation wave E’o° can also be observed sometimes.

Oxidation of the carbon protonated anion may be responsible for this peak.

6.4 Experimental

THF was purified as described in Chapter 3. Anhydrous DMF (Aldrich) was used
as supplied. Supporting electrolyte [(n-Bu)4N](PF(,) (Aldrich) was recrystallized three
times from absolute ethanol and dried in vacuum. A calculated amount of [(n-
Bu)4N](PF6) (to produce 0.1M solution) and a magnetic stirring bar was placed in a main
chamber of an electrochemical cell, and the tetrone (to obtain ca. lmM solution) was
placed in the side chamber. The cell was evacuated on a vacuum line and ca. 2ml of THF
was condensed into the main chamber. When DMF was used as a solvent, it was
transferred to the cell with a syringe inside a glove bag, and then degassed in the cell via
several freeze-pump-thaw cycles. Vacuum distillation of DMF into the cell lowered its
quality, probably through its partial decomposition (to dimethylamine and carbon
monoxide). Cyclic voltammograms were recorded on a BAS CV-SOW Voltammetric
Analyzer with a glassy carbon working electrode, platinum wire counterelectrode and
silver wire as a quasi-reference. Before each experiment the carbon electrode was
polished with an alumina paste, and sonicated. A sweep rate of 0.1V/s was used in most
cases and 100% IR compensation was achieved before each run. Before the tetrone was
dissolved into the electrolyte solution, a few blank sweeps were done until no change in

the background current could be seen. After the runs with tetrone a small amount of

121

ferrocene was added to the solution in the glove bag in order to calibrate the potentials to

the Fesz/FeCpg+ couple.

122

(1)

(2)

(3)

BIBLIOGRAPHY

(a) Southampton Electrochemistry Group. Instrumental methods in
electrochemistry; E. Horwood; Halsted Press: Chichester New York, 1985. (b)
Chambers, J. Q. In The chemistry of the quinonoid compounds; Patai, S., Eds;
Interscience: London ; New York, 1974; p 737.

Almlof, J. E.; Feyereisen, M. W.; Jozefiak, T. H.; Miller, L. L. J. Am. Chem. Soc.
1990, 112, 1206.

Bock, H.; Hierholzer, B.; Jaculi, D. Z.Naturforsch. (B) 1988, 43, 1247.

123

Chapter 7

X-RAY DIF FRACTION STUDIES

7.1 Introduction

No other method of characterization of chemical systems can give a more
complete structural description than single crystal X-ray (or neutron) diffraction. Growths
of single crystals of both neutral and reduced forms of the tetrones were attempted in the
course of this study. In the case of the air stable neutral tetrones, this effort was a
relatively simple task. Crystallizations of the mono- and dianions were more challenging,
due to the low solubility of their metal salts in organic solvents, as well as their air
sensitivity. The solubility problem was solved by synthesis of (t-Bu)3(MeO)42, a
derivative ‘decorated’ with lipophilicw t-Butyl groups. Use of Schlenk techniques allowed
a slow diffusion of hexane into a THF solution of (t-Bu)g(MeO)42K, under oxygen-free
conditions and produced crystals of (t-Bu)g(MeO)42K-(THF)4. Growing of the crystals of

dianions still requires some effort.

7.2 Results and discussion

The most important crystallographic data are summarized in tables 7.1-4 and
Appendix B. Tetrone 2 crystallized from hot acetonitrile as dark red monoclinic needles
in P21/n. The unique atoms of the unit cell comprise half of the tetrone molecule 2, which
contains an inversion center. Molecules of 2 are arranged in 1: stacks with an

intermolecular separation equal to b (4.61 A).

124

Table 7.1. Selected crystallographic data for 2, (NMe2)42-(MeZSO)3, (t-

Bu)g(MeO)42K-(C4H30)4 and (t—Bu)g(MeO)42Na°(C6H1002);.

 

 

 

2 (NM62)42 (l-BU)3(MCO)42K (I-BU)3(MCO)42N3
'(M6230)3 '(C4H80)4- ’(C6H1002)4
chem formula C32H2004 C46H53N4O7S3 C34H1240|2K C163H264032Na2
fvv 468.48 875.16 1364.99 2841.90
cryst syst Monoclinic Monoclinic Triclinic Triclinic
space group P2l/n P21/n P1 P1
3., A (Mo K01) 0.7107 0.7107 0.7107 0.7107
a, A 13.8607(14) 21.092(4) 9.5055(2) 16.1411(4)
b, A 4.6091(5) 8.0588(16) 13.0026(2) 16.3146(2)
c, A 19. 1 2 1 (2) 26.781(5) 17.1668(2) 18.8482(4)
01, deg 90 90 77.2809(11) 64.4540(10)
[3, deg 109.922(2) 101.31(3) 85.5001(12) 72.4140(10)
y, deg 90 9O 77.64l7(10) 88.4470(10)
V, A 1148.4(2) 4463.7(15) 2020.64(7) 4238.05( 15)
Z 2 4 1 1
dew, g cm'3 1.355 1.302 1.384 1.114
p, mm" 0.089 0.176 0.182 0.085
T, K 173(2) 293(2) 293(2) 293(2)
R 0.0493 0.0690 0.0844 0.0728
pr2 0.0952 0.1743 0.2430 0.2082
GOF 1.046 0.856 0.910 0.932
R a R
b

Figure 7.1 Symbols of bonds and dihedrals used in tables 7.2 and 7.3.

125

Table 7.2 Selected bond distances for tetrones and their monoanions.

 

 

(NMe2)42 (t-Bu)3(MeO)42Na (t-Bu)g(MeO)42K
C4-C5 1.480(3) Cll-Cl 1.464(4) C11-C12 1.481(9) C7-C8 1.477(7)
C4-C11 1.472(3) C11-C2 1.453(4) C11-C28 1.502(9) C7-C28 1.465(8)

C30-C3 1.458(4) C44-C45 1.477(9) C44-C4 1.482(7)
C30-C4 1.475(4) C44-C61 1.463(9) C44-C6 1.457(8)
C87-C88 1.490(9)
C87-C104 1.469(8)
C120-C121 1.482(8)
C120-C142 1.486(9)
b C3-C4 1.388(3) C1-C11 1.440(4) C2-C11 1.393(9) Cl-C44 1.440(8)
C4-C30 1.429(4) C5-C44 1.442(8) C4-C7 1.398(8)
C78-C87 1.442(8)
C81-C120 1.408(9)

c C1-C3A 1.471(3) C1-C2 1.455(4) C1-C2 1.478(8) C1-C6 1.470(7)

C2-C3 1.476(3) C1-C6 1.459(4) C2-C3 1.486(9) C1-C2 1.434(8)
C4-C3 1.464(4) C4-C5 1.453(9) C4-C5 1.464(9)
C4-C5 1.477(4) C5-C6 1.452(9) C3-C4 1.449(7)

C77-C78 1.457(8)

C78-C79 1.478(8)

C80-C81 1.479(9)

C81-C82 1.451(9)
d C1-C2 1.540(3) C2-C3 1.559(4) C1-C6 1.529(9) C2-C3 1.547(8)
C5-C6 1.545(5) C3-C4 1.549(8) C5-C6 1.492(8)

C77-C82 1.512(9)

C79-C80 1.572(8)

e Cl-Ol 1.226(2) C2-07 1.240(4) C1-O7 1.219(7) C2-025 1.236(7)

C2-02 1.212(2) C3-O8 1.237(4) C3-OlO 1.207(7) C3-027 1.228(7)
C5-09 1.231(4) C4-O9 1.258(7) C5-024 1.229(7)
C6-010 1.234(4) C6-08 1.263(8) C6-026 1.233(7)

C77-O84 1.254(7)

C79-085 1.228(7)

C80-O86 1.219(7)

C82-083 1.245(8)

 

126

Table 7.3 Selected dihedrals for tetrones and their monoanions.

 

 

2 (NMe2)42 (t-Bu)g(MeO)42K (t-Bu)g(MeO)42Na
a Ol-Cl- 26.9 O7-C2- -20.2 OZ4—C5- -1.7 O8-C6— 8.2
C2-02 C3-08 C6-O26 Cl-O7
O9-C5- -20.9 O27-C3- 3.6 OlO-C3- -6.2
C6-OlO C2-025 C4-O9
O85-C79- 7.8
C80-O86
O83-C82- -6.8
C77-O84
[3 O2-C2- -26.2 OlO-C6- 15.2 024-C5- -0.8 O7-C1- -14.0
C3-C4 C 1-C1 1 C4-C7 C2-C11
OlA-ClA- 37.6 Cll-Cl- 10.4 C7-C4- -11.6 C11-C2- 2.6
C3-C4 C2-O7 C3-027 C3-OlO
O8-C3- 8.4 025-C2- 2.1 O9-C4- 14.4
C4-C3O C1-C44 C5-C44
C30-C4- 3.3 C44-C1- 8.2 C44-C5- -5.3
C5-O9 C6-O26 C6-O8
O86-C80- -3.5
C81-C120
C120-C81- 13.1
C82-O83
O84-C77- 6.5
C78-C87
C87-C78- -16.9
C79-O85

 

127

Table 7.4 Angles of the mean ring planes vs. the planes of the methylene centers of

tetrones and their monoanions.

 

 

2 C[3-5,11]
C[5,10] 38.8
C[11,16] 43.9
C[1-3,1A-3A] 35.0

 

 

(NMez)42 C[4,30,3 1,40] C[1,1 1,12,21]
C[40-45] 36.4

C[31-36] 31.3

C[1-6] 43.1 30.4
C[12-17] 34.0
C[21-26] 28.5

 

(t-Bu)g(MeO)42K

C[4,18,28,7] C[1,44,45,61]

 

C[8-13]
C[28-30,35,38,43]
C[1-6]

C[45-50]

C[61-6]

30.5

33.5

40.9 41.3
34.6
29.8

 

(t-Bu)3(MeO)42Na

C[2,11,28,12] C[5,44,45,61]

 

C[12-17]
C[28-33]
C[1-6]
C[45-50]
C[61-6]

31.5

36.8

36.1 35.9
38.6
31.1

 

(t-Bu)g(MeO)42Na

C[78,87,88,104] C[120,121,142,81]

 

C[88-93]
C[104-109]
C[77-82]

C[121-126]
C[137-142]

37.5

28.3

37.1 37.5
28.2
38.2

 

128

Tetrone (N Me2)42 crystallized from DMSO as dark blue blocks of
(NMe2)42-(DMSO)3. The unique atoms of the unit cell include one molecule of (NMe2)42
and three solvent molecules of DMSO. Molecules of (NMe2)42 are arranged into sheets
with relatively short intermolecular contacts (OS-N37A 3.256A and OlO-N27B 3.260A)
between two of their carbonyl oxygens and nitrogen atoms of two neighboring tetrones.
Additionally two of its own nitrogen atoms (N37 and N27) interacts in similar manner
with carbonyl oxygens of the another two neighbors. The change of the packing mode of
2 vs. (N Me2)42 may be explained by the transfer of charge between the electron-donating
nitrogens and the electron-accepting oxygen and electrostatic polarization of the
molecule. Intermolecular coulombic interactions in the crystal of (N Me2)42 also explain
its, contrasting to 2, low solubility in non polar solvents. The geometry of single
molecules of 2 and (N Me2)42 are presented on figure 7.2 and 7.3. The central 3,6-
dimethylene-l,2,4,5-tetraoxocyclohexane ring of 2 is puckered into a chair-like
conformation, with the unique dicarbonyl dihedral OCCO angle being 26.9°. These
distortions are probably driven by coulombic repulsion between the exocyclic C=O bond
dipoles. The tetrone ring of (N Me2)42 is slightly different than that of 2. Instead of
adopting a pseudo-chair conformation, this molecule twists around the axis defined by

the two methylene carbons. The values of the OCCO dicarbonyl dihedral angles (20.9°

and 202°) are however similar to that of 2. Aryl and tetroxalane rings of 2 and (NMe2)42,
which are connected to the same methylene carbon centers form a propeller-like
arrangement. The steric and electronic differences between the phenyl rings and the one
of tetrone, which is formally doubly bonded to the methylene center, do not seem to have

great influence on their out-of-plane twists.

129

 

Figure 7.2 ORTEP drawing (50% probability) of 2. Hydrogen atoms are not shown.

130

‘5 C19

N18
015 ( "C14
C16
6" C13 039
C17 i ,,~ ‘
h
.1 a "
i {/4
At 024 . .
\l P ‘w a 010 c
029 i" «s 022
N27 023
g‘\
C28

 

Figure 7.3 ORTEP drawing (50% probability) of (NMe2)42°(Me2SO)3. Hydrogen atoms

and DMSO molecules are not shown.

131

The angles defined as that between mean plane of all carbons of the rings and mean plane
of four carbons of the ‘methylene center’ are around 30°-40° (table 7.4).

Slow diffusion of n-pentane into a THF solution of (t-Bu)g(MeO)42K, under
oxygen-free conditions produced purple needle-like triclinic (P1) crystals of (t-
Bu)3(MeO)42K-(THF)4 with one molecule of the complex in the cell space group. A
similar technique (using n-pentanefDME) produced crystals of (t-
Bu)g(MeO)42Na-(DME)4’ with the same color and symmetry but two unique molecules
of the complex in asymmetric unit. In both of the above structures, the organic anions
link alkali metal cation to form l-D chains (figures 7.4 and 7.5). In these chains, the
anions display the hoped-for chelating and bridging modes of coordination. In (t-
BU)3(MCO)42K each potassium cations in the chain forms a pseudo-cubic sphere
coordinating with four carbonyl oxygens from two tetrone molecules and four THF
oxygens. Periodicity of the crystal implies that the central ‘tetrone’ rings of all of the
anions are parallel to each other. There is however a 'zig-zag' angle of 148.2 ° between
mean planes defined by four oxygens of tetrone rings and four carbonyl oxygens
coordinated to the potassium cation. The sodium salt forms similar structure. In this case,
however, there are two independent alternating monoanions in the chain and the pseudo-
cubic coordination sphere of the sodium cation (containing four tetrone oxygens four
oxygens of two chelated DME molecules) is distorted to form a pseudo square antiprism.
This distortion rotates plane of one of the unique monoanions around the chain axis so

the angle between its central mean planes and that of the neighbor is 376°.

 

. The crystal of the sodium salt was grown by Robert Gentner

132

 
 
   
    
 

C[28-30,35.
38,43]

4

C[45-50]

C[61 -66]

Figure 7.4 ORTEP drawing (50% probability) of chains formed in the solid state by
radical anion salt (t-Bu)g(MeO)42K-(THF)4 with one unique molecule of
complex. Hydrogen atoms, THF molecules (excluding coordinated oxygen

atoms) t-butyl and metoxy fragments, are not shown.

133

01104-109]

 

C[121-126]
C[61-66]

Figure 7.5 ORTEP drawing (50% probability) of chains formed in the solid state by
radical anion salt (t-Bu)g(MeO)42Na-(DME)4 with two unique molecules of
complex. Hydrogen atoms, DME molecules (excluding coordinated oxygen

atoms), t-butyl and metoxy fragments are not shown.

134

In addition to four unique coordinated molecules of DME there are also four
noncoordinated ones located in the cavities between the chains. Similar modes of
coordination found in both crystals confirms, as expected, that the negatively charged
pocket formed by two oxygen atoms on one side of the anion is the preferred place for
metal cation binding. Both potassium and sodium salts have also polymeric nature. It
may be attributed not only to the design of anion but also to relatively low coordinating
power of solvents used for crystallization. THF and DME are probably too weak Lewis
bases to compete for coordination sphere with chelating anions of 2. Use of strongly
donating solvent could break chains into separate ion aggregates. Such behavior has been
observed for potassium radical anion salt of 2,3-bis(2-pyridyl)quinoxaline, which from
THF crystallized as 1-D chains of (dpq)K-(THF)2 but from methylamine as discrete
solvent separated ion-pair dimmers of (dpq)K°(MeNH2)2.l Obtained organometallic
structures have charge balance of the coordination equal to zero. We hope that increasing
of the cation charge will force more of the anions to coordinate to it in order to preserve
zero or as close to zero as possible balance. This in turn may induce formation of 2-D
networks based on chessboard or honeycomb motifs.

Reduction of the tetrone notably flattens out its central ring and the values
dicarbonyl dihedrals are now below 82° (table 7.3). There is however not much change
in the propeller-like arrangements of the rings with angles of twist retaining values of 30-
40° (table 7.4). Selected bond lengths of neutral and reduced forms of tetrone are
presented in table 7.2. Relatively low changes of the bond lenghts of the monoanions vs.
tetrones does not allow for unambiguously localize of distribution of the unpaired

electron. Despite repeated attempts, so far we were unable to grow crystals of dianions of

135

2 and monovalent alkali metal cations suitable for X-ray analysis. Hopefully use of

doubly divalent cations will lead to chains of dianions analogous to that of monoanions.

7.3 Experimental

Crystals of compound 2 were grown by slow cooling of its acetonitrile solution.
Crystals of (N Me2)42-((CH3)ZSO)3 formed over time from oversaturated DMSO solution
of (N Me2)42. To produce crystals of (t-Bu)g(MeO)42K-(C4HgO)4, 50 mg of tetrone was
dissolved in 30-50 ml of degassed THF and reacted wit potassium mirror following the
procedure described for preparation of EPR samples. Solution of the potassium salt of
anion was then transferred to two 9mm diameter Pyrex glass tubes attached to the K-cell.
Pentane was condensed over it and and allowed to diffuse for a couple of days in room
temperature. Crystals of (t-Bu)g(MeO)42Na-(C4H1002).; were grown similarly but DME

was used instead of THF.

136

BIBLIOGRAPHY

(1) Ichimura, A. 8.; Szajek, L. P.; Xie, Q. S.; Huang, R. H.; Huang, S. Z.; Wagner,
M. J.; Dye, J. L.; Jackson, J. E. J. Phys. Chem. B 1998, 102, 11029.

137

Chapter 8

ELECTRONIC SPECTROSCOPY

8.1 Introduction

The tetrones and their anions described in earlier chapters were also characterized
by means of electronic spectroscopy. UV-VIS (ultraviolet and visible) spectra in the
range of 200-1000 nm were recorded for neutral tetrones in acetonitrile. In addition, their
reduction with in-situ prepared [K+@C(2.2.2)]K' in THF was monitored by VIS-NIR

(visible and near-infrared) spectroscopy (400-2000 nm).

8.2 Results

Neutral tetrones. The electronic absorption profiles of neutral tetrones in the UV-
VIS region (table 8.1, figure 8.1) comprise of a set of peaks in the UV region and intense
visible absorption bands. In the case of 2, the two bands coalesce into a single broad
band.

Monoanions and dianions. VIS-NIR spectra of mono- and dianions prepared by
reduction of tetrones with [K@c(2.2.2)]+K' in THF were also recorded (figures 8.2, 8.3;
table 8.1). Solutions of 2' are dark wine-red; (MeO)42' and (t-Bu)3(MeO)42'are purple
(similar to aqueous solutions of permanganates) while solutions of (MezN)42' are green.
In addition to the bands in the visible region, all monoanions possess a distinctive broad
absorption band in the near infrared region, with the maximums at 1000, 1100, 1150,
1500 nm for 2’, (MeO)42', (t-Bu)g(MeO)42', and (MezN)42', respectively. The main
absorptions in the VIS-NIR spectra of the dianions are concentrated around 800 nm and

138

are much less affected by the phenyl rings' substituents, so the color of all dianion

solutions is emerald-green.

Table 8.1. Absorption maxima of tetrones their mono- and dianions.

 

 

 

Xmax (nm)
Tetrone Neutral Monoanion Dianion
(MeCN) (THF) (THF)
2 354 430 485,
525 800
1000
(MeO)42 498, 530 745
580 (shoulder) 600 (shoulder)
1100
(t-Bu)8(MeO)42 480 525, 770
610 (shoulder)
1150
(MezN)42 602, 425, 770
674 635,
1500

 

8.3 Discussion

Neutral tetrones. Since all tetrones possess electron-donor and electron-acceptor

moieties, the visible wavelength absorption may be assigned to an intrarnolecular charge-

traIleer transition from the donor diarylmethylene fragment to the acceptor tetrone unit.

The bathochromic shift in the series 2, (MeO)42 and (MezN)42 can be explained by the

increase in the electron-donating abilities of substituents in the order —H, -OMe, -N(Me)2.

A similar trend can be seen for example in the case of substituted diphenylmethyl

carbocations (1,“, = 440(H-), 507(MeO-), 610(Me2N-) m)-'

139

| 10000

(MezN)42

 

(MeO)42

 

(t—Bu)3(MeO)42

 

2

 

 

 

r I l I l I f 1
200 300 400 500 600 700 800 900 1000

A1mm)

Figure 8.1. UV-Vis absorption spectra of neutral tetrones (acetonitrile)

140

    

(t—Bu)3(MeO)42'

2.

 

 

 

I I l I I I I I

400 600 800 1000 1200 1400 1600 1800 2000

A(nun)

Figure 8.2. Vis-NIR spectra of the THF solutions of monoanions.

141

 

 

 

 

 

A (MezN)422'
(Me0)422'

(t-Bu)3(MeO)422‘

22.

 

400 600 800 1000 1200 1400 1600 1800 2000

A(nm

Figure 8.3. Vis-NIR spectra of THF solutions of dianions

142

The increase in donating abilities not only changes the energy of the transition but
also its intensity. The extinction coefficient of (MezN)42 is very high (ca. 70 000) making
that compound an intense blue dye, while the absorptions of (MeO)42 and 2 (purple and
red) are weaker. The charge-transfer band of (MezN)42 possesses two relatively close
maxima at 602 and 674 nm. In (Me0)42, this splitting is much less pronounced and in
addition to one maximum (498 nm) there is only a shoulder at 580 nm. Compound 2 has
only one maximum (3 54 nm) but a broad plateau which then tails over 500nm. Splitting
of the absorption may be attributed to the near degeneracy of LUMO and LUMO+1.

Monoanions. The most distinctive absorption of monoanions is a broad band in
the near-IR region. The energy of this transition decreases with increasing electron-
donating power of the aryl substituents, indicating that the electron is excited from an
orbital conjugated to them into another that is relatively less affected. This correlation

and the low energy of the transition would be in a good agreement with a SOMO to

LUMO transition.

8.4 Experimental

UV-vis spectra of the tetrones were recorded with UNICAM UV2 in acetonitrile
as the solvent. The mono- and dianions of the tetrones were prepared in THF in the
presence of the C(2.2.2) cryptand in the manner described in section 4.4. The progress of
reduction was monitored within the optical cell attached to the sidearm with a Guided

Wave Model 260 fiber-optic spectrophotometer.

143

BIBLIOGRAPHY

(1) Deno, N. C.; Jaruzelski, J. J .; Schriesheim, A. J. Am. Chem. Soc. 1954, 77, 3044.

144

Chapter 9

SUMARY AND CONCLUSIONS

The structures of mono and dianions of the terone l were proposed as discrete
building blocks for assembly of molecular magnetic materials at the beginning of this
work. Examination of their idealized and real behaviors composed the rest of it. Three

major questions have been investigated in the course of this study:

0 Can these anions and their neutral precursors be practically synthesized and
handled?
0 May derivatives of 12' have high-spin ground states?

0 Will they tend to crystallize forming the hoped-for extended organometallic

structures?

Chapter 3 has shown that synthesis of the neutral tetrone precursors can be
achieved when aryl rings protect the reactive methylene groups (group of tetrones 2). The
synthetic route developed is relatively straightforward (two-steps), cost effective, and
allows fast preparation of a broad variety of products from easily available synthons.
Substantial progress toward extending the family of tetrones to alkyl substituted
derivatives has also been achieved.

Electrochemical studies have shown that tetrone 2 and its derivatives can be

reversibly reduced in two successive one electron processes to mono and dianions. These

145

results point to the stability of both mono and dianions of 2, at least on the time-scale of
the cyclic voltammetry experiment (seconds to minutes).

With the above facts established, the bulk of the work focused on the preparation
and determination of the ground state of the dianions (with observation of the triplet state
as a main goal). Those studies started with high quality ab-initio CASSCF calculations of
the energies of the lowest singlet and triplet states of derivatives of 12' (chapter 2), The
calculations confirmed what was predicted earlier by qualitative methods in chapter 1,
that the singlet-triplet gap is small in those dianions. They also showed that control over
the ST gap could be achieved by structural modifications. For example, the strong (ca.
10 kcal/mol) calculated preference for the singlet in the noncoordinated, nonsubstituted
dianion 12' is replaced by a 5 kcal/mol preference for the triplet state in the tightly
coordinated one. Despite the relatively high level of these calculations the values of the
absolute numbers obtained must be treated with skepticism. The most important outcome
should be considered to be the understanding of the mechanism of the S-T tuning and the
assessment of its approximate magnitude.

Experimental work was undertaken complementary to the calculations.
Spectroscopic studies could not be performed with exactly the same species as those used
for calculations. The available phenyl substituted derivatives 2 had to be used instead. In
turn, high quality theoretical study could not be repeated on the tetraaryl substituted
derivatives of 2; the large number of atoms in those molecules placed such computations
beyond the practical scope of available quantum chemistry software.

EPR study of the noncoordinated derivatives of 2' prepared by reduction of

tetrones with potassium in the presence of macrocyclic coligands confirmed their

146

expected structure. Unfortunately no triplet signal could be detected for dianions prepared
in a similar manner. NMR studies could not however prove that they are closed shell
singlets.

The failure to observe a triplet signal was not surprising, since the earlier
calculations had already pointed to the necessity of tuning the electronic structure of the
dianions by strong metal cation coordination in order to achieve triplet preference. The
additional electronic perturbations introduced by the presence of the aryl groups may also
contribute to the relative preference for the singlet state.

Solution studies of coordinated complexes (M-2»M)"+ were considered and
attempted but they turned out to be difficult to perform. In addition to solubility
problems, the required stoichiometry and geometry of the complexes formed in solution
could not be ensured. The anions and the cations might prefer unanticipated modes of
association. This fact could not only prevent generation of the required complexes, but
also lead to misleading ‘positive’ results. For example triplet EPR signals were generated
by two monoradicals interacting with each other through Cs+ in the complex (2C32)'. The
presence of such monoradical aggregates could not be excluded in solution even when
great care was taken to control stoichiometry of the reduction since monoanions may be
formed by decomposition of dianions.

An attempt to solve this problem by preparation of ion triples 2(SnBu3)2 was
undertaken (Figure 9.1). UV irradiation of a mixture of hexabutylditin (SnzBu6) and
tetrone 2 was supposed to break weak Sn-Sn bond with formation of the trialkyltin
radicals. Two equivalents of Bu3Sn- radicals were then expected to add to one of the

tetrones with formation of a tightly coordinated complex of 2. The bulky butyl groups of

147

the tin cations were supposed to prevent coordination of two of them to one side of the

anion or coordination of two anions to one cation, ensuring formation of only 2(SnBu3)2.

P Ph

0 O (BU)3SI’I—SI’I(BU)3 O

hv (Bu)3+Sn::
O O O

 

Ph Ph

 

2

Figure 9.1. Photochemical generation of 2(SnBU3)2.

Photolysis of neat 2 and SnzBu6 indeed lead to the mixture of products which
besides strong EPR doublet, produced a triplet signal with a D value of ca. 200 G,
pointing to strong intramolecular interactions of the electrons. These results could not be
accepted as a proof that coordination of 22' leads to generation of triplet however,
because no information concerning the structure of the species responsible for the signal
could be obtained. Consequently these results have not been detailed in this thesis.

To conclude, EPR results alone, without good structural characterization, can not
unambiguously provide an answer to the question concerning the ground state of such
labile complexes. Definitely, an alternative approach must be chosen. Any spectroscopic
study must follow crystallization of the complex and its X-ray structure determination.
Only with such exact structural results in hand will interpretation of EPR or SQUID data
be meaningful.

An attempt to follow these principles has been undertaken in the latter part of

these studies. The problem of low solubility of the complexes has been solved by

148

synthesis of the t—Butyl substituted analog of the tetrone (MeO)42. The lipophilic butyl
groups increased the solubility of both the tetrone and its complexes in the relatively non-
polar but chemically inert and easy to dry ethereal solvents used in the reductions. Then
anaerobic techniques of crystal growing were applied. Despite these efforts, all attempts
to grow crystals of dianion were unsuccessful. The apparent reason was the slow
decomposition of the dianion at room temperature. A temperature dependent kinetic
study of the decomposition should be performed and, if a safe threshold of temperature
can be found, temperature-controlled crystallization technique should be applied in order
to achieve this goal. It is also likely that the stability of dianion salts will be increased by
coordination to some metal cations.

In the meantime the crystallization effort of the air sensitive, but room
temperature stable monoanion salt (t-Bu)g(MeO)42K has been successfully accomplished.
Analogously to dianions, the monoanions are viable building blocks for assembly of
molecular magnets. The ID chain structures of their potassium and sodium salts have
proved their tendency to form extended structures. The question, which remains to be
answered, is whether the dimensionality of the net may be increased.

Only a small number of paramagnetic bridging ligands has been structurally
characterized so far. Most of them are variations of neutral nitroxides or nitronyl-
nitroxides, species of relatively high stability but with modest coordination abilities. The
anions investigated in this work have properties that set them apart from this group and
that may allow their application beyond the field of molecular magnets. Besides being
anionic and having a robust chelating mode of coordination, they posses low and tunable

reduction potentials. This tunability may allow assembly of systems with redox potentials

149

of the monoanions and the transition metal centers they are coordinated to being almost
equal. An example where such equilibrium led to interesting phenomenon is valence
tautomerism found in structurally related semiquinone complexes of transition metals. In
certain temperatures phase transition may be observed where electrons cooperatively
transfer from metal to semiquinone centers. It would be interesting to observe similar
phenomenon in the extended rather then discrete systems and investigate its

consequences.

150

APPENDICES

APPENDIX A

Table A1. Geometry of the unique atoms of 12' and its complexes optimized at indicated
level of geometry (6-31G* basis set)

TC SCF
12'
C1 0.0000000000 1 .4585402082 0.0000000000
C2 1 .2998449695 0.72643 14308 0.0000000000
C3 0.0000000000 2.795 193 1 157 0.0000000000
Ol 2.3581524018 1.3670568670 0.0000000000
H1 0.927171 1617 3.3373483434 0.0000000000
lLiz

C1 0.0000000000 1 .486208421 8 0.0000000000
C2 1.26375 84273 0.7181900799 0.0000000000
C7 0.0000000000 2.8126012595 0.0000000000
OI 2.37085 73392 1.3129967865 0.0000000000
H1 0.923 1996774 3 .3580798268 0.0000000000
Lil 3.6294109899 0.0000000000 00000000000

1(BeF)2
C 1 00000000000 1 .5 1 063 24296 00000000000
C2 1.2318914718 0.7115756470 00000000000
C3 00000000000 28371442752 00000000000
01 2.392461 1694 1.2201339304 00000000000
H1 0.9229260360 3 .3 845033 134 00000000000
Bel 3.4382197582 00000000000 00000000000
F1 4.828373 8502 00000000000 00000000000

1(A1F2)2
C 1 00000000000 1 .4962442533 00000000000
C2 1.2408255712 0.7108880983 00000000000
C3 00000000000 2.8239066235 00000000000
01 2.3859764059 1.2604845714 00000000000
H1 0.9225973324 3.3714061945 00000000000
All 3 .7343 72593 1 00000000000 00000000000
F 1 4.5714039244 00000000000 1 .404391 1054

152

Table A1. (con’t)

C1
C2
C3
01
F1
Lil

C1
C2
C3
01
02
H1
Lil

C1
C2
C3
H1
H2

C1
C2
C3
01
H1

C1
C2
C3
01
H1
L11

00000000000
1.2735121843
00000000000
2.3 866780569
10503264806
3.6551654006

00000000000
1.2709734910
00000000000
2.3981502585
1.1204651797
0.9584113899
3.6317863922

00000000000
1.0092311570
00000000000
0.9186088459
20827383088

00000000000
1.3290811444
00000000000
2.3774333686
0.9283698910

00000000000
1.2742728843
00000000000
2.3807832515
0.9238015580
3.6287246543

F41Li2

1.4729477875
0.7157051338
2.8019780573
1 .3004466127
3.5540764980
00000000000

(OH)41Li2

1 .4664843050
0.7154595384
2.8225747535
1.2970816889
3.5359119172
4.4678444736
00000000000

DMCB

10787892346
00000000000
2.4005871649
2.9579997044
00000000000

ROHF

12'

1.4300354546
0.7189049086
2.7624645029
1 .3 890045378
3.3027945771

lLiz

1.4597239578
0.7151761290
2.7914967525
1.3209204289
3.3360806323
00000000000

153

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

Table A1. (con’t)

C1

C2
C3
01

H1
Bel
F1

C1
C2
C3
01
H1
A11
F1

C1
C2
C3
01
F1
Lil

C1
C2
C3
01
02
H1
Lil

C 1
C2
C3
H1
H2

00000000000
1.2336313938
00000000000
2.3964403160
0.9234001510
3.43 799263 04
4.8294428556

00000000000
1.2388160420
00000000000
2.3860103334
0.9230116297
3.7308259694
4.5714154155

00000000000
1.2718860074
00000000000
2.3863526663
10493539348
3.6509184484

00000000000
1.2593115219
00000000000
2.387603 8265
1.1241263076
0.9601131859
3.6261648579

00000000000
10277765938
00000000000
0.9178218164
2.0986331928

1(BeF)2

1 .4901642007
0.7101789596
2.8248970023
1.2217418445
3.3714283097
00000000000
00000000000

1(A1F2)2

1.4805812799
0.7105045043
2.8162899820
1.2615779183
3.3627576016
00000000000
00000000000

F41Li2

1.4616977906
0.7159750349
2.7969827473
1.3014250648
35470741672
00000000000

(OH)4lLi2

l .475 8668281
0.7186689653
2.8323695130
1.2919942984
3.5474867515
4.4788542793
00000000000

DMCB

1 .0417794666
00000000000
2.3757180133
2.9346318045
00000000000

154

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000
1.4030614736

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000

Table A1.(con’t)

C1
C2
C7
01
H1

C1
C2
C7
01
H1
Lil

C1

C2
C3
01
H1
Bel
Fl

C1
C2
C3
01
H1
All
F1

C1
C2
C3
01
F1
Lil

00000000000
1.2878403067
00000000000
2.3511309375
0.9266028675

00000000000
1.2581525153
00000000000
2.3663778658
0.9231942774
3.6270196004

00000000000
1.2276350128
00000000000
2.3883018786
0.9228164723
3.4350293267
4.8248598191

00000000000
1 .2370605 882
00000000000
2.3824318708
0.9226590394
3.7320189079
4.5692581898

00000000000
1.2700650861
00000000000
2.3838247558
1.051 1862047
3.6542153381

'Ag CAS(2,3)

12'

1.4643995291
0.7341861650
2.8089987094
1.3584657248
3.3518775913

lLiz

1.4889560523
0.7209446130
2.8175192391
1.3106334147
3 .3627989200
00000000000

1(BeF)2

1.5135429884
0.7135499215
2.8411787046
1.2191883246
3 .3 884466798
00000000000
00000000000

1(A1F2)2

1.4991686030
0.7124864545
2.8275648418
1.2593374055
3.3746701952
00000000000
00000000000

F41Li2

1 .475 7968938
0.7170669492
2.8053083922
12988504922
3.5576903021
00000000000

155

00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000
1.4039421416

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

Table A1. (con’t)

C1
C2
C3
01
02
H1
Lil

C1
C2
C3
H1
H2

C1
C2
C3
H1
H2

C1
C2
C3
01
H1

C1
C2
C3
01
H1
Lil

00000000000
1 .2694409284
00000000000
23966448634
1.1211724752
0.9584481125
3.6316457257

00000000000
10169682221
00000000000
0.9185037520
2.0901298564

00000000000
1.0249892104
00000000000
0.9186615356
20962395269

00000000000
1.2655812247
00000000000
2.3490236649
0.9248032859

00000000000
1.2428262829
00000000000
2.3558276543
0.9224347453
3.6258226865

(OH)41Li2

1 .4687524044
0.7161244466
2.8243 162664
1.2961760235
3.5382952889
4.470051 1956
00000000000

‘Ag CAS(6,6)

DMCB

1.0786299819
00000000000
2.4161951606
2.9739018425
00000000000

383., CAS(6,6)

DMCB

10365581099
00000000000
24014242740
2.9580096342
00000000000

'Ag CAS(14,12)

12'

1.4649397806
0.7432509967
2.8483407587
1.3461741037
3.3946503398

lLiz

1.4943237524
0.7256283108
2.8508866342
1.3097708136
3.3964855867
00000000000

156

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

Table A1. (con’t)

C1
C2
C3
01
H1
Bel
Fl

C1
C2
C3
01
H1
A11
F1

C1
C2
C3
01
F1
Lil

C1
C2
C3
01
OZ
H1
Lil

C1
C2
C3
01
H1

00000000000
1.2148486562
00000000000
2.3755365600
0.9226354584
3 .4293 865 l 76
4.8160595613

00000000000
1.2310267488
00000000000
2.3755593767
0.9226181803
3.7304268525
4.5593640311

00000000000
1.2568926215
00000000000
2.3724017689
10629503359
3.6585160270

00000000000
1 .2567600102
00000000000
2.3817774451
1 .1 380228764
0.9603 504400
3 .6408808276

00000000000
1.2482405271
00000000000
2.3466013428
0.9244125411

1(BeF)2

1.5239991145
0.7184371267
2.8750002228
1.2193241104
3.4210357464
00000000000
00000000000

1(A1F2)2

1.5042538191
0.7131461393
2.8513074834
1.2590872679
3.3970304957
00000000000
00000000000

F41Li2

1.4899949986
0.7203755377
2.8347572722
1.2971269441
3.5915778088
00000000000

(OH)4lLi2

1 .4984561995
0.7205389868
2.8549084322
1.2906088981
3.5821861824
4.5100369316
00000000000

332.. CAS(14,12)

12'

1.4746877122
0.7691125605
2.9087307647
1.3325818677
3.4548530733

157

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000
1.4072213033

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000

Table A1. (con’t)

C1
C2
C3
01
H1

C1
C2
C3
01
H1
Lil

C1
C2
C3
01
H1
Bel
F1

C1
C2
C3
01
H1
All
F1

C1
C2
C3
01
F1
Lil

00000000000
1.3195140226
00000000000
2.3779492532
0.9268519725

00000000000
1.2718134185
00000000000
2.3789577155
0.9231985572
3.6324437579

00000000000
1.2324534653
00000000000
2.3940258576
0.9230525819
3.4403335535
4.8295731263

00000000000
1.2372669101
00000000000
2.3832384265
0.9234086492
3.7327178583
4.5683921087

00000000000
1.2717836859
00000000000
2.3854408253
10587872686
3.6606005600

3B3, CAS(14,12)

12-

1.4323836535
0.7177758626
2.7888564213
1.3869955945
3.3321464548

1Li2

1.4579442578
0.7156580354
2.8145907821
1.3231370418
3.3591716132
00000000000

1(BeF)2

1.4885560369
0.7114097219
2.8466127037
1.2231498958
3.3920404412
00000000000
00000000000

1(A1F2)2

1.4771337693
0.7101475026
2.8349531143
1.2614441489
3.3791837973
00000000000
00000000000

F41Li2

1.4620300807
0.7164236700
2.8125790464
1.3034674015
3.5663395965
00000000000

158

00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000
1.4046532163

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

Table A1. (con’t)

C1
C2
C3
01
02
H1
Lil

00000000000
1.2607441 152
00000000000
2.3865308166
1.1352676445
0.9612206034
3.63 80894530

(OH)4lLi2

1.4772347650
0.7194356555
2.8430876436
1.2926178509
3.5661355725
4.4949824284
00000000000

159

00000000000
00000000000
00000000000
00000000000
00000000000
00000000000
00000000000

Table A2. CAS(2,2) Frequencies (cm") of vibrations of 12' and its complexes in singlet
(TCSCF) and triplet (ROHF) state. Starred values are translations and
rotations of molecules

 

 

12‘ 11.12 1(BeF)2

TCSCF ROHF TCSCF ROHF TCSCF ROHF
1 6.43* 4.82* 251* 4.48* 309* 203*
2 082* 040* 219* 015* 026* 078*
3 016* 005* 021* 009* 020* 026*
4 002* 005* 008* 004* 006* 010*
5 278* 382* 005* 051* 1.57* 010*
6 628* 444* 279* 1.96* 1.59* 233*
7 42.26 52.25 52.40 67.07 41.96 46.25
8 46.87 78.26 104.53 112.39 88.98 89.59
9 55.68 85.07 108.62 119.71 94.72 100.54
10 277.71 278.37 158.81 158.86 138.59 145.45
11 362.90 369.43 191.13 194.08 143.09 145.87
12 367.88 378.78 299.06 304.77 151.50 152.13
13 380.98 395.85 322.40 327.53 188.47 189.99
14 421.44 411.10 340.76 339.02 233.82 237.94
15 439.69 440.82 340.81 345.31 298.38 300.11
16 446.83 444.67 400.21 422.39 310.46 321.22
17 460.30 452.35 433.49 441.15 335.40 336.66
18 461.80 461.34 441.40 444.37 389.96 391.31
19 500.96 468.80 458.45 456.58 409.93 407.35
20 593.36 580.91 466.58 466.53 439.90 442.79
21 618.56 681.42 470.19 468.66 459.08 460.04
22 697.58 697.32 582.24 588.47 463.25 491.47
23 700.50 710.34 599.95 600.83 488.42 497.75
24 708.59 761.96 636.17 640.58 508.95 512.07
25 798.05 790.64 666.75 667.72 530.71 533.51
26 829.01 835.65 669.62 675.27 647.94 652.83
27 830.42 855.99 688.78 684.19 650.45 664.90
28 925.81 917.37 693.48 705.15 661.30 665.37
29 928.43 1010.1 1 700.74 709.72 670.99 666.33
30 928.55 1011.57 723.74 763.16 685.27 685.45
31 1092.17 1093.31 823.71 807.63 698.43 699.54
32 1115.72 1100.17 827.27 825.78 703.22 722.70
33 1275.82 1288.75 848.28 853.62 723.77 729.37
34 1327.07 1357.09 935.39 940.36 743.06 739.89
35 1356.99 1360.62 1090.21 1119.28 816.97 790.93
36 1431.00 1413.10 1091.39 1120.17 849.17 845.02
37 1456.77 1459.64 1126.27 1128.75 850.85 859.78
38 1489.36 1481.25 1172.49 1168.96 938.38 947.90
39 1517.07 1518.50 1295.84 1297.80 1143.37 1148.43

160

Table A2. (con’t)

40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

1598.01
1756.61
1766.32
1803.43
1811.44
3338.61
3339.21
3421.64
3421.70

1573.98
1778.78
1796.80
1807.69
1828.64
3339.82
3340.36
3425.86
3425.94

1379.84
1411.33
1476.30
1551.07
1575.00
1596.42
1628.47
1745.89
1767.19
1847.72
1847.98
3372.05
3372.33
3465.65
3465.67

1378.47
1412.72
1507.50
1534.84
1562.27
1574.86
1608.87
1745.54
1753.59
1806.03
1820.53
3372.90
3373.16
3469.52
3469.55

1150.34
1151.20
1205.15
1304.90
1351.86
1360.81
1368.94
1458.96
1537.35
1563.02
1607.11
1633.68
1641.06
1737.82
1741.85
1848.96
1865.34
3370.24
3370.61
3468.92
3468.94

1158.28
1158.84
1204.24
1307.13
1358.22
1362.69
1367.59
1459.97
1554.60
1575.24
1583.51
1597.39
1655.93
1723.36
1731.45
1795.17
1818.99
3370.95
3371.32
3472.52
3472.54

 

161

Table A2. (con’t)

 

 

104.1132)2 F41Li2 (OH)4lLi2

TCSCF ROHF TCSCF ROHF TCSCF ROHF
1 1.07* 457* 1654(1) 2.93 426.44 (1) 458.17 (I)
2 057* 027* 031* 267* 425.39 (1) 457.12 (I)
3 010* 003* 021* 166* 18.76 (I) 23.12 (I)
4 005* 002* 026* 024* 228* 385*
5 1.18* 068* 087* 006* 1.97* 3.03*
6 1.76* 1.60* 272* 005* 1.82* 1.69*
7 31.11 32.64 354* 11.17* 017* 018*
8 62.95 62.74 71.28 75.20 012* 007*
9 73.76 73.93 94.30 93.12 200* 008*
10 81.64 87.80 108.91 106.49 70.77 66.95
11 89.78 91.99 136.15 132.41 86.79 85.39
12 144.33 145.12 140.12 142.02 87.78 87.89
13 176.80 180.09 179.98 181.57 118.47 113.08
14 195.35 196.92 205.06 205.49 146.62 145.35
15 197.61 198.19 223.3 225.41 202.99 201.08
16 198.17 200.39 271.53 276.57 203.73 202.04
17 210.67 210.79 284.62 298.09 251.05 241.24
18 221.67 222.37 312.7 314.61 271.84 241.36
19 279.59 280.99 345.54 349.91 296.75 249.45
20 310.20 313.20 384.96 385.12 300.87 277.87
21 315.84 317.12 422.08 421.90 309.44 318.24
22 326.64 320.91 431.83 432.66 325.79 328.75
23 349.34 351.61 451.64 456.91 358.5 344.34
24 350.57 357.35 489.31 494.25 386.77 385.81
25 437.58 440.53 502.88 503.05 418.18 418.17
26 449.35 456.25 505.59 506.39 437.5 440.14
27 452.56 462.43 508.16 526.06 458.01 457.25
28 461.93 483.68 562.95 566.01 476.68 502.44
29 477.77 485.18 603.00 609.65 513.34 514.11
30 510.15 522.15 638.99 640.78 523.32 521.04
31 601.51 605.98 672.73 679.24 525.11 531.13
32 638.70 639.02 677.05 684.14 574.93 567.31
33 664.40 662.37 685.00 689.22 646.91 648.97
34 666.36 667.27 704.51 710.93 680.67 676.34
35 667.24 668.46 717.91 716.36 707.48 694.99
36 699.83 702.56 721.70 722.84 709.93 708.63
37 725.88 727.59 735.61 733.69 728.95 711.18
38 735.73 741.64 743.34 752.95 734.32 731.57
39 810.96 785.63 932.53 934.50 735.73 742.16
40 834.53 838.43 1018.21 1017.98 768.00 748.51
41 846.99 840.32 1110.47 1110.26 797.56 761.46

162

Table A2. (con’t)

42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66

855.50
870.91

948.29

992.70

993.62

1134.14
1166.53
1167.23
1193.54
1302.96
1341.74
1449.70
1531.43
1560.03
1600.11
1628.82
1630.81
1728.07
1740.28
1838.73
1862.87
3374.71
3374.98
3473.93
3473.93

855.61

874.46

957.42

990.26

991.18

1140.90
1166.18
1166.77
1 195.40
1305.49
1356.46
1449.57
1554.92
1565.95
1570.07
1589.29
1654.63
1718.30
1729.86
1789.61
1819.13
3376.09
3376.50
3478.43
3478.48

1164.59
1309.66
1338.43
1466.27
1468.82
1500.24
1529.38
1619.56
1626.69
1730.30
1762.55
1865.74
1907.97

1170.8

1323.72
1327.53
1474.49
1480.92
1515.58
1522.13
1590.14
1643.96
1737.38
1756.83
1832.65
1873.65

803.60

932.01

1005.23
1086.42
1156.81
1246.02
1256.55
1257.33
1280.71
1338.60
1452.01
1472.78
1488.24
1512.27
1528.31
1586.77
1636.86
1683.99
1717.78
1807.74
1851.47
4114.97
4115.02
4132.66
4132.95

762.51

934.28

995.38

1077.35
1163.72
1240.94
1250.40
1255.55
1306.04
1376.41
1454.03
1478.12
1485.29
1524.64
1536.21
1575.44
1649.01
1715.31
1721.63
1816.76
1857.21
4118.13
4118.27
4135.12
4137.31

 

163

Table B1. Crystal data for 2.

APPENDIX B

 

Identification code
Empirical formula
Formula weight
Temperature
Wavelength

Crystal system
Space group

Unit cell dimensions

Volume

Z

Density (calculated)
Absorption coefficient

F (000)

Crystal size

Theta range for data collection
Index ranges

Reflections collected
Independent reflections
Refinement method

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

Final R indices [I>2sigma(1)]
R indices (all data)
Extinction coefficient
Largest diff. peak and hole

2

C32 H20 04

468.48

173(2) K

0.71073 A

Monoclinic

[92,/n (#14)

a =13.8607(14)A

b = 4.6091(5) A

c = 19.121(2) A

alpha = 90 deg.

beta = 109.922(2) deg.
gamma = 90 deg.
1148.4(2) A3

2

1.355 Mg/m3

0.089 mm"

488

0.315 x 0.180 x 0.075 mm
1.59 to 25.00 deg.

-16Sh$l 8, -6Sk$6, 2451516
5359

2014 [R(int) = 0.0493]
Full-matrix least-squares on F2
2014 / 0 / 204

1.046

R1 = 0.0493, sz = 0.0952
R. = 0.0858, sz = 0.1077
0.020(3)

0.184 and -O.146 6A"

 

164

Table B2. Atomic coordinates (x 104) and equivalent isotropic displacement parameters
(A2 x 103) for 2. U(eq) is defined as one third of the trace of the
orthogonalized Uij tensor.

 

 

x y z U(CQ)
0(1) 824(1) 9390(3) 892(1) 36(1)
C(1) 550(2) 7129(5) 543(1) 26(1)
0(2) -538(1) 5945(3) 1213(1) 40(1)
C(2) -380(2) 5559(5) 635(1) 26(1)
C(3) 2041(2) 3955(4) 22(1) 24(1)
C(4) 2090(2) 3670(4) 4770) 24(1)
C(5) 2803(1) 3551(4) -955(1) 24(1)
C(6) -3691(2) 1844(5) -1 149(1) 28(1)
C(7) 4372(2) 1820(5) -1873(1) 37(1)
C(8) -4198(2) 3546(5) 2406(1) 35(1)
C(9) 2334(2) 5268(5) 2220(1) 32(1)
C(10) 2637(2) 5272(5) 4504(1) 27(1)
C(11) 2553(2) 3676(5) 411(1) 28(1)
C(12) 2143(2) 1959(5) 1044(1) 40(1)
C(13) 2638(3) 1861(7) 1569(2) 58(1)
C(14) -3505(2) 3487(8) 1476(2) 60(1)
C(15) -3908(2) 5186(7) 861(2) 51(1)
C(16) 3449(2) 5251(6) 331(1) 37(1)

 

165

Table B3. Bond lengths [A] and angles [deg] for 2.

 

0(1)-C(1)
C(1)-C(3)#l
C(1)-C(2)
0(2)-C(2)
C(2)-C(3)
C(3)-C(4)
C(3)-C(l)#1
C(4)-C(1 1)
C(4)-C(5)
C(5)—C(10)
C(5)-C(6)
C(6)-C(7)
C(6)-H(6)
C(7)-C(8)
C(7)-H(7)
C(8)-C(9)
C(8)-H(8)
C(9)-C(10)
C(9)-H(9)
C(10)-H(10)
C(1 1 )-C(12)
C(1 1)-C(16)
C(12)-C(13)
C(12)-H(12)
C(13)—C(14)
C(13)-H(13)
C(14)-C(15)
C(14)-H(14)
C(15)-C(16)
C(15)-H05)
C(16)-H(16)

O(1)-C(1)-C(3)#1
0(1)-C(1)-C(2)
C(3)#1-C(1)-C(2)
0(2)-C(2)-C(3)
0(2)-C(2)-C(1)
C(3)-C(2)-C(1)
C(4)-C(3)-C(1)#1
C(4)-C(3)-C(2)
C(l)#1-C(3)-C(2)
C(3)-C(4)-C(1 1)
C(3)-C(4)-C(5)

1.226(2)
1.471(3)
1.540(3)
1.213(2)
1.476(3)
1.388(3)
1.471(3)
1.473(3)
1.480(3)
1.396(3)
1.400(3)
1.384(3)
100(2)
1.378(3)
098(2)
1.378(3)
100(2)
1.380(3)
101(2)
100(2)
1.395(3)
1.401(3)
1.397(4)
100(2)
1.375(4)
093(3)
1.365(4)
103(3)
1.368(3)
108(3)
096(2)

121.9(2)
1 16.9(2)
121.0(2)
126.7(2)
1 170(2)
1 16.1(2)
121.0(2)
122.0(2)
1 16.7(2)
122.3(2)
120.7(2)

166

Table B3.(con’t)

C(1 l)-C(4)-C(5)
C(10)-C(5)-C(6)
C(10)-C(5)-C(4)
C(6)-C(5)-C(4)
C(7)-C(6)-C(5)
C(7)-C(6)—H(6)
C(5)-C(6)—H(6)
C(8)—C(7)-C(6)
C(8)-C(7)-H(7)
C(6)-C(7)-H(7)
C(9)-C(8)-C(7)
C(9)-C(8)—H(8)
C(7)-C(8)-H(8)
C(8)-C(9)-C(10)
C(8)-C(9)-H(9)
C(10)-C(9)-H(9)
C(9)-C(10)-C(5)
C(9)-C(10)-H(10)
C(5)-C(lO)-H(10)
C(12)-C(1 1)-C(16)
C(12)-C(1 1)-C(4)
C(16)-C(1 1)-C(4)
C(11)-C(12)-C(13)
C(1 1)-C(12)-H(12)
C(13)-C(12)-H(12)
C(14)-C(13)-C(12)
C(14)-C(13)-H(13)
C(12)-C(l3)-H(13)
C(15)-C(14)—C(13)
C(15)-C(14)-H(14)
C(13)-C(14)-H(14)
C(14)-C(15)-C(16)
C(14)-C(15)-H(15)
C(16)-C(15)-H(15)
C(15)-C(16)-C(11)
C(15)-C(16)-H(16)
C(1 1 )-C(16)-H(16)

116.9(2)
118.4(2)
120.7(2)
120.9(2)
120.5(2)
119.2(13)
120.3(13)
120.3(2)
120.0(14)
119.7(14)
119.8(2)
120.8(12)
119.5(12)
120.7(2)
121.4(13)
117.9(13)
120.4(2)
119.7(12)
119.9(12)
118.2(2)
120.4(2)
121.2(2)
119.2(3)
117.3(13)
123.5(13)
120.7(3)
128(2)
112(2)
120.5(3)
118(2)
122(2)
119.6(3)

1 17(2)
123(2)
121.8(3)
123.5(13)
114.7(13)

 

Symmetry transformations used to generate equivalent atoms: #1 —x,-y+1,-z.

167

Table B4. Anisotropic displacement parameters (A2 x 103) for 2. The anisotropic
displacement factor exponent takes the form:

—2n2[h2a*2Ul.+...+2hka*b*U12]

 

 

 

 

 

U11 U22 U33 U23 U13 U12
0(1) 33(1) 31(1) 38(1) -13(1) 7(1) -l(l)
C(1) 22(1) 27(1) 24(1) -2(1) -1(1) 4(1)
0(2) 36(1) 52(1) 33(1) -15(1) 12(1) -7(1)
C(2) 23(1) 26(1) 27(1) -4(1) 4(1) 7(1)
C(3) 23(1) 22(1) 24(1) 0(1) 5(1) 3(1)
C(4) 27(1) 17(1) 27(1) 0(1) 7(1) 1(1)
C(S) 21(1) 22(1) 26(1) -1(1) 5(1) 1(1)
C(6) 26(1) 27(1) 33(1) 1(1) 10(1) -4(1)
C(7) 27(1) 39(2) 39(2) -8(1) 4(1) -9(1)
C(8) 31(1) 37(1) 29(1) -3(1) 2(1) 2(1)
C(9) 33(1) 36(1) 26(1) 3(1) 9(1) 1(1)
C(10) 24(1) 28(1) 29(1) 0(1) 8(1) -2(1)
C(11) 29(1) 28(1) 25(1) -2(1) 6(1) -7(1)
C(12) 45(2) 38(2) 30(1) 1(1) 4(1) -11(1)
C(13) 80(2) 61(2) 26(2) 6(1) 9(2) -29(2)
C(14) 56(2) 86(2) 45(2) 24(2) 28(2) -37(2)
C(15) 46(2) 68(2) 44(2) 20(2) 22(2) -1 7(2)
C(16) 32(1) 43(2) 38(2) -5(1) 14(1) -4(1)
Table B5. Hydrogen coordinates ( x 104) and isotropic displacement parameters (A2 x

103) for 2.

X y z U(CQ)

H(6) -3836(16) 598(49) -766(12) 40(6)
H(7) -4980(l 8) 580(51) -2004(13) 48(7)
H(8) -4701(15) 3539(45) -2923(l2) 35(6)
H(9) -3 178(17) 6506(52) -2602(13) 52(7)
H(IO) -2016(15) 6553(45) -1376(11) 32(6)
H(12) -1520(16) 798(50) 1084(12) 41(7)
H(13) -2326(20) 558(60) 1948(15) 60(8)
H(14) -3883(20) 3428(61) 1856(15) 73(9)
H(15) -4589(22) 6417(64) 817(15) 88(10)
H(16) -3715(16) 6330(49) -125(13) 44(7)

 

168

Table B6. Crystal data for (NMe2)42-((CH3)ZSO)3.

 

Identification code
Empirical formula
Formula weight
Temperature
Wavelength

Crystal system
Space group

Unit cell dimensions

Volume

Z

Density (calculated)
Absorption coefficient

F (000)

Crystal size

Theta range for data collection
Limiting indices

Reflections collected
Independent reflections
Refinement method

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

Final R indices [I>25igma(l)]
R indices (all data)
Extinction coefficient
Largest diff. Peak and hole

(NM62)42
C46H58N407S3
875.16

293(2) K

0.71073 A
Monoclinic

PZI/n (#14)

a = 21.092(4) A
b = 8.0588(16) A
c = 26.781(5) A
alpha = 90 deg.
Beta = 101.31(3) deg.
Gamma = 90 deg.
4463.7(15) A3

4

1.302 Mg/m3
0.176 mm"

1840

'1 .37 to 28.30 deg.

-26Shs27, -9_<.k_<_10, 6431335
37199

10489 [R(int) = 0.1328]
Full-matrix least-squares on F2
10489 / 0 / 541

0.856

R. = 0.0690, sz = 0.1743
R) = 0.1556, sz = 0.1968
0.020(3)

1.468 and —0627 e.A'3

 

169

Table B7. Atomic coordinates ( x 104) and equivalent isotropic displacement parameter
(A2 x 103) for (NMe2)42-((CH3)ZSO)3. U(eq) is defined as one third of the
trace of the orthogonalized Uij tensor.

 

 

X y 2 [103(1)
C(1) 5350(1) 3157(4) 2109(1) 21(1)
C(2) 4824(1) 1962(4) 2056(1) 20(1)
C(3) 4304(2) 2133(4) 2394(1) 22(1)
C(4) 4484(1) 2954(4) 2890(1) 20(1)
C(5) 5085(2) 3944(4) 2985(1) 24(1)
C(6) 5424(2) 4313(4) 2536(1) 22(1)
0(7) 4753(1) 772(3) 1757(1) 30(1)
0(8) 3787(1) 1406(3) 2236(1) 35(1)
0(9) 5315(1) 4632(4) 3390(1) 44(1)
0(10) 5753(1) 5590(3) 2571(1) 37(1)
C(11) 5781(1) 3216(4) 1754(1) 21(1)
C(12) 5541(1) 2915(4) 1210(1) 22(1)
C(13) 4912(1) 3410(4) 967(1) 24(1)
C(14) 4691(2) 3236(4) 456(1) 27(1)
C(15) 5083(2) 2486(4) 136(1) 24(1)
C(16) 5705(2) 1905(4) 380(1) 24(1)
C(17) 5921(2) 21 19(4) 897(1) 24(1)
N(18) 4869(1) 2305(4) 2730) 31(1)
C(19) 4218(2) 2851(5) -6l7(1) 39(1)
C(20) 5285(2) 1619(6) -699(1) 44(1)
C(21) 6465(1) 3566(4) 1924(1) 22(1)
C(22) 6798(2) 3140(4) 2422(1) 26(1)
C(23) 7454(2) 3330(4) 2574(1) 26(1)
C(24) 7838(2) 3992(4) 2243(1) 26(1)
C(25) 7507(2) 4514(4) 1748(1) 25(1)
C(26) 6854(2) 4304(4) 1606(1) 24( 1)
N(27) 8493(1) 41 18(4) 2390(1) 29(1)
C(28) 8809(2) 3777(5) 2922(1) 35(1)
C(29) 8930(2) 4434(5) 2028(1) 38(1)
C(30) 4098(2) 2796(4) 3270(1) 22(1)
C(31) 3395(2) 2708(4) 3145(1) 22(1)
C(32) 3031(2) 1809(4) 3448(1) 23(1)
C(33) 2364(2) 1701(4) 3334(1) 24(1)
C(34) 2007(2) 2542(4) 2899(1) 26(1)
C(35) 2363(2) 3483(4) 2602(1) 28(1)
C(36) 3029(2) 3543(4) 2721(1) 24(1)
N(37) 1348(1) 2383(4) 2772(1) 34(1)
C(38) 975(2) 1690(5) 3121(2) 45(1)
C(39) 993(2) 2874(5) 2261(2) 51(1)

170

Table B7. (con’t)

C(40)
C(41)
C(42)
C(43)
C(44)
C(45)
N(46)
C(47)
C(48)
S(49)
0(50)
C(51)
C(52)
S(53)
0(54)
C(55)
C(56)
S(57)
0(58)
C(59)
C(60)

4406(1)
5002(2)
5279(2)
4988(2)
4397(2)
4120(2)
5251(1)
5892(2)
491 1(2)
2730(1)
3080(2)
2462(2)
3348(2)
2099(1)
1654(1)
2379(2)
2853(2)
2273(1)
2863(1)
2006(2)
1620(2)

2694(4)
1840(4)
1686(4)
2408(4)
3295(4)
3399(4)
2259(4)
1509(5)
2850(5)
2157(2)
3827(4)
1793(6)
628(6)

5416(1)
6036(4)
7159(6)
4991(6)
871(2)

246(5)

-773(5)
770(6)

3814(1)
3981(1)
4485(1)
4875(1)
4707(1)
4196(1)
5376(1)
5541(1)
5770(1)
811(1)

865(1)

1387(2)
892(2)

-153(1)
-625(1)
234(2)

-328(2)
-965(1)
390(1)

3401(2)

-639(2)

22(1)
24(1)
25(1)
25(1)
28(1)
25(1)
29(1)
38(1)
37(1)
54(1)
73(1)
57(1)
60(1)
47(1)
66(1)
59(1)
47(1)
55(1)
78(1)
50(1)
56(1)

 

171

Table B8. Bond lengths [A] and angles [deg] for (NMe2)42*((CH3)2SO)3.

 

C(1)-C(1 1)
C(1)-C(2)
C(1)-C(6)
C(2)-0(7)
C(2)-C(3)
C(3)-0(8)
C(3)-C(4)
C(4)-C(30)
C(4)-C(5)
C(5)-0(9)
C(5)-C(6)
C(6)-0(10)
C(1 1)-C(21)
C(1 1)-C(12)
C(12)-C(13)
C(12)-C(17)
C(13)-C(14)
C(14)-C(15)
C(15)-N(18)
C(15)-C(16)
C(16)-C(17)
N(18)-C(20)
N(18)-C(19)
C(21)-C(22)
C(21)-C(26)
C(22)-C(23)
C(23)-C(24)
C(24)-N(27)
C(24)-C(25)
C(25)-C(26)
N(27)-C(28)
N(27)-C(29)
C(30)-C(31)
C(30)-C(40)
C(31)-C(36)
C(31)-C(32)
C(32)-C(33)
C(33)-C(34)
C(34)-N(37)
C(34)-C(35)
C(35)-C(36)
N(37)-C(38)
N(37)-C(39)

1.440(4)
1.455(4)
1.459(4)
1.240(4)
1.559(4)
1.237(4)
1.464(4)
1.429(4)
1.477(4)
1.231(4)
1 545(5)
1.234(4)
1.453(4)
1.464(4)
1 415(4)
1.422(4)
1.363(4)
1.437(5)
1.357(4)
1.426(4)
1.383(4)
1.462(4)
1 .466(4)
1.421(4)
1.422(4)
1.373(4)
1.416(5)
1.364(4)
1.436(4)
1.364(4)
1.476(4)
1.484(4)
1.458(4)
1 475(4)
1.412(4)
1.419(4)
1 382(4)
1.428(4)
1.371(4)
1.416(5)
1.378(4)
1.447(5)
1 479(5)

172

Table B8. (con’t)

C(40)-C(45)
C(40)-C(41)
C(41)-C(42)
C(42)-C(43)
C(43)-N(46)
C(43)-C(44)
C(44)-C(45)
N(46)-C(48)
N(46)-C(47)
S(49)-0(50)
S(49)-C(51)
S(49)-C(52)
S(53)-O(54)
S(53)-C(56)
S(53)-C(55)
S(57)-O(58)
S(57)—C(60)
S(57)-C(59)

C(1 1)-C(1)-C(2)
C(1 l)-C(l)-C(6)
C(2)-C(1)-C(6)
O(7)-C(2)-C(l)
O(7)-C(2)-C(3)
C(1)-C(2)-C(3)
O(8)-C(3)-C(4)
O(8)-C(3)-C(2)
C(4)-C(3)-C(2)
C(30)-C(4)-C(3)
C(30)-C(4)—C(5)
C(3)-C(4)-C(5)
O(9)-C(5)-C(4)
O(9)-C(5)-C(6)
C(4)-C(5)-C(6)
O(10)—C(6)-C(1)
O(10)-C(6)-C(5)
C(1)-C(6)-C(5)
C(1)-C(l 1)-C(21)
C(1)-C(l 1)-C(12)
C(21)-C(1 1)-C(12)
C(13)-C(12)-C(17)
C(13)-C(12)-C(11)
C(17)-C(12)-C(1 1)
C(14)-C(13)-C(12)

1.404(4)
1.427(4)
1.367(4)
1.434(5)
1.353(4)
1.431(4)
1.383(4)
1.466(4)
1.468(4)
1.527(3)
1.769(4)
1.774(4)
1.505(3)
1.776(4)
1.776(5)
1.524(3)
1.773(4)
1.781(4)

121.7(3)
120.8(3)
117.4(3)
124.6(3)
115.8(3)
119.7(3)
125.4(3)
115.6(3)
118.6(3)

121.6(3) .

120.9(3)
117.5(3)
125.0(3)
115.9(3)
118.9(3)
124.4(3)
116.1(3)
119.5(3)
121.1(3)
120.7(3)
118.2(3)
116.3(3)
121.4(3)
122.3(3)
122.5(3)

173

Table BS. (con’t)

C(13)-C(14)-C(15)
N(18)-C(15)-C(16)
N(l8)-C(15)-C(14)
C(16)-C(15)-C(14)
C(17)-C(16)-C(15)
C(16)-C(17)—C(12)
C(15)-N(18)-C(20)
C(15)-N(18)-C(19)
C(20)-N(18)-C(19)
C(22)-C(21)-C(26)
C(22)-C(21)—C(1 1)
C(26)-C(21)-C(1 1)
C(23)-C(22)-C(21)
C(22)-C(23)-C(24)
N(27)—C(24)—C(23)
N(27)-C(24)-C(25)
C(23)-C(24)—C(25)
C(26)-C(25)-C(24)
C(25)-C(26)-C(21)
C(24)-N(27)-C(28)
C(24)-N(27)-C(29)
C(28)-N(27)-C(29)
C(4)-C(30)-C(31)

C(4)-C(30)-C(40)

C(31)-C(30)-C(40)
C(36)-C(31)-C(32)
C(36)-C(31)-C(30)
C(32)-C(31)-C(30)
C(33)-C(32)-C(31)
C(32)-C(33)-C(34)
N(37)-C(34)-C(35)
N(37)-C(34)-C(33)
C(35)-C(34)-C(33)
C(36)-C(35)-C(34)
C(35)-C(36)-C(31)
C(34)-N(37)-C(38)
C(34)-N(37)-C(39)
C(38)-N(37)-C(39)
C(45)-C(40)-C(41)
C(45)-C(40)-C(30)
C(41)-C(40)-C(30)
C(42)-C(41)-C(40)
C(41)-C(42)-C(43)
N(46)-C(43)-C(44)

121.3(3)
121.3(3)
121.9(3)
116.8(3)
120.8(3)
122.3(3)
121.3(3)
120.8(3)
117.8(3)
115.2(3)
121.6(3)
123.2(3)
122.3(3)
121.5(3)
121.3(3)
121.7(3)
117.1(3)
120.1(3)
123.7(3)
120.3(3)
123.3(3)
116.1(3)
122.4(3)
120.5(3)
117.0(3)
115.5(3)
122.2(3)
122.3(3)
123.1(3)
120.1(3)
122.3(3)
120.3(3)
117.3(3)
121.1(3)
122.7(3)
121.9(3)
120.6(3)
117.5(3)
116.4(3)
121.9(3)
121.7(3)
122.1(3)
121.5(3)
121.2(3)

174

Table B8. (con’t)

N(46)-C(43)-C(42) 122.4(3)
C(44)-C(43)-C(42) 1 16.4(3)
C(45)-C(44)-C(43) 120.9(3)
C(44)-C(45)-C(40) 122.6(3)
C(43)-N(46)—C(48) 121.4(3)
C(43)-N(46)-C(47) 120.6(3)
C(48)-N(46)-C(47) 1 180(3)
O(50)-S(49)-C(51) 107.1(2)
0(50)-S(49)-C(52) 105.8(2)
C(51)-S(49)—C(52) 97.5(2)
O(54)-S(53)-C(56) 106.66(19)
O(54)—S(53)-C(55) 108.0(2)
C(56)-S(53)-C(55) 95.4(2)
O(58)-S(57)-C(60) 105.8(2)
0(58)-S(57)-C(59) 107.4(2)
C(60)-S(57)—C(59) 96.4(2)

 

Symmetry transformations used to generate equivalent atoms: #1 -x,-y+1,-z.

175

Table B9. Anisotropic displacement parameters (A2 x 103) for (N Me2)42*((CH3)ZSO)3.
The anisotropic displacement factor exponent takes the form:

—27t2[h2a*2Ul1+...+2hka*b*U12]

 

 

U11 U22 U33 U23 U13 U12
C(1) 19(2) 17(2) 26(2) 2(1) 3(1) 1(1)
C(2) 16(2) 19(2) 23(2) 2(1) 3(1) 3(1)
C(3) 21(2) 15(2) 30(2) 4(1) 4(1) 3(1)
C(4) 18(2) 19(2) 24(2) 1(1) 6(1) 1(1)
C(5) 23(2) 23(2) 27(2) 2(2) 6(1) 0(2)
C(6) 21(2) 18(2) 26(2) 3(1) 2(1) 1(2)
0(7) 25(1) 27(2) 38(1) -12(1) 7(1) -4(1)
0(8) 24(1) 42(2) 40(1) -12(1) 8(1) -10(1)
0(9) 45(2) 56(2) 33(2) -19(1) 14(1) -27(1)
0(10) 50(2) 24(2) 42(2) -10(1) 18(1) -19(1)
C(11) 26(2) 12(2) 26(2) 2(1) 5(1) 1(1)
C(12) 24(2) 18(2) 24(2) 2(1) 7(1) 2(1)
C(13) 23(2) 25(2) 25(2) 1(2) 7(1) 5(2)
C(14) 24(2) 29(2) 30(2) 3(2) 5(1) 7(2)
C(15) 27(2) 18(2) 28(2) -1(2) 6(1) -3(2)
C(16) 27(2) 20(2) 29(2) -4(2) 12(1) 1(2)
C(17) 21(2) 23(2) 27(2) 2(2) 5(1) 2(1)
N(18) 34(2) 32(2) 27(2) -5(1) 5(1) 2(1)
C(19) 40(2) 45(3) 30(2) -3(2) 1(2) 6(2)
C(20) 49(2) 55(3) 29(2) -9(2) 10(2) 2(2)
C(21) 23(2) 17(2) 26(2) 2(1) 7(1) -2(1)
C(22) 31(2) 19(2) 28(2) 3(2) 8(1) 3(2)
C(23) 30(2) 21(2) 25(2) 2(2) 2(1) 1(2)
C(24) 28(2) 12(2) 37(2) -6(2) 7(2) 3(1)
C(25) 28(2) 21(2) 29(2) -1(2) 1 1(1) -4(2)
C(26) 29(2) 19(2) 24(2) -1(2) 6(1) 1(2)
N(27) 21(2) 23(2) 42(2) 0(1) 4(1) -4(1)
C(28) 28(2) 33(2) 41(2) -6(2) 2(2) -4(2)
C(29) 25(2) 32(2) 57(2) 4(2) 8(2) -4(2)
C(30) 27(2) 10(2) 29(2) 2(1) 6(1) 1(1)
C(31) 26(2) 14(2) 25(2) 2(1) 6(1) -1(1)
C(32) 26(2) 20(2) 23(2) 2(2) 4(1) -1(2)
C(33) 28(2) 19(2) 27(2) 2(2) 9(1) —4(2)
C(34) 22(2) 14(2) 43(2) -6(2) 10(2) 0(1)
C(35) 26(2) 23(2) 33(2) 5(2) 3(1) 3(2)
C(36) 28(2) 15(2) 32(2) 7(2) 10(1) 3(2)
N(37) 24(2) 32(2) 45(2) 1(2) 5(1) -1(1)
C(38) 26(2) 51(3) 60(3) -16(2) 17(2) -12(2)
C(39) 31(2) 43(3) 70(3) 1(2) 30(2) 2(2)

176

Table B9. (con’t)

C(40)
C(41)
C(42)
C(43)
C(44)
C(45)
N(46)
C(47)
C(48)
S(49)
0(50)
C(51)
C(52)
S(53)
0(54)
C(55)
C(56)
S(57)
0(58)
C(59)
C(60)

23(2)
24(2)
19(2)
27(2)
30(2)
22(2)
30(2)
32(2)
40(2)
60(1)

113(3)

45(3)
40(2)
35(1)
42(2)
70(3)
45(2)
58(1)
29(2)
67(3)
37(2)

18(2)
23(2)
20(2)
19(2)
28(2)
19(2)
30(2)
46(3)
39(3)
52(1)
29(2)
74(4)
44(3)
45(1)
68(2)
71(4)
50(3)
33(1)
92(3)
34(3)
62(3)

26(2)
26(2)
37(2)
28(2)
28(2)
34(2)
26(2)
33(2)
31(2)
49(1)
82(2)
48(3)
95(4)
64(1)
82(2)
37(2)
49(3)
80(1)

108(3)

47(2)
69(3)

-2(1)
-2(2)
0(2)

-1(2)
-5(2)
-1(2)
-2(1)
-1(2)
-6(2)
2(1)

4(2)

9(2)

-9(3)
16(1)
14(2)
-3(2)
-4(2)

-11(1)
27(2)

0(2)
3(3)

7(1)
7(1)
6(1)
5(1)
9(1)
8(1)
2(1)
0(2)
5(2)
6(1)
34(2)
1(2)
8(2)
18(1)
-4(2)
12(2)
14(2)
26( 1)
1(2)
8(2)
10(2)

-6(1)
-1(2)
1(1)
-7(2)
-2(2)
0(1)
1(1)
-3(2)
-1(2)
17(1)
1(2)
3(2)
10(2)
1(1)
-2(2)
27(3)
1(2)
-3(1)
-1(2)
8(2)
13(2)

 

177

Table BIO. Hydrogen coordinates (x 104) and isotropic displacement parameters (A2 x
103) for (NMe2)42-((CH3)2SO)3.

 

 

x y 1 U(BQ)
H(13) 4639 3871 1164 29
H(14) 4278 3610 313 33
H(16) 5969 1375 189 29
H(17) 6330 1729 1046 29
H(19A) 4150 2629 -976 59
H(19B) 4176 4020 -563 59
H(19C) 3903 2261 -472 59
H(20A) 5053 15 88 -1045 66
H(20B) 5414 515 -5 89 66
H(20C) 5662 2304 -677 66
H(22) 6562 2718 2652 31
H(23) 7651 3017 2902 31
H(25) 7737 4996 1523 30
H(26) 6652 4663 1284 28
H(28A) 9267 3941 2960 53
H(28B) 8725 2650 3006 53
H(28C) 8641 4516 3145 53
H(29A) 9368 4496 2213 57
H(29B) 8813 5464 1854 57
H(29C) 8891 3547 1785 57
H(32) 3252 1268 3737 28
H(33) 2148 1080 3540 29
H(35) 2144 4070 2322 33
H(36) 3246 4160 2514 29
H(38A) 525 1682 2964 67
H(38B) 1037 2352 3425 67
H(38C) 1117 575 3207 67
H(39A) 539 2682 2238 76
H(39B) 1 144 2227 2007 76
H(39C) 1065 4030 2206 76
H(41) 5210 1374 3739 29
H(42) 5664 1 100 4577 30
H(44) 41 96 3 807 4946 33
H(45) 3730 3960 4100 30
H(47A) 6001 1500 5906 57
H(47B) 6207 2144 5409 57
H(47C) 5886 392 5416 57
H(48A) 5170 2634 6100 55
H(48B) 4505 2282 5737 55
H(48C) 4835 4022 5730 55

178

Table B10. (con’t)

H(SIA)
H(SlB)
H(SlC)
H(52A)
H(52B)
H(52C)
H(SSA)
H(5513)
H(55C)
H(56A)
H(56B)
H(56C)
H(59A)
H(59B)
H(59C)
H(60A)
H(60B)
H(6OC)

2114
2813
2314
3559
3160
3658
2030
2534
2723
2809
3172
2984
2308
1588
1979
1675
1609
1221

2538
1976
669
650
-447
855
7606
7990
6822
4043
4766
5934
-918
-506
-1780
1600
-308
962

1411
1670
1394
605
917
1198
375
31
505
-549
-27
-501
-1623
-1599
-1215
-377
-489
-875

85
85
85
90
90
90
88
88
88
71
71
71
75
75
75
84
84
84

 

179

Table B11. Crystal data for (t-Bu)3(MeO)42K-(C4HgO)4.

 

Identification code
Empirical formula
Formula weight
Temperature
Wavelength

Crystal system
Space group

Unit cell dimensions

Volume

Z

Density (calculated)
Absorption coefficient

F (000)

Crystal size

Theta range for data collection
Index ranges

Reflections collected
Independent reflections
Refinement method

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

Final R indices [I>25igma(l)]
R indices (all data)
Extinction coefficient
Largest diff. peak and hole

(t-Bu)3(MeO)42K
C84H124012K
1364.99

293(2) K

0.71073 A

triclinic

P]

a = 9.5055(2) A

b = 13.0026(2) A

c = 17.1668(2) A
alpha = 77.2809(11)
beta = 85.5001(12)
gamma = 77.64l7(10)
2020.64(7) A3

1

1.384 Mg/m3

0.182 mm"

890

1.64 to 28.33 deg.

-12_<.h312, -17£k$17, 2231322
22327

17415 [R(int) = 0.0182]
Full-matrix least-squares on F 2
17415 / 3 / 874

0.91

R1 = 0.0844, wR2 = 0.2430
R1= 0.1437, wR2 = 0.3117

0.619 and -O.458 eA'3

 

180

Table B12. Atomic coordinates ( x 104) and equivalent isotropic displacement parameters
(A2 x 103). U(eq) is defined as one third of the trace of the orthogonalized Uij

 

 

tensor.

x y I U(eq)
K 5254(3) 6177(2) 1615(2) 90(1)
C( 1) 10216(6) 6702(5) 754(3) 35(1)
C(2) 1 1564(6) 6144(5) 1 102(4) 36(1)
C(3) 1 1601(6) 5616(5) 2003(4) 34(1)
C(4) 10281(6) 5637(4) 2486(3) 30(1)
C(5) 8924(7) 6213(5) 21 11(3) 38(2)
C(6) 8876(6) 6729(4) 1244(3) 30( 1)
C(7) 10347(6) 5098(5) 3287(3) 28(1)
C(8) 9434(6) 5561(4) 3913(3) 30(1)
C(9) 8922(7) 4898(5) 4593(3) 35(1)
C(10) 8153(7) 5320(5) 5202(4) 40(2)
C(1 1) 7954(7) 6432(6) 5162(4) 42(2)
C(12) 8256(7) 7131(5) 4442(4) 47(2)
C(13) 9067(7) 6655(5) 3850(4) 43(2)
C(14) 7503(10) 4557(7) 5918(4) 63(2)
C(15) 8125(16) 4616(10) 6699(5) 127(5)
C(16) 7576(14) 3474(7) 5695(6) 1 13(5)
C(17) 6103(15) 4926(8) 6190(9) 167(8)
O(18) 7305(5) 6861(4) 5799(3) 54(1)
C(19) 8277(13) 6963(12) 6330(7) 123(5)
C(20) 7810(10) 8365(6) 4296(6) 78(3)
C(21) 8090(12) 8914(6) 3481(7) 96(3)
C(22) 6336(14) 8709(8) 4587(9) 152(7)
C(23) 8830(30) 8797(9) 4833(11) 253(13)
0(24) 7753(4) 6251(5) 2474(3) 71(2)
O(25) 12752(5) 6133(4) 751(3) 62(2)
0(26) 7682(5) 7124(4) 958(3) 56( 1)
0(27) 12816(4) 5298(4) 2269(2) 48(1)
C(28) 1 1262(6) 4030(4) 3531(3) 30(1)
C(29) 1 1494(6) 3306(4) 3012(3) 32( 1)
C(30) 12181(7) 2218(5) 3245(4) 40(2)
C(31) 12421(10) 1502(6) 2614(4) 60(2)
C(32) 13919(1 1) 780(8) 2671(5) 100(4)
C(33) 12353(12) 2196(7) 1752(4) 92(3)
C(34) 1 1242(15) 831(9) 2700(6) 1 13(4)
C(35) 12630(7) 1890(5) 4025(4) 43(2)
0(36) 13178(8) 780(4) 4303(3) 81(2)
C(37) 1 1814(19) 233(8) 4658(7) 163(7)
C(3 8) 125 50(7) 2573(4) 4560(4) 37( 1)

181

Table BIZ. (con’t)

C(39)
C(40)
C(41)
C(42)
C(43)
C(44)
C(45)
C(46)
C(47)
C(48)
C(49)
C(50)
C(51)
C(52)
C(53)
C(54)
0(55)
C(56)
C(57)
C(58)
C(59)
C(60)
C(61)
C(62)
C(63)
C(64)
C(65)
C(66)
C(67)
C(68)
C(69)
C(70)
0(71)
C(72)
C(73)
C(74)
C(75)
C(76)

13198(9)
12193(19)
1330503)
1485805)
1 1857(6)
10181(6)
9270(6)
8675(7)
7985(7)
7960(8)
8357(7)
9043(6)
8096(8)
8218(17)
6642(1 1)
9308(12)
7421(9)
8304(15)
7275(9)
7031(15)
5956(16)
8098(16)
1 1091(6)

1 1604(6)
12349(6)
12589(6)
12197(7)
1 1457(6)
12837(8)
13012(1 1)
14588(15)
1 1657(14)
13179(5)
12198(12)
12693(1 1)
12840(20)
14291(14)
1 1650(20)

0(101)
C(102)
C(103)
C(104)
C(105)
0(201)

6346(7)

5680(13)
6535(14)
7692(14)
7570(20)
4078(1 1)

2246(5)
1863(17)
3217(7)
1566(12)
3657(5)
7270(4)
8360(4)
8724(5)
9778(5)
10510(4)
10127(5)
9063(5)
10844(5)
10181(8)
1 1635(7)

11524(10)

1 1564(4)
12198(9)
10127(6)
9186(7)

10852(13)

10865(7)
6820(5)
7464(5)
7052(5)
5949(5)
5238(5)
5697(5)
7866(7)
8858(7)
7336(12)
8301(12)
5532(4)
5410(10)
4017(6)
3576(8)
3660(1 1)
3491(9)
4093(6)
3330(9)
2380(9)
2666(14)
3769(13)
8310(9)

5372(4)
6004(5)
5717(5)
5323(8)
4286(3)
68(3)
299(3)
3050(3)
3312(4)
312(4)
3(3)
207(3)
624(4)
1444(5)
537(6)
527(8)
3090(3)
3367(6)
2158(4)
2416(6)
2153(7)
2741(5)
684(3)
3352(3)
3997(4)
3924(4)
-1236(4)
632(3)
2700(4)
2471(6)

2810(11)

331 1(6)
2589(3)
3137(6)
3095(6)
386(7)

3356(8)

3300(19)

2396(7)
2507(1 1)
3084(8)
3347(12)
2848(18)
907(8)

182

56(2)
198(9)
91(3)
157(6)
31(1)
34(1)
29(1)
36(1)
40(2)
42(2)
36(1)
33(1)
50(2)
108(4)
84(3)

1 19(5)
85(2)
109(4)
54(2)
120(5)
213(1 1)
120(4)
36(1)
38(2)
36(1)
37(2)
42(2)
38(2)
57(2)
89(3)
170(8)
172(8)
48(1)
99(4)
73(3)
247(13)
146(6)
321(19)
146(4)
136(6)
127(4)
183(8)
350(20)
166(5)

Table B12. (con’t)

C(202)
C(203)
C(204)

C(205) '

0(301)
C(302)
C(303)
C(304)
C(305)
0(401)
C(402)
C(403)
C(404)
C(405)

5010(1 1)
4330(20)
3014(18)
2756(9)

4168(9)

2846(12)
2692(14)
3510(40)
4831(12)
6165(1 1)
7779(19)
7720(19)
6460(20)
6020(60)

9025(10)
9490(30)
9432(13)
8765(1 1)
6945(10)
7451(12)
7380(20)
6390(30)
6179(13)
5379(9)

4930(20)
5557(15)
6136(16)
6120(20)

514(14)
-98(15)
375(7)
651(12)
3073(7)
3195(10)
4110(12)
4210(20)
3746(8)
65(7)
-300(16)
3145(9)
3232(12)
-519(15)

181(9)
295(18)
137(5)
174(8)
151(4)
144(6)
223(12)
380(20)
131(5)
173(5)
292(19)
164(6)
202(8)
540(40)

 

183

Table B13. Bond lengths [A] and angles [deg] for (t—Bu)g(MeO)42K-(C4H30)4.

 

K-O(101)
K-O(201)
K-O(27)#1
K-O(26)
K-O(301)
K-O(25)#l
K-O(24)
K-O(401)
(3(1)-C(2)
C(1)-C(44)
C(1)-C(6)
C(2)-C(25)
C(2)-C(3)
C(3)-0(27)
C(3)-C(4)
C(4)-C(7)
C(4)-C(5)
C(5)-C(24)
C(5)-C(6)
C(6)-0(26)
C(7)-C(28)
C(7)-C(8)
C(8)-C(13)
C(3)-C(9)
C(9)-C(10)
C(10)-C(1 1)
C(10)-C(14)
C(1 1)-O(18)
C(1 1)-C(12)
C(12)-C(13)
C(12)-C(20)
C(14)-C(17)
C(14)-C(1 5)
C(14)-C(16)
O(18)-C(19)
C(20)-C(22)
C(20)-C(21)
C(20)-C(23)
O(25)-K#2
O(27)-K#2
C(28)-C(43)
C(28)-C(29)
C(29)-C(30)

2.763(8)
2.796(10)
2.845(5)
2.884(5)
2.929(10)
2.917(6)
2.922(6)
3.061(12)
1.434(8)
1.440(8)
1.470(7)
1.236(7)
1.547(8)
1.228(7)
1.449(7)
1.398(8)
1.464(9)
1.229(7)
1.492(8)
1.233(7)
1.465(8)
1.477(7)
1.372(8)
1.41 1(8)
1.374(8)
1.405(9)
1.575(9)
1.380(7)
1.412(9)
1.401(9)
1.538(10)
1.396(13)
1.530(13)
1.525(1 1)
1.393(12)
1.461(13)
1.455(13)
1.645(18)
2.917(6)
2.845(5)
1.399(7)
1.407(8)
1.409(8)

184

Table 813. (con’t)

C(30)-C(35)
C(30)-C(31)
C(31)-C(34)
C(31)-C(33)
C(31)-C(32)
C(35)-C(38)
C(35)-0(36)
0(36)-C(37)
C(38)-C(43)
C(38)-C(39)
C(39)-C(40)
C(39)-C(41)
C(39)-C(42)
C(44)-C(61)
C(44)-C(45)
C(45)-C(50)
C(45)-C(46)
C(46)-C(47)
C(47)-C(48)
C(47)-C(57)
C(48)-0(55)
C(48)-C(49)
C(49)-C(50)
C(49)-C(5 1)
C(51)-C(53)
C(51)-C(52)
C(51)-C(54)
0(55)-C(56)
C(57)-C(59)
C(57)-C(58)
C(57)—C(60)
C(61)-C(62)
C(61)-C(66)
C(62)-C(63)
C(63)-C(64)
C(63)-C(67)
C(64)-0(71)
C(64)-C(65)
C(65)-C(66)
C(65)-C(73)
C(67)-C(68)
C(67)-C(70)
C(67)-C(69)
0(71)-C(72)

1.384(9)
1.552(8)
1.541(13)
1.553(11)
1.526(11)
1.400(8)
1.412(8)
1.624(15)
1.414(8)
1.505(9)
1.463(14)
1.532(10)
1.641(14)
1.457(8)
1.482(7)
1.369(8)
1.391(8)
1.379(8)
1.410(9)
1.583(9)
1.348(7)
1.431(8)
1.378(8)
1.538(9)
1.532(12)
1.479(12)
1.574(12)
1.292(13)
1.399(12)
1.457(11)
1.520(13)
1.385(8)
1.410(8)
1.410(8)
1.382(8)
1.534(9)
1.400(7)
1.409(9)
1.380(8)
1.526(10)
1.472(12)
1.526(13)
1.673(15)
1.430(10)

185

Table 813. (con’t)

C(73)-C(76) 1.426(18)
C(73)-C(75) 1.551(14)
C(73)-C(74) 1.546(16)
0(101)-C(102) 1.261(13)
0(101)-C(105) 1.383(14)
C(102)-C(103) 1.525(18)
C(103)-C(104) 1.370(17)
C(104)-C(105) 1.485(19)
0(201)-C(205) 1.333(14)
0(201)-C(202) 1.444(16)
C(202)-C(203) 124(3)
C(203)-C(204) 129(2)
C(204)-C(205) 1.523(19)
0(301)-C(302) 1.312(13)
0(301)-C(305) 1.436(15)
C(302)-C(303) 155(2)
C(302)-C(304) 202(4)
C(303)-C(304) 133(4)
C(304)-C(305) 144(3)
0(401)-C(405) 122(3)
0(401)-C(402) 165(2)
C(402)-C(403) 150(2)
C(403)-C(404) 127(2)
C(404)-C(405) 126(3)
O(101)-K-O(201) 176.7(5)
O(101)-K-O(27)#1 74.31(19)
O(201)-K-O(27)#1 104.1(3)
O(101)-K-O(26) 10701(19)
O(201)—K-O(26) 74.6(2)
O(27)#l-K-O(26) 178.4(2)
0(101)-100001) 950(4)
O(201)-K-O(30l) 81.8(4)
O(27)#1-K-O(301) 69.7l(l8)
O(26)-K-O(301) 109.2(2)
O(101)-K-O(25)#1 107.6(2)
O(201)-K-O(25)#1 73.1(3)
O(27)#l-K-O(25)#1 53.45(13)
O(26)-K-O(25)#1 126.39(15)
O(301)-K-O(25)#1 107.1(2)
O(101)-K-O(24) 73.1(2)
O(201)-K-O(24) 106.1(3)
O(27)#1-K-O(24) 126.93(16)
O(26)-K-O(24) 5321(14)

186

Table B13. (con’t)

O(30l)-K-O(24)
O(25)#1-K-O(24)
O(101)-K-O(401)
O(201)-K-O(40l)
O(27)#1-K-O(401)
O(26)-K-O(40l)
O(301)-K-O(401)
O(25)#1-K-O(401)
O(24)-K-O(401)
C(2)-C(1)-C(44)
C(2)-C(1)-C(6)
C(44)-C(l)-C(6)
O(25)-C(2)-C(1)
O(25)-C(2)-C(3)
C(1)-C(2)-C(3)
O(27)—C(3)-C(4)
O(27)-C(3)-C(2)
C(4)-C(3)-C(2)
C(7)-C(4)-C(3)
C(7)-C(4)-C(5)
C(3)-C(4)-C(5)
O(24)-C(5)-C(4)
O(24)—C(5)-C(6)
C(4)-C(5)-C(6)
O(26)-C(6)-C(1)
O(26)-C(6)-C(5)
C(1)-C(6)-C(5)
C(4)-C(7)-C(28)
C(4)-C(7)-C(8)
C(28)-C(7)-C(8)
C(13)-C(8)-C(9)
C(13)-C(8)-C(7)
C(9)-C(8)-C(7)
C(10)-C(9)-C(8)
C(9)-C(10)-C(1 1)
C(9)-C(10)-C(14)
C(11)-C(10)-C(14)
C(10)-C(1 1)-O(18)
C(10)-C(1 1 )-C(12)
O(18)—C(11)-C(12)
C(13)-C(12)-C(1 1)
C(13)-C(12)-C(20)
C(1 1)-C(12)-C(20)
C(8)-C(13)-C(12)

72.7(2)

179.3(2)
862(4)

970(4)

106.9(3)
74.1(3)

175.9(3)
68.8(2)

111.4(2)
120.1(5)
120.0(5)
119.9(5)
124.5(5)
115.6(5)
119.5(5)
124.4(6)
114.6(5)
120.6(5)
119.0(5)
122.6(5)
118.4(5)
122.5(5)
115.8(6)
121.5(5)
122.2(5)
117.7(5)
120.0(5)
121.3(5)
121.1(5)
117.5(5)
118.0(5)
120.7(5)
121.4(5)
121.6(5)
118.6(6)
119.7(6)
121.7(6)
119.7(6)
120.5(6)
119.3(6)
116.9(6)
119.1(6)
123.9(6)
122.8(6)

187

Table 313. (con’t)

C(17)-C(14)-C(15)
C(17)-C(14)-C(l6)
C(15)-C(14)-C(16)
C(17)-C(14)—C(10)
C(15)-C(14)-C(10)
C(16)-C(14)-C(10)
C(1 1)-O(18)-C(19)
C(22)-C(20)-C(21)
C(22)-C(20)—C(12)
C(21)-C(20)-C(12)
C(22)-C(20)-C(23)
C(21)-C(20)-C(23)
C(12)-C(20)-C(23)
C(5)-O(24)-K
C(2)-O(25)-K#2
C(6)-O(26)-K
C(3)-O(27)-K#2
C(43)-C(28)-C(29)
C(43)-C(28)-C(7)
C(29)-C(28)-C(7)
C(28)-C(29)-C(30)
C(35)-C(30)-C(29)
C(35)-C(30)-C(31)
C(29)-C(30)-C(31)
C(34)-C(31)-C(33)
C(34)-C(31)-C(30)
C(33)-C(31)-C(30)
C(34)-C(31)-C(32)
C(33)-C(31)-C(32)
C(30)-C(31)-C(32)
C(30)-C(35)-C(38)
C(30)-C(35)-O(36)
C(38)-C(35)-O(36)
C(35)-O(36)-C(37)
C(35)-C(38)-C(43)
C(35)-C(38)-C(39)
C(43)-C(38)—C(39)
C(40)-C(39)-C(38)
C(40)-C(39)-C(41)
C(38)-C(39)-C(41)
C(40)-C(39)-C(42)
C(38)-C(39)-C(42)
C(41)-C(39)—C(42)
C(28)-C(43)-C(38)

91.5(10)
108.7(8)
120.6(9)
117.5(8)
109.2(7)
109.0(6)
1 13.7(6)
113.2(8)
111.6(8)
114.2(7)
105.3(13)
103.7(11)
108.1(8)
118.3(5)
120.0(4)
119.0(4)
124.2(4)
117.2(5)
123.1(5)
1 19.6(5)
122.9(5)
116.1(5)
125.4(6)
118.4(6)
106.9(8)
110.9(7)
111.2(6)
111.1(8)
105.6(8)
1 11.0(6)
124.8(5)
117.3(6)
117.9(6)
107.0(7)
116.1(5)
124.8(5)
119.1(5)
1 127(8)
93.8(9)
112.4(6)
120.2(11)
110.7(7)
105.4(8)
122.5(5)

188

Table B13. (con’t)

C(1)-C(44)-C(61)
C(1)-C(44)-C(45)

C(61)-C(44)-C(45)
C(50)-C(45)-C(46)
C(50)-C(45)-C(44)
C(46)-C(45)-C(44)
C(47)-C(46)-C(45)
C(46)-C(47)-C(48)
C(46)-C(47)-C(57)
C(48)-C(47)—C(57)
O(55)-C(48)-C(47)
O(55)-C(48)-C(49)
C(47)-C(48)-C(49)
C(50)-C(49)-C(48)
C(50)-C(49)-C(51)
C(48)-C(49)-C(51)
C(45)-C(50)-C(49)
C(53)-C(51)-C(52)
C(53)-C(51)-C(49)
C(52)—C(51)-C(49)
C(53)-C(51)-C(54)
C(52)-C(51)-C(54)
C(49)-C(51)-C(54)
C(56)-O(55)-C(48)
C(59)-C(57)-C(58)
C(59)-C(57)-C(60)
C(58)-C(57)-C(60)
C(59)-C(57)-C(47)
C(58)-C(57)-C(47)
C(60)-C(57)-C(47)
C(62)-C(61)-C(66)
C(62)-C(61)-C(44)
C(66)-C(61)-C(44)
C(61)-C(62)-C(63)
C(64)-C(63)-C(62)
C(64)-C(63)-C(67)
C(62)-C(63)-C(67)
O(71)-C(64)-C(63)
O(71)-C(64)-C(65)
C(63)-C(64)-C(65)
C(66)-C(65)-C(64)
C(66)-C(65)-C(73)
C(64)-C(65)-C(73)
C(65)-C(66)-C(61)

121.2(5)
120.5(5)
118.2(5)
117.6(5)
120.8(5)
121.4(5)
122.1(5)
118.1(5)
119.4(6)
122.5(5)
119.0(5)
120.3(5)
120.4(5)
115.9(5)
120.7(5)
123.4(5)
124.4(5)
1 102(8)
112.7(6)
1 107(6)
107.6(7)
105.6(9)
109.7(6)
118.9(9)

108.1(10)

97.9(11)
116.8(8)
112.4(6)
110.2(5)
110.8(7)
117.1(5)
122.1(5)
120.8(5)
123.0(6)
116.4(5)
126.5(6)
117.0(6)
117.6(5)
118.7(5)
123.6(5)
1 16.7(5)
119.4(6)
123.5(6)
122.9(5)

189

Table 813. (con’t)

C(68)-C(67)-C(63)
C(68)-C(67)—C(70)
C(63)-C(67)-C(70)
C(68)-C(67)-C(69)
C(63)-C(67)-C(69)
C(70)-C(67)-C(69)
C(64)-O(71)—C(72)
C(65)-C(73)-C(76)
C(65)-C(73)-C(75)
C(76)-C(73)-C(75)
C(65)-C(73)-C(74)
C(76)-C(73)-C(74)
C(75)-C(73)-C(74)
C(102)-O(101)-C(105)
C(102)-O(101)-K
C(105)-O(101)-K
O(101)-C(102)-C(103)
C(104)-C(103)-C(102)
C(103)-C(104)-C(105)
O(101)—C(lOS)—C(104)
C(205)-0(201)-C(202)
C(205)-O(201)-K
C(202)-O(201)-K
O(201)-C(202)—C(203)
C(204)-C(203)-C(202)
C(203)-C(204)-C(205)
O(201)-C(205)—C(204)
C(302)-O(301)-C(305)
C(302)-O(30l)—K
C(305)-O(301)-K
O(30l)-C(302)-C(303)
0(301)-C(302)-C(304)
C(303)-C(302)-C(304)
C(304)-C(303)-C(302)
C(303)-C(304)-C(305)
C(303)-C(304)-C(302)
C(305)-C(304)-C(302)
O(301)-C(305)-C(304)
C(405)-O(40l)-C(402)
C(405)-O(401)-K
C(402)-O(40l)-K
C(403)-C(402)-O(401)
C(404)-C(403)-C(402)
C(403)-C(404)-C(405)
O(40l)-C(405)-C(404)

113.0(6)
101.7(9)
110.5(6)
97.4(8)
102.5(7)
131.0(10)
117.4(5)
113.1(9)
113.0(8)
118.7(14)
108.1(8)
103.4(15)
98.2(10)
108.9(11)
124.6(7)
125.5(7)
107.8(10)
110.4(10)
99.6(12)
113.0(11)
106.5(10)
130.6(9)
120.1(7)
101.1(13)
122(2)
98.4(15)
104.0(11)
116.2(11)
125.1(9)
108.6(7)
104.8(13)
71.4(13)
41.2(13)
88.6(18)
121(2)
50.2(15)
83.0(14)
89.4(17)
88(2)
111.8(13)
130.5(14)
102.2(13)
106.7(17)
102(3)
130(3)

 

190

Table 814. Anisotropic displacement parameters (A2 x 103) for (t-

Bu)3(MeO)42K*(C4H30)4. The anisotropic displacement factor exponent
takes the form: —21t2[h2a*2U1.+. . .+2hka*b*U12]

 

 

U11 U22 U33 U23 U13 U12
K 29(1) 103(1) 109(1) 38(1) -3(1) -12(1)
C(1) 24(3) 34(3) 34(3) 1 1(2) 2(2) 5(2)
C(2) 21 (3) 43(3) 33(3) 5(3) 6(2) 0(3)
C(3) 21(3) 35(3) 36(3) 5(2) 4(2) 2(2)
C(4) 28(3) 32(3) 27(3) 3(2) 10(2) -4(2)
C(5) 35(3) 44(4) 25(3) 7(2) 6(3) 0(3)
C(6) 27(3) 33(3) 22(3) 1(2) 5(2) -1(2)
C(7) 21(3) 38(3) 22(3) -4(2) 3(2) 3(2)
C(8) 27(3) 26(3) 29(3) 3(2) 7(2) 8(2)
C(9) 51(4) 28(3) 22(3) 2(2) 3(3) 3(3)
C(10) 42(4) 49(4) 25(3) 1(3) 0(3) 32(3)
C(11) 38(3) 56(4) 29(3) -9(3) 4(3) -4(3)
C(12) 34(3) 40(4) 55(4) -3(3) 14(3) 5(3)
C(13) 40(4) 33(3) 48(4) 0(3) 9(3) 3(3)
C(14) 82(5) 81(5) 36(4) -14(3) 27(3) -45(4)
C(15) 198(13) 140(9) 38(4) 9(5) 2(5) -52(9)
C(16) 200(12) 68(6) 79(6) 39(5) 90(7) -74(7)
C(17) 165(1 1) 77(6) 260(15) 68(8) 181(12) 62(7)
0 (18) 51(3) 62(3) 48(3) 27(2) 9(2) 1(3)
C(19) 82(7) 203(15) 111(9) -106(10) 0(7) 37(8)

C(20) 86(7) 38(4) 85(6) -4(4) 26(5) 20(4)

C(21) 1 18(6) 34(3) 121(7) 32(4) 31(5) 3(4)
C(22) 125(9) 68(6) 172(12) 54(7) 83(9) 66(6)
C(23) 490(30) 45(5) 238(16) 32(7) 240(20) 30(10)
0(24) 20(2) 124(5) 39(3) 16(3) 10(2) 15(3)
0(25) 31(3) 92(4) 38(3) 24(2) 8(2) 1(2)
0(26) 25(2) 73(3) 49(3) 17(2) 0(2) 2(2)
0(27) 25(2) 72(3) 30(2) 12(2) 4(2) 2(2)
C(28) 31(3) 34(3) 21(3) 3(2) 5(2) -7(3)
C(29) 36(3) 31(3) 28(3) -7(2) 6(2) 3(3)
C(30) 44(4) 31(3) 46(4) 36(3) 6(3) 3(3)
C(31) 91(6) 47(4) 40(4) 23(3) 39(4) 9(4)
C(32) 1 10(7) 1 11(7) 67(5) 63(5) 23(5) 58(6)
C(33) 142(8) 74(5) 36(4) -9(4) 2(4) 31(5)
C(34) 186(13) 99(7) 84(7) 25(5) 39(7) -86(8)
C(35) 49(4) 34(3) 39(4) 3(3) 32(3) 5(3)
0(36) 141(6) 36(3) 52(3) 3(2) 24(3) 23(3)
C(37) 340(20) 61(5) 108(8) 45(5) 39(10) 323(9)
C(38) 53(4) 28(3) 28(3) -7(2) 6(3) 3(3)

191

Table B14. (con’t)

C(39)
C(40)
C(41)
C(42)
C(43)
C(44)
C(45)
C(46)
C(47)
C(48)
C(49)
C(50)
C(51)
C(52)
C(53)
C(54)
0(55)
C(56)
C(57)
C(58)
C(59)
C(60)
C(61)
C(62)
C(63)
C(64)
C(65)
C(66)
C(67)
C(68)
C(69)
C(70)
0(71)
C(72)
C(73)
C(74)
C(75)
C(76)
0(101)
C(102)
C(103)
C(104)
C(105)
0(201)

82(6)
231(15)
149(9)
128(9)
30(3)
32(3)
23(3)
48(4)
43(4)
69(4)
45(4)
31(3)
64(5)
219(14)
91(7)
98(7)
172(7)
185(1 1)
83(6)
206(12)
217(15)
242(13)
30(3)
27(3)
29(3)
21(3)
40(4)
34(3)
56(4)
108(7)
154(1 1)
172(1 1)
36(2)
77(7)
77(6)
420(30)
109(8)
232(18)
52(4)
81(7)
128(9)
79(7)
250(20)
94(7)

37(4)
350(20)
61(5)
172(12)
30(3)
28(3)
25(3)
29(3)
40(3)
20(3)
38(3)
43(3)
44(4)
73(6)
78(6)
138(9)
22(2)
76(6)
45(4)
55(5)
269(16)
72(5)
43(3)
45(4)
43(3)
46(4)
36(3)
35(3)
86(5)
50(5)
131(1 1)
255(16)
62(3)
155(1 1)
38(4)
71(6)
144(10)
41(6)
88(5)
79(7)
103(7)
174(14)
129(1 1)
128(8)

44(4)
26(4)
70(5)
151(1 1)
25(3)
32(3)
32(3)
30(3)
30(3)
35(3)
23(3)
20(3)
44(4)
43(5)
84(6)
166(1 1)
50(3)
69(6)
28(3)
104(7)
109(8)
56(5)
26(3)
34(3)
33(3)
39(4)
40(3)
27(3)
26(3)
91(7)
244(17)
71(6)
42(3)
83(7)
87(7)
107(8)
105(8)
700(50)
286(12)
271(18)
146(9)
270(20)
650(50)
241(13)

-9(3)
55(7)
26(4)
3700)
3(2)
7(2)
3(2)

3 1(2)
4(3)
-1(2)
2(2)
6(2)
30(3)
37(4)
29(5)
3 15(9)
4(2)
-2(5)
0(3)
20(4)

339(10)

19(4)
8(3)
-2(3)
-4(3)
3 1(3)
-5(3)
10(2)
-2(3)
1(5)
38(1 1)
1 12(8)
36(2)
31(7)
-4(4)
63(6)
21(7)
3404)
8(6)
35(10)
7(6)
58(13)
160(20)
24(8)

192

-25(4)
-30(6)
-51(5)
-102(9)
-7(2)
7(3)
-1(2)
2(3)
-16(3)
-17(3)
-6(3)
-1(2)
-11(3)
-2(6)
-9(5)
-6(7)
-33(4)
~21(6)
-28(4)
-133(8)

342(10)

-57(6)
0(3)
13(3)
10(3)
9(3)
3(3)
8(3)
14(3)
60(6)
158(12)
-56(7)
12(2)
3 1(6)
21(5)
146(12)
38(7)
230(30)
69(5)
31(9)
-52(7)
22(10)
350(30)
-47(7)

1 1(4)
378(16)
-5(5)
91(9)
4(2)
0(2)
2(2)
6(3)
4(3)
2(3)
3(3)
-9(3)
3(3)
30(7)
-7(5)
37(7)
2(3)
39(7)
3(4)
29(6)
199(14)
65(6)
-5(3)
3(3)
-5(3)
4(3)
7(3)
8(3)
26(4)
38(5)
-79(9)
334(12)
3(2)
-4(7)
8(4)
130(1 1)
86(8)
34(8)
34(4)
-4(6)
33(6)
63(8)
308(13)
5(6)

Table B14. (con’t)

C(202)
C(203)
C(204)
C(205)
0(301)
C(302)
C(303)
C(304)
C(305)
0(401)
C(402)
C(403)
C(404)
C(405)

44(5)
130(15)
165(13)
37(4)
81(5)
69(6)
87(8)
320(40)
91(7)

1 18(7)
92(10)
123(13)
168(15)

88(7)
560(60)
157(12)
124(9)
230(1 1)
177(12)
410(30)
380(40)
169(14)
184(9)
310(30)
213(16)
188(16)

1220(110) 220(20)

390(30)
220(20)
105(8)
310(20)
184(9)
214(14)
258(19)
400(50)
96(8)
156(8)
340(30)
128(1 1)
205(16)
140(20)

27(1 1)
60(30)
67(9)
66(1 1)
309(9)
323(12)
280(20)
30(30)
36(8)
6(7)
180(30)
34(1 1)
33(12)
66(18)

-49(9)
55(15)
-40(8)
-15(8)
42(6)
3(7)
21(10)
220(40)
5(6)
38(6)
32(14)
7(9)
33(13)
200(40)

21(5)
380(30)
20(10)
-18(5)
-75(7)
2(7)
34(12)
-160(40)
38(8)
57(6)
1(14)
61(12)
61(13)
320(40)

 

193

Table B15. Hydrogen coordinates ( x 104) and isotropic displacement parameters (A2 x
103) for (t-Bu)g(MeO)42K-(C4H30)4.

 

 

x y l U(C‘I)

H(9) 9110 4157 4630 42

H(13) 9370 7099 3394 51

H(15A) 9154 4386 6674 191
H(ISB) 7729 4154 7137 191
H(15C) 7883 5343 6774 191
H(16A) 8536 3209 5510 170
H(16B) 6914 3561 5279 170
H(16C) 7322 2970 6156 170
H(17A) 5439 4982 5782 250
H(17B) 6034 5621 6313 250
H(17C) 5873 4430 6661 250
H(19A) 7758 7271 6754 . 184
H(19B) 8905 7422 6053 184
H(19C) 8838 6266 6549 184
H(21A) 9066 8649 3320 144
H(2 13) 7947 9674 3457 144
H(2 1 C) 7442 8782 3130 144
H(22A) 6212 8318 5123 227
H(22B) 5673 8570 4248 227
H(22C) 6154 9466 45 81 227
H(23A) 9824 8583 4677 380
H(23B) 8688 8492 53 88 380
H(23C) 8575 9568 4747 380
H(29) 1 l 180 3557 2492 39

H(32A) 14634 1216 2606 150
H(3ZB) 14060 370 2260 150
H(32C) 14006 298 3185 150
H(33A) 13091 2614 1670 139
H(33B) 1 1426 2669 1679 139
H(33C) 12499 1734 1375 139
H(34A) 1 1243 393 3230 169
H(34B) 1 1428 378 2317 169
H(34C) 10319 1305 2609 169
H(3 7A) 12146 -524 4857 245
H(37B) 1 1 146 340 4242 245
H(37C) 1 1344 561 5085 245
H(40A) 12658 1661 6508 297
H(40B) 1 1898 1251 5892 297
H(40C) 1 1363 2427 6029 297
H(41A) 13957 3612 5388 137

194

Table B15. (con’t)

H(4lB)
H(41C)
H(42A)
H(42B)
H(42C)
H(43)

H(46)

H(50)

H(52A)
H(5213)
H(52C)
H(53A)
H(53B)
H(53C)
H(54A)
H(54B)
H(54C)
H(56A)
H(56B)
H(56C)
H(58A)
H(58B)
H(58C)
H(59A)
H(59B)
H(59C)
H(60A)
H(6OB)
H(6OC)
H(62)

H(66)

H(68A)
H(68B)
H(68C)
H(69A)
H(69B)
H(69C)
H(7OA)
11(7013)
H(7OC)
H(72A)
11(7213)
H(72C)
H(74A)

13654
12369
15367
14846
15333
11795
8745
9373
9130
7453
8148
6574
6564
5877
10234
9166
9261
7770
8853
8945
7918
6686
6328
6086
5275
5600
9058
8139
7616
11449
11188
13329
13716
12107
14692
15053
15025
11446
11979
10804
12730
11609
11596
13162

2972
3675
1923
852
1524
4139
8242
8804
9686
9786
10639
12059
12100
11241
11049
11946
11994
12920
11966
12180
8666
9397
8875
11467
10507
11078
10481
11478
11102
8201
5247
9333
8680
9206
6656
7231
7816
7716
8807
8652
5122
6099
4928
2808

6250
5730
4875
5259
5806
4621
-1386
724
1491
1555
1820

909
646
582
931

-1536
-1813
-956
-2405
-2951
-2064
-1966
-l806
-2685
-2829
-2524
-3240
-1377
-170
-2930
-2065
-2269
-2964
-2314
-3217
-3508
-3747
-3062
-3566
-3349
-2864
-85

137
137
235
235
235
37

43

39

162
162
162
126
126
126
178
178
178
163
163
163
180
180
180
319
319
319
181
181
181
46

45

133
133
133
255
255
255
259
259
259
149
149
149
370

195

Table B15. (con’t)

H(74B)
H(74C)
H(75A)
H(75B)
H(75C)
H(76A)
H(76B)
H(76C)
H(IOA)
H(IOB)
H(IOC)
H(IOD)
H(IOE)
H(IOF)
H(IOG)
H(IOH)
H(20A)
H(20B)
H(20C)
H(20D)
H(20B)
H(2OF)
H(20G)
H(ZOH)
H(30A)
H(3OB)
H(30C)
H(3OD)
H(30E)
H(3OF)
H(30G)
H(3OH)
H(40D)
H(40B)
H(40F)
H(40G)
H(4OH)
H(401)

H(4OJ)

H(4OK)

11924
13530
14553
14896
14415
12008
11455
10773
5591
4721
5918
6870
8593
7595
7548
8411
5953
5117
4346
4883
2924
2377
2236
2214
2154
2699
3125
1713
2908
3750
5635
5091
8529
7940
8435
7908
5858
6484
6400
4987

3753
3892
2889
3944
3928
2727
3738
3653
3129
3547
2148
1783
2198
2672
4271
3793
8629
9526

10252

9243
9063

10134

9219
8223
7108
8196
7889
7394
5917
6170
6385
5447
5076
4161
6007
5074
5828
6864
6677
6384

81

10
-1240
-1070
-1919
-1189
-1858
-990
2004
2730
3537
2818
3238
3913
3193
2494
384
838
-171
-544
-597
-269
1002
620
3018
2911
4293
4323
4114
4768
3944
3662
-14
-280
-1245
-1516
-1511
-1519
-365
-551

370
370
219
219
219
481
481
481
164
164
152
152
220
220
423
423
217
217
354
354
165
165
209
209
173
173
268
268
461
461
157
157
351
351
197
197
242
242
646
646

 

196

llllllllllllillllllllll