fl

‘-

LIBRARY
““ihisan State
1 University

~ J

—__~

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

fif‘é— 3;

MSU Is An Affirmative Action/Equal Opportunity Institution
cztclMpma-pd

 

 

 

REDOX PROPERTIES AND CHEMICAL STABILITY OF SOLUBLE
CONDUCTING MATERIALS: THE u-OXO—(TETRA-t-BUTYLPHTHALO
CYANINATO)GERMANIUM AND POLY(3.4-DIBUTYLTHIOPHENE)

By

Line Francoise Le Blevenec

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillement of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1 990

¢55 -— (DO/>6“

ABSTRACT

REDOX PROPERTIES AND CHEMICAL STABILITY OF SOLUBLE
CONDUCTING MATERIALS: THE p-OXO-(TETRA-t—BUTYLPHTHALO
CYANINATO)GERMANIUM AND POLY(3,4-DlBUTYLTHIOPHENE)

By

Line Francoise Le Blevenec

The redox chemistry and chemical stability of soluble conducting
materials such as the already existing II-oxo-(tetra—t-butylphthalocyaninato)
germanium and the newly synthesized poly(3,4-dibutylthiophene) were
characterized using a variety of conventional electrochemical and spectroscopic
techniques.

Controlled potential coulometry (CFC), cyclic (CV), rotating disk (RDE),
rotating ring-disk (RRDE), and differential pulse (DPV) voltammetry data Show
that the germanium polymer can be reduced to any fractional oxidation state
ranging from 0 to 1- and remains chemically stable. The redox response
occurring over a 1500 mV range is indicative of a broad continuous process. The
observed voltammograms are relatively featureless compared to conventional
electroactive compounds but have the characteristics of a Nernstian process. At
a degree of reduction greater than 100% (Le. one electron added per
phthalocyanine site). the polymer quantitatively breaks into electroactive

monomer units. The decomposition was easily monitored using modified RFlDE

technique. Digital simulations of voltammetric responses prove that the
decomposition of the germanium polymer follows an ECEE-type mechanism. A
rate of 20 sec'1 was obtained for the decomposition.

The other soluble conducting polymer under investigation, the
poly(3,4-dibutylthiophene), was synthesized by oxidative polymerization of the
corresponding monomer using CV technique. CPC, CV, and DPV data Show that
this polythiophene derivative can be oxidized to any degree in the 0034 range
both in solution and as thin film coated on an electrode surface. In any form or
state, it is chemically stable. Again a broad, continuous redox response is
observed in solution over a 1000 mV range, resulting in the formation of a
smooth doping profile. The conductivity profile as function of band-filling parallels
the doping profile. A maximum value of 1 S cm"1 was obtained for a 34%
oxidized film using standard four-probe method. Optical absorption
measurements performed on solutions of poly(3,4-dibutylthiophene) oxidized to
various degrees indicate that the nature of the charge carriers is mainly bipolaron

in character. No polaron state was detected, even at low oxidizing levels.

To my father

iv

ACKNOWLEDGMENTS

First. I would like to express my gratitude to my advisers Dr. Eugene
Legoff and Dr. John G. Gaudiello who helped to guide me during the course of
this study. Their principles of independence might have resulted in a little bit of a
frustration sometimes but in the long run it is the best way of obtaining scientific
integrity. I also would like to thank Dr. Alexander I. Popov for his encouragement,
help, and friendship towards a foreigner.

I also wish to thank the members of my group in particular Dave and
Evaldo for their useful comments and discussions as well as Mike (Benz) for
having provided me some thiophene derivative compounds for my study.

I also would like to thank Anne and Marina for their unforgettable
friendships.

A special thank goes to my parents for their guidance, unconditional
support, and love throughout my all life.

Finally, I would like to thank my husband for his friendship. understanding.

and love all along these years.

TABLE OF CONTENTS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

page #
List of Tables - - - - - ................. viii
Listof Schemes -- --- ..................... ix
List of Figures - - - x
Introduction ...... 1
Experimental .. - --- .......................... 23
2.1 Glassware ................... 23
2.1.1 Vacuum lines 23
2.12 Electrochemical cells .......................................... 23
22 Methods of purification ..... _ 25
22.1 Solvents - ....... 25
222 Supporting electrolyte ........................................ - - 25
2.3 Instrumentation -- ............. 26
2.3.1 Chromatographic 26
2.3.2 Spectroscopic- - 26
2.3.3 Electrochemical ....... 27
2.4 Electrochemical techniques -- - -28
2.4.1 Controlled potential coulometry 28
2.4.2 Voltammetry ..... 32
2.4.3 Simulation ........................................................................... 40
2.5 Spectroelectrochemical technique .............................................. 44

vi

 

Probing the chemical smility as function of band filling for

 

the soluble u—Oxo-(tetIa-t-butylphthalocyaninatolgennanium

3.1 Introduction
32 Synthesis
3.3 Electrochemical studies

3.3.1 Cyclic voltammetry

 

3.32 Controlled potential coulometry

 

 

3.3.3 Rotating disk and rotating ring-disk
voltammetry
3.3.4 Differential pulse voltammetry

 

3.4 Electrochemicd characterization of the dihydroxy
(tetra-t-butylphthdocyaninato)gennanium -

a as asses é

 

Studying the electrical conductivity and spectroscopic
properties as function of band-filling for the soluble

 

poly(3,4-dibutylthiophene)

4.1 Introduction
42 Synthesis

 

4.3 Electrochemical dtaracterization

 

4.4 Fledox properties

 

4.5 Optical properties

 

 

4.6 Conductivity

Conclusions

AmendsK

 

List of References

vii

112

112
114
131
139
142
150

158

160

166

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

LIST OF TABLES

 

 

 

page 4*
Pressed powder room temperature electrical
conductivity data for doped and undoped alkyl
substituted u-oxopolymers. 17
Pressed powder room temperature electrical
conductivity data for doped and undoped alkyl
substituted u-oxopolymers. ..... 17
Infra-red spectral data, optical absorption spectral data
and elemental analytical data for the [t-Bu4PcGeO]nand
t-Bu4PcGe(OH)2 materials. ......................................................................... 54-
Controlled potential coulometry data for u-oxo-(tetra-t-
butylphthalocyaninato)germanium. - 80
Room temperature electrical conductivity data for
B-mono and disubsfimted thiophene derivatives 154

 

viii

Scheme l.

Scheme ll.

Scheme III.

Scheme IV.

UST OF SCHEMES

germanium polymer.

 

 

Chemical preparation of 3,4-dibutylthlophene.

reaction for thiophene derivatives.

 

page 1*

Synthesis of the p-oxo-(tetra-t-butylphthalocyaninato)
-- 51
114

Mechanism of the electrochemical polymerization
119
128

Chemical preparation of 3',4'-dibutyl-a-terthiophene.

 

ix

Figure 1.

figure 2

figure 3.

figure 4.

FIQUI’O 5.

figure 6.

figure 8.

figure 9.

UST OF FIGURES

Room temperature electrical conductivity (8 cm") of
common materials. The vertical axis represents the
conductivity scale

pageif

 

 

Process leading to the formation of an energetic band

Schematic structure of peripherally-alkylated
p-oxopolymers [R4PcMo]n M a Si, Ge, Sn;

13

 

Pc - phthalocyarunato; R - t-Bu4, Tms.

Electrochemical cell used in controlled potential
coulometry experiment
A a platinum gauze counter electrode
8 = fine porosity glass frits
C a silver wire reference electrode
0 a platinum gauze working electrode for
CPC
E a platinum disk working electrode for CV
and DPV

3O

 

Electrochemical cell used in rotating disk and rotating
ringdisk voltammetry

A = platinum gauze counter electrode

B a fine porosity glass frits

C‘s platinum disk and ring-disk working

electrodes
0 = silver wire reference electrode
E :- brush contact

39

 

Flow chart explaining the DPV simulation program.

42

 

Fourier-transformed infra-red spectra of [t-Bu4PcGe0]n

56

 

and t-Bu4PcGe(OI-l)2 as Nujol mulls.

Optical absorption spectra of [t-Bu4PcGeO]n and
t-Bu4PcGe(OH)2 as solutions in THF.

 

Figure 10.

FIgure11.

figure 12.

figure 13.

Figure 14.

Cyclic voltammograms for the reduction of a 33. 4 M
solution of [t-Bu PcGeO]n/102 M TBABF4 in THF at a
sweep rate of 10Ii mV sec
a virgin polymer solution
b. two successive scans of the virgin
polymer solution
c. polymer solution after controlled potential
coulometry (CPO) reduction at -1.65 V
(decomposition into monomeric type
units) and oxidation back to its neutral
form

62

 

Current-time curves for the reduction of a 33.4 pM
solution oflt-Bu4PcGeOIn/02 M TBABF4 in THF during
controlled potential electrolysis (CPO).
a polymer solution reduced at -0.55 V
b. polymer solution reduced at -1.05 V
c. polymer solution reduced at -125 V
d polymer solution reduced at -1.45 V

65

 

Current-time curves for the oxidation of reduced
[t--Bu4PcGeO)'{TBABF4 in THF resulting of figure 11

a'. polymer solution (a) re-oxidized at 020 V

b'. polymer solution (b) re-oxidized at 020 V

a'. polymer solution (c) re-oxidized at 020 V

d'. polymer solution ((1) re-oxidized at

0.20 V

67

 

Degree of partial reduction vs potential determined by
controlled potential mcoulometry for the reduction of
[t-48u PcGeO]/I'BAB in THF. Each point ( )
corresponds Po the mean value for a reduction/re-
oxidation cycle. The error bars represent the difference
between the reduction and re-oxidation.

70

 

Optical absorption spectra of [t-Bu4PcGeO]n in THF
after CPC at -1.44 V (96 % reduction) and oxidation
back to its neutral form at 0.20 V. The spectrum is very
similar to the one obtained for the virgin polymer

- ........ 72

 

(Figure 9). - .......

xi

Figure 15.

figure 16.

figure 17.

figure 18.

figure 19.

Fourier-transformed infra-red spectra of
[t-BucheflH/Nujol after CPC at -1.44 V (98 %
reduction) oxidation back to its neutral form at
020 V. The band at 904 cm' is assigned to the

 

asymmetric O-Ge-O strech.

Variations of log (i(t)fI(0)) with time at different potentials
determined from controlled potential coulometry curves
for the reduction of [t-Bu4PcCieO]n/‘l'BABF4 in THF. i(0)
represents the initial current value for each CPO
experiment. The ordinate scale pertains only to the data
obtained at -1.02 V. Each successive curve is shifted
upwards by 1 unit The % reduction of the listed
potentials are as follows: -1.02 V: 50% reduction;

 

-1.44 V: 98% reduction, and -1.65 V: 227% reduction.

Rotating disk electrode voltammograms for the
reduction of a 33.4 M solution of [t-Bu4PcGeOIn/02 M
TBABF4 in THF at various rotation rates. The current of
the blank solution does not change with increasing
rotation rate.

 

Levich plots for the reduction of
[t-Bu4PcGeO]n/'I'BABF4 in THF at various potentials.
The % reduction of the listed potentials are as follows:
064 V (25% reduction); -0.84 V (37% reduction);
-1.04 V (50% reduction); -1.24 V (68% reduction);
-1.44V(98% reduction).

 

Degree of partial reduction vs potential determined by
rotating disk voltammetry for the reduction of
[t-Bu4PcGeO) F4 in THF. Each point (_)
corresponds the mean value for a reduction/re-
oantion cycle. The mean value is calculated by
averaging the current at different rotation rates. The
error bars represent the standard deviation-

 

xii

74

86

89

Figure 20.

Figure 21.

Figure 22.

Figure 23.

Experimental (a.b.c) and simulated (d,e,f) rotating
electrode voltammograms for the reduction of a 33.4
M solution of [t-Bu4PcGeO]n/02 M TBABF4 in THF.
The rotation {ate was 4000 rpm.with a scan increment
of 4 mV sec' .
a.d. disk voltammograms in the absence of
the ring
b,e. ring voltammograms in the absence of
the disk
c,f. ring voltammograms when the disk is
held at -1.47 V and the ring is scanned
from this value to the rest potential

Continuous shielding-collection plots for the reduction of
[t-Bu4Pct'ieO]n/‘I'BABF4 in THF at various disk
potentials. The rotation rate is 2000 rpm for all plots. i

represents the cathodic ring current when the disk Is
held at a particular potential, i 0° is the cathodic ring
current in the absence of reaction at the disk. and N is
the collection efficiency of the electrode assembly-

........

 

Differential pulse voltammograms for the reduction of a
33.4 pM solution of [t-Bu4PcGeO]r/02 M TBABF4 in
THF. The experimental conditions are as foll : pulse
amplitude: 50 mV; pulse width: 50 mV sec; ; pulse
duration: 1000 msec and scan rate: 4 mV sec' .
a, b. virgin polymer solution
c. polymer solution after CPC at -1.45 V

(98% reduction) and oxidation back at

0.20 V

 

Cyclic (a) and differential pulse (b) voltammograms of a
881 IIM solution of t-Bu4PcGe(OH)2/0.2 M TBABF4 in
THF. The experimental conditions were the same as
above (figures 10 and 22). - ..... _

xiii

91

97

Figure 24.

Figure 25.

Figure 26.

Figure 27.

Figure 28.

Figure 29.

Differential pulse voltammograms for the reduction of a
solution of 881 IIM t-Bu4PcGe(OH)2/02 M TBABF4 in
THF. The experimental conditions are identical to the
ones of Figure 22.

a. virgin monomer solution

b. monomer solution after CPC reduction at

-0.50 V and oxidation back at 0.20 V

c. monomer solution after CPC reduction at
-1.00 V and oxidation back at 020 V

d. monomer solution after CPO reduction at
-1.60 V and oxidation back at 020 V

a. monomer solution after CPC reduction at

-1.85 V and oxidation back at 0.20 V

 

Experimental (a) and simulated (b) differential pulse
voltammograms of a 50.1 M solution of [t-
Bu4PcGe01n/02 M TBABF4 in THF. Both the
simulation and experimental conditions are identical to
the ones of Figure 22. The standard potentials used to

model the system are represented by the vertical lines

on the potential axis. -

Degree of partial reduction vs potential determined by
controlled potential coulometry, rotating disk
voltammetry. and digital simulation. -_

 

Simulated continuous shielding-collection plots for the
reduction of [t-Bu PCGOOIHITBABF4 in THF at various
disk potentials. a rotation rate was 2000 rpm. The
simulated data are based on the assumption of the
EnCEE mechanism. The % reduction of the listed
potentials are as follows: -1.45 V = 100% reduction;
-1.64 V = 26% reduction, and —2.16 V = 320%
reduction.

 

1H-NMR spectrum of 3,4-dibutylthiophene in

00300003. The peak at 2.04 ppm is due to the

reSIdual solvent

 

Repetitive cyclic voltammograms for the polymerization
of a 25 mM solution of 3,4-dibutylthiophene/0 M
TBABF4 in CH3NO2 at a sweep rate of 50 mV sec'

 

xiv

102

105

108

110

116

121

Figure 30.

FIgure 31.

Figure 32

figure 33.

Figure 34.

Figure 35.

Figure 36.

figure 37.

Figure 38.

Fourier-transformed infra-red spectrum of poly(3,4-
dibutylthiophene) on KBR pressed pellet.

124

 

1II-NIIIII spectrum of poly(3,4-dibutylthiophene) in

CDCI3. The peak at 724 ppm is due to the solvent .................

Gel permeation chromatogram of the poly(3, 4-

dibutylthiophene) inTHF. The flow rate is 1.5 ml min .......

................. 126

127

 

Repetitive cyclic voltammograms for the polymerization

of a 25 mM solution of 3',4‘-dibutyl-ot-terthiophene/0 02

M TBABF4 in CH3CN at a sweep rate of so mV sec“.

- 130

 

Repetitive cyclic voltammograms for the polymerization
of a 25 mM solution of 3" 4"
dibutquuinquethiophene/O. .02 M TBABF4 in CHSCN at
a sweep rate of 50 mV sec _ .......

........ 133

 

Cyclic voltammograms of poly(3,4-dibutylthiophene) (a),
poly(3'.4'-dibutyl-a-terthiophene) (b), and poly(3' 4"-
dibutquuinquethiophene) (c) as synthesized films“ in

---135

 

CH30N/TBABF4. The sweep rate was 50 mV sec ........

Cyclic (a) and differential pulse (b) voltammograms of a
273 pM solution of poly(3, 4-dibutylthiophene)/02 M
TBABF4 in CH 2C.I The experimental conditions for
differential pulsevo metry were: .1 pulse amplitude =
50 mV; pulse width- 50 mV sec' ; pu duration .
1000 mV, and scan rate . 4 mV sec .for cyclic
voltammetry, the sweep rate was 50 mV sec

 

Degree of partial oxidation vs potential determined by
controlled potential coulometry for poly(3,4-
dibutylthiophene)fI'BABF4 in CH CI solution and as
film. Each point ( ) corresponds o 9 mean value for
an oxidation/Ie-reduction cycle. The error bars
represent the difference between the oxidation and

re-Ieduction. - - ..................................

 

ln-situ absorption spectra of poly(3,4-dibutylthiophene)
as synthesized film and solution in both its neutral

......... 138

..141

- ..... 144

 

(undoped) and oxidized forms.

XV

Figure 39.

Figure 40.

Figure 41.

Figure 42

ln-situ electronic absorption spectra of a 273 M
solution of poly(3,4-dibutylthiophene)/02 M TBABF4

 

oxidized at various degrees in CHZCIZ.

Optical absorption spectra of poly(3,4-dibutylthiohene),
poly(3',4'-dibutyl-a-terthiophene). and poly(3',4"-
dibutquuinquethiophene) as synthesized films.

 

Electrical conductivity vs degree of partial oxidation at
room temperature for the poly(3,4-dibutylthiophene).

 

Electrical conductivity vs degree of partial oxidation at
room temperature for poly(3,4-dibutylthiophene),
poly(3',4'-dibutyl-ot-terthiophene). and poly(3",4'-
dibutquuinquethiophene).

 

xvi

147

149

152

157

1 . INTRODUCTION

Organic materials that become electrically conducting upon oxidation or
reduction (doping) have been a field of great interest over the past two decades‘.
This interest really started with the discovery in the 19705 that oxidized
polyacetylene exhibits a conductivity value in the metallic regimez. It was
reinforced with the possible use of these materials in numerous technological
applications. Such applications include batteries, sensors. electrochromic
displays, solar energy devices, electromagnetic shieldings, antistatic charge
devices, and microelectronic components for computer industry3. These
conducting materials are generally classified into two groups: charge-transfer
salts‘-4 and conducting polymeISI-5. Charge-transfer salts are comprised of
discrete molecular components which have been condensed into a segregated,
stacked framework. A typical example is T'TF-TCNO (tetrathiafulvalene
tetracyanoquinodimethane) complex. Conducting polymers are covalently linked
assemblies based on covalently bonded conjugated hydrocarbons or
heterocycles arranged in a linear network. Among the most studied conducting

polymers are polyacetylene, polypyrrole, and polythiophene.

Based on the electrical conductivity, any common material can be viewed
as belonging to one of the three groups: metal, semiconductor, or insulator
(Figure 1)“. The conductivity of typical metals such as copper is 106 S cm"1 at
room temperature. The conductivity is generally expressed in S cm’1 which is
equivalent to ohm‘1 cm". In terms of band theory, metal is associated with a
partially filled valence band and an empty conduction band. Conduction arises

from the availability of a large number of electrons in the valence band which can

serve as charge carriers and move easily to the close empty band. In metals, the
conductivity increases as temperature decreases; this phenomenon is explained
by the fact that, at low temperatures, few electrons are scattered by lattice
vibrations (called phonons) and, thus, are not perturbed in their conductive

movement.

The conductivity of semiconductors ranges between 10‘5 S cm'1 and
102 S cm". The valence band is completely filled, the conduction band is
completely empty, and the band gap between them is relatively small. In this
case, there are fewer electrons around to serve as charge carriers and the
number of available carriers is strongly dependent on temperature. At higher
temperatures, more electrons can be promoted into the conduction band and

hence conductivity increases with increasing temperature.

Insulators such as Teflon and polystyrene have conductivity less than
10'6 S cm". The only difference between insulators and semiconductors is that
the band gap between the valence and the conduction band is larger in the
insulator group preventing the movement of the electrons from the valence band

to the conduction band.

Neutral or undoped polymers are usually classifiedas insulators due to
their very low conductivity. However, when they become partially oxidized or
reduced, their conductivity extends over a wide range from insulators all the way
to metals. Usually, the conductivity increaSes by several orders of magnitude
upon doping. Recently, a group of researchers at BASF, West Germany,
reported the synthesis of oxidized polyacetylene which exhibits a conductivity of

2.5 105 S cm", comparable to that of copper.

Figure 1. Room temperature electrical conductivity (S cm'1) of common

materials. The vertical axis represents the conductivity scale.

 

 

 

 

 

 

 

_ scraper i .. 106
silver ,.polyacetylene .—
METALS- bismuth "
-TTF-TCNQ
_ ._ olypyrrole
_ fiolythiophene
vpolymetallo-
“ phthalocyanine
SEMI- - . DOPED
germanium «p '
CONDUCTORS silicon ‘ ORGANIC
{' POLYMERS
F— boron
.. l,
+-
polyvinyl +
INSULATORSd chloride
4..
T
_ teflon .ilD-la
polystyrene

FIGURE 1

Charge-transfer salts have typical conductivity ranging from 101 S cm'1
to 105 S cm'1 which is the conductivity in the semiconducting and metallic
regime. Both charge-transfer salts and conducting polymers undergo a metal
insulator transition. Conductivity reaches a maximum at some temperature
usually less than room temperature. Above this point, the temperature
dependence is metallic, that is, the conductivity increases as temperature
decreases. Below it, the temperature dependence is that of a semiconductor, that

is, the conductivity increases with increasing temperature.

The charge-transfer salts and the conducting polymers have several
features in common. They both exhibit anisotropic or quasi-one-dimensional
propertiesG. Electrical conductivity is much greater in one direction than in the
others. In charge-transfer salts, the conductivity is greatest along the stacking
axis of the molecule whereas in conducting polymers the conductivity is greatest
along the chain direction of the polymer. Another common feature of these
materials is that they are ionic in the conducting state. In the case of
charge-transfer salts, charge is transferred between the donor and acceptor
components of the complex and in the case of conducting polymers, charge is
transferred between the polymer chain and a dopant (chemical oxidizing or
reducing agent, or electrode). Along with fractional or partial oxidation state, the
polymers must possess a band-type electronic structure formed by closely
arranged adjacent molecules which interact with each other in order to be

electrically conducting.

The band formation is described by the tight binding theory analysis-
analogous to the Huckel theory for an infinite linear polyene‘9t4ao7. When two

singly occupied 1: atomic orbitals are allowed to interact in a single molecule, two

molecular orbitals are formed: a rt-bonding orbital, completely filled, of lower
energy and a II'-antibonding orbital, completely empty, of higher
energy (Figure 2). If the rr-bonding and f-antibonding orbitals of two molecules,
stacked directly one above the other, are allowed to interact, two sets of two
molecular orbitals are formed and separated in energy by 2t, where t is the
transfer integral (analogous to B in the Huckel theory). If N molecules are allowed
to interact, there will be Nrr states in the highest occupied molecular orbital
(HOMO) band and Nu’ states in the lowest unoccupied molecular orbital (LUMO)
band. A band is formed when the energy difference between the states remains
small. The HOMO band, usually called the valence band, is entirely filled and the
LUMO band, or conduction band, is entirely empty. Into each of these bands 2N
electrons can be placed. A material with a full valence band and an empty
conduction band will be an insulator, or possibly a semiconductor depending on
the size of the gap separating the two bands. The bandwidth of a band,
represented by 41, describes the extent of interaction between adjacent
molecules in the stack. The second property that the charge-transfer salts and
the conducting polymers must have is a fractional oxidation state“. In other
words, they must have depleted bands. This is analogous to the requirement that
the highest occupied molecular orbital of a metal be partially filled. The partial
band-filling can be achieved in several ways. In charge-transfer salts, charge
(electrons) is transferred between the donor and the acceptor molecules.
Conducting polymers can be doped either chemically or electrochemically.
Theoretical and experimental Studies dealing with electronic structure have
provided information on band formation and energy7a-9. However, few studies
have dealt with the formation of a fractional oxidation state. The success of a

partial oxidation state formation depends on a complex and largely uncontrollable

Figure 2. Process leading to the formation of an energetic band.

metflZm
I
f’4~fi

Molecular
orbitals

 

ZB

 

Single
molecule

FIGURE 2

 

 

 

’q"'-“'ll

F—nr-fl' 26' 2t 4t:
\‘uL—A. V

LUMO band

 

HOMO band

 

Stack of Stack of
two many

molecules molecules

variety of factors such as Van der Waals forces, repulsion interactions, ionization

potential-electron affinity, and the kinetics of crystallization.

Charge-transfer salts are usually obtained by induced charge-transfer
between donor and acceptor molecules‘. The donors and acceptors are usually
large planar molecules that can stack on top of each other in a segregated
arrangement. The band formation is due to the overlap of it orbitals of the
molecules in the stack. The segregated arrangement allows the transferred
charge to move easily along the separate donor and acceptor stacks.
Charge-transfer salts must crystallize in a segregated array without any defect to
be highly conducting. Structure defects or distortions can lead to a metal insulator
phase transition, resulting in insulating properties at low temperatures. The
band-filling is determined by the extent of charge-transfer from donor to acceptor
molecules. The degree of band-filling depends on the properties of both donors
and acceptors. The fractional oxidation state can be achieved using chemical or

electrochemical dopants.

Conducting polymers differ from charge-transfer salts in several
waysior". They are chemically more stable and easy to process. In general,
conducting polymers refer to conjugated unsaturated hydrocarbon and
heterocycle systems. They are usually obtained by a chemical coupling reaction
or by electrochemically initiated polymerization of the corresponding monomer.
Most of them possess a band-type electronic stnlcture due to the overlap of
Ir molecular orbitals on adjacent molecules. However, they do not have a
fractional oxidation state in the neutral form. Insulators or semiconductors in the
neutral form, they become electrically conducting upon doping. Doping can be

performed by chemical or electrochemical techniques. In both cases, charge is

10

transferred between the polymeric cation (anion) and a counterion which is the
oxidizing (reducing) form of the reducing (oxidizing) agent5. rr-Conjugated
polymers are widely used because they can be easily oxidized (reduced) and
form a stable cation (anion) rather than undergoing other chemistry. Besides, in
their systems, the Ir-electrons can be easily removed or added without breaking
the 0 bonds. which are primarily responsible for holding the polymer together.
Mild oxidizing and reducing agents such as iodine and lithium are utilized to
ensure that only a charge-transfer reaction occurs and no other chemical
reaction”. The dopant usually perturbs the polymer extensively. Because of its
large size which does allow it to fit into the polymer lattice and because of the
extensive charge-transfer with the polymer chains, the dopant can cause
geometrical changes in the chain. The doping level is small and limited to the
concentration and strength of the chemical dopant. The chemically doped
polymers are generally insoluble which prevents their direct physicochemical
characterization and their electronic properties from being studied“. Doping also
can be performed electrochemically. With electrochemical techniques, synthesis
and doping occur simultaneously. The monomer undergoes polymerization by
applying an oxidation potential at the electrode surface. The resulting polymer is
in its oxidized (cationic) form with counterion derived from the electrolyte solution.
Compared with chemical doping, electrochemical doping is cleaner with fewer
interferring chemical reactions. It is also more precise, enabling the conductivity
of systems to be switched between the conducting and insulating state by simply
applying the appropriate voltage. I-"Inally, electrochemical doping is more
informative. It has been used to study primarily thin films of material coated on an
electrode surface“. Thermodynamic and kinetic data can be obtained from these
studies. However, the redox behavior of modified electrodes can be influenced by

counterion diffusion within the film, as well as material discontinuities and

ll

nonuniformities‘5. 'Also, it is possible to separate Faradaic and capacitive
components of current. Electrochemical doping performed on slurries of insoluble
materials are time consuming, do not provide any information on the time
dependent-doping process, and yield only macroscopic data regarding the
charge-transfer. The observed response is complicated by variations in particle
size and structural changes accompanying the doping process”. To eliminate
such complications regarding the redox process of these materials, it is desirable
to have soluble materials whose redox properties can be studied in solution by

conventional electrochemical and spectroscopic techniques.

The synthesis of such soluble, conducting materials is currently an area
of great interest. It is motivated by the processibility properties required in
technological applications. Recently, soluble polymeric metallophthalo
cyanines‘7P‘r and soluble polythiophene derivatives1 73'0 have been prepared. In
both cases, solubility is insured by the addition of organic functionalities to the

polymer backbone.

Hanack and coworkers at the University of Tubingen, West Germany,
have recently synthesized soluble tetraalkylsubstituted cofacially joined,
Group IV B phthalocyaninato macromolecules [lR4PcMO]n where M = Si, Ge, Sn
Pc= phthalocyaninato, and R = tetra-t-butyl or tetramethylsilyl (Figure 3)‘7qu.
The presence of the bulky substituents give rise to high solubilities in common
organic solvents such as methylene chloride, benzene, acetone, and
tetrahydrofuran. Metallophthalocyanine polymers are very interesting materials
because they combine the characteristics of both charge-transfer salts and

conducting polymers. Similar to charge-transfer salts, the large planar

12

Figure 3. Schematic sthcture of peripherally-alkylated u-oxopolymers
[Fi4PcMo]n M = Si, Ge, Sn; Pc=phthalocyaninato; Fl = t-Bu4,

Tms.

13

 

FIGURE 3

l4

phthalocyanine molecules are stacked on top of each other with small Pc-Pc
lnterspacing distance allowing interactions between molecular orbitals which
leads to the formation of a band-type electronic structure. Similar to conducting
polymers, the phthalocyanine rings are held together in a rigid cofacial
orientation. These macrocyclic metal complexes have been bridged to a large
variety of ligands‘7q-‘3. Pyrazine, tetrazine, cyanide, and isothiocyanate groups
have been used to bridge Fe, Ru, Co, and Rh macrocyclic transition metal
complexes. Other ligands such as oxygen, sulfur, alkylnyl have been used to
bridge Si, Ge, and Sn phthalocyanines.

The particular substituted peripherally alkylated p-oxopolymers (oxygen-
metal bridging) are essentially identical to the unsubstituted analogs whose
electric, optical, and magnetic properties have been extensively studied in the
solid state by Marks and his group1 6'19. Unsubstituted and substituted
p-oxopolymers with M = Si, Ge have been chemically oxidized using halogens
such as iodine and bromine‘7P-rr193‘9. For example, the partial oxidation of
unsubstituted Si and Ge phthalocyanines was accomplished by stirring a mixture
of polymer and the halogen in benzene for two days. In contrast for M = Sn,
doping resulted in destruction of the tin-oxygen bond. This different behavior
towards doping could be explained because tin is significantly larger than Si and
Ge, and did not fit without distortion into the hole of the phthalocyanine molecule,
thus weakening the tin-oxygen bond. Unsubstituted and substituted Si
phthalocyanine polymers also can be oxidized with nitrosonium salts NO+X'
where X‘ = BF4’, PFG', and SbF6’199. X-ray diffraction studies suggested that
the chemical doping was largely inhomogeneous and that there was a change in
crystal structure from orthorhombic to tetragonal upon doping. The nature of the

iodine-doped II-oxopolymer‘s electronic stnicture was probed by electron spin ‘

15

resonance (ESR). The symmetry of the Iineshapes and the g-values were
consistent with it-radical cations, i.e., the unpaired spin density was located in the
molecular orbitals which is predominantly ligand in character. The broad
electronic absorption spectra were also indicative of an oxidation process
occurring at the macrocycle level and the formation of a radical cation. Four
probe Van der Pawl electrical conductivity measurements on the unsubstituted
[MPcO]n powders showed them to be insulators. However, iodine or bromine
doping resulted in substantial increase in conductivity, an increase of several
orders of magnitude leading to the semiconducting or even metallic state. For
example, the conductivity of [PcSiO]n doped to its maximum with iodine
increased from 5.5 10'6 S cm'1 to 1.4 S cm". Conductivity of [PcGeO]n and
[PcSnO]n doped with iodine also increased, respectively, from 2.2 10‘10 S cm‘1
to 1.1 Scm‘1 and from 1.2 10'9 S cm'1 to 6.5 10‘7 S cm". Substituted
II-oxopolymers follow the same trend of increasing conductivity upon chemical
doping (Figure 5). The conductivity of substituted Si, Ga, and Sn phthalocyanine
polymers doped with iodine increased for M = Si from 8.0 S cm‘1 to 2.0 S cm",
for M = Ge from 6.0 10'11 S cm“ to 1.0 10'3 S cm" and from 4.0 10'12 S cm'1
to 1.0 10'6 S cm'1 for M =- Sn. Compared with unsubstituted polymers, the
electrical conductivity values of substituted polymer analogs were very similar
before doping and slightly lower upon doping. The general trend in conductivity
as function of metal is OSiZOGe>OSn. Since the doping process was largely
ligand dominated and relatively insensitive to the identity of the metal, the metal
dependence of the conductivity observed in these polymers was logically
ascribed to stnrctural differences such as how the interplanar separation was
I influenced by metal ionic radius. The conductivity of these materials increased
continuously as the iodine content increased indicating that, at higher partial

oxidation of the polymer, the number of charge carriers increased. However,

16

Table 1. Pressed power room temperature electrical conductivity data for
doped and undoped II-oxopolymers.

Table 2. Pressed powder room temperature electrical conductivity data for

doped and undoped alkyl substituted p—oxopolymers.

17

TABLE 1

 

compoono “at (s cm-l,
[PcSi01n 5.5 10‘6
[(PcSiO)I0.31]n 1.4

[PcGeO]n 2.2 10-10
[(PcGeO)Il.08]n 1.1 10‘1
IPcSnoin 1.2 10‘9
[(PcSnO)IO.35]n 6.5 10’7

IABLI 2

COMPOUND

 

 

[t-BqucSiO]n
[It-Bu4PcSiO)IO.4]n
[t-Bu4PcGeO]n
[(t-Bu4PcGeOlIo.4]n
[t-Bu4PcSnO]n

[It-Bu4PcSnO)Io.9]n

 

8 o 10‘8
2 o 10‘3
6 o 10‘11
1 o 10‘3
4 o 10‘12
1 o 10‘6

18

independent of the metal and the chemical oxidizing power, and the amount of
oxidizing agent added in large excess, the band-filling of these polymers did not
exceed 35%. In other words, it was possible to vary the amount of band-filling up
to 35% by adjusting the amount of single dopant, but it was not possible to
chemically oxidize a polymer from 0 to 100% and tune its properties as function

of band-filling.

To get better control over band-filling and to eliminate the problems
associated with solid state discussed earlier, unsubstituted Si
u-oxophthalocyaninatopolymers have been doped electrochemically“.
Electrochemical doping was performed as slurries. The finely ground polymer
was oxidized in a three-compartment electrochemical cell (Figure 4) while
suspended in a stirred electrochemical solution. Contrary to chemical doping,
electrochemical doping is homogeneous. Initially, the crystal structure of the Si
polymer was orthorhombic. After partial oxidation at 50%, the crystal structure
changed from orthorhombic to tetragonal. Once it was in tetragonal structure, the
polymer can be further oxidized, reversibly reduced, and cycled a number of
times between oxidized and reduced states with no decomposition and no
structural change. It was noted that the oxidation of the tetragonal phase required
less energy than the initial orthorhombic phase. However, attempt of
electrochemical doping of [GePcO]n oxidatively was unsuccessful. As for
chemical doping, electrochemical doping was also accompanied by an increase
in conductivity. The magnitude in conductivity observed at a given doping level
was roughly comparable for the electrochemically and chemically doped Si
polymers. The principal difference was that the level of partial oxidation
accessible for the electrochemically doped polymer was larger. The highest

degree of partial oxidation level, 67%, was obtained with p-toluenesulfonate

19

counterion (TOS')15d. The highest degree of partial oxidation achieved in this
electrochemically doped polymer appeared to be a beneficial consequence of the
unlimited oxidizing power of the potentiostat. Electronic, optical, and magnetic
studies of [SiPcO(BF4)y]n material revealed an evolution in properties from
insulator/semiconductor to metal-like with increasing doping level. With
electrochemical doping, it was thus proven that band-filling could be varied over a
broad range of compositions and although all complications associated with solid
state remained, the results on these systems were very encouraging because it
was the first time that such band-filling tunability has been achieved to a major
extent for a conventional conductor. Along with substituted peripherally-alkylated
It-oxopolymers previously described in detail, other soluble phthalocyanine
polymers such as substituted peripherally-alkylated phthalocyaninato transition
metal complexes [t-Bu4PcRudib]n and [t-Bu4PcRuMe4dibjn, where
dib = p diisocyanobenzene and Me4dib = tetramethyldiisocyanobenzene, have
been synthesized recently and characterized by spectroscopic techniques; but up
to now, no doping has yet been performed on either these substituted or

unsubstituted analog materials.

Recently, several groups have synthesized soluble polythiophenes‘7a'0.
Like polyphthalocyanines, significant interactions exist between the redox sites of
the polymers. However, these are primarily through-bond overlap of the
It systems instead of through-space orbital overlap as in polyphthalocyanines. In
all these soluble thiophene polymers, solubility was achieved through addition of
long linear alkyl chains ranging from butyl to dodecyl, at the 3-position. Blocking
the 3-position prevents crosslinking polymerization and leads to a more
structurally regular polymer by increasing the amount of or or' coupling in the

polymer. Poly(3-alkylthiophenes) can be synthesized chemically or

20

electrochemically. Chemical synthesis is accomplished by oxidative
polymerization of appropriately 3-substituted thiophene using transition metal
halides such as FeCl3 or MoCl5203. Another synthetic route involves the
preparation of 3-substituted 2,5-diiodothiophene followed by Ni-catalyzed
Grignard coupling‘7dr205'd. More often, these polymers are synthesized through
oxidative electrochemical polymerization of their corresponding monomers by
applying a constant current or potential at the anode, leading to the formation of a
film at the electrode surface‘3av‘7‘v20r21. Polymerization occurs via a cation-
radical coupling mechanism that involves dehydrogenation on the 2- and
5-positions of thiophene oligomers. The resulting film is obtained in Its oxidized
form and the overall charge balance is achieved by incorporating an anion
derived from the electrolyte. Because the anions are directly incorporated into the
film, the choice of the anion is important in determining the quality of the film.
Best results were found with the inert, small anions PF6' and BF4'133. The
oxidized polymer can then be reversibly undoped (and subsequently re-doped)
by electrochemical or chemical reduction (re-oxidation). In contrast, the
chemically synthesized polymers are produced in their undoped insulating state
and can be doped chemically or electrochemically to their conducting states.
These polymers are chemically stable in both their neutral and doped states. The
maximum doping level encountered in these polymers was 36% which could be
achieved either by chemical or electrochemical methods”. Upon doping, soluble
poly(3—hexylthiophene) exhibits relatively high conductivity compared with the
insoluble parent polythiophene”. The electrical conductivity can be as high as
95 S cm‘1 for PF6' doped poly(3-hexylthiophene). To obtain such value, the
polymer must be synthesized in a three-compartment oxygen- and moisture-free
. electrochemical cell. Lower conductivity was obtained for chemically synthesized

polymers. The presence of B-substituent has little influence on the conductivity

21

but allows the polymer to remain soluble in its doped form. Since these
polythiophene derivatives are soluble in common organic solvents in both their
neutral and conducting forms, the electronic, optical, and magnetic properties of
these polymers in the undoped and doped states as well as during the daping
process have been studied. In particular, ESR measurements and optical data on
the oxidized polymer in solution helped in elucidating the nature of the charge
storage configurations. The results indicated that doping induced charged kinks
or excitations (polarons and bipolarons). This situation is mainly due to a large
coupling between the lattice deformation and the electronic excitations. The
charges transferred from the dopant molecule produce a local deformation of the
polymer chain which varies the bond lengths. As a result, this distortion affects
the electronic structure of the polymer and leads to the formation of bipolarons
(dications) which are spinless species. In the formation of bipolarons, the valence
band remains full while the conduction band remains empty and low energy
states are created in the gap between the two bands. Another important feature
of poly(3-alkylthiophenes) is that the electronic, optical, and magnetic properties
of these conducting systems can be studied both in solution and in the solid state
on chemically modified electrodes. Thus, the influence of the medium on such

properties resulting from deping can be determined.

It has been already shown that, by adding an alkyl substituent at the
3-position, the amount of or or' coupling in the polymer increases thereby
producing a better film quality without significant decrease in conductivity.
Another way to enhance the selectivity of or a' coupling consists in using
chemically synthesized oligomers (dimer, trimer, etc) instead of a thiophene
monomer as starting materials for the electropolymerlzation. Studies on

electrochemically synthesized polymers derived from thiophene, bithiophene and

22

terthiophene revealed that it is easier to oxidize the material as the chain length
of the starting oligomers increased, but unexpectedly the ease of oxidation was
accompanied by a large decrease in conductivity“. These results were
interpreted by the decreasing reactivity of the radical cations derived from these
three substrates. No result has yet been reported on the study of the soluble alkyl
substituted analogs, probably due to difficulties encountered in synthesizing

these oligomers.

Following a different approach to force polymerization of thiophene in the
2- and 5-position, Elsenbauer and coworkers have used dialkyl thiophenes to
chemically synthesize soluble polydialkylthiophenes. The alkyl groups in position
3 and 4 serve two important functions. They increase greatly the solubility of
these materials and they eliminate all future sites for polymerization
(or B linkages). These materials should be able to undergo oxidation and
reduction with minimum interference from side reactions. All studies conducted
on doped polydialkylthiophenes should then be the result of the doping process

only.

In this study we will take advantage of the good solubility and chemical
stability of the polyalkylated phthalocyanines and thiophenes to study the redox
and physicochemical properties of conducting materials by a variety of
conventional analytical techniques. These methods will allow us to provide
information concerning the properties of these materials as function of
band-filling. These conducting materials will be characterized to a sophisticated
level, never attempted before. This will greatly help in searching for a rational

preparation of materials with specific, predictable, and tunable properties.

2. EXPERIMENTAL

2.1 Glassware
2.1.1 Vacuum lines
High vacuum line
Prior to begining of the electrochemical experiments, the solvents were
transferred into the electrochemical cells via high vacuum line (10'5 torr). The
cells were then backfilled with argon (Matheson 99.95%) to maintain an air- and
moisture-free atmosphere throughout the experiments. The argon was previously
passed through a drying column (magnesium on silica) to remove residual
oxygen and moisture. Between experiments, the solvents were stored under high

vacuum to preserve their high quality.

Mills.

Reactions requiring an air- and moisture-free environment were

performed using standard Schlenk line techniques”.

2.1.2 Electrochemical cells

Different glass, electrochemical cells that are connected directly to a high
vacuum line were constructed and used in this study. Controlled potential
coulometry (Bulk electrolysis) was carried out in a three-compartment cell25. The
counter electrode, a platinum gauze, was isolated from the working electrode
compartment by two fine porosity glass frits. The reference electrode, a silver
wire, was separated from the working electrode by a fine porosity glass frit. The
working electrode was an 8 cm2 platinum gauze electrode. A 0.018 or“2 platinum
disk working electrode (Bioanalytical Systems Inc) also was placed in the working

23

24

electrode compartment to allow voltammetric measurements to be performed
under the same conditions as controlled potential coulometry experiments. The
platinum disk electrode was polished with 0.05 pm Al203 before use (Buekler).
Diffusional voltammetry experiments (cyclic and differential pulse voltammetry)
were carried out in a Single compartment cell. A 0.018 cm2 platinum disk working
electrode, a platinum wire counter electrode, and a silver wire reference
electrode were used in these cells. Arrays of 5 individually platinum wire working
electrodes (0.5 cm long) separated from each other by 80 pm Teflon insulating

spacer were used in the conductivity measuring experiment.

An indium-tin oxide (ITO) conducting glass working electrode, a platinum
flag reference electrode, and a platinum. wire counter electrode were used in the

optical characterization experiment.

Hydrodynamic voltammetry experiments (rotating disk and rotating
ring-disk voltammetry) were performed in a three-compartment cell carefully
. designed to reduce the formation of a vortex”. A silver wire reference electrode
‘ and a platinum gauze counter electrode were placed in sidearm compartments
separated from the working electrode compartment by fine porosity glass frits.
The working electrodes (Pine instruments) were either the platinum rotating disk
electrode with a radius of 0.256 cm or the rotating ring-disk electrode comprising
a platinum disk and a platinum ring with disk radius=0.228 cm, inner ring
radius=0.246 cm, and outer ring radius=0.269 cm, and a collection efficiency (N)
of 23%. This collection efficiency value was determined from the rotating ring-
disk voltammogram of the standard Fe(CN)63‘/Fe(CN)64' redox couple and

confirmed by digital simulation.

25

2.2 Methods of purification
2.2.1 Solvents
Acetic anhydride and pyridine were dried over calcium hydride and
purified by distillation under nitrogen. Methanol was dried with magnesium
tumings and purified by distillation under nitrogen. Quinoline was dried over
barium oxide for two days and then decanted. The decanted quinoline was

vacuum distilled and stored over activated 3A molecular sieves.

Nitromethane (EM Science) was dried over phosphorous pentoxide and
purified by distillation under nitrogen. The resulting solution was then degassed
on the vacuum line by employing several freeze-pump-thaw cycles and then
vacuum-transferred to a greaseless round bottom flask containing activated 3A
molecular slaves and stored on the high vacuum line. Tetrahydrofuran
(HPLC grade, Burdick and Jackson) was dried by addition of a sodium-potassium
alloy and the same procedure as for nitromethane was used to degas. transfer
and store the solvent on the high vacuum line. Acetonitrile (HPLC grade, Burdick
and Jackson) was dried over calcium hydride and then decanted. The decanted
solution was degassed, vacuum-transferred, and stored on the high vacuum line

as above.

2.2.2 Supporting electrolyte
Tetrabutylammonium fiuoroborate (electrometric grade, Southwestern
Analytical Chemicals) was recrystallized from ethyl acetate/diethyl ether

(5:1 volume ratio, 80 0C) and vacuum dried for three days.

26

2.3 Instrumentation
2.3.1 Chromatographic

The molecular weight of the soluble poly(3,4-dibutylthiophene) was
determined by gel permeation chromatography (GPC) using a Waters 840
instrument equipped with an exclusion column in which polystyrene-linked
material was loaded. The elution time was measured using 10‘2 M polymer
solution in tetrahydrofuran by monitoring the refractive index difference between
the polymer solution and the solvent itself. The molecular weight was determined
from the retention time calibration using polystyrene standards whose molecular

weights ranged from 2 103 to 2 106.

2.3.2 Spectroscopic

Proton nuclear magnetic resonance spectra were obtained on either a
Bruker WM-250 (250 MHz), Varian Gemini (300 MHz) or Varian VXR-300
(300 MHz) spectrometers. Chemical shifts are reported in parts per million (6)
using the proton resonance of the residual solvent as the internal reference
(acetone, 8:2.04; chloroform, 6:7.24: tetrahydrofuran, 5=1.73). Fourier-
transforrned infrared spectra were obtained on a Nicolet 740 spectrophotometer
either as Nujol mulls using KBr plates or KBr pressed pellets. Routine UV-Vis
absorption spectra were recorded on a Cary 17 using 1-cm matched quartz cells.
For in-situ spectroelectrochemical experiments, in the solid state, a 1*1*4 cm3
quartz cuvette was fitted with an indium-tin oxide transparent electrode which
functioned as the working electrode, a platinum flag as counter electrode, and a
silver wire as reference electrode. The cell was fitted with a tight Teflon top and
maintained under positive argon pressure during the measurements. In solution,
a simple 1-cm matched quartz cell was used. All UV-Vis NIR spectra were

performed on a Cary 2300.

27

2.3.3 Electrochemical

A Princeton Applied Research (PAR) 273 potentiostat/galvanostat
interfaced to a Zenith 248AT computer with a National Instrument
PC 2A lEEE-48 card for instnrment control and data acquisition was used for the
bulk electrolysis and CV and DPV voltammetry experiments. Experimental results
were recorded on a Hewlett-Packard 7475A plotter. A bioanalytical system 100A
potentiostat. interfaced to a Zenith 158XT computer and a Hewlett-Packard
Colorpro plotter was alternatively used for bulk electrolysis and voltammetry
experiments. A Pine lnstmment AFRDE4 bipotentiostat coupled with a AFMSR
rotator was used for rotating disk voltammetry experiments. A platinum/Teflon
rotating disk and ring-disk electrode, also constructed by Pine Instrument
company. served as working electrode. The voltammograms were recorded on a
Soltec X-Y or x-v1-v2 recorders. The film thicknesses were determined with a
Dektak 11A profiler, and dc electrical conductivities (0) were measured in-situ by
a standard four probe technique at room temperature”. Values of o (S cm")
were calculated from the relation :

c=P/LRe 0)

where :

L = length of a platinum stripping (0.5 cm)

P = gap between two consecutive strippings (80 pm)

,9 a film thickness in pm .

R =- measured resistance in ohm
All electrochemical experiments were performed either in oxygen- and moisture-
free specially designed cells (as described in section 2.1.2) or inside a vacuum
drybox.

28

2.4 Electrochemical techniques
2.4.1 Controlled potential coulometry

Bulk electrolysis or controlled potential coulometry (CPC) was used to
convert compounds from one oxidation state to another”. In a CPC experiment,
the potential of the working electrode is applied to a value such that complete
electrolysis of the desired species occurs. The total charge passed during the
electrolysis is obtained by integrating current over time. Rapid stirring of the
solution was maintained during the course of the experiment to ensure efficient
mass transfer of material to the working electrode. A large surface area of the
working electrode as well as minimum solvent were employed to minimize
electrolysis time. Supporting electrolyte (0.1-0.2 M TBABF4) was placed in each
compartment of the CPC cell (Figure4) and 004-1 mM of the polymer or
09-25 mM of the corresponding repeat unit (monomer or oligomer) were added
only to the working electrode compartment. The cell was then connected to a
high vacuum line and evacuated (10’5 torr). After cooling it with an
isopropanol/dry ice bath, a known amount of solvent was vacuum transferred.
The cell was then backfilled with argon and allowed to warm up to room
temperature prior to use. Usually CPC is employed to determine the number of
moles of material electrolyzed (N) or the number of electrons involved in the
electrolysis process (n) from the amount of charge passed through an
electrochemical cell during electrolysis (Q). Faraday's law correlates these three
variables by the equation :

Q=nFN Q)

where :

Q = charge transferred

n = number of electrons

N = moles of material electrolyzed

29

Figure 4. Electrochemical cell used in controlled potential coulometry
experiment.
A = platinum gauze counter electrode
B = fine porosity glass frits
C = silver wire reference electrode
D = platinum gauze working electrode for CPC
E = platinum disk working electrode for CV and DPV

 

 

 

 

30

 

 

 

 

 

 

 

 

 

 

 

 

 

VACUUM
/\
I“ ‘
’ I
\
’I
‘N
I -_____‘ \
T \
l 1’ n k
l I ‘ \
I / \ \
I \ \
m 4 | S
/ \I r \
I Y I I
I\ I ‘ I
say \vl
.
B B QC

 

 

 

 

l'IGITRII 4

31

The amount of charge used in equation (2) is the total measured charge
subtracted from the charge due to the background current. In this study, CPC
was used mainly to determine the degree of partial reduction as function of
potential. This technique was employed since it is an absolute method for
determining oxidation states which depend only on the potential of the working
electrode. The degree of partial charge transfer (partial oxidation or reduction) is

 

calculated from the equation:

% of charge

charge a $100 (3)
transfer mass of polymer/gfw of repeat unit)*F

A value of 100% represents the addition of one electron per repeat unit. In this
expression no particular degree of polymerization is assumed, but a simple
relationship exists between the amount of charge passed and the number of
moles of redox sites or repeat units. From the shape of the plots of normalized
currents versus time, information on chemical stability and mechanism during

electrolysis was also obtained.

2.4.2 Voltammetry
Mvoltammetrv

One of the most widely used techniques in electrochemistry is cyclic
voltammetry (CV)29. It Is attractive for its ability to diagnose reaction
intermediates which undergo further chemical reactions and to study the products
and kinetics of these reactions. During CV, the potential of the working electrode
is scanned from some initial potential to a switching potential value and scanned
back to the same initial potential value at a constant sweep rate. In a CV
experiment, for the study of the It-oxo-(tetra-t-butylphthalocyaninato)germanium,
0.1-0.2M of the supporting electrolyte, TBABF4, was placed in a single

32

compartment cell. The polymer (10—40 IIM) was placed in a "dumpster”, a
rotatable reservoir allowing the addition of the polymer in the cell at any time.
With the dumpster, voltammograms of only the blank solution could be obtained
and were subtracted from the sample voltammograms to obtain background-
corrected voltammograms. The background voltammograms also served in
determining the potential range from the solvent which also was used for the
entire experiment. From the current-potential curve response, several variables
of the system studied can be determined. The number of electrons (n) transferred
during the electrode reaction, for a reversible couple, can be calculated from the
separation between anodic and cathodic peak potentials29. In the study of the
polydialkylthiophenes, cyclic voltammetry was used to prepare the polymers.
Polymerizaton was carried out at 5 0C in electrolytic solutions of nitromethane or
acetonitrile containing 0.2 M supporting electrolyte (TBABF4). Under these
conditions, a polymer film was deposited on the ITO electrode surface. Good film
quality and high conductivity were achieved by sweeping the electrode potential
for a few minutes at a sweep rate of 50 mV sec‘1 which corresponds to few
micron thickness. As synthesized, the polymer was in its oxidized form. Neutral
(or undoped) polymer was easily obtained by reversing the direction of the
potential sweep. In order to increase the amount of polymer formed,
polymerization was performed in the CPC cell using big platinum gauze electrode
(10 cm2) as working and counter electrodes. The chemical stability of reactants
and products and kinetic information regarding the electron transfer reaction also
can be evaluated by changing the sweep rate. In the study of the p-oxo-(tetra-t-
butylphthalocyaninato)germanium, CV was mainly used to probe the products of
the redox-induced reaction and to study their chemical stability. CV also was

used to determine n.

33

Rotating disk and rotating ring-gisk voltammetry
Rotating disk (RDE) and rotating ring-disk (RRDE) voltammetry are

among the most useful techniques for studying electrode reactions. In RDE, the
working electrode is connected to a motor which spins the electrode and causes
the solution to move to the electrode surface by convective mass transport.
Unlike most voltammetric techniques, which are performed in motionless solution
and where diffusion is the main mode of mass transport, RDE is performed in a
continuously stirred solution and convection is favoured over diffusion due to a
greater mass transport rat929. Because the electrode moves constantly with
respect to the solution and at a rapid rotation rate, fresh solution is continuously
brought up to the electrode surface. The rotation of the working electrode and a
slow potential scan rate (10 mV sec“) allow these experiments to be performed
under steady state. In a steady state experiment, the current at a given potential
is independent of both the scan direction and time. Consequently, preelectrolysis,
which is a problem encountered when using the DPV technique, does not occur.
In this study, the current-potential response of the RDE voltammograms for the
It-Oxo-(tetra-t-butylphthalocyaninato)gerrnanium was used to obtain the deping
profile of the germanium polymer as function of potential. In RDE voltammetry,
the current at any potential is given by the equation :

I = 0.62 n F A 002/3 v‘1/6 (01/2 [co' - Co(y=0)] (4)

where :

0) = angular velocity of the disk electrode

((0 = 2 It I with f = rotation rate)
v a kinematic viscosity of the solvent
(v = h Id with d =- density)

Co' = bulk concentration of the electroactive material

34

Co(y=0) a concentration of the electroactive material at the electrode

surface
Do a diffusion coefficient of the material
A a area of the disk electrode (0.163 cm2)
n = number of electrons

Using equation (4), the current at the potential corresponding to 100% reduction,
iI, is given by the equation :

i. =- 0.62 F A 002/3 v'“6 tel/2 00* (5)
Dividing equation (4) by equation (5) yields to equation (6) which gives rise to the
expression of the fractional concentration of the material undergoing electron
transfer reaction.

i 00' - Coly=0) . (6)

 

*

i. Co
Derived from equation (6), the percent reduction at any potential is given by the
equation:

% reduction = i / i. * 100 (7)
The RDE technique also was used to determine if there was any kinetic limitation
accompanying the doping process. For a totally reversible system, the plot of i vs
(01/2 is predicted to be linear with zero intercept as expected from the
current-potential curve at the rotating disk electrode. Any deviation from linearity
suggests some kinetic step involved in the electron transfer reaction. Such
behavior is observed for an electrode reaction with slow kinetics, i.e., the rate of

electron transfer is sufficiently slow to actas a limiting factor.

Diffusion coefficients of both the neutral germanium polymer and the

monomer were also estimated from the current-potential curves of the RDE

35

voltammogram and used in the calculation of the degree of polymerization of the
p-oxo-(tetra-t-butylphthalocyaninato)germanium material. The relationship
between molecular weight (M) and diffusion coefficient (D) for macromolecules is
well known“. For coil-shaped polymers, the diffusion coefficient is predicted to
vary with (molecular weight)'°-55. For thin rod-shaped polymers, D is
proportional to M038. Spectroscopic data of the short length oligomeric silicon
and germanium phthalocyanines (n =- 1 to 4) prove conclusively that these
compounds have a stacked ring rod-like structure in solution. Assuming that the
u-oxo-(tetra-t-butylphthalocyaninato)germanium polymer maintains this rod-like
structure in solution, the molecular weight of the polymer can be calculated from
the equation :
I)p = (Mm / Mp)°-31 0m (8)

where :

Dp, Dm a: diffusion of the polymer and corresponding monomer

Mp, Mm :- molecular weight of the polymer and monomer

(987.90 9 mole'1)

Knowing the molecular weight of the polymer and of its repeat unit, the degree of
polymerization for the soluble germanium polymer can be easily determined. The
RDE technique is very attractive because it provides the information contained in
a series of CPC experiments in a few minutes instead of few hours and without

degradation of the solution.

The rotating ring-disk voltammetry is an extension of the previously
discussed rotating disk voltammetry in that the disk electrode is encircled by a
ring which functions as a second working electrode. The well defined steady
state response of the technique along with the ability to control and examine two

independent electrode processes permitted the quantitative characterization of

36

multi-site redox systems. The stability of electroactive compounds as a function
of potential was easily followed via RRDE by performing continuous
shielding-collection experiments. In RRDE, the disk electrode is poised at a
potential where polymer reduction (oxidation) occurs and the potential of the ring
is scanned from the disk potential towards more positive (negative) values.
Initially, the flux of neutral polymer to the ring is decreased by its consumption at
the disk and the current observed at the ring (iRcv a cathodic current)
“Rat an anodic current) is reduced compared to that observed when no reaction
is occurring at the disk (iRc°)(iRa°). This experimental configuration is the
conventional shielding in RRDE experiment. At ring potentials sufficiently positive
(negative) relative to the disk value, where only oxidation (reduction) of the
reduced (oxidized) polymer back to neutral polymer can occur, the ring current
arises from the collection of disk generated material (an anodic current)
(a cathodic current). Under these conditions, the conventional RRDE experiment
can be conducted. At ring potentials between these two limiting cases,the new
current observed at the ring is composed of both anodic and cathodic
components (both shielding and collection type processes).

For a chemically stable system, the relation between iR, inc, and id is given by

the equation :

. 0-.
N: 'R IR (9)

'd
in, iRO = ring current observed at a specific potential with and

 

without a reaction occurring at the ring

current observed at the disk

6...
II

collection efficiency of the electrode assembly, a constant

2
II

dependent only on electrode geometry.

37

Hence, (iRO-iR) / id will be a constant value regardless of the values of id and RR-
Deviations from N imply chemical complications and can be used to monitor the
potential required to initiate the process. In both RDE and RRDE experiments,
0.2 M supporting electrolyte TBABF4 was placed in a three-compartment cell
(Figure 5). A known amount of the germanium polymer (30-35 uM) was added to
the working electrode compartment. The solvent used in these experiments was
vacuum transferred in a tube following regular procedure described previously
and then pipetted into the cell which is already inside an inert atmosphere
drybox. The rotator along with the bipotentiostat were also placed in drybox

allowing the experiment to be performed in an inert atmosphere.

Differential pulse voltammetgy
Differential pulse voltammetry (DPV) is used to characterize the redox

response of the polymer. In this technique, a small amplitude potential step pulse
is applied to the system and the derivative current is recorded as a function of
potential. The derivative current is comprised of the current measured at the end
of the pulse subtracted from that just before the pulse. The cell and procedure
used in the DPV experiment were identical to the one used for CV studies. Due
to its high sensitivity and very good resolution, DPV has often been used in the
detection of very low concentrations of analyte. In this study, DPV was used for
the fine characterization of the redox response of the germanium polymer and
corresponding monomer because unlike the featureless response observed for
most voltammetric techniques, DPV is extremely sensitive to small changes in
the doping profile29. DPV, taken after each CPC‘ cycle, i.e., oxidation (reduction)
followed by reduction back (re-oxidation) provided information on the stability of
the material on the CPC timescale. The shape and magnitude of the DPV

voltammogram along with other parameters specific to the technique and system

38

Figure 5. Electrochemical cell used in rotating disk and rotating ring-disk
voltammetry
A = platinum gauze counter electrode
B = fine porosity glass frits
C = platinum disk and ring-disk working electrodes
D = Silver wire reference electrode
E = ansh contact

ROTATOR

 

 

39

 

 

 

 

 

 

 

 

 

 

FIGURE 5

4O

employed were finally used in modelling the current-potential DPV response of
the germanium providing information on the electronic interactions between

redox sites within the polymer.

2.4.3 Digital simulation

Digital simulations are numerical methods which have been found very
useful in elucidating the mechanism of complicated redox events. In this study,
digital simulation of differential pulse and rotating ring-disk voltammetry
experiments were undertaken to model the current-potential response of the
germanium polymer. lnforrnation on the electronic interactions of the redox sites
within the polymer was provided by estimating the standard potential of each site.
The current-potential response also provided information on the chain length of
the polymer. The ultimate goal of digital simulation will be to predict the
electrochemical behavior of any conducting polymer from corresponding
oligomers (few unit long). In this model, the polymer can be viewed as an array of
multiple redox sites undergoing sequential reversible Nernstian electron transfer
in equilibrium with one another (see Appendix 1). Both the DPV and RRDE
simulation programs are based on the method of finite differences”. The solution
near the electrode surface is divided into a number of volumes of equal size. A
flow chart illustrating the general procedure employed to simulate DPV
voltammogram is shown in Figure 6. The first step consists in setting up the bulk
concentration of the solution prior the experiment. Initially, the solution is
assumed to contain only one species in a particular reduction state in the first
volume element. The initial bulk concentration value was chosen to be identical
to the experimental one of Figure 22a. Next, steps were taken to set up the
apparent bulk concentration of the solution for the base potential value and the

surface boundary conditions at the working electrode for the pulse potential

41

Figure 6. Flow chart explaining the DPV simulation program.

 

NO

42

 

 

INITIAL ARRAYS
t < 0

 

 

I

i l

 

 

CALCULATE THE BASE
POTENTIAL VALUE

 

 

I

 

 

 

SET UP APPARENT BULK
CONCENTRATION

 

 

I

 

 

 

CALCULATE THE PULSE
POTENTIAL VALUE

 

 

I

 

 

CALCULATE AND ESTABLISH
SURFACE BOUNDARY CONDITIONS

 

 

I

 

 

DIFFUSION OF MATERIAL TO
AND AWAY FROM THE ELECTRODE

 

 

V

 

 

 

CALCULATE CURRENT

 

 

V

 

 

 

 

 

TEST TO SEE IF EXPERIMENT
IS COMPLETE

YES

 

 

FIGURE 6

 

END

 

 

43

value. The ratio of the concentration of reactant over product at the electrode
surface follows the Nemst equation and is a function of electrode potential. The
number of sites in the polymer that are oxidized or reduced depends only on the
standard potential of each site and the potential of the working electrode. From
the diffusion of material to and from the electrode and the number of sites of
polymer being oxidized or reduced, a dimensionless current was calculated that
ends the first interaction. The current is derived from the first Fick's law of
diffusion which relates flux of species to concentration gradient. The program
then goes back to the calculation of the next base potential value and calculates
the new dimensionless current. Continuing the process generates the current
profile of the solution as function of time. The program ends when the final
potential of the experiment is reached. The procedure employed for the RDE
simulation is similar to the DPV one with the difference that the mass transport
equation is comprised of a diffusion and a convection term. These procedures
are essentially identical to those used to model single charge transfer systems.
The main difference between the models employed for these multicomponent
systems and the others described in the literature were the equations used to .
describe the initial and surface boundary conditions and the dimensionless -
current. Several parameters of the system studied are necessary for the
simulation. The standard potential of each redox site. the dimensionless diffusion
coefficient of each site, the diffusion coefficient, and the concentration of the bulk
solution along with parameters specific to the technique were entered as input in
the computer program. In addition, the doping profile of the germanium polymer
was easily determined from the simulated DPV voltammogram. From the RRDE
simulation, the chemical stability of this polymer was monitored as function of
potential. Both the mechanism and rate accompanying homogeneous reactions

were accurately determined.

44

2.5 Spectroelectrochemical technique

Experimental and theoretical studies have shown that the doping of
organic conducting polymers induces a charge transfer along the polymer chain
which results in significant local modifications of the chain geometry‘av43v5o33.
These modifications markedly affect the electronic structure of the polymer by
inducing localized electronic states in the gap between the valence and the
conduction band. They also lead to the possible formation of solitons, polarons,
and/or bipolarons, depending on the nature and/or the doping level of the
polymer‘03v17kv33. For conducting polymers possessing a degenerate ground
state such as polyacetylene, solitons are the most important charge defects
encountered in these polymers33°. Forthose without a degenerate ground state
such as polypyrrole or polythiophene, polarons and/or bipolarons are the
dominant charge-storage configurations33i'k. From this point and below, the
discussion focusses on the polaron and bipolaron concepts. A common way to
study the existence of polarons and bipolaron is by investigating the optical
absorption spectra of conducting polymers. Shown in Figure 7 are the energy
level diagrams and the transitions expected for these two types of excitations. In
the case of a hole polaron, four transitions can be expected from the valence
band and the gap level to the conduction band and the upper level. For a chain
with a bipolaron, transitions from the lower gap state are not possible because it
is empty and only two transitions remain between the valence band and the two
levels in the gap for a total of three transitions with the interband transition. All
these features appear in the absorption spectra of conducting polymers because
the number and energy of the transitions are different according to whether
polaron or bipolaron are present or not. Optical studies on polythiophene and

poly(3-alkylthiophene) have provided evidence that, upon doping, the majority of

45

Figure 7. Energy level diagrams showing allowed optical transitions for

polythiophene chain with hole polaron and bipolaron.

 

46

 

 

 

 

 

 

 

 

 

 

 

9 EH.-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

47

the charge is stored in bipolaron33‘J. In this study, absorption speCtroscopy was
primary used to characterize the electronic processes that occur in the
poly(3,4-dibutylthiophene) in the undoped and doped states as well as during
doping. Since this polythiophene derivative is soluble in common organic
solvents, in both neutral and concluding forms, optical properties, as well as
charge-storage configurations of the polymer, were studied _ in solution and
correlated with the ones obtained in the solid state. The changes observed in the
optical spectra of such conducting polymers will serve in elucidating the
mechanism of doping and the nature of the charge-storage species in the
polythiophene chain. Finally, comparison of the optical absorption spectra of
several similar polydialkylthiophenes using oligomers of different chain lengths
for the polymerization process will provide information on the amount of
conjugation within the polymer chain in our search for optimized polymerization

conditions in order to synthesize the best conducting polymers.

3. PROBING THE CHEMICAL STABILITY AS FUNCTION OF BAND-
FILLING FOR THE SOLUBLE CONDUCTING u-OXO-(TETFIA-t-
BUTYLPHTHALOCYANINATO)GERMANIUM

3.1 Introduction

In our group, we are interested in the study of materials containing
multiple interacting redox center531-34 . Understanding the factors that govern the
redox properties of these interacting centers is of importance in many chemical
systems. The inability to accurately control and determine such factors as
composition, structure, chain length, and degree of band-filling have slowed
down the construction of new conducting materials for specific applications. Our
efforts in understanding how the properties of conducting materials can be tuned
for specific applications have let us to study the soluble, already existing
u-oxo-(tetra-t-butylphthalocyaninato)germanium system. As previously described,
this material consists of covalently linked assemblies arranged in a stacked
architecture. It has been shown that upon chemical doping this material becomes
electrically conducting suggesting a large amount of electronic interaction
between the adjacent redox centers3°ao35. In this study, the redox
electrochemistry and chemical stability of the soluble u-oxo-(tetra-t-
butylphthalocyaninato)germanium will be investigated using a large variety of
conventional electrochemical and analytical techniques. Particular efforts will
focus on band-filling and the changes in the electrochemical properties of such a
system that occur upon variations in band-filling. Electrochemical techniques will
be employed to quickly and easily determine the degree of band-filling-potential
profile. It will be the first time that a wide range in degree of band-filling is

obtained for the doping of any conducting system. With the addition of

48

49

spectrochemical techniques, the chemical stability at any degree of band-filling
will be determined. Modified conventional electrochemical techniques also will be
developed to probe quickly and easily the chemical stability of such a system as
function of band-filling without degrading the solution used for the measurements.
These in-depth and novel probings of the redox stability of conducting materials
is unobtainable in solid state using either chemical or electrochemical techniques.
Electrochemical studies also will be performed to obtain information on the
energetics (potentials required to reach the different oxidation states) and kinetics
(rate of the electron transfer) for the doping of this material. The ease and rate at
which a material donates or accepts charge is an important consideration in
designing systems for specific applications. Computer simulations of
electrochemical techniques will be developed to observe the amount of
interaction between the redox sites. For example, the standard potential of each
individual redox center will be estimated by these methods. Digital simulations of
electrochemical techniques also will be used to prove the type of mechanism
occurring at the electrode surface during the doping process. The rate constants
of the homogeneous reactions resulting from the doping process of the system
also will be estimated by simulation. Redox information obtained from this study
will allow one to predict the electrochemical properties of conducting materials as
band-filling is systematically altered. This will aid greatly in the characterization

and design of new conducting systems.

3.2 Synthesis

Both the dihydroxy(tetra-t-butylphthalocyaninato)germanium monomer
and the p-oxo-(tetra-t-butylphthalocyaninato)germanium polymer were prepared
according to the method described by Hanack et al‘7q-fi35. The synthetic route

leading to the formation of the polymer is outlined in scheme l. Ten steps

50

Scheme l. Synthesis of the u-oxo-(tetra-t-butylphthalocyaninato)germanium

polymer.

 

 

a.

51

 

     

 

CI

Cl

0
I
co
m g '—
c' c I NH:
I I
O O
NH
I

1* fife.

[t-Bu4PcGeO]n

>51,
>33:

(CH3)3CC1 / FeCl3 g. POCl3 / pyridine
KMnO4 / aq. pyridine h. anhydrous NH3 / MeOH
EtOH i . G€Cl4

acetic anhydride / heat j. NaOH / H20 / pyridine

urea k. 340 °C / 10 hours / 10"5 torr
aq. NH3

SCHEME I

52

were needed from the commercially available o-xylene. The key step involves the
formation of the Po macrocycle from the reaction of the substituted
1,3-diiminoisoindoline with germanium tetrachloride. The product of this reaction,
the trans-dichloro(tetra-t-butylphthalocyaninato)germanium, t-Bu4PcGeCl2, was
then easily hydrolyzed in basic medium to form the dihydroxy(tetra-t—
butylphthalocyaninato)germanium t-Bu4PcGe(OH)2. Dehydration of
t-Bu4PcGe(OH)2 carried out in a tube furnace at 340 °C for 10 hours under high
vacuum yielded the u-oxo-(tetra-t-butylphthalocyaninato)germanium polymer.
The polymer backbone consists of Pc macrocycle subunits held together via
linear Ge-O-Ge covalent bonds in a stacking arrangement. The polymer was
characterized by IR, UV-Vis and elemental analysis. Data for the u-oxo-(tetra-t-
butylphthalocyaninato)germanium polymer and the corresponding monomeric
dihydroxide are summarized in Table 3. The IR spectrum of the polymer in
comparison with the spectrum of the monomeric dihydroxide shows the absence
of the symmetric Ge-O stretch at 646 cm‘1 and the appearance of the
antisymmetric O-Ge-O stretch at 893 cm‘1 (Figure 8). The absorption bands
observed in the 600 nm and 300 nm region for both the polymer and the
monomeric dihydroxide can be assigned to the Q-like and B-like 1c - 1!. electronic
transitions, respectively (Figure 9). Good agreement was observed between the
experimental and calculated elemental composition for both the polymer and
monomer. Good agreement also was observed for data presented in this study
and those reported in the Iiterature‘7qv'. The degree of polymerization (n) for the
germanium polymer was taken as 15 based on the analysis of diffusion
coefficient data (vida infra) from RDE voltammograms of both the polymer and
monomer, and confirmed by the simulated voltammetric results of the polymer.
Hanack and coworkers estimated the degree of polymerization for the

corresponding silicon polymer to be 2517'. Marks et al

53

 

 

 

 

 

 

l

I
Table 3. Infra-red spectral, optical absorption spectral, and elemental 1

analytical data for the [t-Bu4PcGeO]n and t-Bu4PcGe(OH)2

materials.

m>.mfi .Hv.m .HH.mm omm com owe “cacao:
mv.mH .Hm.m .qp.mm omm . omm mam naeaaom

 

54

Ancsomvz.m.o ncmnlm ncmnlo .o-mo-o.> .o-ww.> unconsoo

Awe mnmaamcm Hmucmemam Ascv Acme. mn>l>= Anise. .Hoflszv mH

 

 

 

M Human.

55

Figure 8. Fourier-transformed infra-red spectra of [t-Bu4PcGe01n and

t-Bu4PcGe(OH)2 as Nujol mulls.

 

stronsmitfonce

 

4000

 

 

ll

3100

56

 

 

 

t‘BU4PCGB‘OHIZ

[t-Bu4PcGeO]n I

 

 

 

 

 

 

2200

 

 

 

41‘
I 300

wovenumber (cm- 1)

I'IGURI 8

:1
400

57

Figure 9. Optical absorption spectra of [t-Bu4PcGeO]n and t-Bu4PcGe(OH)2

as solutions in THF.

 

58

a gnh

A55 £95.20;
0mm, OWN 0mm 0mm

)3

mimovmoomvsmlu

 

 

 

 

eouquosqo

c_omuomv:mlo_

 

59

characterized the analogous unsubstituted silicon and germanium polymer
[MPcO]n, and found that the germanium polymer has a n value approximatively
40% that of the silicon one35. Thus, a value of 15 for the soluble germanium
polymer seemed probable and was used to calculate the concentration of the
solutions. The p-oxo-(tetra-t-butylphthalocyaninato)germanium is soluble in
benzene, methylene chloride, and tetrahydrofuran. The good solubility arises
from the presence of the tetra-t-butyl substituents. As a consequence, the
reductive properties and chemical stability of the p-oxo-(tetra—t-
butylphthalocyaninato)germanium were investigated in solution, thus facilitating

the study over that in the solid state.

3.3 Electrochemical studies

Stacked phthalocyanine systems are unique in that both types of doping
(p and n) are theoretically possible. For example, extensive studies on the
metallophthalocyanine polymer [SiPchn have shown that this material can be
reversibly oxidized by electrochemical means as a slurry”. Reductive studies,
even if not so conclusive, have been shown to be possible‘9ei36. In this study,
only the reductive redox properties of the substituted metallophthalocyanine
polymer [t--Bu4PcGeO]n will be investigated in details. Attempts to
electrochemically oxidize solutions of this material have been unsuccessful due
to the degradation of this material in the electrochemical solvents used
(chlorinated solvents). A solute-solvent interaction occurs resulting in the partial
fragmentation of the O-Ge-O bond. In contrast, the reduction of the polymer
solutions were stable in the electrochemical experiment timescale. All

electrochemical experiments dealing with the reduction of the u-oxo-(tetra-t-

60

butylphthalocyaninato)germanium polymer will be performed in tetrahydrofuran
(THF). The working potential range in THF with 0.2 M TBABF4 is 2.2 V starting at
+0.05 V vs SSCE.

3.3.1 Cyclic voltammetry

Shown in Figure 10 is a series of cyclic voltammograms of a 33.4 uM
solution of the [t-Bu4PcGeO]n polymer (0.413 mg ml") at 100 mV sec". Curve
10a is the CV response observed when the potential of the working electrode is
scanned between +0.05 V and -1.45 V. A broad rather featureless response is
obtained comparable to the featureless capacitive charging current of a blank
solution containing only supporting electrolyte. This broad featureless response
of the polymer was expected. It can be explained by the small potential difference
between successive reductions resulting in incompletely resolved reduction
waves and causing coalescence into one, multiple electron wave. This type of
response is characteristic of a material with a lot of interactions between the
redox sites“. For a redox system comprised of non-interacting redox centers,
the shape of the current-potential response would be identical with the one of the
monomer, and the standard potentials for both systems would be the same3°av37.
However, the magnitude of the current would depend on the total number of
redox centers. The shape of the response is independent of sweep rate (v) but its
magnitude, the current, is proportional to v"2 in the
10-200 mV sec'1 range investigated. The response also remains unchanged
upon repeated cycling suggesting that the products resulting of the reduction and
re-oxidation of the polymer are chemically stable in the CV timescale. As
described below, controlled potential coulometry data prove that the polymer is
being cycled between the neutral and 1-state (one electron added per Pc site)

over this 1500 mV potential range. For comparison, the cyclic voltammogram of

61

Figure 10. Cyclic voltammograms for the reduction of a 33.4 uM solution of
[t-Bu4PcGe01n/02 M TBABF4 in THF at a sweep rate of
100 mV sec".
a. virgin polymer solution
b. two successive scans of the virgin polymer solution
c. polymer solution after controlled potential coulometry
(CPC) reduction at -1.65 V (decomposition into
monomeric type units) and oxidation back to its

neutral form

62

ca HKDGHh

Comm 9 3:5 2:58“.
o I ' I ' O ' O 0 i O ' O | O
m N 8.. 0on cm 0 nm N om: om.o pm 0

\ H”

 

 

 

 

li-l

 

 

 

 

63

conventional Nernstian one-electron redox couples (including electroactive
polymers having little or no interactions between the redox sites) exhibits a
change in oxidation state over only 200 mV potential range. Shown in Figure 10b
are the consecutive voltammograms observed when scanned between +0.05 V
and -1.95 V. On the first scan, a broad featureless response is again initially
seen. At approximatively -1.70 V, a huge increase in current occurs and upon
potential reversal, two small oxidations waves appear near -O.65 V and -0.25 V.
On the second scan, two new small reduction waves that are the corresponding
of the oxidation waves are seen. Upon potential reversal at -1.20 V, the two
oxidation waves are again observed. These two processes are not independent,
they are part of the large process occurring at -1.70 V because they do not
appear when reversing the scan before reaching this potential. This behavior
suggests that the polymer is unstable if reduced at potentials more negative than
-1.45 V. The polymer appears to be decomposing into an electroactive species at

more negative potentials.

3.3.2 Controlled potential coulometry

To accurately determine the percent reduction vs potential profile for the
u-oxo-(tetra-t-butylphthalocyaninato)germanium, a series of CPC experiments
were performed at various potentials. The current-time responses of the polymer
reduced at different potentials and oxidized back at 0.20 V are shown in
Figures 11 and 12. The potential of re-oxidation was set up to 50 mV beyond the
rest potential to ensure that the polymer was back in its neutral form. The current
was monitored until it decayed to a constant limiting background value, less than
1% of the initial current (10'6 A cm'z). Electrolysis times were typically 1 h to
11’2 h. The current was then integrated over time to obtain the total amount of

charge transferred during the CPC process. The net amount of charge used in

64

Figure 11. Current-time curves for the reduction of a 33.4 uM solution of
[t—Bu4PcGeO]n/0.2 M TBABF4 in THF during controlled potential
electrolysis (CPC).

a polymer solution reduced at -0.55 V
b. polymer solution reduced at -1.05 V

O

polymer solution reduced at -1.25 V

d. polymer solution reduced at -1.45 V

 

65

 

0V EMF

#

<
E
in

L

 

HH "MDUHh

AEEV 0E:
ON
4..

 

 

 

 

 

 

66

Figure 12. Current-time curves for the oxidation of reduced
[t-Bu4PcGeO]n/TBABF4 in THF resulting of Figure 11.
a'. polymer solution (a) re-oxidized at 0.20 V
b'. polymer solution (b) re-oxidized at 0.20 V
c'. polymer solution (c) re-oxidized at 0.20 V

d'. polymer solution (d) re-oxidized at 0.20 V

 

67

NH MMDOHh

AEEV 05:
mm 0N im— pl—

 

 

-lln

HO

.6

.D

.m

 

 

68-

the determination of the percent reduction (equation 3) was calculated by
subtracting the product of the steady state background current and time of the
experiment from the total amount of charge measured. Shown in Figure 13 is a
plot of the percent reduction of the germanium polymer as function of potential.
The percent reduction, at a given potential, is determined by averaging the
amount of charge passed for the initial reduction of the neutral polymer and
oxidation back to the neutral form at 0.20 V. The error bars represent the
difference between these two values. The small difference between the two
values can be attributed to the variation in potential of, the quasi-reference
electrode in the course of the experiment. This small difference suggests that the
reduced form of the polymer at all degrees of partial reduction is stable on the
CPC timescale, i.e., several hours. Electrochemical and spectroscopic
characterization of the cycled neutral polymer also support this assumption. The
CV and DPV voltammograms of the polymer obtained after each reduction/re-
oxidation cycle appear identical to that of the virgin material (Figures 10a and
22c). The UV-Vis absorption spectra of the neutral polymer taken after each
cycle are very similar to those of the virgin polymer (Figure 9). Figure 14
represents the UV-Vis absorption spectrum of the neutral polymer after being
reduced at 100% and oxidized back to its neutral form. The presence of intense
peaks at 621 nm and 334 nm proves that the phthalocyanine ring remains intact
and the absence of new peaks confirms the good stability of the cycled polymer.
The corresponding infra-red spectrum (Figure 15) exhibits the asymmetric
O-Ge-O band at 904 cm", but does not show the symmetric Ge-O band in the
600 cm'1 region characteristic of the monomer dihydroxide. This suggests that
the cycled polymer does not decompose into monomeric type units. If the
polymer was degrading into molecular species or small oligomers, the smooth

doping profile would not be observed. Instead, a sharp change in the percent

Figure 13.

69

Degree of partial reduction vs potential determined by controlled
potential coulometry for the reduction of [t-Bu4PcGernfTBABF4
in THF. Each point (0) corresponds to the mean value for a
reduction/re-oxidatlon cycle. The error bars represent the

difference between the reduction and re-oxidation.

70

 

00.—l

0021

NH HMDUHh

00.0l

Comm m> 3.96 32:22.

00.0

no
to.
.8
ion
r2.
tom
-8
-2
row
.8 .
loo.

uouonpeJ as

 

 

71

Figure 14. Optical absorption spectra of [t-Bu4PcGeO]n in THF after CPC at
-1.44 V (98 °/o reduction) and oxidation back to its neutral form at
0.20 V. The spectrum is very similar to the one obtained for the

virgin polymer (Figure 9).

 

72

vn flflDUHh

AEcv 50:23.;
00m 0m¢ 0mm pwm

I

 

eouquosqo

 

 

Figure 15.

73

Fourier-transformed infra-red spectra of [t-Bu4PcGeO]n/Nujol after
CPC at 4.44 V (98 % reduction) and oxidation back to its neutral
form at 0.20 V. The band at 904 cm'1 is assigned to the

asymmetric O-Ge-O strech.

 

74

0m;V

A
EM
3

 

 

 

ma MMDGHh

00NN
llJll

lEov LonEacgoB

00—m

 

 

00V

0

l

 

eououjwsumj'x

75

reduction would be seen. The smooth dOping profile suggests the formation of an
energetic band from which charge can be continuously and cyclicly added. It is
the first time than such behavior has been observed in the solution redox
chemistry. Additional proof of chemical stability was obtained from a series of
plots of normalized currents vs time at different potentials. The current during a
CPC experiment is expected to decay exponentially29 . Shown in Figure 16 is a
series of log(i(t)/i(0)) vs time plots for the reduction of the polymer. In each CPC
experiment, the current, at any time, is normalized to the initial current value. The
data obtained at -1.02 V (50% reduction) and -1.44 V (98% reduction) exhibit the
linear behavior of a chemically stable redox system. When bulk electrolysis is
performed at -1.65 V, a potential corresponding to halfway through the large
reduction wave of the CV in Figure 9b, the data clearly deviate from linearity. At
this potential, the polymer is 227% reduced with a collection efficiency of 102%.
This increase in potential of only 200 mV results in more than doubling of the
degree of reduction. Although no electroactivity is gained or lost, the nature of the
electroactive species has changed. Shown in Figure 10c is the cyclic
voltammogram of a solution after being reduced at -1.65 V and oxidized back at
0.20 V. The response is no longer the broad featureless wave initially observed
for the virgin polymer (Figure 103) but is similar to the one usually obtained for a
conventional redox couple. The first three reduction waves at Epcl = 0.35 V,
Epc2 = -0.83 V and Epc3 = -1.68 V are characteristic of consecutive Nernstian
one-electron reduction-processes resulting in the formation of a chemically stable
product”. The difference between the cathodic and anodic peak potentials for
each wave (AEp) is approximatively 75 mV, close to the 60 mV expected for such
a process. Deviations of AEp from the Nernstian value of 60 mV probably can be
attributed to effects of uncompensated resistance in this highly resistive solvent.

Further information about the electrochemical process was obtained by

Figure 16.

76

Variations of log (i(t)/i(0)) with time at different potentials
determined from controlled potential coulometry curves for the
reduction of [t-Bu4PcGeO]n/TBABF4 in THF. i(0) represents the
initial current value for each CPC experiment. The ordinate scale
pertains only to the data obtained at -1.02 V. Each successive
curve is shifted upwards by 1 unit. The % reduction of the listed
potentials are as follows: -1.02 V: 50% reduction, -1.44 V: 98%
reduction, and -1.65 V: 227% reduction.

77

0H flflDOHh

:E E.
C 00...:

 

 

> N0.—l

> vvél

.> 3.7

[MI-

INI

 

 

(Cl/05°!

78

investigating the dependence of peak currents and potentials on scan rate. The
peak potentials are invariant with the scan rate (v), and the magnitude of the
anodic and cathodic peak currents are the same in the 10-200 mV range and
increase as v1/2. These results demonstrate that the cycled polymer is stable in
this potential range and the waves are diffusion-limited. The fourth reduction
wave at E904 = -2.09 V, near the background limit of the solvent, is well defined
and lacks the corresponding oxidation wave upon potential reversal is unstable
on the CV timescale, i.e., a few minutes. Sequential CPC measurements
performed at potentials 200 mV more negative than each of the first three
reduction waves also suggest that the first three processes are consecutive
one-electron reductions to form stable products. For each CPC experiment
performed at these three reduction/oxidation processes, the same amount of
charge was passed (reduction) and collected (oxidation back to the neutral
material) as expected for 100% reduction of the polymer. These results are
summarized in Table 4 and suggest that the polymer has fragmented into a
mixture of monomeric or oligomeric species without any loss in electroactivity. As
will be shown later via the redox characterization of dihydroxy(tetra-t-
butylphthalocyaninato)germanium , the corresponding polymer, when it is more
than 100% reduced, quantitatively decomposes into monomeric type units that

are electrochemically and spectroscopically identical to it.

The plots of log(i(t)fi(0)) vs time at different potentials also serve in
elucidating the type of reactive mechanism occurring at the electrode during the
CPC process. In the late 1970s, Dr. Louis Meites and his group at the
Polytechnic Institute of Brooklyn reported a theoretical description of the
current-time curves obtained during a controlled potential electrolysis for different

reactions such as EC (electron transfer at the electrode surface followed by

79

Table 4. Controlled potential coulometry data for p-oxo-(tetra-t-butyl

phthalocyaninato)germanium.

80

.> mm.Hi um coauuscou mcfizoaaow muauuo

mam
00m
mm
mNN
mm

mam
mom
pm
.FNN
mm

noeaaom on» we cofluflmOQEoooo n

om.O\ mm.Hl
o~.O\ oa.nl
o~.0\ on.o-
o~.0\nmm.H-
ow.O\ m¢.H-

 

COfiUMUHXO UCOOHOQ

 

cofluoscou ucmuumm

v Hflmdfi

A>v maowo coflumofixo\cowuoscou

HON Hflflucwuom

 

81

homogeneous chemical reaction) and ECE (EC reaction followed by another
electron transfer reaction)”. By comparing the log(i(t)fi(0)) vs time plots in this
study with the ones discussed in Meites' work, it appears that the type of reaction
occurring during the electrolysis of the germanium polymer can be described as
an ECE-type mechanism :
O1 + n1 9 <==> R1
Rl

 

> 02
02 + n2 9 <——-—> R2
where :
O1, 02 = oxidized species
R1, R2 . reduced species
n1, n2 :- number of electrons transferred in the process
The plots of log (i(t)/i(0)) vs time at -1.02 V and -1.44 V are concave-downward
curves and according to Meites results, the chemical step can be identified with
certainty in this case. In contrast, the plot at -1.65 V is a concave-upward curve.
Meites explained that the small and slowly decreasing current observed was
probably due to a very slow transformation of R1 into 02. By including additional
voltammetry results for the polymer and the corresponding monomer, the
mechanism was refined into an EnCEE mechanism:
On
[t-Bu4PcGeO]n + n e' > n t-Bu4PcGeO'
n t-Bu4PcGeO‘ + n e' > n t-Bu4PcGeOZ'
n t-Bu4PcGeOZ'4- n e' > n t-Bu4PcGeO3’

-j R- + e'<=> On-l -j Rj+1 (1 San-1)

 

 

 

For the first series of electron transfer processes, 15 different E0 values are
needed to respond for the complete reduction of the germanium polymer. During
the chemical step, the polymer breaks into n number of monomeric units in the

anion form. These anions can then be further reduced into dianions and trianions.

82

3.3.3 Rotating disk and rotating ring-disk voltammetry

The complex nature of the redox response for the polymer and its
chemical stability as a function of the degree of reduction was investigated in
details through the use of rotating disk and rotating ring-disk electrode
voltammetry. Shown in Figure 17 is a series of RDE voltammograms at different
rotation rates for the reduction of the germanium polymer. For comparison, the
RDE voltammogram of the blank solution containing only the supporting
electrolyte at a rate of 700 rpm over the same potential range is included. The
results correlate well with the CV data set out in Figure 10. The onset of current
begins near -0.40 V and steadily increases to near -1.65 V where a large
increase in current is again observed. Based on the CPC and CV results
discussed above, this increase is believed to be due to the reduction of the
decomposition product of the polymer following 100% reduction. The 1.25 V
range observed before the onset of the large increase in current is nearly
identical to the potential increment seen in the CPC results (Figure 13). Thus the
current response between -0.40 V and -1.65 V is believed to arise from the one-
electron reduction of the polymer. The slope of the RDE response in this region is
independent of angular velocity (0)) and the current magnitude at any potential
within this range increases with m1/2. Levich plots, i.e., i(E) vs 001/2, obtained for
several potentials within the same range as above, yield straight lines with
intersect close to zero (Figure 18) suggesting that there is no kinetic limitation
towards the reduction process323v40. The small deviation from zero intercept can
be attributed to higher background currents at high potentials. This behavior
implies that the reduction process of the polymer within this potential range is
mass-transfer limited and that there are no surface effects (adsorption)

influencing the electron transfer process. The reduction of the polymer is simply a

83

Figure 17. Rotating disk electrode voltammograms for the reduction of a
33.4 uM solution of [t-Bu4PcGeO]n/0.2 M TBABF4 in THF at
various rotation rates. The current of the blank solution does not

change with increasing rotation rate.

84

 

00.nl
c0220» _ 3:03

5.... ooom

E9. oow

E9. 005
EB oou.

EB ooou

EB ooon

EB 82.
EB ooom

ha IMDGHh

00mm m> «:03 3.2.200
00.Nl 00.—l
_ J

 

\

3+

j.

00.0

 

 

Figure 18.

85

Levich plots for the reduction of [t-Bu4PcGeO]n/TBABF4 in THF at
various potentials. The % reduction of the listed potentials are as
follows: 064 V (25% reduction); -0.84 V (37% reduction); -1.04 V
(50% reduction); -1.24 V (68% reduction); -1.44 V (98%

reduction).

86

ma MMDOHh

AN\—loomv 3.00.3 3.30:0. 00. can

 

 

 
  

mm om m. o. m o
_ p _ — —o
l l m.
Pllb\0l\\d\1\10\¢\b \\H\H“
\ \ \
\ \
\ \
\ IO—
\
> ¢0.0| «O i [ON
> ¢m.ol Lo 0
> #Oél «O d
> .vN.—I “O D [on
> ¢¢.—l HO O
10¢

 

 

(v11) iuauno nsgp

87

Nernstian process. Shown in Figure 19 is the percent reduction vs potential
profile for the germanium within the 1.45 V potential range. The percent
reduction, at any potential, is determined by normalizing the RDE current at a
given potential to the RDE current at -1.45 V taken as representing 100%
reduction and multiplying by 100 (equation 7). This particular potential was
chosen based on results of CPC experiments. In this doping profile, the error
bars represent the differences in percent reduction calculated from the RDE
voltammograms of Figure 17 for the same potential at different rotation rates. As
it is observed in Figure 19, the doping profile obtained from RDE data maps out
the percent reduction vs potential plot obtained from CPC results. The excellent
agreement between coulometric and voltammetric data suggest that the
potentials used in CPC and RDE experiments represent the thermodynamic
values for which the reduction process occurs. Using a kinematic viscosity value
of 6.2 10"3'cm2 'sec'1 and a disk electrode area value of 0.163 cm2, a value of
6.6 10'6 cm2 sec'1 for the diffusion coefficient of the neutral germanium polymer
was obtained from the RDE voltammograms. This value is compatible with that of
the silicon analog 3.0 10'6 cm2 sec'1 obtained in 1,1,2,2 tetrachloroethane“, a
solvent slightly more viscous than tetrahydrofuran in which the diffusion
coefficient of the germanium polymer was measured. Shown in Figure 20a,b,c
are a series of RDE, RRE and RRDE voltammograms for the reduction of the
germanium polymer over the potential region in which it is stable (0.00 V to
-1.45 V). As expected, the RDE and RRDE waves are identical in shape and
differ only in magnitude due to the difference in electrode area between that of
disk and that of ring-disk. The RRDE data (Figure 20c) represent a continuous
shielding-collection experiment in which the disk is held at -1.45 V, resulting in

100% reduction of the polymer and the ring is scanned from -1.45 to 0.00 V.

Figure 19.

88

Degree of partial reduction vs potential determined by rotating disk
voltammetry for the reduction of [t-Bu4PcGern/TBABF4 in THF.
Each point (0) corresponds to the mean value for a reduction/re-
oxidation cycle. The mean value is calculated by averaging the
current at different rotation rates. The error bars represent the

standard deviation.

89

 

0H Hmbqu

Comm .m> m:o>v 32:22.
cm. _I 00.. .l 01mm0l 2.0.0

10—
ION
ion
10¢
tom
100

uouonpeJ as

105
100
100

loop

 

 

Figure 20.

90

Experimental (a.b.c) and simulated (d,e,f) rotating disk electrode
voltammongrams of a 50.1 uM solution of [t-Bu4PcGeOJn/02 M
TBABF4 in THF. Both the simulation and experimental conditions
are identical to the ones of Figure 17. The rotation rate was 4000
rpm.
a,d. disk voltammograms in the absence of the ring
b,e. ring voltammograms in the absence of the disk
c,f. ring voltammograms when the disk is held at -1.45V
and the ring is scanned from this potential to the rest

potential

91

ON IKDGHh

Comm 9 £90 ézzmzoa

 

 

 

0m_._l 0021 omodl 00.0 0m_._l 00...! omhol 00.0
a. a.
o. o.
m o
m n
ll_l|l ||_.|
<1 0. <3. o.
Ll LI
6 m
nonmassfim amucmeflummxm

 

 

92

Initially, the ring current is cathodic, the ring being shielded by the disk resulting
from the reduction of the polymer. As the potential is scanned towards more
positive values, the current eventually becomes anodic and reaches a limiting
value resulting from the collection (oxidation back to neutral polymer) of the
reduced polymer. As predicted by equation 9, the difference between i9 and iRO
is a constant at any potential and is equal to id * N. These results, shown in
Figure 21, represent the [(i3 - iR°)/id] - N term vs potential plot. This plot
illustrates the applicability and utility of these types of experiments for quickly
monitoring the stability of multi-site redox systems. At potentials more negative
than -1.45 V, the CV and CFO results suggest that the polymer is unstable upon
reduction. Continuous shielding-collection experiments results also suggest this
behavior. The RRDE data obtained at disk potentials of -1.64 V and -2.19 V
clearly show that the values of (ino - iR) / id is neither equal to N nor constant
over the entire potential range. Only at the limiting potentials, 0.00 V and the
value at which the disk is set, is the difference equal to N. The ring current (iRO)
observed during the RRDE experiments is greater than expected which implies
the inclusion of an additional cathodic current from the reductive process. At ring
potentials sufficiently positive of the disk potential, near 0.00 V, the values of
“HO - iR) / id again is equal to N, suggesting that all the electroactive material
can be collected. This correlates well with the CPC results obtained at -1.65 V
and the resulting CV characterization of the decomposed polymer set out in
Figure 10c. It supports the assumption that no electroactivity or redox sites are
lost, just that the nature of the electroactive species has changed. As will be
shown via digital simulations of the RRDE experiments, the mechanism and rates

of the homogeneous reactions can be elucidated.

Figure 21.

93

Continuous shielding-collection plots for the reduction of
[t-Bu4PcGeO]nfTBABF4 in THF at various disk potentials. The
rotation rate is 2000 rpm for all plots. iRc represents the cathodic
ring current when the disk is hold at a particular potential, iRco is
the cathodic ring current in the absence of reaction at the disk and
N is the collection efficiency of the electrode assembly. The %
reduction of the listed potentials are as follows: 085 V (41%);

-1 .45 v (100%); -2 v (310%)

94'

 

AN 30H.-

Amomm m> 3.05 3.23.2.

 

 

00.Nl 0mw—l big—l 0m.0l 00.0 961
fillql q _ llJl .JIIIfi
>mm.l Lo xflo D
n 0 >91: so in a rotor
a >Nl so in o
o n o n a
o .
D I. no 0|.
0
0 fi
0 a a c a a a a n n
n D a a I? a
o .. cod
0 o o o a a a
o
r 00.0

 

 

N - p! / (093! - an!)

95

3.3.4 Differential pulse voltammetry

The rather featureless responses obtained from the CPC, CV, and RDE
experiments have led to use differential pulse voltammetry to more accurately
probe the energetics of the doping process. The derivative output of DPV makes
it extremely sensitive to small changes in the redox profile of the polymer“.
Since the pulse sequence employed for DPV effectively performs a pre-
electrolysis of the electroactive species near the surface of the electrode before
the current is measured, the voltammogram illustrates the ease with which the
polymer can go from one oxidation state to another. Shown in Figure 22a is the
DPV for the reduction of the polymer over the potential range in which it is stable.
The voltammogram represents the conversion of the polymer from neutral
material to 1-oxidation state, one electron added per Pc site. The DPV data more
clearly show that the reduction of the polymer is not a smooth continuous
process but in fact varies greatly depending on potentials. As it will be seen later
with the simulated DPV response of the polymer, a large increase in current
arises from strong interactions between redox sites within the polymer. This
occurs when the standard potentials of redox sites are very close from one
another. The large dip observed in current near -1.03 V corresponds to the
current plateau seen in the RDE voltammogram. This dip located at
approximatively 50% reduction may imply some inherent stability at this particular
percent reduction“. Theoretical and experimental studies on the silicon analog
have suggested such behavior“. At potentials more negative than -1.45 V, two
additional reduction peaks are observed at -1.79 V and -2.19 V (Figure 22b). The
magnitude, shape, and potential of these peaks are not well defined as the
response up to -1.45 V. This was expected considering the instability of the
polymer at these potentials. The polymer decomposition tends to cause a

cumulative depletion of electroactive material near the electrode surface which

Figure 22.

96

Differential pulse voltammograms for the reduction of a 33.4 pM
solution of [t-Bu4PcGern/02 M TBABF4 in THF. The
experimental conditions are as follows: pulse amplitude: 50 mV;
pulse width: 50 mV sec“; pulse duration: 1000 msec and scan
rate: 4 mV sec".

a, b. virgin polymer solution

c. polymer solution after CPC reduction at -1.45 V (98%

reduction) and oxidation back at 0.20 V

91

 

 

 

 

 

 

 

NN 50H.-

Comm m> «:96 _o_E2oo
. l 0 . 0 .Nl 0m._l

 

 

 

l-é-l
.I

(n no

I.

 

    
 

 

 

98

severely alters the voltammograms. The two new peaks are believed to arise
from the decomposition products and are similar to those observed in the DPV of
the dihydroxy(tetra-t-butylphthalocyaninato)germanium monomer. DPV taken
after each CPC cycle along the doping profile confirm that the polymer is stable
up to -1.45 V and decomposes into monomeric type units at more negative

potentials (Figure 22c).

3.4 Electrochemical characterization of dihydroxy(tetra-t-butylphthalo
cyaninato)germanium

The decomposition of the polymer upon reduction of more than 100% to
monomeric type redox species leads to a detailed investigation of
dihydroxy(tetra-t-butylphthalocyaninato)germanium, the monomeric precursor to
the polymer. Shown in figure 23 are the CV and DPV voltammograms for its
reduction. The first three peaks at E1,2 = -0.29 V, —0.77 V and -1.63 V are
characteristic of consecutive one-electron Nernstian reductions. Reversibility of
reactions assigned by these peaks and stability of the resulting products are
indicated by peak splitting of about 60 mV, peak current ratio of approximate
unity and constant ipc / v”2 values for potential scan rates between
50 mV sec'1 and 200 mV sec". The product of the fourth reduction appears to
be unstable on the voltammetric timescale. The oxidation wave following
potential reversal is absent for that process. CPC studies also indicate that the
polymer can be reduced stepwise to form a stable anion, dianion and trianion.
Each process gives a n value of 1.0 +0.01 for both the initial reduction and
oxidation back to the starting material. DPV taken after each CPC cycle
(figure 24) are identical to those of the virgin material (figure 22b), indicating

stability of all four oxidation states (0 to 3-) on the CPC timescale.

99

Figure 23. Cyclic (a) and differential pulse (b) voltammogram of a 881 uM
solution of t-Bu4PcGe(OH)2/0.2 M TBABF4 in THF. The
experimental conditions were the same as above (Figures 10

and 22).

100

MN HfiDGHh

Comm m> «:30 32.530

III

)3
w:

 

 

 

 

 

 

00.nl 00.Nl 00.. pl

5

 

E

\

 

 

 

31

l.

 

0v.Nl 0v.. —l

0v.0l 0.0.0.

Hi

 

 

101

Figure 24. Differential pulse voltammograms for the reduction of a solution of
881 uM t-Bu4PcGe(OH)2/O.2 M TBABF4 in THF. The
experimental conditions are identical to the ones of figure 22.

a. virgin monomer solution

b. monomer solution after CPC reduction at -0.50 V and
oxidation back at 0.20 V

c. monomer solution after CPC reduction at -1.00 V and
oxidation back at 0.20 V

d. monomer solution after CPC reduction at -1.70 V and
oxidation back at 0.20 V

e. monomer solution after CPC reduction at -1.85 V and

oxidation back at 0.20 V

102

00.Nl 0m.—l
allllrfi

 

 

VN HKDGHh

Comm n> 33$ 3.3030

l—H
l—H

 

00.0l 0v. 0 .Nl 0m..—l

00.0l 030

 

 

103

A diffusion coefficient of 7.5 10'7 cm2 sec'1 for the neutral monomer was
calculated from its RDE voltammogram. The value is similar to that measured for
other monomeric phthalocyanines taking into account differences in solvent
viscosities. From the diffusion coefficient of the monomer and that of the
corresponding polymer a degree of polymerization of 15 was found for the u—oxo-
(tetra-t-butylphthalocyaninato)germanium. This value of degree of polymerization
along with other parameters provided a basis for modelling the current-potential

DPV response of the polymer.

3.4 Digital simulations

To help elucidate the complex reductive chemistry of the u—oxo-(tetra-t-
butylphthalocyaninato)germanium, digital simulations of the electrochemical
response were undertaken. Controlled potential coulometry and rotating disk
voltammetry data suggest that the polymer undergoes Nernstian reduction
responses at all potentials over the 1.45 V range in which it is chemically stable
and that at -1.45 V each redox site comprising the polymer has undergone a
one-electron reduction. Therefore, the reduction of the germanium polymer was
modelled as a series of Nernstian one-electron redox couple reactions in
equilibrium with one another. Shown in figure 25b is the simulated differential
pulse voltammogram for the reduction of the polymer over which it is stable. The
parameters used in the simulations such as the electrode area, concentration of
the polymer along with other parameters specific to the technique employed were
chosen as identical to the ones used to obtain the experimental data. The degree
of polymerization was taken as 15 based on the diffusion coefficient of the
polymer. The standard potentials (El-0) were carefully chosen to "map out“ the

experimental response obtained for the reduction of the germanium polymer. The

Figure 25.

104

Experimental (a) and simulated (b) differential pulse
voltammongrams of a 50.1 uM solution of [t-Bu4PcGeO]n/O.2 M
TBABF 4 in THF. Both the simulation and experimental conditions
are identical to the ones of Figure 22. The standard potentials used
to model the system are represented by the vertical lines on the

potential axis.

105

 

 

 

mN HMDOHh

60mm 2 3.9.0 05505..

 

.1

<0 0mm

I.

 

 

 

106

values of these standard potentials are represented as vertical bars on the
potential axis. Both the magnitude and the shape of the simulated differential
pulse voltammogram are similar with the observed one (Figure 25a). The huge
peak seen at -1.65 V appears to be formed by several closely spaced redox
centers. It was observed that small variations in the standard potential values
caused significent changes in the shape and magnitude of the simulated current
response. The fact that the standard potential values are very close to one
another indicate the existence of a large amount of interaction between the redox
sites. From the simulated differential pulse voltammogram, the percent reduction
vs potential profile was determined for the germanium polymer. Simulated data

closely match the experimental CPC and RDE results (Figure 26).

Digital simulations of the electrochemical response were also used to
determine the mechanism and rate of the decomposition reaction of the
germanium polymer. Shown in Figure 20d,e,f are the simulated RDE, RRE and
RRDE voltammograms over the potential range in which the polymer is stable.
These simulated voltammograms are nearly identical in shape and magnitude
with the those obtained experimentally (Figure 20a,b,c). As predicted by
equation 9, (in0 - iR ) / id is a constant at all potentials over the 1.45 V range and
equal to N suggesting that the polymer is chemically stable at potentials positive
of -1.45 V (Figure 27a). When including the decomposition process (EnCEE
mechanism) into the simulation program, the (iR0 - iR) / id expression deviates
from N. Shown in Figure 27b,c is the [(190 - iR) lid] “-N term vs. potential plot
obtained from the simulated shielding-collection experiments for a decomposition
rate constant of 20 sec'1 at a disk potential of -1.65 V (216% reduction) and
-2.16 V (300% reduction). At -1.65 V, the monomeric anion resulting from the

107

Figure 26. Degree of partial reduction vs potential determined by controlled

potential coulometry, rotating disk voltametry and digital simulation.

108

 

on amoouu

Comm 2 £20 .2223
lobtl 0m.0l 00.0 00

   

 

 

030 000 0 ion
220 mom 0 low
030 00.335.» >00 i iOh

uogionpeJ at

 

 

Figure 27.

109

Simulated continuous shielding-collection plots for the reduction of
[t-Bu4PcGeO]n/TBABF4 in THRat various disk potentials. The
rotation rate was 2000 rpm. Then simulated data are based on the
assumption of the EnCEE mechanism. The % reduction of the
listed potentials are as follows: -1.45V = 100% reduction; -1.64V =
226% reduction and -2.16V = 300% reduction.

110

hN NKDUHE

00mm 2. £20 2:208...

 

0m.Nl 00.N.i 0m.—l 00.—l 0m.0l
.IIJIIIIIIJII IJI ll:
> m—Nl 3 0.03
> toél 3 033
> mY—l 3 0.03
I
I II .
_I .-
I II
I
I «III I

I <<<< 4 I I

< I I

iiiIiI iiiiiiiiii «lab

 

00.

I
‘

 

 

m0.0

070

m—.0

0N.0

mN.0

0m.0

Pi / (38! - 0°30

111

decomposition of the polymer is reduced into the dianion and at -2.16 V it is
further reduced into the trianion. Good agreement was observed between
experimental and simulated results. This allows a better understanding of the

reductive chemistry of the u-oxo-(tetra-t-butylphthalocyaninato)germanium.

4. STUDYING THE ELECTRICAL CONDUCTIVITY AND
SPECTROSCOPIC PROPERTIES AS FUNCTION OF BAND FILLING
FOR THE SOLUBLE POLY(3,4-DIBUTYLTHIOPHENE)

4.1 Introduction

In our laboratory, we also are interesting in the synthesis and
characterization of soluble analogs of conventional conducting organic polymers.
Our efforts have focussed on thiophene derivative systems. Polymers of this type
have been extensively studied in the solid state‘“~‘7a'P-45. The majority of the
work has been performed on thin films attached directly to the electrode surface
and free-standing films. These materials are generally prepared by simply
oxidizing the corresponding monomer. Polymerization occurs via a cation radical
coupling mechanism that involves elimination of predominamly or protons. Chain
growth is believed to be terminated by either steric hindrance or and group
unreactivity12»14. The factors dictating the nature of the resulting materials remain
poorly understood and largely uncontrolable. In addition, they are significantly
influenced by the complications arising from the solid state. Little, if any, influence

over the characteristics of these polymers has been possible.

Substituted polythiophenes that are soluble in both aqueous and
nonaqueous solvents have recently appeared‘7b'q. Solubility in organic solvents
is achieved by the presence of alkyl or substituted alkyl chain on 3-position of the
thi0phene ring. However, addition of alkyl group on 3-position does not prevent
crosslinking polymerization to occur. In order to eliminate this undesirable effect,
polydialkylthiophenes have been synthesized. The presence of alkyl groups in

3— and 4»position forces polymerization to occur selectively in 2- and 5-position,

112

113

leading to a more structurally regular polymer with mainly a a' linkages. Another
way of to enhance the selectivity of these or a' linkages consists in synthesizing
polydisubstitutedthiophenes by polymerization of B-disubstituted oligomeric
thiophene instead of the common monomer. The presence of the alkyl groups in
3- and 4-position also improves the solubility of these polymers, allowing them to
be studied in solution. In this study, the redox properties of
poly(3,4—dibutylthiophene) will be investigated. Since this new polymer is soluble
both in its oxidized and neutral (undoped) form, the study will be performed in
solution using a variety of conventional analytical techniques. Electrochemical
techniques (CV, DPV, and CPC) will be employed to characterize this polymer
when it is oxidized to various degrees. Unlike doping performed on thin films
where the degree of band-filling is influenced or limited by the structure of the
material, this soluble polymer may be able to be oxidized to any level ranging
from O to 100%. Electrical and Optical measurements will be used to investigate
how the collective properties of this conducting polymer vary as function of band-
filling. Optical absorption studies have proven useful in characterizing the basic
electronic stnicture and the nature of the charge carriers of conducting organic
polymers. The appearance of mid-bandgap transitions as degree of band-filling is
varied will provide direct evidence for polarons and/or bipolarons. Conductivity
also will be performed when varying the degree of band-filling. Information on
chemical stability upon doping will be provided using a combination of
electrochemical and spectroscopic techniques. It will be the first time that the
same probe system will be simultaneously investigated in solution and on an
electrode surface. Finally, comparison with other derivative thiophene systems
such as poly(3',4'-dibutyl-a-terthiophene) and poly(3",4"-dibutquuinquethiophene)
which differ in the number of thiophenes present in the repeat unit will help in

searching for new conducting materials with predictable and tunable properties.

114

4.2 Synthesis
Synthesis of mmmenc and polymeric 3,4-dibu_tylthioghene

3,4-Dibutylthiophene was prepared according to the method described
by Kumada et al“. The stepwise synthesis is outlined in Scheme 2.
Commercially available thiophene was easily brominated and then selectively
debrominated to form 3,4-dibromothiophene47. This halothiophene, in the
presence of catalytic amounts of a nickel-phosphine complex was readily
alkylated vith butyl magnesium bromide (Grignard reagent) resulting in the

formation of the 3,4-dibutylthiophene, the basic building block of the polymer.

Scheme ll. Chemical preparation of 3,4-dibutylthiophene

Br Br Bu Bu
</ \> _. fl _. fl \
Br Br
S S S
a. sz/CHCIS
b. f‘lBULi/Etzo

c. 2 BuMgBr/Ni(dppp)Cl2 (dppp = 1,3-bis(diphenylphosphino)propane)

Very pure 3,4-dibutylthiophene was obtained by following this route. The
1H-NMR spectrum of this compoundshows peaks at 0.94, 1.40, and
1.61 ppm which were assigned to the methyl and methylene groups of the butyl
substituent. A peak much further downfield at 7.00 ppm assigned to the aromatic

hydrogen located at the a—positions of the thiophene (Figure 28).

The synthesis of poly(3,4-dibutylthiophene) was achieved by

electrochemical oxidative polymerization of the above described monomer. The

115

Figure 28. 1 H-NMR spectrum of 3.4 Dibutylthiophene in 00300003. The
peak at 2.04 ppm is due to the residual solvent.

116

 

0N “dawnh

AE00V :20 3033.3

 

117

electrochemical approach was preferred over the synthetic one because of its
simpler and cleaner method of preparation. With chemical synthesis, low
molecular weight polymers, stnictural defects, and/or impurities are generally
obtained. The mechanism proposed for the disubstituted thiophene
polymerization is a general scheme for the electrochemical process of
heterocycles (Scheme 3)22. The coupling step of two identical, monomer radical
cations produces an intermediate dication which, upon loss of two protons, yields
the neutral dimer (step 2). The high concentration of the monomer radical cation
maintained near the electrode surface allows for the continued coupling with the
oligomer radical cation (step 3 and 4). The overall reaction is referred to as an
E(CE)n type mechanism, i.e., an electron transfer reaction (E) followed by a
cascade of chemical (C) and g electron transfer (E) reactions.
Poly(3,4-dibutylthiophene) was electrochemically generated in nitromethane
containing 2.5 10'2 M monomer and 2.5 10'2 M supporting electrolyte (TBABF4).
The reaction vessel was a one-compartment cell as described previously using a
platinum flag electrode as working electrode. A potential sweep technique (CV)
was used for the anodic film growth. The growth of the disubstituted
polythiophene film is illustrated through a series of repetitive cyclic
voltammograms (Figure 29). During the first sweep, a broad, irreversible
oxidation peak which corresponds to the formation of the monomer radical cation
is observed at Epa = 1.77 V. The amount of energy required for the oxidation of
the Misubstituted thiophene is about the same as that for the oxidation of the
unsubstituted thiophene, but much higher than that for the oxidation of the
B—monosubstituted thiophene“. In other words, the steric hindrance effect is
more important than the electronic effect for B-disubstituted thiophene, thus
rendering polymerization more difficult“. The fact that the peak is irreversible

suggests that the intermediate species was unstable and rather reactive5°. At a

118

Scheme Ill. Mechanism of the electrochemical polymerization reaction for

thiophene deriviatives.

119

R R R
4) / \ / \ .4. + 2/ \S -2H+I
S 3

overall reaction

RR

R R R
S S

SCHEME Ill

\

/ \
s

/
s

\/\

+ (2x)H+ + (2x)e‘

120

Figure 29. Repetitive cyclic voltammograms for the polymerization of a 25
mM solution of 3,4-Dibutylthiophene/0.02M TBAGF4 in CH3N02

at a sweep rate of 50 mV sec".

121

0N.

 

0N "NDUHH

Comm 9. 3.9.0 05208..
9..
II.

00.0

 

 

 

 

122

few millivolts more anodic than the oxidation potential of the monomer, the
potential scan was reversed, and a reduction peak assigned to the reduction of
the oxidized polymer is observed at Epc = 1.04 V. In successive cyclic
voltammograms, the intensities of the reduction and corresponding oxidation
peaks of the polymer regularly increase with the number of cycles revealing the
continuous growth of the polymer film without any loss in electroactivity. This was
confirmed by observing the electrode surface which became progressively
covered with thicker and thicker polymer films. Good quality film and high
conductivity were obtained for polymer film thickness in the 1-3 pm range as
measured using a profiler. The redox process occurring during the
electropolymerization was accompanied by a reversible change in color. The
polymer, yellowish in its neutral form, turned transparent dark blue when
oxidized. This is due to a solvatochromic effect involving a conformational
change of the polymer backbone during the charging/discharging process“? In
the charging step, the molecular chain with its twisted units becomes less planar,
and after discharging, the system returns to its original state of ordered twisting.
After electrochemical synthesis, the polymer was peeled from the electrode and
washed with acetonitrile to remove any residual monomer, small size oligomers,
or supporting electrolyte and then dried under high vacuum at room temperature
for several hours. The resulting polymer was characterized by its FT IR and NMR
spectra (Figures 30 and 31). It appears to have a regular linear structure with
predominant a a' linkages. This hypothesis was supported by the lack of
polymerization in 3- and 4-position due to the blockage of these two positions by
butyl groups which would require too much energy for the CC bond between the
thiophene and the butyl group to be broken. The addition of the butyl substituents
on the thiophene ring mainly served in improving the solubility of the

polythiophenes. Poly(3,4-dibutylthiophene) was found to be soluble in a large

123

Figure 30. Fourier transformed infra-red spectrum of poly

(3,4-Dibutylthiophene) on KBr pressed pellet.

 

124

00

A
m

—
—

on ENDGHh

1.03 000800033

 

l—Ir ,

 

 

 

00NN 00 —m

 

 

 

 

eouomwsumi as

125

Figure 31. 1H-NMR spectrum of poly(3,4-dibutylthiophene) in opera. The
peak at 7.24 ppm is due to the solvent.

126

 

 

 

 

 

Hn HMDOHH

AEO0V :30 38503

0 m
_ q

R

 

0
.i

0—

127

variety of organic solvents such as toluene, acetone, methylene chloride, and
tetrahydrofuran, thus rendering possible the study of the electrical and
physicochemical properties of this polymer in solution. Determination of the
molecular weight of this electrochemically synthesized polymer by gel permeation
chromatography using THF as eluent gave a mean value of 2000 against
polystyrene standard which corresponds to a polymer of 10 units length
(figure 32). The sharpness of the chromatographic peak suggests that the

synthesized poly(3,4—dibutylthiophene) has a narrow molecular weight

distribution.

Figure 32. Gel permeation chromatogram of the poly(3,4-dibutyl

thiophene) in THF. The flow rate is 1.5 mi min".

 

 

 

 

104 103 102

 

128

Synthesis of oligomeric and polymeric 3',4'-dibugl-a-terthioghene
3',4'-Dibutyl-or-terthiophene was prepared following a method developed

in our laboratory by M. Benz. 3,4-dibutylthiophene was first brominated using
tetramethyl ammonium tribromide (TMAT). The resulting brominated thiophene
reacted with two equivalents of thienyl magnesium bromide in presence of a

nickel based catalysis to form 3',4'-dibutyl-a-terthiophene (Scheme 4).
SchemeIV. Chemical preparation of 3',4'-dibutyl-oi-terthiophene

Br Br Br Br 3
</ \> _a_.,. A _b_.... <7 \> _c_.
8 Br 3 Br 8

a. TMAT/CHaCOzH/CHchZ
b. 2 C4H38MgBr/Pd(dppp)Cl2/Et20

U

Bu
<7 \>
S

Electrochemical polymerization of 3',4'-dibutyl-a-terthiophene was carried
out in acetonitrile using 5 10'3 M monomer and 2.5 10'2 M supporting electrolyte
(TBABF4). Figure 33 represents the successive cyclic voltammograms obtained
in the course of the polymerization process. A comparison with the cyclic
voltammograms obtained during the polymerization of the monomeric
disubstituted thiophene shows a similar shape. As expected, the oxidation peak
of the starting material is broader for longer chain length oligomers. This could be
explained by the closeness of the 15° values of the disubstituted and
unsubstituted thiophenes. It also could be explained by a poor swelling of the
material in acetonitrile. In addition, a decrease in the Epa value of the starting
material is observed. This indicates that polymerization of longer chain length

oligomers is energetically favored. The polymerization process also was

129

Figure 33. Repetitive cyclic voltammograms for the polymerization of a 25
mM solution of 3',4'-DibutyI-a-lerthiophene/0.02 M TBABF4 in
CH3CN at a sweep rate of 50 mV sec".

130

1 .

 

mm HKDOHE

0.0mm as 3.20 05208..
om... la...— owr 3..

3—l

L

 

 

 

 

 

131

accompanied by a change in color from a bright red in the neutral state to a dark

green in the oxidized state.

Synthesis of oligomeric and polymeric 3".4"-dibu_tylguinguethiophene
The last compounds studied in the disubstituted thiophene series were

 

 

3",4"-dibutquuinquethiophene and its corresponding polymer.
3",4"-dibutquuinquethiophene was synthesized following the same procedure as
that employed for the preparation of 3',4'-dibutyl-a-terthiophene. This latter
oligomer served as starting material. The electropolymerlzation of
3",4'-dibutquuinquethiophene also was performed in acetonitrile using 10‘3 M
monomer and 2.5 10’2 M supporting electrolyte (T BABF4). Contrary to the cyclic
voltammograms obtained during the polymerization of the substituted
terthiophene, three distinctive oxidation waves accompanied with the
corresponding, but not as well defined, reduction waves are observed
(figure 34). These waves are indicative of different redox processes. As
expected, the oxidation of this oligomer started at a lower potential than those of
smaller size oligomers. This confirms the idea that longer oligomers are easier to
oxidize. The resulting polymer exhibits a reversible electrochromic behavior as it
was cycled between a bright red color in the neutral state and a dark green color

in the oxidized state.

4.3 Electrochemical characterization

The synthesized poly(3',4'-dibutylthiophene) film was characterized by its
cyclic voltammogram (Figure 35a). A sharp intensive anodic wave is observed at
1.39 V, followed in the reverse scan by a broad peak potential-shift cathodic

wave of low peak current height. The peak height of the anodic and

132

Figure 34. Repetitive cyclic volammogrames for the polymerization of a 2.5
mM solution of 3",4"-dibutquuinquethiophene/0.02M TBABF4 in
CH3CN at a sweep rate of 50 mV sec".

133

m

 

vm HKDGHh

00mm 2 3.9.0 .3508“.

 

 

n+0 0qu.—

 

31

 

 

 

Figure 35.

134

Cyclic voltammograms of poly (3,4—Dibutylthiophene) (a), poly
(3',4'-Dibutyl-a-tethiophene) (b) and poly 3",4'-
Dibutyloquinquethiophene) (c) as synthesized films in
CH30NI'I'BABF4. The sweep rate was 50 mV sec“.

 

135

 

poly(3,4-dibutylthiophene)

 

 

poly(3',4'-dibutyl-a-terthiophen'e)

  

 

 

 

 

poly(3".4"-dibutquuinquethiophene)

 

 

 

 

 

 

__l_
12 ”A
_L_
l 4_ i
1.70 1.20 0.70 0.20

POTENTIAL (volts vs SSCE)

I'IGURI 35

136

cathodic waves are proportional to the sweep rate (v). This is characteristic of a
thin cell behavior34a. It is less common than the linear ip vs v“2 relationship
observed for electroactive ions freely diffusing in homogeneous solution.
Deviation from AE = 0 was interpreted as the result of nonequilibrium between
reduced and oxidized state caused by either uncompensated film resistance or,
more probably, steric effects leading to conformational changes during the
charging and discharging processes“. Cyclic voltammograms of
poly(3',4'-dibutyl-c-terthiophene) and poly(3",4"-dibutquuinquethiophene) exhibit
similar shapes (figures 35b and c). The addition of the broad anodic tail in these
voltammograms was interpreted as a pure capacitance current due to large inner
surfaces of these polymersZ4C-52. Consequently, this increase in the charging
current induces a higher reactivity of the system. As already observed for the
oligomers compared with the monomeric disubstituted thiophene, partially
disubstituted thiophene polymers having longer repeat units are easier to oxidize

when coated as thin film on the electrode surface.

Since poly(3,4-dibutylthiophene) is soluble in common organic solvents in
both its oxidized and neutral form, the electrochemical behavior of this polymer
also was studied in solution. Shown in figure 36 are the cyclic and differential
pulse voltammograms of this virgin polymer in methylene chloride. Only the
oxidative properties will be investigated since no reduction process is seen in the
negative potential window of the solvent. The polymer exhibits a broad
continuous oxidation process over a 1000 mV range. This is characteristic of a
band-type electronic structure material having a large amount of interaction
between the redox sites comprising the assembly of the polymer3°a»35. The
shape of its cyclic voltammogram does not change when varying the sweep rate

and the peak heights of the anodic and cathodic waves are proportional to v1 ’2

Figure 36.

137

Cylic and differential pulse voltammograms of a 273 (M solution
of poly (3,4—dibutylthiophene)/0.2 M TBABF4 in CHZCIZ. The
experimental conditions for differential pulse voltammetry were:
pulse amplitude: 50 mV; pulse width: 50 mV sec"; pulse
duration = 1000 mV sec'1 and scan rate = 4 mV sec". for cyclic

voltammetry, the sweep rate was 50 mV sec'1 .

138

0N.0l

 

00.0

 

on HKDOHh

Comm m> 0:05 3:330

err

|_|
<3. no
LI

0N.0l

 

l-é—i ,

 

 

E

 

 

139

suggesting a diffusional behavior contrary to the one observed in the solid state.
The polymer also can be cycled repeatedly between the oxidized and the neutral
state with no appreciable decomposition of the material rendering possible the
study of the redox properties of the poly(3,4-dibutylthiophene) in the solution

chemistry.

4.4 Redox properties

Previous studies on the soluble polyalkylthiophenes have shown that
these polymers, when coated as thin films on the electrode surface, usually can
be oxidized from 0 to 30%22. The maximum doping level attained in the solid
state can vary significantly, depending on the nature of the counterion and
structural properties of the film‘33v22148. In solution, these complications do not
occur, thus rendering possible the d0ping of these polymers at higher levels. In
order to determine the doping profile as function of potential for
poly(3,4-dibutylthiophene) in solution, a series of controlled potential coulometry
experiments was performed in methylene chloride. As was expected from
theory49, current curves which decay exponentially with time to the background
current value were obtained except for the curve of the re-reduction of the 34%
oxidized solution. In this case, the polymer remains partially oxidized. It can not
be fully re-reduced even after several hours of the undoping process. The
percent oxidation-potential plot determined from this series of CPC experiments
for the poly(3,4-dibutylthiophene) is shown in Figure 37. Each point of the graph
represents the percent oxidation determined by averaging the amount of charge
passed for the oxidation of the) neutral polymer and the reduction back to its
neutral state both in solution and in the solid state. The error bars depict the
differences between these values. A smooth doping profile is seen, suggesting

that there is a continuum or band into which charge can be added. The

 

Figure 37.

140

Degree of partial oxidation vs. potential determined by controlled
potential coulometry for poly(3,4-dibutylthiophene)/TBABF4 in
CH2Cl2 solution and as film. Each point ( ) corresponds to the
mean value for an oxidation/re-reduction cycle. The error bars

represent the difference between the oxidation and re-reduciton.

141

Lab

#0 “MDGHh

Comm m> 33$ 3:330

 

 

000 £03 pm..—

 

 

00.N

l
O

r
O
N

l
O .
P0

10v

 

 

431’

row
{4
1111111.
no
idua‘ll

u000le as

142

differences in the percent oxidation at a given potential indicate that the polymer
is stable up to a value of 30% and, that above this level, some instability inherent
to the polymer is observed. This was proved by looking at the number of
coulombs passed for the oxidation which differs largely from that for the re-
reduction. It appears that the reduction of the doped polymer, when it is more
than 30% oxidized, is not complete. In other words, access to the bipolaron band
for external electrons is more difficult. Similar behavior was found for
poly(3,4-dibutylthiophene) film. Tourillon and Gamier observed the same
phenomenon and concluded that the doping process is more difficult to achieve
for polydialkylthiophenes because of the steric hindrance effect of the bulky

substituents on the thiophene ring49.

4.5 Optical properties

Figure 38 represents the visible absorption spectra of both the solution
and film of the oxidized and reduced state of mly(3,4~dibutylthiophene). In both
cases, the neutral (undoped) polymer has a strong absorption band in the
320 nm region which corresponds to the 1: - 1t. transition from the valence band
to the conduction band. Another common feature of the polymer film and its
solution is that upon doping, the interband transition decreases and
simultaneously a new absorption peak appears at higher energy. These
similarities in absorption spectra suggest that the electronic structure of
poly(3,4-dibutylthiophene) is unchanged after dissolution. The shift of the
absorption band corresponding to the interband transition implies a
conformational change in solution, presumably to a more disordered

conformation.

 

Figure 38.

143

ln-situ optical absorption spectra of poly(3,4-dibutylthiophene) as
synthesized film and solution in both its neutral (undoped) and

oxidized form.

 

 

 

 

144

O
-|o
CD

on

3.30 50.3333

010 m

HMDUHh

0%0

O
O
(D

 

 

 

00000

 

Hmuuscc

can 0.8

.3.0

 

 

-1

 

00000

5.5.8

 

Hmuuscc

q

00m

 

 

 

145

In order to determine the nature of the charge-storage species in the
polymer chain during the doping process, a series of electronic absorption
spectra of the poly(3,4.dibutylthiophene) in dilute methylene chloride solution was
obtained at various doping levels (y) (Figure 39). In the neutral state, a peak
corresponding to the interband transition is observed at 3.9 eV. Upon doping,
(0.04syso.34), a defined peak is observed at 3.9 eV along with two broad peaks
at 2 and 1 eV. This is consistent with the concept of doping through bipolaron. As
the doping level increases, the subgap absorptions grow at the expense of the
interband transition. The high dopant level data were consistent with previously
reported optical spectra of heavily doped polythiophenes and their
defivatives‘ka-wfl. This suggests that, at high doping levels, charge is

predominantly stored in bipolaron.

It is well known that the existence of n-conjugation in conducting
polymers is implied by their optical spectra. In order to evaluate the amount of
conjugation in polydisubstitutedthiophenes, the absorption spectra of the neutral
poly(3.4-dibutylthiophene), poly(3',4'-dibutyl-<x-terthiophene) and poly(3',4"-
dibutquuinquethiophene) films were recorded and compared (Figure 40). These
spectra reveal a slight shift of the absorption maximum of poly(3',4'vdibutyl-a-
terthiophene) compared with poly(3,4-dibutylthiophene) towards lower
wavelength, indicating a lower conjugation for the latter polymer.
This shift is even more strongly marked in the case of
poly(a",4"-dibutquuinquethiophene). The displacement of the absorption maxima
indicates that the conjugation in these polymers increases when increasing chain
length of the starting oligomer with unsubstituted thiophene units. As a result, a
better conducting polymer is formed. Further support for this conclusion will be

provided by the conductivity data of these polydisubstitutedthiophenes.

 

146

Figure 39. ln-situ electronic absorption spectra of a 273 uM solution of

poly(3,4-dibutylthiophene)/O.2 M TBABF4 oxidized at various
degrees in CHZCIZ.

 

147

an HMDOHH

gov >925

 

 

 

swan»
_Qou>

enou>

 

 

 

 

 

 

r
F

 

('n'o) eouquosqo

148

Figure 40. Optical absorption spectra of poly(3,4-dibutylthiophene),
poly(3'4'-dibutyl-a-terthiophene). and poly(3",4"-dibutquuinque

thiophene) as synthesized films.

 

 

149

 

 

0v INDUHh

A55 50:22.63
con owl...“

IIDD’D

 

A29.83sac:_na_33_u-.v..m:_oa
85:82:23-3.3octagon. .l
858253.30-...928 .l

 

T
F

 

 

('n°o) eouquosqo

150

4.6 Conductivity

The study of the electrical properties of poly(3,4-dibutylthiophene) has
been carried out as function of the doping level (y). The results of the electrical
conductivity data at room temperature (031’) are summarized in Figure 41. The
conductivity increases dramatically with doping in a manner typical of conducting
polymers, i.e., the profile of conductivity upon daping corresponds to a smooth,
continuous process. The polymer is insulating in its neutral form. Upon doping,
conductivity increases up to 9 orders of magnitude. The highest conductivity of
1 S cm'1 was obtained for a maximum doping level of 0.34. The comparison of
the conductivity values of highly doped poly(3,4-dibutylthiophene) with other
derivative polythiophenes in Table 5 allows some comments to be made.
Contrary to what could be expected, polyalkylthiophenes and
polydialkylthiophenes have, in general, lower electrical conductivity (except for
poly(3-methylthiophene)) than the parent polythiophene in their oxidized state.
This can be explained by the fact that, when bulky groups are substituted on B-
carbon atoms in the thiophene unit, steric hindrance forcing adjacent rings to
twist out of coplanarity increases, leading to less conjugation in the polymeric
chain and lower conductivity on doping. This effect is even greater for
polydialkylthiophenes possessing large substitutent groups. Another interesting
observation concerns polythiophene and their derivatives synthesized from
oligomers of different chain lengths instead of the usual monomer. When going
from polythiophene to polytrithiophene, the conductivity decreases upon
doping24avc. In other words, the use of oligomers of increasing chain length as
starting materials for the polymerization of unsubstituted polythiophenes leads to
less conducting polymers with a lower degree of conjugation. However, if the
starting material contains both unsubstituted and substituted thiophene units. the

conductivity increases with the number of thiophene units toward the value

151

Figure 41. Electrical conductivity vs. degree of partial oxidation at room

temperature for the poly(3,4-dibutylthiophene).

152

Ian!

00.0

 

D
C

 

Acozono .0212. .o 3.503 >

35

av HMDUHH

0N6

1
q

and
all

ovd

_

 

 

odl

«a ‘t as
O O O

(l..t.uo s) 18 o

“9
o

o.—

 

153

Table 5. Room temperature electrical conducitivity data for B-mono and

-disubstituted thiophene derivatives.

 

 

154

 

om £85 2%: 83:30: AcamcooEumsocSviusflnl..v ...2 Son
om :35 .59 83188 H AmcmcooEfimulallusfloi.v :2 30m
ooava.o czocxcs Amcmnaoflnuumui=~..N.NV>H0Q
Eifm 565:: Amcmnaofifinl.m.~;3a
H :2: the 52:20: 352823338; .2 Soc
513 c_om.fmmomo$ asmso: 132323335048.38..
mzov £25sz 82115: 652851304538
9.33 53.0238 3:00: .mcmfiofiuifimelmvioo
:13 £852.38 83.6: 3933538
.48 2
>ufl>fluus0cou cofluflmoaeoo umE>Hom

 

 

m Hflmdfi

155

obtained for the unsubstituted polythiophene, synthesized from the monomer.
Similar behavior was observed for copolymers based on polyalkylthiophene and
polythiophene‘7". Plots of conductivity as function of doping for the
poly(3',4'-dibutyl-a-terthiophene) and poly(3",4"-dibutquuinquethiophene) indicate
that these polythiophenes also have a smooth continuous
conductivity-doping profile starting at 17% and levelling off around 30%
(Figure 42). A higher conductivity value is reached at lower doping levels.
Compared with that of the parent thiophene, the conductivity increases by more

than 10-fold, even at low doping levels.

156

Figure 42. Electrical conductivity vs degree of partial oxidation at room
temperature for poly(3,4-dibutylthiophene), poly(3',4'-dibutyl-a-
terthiophene), and poly(3'4"-dibutquuinquethiophene).

157

 

 

Nv HMDQHh

AcozoExo .0223 v3 09503 >
o...o ,- ion... - oumo . and 010 no

1.“ n5. 0

 

 

'4

to—

wow

'4

U

r
O
m

182825353923691.228 4

I
O
V.

8:282:23330$3828 a

653253.364.928 o

 

 

(l-wo s) “a

5. CONCLUSION

Conventional and unconventional electrochemical techniques can be
used to characterize soluble conducting polymers. Contrary to the
electrochemical response observed for conventional electroactive compounds,
conducting polymers exhibit a broad, continuous response in solution. This
behavior is characteristic of a material having a large amount of interaction
between its redox centers. Despite the complications accompanying the
electrochemical study of such systems, it has been possible to accurately probe
the redox properties of these conducting materials to a saphisticated level never

attempted before.

The partial reduction of the u—oxo-(tetra-t-butylphthalocyaninato)
germanium can be accurately and systematically varied to any specific level,
ranging from O to 1- by simply changing the potential of the working electrode.
The reduction process can be described as a simple charge-transfer reaction
occurring at the electrode surface, only limited by the mass transport of the
material to the electrode. The germanium polymer is chemically stable at all
degrees of partial reduction and remains stable upon repeating reduction/re-
oxidation cycles. At greater than 100% reduction, the polymer quantitatively
fragments into electroactive monomeric species. This decomposition can be
easily, quickly, and precisely monitored using modified rotating ring—disk
voltammetry technique. Besides the advantages of easiness, rapidity, and
precision offered by this technique, the solution remains intact during the entire

experiment. Digital simulations of the electrochemical responses indicate that the

158

a‘. "5““.

 

*3 In .1 -. .1

159

reduction of the germanium polymer can be modeled as a series of reversible
redox couples in equilibrium undergoing sequential one-electron transfers. From
the simulations of the conventional voltammetry response, the standard
potentials of the redox sites comprising the polymer are determined along with an
estimation of the amount of interaction between the redox sites. From the
simulations of the unconventional voltammetry responses, the mechanism of the
decomposition is elucidated and the rate constant of the decomposition

measured.

In a fashion similar to the u—oxo—(tetra-t-butylphthalocyaninato)
germanium, soluble poly(3,4-dibutylthiophene) can be oxidized to any level
between 0 and 0.34 and remains chemically stable. Similar behavior is observed
for the corresponding film. It is the first time that the same probe has been
investigated simultaneously in solution and in the solid state. Conductivity
measurements of the polythiophene derivative oxidized to various degrees are
directly related to potential, thus allowing the determination of the conductivity
profile as function of band-filling. The optical properties of this polymer as
function of band-filling are studied, using in-situ experiment combining
electrochemical and spectroscopic techniques. lnforrnation on the nature of the

charge carriers during the redox process is provided in this experiment.

Such study is valuable in the long term goal for developing new material

with predictable properties.

 

APPENDIX 1

The digital simulation models employed for both differential pulse and
rotating ring-disk voltammetry techniques are based on the method of finite-
differencesaz. The solution near the electrode surface is divided into volume
elements of equal size. The movement of species between adjacent volumes is
used to mimique diffusion. Parameters are handled in dimensionless form in
accordance with standard procedure. The main difference between this model
and the others discussed in the literature are the equations used to describe the
surface boundary conditions and the dimensionless current for a multi-

component system such as the conducting polymer.

In these models, it is assumed that each phtalocyanine ring (Pc) is an
active redox site having its own and unique standard potential (Ejo) in
equilibrium with the other sites comprising the assembly of the polymer“. Based
on the CPC and RDE data (Nernstian one electron transfer, one electron
removed per Pc site at 100% reduction), the reduction of the polymer can be
viewed as a system undergoing sequential Nernstian one-electron transfer:

an +e' <=> Rn-l 01 E10

RM 01 + e' <=> RM 02 £20

Rn_2 02 +e' <=> Rn_3 03 E30

etc...

where :

R = neutral species

0 = oxidized species

n = length of the polymer

160

 

161

The polymer was found to be comprised of 15 redox sites, based on the
results obtained from the RDE voltammetry experiments. To generalize the
computational process, the concentration of any form of the polymer, regardless
of what redox state it is in, is normalized with respect to the initial concentration.
Since the system is reversible, the concentration at the electrode surface is
governed by the Nemst equation.

The fractional concentration of each species at a given potential is
derived from the potential-dependent equilibrium constant describing each redox

reaction. The equilibrium constants (Kn) of each redox site are :

 

 

 

 

 

 

[a _ o 1 RT
K1: "1 1 =exp(E—E1°)*
[Rn] n1F
[R , o 1 RT
K2: "2 2 =9Xp(E'E20)*‘
[Fin] n2F
[R , 0 RT
K3= "3 3] =exp(E-E3°)*
etc...

The initial bulk concentration [Onlo calculated from the mass balance is :
[Rnlo = [Fin] + [Rn-1 01] + [Rn-2 02] + [Rn-3 03] + .. + [On]

Using this new expression for [Rnlo and including the equilibrium
constants in the expression of concentrations, the fractional concentrations (an)

of each species at the electrode surface are :

 

 

 

 

[Rn] 1
“0 [ano 1+K1+K1K2+K1K2K3-r-...+K1K2K3...Kn
[Rn-1 01] K1
a1 = a

[ano 1+K1+K1K2+K1K2K3+...+K1K2K3...Kn

162

[Rn-2021 We
a = =
2 [Rnlo 1+K1+K1K2+K1K2K3+...+K1K2K3...Kn

 

 

etc...
with (104-a1 +or2+...+an = 1
This model is analogous to the one used for the dissociation of a
polyfunctional acid or base3°b. The real modifications made in the DPV and
RRDE simulation programs concern the flux expression used in the
determination of current. It is important to note here that, even if the DPV and
RRDE simulation programs are constructed following the same scheme, these
two techniques will be treated separatively due to their different modes of mass
transfer. In DPV, mass transfer occurs only by diffusion, linear in character. In
RRDE, a radial term is added to the linear component of the diffusion and a
convection term is added to the diffusion one in the mass transfer equation. For
an easy understanding, the simulation program of DPV is described first, below,
and in detail and will serve as a basis for the more complicated RRDE simulation

program.

Assuming that diffusion is the only mode of mass transfer and that
diffusion is linear, the general expression for the flux of species j (Fj) at the

electrode surface is the following:

-i 5C1 (Xpt)
Pr = njm = ‘01 5,, x=0

where:

 

 

i = current
nj = number of electrons transferred

F = Faraday's constant

163

A = area of the electrode

01- = diffusion coefficient

[Ely—'2. ”0 = concentration gradient at the electrode surface
it

In simulation term, the flux expression (Fj) becomes:

F] = 0M] [C(j+1,t) - C(i.t)l

where:
DMj = dimensionless diffusion current of species j
C(j,t) = concentration of species j in the j volume element

C(j+1,t) = concentration of species j+1 in the j+1 volume element

For the model that has been developed in this study, the expression for
the individual flux (Fj) can be written as:

F] = DMj [C(j+1,t) - aim]

From the flux, the dimensionless current is calculated. It is the sum of the
individual flux terms scaled to some proportionality constant depending only on

the electrochemical technique used. 1
n

dimensionless current: 2 Z (n+1-k) * thk) * C)
where: j=1 k=1
n a number of redox sites

C = simulation constant (DMj, Ax, At)

Likewise the percent of charge transfer at a given potential is:
n J

(Rn- 0]
Z -‘-‘-m.i. Z “k *
j=1 k=1
where:

nk'z total number of electrons involved in the j redox step

The other variables are defined as before.

 

164.

A problem encountered with the computation of the flux terms is that, as
the number of sites increases and/or the the difference between the standard
potential of each redox couple becomes large, the common dominator used to
calculate the individual fluxes becomes unmangeable. For a standard potential
more negative than the applied potential, the equilibrium constant quickly
becomes extremely small. Conversely, for a standard potential more positive
than the applied potential, the equilibrium constant becomes exceedingly large.
When the number of sites is large, the product of the equilibrium constants used
to calculate the fractional concentrations at the electrode surface becomes
difficult to handle. To alleviate this problem, a ”sliding potential window” is used
to determine which equilibrium constants are needed to describe the entire
system. If the applied potential is 300 mV more negative than the E0 value, the
redox reaction is assumed to have gone to completion and the fractional
concentration of all species needed to generate that component are set to zero.
Likewise, if the applied potential is 300 mV more positive than the E0 value, the
species of interest has not been formed yet, and again its fractional concentration
is set to zero. A value of 300 mV was chosen since it represents, for a one-
electron process, a ratio of 10'5 for one form of the couple over the other
(0.3RT/F = log (Ox/Red». This procedure has the additional benefit to effectively
cutting down the number of equilibrium constants that are needed to characterize

the entire system, resulting in shorter processing time.

For the RRDE, the following model is used. The solution is divided into
thin parallel layers and each layer into a cylindrical box and a series of concentric
annular boxes. In this model, several regions are considered: the disk, the
insulating gap between the disk and the ring, the ring and the outer insulation.

Unlike previous work, the number of boxes assigned to the disk or ring is no

165

longer a constant, but depends on the diameter of the electrode considered. For
each region the flux is first calculated from the mass transfer equation which
includes terms of diffusion and convection. Then the disk and ring currents are
calculated in a manner similar to the one described earlier. The ring current also
is calculated when the disk is held at a certain potential. Finally the decompsition
of the germanium is monitored by simulation of the continuous shielding-
collection experiments. The decompositon is considered as a first order
homogeneous chemical reaction, i.e., the polymer decomposes totally into the
corresponding monomeric type units, in anion form which are further reduced into
dianions and trianions. This decomposition is part of the EnCEE mechanism
described earlier. Thus, a new set of fractional concentrations taking into account
the decomposition process are calculated from which the flux and current are
derived. From this simulation, a rate constant of 20 sec‘1 was obtained for the

decompostion of the germanium polymer.

J-J'

 

(b)
(c)
(d)
(e)
(f)
(g)
(h)

(i)

(a)

(b)
(c)

(d)

 

REFERENCES

Skotheim, T.A., Ed. "Handbook of Conducting Polymers”, (Dekker,
New York, 1986) Vols. 1-2. .

Ferraro, J.R.; Williams, J.M. “Introduction to Synthetic Electrical
Conductors", (Academic Press, New York, 1987).

Skotheim, T.A., Ed. "Electroresponsive Molecular and Polymeric
Systems", (Dekker, New York, 1988) Vol. 1.

Miller, J.S., Ed. ”Extended Linear Chain Compounds", (Plenum
Press, New York, 1982) Vols. 1-3.

Ward, M.W. "Electroanalytical Chemistry"; Bard, A.J., Ed. (Dekker,
NY, 1989) Vol. 16, 181-312.

Miller, J.S.; Epstein, A.J. ”Progress in Inorganic Chemistry”;
Lippard, S.J. Ed. (Wiley, New York, 1976) Vol. 20, 1-151.
Proceedings of the International Conference on Science and
Technology of Synthetic Metals (ICSM '86), Synth. Men, 1987, 17.
Proceedings of the International Conference on Science and
Technology of Synthetic Metals (ICSM '88), Synth. Met., 1988-
1989, 27-29.

Epstein, A.J.; Conwell, E.M. Eds. "Proceedings of the International
Conference on Low-Dimensional Conductors", Boulder Colorado,
Au ust 9-16, 1981, Mol. Cryst. Liq. Cryst.,77, 79, 83, 85, 86, Parts
A- (1981-1982).

Chiang, C.K.; Fincher, C.Fl. Jr.; Park, Y.W.; He er, A.J.;
Shirakawa, H.; Louis, E.J.; Gau, S.C.; MacDiarmid, A.G. hys. Rev.
Left, 1977. 39. 1098-1101.

Shirakawa, H.; Louis, E.J.; MacDiarmid, A.G.; Chiang, C.K.;
Heeger, A.J. J. Chem. Soc. Chem. Commun., 1977, 578-580.
Chiang, C.K.; Druy, M.A.; Gau, S.C.; Hee er, A.J.; Louis, E.J.;
MacDiarmid, A.G.; Park, Y.W. J. Am. Chem. ca, 1978, 100, 1013-
1015.

Chiang, C.K.; Park, Y.W.; Heeger, A.J.; Shirakawa, H.; Louis, E.J.,
MacDiarmid, A.G. J. Chem. Phys, 1978, 69, 5098-5104.

Albery, W.J.; Bartlett, P.N.; Cass, A.E.G.; Craston, D.H.; Haygett,
86.0. J. Chem. Soc. Faraday Trans. l, 1986, 82, 1033-1050.
Heeger, A.J.; Moseo, D.; Sinclair, M. Synth. Met, 1986, 15, 95-104.
Alper, J. Science, 1989, 246, 208-210.

Duke, C.B. Synth. Met., 1987, 21, 5-12.

lnganos, 0.; Lundstrom, l. Synth. Met, 1987, 21, 13-19.
MacDiarmid, A.G. Synth. Met, 1987, 21, 79—83.

oowan, o.o.; Wiygul, F.M. chem. Eng. News, 1986, 64, 28-45.
Bryce, M. H.; Murphy, LC. Nature, 1984, 309, 119-126.
Wudl, F. Acc. Chem. Res, 1984, 17, 227-232.

Scrosati, B. Prog. Solid State Chem, 1988, 18, 1-77.

Wegner, G. Angew. Chem. Int. Ed. Eng., 1981, 20, 361-381.
Reynolds, J.R. Chemfech., 1988, 18, 440.447.

166

10.

11.

12.

13.

14.

(d)
(9)

167

Heinze, J.; Storzbach, M.; Mortensen J. Ber. Bungsenges. Phys.
Chem, 1987, 91, 960-961.
Heinze, J. Top. Curr. Chem, 1990, 152, 1-47.

Greene, R.L.; Street, G.B. Science, 1984, 226, 651-656.

 

a
b
c

13

Hoffmann, R. Angew. Chem. Int. Ed. En ., 1987, 26, 846.
Duke, B.J.; O'leary, B. J. Chem. Ed., 1 , 65, 319-321.
Duke, B.J.; O'leary, B. J. Chem. Ed., 1988,65, 379-383.

Marks, T.J. Science, 1985, 227, 881-889.
Marks, T.J. Angew. Chem. Int. Ed. Eng., 1990, 29, 857-890.

Whangbo, M.-H. Acc. Chem. Res., 1983, 16, 95-101.

(a)

a)
”l

c
d)

i
r

Sato, M.; Tanaka, S.; Kaeriyama, K. J. Chem. Soc. Chem.
Commun., 1985, 713-714.

Satoh, M.; Kaneto, K.; Yoshino, K. Jpn. J. Appl. Phys., 1985, 24,
L423-450.

Roneali, J.; Gamier, F. Nouv. J. Chim., 1986, 10, 237-240.

Roneali, J.; Gamier, F.; Lemaire, M.; Garreau, R. Synth. Met,
1986. 15, 323-331.

Grant, P.M. Phys. Rev. 8.: Conden. Matter, 1982, 26, 6888-6895.
Grant, P.M. Bull. Am. Ph 3. Sec, 1981, 26, 213-217.

Weger, M.; Kaveh, M.; utfreund, H. Solid State Commun., 1981,
37, 421-426.

Gutfreund, H.; Hartstein, C.; Weger, M. Solid State Commun.,
1980, 36, 647-651.

099:9", E.M. Phys. Rev. B: Condens. Matter, 1980, 22, 1761-
1 .

Epstein, A.J.; Conwell, E.M.; Miller, J.S. Ann. NY. Acad. Sci, 1978,
313, 183-209.

Mizutari, F.; lijima, S.; Tanabe, Y.; Tsuda, K. J. Chem. Soc. Chem.
Commun., 1985, 1728-1729.

Fan, F.-R.; Bard, A.J. J. Electrochem. Soc, 1986, 133, 301-309.
Pegner, R.M.; Martin, C.R., J. Electrochem. Soc., 1986, 133, 310-
31 .

Iyoda, T.; Dhtani, A., Shimidzu, T.; Honda, K. Chem. Lett., 1986,
687-690.

Chung, T.C.; MacDiarmid, A.G.; Feldblum A.; Heeger, A.J. J. Chem.
Phys., 1981, 74, 5504-5507.

131
131
r

c
d)

Diaz, A.F.; Lacroix, J.C. New J. Chem, 1 988, 12, 171 -1 80.
Wudl, F. lsreal J. Chem, 1 986, 27, 289-292.

Diaz, A.F.; Logan, A.J. J. Electroanal. Chem, 1980, 111, 111-114.
Diaz, A.F.; Clark, T.C. J. Electroanal. Chem, 1980, 111, 115-117.
Diaz, A.F.; Castillo, J.I.; Logan, J.A.; Lee, W.Y. J. Electroanal.
Chem, 1981 , 129, 115-132.

Diaz, A.F.; Martinez, A.; Kanazawa, K.K. J. Electroanal. Chem,
1981. 130, 181-187.

15.

16.

17.

(9)
(f)
(9)
(h)

(J)

(a)
(b)
(C)

(a)
(b)
(C)
(d)

(9)
(f)

QOO'DJ

(9)
(f)

(i)
(k)
(I)

168

Bull, R.A.; Fan, F.-R. F.; Bard, A.J.; Castillo, J; Kanazawa, K.K.;
Logan, J.A. J. electroanal. Chem, 1982, 129, 1009-1015.

Diaz, A.F.; Castillo, J.; Kanazawa, K.K.; Logan, J.A. J. Electroanal.
Chem, 1982, 133, 233-239.

Tognllon, G.; Gamier, F. J. Electroanal. Chem, 1982, 135, 173-
Gamier, F.; Tourillon, G.; Gazard, M.; Dubois, J.C. J. Electroanal.
Chem, 1983, 148, 299.

Genies, E.M.; Bidan, G.; Diaz, A.F. J. Electroanal. Chem, 1983,
149,101-113.

Waltman, B.J.; Bargon, J.; Diaz, A.F. J. Phys. chem, 1983, 87,
1459-1463.

Feldberg, S.W. J. Am. Chem. Soc., 1984, 106, 4671-4674 and
references therein.

Murray, R.W. 'Electroanalytical Chemistry”; Bard, A.J. Ed., (Dekker,
NY, 1986 Vol. 13.

Chung, .C.; MacDiarmid, A.G.; Feldblum, A.; Heeger, A.J. J.
Chem. Phys., 1981, 74, 5504-5507.

Gaudiello, J.; Ma , H.O.; McCarthy, W.J.; Moguel, M.K.;
Kannewurf, C.R.; Ma s, T.J. S nth. Met, 1986, 15, 115-128.
Gaudiello, J.; Almeida, M.; arks, T.J.; McCarthy, W.J.; Butler,
J.C.; Kannewurf, C.R. J. Phys. chem, 1985, 90, 4917-4920.
Almeida, M.; Gaudiello, J.; Butler, J.C.; Marcy, H.O.; Kannewurf,
C.R.; Marks, T.J. Synth. Met, 1988, 27, 261-266.

Kellogg. G.E.; Gaudiello, J.; Schlueter, J.A.; Tetrick, S.M.; Marks.
T.J.; Mar , H.O.; McCarthy, W.J.; Kannewurf, C.R. Synth. Met,
1989,29, 15-F24.

Gaudiello, J.W.; Kellogg. G.E.; Tetrick, S.M.; Marks, T.J. J. Am.
Chem. Soc., 1989, 111, 5259-52271.

Almeida, M.; Gaudiello, J.G.; Kellogg, E.K.; Tetrick. S.M.; Marcy,
H.O.; McCarthy, W.J.; Butter, J.C.; Kannewurf, C.R.; Marks, T.J. J.
Am. Chem. Soc., 1989, 111, 5271-5284.

Frommer, J.E. Acc. Chem. Res, 1986, 19, 2-9.

Li, S.; Cao, Y.; Xue, Z. Synth. Met, 1987, 20, 141-149.

Jiang, H.; Don . S. Synth. Met, 1988, 24, 255-265.

I1El7saenbaumer, .L.; Jen, K.Y.; Ododi, Fl. synth. Met, 1986, 15, 169-
Elsenbaumer, R.L.; Jen, K.Y.; Miller, G.G.; Shacklelte, LW. Synth.
Met, 1987, 18, 277-282.

Patil, A.D.; lkenove, Y.; Basecu, N.; Colaneci, N.; Chen, J.; Wudl,
F.; Heeger, A.J. Synth. Met, 1987, 20, 151-159.

Hotta, S. Ssynth. Met, 1987,22, 103-113.

Hotta, S.; oga, M.; Sonada, N. Synth. Met, 1988, 16, 267-279.
Ruiz, J.P.; Nayak, K.; Marynick, D.S.; Reynolds, J.R.
Macromolecules, 1 989, 22, 1 231 -1 238.

Rughooputh, S.D.D.V.; Novak, M.; Hotta, S.; Heeger, A.J.; Wudl, F.
Synth. Met, 1987, 21, 41-50.

Sato, M.-A.; Tanaka, S.; Rughooputh, S.D.D.V.; Heeger, A.J.;
Wudl, F. Macromolecules, 1987, 20, 212-215.

Roncoli, J.; Garreau, Ft; Yassar, A.; Ma ue, P.; Gamier, F.;
Lemaire, M. J. Phys. Chem, 1987, 91, 6706-6 14.

18.

19.

20.

21.

m)

(o)
(p)
(q)

(r)
(a)

(a)
(b)
(c)
(d)

(6)

(f)

(9)
(h)
(i)

(a)
(b)
(C)
(d)

(a)
(b)

169

Nalwa, H.S. Synth. Met, 1990, 35, 387-391.

Patil, A.D.; lkenove, Y; Wudl, F.; Heeger, A.J. J. Am. Chem. Soc.,
1987. 109. 1858-1859.

Patil, A.D.; lkenove, Y.; Wudl, F.; Heeger, A.J.; Synth. Met, 1987,
20, 151 -159 and references therein.

$5235 Pawlowowski, G.; Hanack, M. Z Naturforsch., 1983, 386,
Hanack, M.; Datz, A.; Faye, R.; Fischer, K.; Keppeler, V.; Kach, J.;
Metz, J.; Mezger, M.; Schneider, 0.; Schulze, H. "Handbook of
Conducting Polymers"; Skelheim, T.A., Ed., (Dekker, NY, 1986)
Vol. 1 and references therein.

Hanack, M.; Vermehren, P. Synth. Met, 1989, 32, 257-261.

Hanack, M.; Mitulla, K.; Pawlowski, G.; Subramanian, L.R. J.
Organometal. Chem, 1981, 204, 315-325.

Hanack, M. Mol. Cryst. Liq. C st, 1988, 160, 133-137.

Hanack, M.; Lan e, A.; Rein, .; Babnisch, R.; renz, G.; Leverenz,
A. Synth. Met, 1 89, 29, 71-78.

Diel, B.D.; Inabe, T.; Lyding, J.W.; Schoch, Jr., K.F.; Kannewurf,
C.R.; Marks, T.J. J. Am. Chem. Soc., 1983, 105, 1551-1567.
Petersen, J.L.; Schramm, C.S.; Stojakovic, D.,R.; Hoffman, B.M.;
Marks, T.J. J. Am. Chem. Soc., 1977, 99, 286-288.

Schramm, C.S.; Scaringe, R.P.; Sko'akovic, D.R.; Hoffman, B.M.;
Ibers, J.A.; Marks, T.J. J. Am Chem. as, 1980, 102, 6702-6713.
Diel, B.D.; Inake, T.; Jag i, N.K.; Lyding, J.W.; Schneider, 0.;
Hannack, M.; Kannewurf, .R.; Marks, T. ., Schwartz L.H. J. Am.
Chem. Soc., 1984, 106, 3207-3214.

Inabe, T.; Gaudiello, J.G.; uel, M.K.; Lyding, J.W.; Burton, R.L.;
McCarthy, W.J.; Kannewurf, .R.; Marks, T.J. J. Am. Chem. Soc.,
1986, 108, 7595-7608.

Marks, T.J.; Schoch, K.F. Jr.; Kundalkar, B.R. Synth. Met, 1979-
1980, 1, 337-347.

Brant, P; Nohr, R.S.; Wynne, K.J.; Weber, D.C. Mol. Cryst, 1982,
81, 255-263.

Dirk, C.W.; Inabe, T.; Schoch Jr., K.F.; Marks, T.J. J. Am. Chem.
Soc., 1983, 105, 1539-1550.

Dirk, C.W.; Marks, T.J. Inorg. Chem, 1984, 23, 4325-4332.

Sugismoto, R.; Takeda, S.; Gu, M.B.; Yoshino, K. Chem. Express,
19 , 1. 635-642.

Ega-rggr, J.; Huddleston, P.R.; Wood, ML Syn. Commun., 1975, 5,
Kobayashi, M.; Chen, J.; Moraes, T.-C.; Heeger, A.J.; Wudl, F.
Synth. Met, 1984, 9, 77-86.

Yen, K.Y.; Miller, G.G.; Elsenbaumer, R.L. J. Chem. Soc. Chem.
Commun., 1986, 1346-1347. .

Genies, E.M.; Bidan, G.; Diaz, A.F. J. Electroanal. Chem. lnterfacial
Electrochem, 1983, 149, 101-113.

Fuyimoto, H.; Nagashima, V.; lnokrechi, H.; Scki, K.; Nakahara, H.;
Hakayana, J.; Hoshino, M.; Fukudo, K. J. Chem. Phys, 1988, 89,

. 1198-1199.

22.
23.
24.

25.

26.

27.

28.

29.

30.

31.

170

Watman, B.J.; Baugen, J. Can. J. chem, 1986,64, 76-95.
Sato, M.-A.; Tanaka, 8.; Kaeriyama, K. Synth. Met, 1987, 18, 229-232.

a
b
C)

(d)

 

(9)

Roncalier, J.; Gamier, F. Synth. Met, 1986, 15, 323-331.

Krische, l.; Zagorska, M. Synth. Met, 1989, 33, 257-267.

Heinze, J.; Martensen, J.; Hinkelmann, K. Synth. Met, 1987, 21,
209-214.

Kulszewicz-Bajer, l.; Za orska, M.; Genoud, F.; Kruszka, J.;
Sggcka, K; Pron, A.; Nec tschein, M. Synth. Met, 1990, 35, 129-
1

Funt, L.; Lowen, S.V. Synth. Met, 1985, 11, 129-137.

Schriver, D.E. 'T he Manipulation of Air-Sensitive Compounds", (McGraw-
Hill, New York, NY 1969).

Smith, W.H.; Bard, A.J. J. Am. Chem. Soc., 1975, 97, 5203-5210.

Adams, R.N. "Electrochemistry at Solid Electrodes”, (Marcel Dekker, NY,
1969) and references therein.

(a)
(b)

(a)
(b)
(cl
(0)
(e)

(a)
(b)

Genies, E..;M Hany, P.; Lapkowski, M.; Santier, Ch.; Olmedo, L.
Synth. Met, 1988, 25, 29-37.

Olmedo, L.; Chanteloube, l.; Germain, A.; Petit, M. Synth. Met,
1989.30.159-172.

3am, A9..é.;)Faulkner, C.R., ”Electrochemical Methods", (Wiley, New
0 , 1 0.

Rieger, P.H., "Electrochemistry', (Prentice-Hill, Englewood Cliffs,
New Jersey 1987).

Kissenger, P.T.; Heiniman, W.R., Eds. “Laboratory Techniques in
electroanalytical Chemistry", (Marcel Dekker, New York, 1984).
Harrar, J. E. "Electroanalytical Chemistry"; Bard, A.J. Ed., (Dekker,
NY, 1975) Vol 8,1-167.

Bard, A..J; Santhanam, K. S. "Electroanalytical Chemistry" A.J.
Bard, Ed. (Marcel Dekker, NY, 1979) Vol 4, 215-315.

Flanagan, J.B.; Margel, S.; Bard, A.J.; Anson, F.C. J. Am. Chem.
Soc., 1979, 100, 6248-6253.

Tanford, 0., "Physical Chemistry of Macromolecules”, (Wiley, New
York. 1961).

Janson, T.R.; Kane, A.R.; sullivan, J.F.; Knox, K.; Kenney, M.E. J.
Am. Chem. Soc., 1969, 91, 5210- 5214.

Kane, A..;R Sullivan, J.F,; Kenney, D.I-I.; Kenney, M..E Inorg.
Chem.,1970,9, 1445-1448.

Hush, N. S.; Woolsey, l. S. Molecular Phys, 1971, 21, 465-474.
Mooney, J. R.; Choy, C..;K Knox, K.; Kenney, M..E J. Am. Chem.
Soc., 1975, 97, 3033- 3038.

Ciliberto, E.; Doris, K.A.; Pietro, W.J.; Reisner, G.M.; Ellis, D.E.;
Fragala, l.; Herbstein, F.M.; Ratner, M.A.; Marks, T.J. J. Am. Chem.
Soc., 1984, 106, 7748-7761.

Simiic-Glavaski, 8.; Tanaka, A..;A Kenney, M. E, Yeager, F. J.
Electroanal. Chem, 1 987, 229, 285-296.

32.

33.

34.

35.
36.

(9)
(h)
(i)

(a)
(b)

(c)

(d)

(d)
(6)
9)
(h)
(i)
(1)

(a)
(b)
(c)
(d)
(e)
(f)

171

Mezza, T.M.; Armstrong, N.R.; Ritter, G.W.; lafalice, J.P.; Kenney,
M.E. J. Electroanal. Chem, 1982, 137, 227-237.

DeWulf, D.W.; Leland, J.K.; Wheeler, B.L.; Bard, A.J.; Batzel, D.A.;
Dinninn , D.R.; Kenneg, M.E. Inorg. Chem, 1987, 26, 266.

Doris, .A.; Ciliberto, .; Fragala, .; Ratner, M.A.; Marks, T.J. lsreal
J. Chem, 1986, 27, 337-346.

Feldberg, S.W. ”Electroanalytical Chemistry”; Bard, A.J. ed.,
(Marcel Dekker, New York, 1964) Vol. 3, 271.

Prater, K. B. ”Computers in Chemist and Instmmentation";
Mattson J.S.; Mark, H.B.; MacDonald, H. . Eds, (Marcel Dekker,
New York, 1972) Vol. 2.

Malloy, J.T. ”Laboratory Techniques in Electroanalytical Chemistry",
Kigsg‘inger, P.T.; Heineman, W.R., Eds, (Marcel Dekker, New York,
Britz, D. "Di ital Simulations in Electrochemistry, lecture notes in
chemistry”, ( pringer-Veilaz, Heidleberg, 1981).

Bredas, J.L.; Street, G.B. Acc. Chem. Res, 1985, 18, 309-315.
Bredas, J.L.; Themans, 8; Andre, J.M.; Chance, R.R.; Silbey R.
Synth. Met, 1984, 9, 265-276.

Su, W.P.; Schrieffer, J.R.; Heeger, A.J. Phys. Rev. Letts., 1979, 42,
1698-1701.

Sato, M.; Tanaka, S.; Kaeriyama, K. J. Chem. Soc., Chem.
Commun., 1985, 713-715; S nth. Mat, 1986, 14, 279-288.

Bertho, D.; Jouanin, C. Synt . Met, 1988, 24, 179-182.

Patil, A.D.; Heeger, A.J.; Wudl, F. chem. Rev., 1988, 88, 183-200.
Novak, M.; Rughooputh, S.D.D.V.; Hotta, S.; Heeger, A.J.
Macromolecules, 1 987, 20, 965-968.

Nowak, M.J.; Spiegel, D.; Hotta, S.; Heeger, A.J.; Pincus, P.A.
Macromolecules, 1989, 22, 2917-2926.

Colaneiri, N.; Newak, M.J.; Spiegel, D.; Hotta, S.; Heeger, A.J. J.
Phys. Rev. B, 1987, 36, 7966-7968.

Scott, J.C.; Bredas, J.L; Yakuski, K.; Pfluger, F.; Street, 68
Synth. Met, 1984, 9, 165-172.

Murray, R.W. ”Electroanalytical Chemistry", Bard, A.J. Ed. (Dekker,
New York, 1984) Vol. 13 and references therein.

Morrison, Jr., W.H.; Krogsrud, S.; Hendrickson, D.N. Inorg. Chem,
1973. 12, 1998-2004.

Brown, G.M.; Meyer, T.J.; Cowan, D.O.; LeVanda, C.; Kaufman, E.;
Roling, P.V.; Rausch, M.D. Inorg. Chem, 1975, 14, 506-511.
LeVanda, C.; Bechgard, K.; Cowan, D.O.; Rausch, MD. J. Am.
Chem. Soc., 1977, 99, 2964-2967.

Debois, M.H.; Astruc, D.; Guillin, J.; Maroit, J.P.; Varret, F. J. Am.
Chem. Soc., 1985, 107, 5280-5282 and references therein.

Van Order, Jr., N.; Gergei, W.E.; Bitterwolf, T.E.; Rhein old, A.L. J.
Am. Chem. Soc, 1987, 109, 5680-5690 and referencest erein.

Yap, W.T.; Durst, R.A. J. Electroanal. Chem, 1981, 30, 3-8.

For preparin the precursors of this polymer, see: Hanack, M.; Metz, J.;
Pawlowski, . Chem. Ber., 1982, 115, 2836-2853.

37.
38.
39.
40.
41 .

42.
43.
44.

45.
46.

47.

48.

49.

172

Marks T.J., private communication.

Ammar, F.; Saveant, J.M. J. Electroanal. Chem, 1973, 47, 215-221.
Nicholson, R.S.; Chain, l. Anal. Chem, 1961, 36, 706-723.

Karp, S.; Meites, L. J. Electroanal. Chem, 1968, 17, 253-265.

Levich, V.G. 'Physicochemical Hydrodynamics“, (Prentice-Hill, Englewood
Cliffs, NJ 1962).

Gale, D.C.; Gaudiello, J.G. J. Am. Chem. Soc., submitted for publication.
Parry, E.P.; Osteryoung, R.A. Anal. Chem, 1965,37, 1634-1637.

(a) 186,955,730; Schrieffer, J.R.; Heeger, A.J. Phys. Rev. Lett, 1979, 42,
- 1.

b) Rice, M.J. Ph 5. Lett, 1979, A-71, 152.

c) Schrieffer, J. . Mol. Cryst. Liq. Cryst, 1981, 77, 201-207.

Gale, D.C., Master Dissertation, Michigan State University (1989).

(a) Thackeray, J.W.; White, H.S.; Wrighton, M.S. J. Phys. Chem,
1985, 89, 5133-5140.

(b) Mallouk, T.E.; Cammarata, V.; Crayston, J.A.; Wrighton, M.S. J.
Phys. Chem, 1986, 90, 2150-2156.

(c) 'é’ggeray, J.W.; Wrighton, M.S. J. Phys. Chem, 1986, 90, 6674-

(d; Wrighton, M.S. Comments Inorg. Chem, 1985.4, 269-294.

e White, N.S.; Kittlesen, G.P.; Wrighton, M.S. J. Am. Chem. Soc.,
1984, 106, 5375-5377.

(1) Kittlesen, H.; White, S.; Wrighton, M.S. J. Am. Chem. Soc., 1984,
106, 7389-7396.

(9) Paul, E.W7.; Ricco, A.J.; Wrighton, Ms. J. Phys. Chem, 1985, "89,
1441-144 .

(h) Kittlesen, G.P.; White, H.S.; Wrighton, M.S. J. Am. Chem. Soc.,
1985, 107, 7373-7377.

8; Ofer, D.; Wrighton, Ms. J. Am. Chem. Soc., 1988, 10, 6667-6668.

' Crczaoks, R.M.; Chyan, O.M.R.; Wrighton, M.S. Chem. Maker, 1989,
1, -4.

Tamao, K.; Kodama, S.; Nakajima, I.; Minata, A.; Suzuki, K.; Kumada, M.
Tetrahedron, 1982, 38, 3347-3354.

For preparing the precursors of this monomer see:

a Janda, M.; Srogl, J.; Shbor, l.; Nemec, M.; Vopatma, P. Synthesis,
1972, 10, 545-547. ‘

(b) Lawesson, S.A. Arkiv Kemi, 1957, 1, 325-336.

c Aiello, E.; Dattolo, G.; Cirrincione, G.; Almerico, A.M.; Asdia, D. J.
Heterocyclic Chem, 1 982, 19, 977-979.

Tourillon, G. "Handbook of Conducting Polymers", Skotheim, T.A. Ed.,
(Dekker, NY, 1986) Vol. 1 and references therein.

WW*fl-*A““Ef

50

51.
52.

53.
54.

173

Tourillon, G.; Gamier, F. J. Electroanal. Chem, 1984, 161, 51-58.
Dian, G.; Barley, G.; Decroix, B. Synth. Met, 1986, 13, 281-289.

33
c)

(d)

 

Daum, P.; Murra , R.W. J. Phys. Chem, 1981, 85. 389-396.

Perce, P.J.; Ba , A.J. J. Electroanal. Chem, 1980, 114, 89-115.

ggronzier, A.; Latour, J.M. J. Electroanal. Chem.,1987, 224, 295-
1.

Abruna, H.D.; Meyer, T.J.; Murray, R.W. Inorg. Chem, 1979, 11,

3233-3240.

Feldberg, S.W. J. Am. Chem. Soc., 1984, 106, 4671-4674.

(a)
(b)

(C)
(d)

Mezza, T.M.; Armstrong, N.R.; Ritter, G.W.; lafalice, J.P.; Kenney,
M.E. J. Electroanal. Chem, 1982, 37, 227-237.

Wheeler, B.L.; Nagasubramanian, G.; Bard, A.J.; Schechtman,
L.A.; Dininny, D.R.; Kenney, M.E. J. Am. Chem. Soc., 1984, 106,
7404-7410.

Dewulf, D.W.; Leland, J.K.; Wheeler, B.L.; Bard, A.J.; Batzel, D.A.;
Dininny, D.R.; Kenney, M.E. Inorg. Chem, 1987, 261, 260-270.
Simic-Glavaoki, 8.; Tanaka, A.A.; Kenney, M.E.; Yeager, E. J.
Electroanal. Chem, 1 987, 229, 285-296.