This is to certify that the
thesis entitled
Synthesis and Spectrosc0py of Electron Exchange
Coupled Semiquinone Complexes of Chromium and Nickel

presented by

Richard Tirasak Praseuth

has been accepted towards fulfillment
of the requirements for

M. S . degree in Chemistry

 

 

QM /¢ 777 (L
fl Major professor

 

Date 7/Y /z.u1

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

LIBRARY
Michigan State
University

 

 

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

 

APR172

Bil

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DaieDue.965-p.15

SYNTHESIS AND SPECTROSCOPY OF ELECTRON EXCHANGE COUPLED

SEMIQUINONE COMPLEXES OF CHROMIUM AND NICKEL

By

Richard T. Praseuth

A THESIS

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

MASTER OF SCIENCE
Department of Chemistry

2002

ABSTRACT
SYNTHESIS AND SPECTROSCOPY OF ELECTRON EXCHANGE COUPLED
SEMIQUINONE COMPLEXES OF CHROMIUM AND NICKEL
By

Richard T. Praseuth

In this thesis, the impact of electron exchange coupling on the physical properties
of nickel-semiquinone and chromium-semiquinone complexes is investigated. Following
a brief discussion of magnetism in general, the synthesis, structure, and physical
properties of [Ni(tren)(3,5-DTBSQ)](PF6) (where 3,5-DTBSQ is 3,5-di-tert-
butylsemiquinone) are discussed. A group theoretical analysis demonstrates that the spin
forbidden transition 1E <— 3A2 transition in NiII should be altered due the presence of
ferromagnetic coupling between Nill and a semiquinone ligand. In particular, it is shown
that a transition from the ground state to the doublet low-lying excited state could be
accomplished thermally. However, variable-temperature electronic absorption data failed
to show the expected thermochromic effect. The magnetic properties of Cr(3,5-DTBSQ)3
and Cr(PhenSQ)3 are addressed to assess whether PhenSQ might couple more weakly to
a metal ion and yield the thermochromic response that could not be achieved with the Ni-
3,5-DTBSQ complex. In addition, the results of a density functional theory study of
phenanthrenequinone and phenanthrenesemiquinone are discussed. The primary goals of
this study were to explore the spatial distribution of the molecular orbitals and the relative
ordering of energy levels of the molecular orbitals upon reduction from the quinone to the

semiquinone redox state. Finally, future directions of this project are briefly discussed.

In loving memory of my grandmother, Judy Wong Sher.

iii

ACKNOWLEDGEMENTS

First, I would like to take this opportunity to thank Professor James McCusker for
all his support and guidance throughout the past two years at Michigan State University.
In addition, I would like to thank the members of the McCusker Group; past and present.
They have all been very helpful with my synthetic inquiries and great companions on the
frequent trips down to Grand River (Grand Lake?) for lunch and coffee at Starbucks.
Specifically, I would like to thank Jeff and Laura for reading and editing the multiple
chapters of my thesis. Also, if it weren’t for Jeff’ s expertise in x-ray crystallography, my
thesis would have been void of any crystal structures and twenty pages shorter. Thank
you very much Jeff.

Next, I would like to extend sincere thanks to Donna and Mike. You guys have
been more than generous with me and over the past year or two, you guys have proven to
be my “true” friends. Thank you very much for everything! You guys are great friends
and you made life in East Lansing bearable. I will never forget the two of you.
Furthermore, I would like to thank Javier for his friendship and willingness to skip work
midday to play basketball with me. It was great playing ball with you and I hope we can
do it again in the future.

Last, but not least, I would like to thank my family. I am very grateful for the
encouragement and constant instilment of confidence in me to strive for my goals. I’m
not sure what path I’ll choose in the future but I’m confident that from the support and
love given to me by my family, I will always be successful. Many thanks to my brothers

(Mike, David, and Alex) for being the closest and best of friends to me. You guys have

iv

been my role models since day one and the person I am today, good and bad, can be
accredited to you guys. Thanks also to my sister-in-laws, Francine and Cheri, for being
the sisters I’ve never had. I’ve enjoyed the many conversations we’ve had and semi-

twisted humor we’ve shared.

TABLE OF CONTENTS

LIST OF TABLES .............................................................................................................. x
LIST OF FIGURES .......................................................................................................... xii
LIST OF ABBREVIATIONS ......................................................................................... xvii

CHAPTER ONE: Synopsis of Electron Exchange Coupling and the Photophysics

of Transition Metal Complexes ..................................................... 1
1.1 Introduction ......................................................................................................... 1
1.2 Definition of Electron Exchange Coupling ......................................................... 2
1.3 Determination of the Exchange Coupling Constant, J ........................................ 3
1.4 Effects of Electron Exchange Coupling on Ground State Properties ................. 7
1.5 Use of Quinones in Metal Compounds ............................................................... 8
1.6 Research Objectives .......................................................................................... 10
1.7 References and Notes ........................................................................................ 16

CHAPTER TWO: Electron Exchange and the Photophysical Properties of a

Nickel Monosemiquinone Complex: Synthesis and

Spectroscopy ................................................................................. 19

2.1 Introduction ..................................................................................................... 19
2.21 Experimental Section ...................................................................................... 22
2.2.1 General ............................................................................................. 22

vi

2.2.2 Synthesis of Ni(tren)Br2 ................................................................... 22

2.2.3 Synthesis of [Ni(tren)(3,5-DTBSQ)](PF6) ....................................... 24
2.2.4 X-ray Structure Determination ........................................................ 24
2.2.5 Cyclic Voltammetry ......................................................................... 25
2.2.6 Variable-Temperature Magnetic Susceptibility ............................... 30
2.3 Results and Discussion ................................................................................... 30
2.3.1 Synthesis .......................................................................................... 30

2.3.2 Description of the X-Ray Structure of [Ni(tren)(3,5-DTBSQ)](PF6)

.................................................................................................................... 31

2.3.3 Electrochemistry .............................................................................. 34

2.3.4 Magnetic Susceptibility ................................................................... 36

2.3.5 Electronic Absorption Spectra ......................................................... 42

2.4 Concluding Comments .................................................................................... 44
2.5 References and Notes ...................................................................................... 47

CHAPTER THREE: Synthesis and Magnetic Properties of Cr(3,5-DTBSQ)3 and

Cr(PhenSQ)3 .............................................................................. 49

3.1 Introduction ..................................................................................................... 49
3.2 Experimental Section ...................................................................................... 50
3.2.1 General ............................................................................................. 50

3.2.2 Synthesis of Cr(3,5-DTBSQ)3 ......................................................... 50

3.2.3 Synthesis of Cr(PhenSQ)3 ................................................................ 51

3.2.4 Variable-Temperature Magnetic Susceptibility ............................... 51

vii

3.3 Results and Discussion ................................................................................... 51
3.3.1 Magnetism of Cr(3,5-DTBSQ)3 ....................................................... 51

3.3.2 Magnetism of Cr(PhenSQ)3 and Comparison with Cr(3,5-DTBSQ)3

.................................................................................................................... 55
3.4 References and Notes ...................................................................................... 59
CHAPTER FOUR: Density Functional Theory Analysis of the Electronic
Structures of Phenanthrenequinone and
Phenanthrenesemiquinone .......................................................... 61
4.1 Introduction ..................................................................................................... 61
4.2 Computational Methods .................................................................................. 63
4.2.1 General Methods .............................................................................. 63
4.2.2 Geometry Optimizations .................................................................. 63
4.2.3 Single Point Calculations ................................................................. 64
4.3 Results and Discussion ................................................................................... 67
4.3.1 Optimized Geometries of PhenQ and PhenSQ ................................ 67
4.3.2 Molecular Orbitals of Phenanthrenequinone (PhenQ) ..................... 70
4.3.3 Molecular Orbitals of Phenanthrenesemiquinone (PhenSQ) ........... 76
4.3.4 Energy Spacing of the Molecular Orbitals of PhenQ and
PhenSQ ............................................................................................ 79
4.3.5 Natural Population Analysis (NPA) Charge Densities and Spin
Densities ........................................................................................... 81
4.4 Concluding Remarks ....................................................................................... 83

viii

4.5 Future Directions ............................................................................................ 86

4.6 References ....................................................................................................... 88

APPENDIX: Tables of Bond Lengths, Bond Angles, and Anisotropic
Thermal Factors, for the Crystal Structure of

[Ni(tren)(3,5-DTBSQ)](PF6) Presented in Chapter 2 ......................... 91

LIST OF TABLES

CHAPTER TWO:
Table 2.1. Crystal Data and Structure Refinement for [Ni(tren)(3,5-
DTBSQ)](PF6)*AcOEt ........................................................................ 26
Table 2.2. Atomic Coordinates [ x 10"] and Equivalent Isotropic Displacement
Parameters [A2 x 103] for [Ni(tren)(3,5-DTBSQ)](PF6) ...................... 27
Table 2.3. Selected Bond Lengths and Bond Angles for [Ni(tren)(3,5-

DTBSQ)](PF6) ..................................................................................... 32

CHAPTER FOUR:
Table 4.1. Total Self-Consistent Field Energies (Hartree) Obtained at the
B3LYP/6-311G** Level Using the Optimized Geometries of PhenQ
and PhenSQ .......................................................................................... 66
Table 4.2. Selected Optimized Bond Lengths (A) for Phenanthrenequinone
(PhenQ) and Phenanthrenesemiquinone (PhenSQ) Obtained from Self-
Consistent Field (SCF) Geometry Optimizations at the B3LYP/6-31G*
Level ...................................................................................................... 69
Table 4.3. Energies and Percent Atomic Contributions to the Frontier Molecular
Orbitals of Phenanthrenequinone (PhenQ) Obtained at the U-B3LYP/6-

311G** Level ........................................................................................ 71

Table 4.4. Energies and Percent Atomic Contributions to the Frontier Molecular
Orbitals of Phenanthrenesemiquinone (PhenSQ) Obtained at the U-
B3LYP/6-311G** Level ....................................................................... 77

Table 4.5. Absolute and Relative Orbital Energies (CV) of Phenanthrenequinone
and Phenanthrenesemiquinone Calculated at the U-B3LYP/6-311G**
Level ...................................................................................................... 80

Table 4.6. NPA Atomic Charge Densities of Phenanthrenequinone and
Phenanthrenesemiquinone Calculated at the U-B3LYP/6-311G** Level
............................................................................................................... 82

Table 4.7. NPA Atomic Spin Densities and Mulliken Net Spin Densities for

Phenanthrenequinone and Phenanthrenesemiquinone Obtained at the U-

B3LYP/6-311G** Level ....................................................................... 84

APPENDIX:
Table A1. Bond Distances (A) for [N i(tren)(3,5-DTBSQ)](PF 6) ......................... 92
Table A2. Bond angles (°) for [N i(tren)(3,5-DTBSQ)](PF 6) ............................... 95

Table A3. Anisotropic Displacement Parameters [A2 x 103] for [N i(tren)(3,5-
DTBSQ)](PF6). The Anisotropic Displacement Factor Exponent Takes

the Form: -2p2[(ha*)2Ul 1 +...+2hka*b*Ul2] ................................. 101

xi

LIST OF FIGURES

CHAPTER ONE:

Figure 1.1.

Figure 1.2.

Figure 1.3

Figure 1.4.

Schematic representation of a two S = ‘/2 spin system in the absence
and presence of electron exchange coupling. The electronic state
ordering on the right is appropriate for an antiferromagnetically
coupled system (i.e., J < O). ................................................................... 6
The redox chemistry of quinones and spin states corresponding to

the respective redox states ..................................................................... 9

. Examples of commonly studied o-semiquinone ligands:3,6-di-tert-

butquuinone (3,6-DTBQ), 3,5-di-tert-butquuinone (3,5-DTBQ),
tetrachloro-1,2-quinone (TCQ), 9,10-phenanthrenequinone (PhenQ)

.............................................................................................................. 11
Results of a ligand field analysis for a Cr(III)-quinone dyad applied by
Wheeler et. al. The first column indicates two of the lowest-lying field
states present in a (13 metal ion in 0;, symmetry. The second column
represents the splitting upon lowering the symmetry from Oh to CZV,
which corresponds to [Cr(tren)(3,6-DTBSQCat)]+. The third column
represents the splitting caused by the coupling between the unpaired
spins of the metal with an S = ‘/2 spin of b; symmetry. This

corresponds to the complex, [Cr(tren)(3,6-DTBSQ)]2+. The relative

xii

ordering of the spin-coupled states for the Cr-SQ diagram assumes

antiferromagnetic coupling in the excited states. Other higher-lying

excited states that are present have been omitted for clarity ............ 13
Figure 1.5. Tanabe-Sugano diagram for a (18 metal complex in octahedral

symmetry .............................................................................................. 1 4

CHAPTER TWO:

Figure 2.1. Room temperature absorption spectra of a) [Cr(tren)(3,6-
DTBCat)](PF6) and b) [Cr(tren)(3,6-DTBSQ)](PF6)2 measured in 4:1
EtOH/MeOH ........................................................................................ 21

Figure 2.2. Results of a ligand field analysis for a Ni(II)-quinone dyad. The first
column indicates two of the lowest-lying ligand field states present in
a d8 metal ion in 0 symmetry. The second column represents the
splitting upon lowering the symmetry from O to C2,, which
corresponds to a non-exchange coupled Ni(II) complex. The third
column represents the splitting caused by the coupling between the
unpaired spins of Ni(lI) with the semiquinone ligand (S = 1/2) of b.
symmetry. This corresponds to the complex, [Ni(tren)(3,5-DTBSQ)]+.
Other higher-lying excited states that are present have been omitted for
clarity ................................................................................................... 23

Figure 2.3. Drawing and labeling scheme of the two crystallographically unique

[Ni(tren)(3,5-DTBSQ)]Jr cations determined by single-crystal x-ray

xiii

diffraction. The structural and crystallographic details are summarized
in Tables 2.1 through 2.3 ..................................................................... 33
Figure 2.4. Cyclic voltarnmogram of [Ni(tren)(3,5-DTBSQ)](PF6) in deaerated
dichloromethane ................................................................................... 35
Figure 2.5. The redox chemistry of [Ni(tren)(3,5-DTBSQ)]+ .............................. 37
Figure 2.6. Plot of effective magnetic moment versus temperature for
[Ni(tren)(3,5-DTBSQ)](PF6) at 2.5 T with (squares) and without
(triangles) the correction for temperature-independent paramagnetism
(TIP) ..................................................................................................... 39
Figure 2.7 Schematic drawing of the effect of the zero-field splitting on the S =
3/2 ground state of [Ni(tren)(3,5-DTBSQ)]+. As drawn, D > 0; if D <
0, then the :t 3/2 levels would be lower in energy ............................... 41
Figure 2.8 Room-temperature absorption spectrum of
[Ni(tren)(3,5-DTBSQ)](PF6) measured in dichloromethane ............... 43
Figure 2.9 Variable-temperature electronic absorption spectra of [Ni(tren)(3,5-

DTBSQ)](PF6) in dimethylformamide solution ................................... 45

CHAPTER THREE:

Figure 3.1. Plot of effective magnetic moment versus temperature for Cr(3,5-
DTBSQ)3 (circles). The solid line represents the fit to a theoretical

model; g = 2.00, TIP = 150 x 10'6 cm3/mol. and J = -440 cm". .......... 52

xiv

Figure 3.2. Energy levels for Cr(SQ)3 complexes resulting from intramolecular

antiferromagnetic exchange between the unpaired electrons of the o-

semiquinone ligands and the Cr(III) ion .............................................. 56
Figure 3.3. Plots of effective magnetic moment versus temperature for
Cr(PhenSQ)3 (circles) and Cr(3,5-DTBSQ)3 (triangles) ...................... 58
CHAPTER FOUR:
Figure 4.1. Labeling scheme and coordinate system used for

Figure 4.2.

Figure 4.3.

Figure 4.4.

Figure 4.5.

Figure 4.6.

Figure 4.7.

Figure 4.8.

phenanthrenequinone and phenanthrenesemiquinone. The peripheral
carbons are defined as C2, C3, C4, C5, C11, C12, C13, and C14 ....... 65
The redox states of phenanthrenequinone .......................................... 68
MO 52a of phenanthrenequinone. The orbital density is localized
above and below the plane of the ring ................................................ 72
MO 53a of phenanthrenequinone. The orbital lies in the plane of the
ring ...................................................................................................... 72
MO 540 (HOMO) of phenanthrenequinone. The orbital density lies
above and below the plane of the ring ................................................ 74
MO 550. (LUMO) 0f phenanthrenequinone. The orbital density lies
above and below the plane of the ring ................................................ 74
MO 56a of phenanthrenequinone. The orbital density lies above and
below the plane of the ring .................................................................. 75
MO 550. (HOMO) of phenanthrenesemiquinone. The orbital density

lies above and below the plane of the ring .......................................... 78

XV

Figure 4.9. MO 558 (LUMO) of phenanthrenesemiquinone. The orbital density

lies above and below the plane of the ring .......................................... 78

xvi

3,5-DTBCat
3,5-DTBSQ
3,6-DTBSQ
AcOEt

Cat

CTH

DF T

NPA
PhenSQ
SCF

SQ

TCSQ

List of Abbreviations

3,5-di-tert-butylcatechol
3,5-di-tert-butylsemiquinone
3,6-di-tert-butylsemiquinone
Ethyl acetate

Catechol
5,7,7,]2,14,14-hexamethyl-l ,4,8,1 l-tetraazacyclodecane
Density fimctional theory
Natural population analysis

9, 1 O-phenanthrenesemiquinone
Self-consistent field
Semiquinone

Tetrachloro-l ,2—semiquinone

xvii

CHAPTER ONE
Synopsis of Electron Exchange Coupling and the Physical Properties of Transition

Metal Complexes

1.1 Introduction

For many years, chemists have been interested in understanding the physical
properties, and ultimately the reactivity, of transition metal complexes. These properties
can often be very complex and require an understanding of the electronic structure of the
compound in question. One aspect of molecular electronic structure that has governed
considerable attention relates to magnetism. In general, the arrangement of electrons in
the valence orbitals of a molecule presents two types of magnetic behavior:
paramagnetism or diamagnetism. The compound is defined as paramagnetic if it contains
any unpaired electrons, whereas, if the electrons are all paired, the molecule is said to be
diamagnetic. Depending on the type of complex, other factors may also be present,
causing perturbations in the overall electronic structure. In addition, situations may arise
where multiple paramagnetic centers are present in a single compound. To a large extent,
the electronic structure of this type of system can be perturbed by the mutual interaction
of the unpaired spins. The coupling between these spin centers, which occurs in the
absence of any external field (e.g., magnetic or electronic), can vary in magnitude and the
resulting physical properties of the chemical system will differ depending on the nature

of the interaction.1 This interaction (i.e., electron exchange coupling) has become an

increasingly important area of study for many disciplines of chemistry.2'7

1.2 Definition of Electron Exchange Coupling

Electron exchange coupling is an electrostatic interaction that arises whenever
two or more paramagnetic centers are in close proximity. In order for this interaction to
occur in any molecular system, two general criteria must be met: First, the complex must
contain two or more species with unpaired electrons; Second, there must be a pathway
through which the spins can interact. In addition, the spin orbitals must be reasonably
similar energetically.8 The most common type of exchange observed is intramolecular
exchange, which is exchange coupling between spin centers within the same molecule.
This mode of coupling can occur via two mechanisms, either through direct overlap
between the two spin centers or through a diamagnetic bridge, defined as
“superexchange”.9 In the case of superexchange, a metal ion can act as the bridge, which
provides a mode of interaction between the radical ligands. There are several examples of
metal mediated coupling between radical ligands in the literature.“"'2 For example,
Adams et. al.'3 reported the synthesis and characterization of the gallium complex,
Ga(3,5-DTBSQ)3 (3,5-DTBSQ = 3,5-di-tert-butylsemiquinone) which exhibits
superexchange between the semiquinone ligands through the diamagnetic gallium(III)
metal.

To a lesser extent, intermolecular exchange is also observed which is defined as
exchange between separate molecules.”IS Although this type of coupling occurs, it is

quite often very weak. For example, complexes synthesized by Schulz et al.'4 have shown

very weak intermolecular exchange coupling between separate Cu(II) centers mediated
by hydrogen bonding. In general, intermolecular exchange coupling is orders of
magnitude smaller than intramolecular exchange, and therefore will not be discussed
further.

Another aspect of exchange that must be considered is the nature of spin
alignment. This is a consequence of the details of the interactions between orbitals
containing the unpaired electrons. If the electrons couple such that their spins are aligned
antiparallel, the interaction is said to be antiferromagnetic. Orthogonality between
magnetic orbitals, leading to a spin-aligned configuration is called a ferromagnetic
interaction.’ This difference in coupling and the resulting perturbations in the physical

properties of the compound will be discussed further in the following sections.

1.3 Determination of the Exchange Coupling Constant, J

The basis of understanding how exchange coupling affects the properties of a
system requires knowledge concerning the strength of the interaction between the spin
centers. In general, the coupling can be described by the spin Hamiltonian (H)16 of the

form shown in Equation 1.1, where S; and Sj represent the quantum mechanical

H = -2 2i 2.1 Jij Si'Sj (1.1)

h

spin operators on the it and j1h paramagnetic site, and J is the exchange coupling

constant which quantifies the magnitude of the coupling between the spin centers. Based

on Equation 1.1, a negative value of J represents an antiferromagnetically coupled system
(spins are aligned antiparallel), whereas a positive value represents a ferromagnetic
system (spins are aligned parallel).

A simple but illustrative example of exchange coupling can be envisioned
between two paramagnetic centers, each with S = 1/2 . Based on Equation 1.1, the

Hamiltonian that describes this system is given below in Equation 1.2.
HZ-ZJQ SI'SZ (1.2)

Using the Kambe approximation'7 and defining the total spin of the complex, ST, as ST =

S] + 82, an operator-equivalent form of Equation 1.2 can be derived (Equation 1.3).
H=-Ji2 [ST2 '512-522] (13)

From Equation 1.3, the corresponding eigenvalue expression can be generated, Equation

1.4, and the two eigenvalues (ESTzo and Esp.) can be easily obtained as +1.5 J and — 0.5 J,
E(S) = -J [ST(ST+1)'SI(SI+1 )-Sz(Sz+1)l (1-4)
respectively. The energy that separates the two states is given in Equation 1.5.

AB = E(S'|~—-0) — E(S'r=l) = 1.5J— (-0.5J) = 2J (1.5)

The resulting energy levels (i.e., spin ladder) are shown in Figure 1.1. The left
side of the figure depicts two S = '/2 spin centers prior to exchange interactions. The two
possible spin interactions, ST = 0 and ST = 1, are degenerate prior to coupling. Upon
introduction of exchange coupling interactions, the degeneracy of the spins, ST = 0 and ST
= 1, is removed and the states are separated by 2.] (the energies of these electronic states
are given by Equation 1.4).

For molecular systems, the magnitude of exchange coupling (J) is often
determined by measuring the bulk magnetization of a sample as a function of
temperature. Magnetic susceptibility is determined from a Boltzman distribution over the
available spin states of the system. The general expression for magnetic susceptibility is
given by Equation 1.6.18 This can be further simplified to a more useful form, given in
Equation 1.7”, where N is Avogadro’s number, g is the gyromagnetic factor of the
electron, B is the electron Bohr magneton, k3 is the Boltzman constant, T is temperature,
S is the total spin quantum number of a given spin state, and E(S) is the energy associated
with that spin state (e. g., Equation 1.4). The summation is made over all the spin states in

the system.

2 2 2 ZS Ms’expt-E(s>/kBTI
Ng B S Ms=-S

knT Z (28 + 1) eXp[-E(S)/kBTI
s

(1.6)

 

x:

 

ST=1(E=-0.SJ)
iSli=lSZ =1/2

E [=]

 

 

ST=0(E=+1.SJ)

Non-Exchange Exchange
Coupled System Coupled System
J = 0 J ¢ 0

Figure 1.1. Schematic representation of a two S = ‘/2 spin system in the absence
and presence of exchange coupling. The electronic state ordering on the right is

appropriate for an antiferromagnetically coupled system (i.e., J < O).

Ngzfiz ZS) S(S+1)(ZS+l)exp[-E(S)/kBT]

x 2 (1.7)

 

3kBT 2 (23+ l)exp[-E(S)/kBTI
s

The variable temperature magnetic susceptibility data obtained from experiment can then

be fit to Equations 1.6 or 1.7 to obtain a value for the exchange coupling constant, J.'9
1.4 Effects of Electron Exchange Coupling on Ground State Properties

As mentioned above, a molecule’s physical properties are largely determined by
its electronic structure. This can be altered dramatically by the introduction of exchange
interactions, which will manifest themselves in the ground state properties of the
complex. The most obvious effect can be observed in the magnetism of these complexes.
For example, in the absence of exchange coupling, a molecule which contains two high
spin Fe(III) centers Should have a magnetic moment20 of approximately 8.37 ha. The
magnetic moment can deviate substantially if the two metals are coupled. Our group has
found that two antiferromagnetically coupled Fe(III) metal centers bridged by an oxo
group, can exhibit an effective magnetic moment of 2.45 in; (at 300K) as seen in the
complex F ezO(OzCCH2F )2(Tp)2 (Tp = 1-hydro(trispyrazolyl)borate).2| This demonstrates
the dramatic effect of an interaction that is nominally ca. 100 cm'l in magnitude.

The absorptive properties of a molecular complex can also be perturbed by
exchange interactions. In the high spin non-exchange coupled Fe(III) dinuclear cluster

mentioned above, the ground state of each Fe(III) atom in octahedral geometry would be

6A.. By examination of the Tanabe-Sugano22 diagram, it is clear that there are no spin-
allowed ligand-field transitions from the ground state. However, upon the introduction of
exchange coupling, the overall spin states of both the ground state and the excited states
of the system are altered. Therefore, it is possible for excitations which were formally
forbidden in the absence of exchange to become allowed upon introduction of exchange

coupling interactions.

1.5 Use of Quinones in Model Compounds

Quinones have attracted interest from a number of researchers because of their
“non-innocent” behavior as ligands with respect to their oxidation states. In general,
quinones possess three stable oxidation states in which the semiquinone redox state is
paramagnetic with S = l/2 (Figure 1.2). A large number of complexes containing quinones
have been synthesized and exploited as models to study the effects of electron exchange
coupling on the physical properties of molecules. The type of exchange-coupled systems
studied have varied from simple complexes consisting of a paramagnetic metal ion

23-26
d

chelated by a single semiquinone Iigan to diamagnetic metals chelated by multiple

”'2‘”. A third type of system, which is more complicated, consists

semiquinone ligands
of multiple semiquinone ligands chelated to a paramagnetic metal ion28'30. This type of

system involves two modes of exchange; direct exchange between the metal and the

semlquinone llgands, and superexchange between the semlquinone ligands.

 

Quinone Semiquinone Catecholate

S=0 S='/2 S=0

Figure 1.2. The redox chemistry of quinones and spin states corresponding to

the respective redox states.

The advantage of using a redox-active ligand such as a quinone lies in the ability
to effectively turn exchange coupling on or off in a given system without making
substantial changes in the overall composition of the molecule. This can be accomplished
by oxidizing the catecholate ligand (S = O) to the semiquinone form (S = 1/2).3 I A few

examples of the widely studied ortho-quinone ligands are listed in Figure 1.3.

1.6 Research Objectives

Previous studies have investigated the effects of exchange coupling on the
spectroscopic properties of Cr'”-semiquinone23‘24 and Crm-catecholate systems.23 These
systems represent excellent models for the systematic study of the effect of electron
exchange coupling on the physical and photophysical properties of metal complexes.
Focusing on [Cr(tren)(3,6-DTBSQ)]2+, where tren is tris-2-(aminoethyl)amine, complex,
the magnetic susceptibility data confirmed the presence of antiferromagnetic coupling
between the metal and the semiquinone ligand.23 A qualitative ligand-field theory
analysis of the electronic structure applied by Wheeler et al. to both the catecholate and
semiquinone complexes accounted for the optical properties of the semiquinone complex.
The ground state of the catecholate complex was found to be 4B2, which splits into 3A2
and 5A2 upon introduction of electron exchange with the semiquinone. This is illustrated

III

in Figure 1.4, along with the term states of the Cr - catechol complex in C2v symmetry

10

3,6-DTBQ 3,5-DTBQ
CI
Cl 0 0
Cl 0 0
Cl
TCQ PhenQ

Figure 1.3. Examples of commonly studied o-quinone ligands: 3,6-di-tert-butquuinone
(3,6-DTBQ), 3,5-di-tert-butquuinone (3,5-DTBQ), tetrachloro-l,2-quinone (TCQ),

9,1 O-phenanthrenequinone (PhenQ).

11

and their parent terms in Oh.24 Essentially, electron exchange interactions between
Cr(III)-SQ creates two new transitions (3A2 «— 3A2, z-polarized, and 3B] <— 3A2, y-
polarized ) from the ground state that were formally spin forbidden in the catecholate
complex}!3

Examination of the Tanabe-Sugano diagram22 of a d8 metal ion in octahedral
symmetry (Figure 1.5) reveals that the ground state of Ni(II) is 3A; with a low-lying lE
excited state. Given the fact that nickel(II)-semiquinone complexes are known to exhibit
strong ferromagnetic interactions’z’33 between the metal and semiquinone ligand, it
should be possible to produce spin allowed transitions analogous to the Cr(III)-
semiquinone complexes but from thermally accessed excited states.
In this thesis, the impact of electron exchange coupling on the physical properties of
nickel—semiquinone and chromium-semiquinone complexes is investigated. Specifically,
the idea of thermochromism that arises as a result of exchange interactions is probed.
The model complexes that will be used to study this feature are Ni(II)-semiquinone
complexes. In Chapter Two, the synthesis and physical properties of [Ni(tren)(3,5-
DTBSQ)](PF6) are presented. Variable-temperature electronic absorption spectroscopy
was used as a probe for the detection of thermochromism. As mentioned earlier, the
ground state of Ni" is 3A2 with a low-lying lE state. The transition from the ground state
to the IE excited state is formally spin forbidden but the presence of ferromagnetic
coupling between Nill and a semiquinone ligand should produce a spin allowed transition
analogous to the [Cr(tren)(3,6-DTBSQ)]2+ complex but from a thermally-accessed

excited state. This idea of thermochromism will be explored in Chapter 2.

l2

"
’-
’
_’
’
’

 

 

 

 

 

2Eg ://’/ —"' 1”‘2
\\\T‘\\ 2A1 ’’’’’’ T— 3B:
E TTTTTTTTT ____ IBl
X x
432 _____ 5A

4A,, —— —————————— —_=:::::::" '17 2

________ 3A2
C13+ (Oh, d3) Cr-Cat (C2,) Cr-SQ (sz)

Figure 1.4. Results of a ligand field analysis for a Cr(III)-quinone dyad applied by
Wheeler et al.“. The first column indicates two of the lowest-lying ligand field states
present in a (13 metal ion in Oh symmetry. The second column represents the splitting
upon lowering the symmetry from Oh to C2,., which corresponds to [Cr(tren)(3,6-
DTBCat)]+. The third column represents the splitting caused by the coupling between the
unpaired spins of the metal with an S = '/2 spin of b1 symmetry. This corresponds to the
complex, [Cr(tren)(3,6-DTBSQ)]2+. The relative ordering of the spin-coupled states for
the Cr-SQ diagram assumes antiferromagnetic coupling in the excited states. Other

higher-lying excited states that are present have been omitted for clarity.

l3

 

 

 

 

l
A ’72 31463;)
i '1
60« E '
I .Tz
S
‘r
so-« '
3 s 3
40d / T,(d‘d,,l
3
w #‘A'
30*
'G
204 '2
JP
‘0
10‘
3F 1 l ‘1 *2“?ng
1 2 3
Dq /8

Figure 1.5. Tanabe-Sugano diagram for a (18 metal complex in octahedral symmetry.22

14

In Chapter Three, the magnetic properties of Cr(3,5-DTBSQ)3 and Cr(PhenSQ)3
are described. This study was performed in search of a semiquinone ligand that couples
more weakly to a metal ion than 3,5-DTBSQ. The magnitude of exchange coupling for
Cr(3,5-DTBSQ)3 was determined and compared with the value known for the complex
Cr(PhenSQ)3. These data provide an indication of the relative strength of interaction of
the two semiquinones to the same metal center, in this case Cr(III). The relative
magnitudes of the coupling constants between Ni(II) and these semiquinone ligands
could then be inferred given similar geometric environments.

In Chapter Four, the results of an ab-initio density functional theory study
performed on phenanthrenequinone and phenanthrenesemiquinone are discussed. The
primary goals of this study were to explore the spatial distribution of the molecular
orbitals and the relative ordering of energy levels of the molecular orbitals upon
reduction from the quinone to the semiquinone redox state. In addition, the results found
in this study are compared to those found for 3,S-di-tert-butquuinone, providing insight
into the differences between the two quinones. Finally, future directions of this project

will be discussed briefly.

15

1.7 References and Notes

(1)

(2)

(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)

(14)

(15)

The coupling between the two unpaired spins can either be ferromagnetic or
antiferromagnetic which can result in significant differences in the physical
properties (e. g., magnetism) of the system.

Schugar, H. J .; Rossman, G. R.; Barraclogh, C. G.; Gray, H. B. J. Am. Chem. Soc.
1972, 94, 100.

Stenkamp, R. E.; Sieker, L. C.; Jensen, L. H. J. M01. Biol. 1978, 126, 457.
Sheriff, 8.; Hendrickson, W. A.; Smith, J. L. J. Mol. Biol. 1987, 197, 273.
Pierpont, C. G.; Attia, A. S. Collect. Czech. Chem. Commun. 2001, 66, 33.
Miller, J. S.; Epstein, A. J.; Reiff, W. M. Acc. Chem. Res. 1988, 21, 114.
Iwamura, H. Adv. Phys. Org. Chem. 1990, 26, 179.

Thoman, M.; Kahn, O.; Guilheim, J.; Varret, F. Inorg. Chem. 1994, 33

Hay, P. J.; Thibeault, J. C .; Hoffman, R. J. Am. Chem. Soc. 1975, 97, 4884.
Bhatttacharya, S.; Pierpont, C. G. Inorg. Chem. 1994, 33, 6038.

Lange, C. W.; Pierpont, C. G. Inorg. Chim. Acta. 1997, 263, 219.

Lange, C. W.; Conklin, B. J .; Pierpont, C. G. Inorg. Chem. 1994, 33, 1276.
Adams, D. M.; Rheingold, A. L.; Dei, A.; Hendrickson, D. N. Angew. Chem, Int.
Ed. Engl. 1994, 32, 391.

Schulz, D.; Weyhermuller, T.; Wieghardt, K.; Butzlaff, C.; Trautwein, A. X.
Inorg. Chim. Acta. 1996, 246, 387.

Tosik, A.; Maniukiewicz, W.; Bukowska-Strzyzewska, M.; Mrozinski, J.; Sigalas,

M. P.; Tsipis, C. A. Inorg. Chim. Acta. 1991, 190, 193.

16

(16)
(17)
(18)

(19)

(20)

(21)

(22)

(23)

(24)

(25)

(26)

(27)

(28)

Ginsberg, A. P. J. Am. Chem. Soc. 1980, 102, 1 11.

Kambe, K. J. Phys. Soc. Jpn. 1950, 48, 5.

Kahn, 0. Molecular Magnetism New York, NY, 1993.

Parameters for paramagnetic impurities and temperure-independent
paramagnetism (TIP) are included in the determination of J.

The effective magnetic moment of two non-interacting spins can be calculated
using the equation, Iten=[n.(n.+2)+n2(n2+2)]”2 (n, = number of unpaired
electrons).

Weldon, B. T.; Wheeler, D. E.; Kirby, J. P.; McCusker, J. K. Inorg. Chem. 2001,
40, 6802.

Tanabe, Y.; Sugano, S. J. Phys. Soc. Japan 1954, 753, 766.

Wheeler, D. E.; McCusker, J. K. Inorg. Chem. 1998, 37, 2296.

Benelli, C.; Dei, A.; Gatteschi, D.; Gudel, H.; Pardi, L. Inorg. Chem. 1989, 28,
3089.

Kessel, S. L.; Emberson, R. M.; Debrunner, P. G.; Hendrickson, D. N. Inorg.
Chem. 1980, 19, 1170.

Ruf, M.; Noll, B. C.; Groner, M. D.; Yee, G. T.; Pierpont, C. G. Inorg. Chem.
1997, 36, 4860.

Bruni, S.; Caneshi, A.; Cariati. F.; Delfs, C.; Dei, A.; Gatteschi, D. J. Am. Chem.
Soc. 1994, 116, 1388.

Buchanan, R. M.; Kessel, S. L.; Downs, H. H.; Pierpont, C. G.; Hendrickson, D.

N. J. Am. Chem. Soc. 1978, 100, 7894.

17

(29)

(30)

(31)

(32)

(33)

Attia, A. S.; Conklin, B. J.; Lange, C. W.; Pierpont, C. G. Inorg. Chem. 1996, 35,
1033.

Lynch, M. W.; Buchanan, R. M.; Pierpont, C. G.; Hendrickson, D. N. Inorg.
Chem. 1981, 20, 1038.

Further oxidation to the quinone redox state is not desirable because the quinone
binds very weakly to metal ions and would readily dissociate.

Benelli, C.; Dei, A.; Gatteschi, D.; Pardi, L. Inorg. Chem. 1988, 27, 2831.

Benelli, C.; Dei, A.; Gatteschi, D.; Pardi, L. Inorg. Chem. 1989, 28, I476.

18

CHAPTER TWO
Synthesis and Physical Properties of a

Nickel-Monosemiquinone Complex

2.1 Introduction

A large number of complexes containing a single radical semiquinone ligand
chelated to a paramagnetic metal ion have been studied. In general, they all exhibit strong
exchange interactions between the metal ion and semiquinone ligand. Previous studies
have investigated the effects of exchange coupling on the spectroscopic properties of

[II-catecholate systems."2 These systems represent excellent

Crm-semiquinone and Cr
models for the systematic study of the effect of electron exchange coupling on the
physical and photophysical properties of metal complexes. Benelli and co-workersl were
the first to report such a complex in 1989 in which the synthesis, magnetic, and electronic
absorption properties of [Cr(CTH)(3,5-DTBSQ)]2+ (CTH = 5,7,7,12,14,14-hexamethyl-
1,4,8,11-tetraazacyclodecane) were reported. Later, our group developed a related dyad
system using 3,6-DTBSQ and the tetradentate capping ligand, tris-(2-aminoethyl)amine
(tren).2 In addition, the non-exchange coupled catecholate analogue, [Cr(tren)(3,6-
DTBCat)]+, was synthesized for comparison.2 These compounds and their physical
properties were extensively characterized. Magnetic data revealed the presence of strong

antiferromagnetic coupling in [Cr(tren)(3,6-DTBSQ)]2+, however, the most pronounced

perturbation was observed in the absorption spectrum (Figure 2.1). The semiquinone

19

complex exhibited additional peaks in the absorption spectrum which were absent in the
non-exchange coupled catecholate complex. These peaks were qualitatively accounted
for using a simple group theoretical analysis. The ground state of the catecholate complex
is 4B2. Upon introduction of electron exchange in the semiquinone complex, this state
splits into 3A2 and 5A2 with the 3A2 term lying lowest in energy. This effectively creates
two new transitions (3A2 «— 3A2 and 3B. <— 3A2) from the ground state which were
formally spin forbidden in the catecholate complex.2

By analogy, an extension of this study can be made using Ni(lI). An examination
of the Tanabe-Sugano diagram of a (18 metal ion in octahedral symmetry, as seen in
Figure 1.4, reveals that the ground state of Ni(II) is 3A2 with a low-lying I E excited state.
Pseudooctahedral nickel(II) complexes have been found to couple ferromagnetically to
semiquinone ligands."’4 The difference in coupling between Cr-SQ and Ni-SQ can be
rationalized in terms of the molecular orbitals involved. MO calculations have shown that
the single unpaired electron in o-semiquinones resides in a rt—type molecular orbital."7 In
pseudooctahedral complexes, the metal eg orbitals are considered o-antibonding orbitals
while the metal tgg orbitals are of the n-antibonding type. It follows, then, that the
chromium(III)-SQ complexes would be expected to exhibit an antiferromagnetic
exchange interaction because one of the three unpaired electrons of the Cr(III) metal
resides in an orbital having the same symmetry as the magnetic orbital of the
semiquinone ligand.8 In contrast, the o-type magnetic orbitals of the Ni(lI) ion are
orthogonal to that of the semiquinone ligand, giving rise to ferromagnetic coupling in the
Ni(II)—SQ system. This was the result found by Benelli et al.3 for [Ni(CTH)(3,5-

DTBSQ)]+.

20

 

6000

5000

l

4000

l

3000 -

b)

Molar Absorpti» ity (M.l cmcl)

V

2000

 

1000

 

 

 

0 1 L !:=1¥ 1
30000 25000 20000 1 5000 10000

. -1
Energy (cm )

Figure 2.1. Room temperature absorption spectra2 of a) [Cr(tren)(3,6-DTBCat)](PF6) and

b) [Cr(tren)(3,6-DTBSQ)](PF6)2 measured in 4:1 EtOH/MeOH.

21

In terms of optical properties, spin-allowed transitions analogous to the Cr(III)-
semiquinone complexes should be possible as a result of exchange coupling in the Ni-SQ
complex. However, due to the ferromagnetic coupling, this transition will originate from
the higher lying term of the ground—state spin ladder. A schematic diagram is shown in
Figure 2.2. If the doublet state (2A2) formed by exchange interactions is thermally

accessible, the compound should be thermochromic.

2.2. Experimental Section

2.2.1. General

All reagents and materials were obtained from commercial sources and were used
as received except for the ligand tris(2-aminoethyl)amine (tren), which was vacuum
distilled prior to use. Solvents were purchased from either Aldrich Chemical Co. or
Strem. Elemental analysis was obtained through the Analytical Facilities of Michigan

State University.

2.2.2. Synthesis of Ni(tren)Br2

This compound was prepared by a method analogous to that reported previously
for Ni(tren)C12.9 Tren (0.350 g, 2.40 mmol) was dissolved in 95% ethanol (25 mL) and
placed into a 125 mL round bottom flask. NiBrz - tzO (0.500 g, 2.30 mmol) was added

to the tren solution and the resulting blue solution was stirred for one hour. The volume

22

 

 

 

Ix)

’

 

 

 

 

’-
3A‘) 382 ””””
““
~~~~

4A2

 

 

Ni2+ (Oh, d8) Ni2+ Ni2+
(CZV, no exchange) (C2,, w/ exchange)

Figure 2.2. Results of a ligand field analysis for a Ni(II)-quinone dyad. The first column
indicates two of the lowest-lying ligand field states present in a d8 metal ion in 0
symmetry. The second column represents the splitting upon lowering the symmetry from
O to C2,,, which corresponds to a non-exchange coupled Ni(II) complex. The third
column represents the splitting caused by the coupling between the unpaired spins of
Ni(II) with the semiquinone ligand (S = '/2) of b. symmetry. This corresponds to the
complex, [N i(tren)(3,5-DTBSQ)]+. Other higher-lying excited states that are present have

been omitted for clarity.

23

of the solution was reduced to ca. 5 mL by rotary evaporation and the resulting blue solid

was collected by filtration and washed with absolute ethanol (3 x 25 mL).

2.2.3. Synthesis of [Ni(tren)(3,5-DTBSQ)](PF6)

The following synthesis is a slight modification of the method reported by Benelli
et al.3 In an inert atmosphere box, 3,5-DTBQ (0.100 g, 0.450 mmol) was reduced to 3,5-
DTBCat by addition of excess Na metal to a solution of the quinone in THF (10 mL).
This solution was then filtered and cannulated over to a solution of Ni(tren)Br2 (0.165 g,
0.450 mmol) dissolved in degassed methanol (25 mL). The resulting solution was stirred
for l h at room temperature and then removed from the glovebox, whereupon an
immediate color change from light violet to dark brown was noted. The volume of the
solution was reduced to approximately 5 mL by rotary evaporation followed by addition
of a saturated aqueous solution of excess potassium hexafluorophosphate which induced
the precipitation of a black-green solid. The solid was collected and redissolved in
dichloromethane. The product was precipitated by addition of hexanes to this
dichloromethane solution and isolated by filtration. Yield: 0.187 g (72.8%). Anal. Calcd

for NiCzoOzN4H3gPF6: C, 42.13; H, 6.72; N, 9.83. Found: C, 42.15; H, 6.74; N, 9.40.

2.2.4. X-ray Structure Determination

A single-crystal x-ray structure determination of [Ni(tren)(3,5-DTBSQ)](PF6) was

carried out at the x-ray facility of the Department of Chemistry, Michigan State

24

University by Dr. Jeffrey Bodwin. A blue block crystal of [Ni(tren)(3,5-DTBSQ)](PF6)
was obtained by layering hexanes onto an ethyl acetate solution of the compound. One
crystal having approximate dimensions of 0.35 x 0.20 x 0.12 mm was mounted on a glass
fiber. Diffraction data were collected on a Siemens SMART diffractometer with graphite
monochromated Mo Ka X-ray radiation. Data were collected at 173 K. Details of crystal
and experimental data are given in Table 2.1, and positional parameters are listed in
Table 2.2. Cell parameters were obtained from 8996 independent reflections in the range
1.74° < 0 < 23.29°. Absorption corrections were performed using SAD.ABS. The
structural parameters were refined by using version 5.1 SHELXTL package.lo Disorder
present in the PF,‘ anions, ethyl acetate solvates, and carbons 15 and 16 within the
capping tren ligand, were modelled as two components in each case and the occupancies
independently refined. The final least-squares refinement, performed by introducing

hydrogen atoms in idealized positions. converged with R = 0.0543 (Rw = 0.1430).

2.2.5 Cyclic Voltammetry

Electrochemical measurements were carried out inside an Nz-filled glovebox
using a BAS CV-50W electrochemical analyzer. The compound was dissolved in
distilled and deaerated CH2C12 containing NBu4PF6 (ca. 0.1 M) as the supporting
electrolyte. A standard three-electrode setup was used with a platinum working electrode,
platinum wire counter electrode, and a Ag/AgNO3 electrode as reference. The

voltammogram was recorded at a scan rate of 100 mV/s. The reported potentials can be

25

Table 2.1 Crystal Data and Structure Refinement for [Ni(tren)(3,5-

 

DTBSQ)](PF6)*ACOEt
Empirical formula C2...H4(,F(,N4Ni,O4Pl
Formula weight 526.66
Temperature (K) 173(2)
Wavelength (A) 0.71073
Crystal system Triclinic
Space group P1
Unit cell dimensions a =1 1.954(2) A alpha =100.85(3)°
b =12.090(2) A beta =97.01(3)°
c =22.489(5) A gamma =98.65(3)°
Volume (A3) 3117.7(11)
Z 5
Density (calculated) (Mgm3) 1.403
Absorption coefficient (mm") 0.744
F(000) 1388
Crystal size (mm) 0.35 x 0.20 x 0.12
q range for data collection (deg) 1.74 to 23.29
Limiting indices -13<h<l3,-13<k<l3,-25<l<24
Reflections collected 27106
Independent reflections 8996 (R,m = 0.0900)
Completeness to q = 23.29° 99.90%
Refinement method Full-matrix least-squares on F2
Data / restraints / parameters 8996 / 0 / 832
Goodness-of-fit on F2 1.034
Final R indices [l>Zs(1)] R1 = 0.0543, wR2 = 0.1430
R indices (all data) R1 = 0.0680, W = 0.1529
Largest diff. peak and hole (eA'3) 0.892 and -0.546

 

26

Table 2.2. Atomic Coordinates [ x 104] and Equivalent Isotropic Displacement

Parameters [A2 x 103] for [Ni(tren)(3,5-DTBSQ)](PF6).

 

 

x 3” Z U (6(1)a
Ni(l) 854(1) 3789(1) -1797(1) 26(1)
0( 1) -222(2) 2844(2) -2586(1) 31(1)
0(2) 1647(2) 4422(2) -2438(1) 29(1)
C( 1) 140(3) 3096(3) -3068(2) 24(1)
C(2) 1 184(3) 3940(3) -2987(2) 25(1)
C(3) 1635(3) 4185(3) -3522(2) 26(1)
C(4) 1032(3) 3634(3) -4085(2) 29(1)
C(5) -3(3) 2817(3) «4172(2) 27(1)
C(6) -43 1(3) 2568(3) -3666(2) 27(1)
C(7) 2758(3) 5036(3) -3449(2) 31(1)
C( 8) 373 1(3) 4606(4) -3094(2) 40(1)
C(9) 2620(4) 6225(3) -3 106(2) 40(1 )
C(10) 31 16(4) 5173(4) -4067(2) 47(1)
C(1 1) -573(3) 2239(3) -4827(2) 30(1)
C(12) - 1609(4) 1319(4) -4834(2) 49(1)
C(13) ~977(4) 3125(4) -5163(2) 47(1)
C(14) 283(4) 1661(4) -5174(2) 43(1)
N(l) 94(2) 3108(2) -1 125(1) 27(1)
N(3) 1716(3) 2381(3) -1833(1) 34(1)
N(4) -452(3) 4796(3) -l742(1) 35(1)
N(2) 1925(17) 4807(16) -1063(9) 36(8)
C(15) 670(20) 3680(20) -472(15) 22(5)
C(16) 1371(9) 4872(9) -503(4) 38(3)
N(2') 1995(19) 4876(15) -1036(10) 30(7)
C(15') 730(20) 3780(30) -556(16) 37(9)
C(16') 1937(8) 4321(9) -534(4) 34(3)
C(17) 199(3) 1882(3) -1246(2) 39(1)
C(18) 1390(4) 1733(3) -1370(2) 41(1)

 

27

aU(eq) is defined as 1/3 of the trace of the orthogonalized Uij tensor.

Table 2.2. continued

 

 

x y 2 U (eq)

C(19) -1 121(3) 3267(3) 4221(2) 36(1)
C(20) -1 193(3) 4473(3) 4295(2) 39(1)
Ni(2) 3634(1) 2231(1) 1599(1) 26(1)
0(11) 4414(2) 1642(2) 2327(1) 31(1)
0(12) 2992(2) 3124(2) 2294(1) 32(1)
C(21) 41 13(3) 2090(3) 2839(2) 25(1)
C(22) 3315(3) 2898(3) 2815(2) 25(1)
C(23) 2934(3) 3393(3) 3378(2) 27( 1)
C(24) 3380(3) 3100(3) 3902(2) 30(1)
C(25) 4171(3) 2330(3) 3936(2) 27(1)
C(26) 4524(3) 1840(3) 3406(2) 27(1)
C(27) 2071(3) 4202(3) 3358(2) 34(1)
C(28) 960(3) 3590(4) 2932(2) 45(1)
C(29) 2586(4) 5256(3) 3126(2) 43(1)
C(30) 1767(4) 4624(4) 3994(2) 57(1)
C(31) 4584(3) 2085(3) 4565(2) 32( 1)
C(32) 5232(4) 3182(4) 4997(2) 59(1)
C(33) 5388(4) 1214(5) 4521(2) 63(2)
C(34) 3568(4) 1594(4) 4840(2) 54(1)
N(l 1) 4266(3) 1243(3) 895(1) 33(1)
N(12) 2334(3) 785(3) 1434(2) 41(1)
N(13) 5227(3) 3332(3) 1684(2) 40(1)
N(14) 2882(3) 2905(3) 893(2) 43(1)
C(35) 3868(4) 49(3) 946(2) 48(1)
C(36) 2592(4) -l48(4) 969(2) 58(1)
C(37) 5520(3) 1567(4) 1039(2) 45(1)
C(38) 5881(4) 2847(4) 1215(2) 48(1)
C(39) 3826(4) 1496(4) 298(2) 48(1)
C(40) 2791(6) 2021(6) 329(2) 87(2)
P(1) 9215(1) 11862(1) 819(1) 38(1)
F( 1) 8663(4) 12453(5) 324(2) 137(2)
F(2) 9722(3) 11 1 19(3) 309(2) 109(1)
1(3) 9776(3) 11293(3) 1317(2) 99(1)
F(4) 10361(2) 12793(2) 928(2) 82(1)
F(5) 8097(3) 10942(3) 717(2) 91(1)

 

28

Table 2.2. continued

 

 

x y 2 U (eq)
F(6) 8738(2) 12671(2) 1330(2) 73(1)
P(2) 4831(1) 6483(1) 1124(1) 78(1)
F(l 1) 4204(8) 5869(12) 1675(6) 120(4)
F( 12) 5642(17) 5830(30) 774(13) 243(13)
F(13) 3922(14) 7268(11) 1220(11) 165(11)
F(14) 5593(1 1) 6480(30) 1702(9) 170(9)
F(15) 5670(1 l) 7568(16) 1069(9) 121(6)
F(16) 4127(17) 5553(19) 722(7) 201(12)
F(l 1') 5565(11) 7501(18) 1627(17) 252(14)
F(12') 5734(9) 5748(13) 1317(11) 119(6)
F(13') 4059(13) 7010(20) 1475(7) 190(11)
F(14') 3933(13) 5359(11) 932(10) 151(8)
F(15') 5492(16) 6750(30) 667(12) 230(11)
F(16') 4168(8) 6679(1 1) 470(5) 137(5)
O(51) 2061(4) -563(4) 2357(2) 103(2)
O(52) 2843(3) -915(3) 3234(2) 64(1)
C(51) 1459(5) 233(5) 3279(3) 77(2)
C(52) 2136(5) -444(4) 2902(3) 62(1)
C(53) 3513(6) -1628(7) 2866(4) 104(2)
C(54) 4469(8) - 1 748(9) 3191(6) 276(10)
0(53) 7365(4) 614(3) 3580(2) 75(1)
C(55) 7703(5) -1285(5) 3572(3) 73(2)
C(56) 7745(4) -207(5) 3353(2) 57(1)
0(54) 8395(4) - 1 74(6) 2901(3) 33(2)
C(57) 8481(9) 809(8) 2634(4) 40(3)
C(58) 7432(9) 900(10) 2233(5) 44(3)
0(54') 7810(60) -420(30) 2772( 12) 189(16)
C(57') 7180(30) 470(30) 2379( 16) 124(13)
C(58') 8270(40) 1 130(30) 2550(30) 240(40)

 

29

converted from reference to Ag/AgNO3 to SCE by comparison to ferrocene measured

under identical conditions.H

2.2.6 Variable-Temperature Magnetic Susceptibility

Magnetic susceptibility data were collected using a Quantum Design MPMS
SQUID magnetometer interfaced to an IBM PC. The apparatus was calibrated by
measuring the magnetic susceptibility of [(CH3)2NHCHzCHzNH(CH3)2]CuC14l2 at
several temperatures between 5 and 380 K. Data were collected at two magnetic field

strengths of IT and 2.5T and corrected for diamagnetism using Pascal’s constants.l3

2.3 Results and Discussion

2.3.1. Synthesis

Previous studies of the Cr-SQ dyad [Cr(tren)(3,6-DTBSQ)](PF6)2 by our group
led us to the idea of studying a related Ni-SQ dyad. A thorough search of the literature
yielded only two examples of this motif for nickel(ll).3‘4 Benelli, et a1. 3‘4 reported the
synthesis and characterization of [Ni(CTH)(3,5-DTBQ)]Y (Y = C104, PF6, BPh4) and
[Ni(CTH)(TCSQ)](ClO4), where 3,5-DTBSQ is 3,5-di-tert-buty1orthosemiquinone,
TCSQ is 3,4,5,6-tetrachloroorthosemiquinone, and CTH is 5,7,7,12,14,14-hexamethyl-
1,4,8,1l-tetraazacyclodecane, a tetradentate macrocycle. The system involving the 3,5-

DTBSQ radical seemed to be a reasonable starting point for our research. The synthesis

30

of the title compound involves cannulation of a THF solution of 3,5-DTBCat to a
methanol solution of Ni(tren)Br2 followed by air oxidation. A black-green material was
obtained by reducing the volume of the solution to approximately 5 mL followed addition
of a saturated aqueous solution of KPF6. The solid obtained analyzed satisfactorily for

[Ni(tren)(3,5-DTBSQ)](PF(,).

2.3.2. Description of the X-ray Structure of lNi(tren)(3,5-DTBSQ)](PF6)

[Ni(tren)(3,5-DTBSQ)](PF6) crystallizes as an ethyl acetate solvate in the triclinic
space group PI. The unit cell is comprised of two crystallographically unique

[Ni(tren)(3,5-DTBSQ)] cations, two PF6' anions, and two ethyl acetate solvates. The
crystallographic details are given in Table 2.1, with positional parameters and bond
distances and angles given in Tables 2.2 and 2.3, respectively. Additional details of the
structure are found in the appendix. An ORTEP drawing of the cation is depicted in
Figure 2.3, which confirms the monomeric nature of [Ni(tren)(3,5-DTBSQ)](PF6). The
geometry of the compound is best described as a distorted octahedron with the nickel
coordinated by four aliphatic nitrogen atoms from the tren and two oxygen donors from
the o-semiquinone. The Ni-N bond distances are all different; Ni(1)-N(3) and Ni(1)-N(4)
at 2.113(3) and 2.120(3) A, respectively, are significantly longer than the Ni(1)-N(1) and
Ni(1)-N(2) bonds (2.091(3) and 2.032(19), respectively). The Ni-O bond distances,
Ni(1)-O(1) and Ni(1)-O(2) are 2.085(3) A and 2.022(2) A, respectively. These values

compare well with those reported by Benelli et. al. for [Ni(CTH)(3,5-DTBSQ)]+.3

31

Table 2.3. Selected Bond Lengths and Bond Angles for [Ni(tren)(3.5-DTBSQ)](PF(,)

 

Ni( 1 )-O(2)
Ni(1)—N(l )
Ni( 1 )-N(3)
0( I )-C( I)
C(1 )-C(2)
C(2)-C(3)
C(4)-C(5)

Ni(2)-O(l 1)
Ni(2)-N(l 1)
Ni(2)-N( 1 3)
0(1 1)-C(21)
C(21)-C(22)
C(22)-C(23)
C(24)-C(25)

O(2)-Ni(l)-N(2)
O(2)-Ni( 1 )-O( 1)
N(2)-Ni( 1 )-O( 1)
O(2)-Ni(1)-N(1)
N(2)-Ni(l)-N(1)
O(l)-Ni(l)-N(1)
O(2)-Ni(l)-N(3)
N(2)-Ni(l)-N(3)
O(1)—Ni(1)—N(3)
N(1)-Ni(l)-N(3)
O(2)-Ni( 1 )-N(4)
N(2)-Ni(1)-N(4)
O( 1 )-Ni(1)-N(4)
N(l)-Ni(l)-N(4)
N(3)-Ni(1)-N(4)

2.022(2)
2.091(3)
2.113(3)
1.286(4)
1.458(5)
1.438(5)
1.430(5)

2.079(2)
2.088(3)
2.1 19(3)
1.294(4)
1.468(5)
1.453(5)
1.429(5)

Bond Lengths (A)
Ni(l)—O(l)
Ni(l )-N(2)
Ni(1)-N(4)
O(2)-C(2)
C(1)-C(6)
C(3)-C(4)
C(5)—C(6)
Ni(2)—O(12)
Ni(2)-N(12)
Ni(2)-N(14)
0(12)-C(22)
C(21)-C(26)
C(23)-C(24)
C(25)-C(26)
Bond Angles (deg)
95.8(6) 0(12)-Ni(2)-O(1 1)
8049(9) 0(12)-Ni(2)-N(14)
175.7(6) 0(1 1)-Ni(2)-N(14)
177.64(11) O(12)-Ni(2)-N(1 1)
83.2(6) O(l l)-Ni(2)-N(1 1)
100.54(11) N(14)-Ni(2)-N(l 1)
94.62(1 1) 0(12)-Ni(2)-N(12)
95.6(6) O(11)-Ni(2)-N(12)
86.94(1 1) N(14)-Ni(2)-N(12)
83.33(12) N(l 1)-Ni(2)-N(12)
99.37(11) O(12)-Ni(2)-N(l3)
94.7(5) O(1 l)-Ni(2)-N(13)
83.77(11) N(14)-Ni(2)-N(l3)
8287(12) N(l l)-Ni(2)-N(13)
161.66(12) N(12)-Ni(2)-N(13)

2.085(3)
2.032( 19)
2.120(3)
1.279(4)
1.415(5)
1.372(5)
1.369(5)

2.025(2)
2.099(3)
2.086(3)
1.278(4)
1.412(5)
1.363(5)
1.371(5)

8052(9)

97.78(12)
177.18(12)
177.43(11)
98.00(11)
83.78(13)
95.07(12)
88.60(12)
93.80(14)
82.77(13)
98.82(12)
85.16(12)
9291(14)
83.1 1(13)
163.61(13)

 

32

 

          
   

’3’ lT111121 ¥7\
' S. 111 (”I
0(2) C181 ‘7 4
“111. H 01101
I, s
(7). ‘,c131
111111
.‘\ '
/5(.:193 13’ - @4141
wcnai (IE '49 '
01171 0111 - \
C161 (53 'b“ 01131
5;?
C1111!
~l
r’v
\\\\
C1129 c1141
:9

Figure 2.3. Drawing and labeling scheme of the two crystallographically unique
[Ni(tren)(3,5-DTBSQ)]+ cations determined by single-crystal x-ray diffraction. The

structural and crystallographic details are summarized in Tables 2.1 through 2.3.

33

Overall, however, the local symmetry around the Ni(II) can still be reasonably
approximated as sz.

Given the empirical formula, charge balance clearly requires an oxidation state of
—1 for this ligand. However, the oxidation state of the o-quinone bound to the metal is
also apparent by examining the bond lengths within the ring of the ligand.'4 In general, in
the catecholate form all the C-C bond distances are nearly identical due to the aromatic
nature of the ligand. In contrast, the semiquinone redox state exhibits alternating short
and long C-C bond distances because of the more localized nature of the double bonds
within the ring. Given this, the semiquinone redox state is clearly evident. For example,
C(3)-C(4) and C(5)-C(6) are found to be 1.372(5) A and 1.369(5) A, respectively,
whereas C(l)-C(2), C(2)-C(3), C(4)-C(5), and C(1)-C(6) are all on average 1.435 i 0.005
A. The C-O bond distances also support the semiquinone redox state, with C(1)-O(1) and
C(2)-0(2) distances of 1.286(4) and 1.279(4) A, respectively. The shorter C(2)-0(2)
bond suggests more double-bond character than the C(1)-O( 1) bond. However, these

differences are only marginal given the standard deviations quoted.

2.3.3. Electrochemistry

The cyclic voltammogram of a deaerated dichloromethane solution of
[Ni(tren)(3,5-DTBSQ)](PF6), depicted in Figure 2.4, reveals two redox couples at -0.739
and +0.104 V (vs Ag/AgNO3). The two redox processes can be reasonably attributed to
the SQ'l/Catz' and QO/SQ'I couples based on several considerations. First, the oxidation

and reduction processes involving the nicke1(II) ion are expected to occur at significantly

34

1.596

 

1.086

.01
ii,’
, .44

Current, uA
O
O

-5.0e-7

V I Y

-1.0e-6

' I

 

 

 

_1&6 IxtlnLLIlllllll144L1111111111111111111111
400 200 0 20) 400 600 6(1) 411])

mV vs. Ag/AgNO3

Figure 2.4. Cyclic voltammogram of [Ni(tren)(3,5-DTBSQ)](PF6) in deaerated

dichloromethane.

35

more positive or more negative potentials, respectively.”l6 Second, the cyclic
voltammogram of 3,5-DTBSQl7 in acetonitrile exhibits a redox couple corresponding to
3,5-DTBQ —> 3,5-DTBSQ at —0.955 V vs Ag/AgNOg. Finally, these potentials are
similar to those observed for the analogous CrIll complex. The one-electron wave at -
0.739 V is reversible with a peak separation of 66 mV, whereas the one-electron wave at
0.104 V is at best quasi-reversible. This is not surprising due to the fact that the fully
oxidized quinone is expected to be a weakly binding ligand. Formation of this species
following oxidation of the semiquinone complex will likely result in ligand dissociation,
hence an irreversible wave in the cyclic voltammogram. This hypothesis is supported by
the fact that the oxidation wave becomes more reversible at faster scan rates. The overall
redox chemistry of [Ni(tren)(3,5-DTBSQ)](PF6) is summarized in Figure 2.5. A shoulder
seen in the second wave is attributed to the reduction of free quinone which is liberated

from the complex upon oxidation of [Ni(tren)(3,5-DTBSQ)]+ to [Ni(tren)(3,5-DTBQ)]2+.
2.3.4. Magnetic Susceptibility

Electron exchange within the ground state configuration of this class of molecules
can be quantified by variable-temperature magnetic susceptibility measurements.
Electron exchange in low-symmetry complexes can be described using a simple

Heisenberg exchange Hamiltonian of the form in Equation 2.1,

H = -2 :1 2.1 Jij SVSJ' (2.1)

36

 

 

 

 

[Ni(tren)(3,5-DTBSQ)]+ A ‘ [Ni(tren)(3,5-DTBCatH
- e-
El/z = - 0.739 V
+ e'
[Ni(tren)(3,5-DTBQ)]2+ ‘ ‘ [Ni(tren)(3,5-DTBSQ)?
- e-
El/z = + 0.104 V

Figure 2.5. The redox chemistry of [Ni(tren)(3,5-DTBSQ)]+.

37

where S, and Sj represent the quantum mechanical spin operators on the ith and j‘h (e. g.,
the Ni" ion and the semiquinone radical) paramagnetic site respectively. Jij is the
exchange coupling constant, a scalar quantity which gauges the magnitude of the
coupling between the spin centers. Defining the total spin of the system, ST, as ST = S. +

S; and using the Kambe approximationl8 the operator-equivalent expression given in

Equation 2.2 can be derived,

H = -J [ST2 - 5.2 — 822] (2.2)

and the data analyzed as described in Chapter 1.

Variable temperature data were collected at magnetic field strengths of l T and
2.5 T. A Plot of the effective magnetic moment versus temperature in the range from 5 -
380 K for solid samples of [Ni(tren)(3,5-DTBSQ)](PF6) is given in Figure 2.6 (data at 2.5
T). In addition, the data were corrected for temperature-independent paramagnetisml9 and
shown in these figures.

At 2.5 T and 380 K, the effective magnetic moment is 3.57 113 and approaches
3.83 1.13 at 100 K, which is a value close to the S = 3/2 moment (um = 3.87 1113) expected
for a ferromagnetically coupled nickel(lI)-SQ dyad. The magnetic moment then
decreases to a value of 3.67 113 at 10 K and continues to decline. The decline in effective
moment at room temperature from the S = 3/2 moment is due to the population of the
doublet excited state, 2A2 and suggests that the exchange constant, J, is on the order of a

couple of hundred wavenumbers.

38

 

4.0
3.9 ~
3.8 ~ A
0‘ I ' A ‘ A
I A
3.7 3
3.6 ~

315-

Effective Magnetic Moment

314+

 

 

 

3.3 I I I
0 100 200 300 400

Temperature (K)

Figure 2.6. Plot of effective magnetic moment versus temperature for [Ni(tren)(3,5-
DTBSQ)](PF(,) at 2.5 T with (squares) and without (triangles) the correction for

temperature-independent paramagnetism (TIP).

39

The drop in um below 100 K can be explained by considering the phenomenon of
zero-field splitting. A diagram which illustrates the effect of zero-field splitting of the S =
3/2 state of [Ni(tren)(3,5-DTBSQ)]+ is presented in Figure 2.7. Based on symmetry and
the degeneracy the point group (sz) can support, the fourfold degeneracy of the S = 3/2
state, (m3 = i 1/2, i 3/2), is partially lifted with the m5 = :t 3/2 and i 1/2 levels separated
by an amount 2D. The zero-field parameter, D, not only has magnitude but also sign. As
drawn, D > 0; if < 0, then the :1: 3/2 levels would be lower in energy. From Figure 2.5
and 2.6, we can infer that D is positive in the case of [Ni(tren)(3,5-DTBSQ)](PF6)
because the effective magnetic moment drops significantly below the value expected for
S = 3/2.

Due to the presence of zero-field splitting, the Hamiltonian presented in Equation
2.1 must be adjusted to include this effect. In the case of [Ni(tren)(3,5-DTBSQ)](PF6),

the Hamiltonian becomes

H = ~4st — s3 — 822] + D[SZ2 — 1/3sT2] (2.2)

where D is the zero-field splitting parameter and 82 is the ms value of the sublevel. The
magnetic data indicates that the zero-field splitting parameter is quite large due to the
appearance of zero-field splitting effects at approximately 80 K. This result suggests that
D is comparable enough in magnitude to J so as to make questionable the use of the Van
Vleck equation to fit the data. A full matrix approach to analyzing these data is currently

being pursued.

40

sT = 1/2

 

 

[—] 1nS = :1: 3/2

 

V
"‘2 I 2D
S... = 3/2 [=1 ms = i “2
. all .11
N1(II) N1 -SQ N1 -SQ
D = 0 D 75 0

Figure 2.7. Schematic drawing of the effect of zero field splitting on the S = 3/2 ground
state of [Ni(tren)(3,5-DTBSQ)]+. As drawn, D > 0; if D < 0, then the t 3/2 levels would

be lower in energy.

41

2.3.5. Electronic Absorption Spectra

The absorption spectra of non-exchange coupled six coordinate nickel(II)
complexes are known to exhibit three spin allowed transitions to the 3T2, 3T. and 3T.
levels. These occur in the range 7000 - 13,000, 11,000 - 20,000, and 19,000 - 27,000 cm"
regions, respectively. In addition, a spin forbidden band to lEg is also observed at
approximately 15,000 cm'l (~ 670 nm). As seen in Figure 2.2, the transition from low-
lying doublet state to the next excited state in the exchange coupled Ni(II)-SQ complex is
expected to be a few hundred cm"l different in energy than 'Eg <— 3A2. Therefore, we
would expect the transition from the low-lying doublet excited state to the next doublet
excited state (2B1 <— 2A2) to occur around the same region in the case of the exchange-
coupled Ni(II)-semiquinone complex.

The room-temperature electronic absorption spectrum of a dichloromethane
solution of [Ni(tren)(3,5-DTBSQ)](PF(,) (Figure 2.8) shows bands at 43500 (e = 8200
M"om"), 31700 (e = 9100 M'cm"), and 22700 om‘I (e = 1000 M'om") with a notable
shoulder at 28200 cm". No transitions assignable to d-d states are observed.20 This
spectrum is similar to that reported by Benelli et al. for [Ni(CTH)(3,5-DTBSQ)]+.3 The
similarity of this spectrum to that reported for [Zn(S,S-CTH)(3,5-DTBSQ)](PF(,)4 and the
free 3,5-DTBSQ21 ligand indicates that these bands are likely due to internal ligand
transitions involving the 3,5-DTBSQ ligand.

The magnetic data obtained for [Ni(tren)(3,5-DTBSQ)](PF(,) shows evidence for
temperature dependence in the magnetic moment. This led us to pursue variable-

temperature absorption studies in hope of observing a thermochromic response. The

42

 

 

 

10000

/\ 8000 1
“a
_0

E; 6000-!
.2»
22
e
8

.2 4000 -
1<
E
O
2

2000 4

0 I I f 1 I I j

 

45000 40000 35000 30000 25000 20000 15000

Energy (cm")

Figure 2.8. Room-temperature absorption spectrum of [Ni(tren)(3,5-DTBSQ)](PF6)

measured in dichloromethane.

43

variable-temperature absorption spectra were all measured in dimethylformamide. An
initial absorption spectrum was obtained at room temperature (~ 298 K) followed by
acquisition of multiple absorption spectra at higher temperatures. The selected spectra are
depicted below in Figure 2.9. Due to the limitations of the apparatus, we were unable to
reach temperatures greater than 370 K. At the temperatures that we were able to access,
the spectra showed no clear evidence of the expected absorption at ~ 670 nm.

Our inability to observe the expected thermochromic behavior can be attributed to
the following. The transitions 2A2 <— 2A2 and 2B. 1— 2A; are expected to have extinction
coefficients of approximately 102 M"lcm'l (based on the fact that this transition is a d-d
based transition which is exchange enhanced). As seen in Figure 2.9, the extinction
coefficient of the band in the region is roughly 400 M'lcm‘l. Therefore, the expected band
(2A2 <— 2A2 and 2B1 1— 2A2) may be obscured by this ligand-based absorption feature.
This problem is likely compounded by the fact that the coupling between the Ni(II) and
3,5-DTBSQ is on the order of 150 cm". At 320 K, this will likely lead to a ~11 %
population of the 2A2 state, further attenuating the observed intensity of the expected

absorption.

2.4 Concluding Comments

This study documents the synthetic, structural, electronic, and optical properties

of the pseudooctahedral [Ni(tren)(3,5-DTBSQ)](PF6) complex. Our goal was to utilize

44

 

 

 

 

 

 

 

 

1000
800

'E

_Q

g 600

2:

IE

8-

8 400

.9

<

ta

'3

E 200
0.0 H7 1 1 I l I

 

24000 22000 20000 18000 16000 14000

Energy (cm")

Figure 2.9. Variable-temperature electronic absorption spectra of [Ni(tren)(3,5-

DTBSQ)](PF6) in dimethylformamide solution.

45

this metal-semiquinone dyad to produce a thermochromic compound. Magnetic
susceptibility data confirm the expected presence of a strong ferromagnetic exchange
interaction between the metal and semiquinone radical anion. The data also exhibited
temperature dependence in the effective magnetic moment, however, variable-
temperature absorption spectroscopy failed to show clear evidence of the expected
transition from the lowest-lying doublet excited state. It was suggested that the exchange
interaction might be too large, resulting in too great an energy separation between the
ground state and the lowest-lying doublet excited state from which the new transition(s)
is expected. One possibility to circumvent this is to employ a different semiquinone
ligand which will couple more weakly to the Ni(II) center. This notion is explored in the

next two chapters.

46

2.5 References and Notes

(1)

(2)
(3)
(4)
(5)

(6)
(7)
(8)

(9)
(10)
(11)

(12)

(13)

(14)

(15)

(16)

Benelli, C.; Dei, A.; Gatteschi, D.; Gudel, H. U.; Pardi, L. Inorg. Chem. 1989, 28,
3089.

Wheeler, D. E.; McCusker, J. K. Inorg. Chem. 1998, 37, 2296.

Benelli, C.; Dei, A.; Gatteschi, D.; Pardi, L. Inorg. Chem. 1988, 27, 2831.

Benelli, C.; Dei, A.; Gatteschi, D.; Pardi, L. Inorg. Chem. 1989, 28, 1476.
Wheeler, D. E.; Rodriguez, J. H.; McCusker, J. K. J. Phys. Chem. A 1999, 103,
4101.

Pilar, J. J. Phys. Chem. 1970, 74, 4029.

Kuwata, K.; Shimizu, Y. Bull. Chem. Soc. Jpn. 1969,42, 864. '

Rodriguez, J. H.; Wheeler, D. E.; McCusker, J. K. J. Am. Chem. Soc. 1998, 120,
12051.

Perkins, C. M. In;; University of Washington, 1980.

XPREP; 5.03 ed.; Siemens Industrial Automation, Inc.: Madison, 1995.

Barrett, W. C.; Johnson, H. W.; Sawyer, D. T. Anal. Chem. 1984, 56, 1890.
Brown, D. 8.; Crawford, V. H.; Hall, J. W.; Hatfield, W. E. J. Phys. Chem. 1977,
81, 1303.

Boudreaux, E. A.; Mulay, L. N. Theory and Applications of Molecular
Paramagnetism; John Wiley and Sons: New York, 1976.

Pierpont, C. G.; Lange, C. W. Prog. Inorg. Chem. 1994, 41 . 331.

Nag, K.; Chakravorty, A. Coord. Chem. Rev. 1980, 33, 87.

Olson, D. C.; Vasilevskij, J. Inorg. Chem. 1969, 8, 161 l.

47

(17)

(18)

(19)

(20)

(21)

Lehman, M. W.; Evans, D. H. J. Electroanal. Chem. 2001, 500, 12.

Kambe, K. J. Phys. Soc. Jpn. 1950, 48, 5.

The temperature-independent paramagnetism (TIP) value for this complex was
assumed to be 250 x 106 cm3/mol

The d-d transition ('E <— 3A2) is known to occur at ~650 nm for non-exchange

coupled octahedral complexes.

Stallings, M. D.; Morrison, M. M.; Sawyer, D. T. Inorg. Chem. 1981, 20, 2655.

48

CHAPTER THREE

Synthesis and Magnetic Properties of Cr(3,5-DTBSQ)3 and Cr(PhenSQ)3

3.1 Introduction

The synthesis and characterization of the complex, [Ni(tren)(3,5-DTBSQ)](PF6)
was explored in the previous chapter. The function of that complex was to study
thermochromism, which is expected to arise due to exchange coupling between the Ni(II)
and semiquinone ligand. The results obtained from variable-temperature absorption
studies of the complex, [Ni(tren)(3,5-DTBSQ)](PF6), failed to show clear evidence of
thermochromic behavior. It was suggested that the energy gap (AB = 3.1) between the
ground state and the low-lying excited state might be too large to produce enough thermal
population of the doublet excited state. In order to decrease this gap, a new quinone must
be chosen which couples more weakly to the Ni(II).

A large number of complexes containing multiple radical semiquinone ligands
chelated to paramagnetic metal ions have been studied. Specifically, the preparations of
Cr(3,5-DTBSQ)31'8 and Cr(PhenSQ)33’8'” (PhenSQ = phenanthrenesemiquinone) have
been reported along with partial characterization of their magnetic properties. These
studies indicate the presence of strong antiferromagnetic coupling between the metal and
radical semiquinone ligands in both molecules. In this chapter, the magnetic properties of
Cr(3,5-DTBSQ)3 and Cr(PhenSQ)3 will be examined in more detail. Specifically, the

magnetic susceptibility data will be acquired and fit to the Van Vleck equation in order to

49

extract the coupling constant between Cr-SQ. Values obtained for J will give us insight
into the possible difference in coupling between PhenSQ and 3,5-DTBSQ given a
common environment (i.e., coupling to Cr(III)) and provide an indication as to whether

PhenSQ might be a more suitable ligand to achieve thermochromism in a Ni(II) system.

3.2 Experimental Section

3.2.1. General

All reagents and materials were obtained from commercial sources and were used
as received. Solvents were purchased from either Aldrich Chemical Co. or Strem.
Elemental analyses were obtained through the Analytical Facilities of Michigan State

University.

3.2.2 Synthesis of Cr(3,5-DTBSQ)3

Preparative procedures for this complex have previously been reported.3’l2 The
dark violet crystals were washed with absolute ethanol (3 x 25 mL) and recrystallized
from 1:1 toluene/ethanol. Anal. Calcd. CrC42H(,OO6: C, 70.76; H, 8.48; N, 0.00. Found: C,

71.36; H, 8.62; N, 0.14.

50

3.2.3 Synthesis of Cr(PhenSQ)3

The synthesis of this compound has been previously reported.3 ’9‘10‘12 The dark
violet crystals that precipitated upon cooling were collected and washed with toluene (3 x
25 mL) followed by ethanol (3 x 25 mL). Anal. Calcd. CrC42H24O6: C, 74.55; H, 3.58; N,

0.00. Found: C, 75.50; H, 3.57; N, 0.10.

3.2.4 Variable-Temperature Magnetic Susceptibility

Magnetic susceptibility data were collected as described for [Ni(tren)(3,5-
DTBSQ)](PF6) in Chapter 2. Data were collected at a magnetic field strength of 2.5 T and
corrected for diamagnetism using Pascal’s constants.'3 The data were fit to Equations 1.7

using a program of local origin to obtain a value for the exchange coupling constant, J.

3.3 Results and Discussion

3.3.1 Magnetism of Cr(3,5-DTBSQ)3

Variable-temperature magnetic susceptibility data for Cr(3,5-DTBSQ)3 is

illustrated in Figure 3.1 as a plot of the effective magnetic moment versus temperature.

At 380 K, the effective magnetic moment is 1.13 113 and decreases to a value of 0.17 113 at

5 K. This is significantly lower than the spin-only value for a system comprised of non-

51

 

Effective Magnetic Moment

 

 

 

 

00 T .l 1 1 1 1 1 1 1 1 l 1 A 1 1 l 1 I l 1
0 100 200 300 400

Temperature (K)

Figure 3.1. Plot of effective magnetic moment versus temperature for Cr(3,5-DTBSQ)3
(circles). The solid line represents the fit to a theoretical model; g = 2.00, TIP = 150 x 10’

6 cm3/mol, and J = -440 cm".

52

exchange coupled Cr(III) and three radical semiquinones, implying that the coupling
between the metal and the ligands is antiferromagnetic in nature.
The exchange interactions within this molecule can be described by a simple

Heisenberg exchange Hamiltonian shown below in Equation 3.1,

H = -2 X; 25.1” Si'Sj (3.1)

where the terms are those previously described in Chapters 1 and 2.

In the case of Cr(3,5-DTBSQ)3, or Cr(SQ)3 in general, there are two exchange
interactions present: direct exchange between the metal and each semiquinone ligand, as
well as superexchange between the semiquinone ligands. The complete Hamiltonian

accounting for all pairwise interactions can be written as

H = -2J [SCr'SSQI 1’ SCr'SSQZ ‘1" SCr'SSQ31

~2J’ [S sol'Ssoz ‘1' S soz'Ss03 1’ S SQI'SSQ3] (32)

where due to the D3 symmetry of the molecule, JCr-SQ| = JCpSQZ = JCr-SQ3 = J and JSQ]-SQZ
= JSQ2-SQ3 = JSQ1_SQ3 = J’. The magnitude of direct exchange is expected to be much larger
than that of superexchange.l4 For example, the exchange constant corresponding to
superexchange between the 3,5-DTBSQ ligands in the compound Ga(3,5-DTBSQ)3 was

' typical of metal-

deterrnined to be 7.8 cm'1 compared to several hundred cm'
semiquinone coupling.15 Therefore, as first approximation, the only exchange parameter

that will be considered is the Cr-SQ interaction. The spin Hamiltonian now becomes

53

H = -2J [SCr'SSQI + SCr'SSQZ + SCr'Ssosl (33)

Prior to employing the Kambe approximation, we define a general spin operator, SA

(Equation 3.4).

SA = 8801 + Ssoz + SSQ3 (3.4)

Squaring Equation 3.4 affords Equation 3.5.

3A2 = Ssol2 + Ssoz2 + Ssos2 + 25 sol'Ssoz + ZS soz'Ss03 + ZS SQI'SSQS (3-5)

In addition, the product 80's,, is given in Equation 3.6.

SCr'SA = SCr'Ssol + SCr'Ssoz + SCr'SSQ3 (3-6)

Therefore, by applying the Kambe approximation[6 and using the fact, ST = Scr + SSQ] +

8502 + 8503 (or ST = 80 + S A), the square of total spin operator can be written as
5T2 = SCr2 + SAZ + 25Cr'SA (3-7)

or simply

sees. = 1/21812 - so} - 5,2) (3.8)

54

Substituting Equation 3.8 into Equation 3.6 and 3.3 affords Equation 3.9.

H = 41(st — sf — s02] (3.9)

From this equation, the eigenvalue expression can be generated, Equation 3.10, and the

eigenvalues can be obtained.

E(S) = -J [ST(ST+1)‘SA(SA+1 )-Scl»(Sc1+1)1 (3-10)

The resulting energy levels for Cr(3,5-DTBSQ)3, or any analogous Cr(SQ)3 complex, are
depicted in Figure 3.2. The magnetic properties of Cr(SQ)3 as a function of temperature
reflect a Boltzman distribution over the thermally accessible spin states shown in Figure
3.2. The general expression that describes magnetic susceptibility is given by Equation
1.7. The variable temperature magnetic susceptibility data were fit to this expression
using the eigenvalues of Equation 3.10. The solid line in Figure 3.1 corresponds to the
best fit with parameters of g = 2.00 and J = -440 cm"; temperature independent

paramagnetism (TIP) was fixed at 150 x 10'6 cm3/mol.

3.3.2 Magnetism of Cr(PhenSQ); and Comparison with Cr(3,5-DTBSQ)3

The variable-temperature magnetic susceptibility data obtained by other workers

for Cr(PhenSQ)3 displayed residual paramagnetism at higher temperaturesg‘9 At 286 K,

Cr(PhenSQ)3 was found to have a magnetic moment value of 1.15 1.13, which decreases to

55

S=3(E=-9/2J)

[= S=2(E=-3/2J)

S=3/2

E S=2(E=3/2J)

=] S=1(E=5/2J)

S=1(E=11/2J)

 

S=0(E=15/2J)

Cr (111) Cr(SQ)3

Figure 3.2 Energy levels for Cr(SQ)3 complexes resulting from intramolecular
antiferromagnetic exchange between the unpaired electrons of the o-semiquinone ligands

and the Cr(III) ion.

56

0.30 113 at 4.2K.8 This is consistent with the formulation of an antiferromagnetically
coupled Cr(PhenSQ)3 system. The exchange parameter for this compound was
determined by Lynch et al. to be —3 50 cm".ll

The variable-temperature magnetic susceptibility data we obtained for this
complex are summarized in Figure 3.3 as a plot of the effective magnetic moment versus
temperature in the range 5 — 380 K. An attempt to fit the magnetic susceptibility data to
Equations 1.7 was hindered by the inability to accurately compensate for background
contributions. We can, however, infer that the coupling between Cr(III)-PhenSQ is
weaker than Cr(III)-3,5-DTBSQ from the data due to the fact that the effective moment
of the former complex is greater than that of the latter complex (Figure 3.3).

The coupling constant obtained for Cr(3,5-DTBSQ)3 in this study clearly
indicates that 3,5-DTBSQ couples more strongly than PhenSQ to Cr(III). Although the
strength of exchange interactions are affected by numerous factors, including orbital
energetics, it is nonetheless reasonable to expect that the coupling between PhenSQ and a
given metal ion should be smaller than that observed between 3,5-DTBSQ and that same
metal, given the same geometric environment. The phenanthrenesemiquinone version of
[Ni(tren)(3,5-DTBSQ)]+ is therefore expected to exhibit a smaller coupling constant and a

greater likelihood of exhibiting thermochromic behavior.

57

215

 

 

 

 

*, 2111
113 o
5
2 . o '
£2 ‘15 ‘ 1.
... o
o o
c 0
g o"
a, 1.0~ o
o“ .
e l .
UJ A A
A A ‘
AA“‘
0.0 I I I
0 100 200 300 400

Temperature (K)

Figure 3.3 Plots of effective magnetic moment versus temperature for Cr(PhenSQ)3

(circles) and Cr(3,5-DTBSQ)3 (triangles).

58

3.4 References and Notes

(1)

(2)

(3)

(4)

(5)

(6)

(7)
(8)

(9)

(10)
(11)

(12)
(13)

(14)

Abakumov, G. A. Izv. Akad. Nauk, Ser. Khim. 1977, 7, 1642.

Danek, M.; Vicek, A., Jr. Proc. Conf. Coord. Chem. 1989

Downs, H. H.; Buchanan, R. M.; Pierpont, C. G. Inorg. Chem. 1979, 18, 1736.
Glushakova, V. N.; Skorodumova, N. A.; Mitin, A. V.; Nevodchikov, V. 1.;
Abukumova, L. G.; Lovanov, A. V. Izv. Akad. Nauk, Ser. Khim. 1994, 2, 327.
Lovanov, A. V.; Abakumov, G. A.; Razuvaev, G. A. Dokl. Akad. Nauk SSR 1977,
235, 824.

Sofen, S. R.; Ware, D. C.; Cooper, S. R.; Raymond, K. N. Inorg. Chem. 1979, 18,
234.

Vclek, A., Jr. Inorg. Chem. 1986, 25, 522.

Buchanan, R. M.; Downs, H. H.; Shorthill, W. B.; Pierpont, C. G.; Kessel, S. L.;
Hendrickson, D. N. J. Am. Chem. Soc. 1978, 100, 7894.

Buchanan, R. M.; Downs, H. H.; Shorthill, W. B.; Pierpont, C. G.; Kessel, S. L.;
Hendrickson, D. N. J. Am. Chem. Soc. 1978, 100, 4318.

Vlckova, B.; Stepanek, J. Proc. Conf Coord. Chem. 1987, 11, 463.

Lynch, M. W.; Buchanan, R. M.; Pierpont, C. G.; Hendrickson, D. N. Inorg.
Chem. 1981, 20, 1038.

Brown, D. G.; Johnson, L. Z. Naturforsch 1979, 34b, 712.

Boudreaux, E. A.; Mulay, L. N. Theory and Applications of Molecular
Paramagnetism; John Wiley and Sons: New York, 1976.

Pierpont, C. G.; Attia, A. S. Collect. Czech. Chem. Commun. 2001, 66, 33.

59

(15) Adams, D. M.; Rheingold, A. L.; Dei, A.; Hendrickson, D. N. Angew. Chem, Int.
Ed. Engl. 1994, 32, 391.

(16) Kambe, K. J. J. Phys. Soc. Jpn. 1950, 48, 5.

60

CHAPTER FOUR
Density Functional Theory Analysis of the Electronic Structures of

Phenanthrenequinone and Phenanthrenesemiquinone

4.1 Introduction

Quinones are commonly exploited for their innate redox prOperties.l Upon
reduction to the semiquinone redox state, the diamagnetic quinone (S = 0) becomes
paramagnetic (S = 1/2). Further reduction yields the diamagnetic (S = 0) catecholate
form. All three redox states are known to be stable; however, the quinone redox state is
not as useful for coordination chemistry due to its poor binding affinity to a metal center.
Metal-semiquinone complexes have been utilized as models by many inorganic chemists
in the study of simple intramolecular exchange coupling. For example, several complexes
of metals ranging from chromium, copper, zinc, to nickel among others have been
studied.2'l3 Common semiquinone ligands include tetrachlorobenzosemiquinone (TCSQ),
3,5-di-tert-butylsemiquinone (3,5-DTBSQ), 3,6-di-tert-butylsemiquinone (3,6-DTBSQ),
and phenanthrenesemiquinone (PhenSQ). Our interest in these systems stems from the
opportunity they afford for effectively turning on (SQ) or off (Cat) exchange coupling by
converting from one redox state to the other.2

Theoretical studies have also been performed on quinones and metal-quinone

complexesl‘m6 Fenske-Hall calculations, which is a nonempirical molecular orbital

61

method, have been reported on the free quinone ligand for inclusion in a study of the
bonding present in metal-quinone complexes.l4 Bianchini, et al.”, performed extended
Hfickel and fragment molecular orbital calculations on the unsubstituted quinone for a
study involving Co-quinone complexes. This study presented qualitative results that were
used to describe magnetic properties as well as a mechanism for electron transfer in these
compounds. In addition, density functional theory (DFT) has been applied to the free
semiquinone ligand, which proved useful in the analysis of the electronic structure
variations across the redox series of the 3,6-di-tert-butylbenzoquinone and 3,5-di-tert—
butylbenzoquinone.18 The results found in this study detailed the molecular orbital
compositions, overall relative ordering of the molecular orbitals, localization of the NPA
charge densities, and localization of the unpaired spin densities. The calculations carried
out by Rodriguez, et al. detailed the mechanism of exchange between chromium(III) and
a single semiquinone ligand.16

The results of the previous chapter suggest that phenanthrenesemiquinone might
afford a smaller exchange constant when bound to Ni(II) than does 3,5-DTBSQ. By
analogy to previous work from our lab,””18 this chapter describes a computational study
of both phenanthrenequinone and phenanthrenesemiquinone in order to gain insight into
their electronic structures. The primary goals of this study will be to explore the spatial
distribution of the molecular orbitals and the relative ordering of energy levels of the
molecular orbitals upon reduction from the quinone to the semiquinone redox state. The

catechol redox state will not be considered.

62

4.2 Computational Methods

4.2.1 General Methods

The self-consistent field (SCF) density functional calculations were performed
using the Gaussian 98 program. '9 The two different hybrid functionals used were the
BLYP functional and the B3LYP functional. The BLYP functional combines the
gradient-corrected Becke (B) exchange with the nonlocal correlation functional of Lee,
Yang, and Parr (LYP).20 The hybrid B3LYP functional combines the three parameter

2"23 with the LYP correlation functional. All calculations were done

exchange of Becke
using tight convergence criteria.24 The atomic charge analyses were performed using the

natural population analysis (N PA) framework developed by Weinhold et al.25'27

4.2.2. Geometry Optimizations

The optimization was carried out in three steps. First, the geometry was optimized
using restricted Hartree-Fock (RHF) methods for phenanthrenequinone (PhenQ) and
restricted open shell Hartree-Fock (ROHF) for phenanthrenesemiquinone (PhenSQ). An
STO-3G** basis set was used in each case. No geometrical constraints were imposed on
the molecule. The starting geometry used for the first optimization was obtained from an
X-ray structure that has been reported for phenanthrenequinone.28 There were several
forms of the crystal reported28 but the exact form chosen for this calculation was the 01-

form of the PhenQ crystal. The labeling scheme and coordinate system used for both

63

molecules is shown in Figure 4.1. After the first optimization, a second optimization was
performed at the BLYP/3-21G** level using restricted density functional theory (RDFT)
and restricted open-shell density functional theory (RODF T) for the quinone and
semiquinone forms, respectively. The final optimization was performed at the B3LYP/6-
310* level using either RDF T or RODFT. The optimized geometry obtained at this level

was then used in subsequent single point calculations.

4.2.3. Single Point Calculation

The calculations carried out for the singlet PhenQ incorporated the B3LYP
functional and a 6-311G** basis set. The molecular charge was designated to be zero.
The calculations performed for the doublet PhenSQ involved a similar functional and
basis set, however, the molecular charge in this case was assumed to be —1. Separate
geometry optimizations and single-point calculations were carried out on PhenQ
assuming singlet and triplet configurations. The optimized structure obtained for the
singlet configuration produced a lower total SCF energy relative to the triplet, confirming
that the ground state of phenanthrenequinone is a singlet. Similar calculations were
performed for the semiquinone, assuming doublet and quartet configurations: the total
SCF energy of the doublet was significantly lower than the quartet. The total SCF
energies obtained for both the quinone and the semiquinone molecules are given in Table

4.1.

64

/C9—C1O y
\
C14—C8 Cl—Cz
/ \ \ ,
C13 C7—C6 Cs
\ \
C12—C11 C5‘—'C4

Figure 4.1. The labeling scheme and coordinate system used for phenanthrenequinone
and phenanthrenesemiquinone. The peripheral carbons are defined as C2, C3, C4, C5,

C11, C12, C13, and C14.

65

Table 4.1. Total Self-Consistent Field Energies (Hartree) Obtained at the B3LYP/6-

311G** Level Using the Optimized Geometries of PhenQ and PhenSQ

 

 

Molecule 23 + 1 Energy
Phenanthrenequinone l -688.924912
3 -688.856575
Phenanthrenesemiquinone 2 -688.987181
4 -688.86521 l

 

66

4.3 Results and Discussion

4.3.1. Optimized Geometries of PhenQ and PhenSQ

The redox states of phenanthrenequinone are depicted in Figure 4.2. The
structural parameters obtained from the optimization of PhenQ and PhenSQ are listed in
Table 4.2. The bond lengths obtained for PhenQ are reasonably similar to those reported
for the X-ray crystal structure.28 The trends observed in the optimized geometry of PhenQ
are similar to those found for other quinones. The longest bond in the molecule is
between C9-C10, which is 1.545 A. The bonds between Cl-C6 and C7-C8 exhibit
lengths of 1.416 A which are the shortest bond lengths within the ring. These results
reflect the bonding characteristics expected for quinones where the shortest bonds in the
optimized structure are designated in the literature29 as those having the most double
bond character within the ring. The bonds between C9-Ol and C10-O2 exhibit distances
of 1.219 A, typical ofa C=O bond.

Reduction of the quinone to the semiquinone redox state is accompanied by
appreciable geometric changes. In general, one observes a shortening of the single bonds
and elongation of the double bonds of the quinone upon addition of an electron. For
example, the bond between C9-C10 shortens by approximately 0.06 A and the bonds
between Cl-C6 and C7-C8 lengthen by approximately 0.01 A. Another noteworthy
change is the lengthening of the carbon-oxygen bonds by approximately 0.04 A. This is
reasonable considering that a one-electron reduction eliminates one C-O double bond to

yield a double bond which is delocalized over two carbon-oxygen pairs. These structural

67

 

O O
o o-
Phenanthreneq uinone Phenanthrenesemiquinone Phenanthrenecathecholate
(PhenQ) (PhenSQ) (PhenCat)

Figure 4.2. The redox states of phenanthrenequinone.

68

Table 4.2. Selected Bond Lengths (A) for Phenanthrenequinone (PhenQ) and
Phenanthrenesemiquinone (PhenSQ) Obtained from Experiment and Self-Consistent

Field (SCF) Geometry Optimizations at the B3LYP/6-31G* Level

 

 

Bond PhenQ (exp )3 PhenQ (calc) PhenSQ (calc)
C(1 )-C(2) 1.39 1.402 1.413
C(2)-C(3) 1.34 1.389 1.383
C(3)-C(4) 1.38 1.398 1.408
C(4)-C(5) 1.38 1.393 1.387
C(5)-C(6) 1.39 1.405 1.413
C(1)-C(6) 1.41 1.416 1.427
C(6)-C(7) l .48 1 .485 1.462
C(7)-C(8) 1.41 1.416 1.427
C(8)-C(9) 1.47 1.484 1.477

C(9)-C(10) 1.51 1.545 1.481
C(1)-C(10) 1.47 1.484 1.477
C(7)-C(l 1) 1.39 1.405 1.413
C(11)-C(12) 1.38 1.393 1.387
C(12)-C(13) 1.38 1.398 1.408
C(13)-C(14) 1.34 1.389 1.383
C(8)-C(14) 1.39 1.402 1.413

C(9)-0(1) 1.25 1.219 1.255
C(10)-O(2) 1.25 1.219 1.255

 

a Matsuzaki, S.Y., Gotoh, M., Kuboyama, A. Mol. Cryst. Liq. Cryst, 1987, 142, I27

69

changes are also in accord with qualitative expectations based on the character of the

molecular orbitals involved (vide infra).

4.3.2. Molecular Orbitals of Phenanthrenequinone (PhenQ)

The energies and atomic contributions of the highest occupied and lowest
unoccupied molecular orbitals (MOS) 52-56 are summarized in Table 4.3. The 01 and B
orbitals (1 e' orbitals of differing spin) of PhenQ were found to be identical in energy and
composition, as expected for this closed-shell singlet. The highest molecular orbital
(HOMO) and lowest unoccupied molecular orbital (LUMO) of phenanthrenequinone are
MO 5401,13 and MO 5501,13 respectively.

The occupied molecular orbitals 5201,13 are mostly comprised of orbital
contribution from C1, C8, and the carbons of the peripheral rings (C2, C4, C5, C11, C12,
C14). They contribute approximately 11 % each to these molecular orbitals. Figure 4.3
shows the spatial representation of MO 5201. From this figure, it is apparent that the
oxygen atoms are relatively isolated from each other, exhibiting only very weak 7: anti-
bonding interactions. It is also apparent that much of the interaction occurs between the
carbons within the peripheral rings. For example, it bonding interactions is observed in
the following carbon pairs, C1-C2, C4-C5, C1 1-C12, and C8-C14 (Figure 4.3).

The next highest occupied molecular orbitals, 5301,11, are composed mostly of p,
and py orbitals on the oxygens (ca. 32 % each). As shown in Table 4.3, additional
contributions (of less than ca. 10%) are provided by C1, C8, C9, and C10. The

contributions from the carbons of the peripheral rings are negligible in this case. The

70

Table 4.3. Energies and Percent Atomic Contributions to the Frontier Molecular Orbitals

of Phenanthrenequinone (PhenQ) Obtained at the U-B3LYP/6-31 10’” Level

 

 

Orbital" 5201,13 5301,13 5411,13 5501,13 5611,13
Typeb O O O V V
Energy (eV) -7.515 -6.871 -6.853 -3.253 -1.679
%Cl 11.12 8.41 7.26 3.31 1.17
% C2 11.31 1.78 3.03 4.04 10.12
% C3 0.02 0.19 15.99 1.36 13.12
% C4 11.22 0.07 1.63 5.77 0.24
% C5 10.83 0.45 8.36 0.14 15.11
% C6 0.00 0.64 l 1.97 4.73 8.68
% C7 0.00 0.73 11.97 4.73 8.68
%C8 11.12 8.37 7.26 3.31 1.17
% C9 0.01 5.85 0.03 12.04 1.04
% C10 0.01 5.71 0.03 12.04 1.04
% C11 10.83 0.39 8.36 0.14 15.11
% C12 11.22 0.10 1.63 5.77 0.24
% C13 0.02 0.24 15.99 1.36 13.12
% C14 11.31 1.68 3.03 4.04 10.12
% O] 5.49 32.19 1.73 18.61 0.53
% 02 5.49 32.11 1.73 18.61 0.53

 

a The a and [3 orbitals for this molecule were found to be identical energetically and in

composition. b V = virtual (unoccupied); O = occupied

71

 

Figure 4.3. MO 5211 of phenanthrenequinone. The orbital density is localized above and

below the plane of the ring.

 

Figure 4.4. MO 530 of phenanthrenequinone. The orbital lies in the plane of the ring.

72

principal interaction, as depicted in Figure 4.4, is the rt—antibonding interaction occurring
between the oxygen and carbon orbitals.

The HOMOs for PhenQ are MOS 5401,13, which are depicted in Figure 4.5. They
are comprised mainly of contributions from the carbons (C3, C13, C5, C11) of the
peripheral rings. Also significant are the contributions from C6 and C7 (ca. 12 %). The
oxygens, 01 and 02, contribute negligibly (1.73 %) to these molecular orbitals; C9 and
C10 contribute less than 1 %. Table 4.3 lists the other contributions to these molecular
orbitals. Interestingly, this result differs from what was found previously'8 for the
HOMOS of 3,6-di-t'ert-butquuinone (3,6-DTBQ). Wheeler et al. found that the HOMOs
of 3,6-DTBQ were mainly composed of contributions from the oxygen atoms (18.5 %
each) while the contributions from the carbons of the ring were less significant.

The quinone LUMOs (5501,13, Figure 4.6) are composed mostly of pZ orbitals on
the oxygen atoms, contributing 18.61 % each. The next major contributors are C9 and
C10, (ca. 12 %). As seen in Figure 4.6, n-antibonding interactions are present between
the oxygen (01 and 02) and carbon (C9 and C10) atoms. There are also weak n-bonding
interactions between C1, C8, C9, and C10.

The last molecular orbitals that will be considered for phenanthrenequinone are
the degenerate MOS 5601,13. Main contributions come from the pZ orbitals of C5 and C11
(15.11 %). Carbons 3 and 13 each contribute 13.12 % whereas C2 and C14 contribute
10.12 %. In addition, C6 and C7 contribute ca. 9 % to these molecular orbitals. The other
carbons in the ring and peripheral rings contribute less than 1 %. Overall, the main
interactions in these molecular orbitals are thus strong n-antibonding interactions between

the carbons in the peripheral rings. This is clearly seen in Figure 4.7.

73

 

 

Figure 45. MO 5411 (HOMO) of phenanthrenequinone. The orbital density lies above

and below the plane of the ring.

 

Figure 4.6. MO 5501 (LUMO) of phenanthrenequinone. The orbital density lies above

and below the plane ofthe ring.

74

.—

 

Figure 4.7. MO 5601 of phenanthrenequinone. The orbital density lies above and below

the plane of the ring.

75

4.3.3. Molecular Orbitals of Phenanthrenesemiquinone (PhenSQ)

The next molecule examined here is PhenSQ. Table 4.4 lists the energies and
composition of orbitals 52-56. For the purpose of discussion, only the similarities and
differences of the orbitals of this molecule with those of phenanthrenequinone will be
discussed rather than the details of every orbital.

Upon reduction, the quinone undergoes significant changes in geometry. These
changes are in accord with the bonding and antibonding interactions present in the lowest
unoccupied molecular orbital (LUMO) of the PhenQ which becomes occupied upon
reduction. The principal interaction in the LUMO between C1-C6 and C7-C8, as depicted
in Figure 4.6, is n-antibonding which contributes to the lengthening of the bond upon
reduction. In contrast, the principle interaction between C9-C10 is at-bonding which
results in shortening of the bond upon reduction.

Molecular orbital 5501 (HOMO) of phenanthrenesemiquinone bears a remarkable
resemblence to molecular orbital 55 (LUMO) of phenanthrenequinone. Despite the fact
that MO 55 is unoccupied in PhenQ and occupied in PhenSQ, its spatial distribution
remains virtually unchanged. Therefore, the Aufbau principle applies in this system,
where an electron placed in the LUMO of PhenQ represents, to a good approximation,
the correct valence electronic structure of the one-electron reduced species. However, it is
interesting to note that several of the lower energy orbital compositions of PhenQ and
PhenSQ are quite different. This is in stark contrast to what was found for the ligands 3,5-
DTBQ and 3,6-DTBQ,'8 the spatial distributions of most of the higher energy occupied

molecular orbitals remained virtually identical upon reduction of the quinone to

76

Table 4.4. Energies and Percent Atomic Contributions to the Frontier Molecular Orbitals

of Phenanthrenesemiquinone Obtained at the U-B3LYP/6-311G** Level

 

 

Orbital 52 01 52 13 53 01 53 13 54 01 54 13 55 01 55 13 56 01 56 13
Type“ O O O O O O O V V V
Energy -2.884 -2.697 -2.705 -2.455 -2.033 -1.859 -0.216 1.542 2.305 2.386
(eV)

% C] 2.41 2.25 10.95 12.06 8.06 7.76 3.43 8.02 53.58 0.20
% C2 0.58 1.00 0.65 0.09 1.12 1.62 3.51 1.99 15.27 13.33
% C3 0.07 0.17 14.06 10.14 0.06 0.08 1.48 3.62 0.04 10.49
% C4 0.04 0.08 2.83 4.87 0.08 0.09 5.40 6.81 0.08 1.21
% C5 0.31 0.28 5.93 2.92 0.33 0.38 0.19 0.05 0.04 15.46
% C6 0.78 1.18 11.70 10.11 0.43 0.65 4.88 6.06 15.35 7.17
% C7 0.78 1.18 11.70 10.11 0.43 0.62 4.88 6.06 0.06 7.17
% C8 2.41 2.25 10.95 12.06 8.06 7.72 3.43 8.02 0.02 0.20
% C9 4.12 4.04 0.02 0.72 4.69 4.54 11.60 6.08 0.26 1.15
% C10 4.12 4.04 0.02 0.72 4.69 4.59 11.60 6.08 14.33 1.15
% C1 1 0.31 0.28 5.93 2.92 0.33 0.39 0.19 0.05 0.06 15.46
% C12 0.04 0.08 2.83 4.87 0.08 0.09 5.40 6.81 0.01 1.21
% C13 0.07 0.17 14.06 10.14 0.06 0.08 1.48 3.62 0.00 10.49
% C14 0.58 1.00 0.65 0.09 1.12 1.64 3.51 1.99 0.00 13.33
% 01 41.45 40.54 3.85 9.08 34.76 34.17 19.52 17.38 0.08 0.99
% 02 41.45 40.54 3.85 9.08 34.76 34.18 19.52 17.38 0.60 0.99

 

a O = Occupied; V = Virtual (unoccupied)

77

 

 

Figure 4.8. MO 5511 (HOMO) of phennntL The orbital density lies

above and below the plane of the ring.

 

Figure 4.9. MO 5513 (LUMO) of phennnt‘ The orbital density lies

above and below the plane of the ring.

78

semiquinone. For example, a comparison of Table 4.3 and Table 4.4 reveals that the
orbital contributions to MO 52 from 01 and 02 increases from 5.49 % to approximately
41 % upon reduction of PhenQ to PhenSQ. Changes in percent contributions are also
observed in other molecular orbitals. Since many of the physical properties of the
quinone systems are linked to the HOMO, it is not clear how these changes might impact

the behavior of the phenanthrenequinone (e.g., reactivity)

4.3.4. Energy Spacing of the Molecular Orbitals of PhenQ and PhenSQ

The absolute and relative orbital energies of PhenQ and PhenSQ are listed in
Table 4.5. Since a direct comparison between the absolute energies of the orbitals of the
two redox states is not possible due to the difference in the total electron count between
the two molecules, only the relative energies of the orbitals will be examined.

In PhenQ the energy spacing between orbitals 5201 and 5301 is 0.644 eV. A much
smaller energy difference of 0.018 eV is found between molecular orbitals 5301 and 5401.
The energy difference between MO 5401 and MO 5501 is 3.600 eV (HOMO-LUMO gap)
which is larger than that found for 3,6-DTBQ of 3.230 eV for the HOMO-LUMO gap.l8

Addition of an electron to PhenQ results in significant changes in the relative
energy difference between the various molecular orbitals. Orbital 5201 is now just 0.179
eV below MO 5301. In contrast, the energy spacing between MO 5401 and MO 5501
increases by ca. 0.66 eV. An increase is also observed in the energy spacing of MOS 5501

and 5601. Molecular orbitals 5401 and 5501, which correspond to the HOMO-LUMO gap in

79

Table 4.5. Absolute and Relative Orbital Energies (eV) of Phenanthrenequinone and

Phenanthrenesemiquinone Calculated at the U-B3LYP/6-3l 16““ Level

 

 

Orbital PhenQ PhenSQ
5211 -7.515 -2.884
5213 -7.515 -2.697
5311 -6.871 -2.705
5313 -6.871 -2.455
5411 -6.853 -2.033
5413 -6.853 -1.859
5511 -3.253 -0.216
5513 -3.253 1.542
5611 -1.679 2.305
5613 -1.679 2.386

5311 - 5211 0.644 0.179
5411 - 5311 0.019 0.672
5511 - 5411 3.600 (LUMO — HOMO) 1.817
5513 - 5511 0.000 1.758 (LUMO — HOMO)
5611 - 5511 1.574 2.521

 

80

PhenQ, are separated by only 1.817 eV in PhenSQ. The HOMO-LUMO (5513 - 5501) gap

in PhenSQ is only 1.817 eV which is substantially less than that observed in PhenQ.

4.3.5. Natural Population Analysis (N PA) Charge Densities and Spin Densities

The last characteristics to be considered are the charge and spin distributions of
the ligand and how they change upon reduction. NPA charge densities of PhenQ and
PhenSQ are tabulated in Table 4.6. Due to the difference in electronegativity between
oxygen and carbon, we expected most of the negative charge density to reside on the
oxygen atoms. This was precisely the result found from the analysis. For example, the
negative charge density of PhenQ at O] and 02 is —0.454 whereas the carbon atoms
(except for C9 and C10) exhibit charge density values of less than —0.09. Positive charge
density is localized on the carbons atoms (C9 and C10) which are bonded to the oxygen
atoms. This is expected due to the polarity of the C-0 bond. The absolute charge
densities of these carbon atoms are ca. +0.414. Upon reduction, the oxygen atoms
experience an increase of approximately 0.10 in the magnitude of negative charge
density. Most of the carbon atoms experience an increase in net negative charge (albeit
less than the oxygen atoms), with the exception of C1 and C8 which exhibit a slight
decrease in negative charge density. A significant decrease in positive charge density is
observed for C9 and C10, although they maintain a net positive charge density of +
0.270. These trends are in accord with what was found for reduction of 3,5-DTBQ and
3,6-DTBQ to the semiquinone redox states.'8 However, the charge density found for the

oxygen atoms in phenanthrenesemiquinone is approximately 0.1 less negative than

81

 

Table

4.6. NPA Atomic

Charge

Densities

of Phenanthrenequinone

Phenanthrenesemiquinone Calculated at the U-B3LYP/6-311G** Level

and

 

 

Atom PhenQ PhenSQ
CI -0.085 -0.065
C2 -0.020 -0.054
C3 -0.050 -0.089
C4 -0.039 -0.088
C5 -0.052 -0.073
C6 0.010 -0.018
C7 -0.003 -0.017
C8 ~0.078 -0.066
C9 0.413 0.270

C10 0.41 5 0.269
C1 1 -0.057 -0.073
C12 -0.031 -0.088
C13 -0.052 -0.090
C14 -0.016 -0.054
01 -0.454 -0.559
02 -0.453 -0.559

 

82

either 3,5-DTBSQ or 3,6-DTBSQ. This indicates smaller net negative charge densities at
the oxygen atoms of PhenSQ. It is interesting to speculate what this result may imply
about reactivity, specifically in terms of a decrease in nucleophilicity realtive to 3,5-
DTBSQ.

Information on the spin densities of PhenSQ are listed in Table 4.7. The principle
results found in the analysis of PhenSQ is that most of the spin density is located on O]
and 02. This result is in accord with what was found for 3,6-DTBQ and 3,6-DTBSQ.18
The Mulliken spin density values for 01 and 02 are 0.236 each. It appears as though the
carbon atoms attached to the oxygen atoms have more spin density than the other carbon
atoms; i.e., C9 and C10 have values of + 0.109. The finding that the majority of the spin
density of PhenSQ resides on the oxygen atoms indicates that the mechanism for

exchange in metal-phenanthrenesemiquinone complexes is likely to be direct exchange.

4.4. Concluding Remarks

Nonlocal gradient-correlated density functional theory was applied to the study of
PhenQ and PhenSQ. The optimized geometry of PhenQ at the U-B3LYP/6-3IG* level
yielded a structure that is similar to the x-ray structure reported in the literature. Upon
reduction of PhenQ to PhenSQ, single bonds contracted while double bonds elongated in
accord with expectations from simple resonance pictures of the 71 system. For example,
the C-0 bond distances increase from a length characteristic of a double bond to one that
is intermediate between a double bond and a single bond. This is in accord with what is

- - 29
expected for semrqumones.

83

Table 4.7. NPA Atomic Spin Densities and Mulliken Net Spin Densities for

Phenanthrenequinone

311G** Level

and Phenanthrenesemiquinone Obtained at the U-B3LYP/6-

 

 

Atom Mulliken Spin Densities NPA Atomic Spin Densities
C1 0.031 0.035
C2 0.051 0.042
C3 0.008 0.010
C4 0.083 0.076
C5 -0.014 -0.015
C6 0.051 0.051
C7 0.052 0.052
C8 0.031 0.034
C9 0.109 0.105

C10 0.108 0.105

C11 -0.015 -0.016

C12 0.084 0.076

C13 0.007 0.009

C14 0.051 0.042
01 0.236 0.238
O2 0.236 0.239

 

84

The electronic structures of PhenQ and PhenSQ were calculated at the U-
B3LYP/6-31 10““ level using the optimized structures described above. The only
similarities found between the molecular orbitals of PhenQ and PhenSQ is the HOMO of
phenanthrenesemiquinone (MO 5511) and M05 5511,13 of phenanthrenequinone (LUMO).
Hence, the Aufbau principle is applicable in the sense that the LUMO of PhenQ and the
HOMO of PhenSQ are nearly identical in composition. The lower energy orbital
compositions of PhenSQ differed substantially from PhenQ. This result is in contrast with
what was found for the ligands, 3,5-DTBQ and 3,6-DTBQ.18 However, it is not clear if
these differences are chemically significant because many of the physical properties of
the quinone system are linked to the HOMO and not the lower energy orbitals.

Lastly, the natural population analysis (NPA) charge densities reveal that the
majority of negative charge density resides on the oxygen atoms while the positive
charge density resides on C9 and C10. This result signifies the polarity of the C-0 bond
and indicates the importance of electronegativity differences in the localization of net
charge within the molecule. Overall, the negative charge density of the atoms increase
upon reduction, with the oxygen atoms showing the largest increase. NPA atomic spin
densities indicate that the unpaired spin density in PhenSQ is largely localized on the
oxygen atoms. The remaining carbon atoms contain smaller non-zero values of spin
density, with C9 and C10 having the largest values of + 0.109. This implies that the
mechanism of exchange between phenanthrenesemiquinone and a metal ion is expected
to be through direct exchange due to the localization of the majority of the spin density
on the oxygen atoms which will be binding directly to the metal ion. It can be inferred

that a Ni(II)-PhenSQ complex would be expected to exhibit ferromagnetic coupling due

85

to the fact that the o-type magnetic orbitals of the Ni(II) ion are orthogonal to that of the
semiquinone ligand. These implications can be further studied in future experiments on

Ni(II)—SQ systems employing similar calculations.

4.5 Future Directions

A similar reaction (Reaction 2) outlined in Chapter 2 was applied to the synthesis
of the [Ni(tren)(PhenSQ)]+. However, the electronic spectrum of the material obtained
matched that of PhenQ. This implies that either PhenSQ is not binding or that the
product, [Ni(tren)(PhenSQ)]+, is unstable and PhenSQ dissociates and/or oxidizes to the
quinone redox state upon exposure to air. A problem encountered in this synthesis was
the reduction of phenanthrenequinone to phenanthrenecatechol. It appears as though the
reduction is not complete using either sodium metal or potassium metal. Due to this, the
likely cause of the failed reaction is the binding of the semiquinone to the metal ion. A
possible method to circumvent this is to oxidatively bind the quinone to a Ni complex
(d'O) prior to any other steps. Another possible route that can be taken involves, first,
making Ni(PMe3)2(PhenSQ) by the direct oxidative addition reaction between Ni(PMe3)4
and PhenQ.”3 ' The next step would just involve the reaction of Ni(PMe3)2(PhenSQ) with
tren, which should displace the trimethylphosphine ligands. The latter reaction is
entropically favored and should proceed to yield [Ni(tren)(PhenSQ)]. This can then be
oxidized to give [Ni(tren)(PhenSQ)](PFb) by stoichiometric addition of AgPF6.

Although it is not possible to formulate a direct comparison between the absolute

energies of the orbitals of phenanthrenequinone, 3,5-DTBQ, and 3.6-DTBQ, the

86

qualitative differences in the localization of the molecular orbitals are noted above along
with differences in the charge density localization. These dissimilarities are noteworthy
and could be chemically significant. For example, we found that there was less negative
charge density localization on the oxygen atoms of PhenSQ than either 3,5-DTBSQ or
3,6-DTBSQ. This result explains and supports our finding that the

phenanthrenesemiquinone ligand exhibits weaker binding affinity to metal centers.

87

4.5. References

(1)
(2)
(3)
(4)
(5)
(6)

(7)
(8)

(9)

(10)
(11)
(12)

(13)

(14)
(15)
(16)

The Chemistry of the Quinonoid Compounds; Wiley: New York, 1974.

Wheeler, D. E.; Mccusker, J. K. Inorg. Chem. 1998, 37, 2296.

Benelli, C.; Dei, A.; Gatteschi, D.; Pardi, L. Inorg. Chem. 1990, 29, 3409.

Lange, C. W.; Conklin, B. J .; Pierpont, C. G. Inorg. Chem. 1994, 33, 1276.
Benelli, C.; Dei, A.; Gatteschi, D.; Pardi, L. Inorg. Chem. 1988, 2 7, 2831.

Benelli, C.; Dei, A.; Gatteschi, D.; Gudel, H. U.; Pardi, L. Inorg. Chem. 1989, 28,
3089.

Buchanan, R. M.; Claflin, J.; Pierpont, C. G. Inorg. Chem. 1983, 22, 2552.

Kessel, S. L.; Emberson, R. M.; Debrunner, P. G.; Hendrickson, D. N. Inorg.
Chem. 1980, 19, 1170.

Attia, A. S.; Pierpont, C. G. Inorg. Chem. 1995, 34, 1172.

Attia, A. S.; Pierpont, C. G. Inorg. Chem. 1997, 36, 6184.

Attia, A. S.; Bhattacharya, S.; Pierpont, C. G. Inorg. Chem. 1995, 34, 4427.

Kahn, O.; Prins, R.; Redijk, J.; Thompson, J. Inorg. Chem. 1987, 26, 3557.

Ruf, M.; Noll, B. C.; Groner, M. D.; Yee, G. T.; Pierpont, C. G. Inorg. Chem.
1997, 36, 4860.

Gordon, D. J .; Fenske, R. F. Inorg. Chem. 1982, 21, 2907.

Gordon, D. J.; Fenske, R. F. Inorg. Chem. 1982, 21, 2916.

Rodriguez, J. H.; Wheeler, D. E.; Mccusker, J. K. J. Am. Chem. Soc. 1998, 120,

12051.

88

(l7) Bianchini, C.; Masi, D.; Mealli, C.; Meli, A.; Martini, G.; Laschi, F.; Zanello, P.
Inorg. Chem. 1987, 26, 3683.

(18) Wheeler, D. E.; Rodriguez, J. H.; Mccusker, J. K. J. Phys. Chem. 1999, 103,
4101.

(19) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.; Stratmann, R. E.;
Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain,
M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.;
Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.;
Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J .; Ortiz, J. V.; Stefanov, B. B.; Lin, G.; Liashenko,
A.; Piskorz, P.; Komaromi, 1.; Gomperts, R.; Martin, R. L.; Fox, D. J .; Keith, T.;
Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.;
Gill, P. M. W.; Johnson, 8.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.;
Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.4;
Gaussian, Inc.: Pittsburgh, PA, 1998.

(20) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.

(21) Becke, A. D. Phys. Rev. A 1988, 38, 3098.

(22) Becke, A. D. J. Chem. Phys. 1993, 98, 1372.

(23) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.

(24) F risch, M. J .; Frisch, A. Gaussian 98 User's Reference; Gaussian, Inc.: Pittsburgh,
1998.

(25) Reed, A. E.; Wienhold, R. J. J. Chem. Phys. 1983, 78. 4066.

89

(26)

(27)

(28)

(29)

(30)

(31)

Reed, A. E.; Weinstock, R. B.; Wienhold, R. J. J. Chem. Phys. 1985, 83, 735.
Glendening, E. D.; Reed, A. E.; Carpenter, J. E.; Weinhold, F. NBO 3.1,
Theoretical Chemistry Institute, University of Wisconsin: Madison, WI, 1996.
Matsuzaki, S. Y.; Gotoh, M.; Kuboyama, A. Mol. Cryst. Liq. Crst. 1987, I42,
127.

Pierpont, C. G.; Attia, A. S. Collect. Czech. Chem. Commun. 2001, 66, 33.

Klein, H. F.; Karsch, H. H. Chem. Ber. 1976, 109, 2515.

Klein, H. F.; Auer, E.; Dal, A.; Lemke, U.; Lemke, M.; Jung, T.; Rohr, C.; F lorke,

U.; Haupt, H. Inorg. Chim. Acta 1999, 287, 167.

90

Appendix
Tables of Bond Lengths, Bond Angles, and Anisotropic Thermal Factors

for the Crystal Structure of [Ni(tren)(3,5-DTBSQ)](PF6)

Presented in Chapter 2

91

Table A.l. Bond Distances (A) for [Ni(tren)(3,5-DTBSQ)](PF6)

 

 

Bond Bond Distance (A) Bond Bond Distance (A)
Ni(l)-O(2) 2.022(2) N(1)-C(15) 1.53(3)
Ni(1)-N(2) 2.032(19) N(3)-C(18) 1.471(5)
Ni(1)-O(1) 2.085(3) N(4)-C(20) 1.488(5)
Ni(1)-N(l) 2.091(3) N(2)-C(16) 149(2)
Ni(l)-N(3) 2.113(3) C(15)-C(l6) 157(3)
Ni(l)-N(4) 2.120(3) N(2')-C(16) 142(2)
Ni(1)-N(2') 2.136(17) C(15')-C(l6') 1.48(3)
O(l)-C(1) 1.286(4) C(17)-C(18) 1.513(6)
C(2)-C(2) 1.279(4) C(19)-C(20) 1.51 1(5)
C(1)-C(6) 1.415(5) Ni(2)-O(12) 2.025(2)
C(l)-C(2) 1.458(5) Ni(2)-O(l 1) 2.079(2)
C(2)-C(3) 1.438(5) Ni(2)-N(14) 2.086(3)
C(3)-C(4) 1.372(5) Ni(2)-N(l l) 2.088(3)
C(3)-C(7) 1.535(5) Ni(2)-N(12) 2.099(3)
C(4)-C(5) 1.430(5) Ni(2)-N(13) 2.119(3)
C(5)-C(6) 1.369(5) 0(1 1)-C(21) 1.294(4)
C(5)-C(11) 1.532(5) 0(12)-C(22) 1.278(4)
C(7)-C( 10) 1.534(5) C(21)-C(26) 1.412(5)
C(7)-C(8) 1.540(5) C(21)-C(22) 1.468(5)
C(7)-C(9) 1.541(5) C(22)-C(23) 1.453(5)

C(11)-C(13) 1.525(5) C(23)-C(24) 1.363(5)
C(l l)-C(12) 1.531(6) C(23)-C(27) 1.526(5)
C(l l)-C( 14) 1.533(5) C(24)-C(25) 1.429(5)
N(1)-C(15') 1.4413) C(25)-C(26) 1.371(5)
N( l )-C( l 7) 1.484(5) C(25)-C(31) 1.539(5)
N(1)-C(19) 1.489(5) C(27)-C(30) 1.531(5)

 

92

Table A.l. continued

 

 

Bond Bond Distance (A) Bond Bond Distance (A)
C(27)-C(28) 1.536(6) P(2)-F(13) 1.556(13)
C(27)-C(29) 1.538(6) P(2)-F(15) 1 .561(13)
C(31)-C(34) 1.522(5) P(2)-F(12') 1.569(9)
C(31)-C(32) 1.525(6) P(2)-F(1 1') 1.572(15)
C(31)-C(33) 1.526(6) P(2)-F(16') 1.659(8)
N(l l)-C(37) 1.470(5) P(2)-F(l l) 1.747(10)
N(11)-C(35) 1.482(5) F(11)-F(13') 156(2)
N(l 1)-C(39) 1.483(5) F(l l)-F(14') 1.64(2)
N(12)-C(36) 1.483(6) F(l 1)-F(14) 1.703(17)
N( 1 3)-C(3 8) 1.475(5) F(12)-F(15') 122(3)
N(14)-C(40) 1.476(7) F(12)-F(12') 123(3)
C(35)-C(36) 1.517(7) F(12)-F(16) 1.78(3)
C(37)-C(38) 1.505(6) F( l 3)—F(l3') 0.7l(3)
C(39)-C(40) 1 477(7) F(13)-F(16') 1 .78(2)

P(1)-F(3) 1.557(3) F(14)-F(12') 1.17(2)
P( 1 )-1=(2) 1.560(3) F(14)-F(1 1') 128(2)
P(1)-P(1) 1.564(4) F(15)-F(15') 1.18(3)
P(1)-F(5) 1.567(3) F(15)-F(1 1') 129(3)
P(1)-F(6) 1.583(3) F(16)-F(14') 0.62(2)
P(1)-F(4) 1.598(3) F(16)-F(16') 157(2)
P(2)-F(16) 1.392(14) F(l 1')-1=(13') 1.78(2)
P(2)-F(15') 1.426(16) F(15')-F(16') 158(2)
P(2)-F(13') 1.432(12) O(51)-C(52) 1.197(6)
P(2)-F(14) 1.494(1 1) C(52)-C(52) 1.322(6)
P(2)-F(12) 1.540(14) C(52)-C(53) 1.472(7)
P(2)-F(14') 1.553(13) C(51)-C(52) 1.466(8)

 

93

Table A.l. continued

 

Bond

C(53)-C(54)
C(53)-C(56)
C(55)-C(56)
C(56)-0(54')
C(56)-C(54)

Bond Distance (A)

1.322(1 1)
1.206(6)
1.473(7)
130(2)
1.355(9)

Bond

C(54)-C(57)
C(57)-C(58)
C(54)-C(57')
C(57')-C(58')

Bond Distance (A)

1.425(12)
1.483(16)
1.72(6)
139(6)

 

94

Table A.2. Bond Angles (°) for [Ni(tren)(3,5-DTBSQ)](PF6)

 

Atoms

Angle

Atoms

Angle

 

O(2)-Ni(l)-N(2)
O(2)-Ni(1)-O(1)
N(2)-Ni(1)-O(1)
O(2)-Ni(l)-N(1)
N(2)-Ni(1)-N(1)
O(1)-Ni(l)-N(l)
O(2)-Ni(l)-N(3)
N(2)-Ni(1)-N(3)
O(1)—Ni(1)-N(3)
N(l)-Ni(1)-N(3)
O(2)-Ni(1)-N(4)
N(2)-Ni(1)-N(4)
O(1)-Ni(1)-N(4)
N(1)-Ni(1)-N(4)
N(3)-Ni(1)-N(4)
O(2)-Ni( 1 )-N(2')
N(2)-Ni(l)-N(2')
O(1)-Ni(l)—N(2')
N(1)-Ni(1)-N(2')
N(3)-Ni( l)-N(2')
N(4)-Ni(I)-N(2’)
C(1)-O(1)-Ni(l)
C(2)-O(2)-Ni(l)
O(1)-C(1)-C(6)
O(1)-C(1)-C(2)
C(6)-C(1)-C(2)

95.8(6)
8049(9)
175.7(6)
177.64(11)
832(6)
100.54(11)
94.62(11)
95.6(6)
86.94(1 1)
83.33(12)
9937(1 1)
94.7(5)
83.77(1 1)
82.87(12)
161.66(12)
94.8(7)
1.0(13)

174.7(7)
84.3(7)
95.8(6)
94.8(6)
110.6(2)
113.3(2)
122.1(3)
118.1(3)
119.7(3)

C(2)-C(2)-C(3)
C(2)-C(2)-C(l)
C(3)-C(2)-C(1)
C(4)-C(3)-C(2)
C(4)-C(3)-C(7)
C(2)-C(3)-C(7)
C(3)-C(4)-C(5)
C(6)-C(5)-C(4)
C(6)-C(5)-C(l 1)
C(4)-C(5)-C(11)
C(5)-C(6)-C(I)
C(10)-C(7)-C(3)
C(10)-C(7)-C(8)
C(3)-C(7)-C(8)
C(10)-C(7)—C(9)
C(3)-C(7)-C(9)
C(8)-C(7)-C(9)

cu3yculycuz)
C(13)-C(11)-C(5)
C(12)-C(1 1)—C(5)
CU3}CU]}CU4)
C(12)—C(11)-C(14)
C(5)-C(1 1)-C(14)
C(15')-N(1)-C(17)
C(15')-N(1)-C(19)
C(17)-N(1 )-C(19)

123.8(3)
117.4(3)
118.8(3)
117.8(3)
122.4(3)
119.8(3)
124.1(3)
118.5(3)
122.7(3)
118.8(3)
121.0(3)
112.5(3)
107.2(3)
109.5(3)
107.5(3)
109.9(3)
110.3(3)
108.3(3)
110.0(3)
111.7(3)
109.0(3)
107.9(3)
109.8(3)
115.1(13)
113.0(11)
112.1(3)

 

95

Table A.2. continued

 

Atoms

C(lS')-N(1)-C(15)
C(17)-N(1)-C(15)
C(19)—N(1)-C(15)
C(15')-N(1)-Ni(1)
C(17)-N(1)—Ni(1)
C(19)-N(1)-Ni(1)
C(15)-N(1 )-Ni(1)
C(18)—N(3)-Ni(l)
C(20)-N(4)—Ni(1)
C(16)-N(2)-Ni(1)
N(1)-C(15)-C(16)
N(2)-C(16)-C(15)
C( 16')-N(2')-Ni(1 )
N(1)-C( 1 5')-C( 16')
N(2')-C(16')—C(15')
N(1)-C(l7)-C(18)
N(3)-C(18)-C(l7)
N(1)-C(19)-C(20)
N(4)-C(20)-C(19)
O(12)-Ni(2)-O(11)
O(12)—Ni(2)-N(14)
0(1 1 )-N i(2)—N( 14)
0(12)-Ni(2)-N(11)
0(1 1)-Ni(2)-N(1 1)
N(14)-Ni(2)-N(l 1)
O(12)—Ni(2)—N(12)

Angle

9(3)
109.3(12)
1 10.7(1 1)
104.4(14)
105.6(2)
105.6(2)
113.5(13)
109.9(2)
110.1(2)
110.4(12)
107(2)
111.9(15)
106.7(11)
119(2)
108.2(17)
111.0(3)
109.6(3)
110.7(3)
109.9(3)
8052(9)
97.78(12)

177.18(12)
177.43(11)

98.00(11)
83.78(13)
95.07(12)

Atoms

O(11)-Ni(2)-N(12)
N(14)-Ni(2)-N(12)
N(11)-Ni(2)-N(12)
0(12)-Ni(2)-N(13)
O(11)-Ni(2)-N(l3)
N(14)-Ni(2)-N(13)
N(11)-Ni(2)—N(l3)
N(12)-Ni(2)—N(13)
C(21)-O(11)-Ni(2)
C(22)-O(12)-Ni(2)
0(1 1)—C(21)-C(26)
0(1 1 )-C(21 )-C(22)
C(26)-C(21)-C(22)
0(12)-C(22)-C(23)
O(12)-C(22)-C(21)
C(23)-C(22)-C(21)
C(24)-C(23)-C(22)
C(24)-C(23)-C(27)
C(22)-C(23)-C(27)
C(23)-C(24)-C(25)
C(26)-C(25)-C(24)
C(26)-C(25)-C(3 1)
C(24)-C(25)-C(3 1)
C(25)-C(26)-C(21)
C(23)-C(27)-C(30)
C(23)-C(27)-C(28)

Angle

8860(12)
93.80(14)
82.77(13)
9882(12)
85.16(12)
92.91(14)
83.11(13)
163.61(13)
111.3(2)
113.3(2)
123.0(3)
117.0(3)
120.0(3)
123.7(3)
117.8(3)
118.5(3)
117.1(3)
123.6(3)
119.4(3)
125.1(3)
118.6(3)
123.0(3)
118.4(3)
120.7(3)
111.6(3)
110.3(3)

 

96

Table A.2. continued

 

Atoms

Angle

Atoms

Angle

 

C(30)-C(27)-C(28)
C(23)-C(27)—C(29)
C(30)-C(27)-C(29)
C(28)-C(27)-C(29)
C(34)-C(31)—C(32)
C(34)-C(31)—C(33)
C(32)-C(31)-C(33)
C(34)-C(31)-C(25)
C(32)-C(31)-C(25)
C(33)-C(31)-C(25)
C(37)-N(1 1)-C(35)
C(37)-N(1 1)-C(39)
C(35)-N(11)-C(39)
C(37)-N(l 1 )—N 1(2)
C(35)-N(1 1 )-N 1(2)
C(39)-N(1 1)-Ni(2)
C(36)-N(12)-Ni(2)
C(38)-N(13)-Ni(2)
C(40)-N(14)-Ni(2)
N(l 1)-C(35)-C(36)
N(12)-C(36)-C(35)
N(l 1)-C(37)-C(38)
N(13)-C(38)-C(37)
C(40)-C(39)—N(1 1)
C(39)-C(40)-N(l4)
F(3)-P(1)-F(2)

107.9(4)
109.6(3)
107.8(3)
109.6(3)
109.6(4)
107.2(4)
107.6(4)
110.2(3)
110.1(3)
112.1(3)
111.8(3)
111.4(3)
1 13.6(3)
105.6(2)
104.1(2)
109.8(2)
110.6(2)
109.3(2)
105.6(3)
110.0(4)
109.1(3)
111.0(3)
110.0(3)
111.9(4)
112.8(4)
91.2(2)

F(3)-P(1)-F(1)
F(2)-P(1)-F(l)
F(3)«P(1)-F(5)
P(2).P(1)-F(5)
F(1)-P(1)-F(5)
F(3)-P(1)-F(6)
F(2)-P(I)-F(6)
F(1)-P(l)-F(6)
F(5)-P(1)-F(6)
F(3)-P(1 )-1=(4)
F(2)-P(1)-F(4)
F(1)-P(l)-F(4)
F(5)-P(1)-F(4)
F(6)-P(1)-F(4)
F(16)-P(2)-F(15')
F(16)-P(2)-F(13')
F(15')-P(2)-F(13')
F(16)-P(2)-F(14)
F( 1 5')-1>(2)—1=( 14)
F(l3')-P(2)-F(14)
F(16)—P(2)-F(12)
F(15')-P(2)-F(12)
F(13')-P(2)-F(12)
F(14)-P(2)-F(12)
F(16)-P(2)-F(14')
F(15')-P(2)-F(l4')

179.1(3)
892(3)
902(2)

91 .73119)
90.6(3)
90.0(2)

177.16(19)
895(2)

90.82(17)
892(2)

8823(18)
90.0(2)
179.4(2)

8923(16)

93.1(14)
103.8(12)
136.4(17)
128.0(16)
110.7(10)
90.0(13)
74.4(12)
48.6(1 1)
175(2)
87.6(9)
235(9)
115.8(14)

 

97

Table A.2. continued

 

Atoms

mic

Atoms

Angle

 

F( 1 3')-P(2)F( 14')
F(14)-P(2)-F(14')
F(12)-P(2)-F(14')
F(16)-P(2)-F(13)
F(15')-P(2)-F(13)
F(13')-P(2)-F(13)
F(14)-P(2)-F(13)
F(12)-P(2)-F(13)
F(14')-P(2)-F(13)
F(16)-P(2)-F(15)
F(15')-P(2)-F(15)
F(13')-P(2)-F(15)
F(14)-P(2)-F(15)
F(12)-P(2)-F(15)
F(14')-P(2)-F(15)
F(13)—P(2)-F(15)
F(16)-P(2)-F(12')
F(15')-P(2)-F(12')
F(l3')-P(2)-F(12')
F(l4)-P(2)-F(12')
F(12)-P(2)-F(12')
F(14')-P(2)-F(12')
F(13)-P(2)-F(12')
F(15)-P(2)-F(12')
F(16)—P(2)-F(1 1')
F(15')-P(2)-F(1 1')

89.9(10)
1 10.2(12)
86.9(14)
98.0(13)
1 1 1.6(14)
27.3(11)
1 13.4(15)
156.9(17)
94.3(8)
136.5(10)
46.2(10)
100.0(13)
87.4(8)
84.3(14)
159.9(10)
87.0(9)
93.4(14)
88.4(9)
129.3(12)
44.6(8)
46.8(9)
87.9(8)
156.3(12)
99.0(7)
174.6(15)
92.2(11)

F(13')-P(2)-F(l 1')
F(14)-P(2)-F(1 1')
F(12)-P(2)-F(1 1')
F(14')-P(2)-F(1 1')
F(13)-P(2)-F(1 1')
F(15)-P(2)-F(1 1')
F(12')-P(2)-F(l 1')
F(16)-P(2)—F(l6')
F(15')-P(2)-F(16')
F(13')-P(2)-F(16')
F(14)-P(2)-F(16')
F(12)-P(2)—F(16')
F(14')-P(2)-F(16')
F(13)-P(2)-F(16')
F(15)-P(2)-F(16')
F(12‘)-P(2)-F(16')
F(l 1')-P(2)-F(16')
F(16)-P(2)-F(1 1)
F( 15')-P(2)-F(1 1)
F(13')-P(2)-F(1 1)
F(14)-P(2)-F(1 1)
F(12)-P(2)-F(11)
F(14')-P(2)-F(1 1)
F(13)-P(2)-F(1 1)
F(15)-P(2)-F(1 1)
F(12')-P(2)-F(11)

72.4(9)
49.2(10)
109.1(9)
151.2(16)
80.4(10)
48.7(11)
86.4(9)
61 .1(10)
60.9(9)
92.8(8)
169.5(8)
90.4(13)
80.0(7)
672(8)
82.1(6)
136.3(10)
122.3(17)
82.8(7)
165.8(13)
57.7(10)
62.8(6)
117.3(18)
59.3(7)
825(8)
140.5(9)
78.4(6)

 

98

Table A.2. continued

 

Atoms

F(l 1')-P(2)-F(1 1)
F(16')-P(2)-F( 1 1)
F(13')-F(1 1)-F(14')
F(13')-F(11)-F(14)
F(14')-F(1 1)-F(14)
F(13')-F(1 1)-P(2)
F(14')-F(1 1)-1>(2)
F( 14)-F(1 1)-1>(2)
F(15')-F(12)-F(12')
F(15')-F(12)-P(2)
F(12')-F(12)-P(2)
F(15')-F(12)-F(16)
F(12')-F(12)-F(16)
P(2)-F(12)-F(1 6)
F(13')-F(13)-P(2)
F(l3')-F(13)-F(16')
P(2)-F(13)-F(16')
F(12')-F(14)-F(1 1')
F(12')-F(14)-P(2)
F(l 1')-F(14)-P(2)
F(12')-F(14)-F(11)
F(11')-F(14)-F(11)
P(2)-F(14)-F(ll)
F(15')-F(15)-F(1 1')
F(15')-F(15)-P(2)
F(11')-F(15)—P(2)

Angle

91.9(15)
126.9(5)
82.5(8)
78.6(11)
96.7(10)
50.9(5)
54.5(5)
51.3(4)
116.4(19)
60.8(11)
67.8(11)
83.9(14)
90(2)
49.0(8)
66.7(16)
122(2)
592(7)
123(2)
71 .1(10)
68.6(12)
92.2(17)
105.7(14)
65.9(7)
122.2(18)
60.8(10)
66.1(9)

Atoms

F(14')-F(16)-P(2)
F(14')-F(l6)-F(16')
P(2)-F(16)-F(16')
F(l4')-F(l6)-F(12)
P(2)-F(l6)-F(12)
F(16')-F(16)-F(12)
F(14)-F(11')-F(15)
F(14)-F(1 1')-1>(2)
F(15)-F(1 1')-1>(2)
F(l4)-F(1 1')-F(13')
F(15)-F(l 1')-F(13')
P(2)-F(1 1')-F(l3')
F(14)-F(12')-F(12)
F(14)-F(12')-P(2)
F(12)-F(12')-P(2)
F(13)-F(13')-P(2)
F(13)-F(13')-F(l 1)
P(2)-F(13')-F(1 1)
F(13)-F(13')-F(11')
P(2)-F(13')-F(1 1')
F(11)-F(l3')-F(11')
F(16)-F(14')-P(2)
F(16)-F(14')-F(1 1)
P(2)-F(14')-F(1 1)
F(15)-F(15')-F(12)
F(15)-F(15')-P(2)

Angle

93(3)
137(4)
67.9(9)
1 17(4)
56.6(8)

85.3(16)

110.4(17)
622(9)

65.2(11)

83.5(14)

95.4(19)
502(6)

122.0(15)
64.3(7)
65.4(8)

86(2)
145(3)
71.4(9)

99(3)
57.5(7)

91 .2(16)

64(2)
130(3)
66.3(8)
120(3)

73.0(16)

 

99

Table A.2. continued

 

Atoms

F(12)-F(15')-P(2)
F(15)-F(15')-F(16')
F(12)-F(15')-F(16')
P(2)-F(15')-F(16')
F(16)-F(l6')-F(15')
F(16)-F(16')-P(2)
F(15')-F(16')-P(2)
F(16)-F(l6')-F(13)
F(15')-F(16')-F(13)
P(2)-F(16')-F(13)
C(52)-C(52)-C(53)
O(51)—C(52)—O(52)
O(51)-C(52)-C(51)

Angle

70.6(14)
99(2)
107.7(15)
66.9(9)
81.3(11)
51 .0(6)
522(6)
832(9)
94.4(13)
53.6(5)
113.4(5)
123.5(6)
124.4(6)

Atoms

O(52)—C(52)-C(51)
C(54)-C(53)-O(52)
O(53)-C(56)—O(54')
O(53)-C(56)-O(54)
O(54')-C(56)-O(54)
O(53)—C(56)—C(55)
O(54')-C(56)-C(55)
O(54)-C(S6)-C(55)
C(56)-O(54)-C(57)
O(54)-C(57)-C(58)
C(56)-O(54')-C(57')
C(58')-C(57')-O(54')

Angle

112.0(5)
112.5(7)
120.5(12)
121.9(6)
31(3)
125.7(5)
108.6(1!)
112.0(5)
117.8(7)
115.1(9)
114(2)
82(4)

 

100

Table A.3. Anisotropic Displacement Parameters [A2 x 103] for [Ni(tren)(3,5-
DTBSQ)](PF6).The Anisotropic Displacement Factor Exponent Takes the Form: -2p2 [

(ha*)2U11 + + 2hka*b*U12]

 

 

Atom 1111 1122 1133 1123 1113 1112
Ni(l) 26(1) 30(1) 19(1) 4(1) 4(1) 1(1)
0(1) 30(1) 38(2) 23(1) 6(1) 3(1) -211)
0(2) 29(1) 31(1) 23(1) 3(1) 3(1) -4(1)
C(1) 27(2) 25(2) 20(2) 3(2) 5(2) 5(2)
C(2) 27(2) 26(2) 23(2) 3(2) 4(2) 6(2)
C(3) 28(2) 28(2) 23(2) 7(2) 5(2) 5(2)
C(4) 29(2) 35(2) 24(2) 10(2) 7(2) 6(2)
C(5) 25(2) 30(2) 25(2) 4(2) 0(2) 10(2)
C(6) 21(2) 30(2) 27(2) 4(2) 1(2) 1(2)
C(7) 33(2) 35(2) 24(2) 8(2) 3(2) -3(2)
C(8) 32(2) 45(2) 43(2) 10(2) 8(2) 1(2)
C(9) 40(2) 31(2) 45(3) 9(2) 3(2) -1(2)
C(10) 44(3) 53(3) 39(2) 13(2) 13(2) -13(2)
C(1 1) 3 1 (2) 37(2) 20(2) 0(2) -2(2) 6(2)
C(12) 38(2) 66(3) 29(2) —4(2) -3(2) -9(2)
C(13) 60(3) 52(3) 27(2) 1(2) -6(2) 23(2)
C(14) 44(2) 44(2) 35(2) -6(2) 2(2) 8(2)
N(l) 26(2) 30(2) 21(2) 1(1) 2(1) 1(1)
N(3) 32(2) 40(2) 28(2) 1(2) 6(1) 8(2)
N(4) 36(2) 34(2) 31(2) 5(1) -2(1) 3(1)
N(2) 29(10) 55(13) 40(13) 30(9) 19(8) 16(8)
C(15) 25(9) 27(8) 9(8) 5(7) -1(5) -1315)
C(16) 38(6) 44(6) 23(4) -4(4) 5(4) -4(5)
N(2') 34(1 1) 12(8) 24(1 1) -24(6) -8(7) -15(6)
C(15') 56(14) 45(13) 23(1 1) 17(7) 22(7) 20(10)
C(l6') 26(6) 46(6) 26(5) 5(4) 2(4) -1(4)

 

101

Table A.3. continued

 

 

Atom 1111 1122 1133 1123 U13 U12
C(17) 45(3) 32(2) 37(2) 8(2) 9(2) 2(2)
C(18) 56(3) 37(2) 33(2) 8(2) 9(2) 15(2)
C(19) 25(2) 44(2) 39(2) 9(2) 8(2) 3(2)
C(20) 30(2) 46(2) 42(2) 7(2) 8(2) 12(2)
Ni(2) 25(1) 32(1) 21(1) 4(1) 3(1) 5(1)
0(11) 30(1) 38(2) 26(1) 5(1) 4(1) 11(1)
0(12) 36(2) 39(2) 22(1) 8(1) 2(1) 14(1)
C(21) 23(2) 25(2) 25(2) 4(2) 5(2) 1(2)
C(22) 21(2) 24(2) 26(2) 3(2) 3(2) 0(2)
C(23) 26(2) 29(2) 27(2) 4(2) 5(2) 6(2)
C(24) 34(2) 37(2) 20(2) 4(2) 7(2) 9(2)
C(25) 25(2) 28(2) 25(2) 5(2) 0(2) -1(2)
C(26) 19(2) 29(2) 32(2) 7(2) 3(2) 5(2)
C(27) 42(2) 36(2) 26(2) 5(2) 6(2) 17(2)
C(28) 32(2) 51(3) 55(3) 15(2) 8(2) 15(2)
C(29) 45(3) 36(2) 49(3) 7(2) -2(2) 14(2)
C(30) 71(3) 70(3) 42(3) 1 1(2) 16(2) 47(3)
C(31) 25(2) 46(2) 26(2) 14(2) 3(2) 6(2)
C(32) 69(3) 69(3) 27(2) 16(2) 45(2) -16(3)
C(33) 73(3) 95(4) 39(3) 30(3) 10(2) 51(3)
C(34) 36(2) 82(3) 50(3) 37(3) 4(2) 0(2)
N(l 1) 32(2) 36(2) 30(2) 4(1) 8(1) 4( 1)
N(12) 30(2) 54(2) 36(2) 10(2) 5(2) -3(2)
N(13) 37(2) 38(2) 39(2) 8(2) -5(2) 4(2)
N(14) 41(2) 56(2) 35(2) 14(2) 6(2) 16(2)
C(35) 67(3) 30(2) 49(3) -1(2) 26(2) 8(2)

 

102

Table A.3. continued

 

 

Atom U1 1 1122 1133 1123 1113 1112
C136) 72(3) 44(3) 46(3) -2(2) 14(2) -1912)
C137) 34(2) 5213) 54(3) 16(2) 17(2) 1412)
C138) 2812) 57(3) 6013) 18(2) 912) -312)
C139) 5613) 5713) 29(2) 4(2) 12(2) 8(2)
C(40) 105(5) 126(5) 43(3) 24(3) 1(3) 61(4)
P(1) 29(1) 39(1) 4311) 4(1) 6(1) 5(1)
F(1) 136(4) 202(5) 111(3) 103(3) 1313) 6713)
F12) 97(3) 96(3) 1 14(3) -4012) 57(2) 212)
F(3) 84(2) 1 1913) 1 17(3) 71(2) 5(2) 39(2)
F(4) 54(2) 60(2) 118(3) -1012) 3612) -16(1)
F(S) 52(2) 8712) 106(3) -l6(2) 1312) -2712)
F(6) 61(2) 62(2) 91(2) -1212) 39(2) 6(1)
P(2) 3411) 60(1) 141(2) 3211) 111) 811)
F(l 1) 88(7) 160110) 148(1 1) 78(9) 48(6) 45(6)
F(12) 153(15) 320(30) 230(20) -86(18) -l6(13) 166(18)
F(13) 82(8) 80(7) 350(30) 74(13) 3112) 50(6)
F(14) 91(9) 320(30) 139113) 162117) 419) 41(14)
F(15) 74(7) 13411 1) 160113) 8611 1) 0(8) -2217)
F(l6) 178115) 250(20) 84(8) 8000) 25(8) -151114)
F(l 1') 9018) 144113) 430130) -139117) 52(16) -28(9)
F(12') 6015) 126(9) 223117) 115111) 59(9) 51(6)
F( 13') 8718) 340130) 1 17110) -50113) 3017) 63113)
F(14') 108(8) 8716) 260120) 58110) 55110) -32(6)
F(15') 184119) 300(30) 250(20) 210120) 76(16) .44117)
F(16') 8016) 191111) 12719) 74(8) -45(6) -1 716)
0151) 138(4) 9313) 54(3) 3612) -1912) -52(3)

 

103

Table A.3. continued

 

 

Atom 1111 1122 U33 023 1113 1112
0152) 5412) 66(2) 6512) 14(2) -112) -312)
C(51) 61(3) 59(3) 9915) 2913) -2413) -713)
C152) 6313) 49(3) 61(4) 2213 ) -1513) -2313)
C153) 8115) 101(5) 116(6) 015) 1914) 4(4)
C(54) 106(7) 197111) 400(20) -189(12) - 122110) 8818)
0153) 1 1 1(3) 76(2) 46(2) 4(2) 19(2) 49(2)
C155) 80(4) 79(4) 80(4) 39(3) 26(3) 3713)
C156) 6613) 72(3) 39(3) 11(3) 10(2) 30(3)
0154) 23(4) 39(3) 35(3) 2(2) 312) 8(2)
C(57) 4915) 29(6) 40(5) 3(3) 314) 1 1(4)
C(58) 39(6) 5917) 4115) 1514) 1214) 13(5)
0154') 460150) 100117) 60113) 35112) 100120) 150130)
C(57') 100120) 190140) 110120) 60120) 55117) 30120)
C(58') 260160) 22114) 490190) 30120) 300160) 20120)

 

104

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

111111111111211111111111.1111