4

pr.

l'

:‘-.w“.-_

"in.

x..w

 

 

 

ff v‘
m 19' “’

‘ y‘m‘fl

n-

7 .
unafr: 9125a} «

 

.1145.

3%

2!“ - «an. . , - ‘ .‘

.1 - L ‘ "
m "
.. 1 “.ua-‘c'Y
i

we“
2’44?" .-. F . ' “a 41‘“ ‘ MFA?" “
,3 ..

$.93? I-a: ‘»»‘3.;\‘3 c4;

,.. - _...~."-, nu
; 535,1."

 

537'
MI, W” -;,f:Af‘U'
a» 234*. "’1

at: ”iffy
“Ni?! Lar- nII:-

1:3”?

" F

'r--§1‘. Nafifl...“
(r-

wflrz‘fil," 74%
,1 ’2}: “"33

u
" W; "’l‘ 31w 'IFSQwr
éfflg‘V’” “.1" :1» "r 'g'fil')
m2" “747.. .. ..~.i~.e;r+7 9" w
'fifi"; 7"

{MW
I A, ”#374

 

3% 1.32;; x» “a:
pa -- -A .7
.. x. 7.:2— my? .5;
”11"“: «”9933

mm.

   

-:4:r-~I ixi ’7':
t"‘ "2.1:.1. u M h“
,. «n... .1" guru.” '°'
“’3 V

 

-9.

5'

,.'..J': ,3?»
v73uv. ""

“04...? "" x321:

§~h 72151-
7-724.

,u.’

 

:fifls B01671

\ES
MmHlGAN

H \ ll

 

 

F.

8137“; mum W. W. (ill \i \\\\
l \‘f\\\\\\\:1‘lV\\\ l\,i\\,\\‘\\\l\l\m
\\\\\\\\\i \\\ W \\l W \\8g‘\\6é55

\\\

lIBRARY \\ 31293006
Michigan State
University

 

 

This is to certify that the

thesis entitled

Electrochemical Oxidation and Characterization
of fl-Oxo— (Tetra-T-Butylphthalocyaninato)
Silicon

presented by

David Gale

has been accepted towards fulfillment
of the requirements for

Master ' 3 Chemistry

 

 

degree in

W

Major professor

8-‘i~8cl

Date

 

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

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

DATE DUE DATE DUE DATE DUE l

 

    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Minn-av. Action/Equal Opportunity Institution

‘_._——-—-7:

ELECTROCHEMICAL OXIDATION AND CHARACTERIZATION OF

p-OXO-(TETRA-T-BUTYLPHTHALOCYANINATO)SILICON.

by

David C. Gale

A DISSERTATION

Submitted to
Michigan State University
In partial fulfillment of the requirements

for the degree of

MASTER OF SCIENCE

Department of Chemistry

1989

ABSTRACT

ELECTROCHEMICAL OXIDATION AND CHARACTERIZATION OF
p-OXO-(TETRA-T-BUTYLPHTHALOCYANINATO)SILICON.
by

David C. Gale

The oxidative redox properties of u-oxo-(tetra-t-
butylphthalocyaninato)silicon was investigated in solution by
conventional electrochemical techniques. This soluble, covalently-
linked polymer, a known electrical conductor in the solid state when
oxidized (doped), was characterized using controlled potential
coulometry (CPC), rotating disk voltammetry (RDE), differential pulse
voltammetry (DPV), and cyclic voltammetry (CV). Digital simulations of
the voltammetry results were performed to obtain information regarding
the potential of redox sites comprising the polymer. The degree of
partial oxidation can be adjusted to any level ranging from zero to
1001 (from neutral material to one electron removed per phthalocyanine
site). This process occurs over a 1700 mV range and is consistent for
a material with a large amount of interaction between adjacent
molecules. Voltammetry experiments and spectroscopic measurements
indicate that the doped material is stable at all degrees of partial
oxidation. RDE results suggest that there is no kinetic limitation
for the oxidative process, allowing the thermodynamic potentials for

band depletion to be accurately determined.

To My Parents

ACKNOWLEDGEMENTS

The author would like to thank. all of the people who have
provided assistance on this project. I would like to thank my parents
for all of their support, understanding, and help through the years.
I would like to thank my brothers, Peter and Colin, for their advice
and encouragement these past few years. I would also like to thank my
fellow group members for their helpful suggestions and timely advice.
The author greatly appreciates the work of Tom Chapaton regarding the
computer simulations. I would like to thank the rest of the graduate
students in the department for their friendship, support, and external
activities which have made graduate school bearable. Finally, I would
like to thank Professor John Gaudiello, for his support during this
work, for all the knowledge I have obtained in conversations with him,

without his support this project would have never been accomplished.

iv

TABLE OF CONTENTS

List of Tables ........... ........................ .........
List of Schemes.. ....... ..................................
List of Figures...........................................
Introduction..............................................
Experimental.......... ..... ...............................
2.1 Instrumentation ................. ............. ........
2.1.1 Electrochemical...............................
2.1.2 Rotating Electrode Assembly and Bipotentiostat
2.1.3 Spectroscopic Instrumentation.................
2.2 Glassware.......... ....... ............................
2.2.1 Vacuum Lines..................................
2.2.2 Electrochemical Cells.........................
2.3 Chemicals.............................................
2.3.1 Synthesis Solvents............................
2.3.2 Electrochemical Solvents......................
2.3.3 Supporting Electrolyte ..... . ..... .............
2.4 Synthesis.............................................
2.4.1 Trans-dichloro(tetra-t-
butylphthalocyaninato)silicon.................
2.4.2 Dihydroxy(tetra-t-

butylphthalocyaninato)silicon.................

19
19
19
19
20
20
20
20
22
22
22
23
23

23

30

page #

2.4.3 p-oxo-(tetra-t-
butylphthalocyaninato)silicon ...... .... ....... 30
2.4.4 Di-trimethylsiloxy(tetra-t-
butylphthalocyaninato)silicon................. 39
Electrochemical Studies of
p-oxo-(tetra-t-butylphthalocyaninato)silicon.............. 42
3.1 Bulk Oxidation by Controlled Potential Coulometry..... 42
3.1-1'Introduction ................................ .. 42

3.1.2 Procedure,,...

................................ 42

3.1.3 Results '45

3.2 Rotating Disk Voltammetry Studies ......... . ...... ..... 58

3.2.1 Introduction,

................................. 58

3.2.2 Procedure,,,,

................................. 59

3.2.3 Results 59

3.2.4 Levich Plot 64

3.3 Differential Pulse Voltammetry,,,,.

................... 68

3 3-1 Introduction.................................. 68
3.3 2 Procedure..................................... 68
3.3.3 Results........................... ....... ..... 69
3.3.4 Digital DPV Simulation........................ 72
3.4 Cyclic Voltammetry.................................... 84
3.4.1 Introduction ....... ............... ..... . ..... . 84

3'4'2 Procedureooooooooooooooooooooooooooooooooooooo 84

3'4'3 Resultsooooooo00.000000000000000...ooooooooooo 84

vi

page #

3.5 Chronocoulometry................ ...................... 92
3.5.1 Introduction.................................. 92
3.5.2 Procedure..................................... 92
3.5.3 Results ............. ....... .................. . 92

Conclusions............................................... 96

Append1X.................................................. 102

List of ReferenceS................... ............. . ....... 104

vii

LI ST OF TABLES

page #

Table I. Z Oxidation vs. Potential for the Silicon Polymer...... 54

Table II. Digital Simulation Data for the Silicon Polymer....... 74

viii

LIST OF SCHEMES

page #

Scheme I. Synthesis of (t-BuaPcSiO) 26

“'0 ooooooooo o oooooooooo 0

ix

LIST OF FIGURES

Figure page #

Figure 1. Band formation from the mixing of electronic
states. According to Huckel theory, if two atomic orbitals
interact, two molecular orbitals are formed with. an. energy
separation of Zfi. If two molecular orbitals interact, then two
sets of molecular orbitals are formed. The molecular orbitals
in each set are separated in energy by 2t, where t is the
transfer integral (a measure Of the amount of interaction
between adjacent molecules). If n molecules interact, n states
are formed, from each of the x and «* molecular orbitals. Into
each band of N states, can be placed 2N electrons................ 5

Figure 2. Schematic of the peripherally substituted
phthalocyanine polymer (t-BuAPcSiO)n, central metal atom - Si,
the chain length of this polymer is approximately 25 units.17b... 24

Figure 3. 1H NMR spectrum of dichloro(tetra-t-
butylphthalocyaninato)silicon, t-BuaPcSiClz, obtained in CDCl3.
The spectrum is referenced to tetramethylsilane and the peak at
7.24 ppm is due to residual CHC13. Peak assignments are given
in the text ..................................................... 28

Figure 4. UV/Vis spectrum of dihydroxy(tetra-t-
butylphthalocyaninato)silicon, t-BuhPcSi(OH)2, obtained in
CH2C12. Absorbance peaks occur at 679 and 354 nm................ 31

Figure 5. FT-IR spectrum of dihydroxy(tetraet-
butylphthalocyaninato)silicon, t-BuaPcSi(0H)2, obtained as a
nujol mull. The IR spectrum is identical to that reported in
the literature including the OH stretching vibration at 3500

ell-.10 Oo000000000000000.ocoo-000000000000OOOOOOOOOOOOQOOOOOOOOOOOo 33

Figure page #

Figure 6. UV/Vis spectra of the pristine, unoxidized silicon
polymer, (t-BuaPcSiO)n, obtained in CH2012. The absorbance

peaks occur at 619 and 327 nm.............. ........ ..............

Figure 7. FT-IR spectrum of p-oxo(tetra—t-
butylphthalocyaninato)silicon (t-BuaPcSi0)n, obtained as a
nujol mull. The IR spectrum is identical to that reported in
the literature..................................................

Figure 8. 1H NMR spectrum of bis(trimethylsiloxy)(tetra-t-
butylphthalocyaninato)silicon, (t-BuaPcSiOZ)(Si(CH3)3)2,
obtained in CDC13. The spectrum is referenced to
tetramethylsilane and the peak at 7.24 ppm is due to residual
CHCl3. Peak assignments are given in the text..................

Figure 9. All ground glass joint coulometry cell utilized for
the bulk oxidations of the silicon polymer. The counter
electrode (Pt mesh, 2 cm2 in. area) is separated from the
working electrode by two fine porosity glass frits. The Ag
wire reference electrode is separated from the working solution
by' a fine porosity glass frit. The working electrode is
constructed of Pt mesh approximately 8 cm2 in area. A small
area Pt (0.018 cm2) electrode was also placed into the working
solution.

A - ground glass joints, 3 - counter electrode

C - glass frits. D - working electrode

E - small area working electrode, F — reference electrode.......

Figure 10. Current-time curves for the oxidation of neutral
silicon polymer, (t-BuhPcSiO)n, in l,1,2,2-tetrachloroethane /
0.2 M TBABF,‘ at different potentials. The solution
concentration was 0.0288 mM based on a 25 phthalocyanine unit

polymer or 0.72 mM based on a single phthalocyanine moiety. ......

xi

35

37

40

43

46

Figure page #

Figure 11. Current-time curves for the reduction of oxidized
silicon polymer, (t-BuAPcSiO)n, in l,1,2,2-tetrachloroethane /
0.2 M TBABF,‘ at different potentials. The solution
concentration was 0.0288 mM based on a 25 phthalocyanine unit
polymer or 0.72 mM based on a single phthalocyanine moiety. 49

Figure 12. The degree of oxidation vs. potential for the
electrochemical oxidation of the silicon polymer, (t-
BuhPcSiO)n, in 1,1,2,2-tetrachloroethane / 0.2 M TBABFA. Each
point corresponds to the mean value for a complete
oxidation/re-reduction cycle performed by controlled potential
coulometry. The error bars represent the difference between
the oxidation and re-reduction values. Solutions were
typically 0.02 to 0.04 mM based on a 25 phthalocyanine unit
polymer or 0.5 to 1 mM, based on a single phthalocyanine moiety
(0.4 - 0.8 mg polymer/mL). The 2 oxidation values were
determined by relating the mass of material present to the
amount of charge passed at the given potential (see equation

Figure 13. UV/Vis spectra of the extensively cycled silicon
polymer, (t-BuaPcSiO)n, obtained in CHZClz. Absorbance peaks
occur at 625 and 327 nm. The UV/Vis spectra of the extensively
cycled silicon polymer is essentially the same as the pristine

pOIYmBr'O.OOOOOOOOOOOOOOOOOOIOOIOOOOOOOOOOOOOOOOO...0.0.0.0000... 56

Figure 14. Rotating disk voltammograms of a 0.0208 mM solution
of the silicon polymer , (t-BuaPcSiO)n, in l , l , 2 , 2-
tetrachloroethane / 0.2 M TBABFa at a Pt disk electrode at
different rotation rates. The background RDE voltammogram of a
blank solution was obtained at 1800 rpm. The current values
for the background RDE did not vary as a function of rotation

rate.coo...0.0000000000000000...00000000000000000...0000000000000 60

xii

Figure page #

Figure 15. The degree of oxidation vs. potential for the
silicon polymer, (t-BuaPcSiO)n, determined from the RDE
voltammograms. The RDE values were determined by normalizing
the current at any potential to the limiting current at 1.82 V.
The error bars represent the standard deviation of the
calculated 1 oxidation, for the same potential, at different
rotation rates..,,..,,,,,.

....................................... 63

Figure 16. Plot of disk. current at various potentials and
rotation rates vs. w1/2 (w - 2NN, N-rps), for the oxidation of
the silicon polymer, (t-BuaPcSiO)n, in 1,1,2,2-
tetrachloroethane/0.2 M TBABFA............................. ...... 66

Figure 17. Differential pulse voltammogram of a 0.056 mM
solution of silicon polymer, (t-BuaPcSiO)n, in 1,1,2,2-
tetrachloroethane / 0.2 M TBABF4 obtained with a small area
0.018 cm2 Pt electrode. The scan rate was 4 mV/s, pulse
amplitude - 50 mV, pulse width - 50 ms, and the pulse period -

1000 ms.......................................................... 70

Figure 18. Simulated differential pulse voltammogram for the
oxidation of the silicon polymer. The conditions utilized
(scan rate, pulse width, concentration, pulse amplitude, area
of electrode) for the simulation are the same as those of the
experimental data shown in Figure 13. The bars along the X-
axis represent the standard potentials of the redox couples
utilized to obtain the current-potential response (see Table 2
for the exact potentiaIS)........................................ 75

Figure 19. A plot of the difference in standard potential
between successive sites utilized in the DPV, CV, and RDE
simulations of the silicon polymer (see Table 2 for the exact

valueS) ......................................................... 79

xiii

Figure page #

Figure 20. The degree of oxidation vs. potential calculated by
digital simulation using the potentials listed in Table 2. The
1 oxidation is not calculated from a particular technique,
rather it is based on the number of sites oxidized at a
specific potential. The number of sites oxidized is determined
from the potential of the electrode, the standard potential of
the sites, and the Nernst equation (see also Appendix A).,,,,,,,, 80

Figure 21 . The degree of oxidation vs . potential for the
silicon polymer , (t-BuaPcSiO)n, from the CPC data and the
digital Simulation.OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO00...... 82

Figure 22. Cyclic voltammogram of a 0.0432 M solution of
silicon polymer, (t-BuaPcSi0)n, in 1,1,2,2-tetrachloroethane /
0.2 M TB'ABF,‘ obtained with a small area 0.018 cm2 Pt electrode.
The scan rate was 100 mV/s....................................... 85

Figure 23. Cyclic voltammogram of a 1.25 mM solution of
bis(trimethylsiloxy)(tetra-t-butylphthalocyaninato)silicon, (t-
BuhPcSiOZ)(Si(CH3)3)2, the capped monomer in 1,1,2,2-
tetrachloroethane / 0.2 M TBABF“ obtained with a 0.018 cm2 Pt
electrode. The scan rate was 100 mV/s........................... 88

Figure 24. Simulated cyclic voltammogram of the oxidation of
the silicon polymer. The standard potentials used in this
simulation are the same as those used for Figure 13 (see Table
2 for the exact potentials). The conditions utilized (scan
rate, concentration, area of electrode) are the same as those
in Figure 18..................................................... 90

Figure 25. Degree of oxidation vs. potential for the silicon
polymer, (t-BuaPcSiO)n, from the CPC, RDE. digital simulation

results.......................................................... 98

xiv

l . INTRODUCTION

Organic, organometallic, and inorganic materials that can
become electrically conductive upon partial oxidation or reduction

1 In the

(doping) are currently an area of considerable interest.
1960's and 1970's, scientists discovered that doped polymers and
various condensed organic solids could become metal-like electrical

2

conductors. The development of these materials for use as possible

electronic devices has attracted widespread interest and investment

in recent years.3

These doped polymers and condensed organic solids
have properties normally associated with traditional metals, that is,
the electrical conductivity of these materials goes up as the
temperature goes down. These "molecular metals” have generally been
grouped into two separate classes. The first class are the molecular
solids , 1 ’ 4 such as TTF - TCNQ (tetrathiafulvalene -
tetracyanoquinodimethane), which are comprised of discrete molecular
components that have been condensed into a segregated, organized
framework. Molecular solids can usually be redissolved to yield
their neutral molecular components. The second class of "molecular

metals" is the conducting polymers,]"5

such as polyacetylene and
polythiophene, which are based upon covalently bonded aromatic
hydrocarbons or heterocycles. Although there are similarities

between these two classes of materials,6

many of their novel
properties are different. Polymers historically have been utilized

as insulators for a number of applications in the consumer and

electronics field.3 Recently electrically conductive polymers have
been used in the construction of battery anodes and cathodes,
chemical sensors, and electromagnetic interference shielding

3

devices. It has been suggested that these materials will appear in

numerous applications because of their light weight, flexibility, and

electrical properties . 3c

Although it is no longer believed that
these materials will replace conventional metals or semiconductors,
they will find specialty applications due to their unique and novel
properties. Some of the proposed applications for these materials
include use in photoelectrochromic displays, rechargeable batteries,

and microelectronic conductive components.3

It is their unique and
novel properties which need to be studied and comprehended before
more applications can be made.

All materials can be grouped into one of three classes based on
their electrical conductivity, either as a metal, semiconductor, or

insulator.4a

Metals such as copper or silver have. electrical
conductivities of 106 ohm'1 cmil. The electrical conductivity of
metals decreases with increasing temperature. This is due to a
greater number of lattice vibrations or phonons at increased
temperatures which perturb the movement of the mobile charge
carriers. Semiconductors typically have electrical conductivities in
the range from 102 ohm"1 cm'1 to 10-5 ohm'1 cm'l. The electrical
conductivity of semiconductors increases with increasing temperature.
Thermal excitation enables charge carriers to be excited up to the
empty valence band from the occupied conduction band, this process

increases both the number of charge carriers and the electrical

conductivity. The third class of electrical conductors is the

insulators, such as teflon or sulfer, with electrical conductivities
less than 10'6 ohm'1 cm'l. Insulators have completely filled
conduction bands and completely empty valence bands similar to
semiconductors. However, the band gap or energy difference between
the conduction and valence bands is substantial, preventing the
movement of electrons from the conduction .to the valence band.
Neutral or undoped organic polymers are classified as insulators due
to their extremely low electrical conductivity. Doped polymers can
have electrical conductivitiesl‘a'5 ranging from 10'9 to 104 ohm"1 cm'
1, which spans the entire range of conductivity from insulators to
metals. Charge transfer salts typically have conductivities ranging
from 101 ohm"1 cm'1 to 105 ohm'1 cm'1 whidh are the conductivities
associated with traditional semiconductors and metals.4 Both charge
transfer salts and doped polymers have electrical conductivities
associated with semiconductors and insulators, however only from
temperature-conductivity and spectroscopic measurements can their
electrical classification be determined. The electrical conductivity
of doped polymers and most molecular charge-transfer salts goes up as
the temperature goes down. This electrical conductivity vs.
temperature behavior is similar to that observed for metals.

Both molecular solids and doped conducting polymers have some
common features. Both classes of materials exhibit anisotropic, or

low-dimensional properties.6

Electrical conductivity is greater in
one direction than in others. In molecular solids, the direction of
greatest conductivity is parallel to the stacking axis of the
crystal. For conducting polymers, conductivity is greatest along the

chain direction. Another common feature for both class of materials

is that they are partially oxidized or reduced when they are
electrically conductive. For conducting polymers, charge is
transferred between the polymer chain and a dopant (oxidizing or
reducing agent, or electrode). In molecular solids, charge is
transferred between the donor and acceptor components of the
condensed complex. Both classes of "molecular metals" possess two
important prOperties that give rise to their unique characteristics.
First, they have a band type electronic structure formed from the
closely spaced interacting moleculesl‘a'7 and second, a partial or

fractional oxidation state . 6 ' 8

For example, in a single molecule
described by Huckel theory, two single occupied p atomic orbitals can
interact to form two molecular orbitals (Figure l). The two
molecular orbitals are a filled low-energy w-bonding orbital and an
empty high energy x*-antibonding orbital. If two molecules are
stacked directly on top of one another, such that their molecular
orbitals interact, two new filled Tr-bonding molecular orbitals are
formed and are separated in energy by 2t, where t is the tight
binding integral a measure of the interaction between the two
molecular orbitals (analogous to 8, the Hfickel integral). Similar to
the two new u-bonding molecular orbitals, two new empty «*-
antibonding molecular orbitals are formed also separated by an energy
gap of. If a stack of N molecules interact, there will be N a states
in the highest occupied molecular orbital band (HOMO) and N «1" states
in the lowest unoccupied molecular orbital band (LUMO). A band type
electronic structure is formed when the energy difference between

these states becomes small. For a filled band (HOMO) comprised of N

states, there are 2N electrons occupying the band. A material with a

Figure 1. Band formation from the mixing of electronic states.
According to Huckel theory, if two atomic orbitals interact, two
molecular orbitals are formed with an energy separation of 213. If
two molecular orbitals interact, then two sets of molecular orbitals
are formed. The molecular orbitals in each set are separated in
energy by 2t, where t is the transfer integral (a measure of the
amount of interaction between adjacent molecules). If n molecules
interact, n states are formed, from each of the at and 1r."r molecular
orbitals. Into each band of N states, can be placed 2N electrons.

1

Isolated
orbitals

,' 2t
23
F 2:
Single Two stacked
molecule molecules

FIGURE 1

 

 

 

l

LUHO band

HOMO band

 

 

llll

Straight stack

of many
molecules

6:

let

completely filled HOMO band and completely empty LUMO band will be
either a semiconductor or an insulator depending on the energy
separation between the two bands and the temperature. The second
property that conducting polymers and charge transfer salts possess,
is a fractional or partial oxidation state where the HOMO or LUMO
bands are only partially filled. These fractionally filled bands
allow conduction within the band and do not require the transfer of
charge between them. The partial filling of these bands can be
accomplished by several methods. In molecular charge-transfer salts,
charge is usually transferred between the donors and the acceptors.
Conducting polymers can be doped either through chemical or
electrochemical means.

Neutral conjugated organic polymers are generally insulators or

semiconductors due to the lack of charge carriers.5

Some conjugated
polymers have the necessary band type electronic structure arising
from the overlap of molecular orbitals on adjacent. molecules.
However, in a neutral form, the fractional oxidation state is not
present. In order to obtain greater electrical conductivity, the
polymer must be oxidized or reduced. The neutral polymer can be
chemically converted into an ionic compound consisting of a polymeric
cation (for the oxidation case) and a counter-ion which is the
reduced form of the oxidizing agent. Similar results can be obtained
by utilizing a reducing agent which forms a polymeric anion. The
wide use of x-bonded unsaturated polymers for the preparation of
conducting polymers is due to the stability of the cation or anion

1c

forms of these compounds. The removal or addition of 1r electrons

does not affect the bonds (sigma) which are responsible for holding

the polymer together. Mild oxidizing or reducing agents such as
halogens, sodium napthalide, and quinones are utilized to ensure that
only charge-transfer takes place and that no further chemistry
occurs. The dopant or counterion perturbs the polymer structure
extensively due to its large relative size, and the extensive charge
transfer which takes place between the polymer chain and dopant.6
The range of band-filling obtained utilizing chemical dopants is
usually very limited. The chemically doped polymers are also

generally insoluble , 9

which prevents the characterization of their
redox behavior directly. Recently, interest in conducting polymers
has centered on the electropolymerization of heterocycles
(polythiophene, polypyrrole) that do not depend on the oxidation or
reduction of the polymer by a chemical agent. The synthesis and
doping of these materials is accomplished by using a conventional

metallic electrode.9

The level of doping is no longer dependent on
the concentration or oxidative (reductive) strength of the
counterion, and the doping level can be more easily varied.
Electrochemical doping is also cleaner with less interfering chemical
reactions. A major problem associated with electrochemically
synthesized polymers, however is their insolubility.9 Typically
these polymers have been studied as thin films on electrodes.10
Kinetic and thermodynamic data have been obtained from these modified
electrodes. However, the redox behavior of these films has been
shown to be influenced by film discontinuities and non-uniformities,
inseparability of faradaic and capacitive currents, and counterion

diffusion through the film.11 The degree of polymerization of many

of these polymers is unknown, making a quantitative description of

their properties difficult. Polymer properties are controlled by
many factors, including molecular weight, polydispersity, chain
branching, crystallinity, and morphology.5c Structural defects in
the polymer chain can occur because of the formation of charge
carriers, specifically polarons and bipolarons. The efforts to
understand the chemistry and physics of these materials have been
hampered by the insolubility, limited crystallinity, and non-
uniformity of these materials.

The second class of electrically conducting materials are
called charge-transfer salts or molecular solids. These materials
are formed when charge is transferred between donor and acceptor

molecules.4

The donor and acceptor molecules are usually planar
molecules that stack upon one another to form segregated arrays. The
formation of a band-type electronic structure occurs due the overlap
of molecular orbitals of the molecules in the stack.4 These separate
stacks allow the transferred charge to be easily moved along the
separate donor and acceptor stacks. The charge-transfer salts must
crystallize in a systematic fashion without defects. The synthesis
and properties of these materials are influenced by crystal packing
forces, intermolecular interactions, and Van der Waal's forces.
Defects or disorders in the crystal can cause metal-insulator phase
transitions, causing the electrical conductivity to drop to values
associated with insulators.“a The partial filling of the band occurs
due to charge transfer between the donors and acceptors. The degree
of band-filling for molecular solids is solely dependant on the

properties of the donor and acceptor molecules. It is difficult to

compare systems with various levels of band-filling due to

10

differences in elemental composition, stoichiometries, crystal
structure, and band widthsf"128

Very few studies have been performed on non-traditional
electrically conducting materials as a function of band-filling. In
systems where chemical dopants have been utilized, the variation in
band-filling has been limited due to the lack of a wide range of
dopants that can be applied on a single system.13 In charge transfer
salts, the ability to tune or change the amount of band-filling is
dependent on finding a acceptor-donor molecular systems that will
condense together in an organized framework. Recently, through
replacement of a fraction of the donor molecules by a molecule of
similar size and shape, variation in the amount of band-filling for a

d.12 These neutral molecules

charge transfer salt has been obtaine
ensure that the unit cell does not collapse and that the crystal
structure is stabilized. The [NMP]1_X[Phen]x[TCNQ] charge transfer
salt, where NMP-N-methylphenazinium, Phen-phenazine and TCNQ -
tetracyanoquinodimethane, was synthesized with different ratios of
NMP and Phen. Phenazine, a molecule of similar shape and
polarizability to NMP, was substituted for NMP into the charge
transfer salt. The degree of band-filling varied from 25 to 501 with
different ratios of NMP and Phen (fractional charge per cation,
cation-donor, TCNQ-acceptor). This particular experiment was the
first reported variation in band-filling for a molecular solid. This
study also showed the first direct evidence for solitons (structural
defects) in a molecular charge-transfer salt.

Electrochemical doping has recently been applied to vary the

amount of band-filling in conducting polymers. Wrighton and Ofer

11

performed a study on the potential dependence of conductivity of a
thiophene polymer, by constructing a transistor made out of

polymerized thiophene . 14

Two microelectrodes were connected to the
redox active polymer, and a fixed potential difference (drain
voltage, VD) was placed across the electrodes. A third electrode was
placed between the two microelectrodes to measure the current as the
potential (VG) of the polymer was changed. A plot of ID vs. VG
reveals the potential dependence of electrical conductivity for the
thiophene polymer. The thiophene polymer has a window of high
conductivity approximately 1 volt wide. The thiophene polymer is
initially an insulator, becomes highly electrically conductive and
with further increase in potential (VG), becomes an insulator again.
The potential (VG) can then be cycled back to the initial potential
similar to a cyclic voltammetry experiment (vide infra). A
conductivity vs. potential curve of the polymer was obtained. The
conductivity vs. potential (ID vs. VG) curves for the oxidation scan
and the reduction scan are not identical. The differences in the two
curves are attributed to changes in the polymer structure. This
particular study showed that electrical conductivity does depend on
the oxidation of the material. Unfortunately the band-filling as a
function of potential could not be determined from this experimental
setup. The study was performed in the solid state, thus it is
difficult to determine what the nature of the charge carrier in the
thiophene polymer is as a function of potential. The technique
utilized by Wrighton and Ofer also did not permit the determination
of the exact conductivities as a function of potential. The redox

behavior of many electrically conductive polymersl have been studied

12

as films on electrodes. Many of those experiments suffer from the
same difficulties encountered in this study. Film discontinuities,
structural changes, along with the inability to accurately change and
determine the degree of band-filling causes the determination of
physical and chemical properties as a function of band-filling to be
inaccessible from studies utilizing films on electrodes..

Recently, cofacially joined phthalocyanine polymers have been

8.13.15

doped by chemical and electrochemical means. Phthalocyanine

polymers combine characteristics of both molecular charge-transfer

salts and conducting polymers.8

Similar to charge-transfer salts,
the large planar phthalocyanine molecules are stacked upon one
another, permitting the interaction of molecular orbitals and forming
a band-type electronic structure. Similar to conducting polymers,
the planar phthalocyanine rings or moieties are held together in a
rigid-cofacial orientation thru covalent bonds involving the center
atom complexed to the phthalocyanine ring. These macrocyclic metal
complexes have been bridged with a variety of ligands. Pyrazine,
tetrazine, cyanide, and isothiocyanate groups have all been used to
bridge Fe, Ru, Co, and Rh phthalocyaninesla’16 Axial O-M-O (O -
oxygen, M - central metal atom) covalent bonds have been used to
bridge Ge, Si, and Sn phthalocyanines.

Chemical doping of both substituted and unsubstituted Si, Ge,
and Sn phthalocyanine polymers has been accomplished by a number of

groups With iodine, bromine, and nitrosonium salts.13a'e

The partial
oxidation of the unsubstituted Si and Ge phthalocyanine polymers by
12 and Brz was accomplished by stirring the polymer and the

corresponding halogen in benzene for 48 hours. Electrical

l3

conductivity of the unsubstituted phthalocyanine polymer [PcSiO]n,
doped with iodine increased from 5.5 x 10'6 ohm.1 cm'1 to 1.4 ohm-1
cm'l. Electrical conductivity for [PcGeO]n doped with iodine
increased from 2.2 x 10"10 ohm'1 cm'1 to 1.1 x 10 '1 ohm"1 cm'l. The
conductivity of these materials increased as the iodine uptake
increased, indicating that at higher partial oxidation of the
polymer, the number of charge carriers increased. Peripherally
substituted Si and Ge phthalocyanines have also been doped in the

solid state by halogens.17

Peripheral substitution with t-butyl
groups enables these compounds to be soluble in a number of
nonaqueous solvents. Prior to doping, the electrical conductivities
of the substituted and unsubstituted polymers were very similar. The
highest electrical conductivities for the iodine doped substituted
silicon phthalocyanine polymer [(t-BuAPcSiO)IY]n was 2 x 10'3 ohm'1
cm'l, and the substituted germanium phthalocyanine polymer [(t-
BuchGeO)IY]n was 1 x 10'3 ohm'1 cm'l. These values are slightly
lower than the unsubstituted phthalocyanine polymers. The EPR
spectra of the halogen doped substituted silicon polymer shows an
intense signal for a free electron, which supports a ligand centered
oxidation process. Information obtained from the chemical doping of
solid samples does not contain a great deal of information on the
band-filling of these compounds. These experiments require long
doping times to partially oxidize the polymer. Thus, information on
the time dependence of the doping process is lost. With chemical
doping, variation in the amount of band-filling by a single dopant is

possible by increasing or decreasing the amount of dopant. However,

it is not possible to chemically oxidize a polymer from 0 to 100%

l4

homogeneously with a single dopant. Information on the interactive
nature of these systems is difficult to probe and usually is not
obtainable from studies performed in the solid state.

To obtain a greater control over the amount of band filling,
Hanack and Leverenz utilized electrochemical doping in the solid
state for the oxidation of p-pyrazine(phthalocyaninato)iron(II).18
The potential of the electrode can be used to vary the amount of
band-filling. This particular polymer was doped galvanostatically as
a weakly pressed pellet in CH2612. Galvanostatic doping utilizes a
constant current source, the potential of the working electrode
varies to keep the same amount of current flowing at all times. Upon
galvostatically doping, the electrical conductivity of this polymer
increased from 2 x 10'6 ohm'1 cm'1 to 4.7 x 10'2 ohm"1 cm'l. The
difficulty with galvostatic doping in the solid state is the
inability to separate the faradaic and background currents. Also
little information on the rate of electron transfer is obtained from
these experiments. However, the power and utility of electrochemical
doping is shown by the possibility of tuning the amount of band-
filling for a conductive polymer. The greatest electrical
conductivity occurred after approximately 502 oxidation of the
polymer. At 100% oxidation, the electrical conductivity was 3 x 10'2
ohm'1 cmfl, only' slightly lower“ for the 'value obtained. at 50 1
oxidation. Theory predicts that this value should be much lower due
to the energy expense of pairing mobile charge carriers in the
band.7b This points out one of the major problem associated with
galvanostatic doping in the solid state. The calculation of accurate

values of band-filling is difficult, if not impossible. This percent

15

oxidation is probably less than 100% due to the background current
that cannot be subtracted. The insolubility also does not permit the
interaction between states to be probed either.

Electrochemical doping of the unsubstituted silicon
phthalocyanine polymers has been performed by Marks and co-workers.15
The silicon polymer was doped as a slurry in a three compartment
electrochemical cell. Prior to oxidation, the crystal structure of
the silicon polymer was orthorombic. After initial partial oxidation

to 502 at Ea > +1.80 V and reduction back to the neutral polymer at

PP
-0.20 V, the crystal structure of the polymer had changed to a
tetragonal phase. The tetragonal phase could then be partially
oxidized to 501 at a potential of Eapp - +1.35 V. The oxidation of
the tetragonal phase requires far less energy than the original
orthorombic phase. For these slurry doping experiments, the maximum
extent of partial oxidation obtainable is dependent on the supporting
electrolyte used for the experiment. Different counterions provided
different ranges of band filling, the highest partial oxidation, 67%
was obtained with the p-toluenesulfonate (TOS') counterion.15d The
electrochemical slurry doping of the unsubstituted silicon
phthalocyanines once in the tetragonal phase could be cycled from 0%
to 501 and back to 0% oxidation (BFA‘ counterion) without change in
redox behavior. The potential range required to go from 0 to 501
oxidation (BF4’ counterion) began at +0.30 V and ends at +1.20 V.
The doping profile of the polymer occurs smoothly and linearly over a
0.90 V range. The undoping of the 50% (BFa' counterion) oxidized

material again proceeds smoothly and linearly over a 1.0 V range

beginning at +0.90 V and ending at -0.10 V. The smoothness of the

16

doping profiles suggests that there are no additional structural
changes accompanying the oxidation process once the polymer is in the
tetragonal phase. Optical reflectance data for these materials
indicates that at low doping levels (less than 202 oxidation) the
phthalocyanine acts as a p-type (radical cation) semiconductor. .At
slightly higher doping levels the phthalocyanine polymer undergoes an
insulator to metal transition and begins to act like a low-
dimensional molecular metal. Again these materials are doped in the
solid state, prohibiting the acquisition of information on the
interaction between redox sites as well as the initial charge

transfer event .

The electrochemistry of several various length unsubstituted

d . 19 The monomer thru

phthalocyanines oligomers has been reporte
tetramer (l to 4 phthalocyanine rings) materials were capped using t-
butyl-dimethylchlorosilane to increase the compound's solubility in
nonaqueous solvents. The cyclic voltammograms of these compounds
were obtained and compared. For the oligomers, the first oxidation
and reduction process is energetically easier (less positive and
negative respectively), as the number of phthalocyanine rings
increases. In addition, as the oligomer length increases, the
potential difference between successive oxidations becomes smaller.
This is consistent with decreasing electrostatic repulsion due to
delocalization of the charge over a larger aggregate. From the

previous studies involving chemical13 1'5 of

and electrochemical doping
unsubstituted phthalocyanines, it is known that these materials have
a band-type electronic structure. Even with these small oligomers,

the interaction between adjacent macrocycles can be observed. In

17

solution, it appears that the cofacially joined phthalocyanines
polymers retain their rigid rod cofacial arrangement.

Most doping experiments in the solid state have concentrated
only on determining the degree of partial oxidation for a particular
doped system. Very little information has been obtained on the
chemistry and physics of polymers with interactive redox sites.
Solid state doping does not provide detailed information on the
charge transfer event or on the rate of charge-transfer between the
dopant and conducting polymer. The synthesis of electrically
conductive materials that are soluble in conventional solvents is
currently an area of considerable interest.20 The solubility of
these systems would allow for the direct characterization of their
redox properties by standard electrochemical methodology, eliminating
many of the problems associated with solid state doping. Kinetic
information on the doping event (i.e. how fast the material can
shuttle charge) is an important consideration for many of the future
applications of these materials. The substituted phthalocyanine
polymers provide an excellent system to study the redox properties of
a system with a large amount of interaction between adjacent
molecules. From the studies previously performed on substituted and
unsubstituted phthalocyanine polymers, it is known that in the solid
state, these material possess a band-type electronic structure. The
electrochemical studies of the various length oligomers indicate that
these materials do interact in solution as shown by the decrease in
potential required to oxidize successively larger oligomers.
Increases in electrical conductivity (substituted and unsubstituted

phthalocyanine polymers) of several orders of magnitude have been

18

obtained thru the use of chemical or electrochemical (as slurries)
doping. Information. on the doping process is lost due to the
inability to probe the unsubstituted materials. The possibility
exists with a soluble conductive polymer, that kinetic information on
the rate of electron transfer can be obtained. The solubility of
these substituted phthalocyanine polymers could also provide

information on the interaction of molecules in a conductive system.

2. Experimental

2.1 Instrumentation

2.1.1 Electrochemical

A Princeton Applied Research (PAR, Princeton, NJ) Model 273
potentiostat / galvanostat, interfaced to a Zenith (Franklin Park,
IL) Model 248 AT computer with a National Instruments model PCZA
IEEE-488 card for instrument control and data acquisition, was used
for the bulk electrolysis and voltammetry experiments. Experimental
results were plotted on a Hewlett-Packard (Palo Alto, CA) 7475A
plotter. A Bioanalytical Systems (BAS Inc., W. Lafayette, IN) Model
100A potentiostat, interfaced to a Hewlett-Packard Colorpro plotter,
was utilized for voltammetry experiments. A Zenith Model 158 KT
computer was used to store the acquired data. Inert atmosphere
electrochemical studies were performed either in specially designed
cells (described in section 2.2.2) or inside a Vacuum Atmospheres
(Hawthorne, CA) drybox.

2.1.2 Rotating Electrode Assembly and Bipotentiostat

A Pine Instruments (Grove City, PA) Model AFRDE4 bipotentiostat
coupled with a Pine Instruments AFMSR rotator were used to obtain
rotating disk voltammograms. The rotation rate for each voltammogram
was set by manually adjusting a rpm rate knob on the AFSMR rotator.
A Pine Instruments platinum disk and platinum ring-disk electrode

were used to obtain the rotating disk voltammograms. A Soltec (San

19

20

Fernando, CA) X—Y (Model VP-6423S) or a Soltec X-Yl-Yz (Model VP-
64248) recorder was used to record the voltammograms.
2.1.3 Spectroscopic Instrumentation

UV/Vis spectroscopy were performed with a Beckman (Fullerton,
CA) DU-64 spectrophtometer. Spectra were obtained using 1 cm matched
quartz cells. 1H NMR spectra were obtained with a Bruker (Billerca,
MA) WM-ZSO 250 MHz NMR. All of the 1H spectra are referenced to
tetramethylsilane. Fourier transform infrared spectra were obtained
on either a Nicolet (Madison, WI) 5-DX or Nicolet 760 instrument.

All spectra were obtained as nujol mulls using KBr plates.

2.2 Glassware
2.2.1 vacuum Lines
All solvents used in the electrochemical experiments were
transferred directly into the electrochemical cells using a high
vacuum (10'5 torr) dual-manifold vacuum line. After solvent
transfer, the electrochemical cells were backfilled with argon
(Matheson, 99.95%) to ensure their anaerobic integrity. The argon
had been passed through a drying column (Mn on silica) to remove
residual oxygen and moisture. Synthetic steps requiring the
exclusion of either air or moisture, were performed using standard
Schlenk-line techniques.21
2.2.2 Electrochemical Cells
Several types of electrochemical cells that interface directly
to the high vacuum line thru ground glass joints were used in these
studies. Controlled potential coulometry was carried out in an all

122

glass three compartment cel The counter electrode, constructed

21

of Pt gauze, was isolated from the working electrode compartment by
two fine porosity glass frits. The reference electrode, a Ag wire,
was separated from the working cell compartment by a fine porosity
glass frit. The working electrode for bulk coulometry studies was
constructed from Pt gauze and was approximately 8 cm2 in area. A
small area (0.018 cmz) platinum disk working electrode (BAS Inc., W.
Lafayette, IN) was also placed in the working compartment of the cell
to allow voltammetry studies to be performed. The small area Pt
working electrode was polished with 0.05 pm alumina (Buehler, Lake
Bluff, IL) prior to use.

Experiments involving only voltammetric measurements (cyclic
voltammetry, differential pulse voltammetry) were performed in single
compartment cells that also interfaced to a high vacuum line thru a
ground glass joint. A small area (0.018 cm2) platinum disk electrode
(BAS Inc., W. Lafayette, IN), a platinum wire counter electrode, and
a Ag wire reference electrode were used in these cells.

Rotating voltammetry experiments were performed in a three
compartment cell specially designed to reduce the formation of a

vortex . 23

A Ag wire reference electrode and a Pt gauze counter
electrode were placed in sidearm compartments separated from the main

part of the cell by fine porosity glass frits. A Pt ring-disk (Pine

22

Instruments, Grove City, PA) or a Pt disk electrode was used as the

working electrode (area of disk - 0.163 cmz).

2.3 Chemicals
2.3.1 Synthesis Solvents

Pyridine was dried and purified by distillation from CaH2 under
N2. Quinoline was stirred over barium oxide for two days and then
decanted. The decanted quinoline was then vacuum distilled and
stored over activated 3A sieves. Methanol was dried by distilling
from magnesium under N2. Ether was dried and deaerated by distilling
from sodium/benzophenone under N2. Chloroform was dried and purified
by distillation from CaHz under N2.

2.3.2 Electrochemical Solvents

l,1,2,2-tetrachloroethane (Aldrich Chemical, Milwaukee, VI) was
stirred with CaH2 for 48 hours, then vacuum distilled. The resulting
solvent was degassed by six freeze-pump-thaw cycles (liquid N2) on a
high vacuum line, and then vacuum transferred to a specially designed
round bottom solvent bulb containing activated 3A.mmdecular sieves.
For storage the solvent was subsequently freeze-pump-thaw one
additional time to ensure deaeration.

Tetrahydrofuran (HPLC grade, Burdick and Jackson, Muskegon MI)
was dried by addition of sodium-potassium amalgam (NaK). It was
degassed by six cycles of a freeze-pump-thaw sequence.

Methylene chloride (HPLC grade, Burdidk and Jackson, Muskegon
MI.) was degassed by four cycles of a freeze-pump-thaw sequence, and

then dried by stirring for three days with activated 3A sieves. The

23

dried solvent was then transferred to a solvent bulb containing
activated 3A sieves and treated to a final freeze-pump-thaw cycle.
2.3.3 Supporting Electrolyte
Tetrabutylammonium fluoroborate (TBABFA, electrometric grade,
Southwestern Analytical Chemicals, Austin, TX) was recrystallized
from ethyl acetate/diethyl ether and dried under dynamic vacuum (10'5

torr) for 2 days.

2.4 Synthesis
2.4.1 Trans-dichloro(tetra-t-
butylphthalocyaninato)silicon
The p-oxo-(tetra-t-butylphthalocyaninato)silicon polymer (t-
BuaPcSiO)n (Figure 2) was prepared by a modification of the

literature procedural-Ia"c

The synthesis of the polymer requires 10
steps from the initial starting materials. The individual steps are
outlined in Scheme 1. Several problems were encountered in the
synthesis of the dichloro(tetra-t-butylphthalocyaninato)silicon t-
BuaPcSiClz (step 8 in Scheme 1). Following the literature procedure
for the synthesis of t-BuchSiClz (8), 5-t-butyl-l,3-
diiminoisoindoline (5 g, 25 mmol, 95% pure by NMR) and 4.1 mL of
SiClh (Aldrich Chemical Co., Milwaukee, WI) were refluxed for 30
minutes in dry quinoline under argon. The reaction mixture turned
dark green and, after allowing it to cool to room temperature, was
diluted with 400 mL of dry chloroform. HCl gas (Matheson, Lynhurst,
NJ) was bubbled through this solution for 1 hour. It was then

filtered to remove any insoluble material (none was present). The

filtrate was reduced in volume under vacuum to the point where

24

Figure 2. Schematic of the peripherally substituted phthalocyanine

polymer (t-BuAPcSiO)n, central metal atom - Si, the chain length of

this polymer is approximately 25 units.17b

.3 es

es-.-
.3.

as. k.

 

 

2
URE
FIG

SCHEME 1
SYNTHESIS OF (t-Bu4PcSiO)n
4-t-bu l l
o-xylcne t-butyl chloride XL 1041:0120
N2. EIOH
i?

4-t-butylphthalic anhydnde R0 2 (I: _ OH
O c/0 :

acetic snhydride OC"—'0H

O

4-t—buty1phmalic acid
4 l. urea. A
2.1120 ('2'
g\ -—NH2
N - H
— NH
C/ "I "“3 X0 3 2
(ll) 4-t-butylphmalamide

4-t-butyylphthalimidc
NH 6 Q -P°C‘3
(Ii
7 CN
0 C/N . H N"3- Inhydrous 0
|| 3:] CN
N-H McOH
5 'l'butyl‘ l .3’di immo iSOindOlinc 4-t-butylphlhalonitrilc
SiCi4
8 . - \
umohnc
q |N' 350 °C.4H.
,-Bu‘pcsmz i:::::,’> t-Bu,PcSi(OH)2 22$ (t-Bu4PcSiO)“
""40" 10

9

27

crystallization began. The resulting thick green solution was
filtered under argon with great difficulty resulting in the isolation
of a small amount (0.1 g) of a microcrystalline powder. The reported
1H NMR spectrum (CD013) for t-BuhPcSiClz (8) has peaks at 61.8 ppm
(CH3 of the t-butyl group) and multiplets at 8.4 and 9.5 (ring
protons on the phthalocyanine).17c The 1H NMR spectrum (CDC13) of
the isolated material showed a series of peaks in the 81.0 to 2.0 ppm
region, (large singlet at 1.6 ppm, small singlet at 1.8 ppm). Other
peaks were observed at 7.9-8.1 ppm (multiplet), and 8.9-9.0 ppm
(multiplet), no peaks were observed at either 8.4 or 9.5 ppm. The
synthesis of was repeated four more times with similar results.

The 1H NMR spectra and correspondence with Professor M. Hanack
of the Univ. of Tubigen FRG, suggested that the product of the above
reactions was a quinoline adduct of the desired compound. Upon his
advice, the experimental procedure was modified in an attempt to
destroy this adduct . After refluxing 5 - t ~butyl - l , 3 -
diiminoisoindoline (5 g, 25 mmol, 95% pure by NMR) and 4.1 mL of
81014 for 30 minutes in quinoline, the solution was allowed to cool
to room temperature. To this solution 400 mL of concentrated HCl was
added. The resulting mixture was then heated to 50 °C and gently
stirred for one hour. The reaction mixture was then suction filtered
in air and washed with copious amounts of HCl and water. The 1H NMR
spectrum of the resulting product indicated that the desired product
was formed and was approximately 902 pure. The t-BuaPcSiClz (8)
being slightly soluble in ether, was purified via soxlet extraction.

The 1H NMR spectra of the ether extracted material (Figure 3) was

28

Figure 3. 1H NMR spectrum of dichloro(tetra-t-
butylphthalocyaninato)silicon, t-BuAPcSiC12, obtained in CDC13. The
spectrum is referenced to tetramethylsilane and the peak at 7.24 ppm
is due to residual CHC13. Peak assignments are given in the text.

 

 

n gnu—r.—

stmm
0.0 0. — 0.N 0.m 04 0.m 0.0 0.x. 0.0 0.0 0.0—
______p__.LL____Lh_FLe__L_ee_r___.__L_____m_.___p__

 

29

 

 

is I. . 4.

 

 

mze o» omonmmumm
2835822.2360215558341509.855
to 22:88 «22

 

 

30

identical to that reported in the literature and was approximately
981 pure.
2.4.2 Dihydroxy(tetra-t-
butylphthalocyaninato)silicon

The procedure for the conversion of [t-BuaPcSi(Cl)2] (8) to
dihydroxy(tetra-t-butylphthalocyaninato)silicon [t-BuaPcSi(OH)2] (2)
involved a slight modification of the literature method.17c A
suspension of t-BuchSiC12 was heated to 60 °C in 50 mL of a 50/50
mixture of ammonium hydroxide/pyridine in a sealed pressure tube for
24 hours. The resulting material was isolated by suction filtration
and dried under vacuum (10'3 torr). UV/Vis (Figure 4) and FTIR
(Figure 5) spectra of the dihydroxy capped monomer, t-buaPcSi(OH)2,

closely match those reported in the literature.17a

The large
absorbance peaks occur at 679 and 354 run The reported literature
values are 680 and 350 nm.176
2.4.3 p-oxo-(tetra-t-butylphthalocyaninato)silicon

The p-oxo-(tetra-t-butylphthalocyaninato)silicon polymer (19)
was formed by heating [t-BuhPcSi(OH)2] to 350 0C under dynamic vacuum
(10'5 torr) for 4 hours, in a tube furnace (Lindberg, Watertown,
WI).17a UV/Vis (Figure 6) and FTIR (Figure 7) spectra of the silicon
polymer, (t-buaPcSiO)n, closely match those reported in the

17a

literature. The degree of polymerization of the silicon polymer

31

Figure 4. UV/Vis spectrum of dihydroxy(tetra-t-
butylphthalocyaninato)silicon, thBuaPcSi(OH)2, obtained If! CH2C12.
Absorbance peaks occur at 679 and 354 nm.

 

 

32

 

v g9:

A220 1623?;
oodom 006mm oovmmm ooémv oodmm cohomm

 

 

i_||
mm< 00.0

IFI

muzozoz u0_xom0>I zoo_.:m MT: mo (aboumm m_>\>:

 

 

33

Figure 5. FT-IR spectrum of dihydroxy(tetra-t-
butylphthalocyaninato)silicon, t-BuaPcSi(OH)2 , obtained as a nujol
mull. The IR spectrum is identical to that reported in the

literature including the OH stretching vibration at 3500 cm'1

 

 

 

 

8Tr0nmnitt0nco

STrmnmnittmnce

49.167 50.333 07.500 76.667 05.031 95.000 0.0000 15.000 30.000 45.000 00.000 75.000 90.000

40.000

34

hydPOXSC.

 

 

 

 

 

 

#—

4000.0 3000.0 3200.0 3000.0 5400.0 5000.0 {000.0 1200.0 000.00fi400.00
Havonumoor (cm-1)

 

 

 

T hyfll‘OIiO.

l

A.

a

l

l.

.

i; F 4f A» + ' e A» e— #———i
3400.0 3200.9 1177.0 1000.? 003.55 044.44 733.33 022.32 on.” 400.00

unvonunhnr (ch-z)

FIGURE 5

35

Figure 6. UV/Vis spectra of the pristine, unoxidized silicon
polymer, (t-BuaPcSiO)n, obtained in CH2C12. The absorbance peaks
occur at 619 and 327 nm.

36

0559—..—
$20 152503;,

 

 

00.000 00._0mm 00._00m oowwma .00..lemi 00..00N

llfil
mm< 0V0

LI.

muz>mom zoo_i__m z_om_> MI... m0 4.5.0010 m_>\>0

 

 

37

Figure 7. FT-IR spectrum of p-oxo(tetra-t-
butylphthalocyaninato)silicon (t-BuaPcSiO)n, obtained as a nujol
mull. The IR spectrum is identical to that reported in the

literature.

40.000 00.000 70.000 00.000

fifrmnonlttmneo

-0.0000 13.000 00.000

Ifrmnomattoneo
08.000 00.000 00.000 77.000 06.000

43.000

00.000

38

P00 0800ND POLYflIR 00 Jul 00 11:47:10

 

 

 

 

 

W I

4000.0 3000.0 3300.0 5000.0 5400.0 3000.0 1000.0 1300.0 300.00 400.00
wmvonuabor (00-1)

#4

 

00 0000MB POLYNIR 00 Jul 00 11:47:10

 

 

 

 

 

$400.0 “03.3 Tums 1000.7 500.00 544.44 733.33 Tamas 0“.“ 700.00
Novena-oer (cu-1)

FIGURE 7

39

(t-buaPcSiO)n, has been determined by Hanack and coworkers by end-
group analysis to be 25.17b
2.4.4 Bis(trimethylsiloxy)(tetra-t-
butylphthalocyaninato)silicon
Bis(trimethylsiloxy)(tetra-t-butylphthalocyaninato)silicon (t-
BuAPcSi02)(Si(CH3)3)2, the capped monomer, was obtained by refluxing
1.0 mL of trimethylchlorosilane (Fluka Chemie AG, Ronkonkoma, NY),
0.10 grams of dried t-BuaPcSi(OH)2, and 20 mL of pyridine for five
hours. The resulting solution was taken to dryness and the excess
silane was removed in vacuo (10’3 torr). The dried solids were
chromatographed on alumina (activity 1) using toluene-hexane (1:2) as
the eluant. The capped monomer was the first major visible band.
Several indistinct bands came off after main band of capped monomer.
These bands may have been higher oligomers, however their limited
concentration prevented characterization. The 1H NMR of the capped
monomer in CD013 (Figure 8) was 69.55 (m, 3,6-Pc), 8.39 (m, 5-Pc),

1.81 (s, t-bu), -2.78 (s, CH3).

40

Figure 8. 1H NMR spectrum of bis(trimethylsiloxy)(tetra-t—
butylphthalocyaninato)silicon, (t-BuaPcSiOZ)(Si(CH3)3)2, obtained in
CDC13. The spectrum is referenced to tetramethylsilane and the peak

at 7.24 ppm is due to residual CHC13. Peak assignments are given in
the text.

41

ago:
zen.

0.0! 0.Nl 0.—l 0.0 0.— 0.N 0.0 0.v 0.0 0.0 0.5 0.0 0.0 0.0—

»Pmeerhee____:_L»LLFP__PCe_e__.___.eL__C:_:r:ep_e_::_:___

 

 

 

4 is L < c.

 

 

 

020. O» ouozwmmmum
zoo....mA0h<z_z<>oo._<I._.Im._C.DmIhl<mpme>x04>4.mq>h02_mbm_0
mo Inmpowmm mzz

 

 

3. Electrochemical Studies of u-oxo-(tetra-t-

butylphthalocyaninato)silicon

3.1 Bulk Oxidation by Controlled Potential Coulometry

3.1.1 Introduction

Bulk electrolysis is used to convert a compound from one
oxidation state to another, and can be used to determine the number
of electrons involved in the process.24 This is usually accomplished
by controlling the potential of the working electrode and integrating
the resultant current *with time (CPC, Controlled Potential
Coulometry). Rapid stirring of the solution is used to ensure
efficient mass transfer of material to the working electrode.
Controlled potential coulometry (CPC) was employed to determine the
degree of partial oxidation as a function of potential. This
technique was used since it is an absolute electrochemical method for

determining oxidation states.24

All of the polymer in the cell is
converted to a specific oxidation state depending on the potential of
the working electrode.
3.1.2 Procedure

Supporting electrolyte (TBABFA) was weighed out and added to
the three compartments of the CPC cell (Figure 9). The amount of
supporting electrolyte (TBABFQ) in all three compartments was enough
to produce a 0.2 M solution. A known amount of (t-BuchSiO)n polymer
(11.0 to 12.0 mg, 0.55 to 0.60 mg/mL) was added to the working

compartment of the cell. The coulometry cell was then connected to a

high vacuum line and evacuated (10"5 torr). After the entire cell

42

43

Figure 9. All ground glass joint coulometry cell utilized for the
bulk oxidations of the silicon polymer. The counter electrode (Pt
mesh, 2 cm2 in area) is separated from the working electrode by two
fine porosity glass frits. The Ag wire reference electrode is
separated from the working solution by a fine porosity glass frit.
The working electrode is constructed of Pt mesh approximately 8 cm2
in area. A small area Pt (0.018 cm2) electrode was also placed into
the working solution.

A - ground glass joints, B - counter electrode

C - glass frits, D - working electrode

E - small area working electrode, F - reference electrode

44

 

FIGURE 9

45

was cooled with an isopropanol-dry ice bath, a known amount of
solvent was vacuum transferred. The cell was then backfilled with
argon to ensure anaerobic conditions. After allowing the solution to
warm to room temperature, the cell was ready for use.
3.1.3 Results

The CPC oxidation experiments were conducted on neutral polymer
dissolved in l,1,2,2-tetrachloroethane. The potential for the onset
of oxidation of the polymer was determined from rest potentials of
the solution and differential pulse voltammetry (DPV) experiments.
Before the initial bulk electrolysis experiments were run, it was
necessary to ensure that the polymer was in its neutral form.25 This
was accomplished by reducing the initial solution at a potential
where the differential pulse voltammetry (DPV, vide infra) response
exhibited only background current (the region between the onset of
oxidation and reduction). The as-synthesized or virgin material had

typically been oxidized a few percent.25

After this procedure the
potential of polymer solution was approximately zero (0) volts.

To ”map-out" the 1 oxidation vs. potential profile for the
polymer, a series of CPC experiments were performed. The working
electrode potential was set to a value where oxidation of the polymer
would occur. The resulting current was monitored as a function of
time until it decayed to a limiting background value (Figure 10).
Generally, the background current (10'7 A/cmz) was three orders of
magnitude less than the initial current (10'4 A/cmz). The number of
coulombs passed at a given potential was determined by integrating

the current over time. This value is comprised of contributions from

both the oxidation of the polymer and any background processes.

46

Figure 10. Current-time curves for the oxidation of neutral silicon
polymer, (t-BuAPcSiO)n, in l,1,2,2-tetrachloroethane / 0.2 M TBABFA
at different potentials. The solution concentration was 0.0288 mM
based on a 25 phthalocyanine unit polymer or 0.72 mM based on a
single phthalocyanine moiety.

47

 

 

 

 

           

 

 

 

$052.3 us: 2850....
8.02 8.8 8.8 8.9. 8.8 8.0
p r _ _ _ nooluuél
5 B > o 28... 20.2938
m; E > o 29: 20.2020 _ . O
<1 oom n
H
_ Mu
1...
n3 2 > o :9: 292938 .N..
I E > o 29.... 20.2936 .
ad 8 > 0 sex... 20.2938 \V/
N
d
3
no 9 > o 29.... 20.298 we.
. e.
no 8 > 0 20m... zo:<o_xo
1-- . , -- I 08335
$2500 2836 a: to 20595
mi: mod mbammm mzeiezmmmao E52338

 

 

48

Since the current arising from the background oxidations are usually
constant over the lifetime of the experiment, the amount of charge
consumed by these events can be estimated by multiplying the total
time of the experiment by the limiting background current value. The
net number of coulombs for oxidation of the polymer was calculated by
subtracting the number of background coulombs from the number of
gross coulombs recorded. Once the polymer was oxidized, it was then
reduced back to the neutral state (Figure 11) by setting the
potential of the working electrode to a value 50 mV negative of the
rest potential of the neutral polymer. The stability of the polymer
was checked by DPV after this cycle. Once the polymer was back in
the neutral form, another oxidation/re-reduction cycle was run at a
different potential. ,In order to map out the 1 oxidation 'vs.
potential curve (Figure 12), a total of 10 oxidation/re-reduction
cycles were run.

The number of electrons transferred for a simple redox process
can be calculated from Faraday's law.

Q—nFN (1)

where:

Q

charge (coulombs)

number of electrons

:3
I

-1)

u;
I

Faraday's constant (96,485 coulombs mole

Z
I

moles of analyte

The (t-BuaPcSiO)n polymer is comprised of redox active sites coupled

together thru axial covalent bonds. The number of sites in the

49

Figure 11. Current-time curves for the reduction of oxidized
silicon polymer, (t-BuaPcSiO)n, in 1,1,2,2-tetrachloroethane / 0.2 M
TBABF,I at different potentials. The solution concentration was
0.0288 mM based on a 25 phthalocyanine unit polymer or 0.72 mM based
on a single phthalocyanine moiety.

50

 

 

 

 

 

 

 

A8523 02: 3850:
8.83 8.8 8.8 8.8 8.8 8.0
i 2: it: #8388
> o B > we 28... 29588

> o E > no 28... 20:88”. .
O
> o 2 > me :8... 29588 0
Nu
> o B > I 28... 20:88”. mm
N
> o E > 3 :2: 2058mm - l
> o E > ms :9: 20:88”. W)
> o 9 > 5 28... 20.838”. _ mm
<3. 8... - a
_.1_
_ S
(
I nooluué

$258 2830 850:8 mi to 20.888
UT: m8 mbammm uselezmmmao E52828

51

Figure 12. The degree of oxidation_ vs. potential for the
electrochemical oxidation of the silicon polymer, (t-BuaPcSiO)n, in
1,1,2,2-tetrachloroethane / 0.2 M TBABFa. Each point corresponds to
the mean value for a complete oxidation/re-reduction cycle performed
by controlled potential coulometry. The error bars represent the
difference between the oxidation and re-reduction values. Solutions
were typically 0.02 to 0.04 mM based on a 25 phthalocyanine unit
polymer or 0.5 to 1 mM, based on a single phthalocyanine moiety (0.4
- 0.8 mg polymer/mL). The 1 oxidation values were determined by
relating the mass of material present to the amount of charge passed
at the given potential (see equation 2).

52

«3:52..—
Gomm .2 233 2:238

00.0 00.0 00._ 0m.— 00.N
a. I1 a.

 

 

N

28 08 * W
*

fl

toErom .3025 of ._2 32:23. .m> co:on_xon

10—
[ON
low
:9»
10m
:00
non
r00
rom
r00—

 

 

NOLWCJIXOas

53

polymer depends on the degree of polymerization. Oxidation of
neutral polymer causes charge to be removed from these sites.
Results of the coulometry experiments are expressed as percent
oxidation, or percentage of polymer oxidation. Since each
phthalocyanine ring is a site for oxidation, the number of
phthalocyanine moieties in each experiment must be calculated. This
can be done from the number of grams of polymer used, and the
molecular weight (781 grams mole'l) of a single phthalocyanine

unit.17a

From the number of moles of -(t-Bu4PcSi0)- (phthalocyanine
moieties), and the number of coulombs passed, the 1 oxidation of the

material can be determined by:

Qexp(WT/MW*F)-n (2)
where:
Qexp - experimentally determined coulombs
WT - weight of (t-BuaPcSiO)n in grams
MW - molecular weight of -(t-Bu4PcSiO)-
F - Faraday's constant (96,485 coulombs mole'l)
n - fractional number of electrons transferred per site

The 2 oxidation at a specific potential is determined by ratioing the
number of coulombs measured at that potential to those required for
100 1 oxidation. These results are summarized in Table l and Figure
12. By varying the potential of the working electrode, and
calculating the number of. background corrected coulombs at the
different potentials, the percent oxidation at different potentials

was determined. The 1 oxidation vs. potential profile of the polymer

54

TABLE 1

Z OXIDATION VS. POTENTIAL FOR THE SILICON POLYMER

ENTIALa 029 DATAb Ens DAIAC SIMULATION 05:54
0.42 11.5 (i 1.2) 14.3 (i 0.6) 8.5
0.64 27.9 (i 1.0) 21.2 (i 0.6) 18.9
0.82 31.4 (i 1.5) 30.5 (i 1.0) 30.6
1.02 47.5 (i 0.6) 41.6 (i 1.0) 41.9
1.22 55.1 (i 5.4) 52.7 (i 1.2) 50.7
1.42 74.1 (i 2.4) 68.3 (i 1.5) 65.5
1.62 83.2 (i 4.8) 85.1 (i 2.1) 84.6
1.82 93.0 (i 1.1) 1006 99.9
2 02 100.0 (2.8) 100e 100 0

potential in volts (V) vs. aq. SSCE in 1,1,2,2-tetrachloroethane.

average value and the deviation based on the inital oxidation of the
polymer and rereduction back to the neutral form.

average value and the deviation based on the voltammograms shown in
Figure 13. See the text for complete details.

based on 25 redox sites. The standard potentials for each couple
used in the simulation are listed in Table II. See the text for
complete details.

limiting current of the RDE voltammograms were taken as representing
100% oxidation.

55

shown in Figure 12 occurs over a 1700 mV range, which is indicative
of a polymer in which there is a large degree of interaction between
the redox sites. As the error bars illustrate (Figure 12), the
number of coulombs necessary for the oxidization and subsequent re-
reduction back to neutral (t-buPcSiO)n polymer were virtually
identical . The small difference in these values suggests that the
polymer is stable at all degrees of partial oxidation. The smooth
doping profile suggests that there is a continuum or band from which
charge can be removed and that there is no large amount of inherent
stability at any particular degree of partial oxidation. If there
was some instability at a given partial oxidation state, the number
of coulombs for the re-reduction of the oxidized polymer would be
different compared to the initial oxidation.

As stated earlier, the chemical stability at a specific
oxidation state was also monitored by DPV. After an oxidation/re-
reduction cycle, a DPV was run and compared to that of the uncycled
or virgin polymer. The DPV for initially uncycled polymer and the
oxidation/re-reduction cycled polymer were essentially identical even
after 10 cycles. These results suggest that the polymer is stable at
all degrees of oxidation. If the polymer was unstable toward
oxidation at a specific potential, the DPV would be expected to show
a significant change after the oxidation/re-reduction cycle. This
was never observed. The UV/Vis absorption spectra of neutral
uncycled virgin polymer (Figure 6) and extensively oxidation/re-
reduction cycled neutral polymer, including 1001 oxidation (Figure
13), are similar, with no peaks associated with monomeric fragments

(Figure 4). Absorbance peaks in the spectra of the neutral uncycled

56

Figure 13. UV/Vis spectra of the extensively cycled silicon
polymer, (t-BuaPcSiO)n, obtained in CH2C12. Absorbance peaks occur
at 625 and 327 rmL The UV/Vis spectra of the extensively cycled
silicon polymer is essentially the same as the pristine polymer.

57

mm ”.552..—
0220 Eozmdi;

 

 

00.000 09.000 oowmmm oowvmv 00._Nmm 00+00N

 

l—II
mm< 0m.0

LII

«02>..0m zoo_.__m 0040>0 ”2.: m0 <m._.own_m m_>\>0

 

 

58

virgin polymer (Figure 6) occurred at 619 and 327 nm. The literature
reported wavelengths for the same peaks are 612 and 322 nm.17a The
extensively cycled polymer has absorbances at 625 and 327 nm, similar
to the uncycled polymer and to the literature values. The UV/Vis
absorption spectra suggests that the polymer is stable to

oxidation/re-reduction cycles.

3.2 Rotating Disk Voltammetry Studies
3.2.1 Introduction

Conventional voltammetric techniques can also be used to
characterize the redox properties of the polymer. In rotating disk
voltammetry, the working electrode is connected to a motor, which
spins the electrode and causes sample to pass by the electrode due to
forced convection of the solution. Most voltammetric techniques are
performed on quiescent solutions, where diffusion is the main mode of
transport of material to and from the electrode. In RDE voltammetry,
the rotation of the working electrode causes the rate of mass
transfer to be much greater than the rate of diffusion.24 Thus,
during an RDE experiment, electrochemically unperturbed material is
continuously being brought up to the electrode due to the rapid
transfer of sample by forced convection. The rotation of the working
electrode, in conjunction with a slow potential scan rate (10
mv/sec), enables these experiments to be obtained under steady state
conditions. The current at any potential is independent of scan
direction and time. Although currents in rotating disk voltammetry
are typically larger than in other voltammetric techniques (cyclic

voltammetry and differential pulse voltammetry), the total current is

59

small enough that little net electrolysis of the bulk solution
occurs. Rotating disk voltammetry was utilized to determine if there
were any kinetic limitations to the oxidation of the polymer. The
current-potential curve obtained from RDE voltammetry can also be
used to obtain the doping profile of the polymer as a function of
potential. The RDE experiment provided the same information as an
entire series of CPC experiments in a few minutes.
3.2.2 Procedure

Supporting electrolyte (TBABFA) was weighed out and added to
the three compartments of the RDE cell. The concentration of
supporting electrolyte (TBABFA) in each compartment was enough to
produce a 0.2 M solution. A known amount of polymer (11.0-12.0 mg,
0.55-0.60 mg/mL) was weighed out and added to the working
compartment. The typical concentration of the polymer in the cell
was 2.8 x 10'5 M to 3.2 x 10'5 M based on a 25 phthalocyanine moiety
polymer.17b The solvent used for these experiments, 1,1,2,2-
tetrachloroethane, was vacuum transferred into a solvent transfer
tube and then pipetted into the cell in an inert atmosphere glovebag.
Prior to electrochemical measurements, the working compartment was
extensively deaerated by bubbling with argon. The working electrode,
a platinum ring-disk electrode (Pine Instruments, Grove City PA) was
polished with 0.05 pm alumina prior to use.

3.2.3 Results

Shown in Figure 14 are a series of rotating disk electrode
(RDE) voltammograms of the silicon polymer obtained at different
rotation rates. The potential was scanned over the entire range

defined by the CPC results. The starting potential was set negative

60

Figure 14. Rotating disk voltammograms of a 0.0208 mM solution of
the silicon polymer, (t-BuaPcSiO)n, in 1,1,2,2—tetrachloroethane /
0.2 M TBABFA at a Pt disk electrode at different rotation rates.
The background RDE voltammogram of a blank solution was obtained at
1800 rpm. The current values for the background RDE did not vary as
a function of rotation rate.

61

 

Snag—Dar.—
Amomm m> m:o>v ._<:zm._.0n_

00.0 00; 004 cod
4 1.. 0

W

{I‘ i’

< I 2mm 000— x250

2mm 000
2mm 000—
<U._.m.nm 2mm 005w
l_..| 2mm comm
2%. 0000

tmEron. cooEm or: 00 mEotmoEEo:o> v35 0:20.61

 

 

62

of the initial oxidation response observed in DPV and CPC
experiments. The final potential was determined by the oxidation of
the solvent. The increase in current at high positive potentials is
due to this process. The RDE voltammograms occur over an
approximately 1700 mV range, in contrast to a typical rotating disk
response for a single conventional redox couple would require

approximately 200 mV to change its oxidation state?“

The general
equation ‘which describes the current response for rotating disk

voltammetry is:

1 _ -------------------------------------- (3)
1.61 001/3 w‘l/z u1/6

where:

w - angular velocity of the disk (w-ZnN, N-rps)

y - kinematic viscosity of the fluid, cm2 sec"1

1 - current, A

00* - initial solution concentration, mol cm'3

Co(y-O) - solution concentration at the electrode, mol cm.3
Do - diffusion coefficient, cm2 sec'1

A p - area of the electrode

n - number of electrons

The 2 oxidation vs. potential plot (Figure 15) for the polymer
can also be calculated from the rotating disk voltammograms. The
current at 1.82 V (Figure 14) is the limiting current for the
rotating 1disk 'voltammograms. At this potential, the polymer is
assumed to be 1001 oxidized (supported by CPC results). The percent

oxidation at a given potential, can be calculated from the limiting

63

 

sOxidotion vs. Potential for the Silicon Polymer

1000 O
90— IE
80- O RDE Data
70-9 m
60—
50- 8
40- @

30- :3:
204 9
10-

SOXIOATION

@
1 a .1;
2.00 1.50 1.00 0‘é___0"‘0‘. 0 .0

POTENTIAL (Volts vs. SSCE)

 

 

Figure 15. The degree of oxidation vs. potential for the silicon
polymer, (t-BuaPcSiO)n, determined from the RDE voltammograms. The
RDE values were determined by normalizing the current at any
potential to the limiting current at 1.82 V5 The error bars
represent the standard deviation of the calculated 1 oxidation, for
the same potential, at different rotation rates.

64

current and the current at that specific potential. The 1 oxidation
is calculated by dividing the current at the given potential by the
limiting current and multiplying by 100. Consistent 2 oxidation vs.
potential curves are obtained at all rotation rates. The error bars
in Figure 15 represents the standard deviation of the 1 oxidation vs.
potential values calculated from the five individual RDE
voltammograms. In addition, as seen in Figure 15 (2 oxidation vs.
potential for RDE and CPC), there is excellent agreement between the
two inherently different techniques for the determination of the 2
oxidation vs. potential curve. The shape of the RDE voltammograms
are independent of angular frequency (w-2xN, N — rps) and increase in
magnitude with 091/2. This behavior is predicted only for a
reversible, nernstian redox system. The magnitude of the limiting
current is described by the Levich equation.
3.2.4 Levich Plot

If there is no kinetic barrier to electron transfer, the

increase in current for rotating disk voltammetry is dependent only

24a

on the increase in rotation rate (col/2). The equation that

65

describes the limiting current for a rotating disk voltammogram is

given by the Levich equation.

1L - 0.620 n F A 00 002/3 u'1/5 01/2 (4)

w - angular velocity of the disk (w-2aN, N-rps)

u - kinematic viscosity of the fluid, cm2 sec"1

1L - limiting current, A

Co - solution concentration, mol cm.3
Do - diffusion coefficient, cm2 sec’1
A - area of the electrode

n - number of electrons

For a specific redox system, all of the terms in the Levich equation

are constant except 1L and col/2.

1/2.

The current, at any potential,
should vary as w 'A plot of i(E) vs. w1/2 for a reversible system
should yield a straight lines that intersect the origin. A deviation
from a straight line suggests a kinetic limitation or barrier for the
electron transfer reaction. A plot of current vs. wl/Z (Figure 16),
does in fact yield straight lines that intersect the origin at all
potentials for the doping process. The deviation at the high
potential values is probably due to the oxidation of the solvent.

The straight line behavior is indicative of a simple, uncomplicated

charge transfer process, and suggests that there are no 'kinetic

66

Figure 16. Plot of disk current at various potentials and rotation
rates vs. w1/2 (w - 2xN, N-rps), for the oxidation of the silicon
polymer, (t-BuaPcSiO)n, in 1,1,2,2-tetrachloroethane/O.2 M TBABFQ.

67

say—DUE
.000... ”730024. 00 ~50

8.8 8.8 8%. 8.8. 8.m

_

 

 

>>

00.0
00.0
00.0
00.—

00.—

00.—

00.—

 

1w .1.

A.
Q‘ a!

i
«i Q» 4): q-
l‘ 1 a»

*1
L‘
k‘

8‘

>zoo_o> 3.30:4. .m> 29.30

I

l

l

 

 

00.0

00.0—

00.0_

00.0w

(v 11) 1N388n0

68

limitations for the oxidation of the polymer. Thus, charge can be

easily removed from the polymer at all levels of oxidation.

3.3 Differential Pulse Voltammetry
3.3.1 Introduction

Differential pulse voltammetry (DPV), a very sensitive
derivative technique was also used to characterize the redox response
of the silicon polymer. DPV is an attractive technique for studying
multi-electron transfer events due to its high sensitivity and
resolution. Unlike the featureless response observed for most
voltammetric techniques, DPV is extremely sensitive to subtle changes
in the doping profile. DPV was employed in hopes of observing the
individual redox states comprising the polymer.

3.3.2 Procedure

The differential pulse voltammetry experiments were performed
in a single compartment cell. The working electrode was a small area
Pt electrode (BAS Inc., W. Lafayette, IN) that was polished prior to
use with 0.05 pm alumina. Enough supporting electrolyte (TBABFA) was
weighed out and placed in the cell to provide a 0.2 M solution. A
known amount of polymer was weighed out and placed in the cell to
provide a 0.55-0.60 mg polymer/mL solution. The polymer was placed
in a dumpster, a rotatable reservoir that allowed the addition of
polymer at any time. Utilizing this dumpster, voltammograms of the
blank solution could also be obtained. These background voltammograms

could then be subtracted from sample voltammograms to obtain

69

background corrected data. The background voltammograms were also
useful in determining the potential window of the solvent.
3.3.3 Results

Shown in Figure 17 is a DPV of the (t-BuchSiO)n polymer. The
DPV response of the polymer is approximately 1700 mV wide, which is
similar to that obtained by CPC and RDE experiments. In contrast, a
DPV of a single non-interacting redox couple, for example ferrocene,
would typically have a base peak width about 200 mV wide.2l‘a’26 The
1700 mV potential width of the polymer's differential pulse
voltammogram is indicative of a material with multiple interacting
redox sites. These multiple interacting redox sites provide a large
energy range of states from which charge can be removed. The DPV
observed for the polymer does not have a symmetrical response. The
DPV shows an onset of current at 0.15 V, rising to a peak around 0.7
V, where it then dips sharply. .The current levels off at 1.00 V and
then begins to increase sharply to a current maximum at 1.55 V.
After this current maximum, the current decays to background levels,
suggesting that no further oxidation is occurring. The current dip
at 1.00 V occurs after approximately 451 oxidation based on
controlled potential coulometry data. This nonlinear doping profile
is indicative perhaps of some stability or reorganization for the
polymer occurring at or near 452 oxidation.27 The steady increase in
current from 1.00 V to the maximum at 1.55 V is indicative of a
greater density of states in that potential region than in the region
from 0.15 V to 0.95 V. While DPV is one of the most sensitive
electrochemical techniques, it did not provide any information on the

individual redox potentials of states in the polymer. The degree of

70

Figure 17. Differential pulse voltammogram of a 0.056 mM solution
of silicon polymer, (t-BuaPcSiO)n, in 1,1,2,2-tetrachloroethane /
0.2 M TBABFA obtained with a small area 0.018 cm2 Pt electrode. The
scan rate was 4 mV/s, pulse amplitude - 50 mV, pulse width - 50 ms,
and the pulse period - 1000 ms.

71

5.559..—
Gomm 9, 6:60 .3538
8.0.. 8.0 8.0 8.. Pm;

llJlllllAlllllJl-IIIIIJ

_‘

(c on

l—

..oEzom c0050 9.: yo
EotmoEEo:o> 023m 32:83.20

 

 

 

72

polymerization of the silicon polymer (n - 25) has been determined
from end group analysis and is reported in the literature.17b This
information, along with the concentration of the polymer, area of the
electrode, diffusion coefficient of the polymer, and the potential
pulse sequence employed, determine the potential-current response.
These experimental parameters were used as variables in a digital
simulation, so that the experimental current-potential DPV response
of the polymer could be modeled.
3.3.4 Digital DPV Simulation

The model utilized for the digital simulation of the DPV
response of the polymer, is based on sequential electron transfer of
redox couples in equilibrium with one another (see Appendix A). This
is exactly analogous to the sequential proton transfer in polyprotic
acids.28 The ultimate goal of the digital simulation was to estimate
the approximate redox potentials of sites in the polymer, their
distribution, and to obtain some information on the current dip at
1.00 V (corresponding to 451 oxidation). The modeling of the DPV
current potential response should also provide information on the
chainlength of the polymer. The computer program is based on the
finite differences method and initially divides the solution up into

a number of volume elements or boxes.29

The program simulates
diffusion between the volume elements and calculates the surface and
bulk conditions of each redox state of the polymer. The initial and
bulk concentrations are the same as the experimental values in Figure
17. The next step is to set up the conditions at the working

electrode and to calculate the dimensionless current. The program

then goes back and recalculates the diffusion of material to and from

73

the electrode. The program executes these steps continuously until
the final potential of the experiment is reached. The ratio of
reactant and product concentrations at the electrode surface is based
the Nernst equation and is a function of electrode potential. The
number of states in the polymer that are oxidized is dependent solely
on the standard potentials of the sites and the potential of the
working electrode. The current is calculated from the flux of the
polymer at the electrode surface and number of states of the polymer
being oxidized. The potential of each individual redox state in the
polymer was inputted into the computer program (Table 2) along with
the other experiment parameters necessary for the simulation. The
digital simulation utilized the same experimental conditions and
parameters as those used to acquire the DPV shown in Figure 17. The
digital DPV simulation, Figure 18, is very similar to the
experimental DPV of the silicon polymer, Figure 17. The simulation
program was run numerous times on a trial and error basis to achieve
the best fit between the real and simulated data. After the
simulation was run several times, the response of various changes in
the parameters could be predicted. A more rational, calculating
approach was then taken to the refining of the simulated DPV. The
shape of the simulated DPV is extremely sensitive to the standard
potentials chosen for the redox couples used to model the polymer.
Small changes in the distribution of these values caused drastic
changes in the current-potential shape of the simulated DPV. From
the simulation, information is obtained about the standard potential
of the sites in the polymer. The experimental and simulation DPV,

Figures 17 & 18, are very similar in their current-potential shapes.

74-

TABLE II

Digital Simulation Data for the Silicon Polymer

Redo: Site Standard Potential (£0)‘ Potential Difference
Between Successiv
Sites

1 0.275

9.0
2 0.365

8.5
3 0.450

8.0
9 0.530

7.5
5 0.605

7.0
6 0.675

6.9
7 0.744

6.7
8 0.811

6.4
9 0.875

7.0
10 0.945

7.5
11 1.020

11.0
12 1.130

8.0
13 1.210

6.6
14 1.276

5.4
15 1.330

5.0
16 1.380

4.7
17 1.427

4.4
18 1.471

4.2
19 1.513

4.1
20 1.554

4.0
21 1.594

4.0
22 1.634

4.1
23 1.675

4.4
24 1.719

5.5
25 1.774

a in volts (V).

b in Iiilivoltl (IV).

75

Figure 18. Simulated differential pulse voltammogram for the
oxidation of the silicon polymer. The conditions utilized (scan
rate, pulse width, concentration, pulse amplitude, area of
electrode) for the simulation are the same as those of the
experimental data shown in Figure 13. The bars along the X-axis
represent the standard potentials of the redox couples utilized to
obtain the current-potential response (see Table 2 for the exact
potentials).

76

3559—..—
Gomm 2, «=90 2:238

00.0! 00.0 00.0 00»— 00.. 00.N

l1.|||l1||llJJ |..l l.|..J

_.|

(c 0N

l‘

 

toEZom c0050 2.: *o
EotmoEEo:o> 33¢ 325.625 “0223.50

 

 

 

77

Thus, the redox potentials used in the digital simulation
approximately match the states in the polymer. Information obtained
from the distribution of the redox states in the simulation is quite
interesting. Shown in Figure 19 is a plot of the difference in
potential between successive sites used in the DPV simulation. The
difference in potential initially between the sites is approximately
9 mV. The potential difference between successive sites then
steadily decreases to 6.4 mV between sites 8 and 9. The difference
in potential then increases with a substantial gap of 11 mV between
site 11 and 12, this corresponds to a current gap in the simulation
similar to the experimental DPV dip at 1.00 V. After the dip a 1.00
V, the potential difference between successive sites slowly decreases
in a smooth transition. The smallest potential difference is found
between sites 20, 21, and 22 which are only 4 mV apart. The
potential difference for the remaining sites increase slightly to a
final potential difference of 5.5 mV between sites 24 and.25. The 1
oxidation vs. potential for the polymer (Figure 20) was also
calculated based on this sequential electron transfer model. These
values are determined assuming a nernstian distribution of species
dependent on the electrode potential. The concentration of the
unperturbed and oxidized states in the polymer was calculated from
the Nernst equation knowing the redox potential of each site and the
potential of the working electrode. Shown in Figure 21 is a plot of

1 oxidation vs. potential from CPC experiments and the digital DPV

78

Figure 19. A plot of the difference in standard potential between
successive sites utilized in the DPV, CV, and RDE simulations of the
silicon polymer (see Table 2 for the exact values).

 

 

79

a nay—DUE

100202 :5
00.0w 00.0w 00.0. 00.0. 00.0 00.010
0
* *** mm
* ** -
* ** **** -
** I
r0.
* .
10.
toEzoa cooEm of to.
3:0 ozmmooosm c3330 mo:_o> cm 5 3.3.335

 

 

 

(A90 63V

80

Figure 20. The degree of oxidation 'vs. potential calculated by
digital simulation using the potentials listed in Table 2. The 1
oxidation is not calculated from a particular technique, rather it
is based on the number of sites oxidized at a specific potential.
The number of sites oxidized is determined from the potential of the
electrode, the standard potential of the sites, and the Nernst
equation (see also Appendix A).

81

   

 

3.559..—
Gomm .9, 6:60 2:58....

 

00.0 00.0 00.— 00; 00.N
1J.

I...

8.635%

$8300 .3220 of LB _o:c2om .m> cozoExou

:0—
ION
:00
10.4
r00
r00
105
T00
r00
100—

 

 

NOIlVCJIXOas

82

Figure 21. The degree of oxidation vs. potential for the silicon
polymer, (t-BuaPcSiO)n, from the CPC data and the digital
simulation.

83

 

Guy—DUE
Comm .9 £60 55.58..

00.0 00.0 00.. JPMJ 00.N

28 80 * m
8.3.565 >8|

w

.3530... c0050 2.: ..o.. 32:20.". .m> cot—62x0...

:0—
[ON
100
low
r00
:00
[on
r.00
:00
:00—

 

 

NOLW'CIIXOas

84

simulation. There is excellent agreement between the two sets of

data.

3.4 Cyclic Voltammetry
3.4.1 Introduction
Cyclic voltammetry (CV) has also been utilized in the study of
the silicon polymer. In this particular technique, the potential is
cycled from some initial potential to a switching potential and then
back to the initial potential at a constant sweep rate. From the
cyclic voltammogram or current-potential response, several different
characteristics of the system being probed can be determined. From
the peak splitting of the anodic and cathodic peaks, the number of
electrons involved in the redox event can be determined.24 The
stability of reactants and/or products, and kinetic information
regarding the electron-transfer reaction can be probed by varying the
sweep rate. Products of a redox event that decompose can be
”captured” and potentials required for their oxidation or reduction
can be determined. These traits have made cyclic voltammetry one of
the most widely used electrochemical techniques.
3.4.2 Procedure
Cyclic voltammetry studies were carried out in a one
compartment cell, identical to that used for DPV studies. The
procedure for the experiment is the same as for DPV (see section
3.3.2).
3.4.3 Results
Shown in Figure 22, is a typical cyclic voltammogram (CV) of

the silicon polymer. The current is relatively high, however, there

85

Figure 22. Cyclic voltammogram of a 0.0432 mM solution of silicon
polymer, (t-BuchSiO)n, in l,1,2,2-tetrachloroethane / 0.2 M TBABFa
obtained with a small area 0.018 cm2 Pt electrode. The scan rate
was 100 mV/s.

 

86

 

«um—~52..—
Gomm ....> 6:20 2:238

 

 

 

 

toEEom c0050 9.: 0o EotmoEEozo> 0:96

88.. 8.0 8.0 8... 8.. 8.~
._
8_
. I._|
<1oflo
Ll

 

I315

'58:;

lice:

 

87

are almost no discernable features. The CV is extremely smooth and
broad, very similar to a background voltammogram. In contrast,
observe the CV of the bis(trimethylsily1oxy) (tetra- t -
butylph thalocyaninato ) s i 1 icon , the capped -monomer [ (t -
BuAPcSiOZ)(Si(CH3)3)2], Figure 23. The CV of the capped monomer
shows two reversible oxidation waves. The CV of the monomer shows
the typical current-potential shape, a large anodic current
(oxidation) and after potential reversal, a large cathodic current
(reduction). The peak splitting of the wave should only be
approximately 60 mV/n apart for a reversible system, and in fact the

peak splitting is close to this value,24a

The capped monomer
actually undergoes two one electron oxidations separated by 900 mV,
forming the stable monocation and dication as shown by the
reversibility of the CV. The CV of the polymer can be digitally
simulated using a program very similar to that used for the digitally
simulated DPV (see section 3.3.4). Both electrochemical simulation
programs are structured the same way, utilizing the finite
differences method and modeled based on sequential electron transfer
between states in equilibrium with one another. In Figure 24, the
digitally simulated current-potential curve of the polymer, shows a
CV which is very similar to the experimentally observed voltammogram.

The potentials used for the simulation are the same values used for

the DPV simulation (Table 2). The CV shows very little information

..__.—.-‘F~'- 9 M- A- -

88

Figure 23. Cyclic voltammogram of a 1.25 mM solution of
bis(trimethylsiloxy)(tetra-t-butylphthalocyaninato)silicon, (t-
BuaPcSiOZ)(Si(CH3)3)2, the capped monomer in 1,1,2,2-

tetrachloroethane / 0.2 M. TBABFQ obtained with a 0.018 cm2 Pt
electrode. The scan rate was 100 mV/s.

 

89

 

 

mum—~52..—
Gomm 9 ”=60 2:228

00.0 00.0 00.. 00.— 00.N
I7 I.»

JIIIIJI JJ

 

 

|_|
<1 o.~
Ll

.5893: c850 toaaoo or: 00 EotmoEEozo> 0:96

 

 

90

Figure 24. Simulated cyclic voltammogram of the oxidation of the
silicon polymer. The standard potentials used in this simulation
are the same as those used for Figure 13 (see Table 2 for the exact
potentials). The conditions utilized (scan rate, concentration,
area of electrode) are the same as those in Figure 18.

91

 

quay—DUE
Comm 9, «:60 8:55..

 

 

 

 

 

00.0.. .0 . 0b.. 00. _ 00.0
fill—I . . .JF JIJ
..
0 _ / l||\li
.I_|l
<1oflo
l_|

3810.. c0050 2.: 0o EotmoEEozo> 0:950 0223.50

 

 

92

due to the close proximity of the redox states in the polymer. The

broad featureless CV is actually the response expected.

3.5 Chronocoulometry
3.5.1 Introduction
Chronocoulometry is an electrochemical technique where the
potential of the working electrode is stepped from an initial
potential where no faradaic current flows to a final value where

electron transfer occurs.248

In most cases the potential of the
electrode is set well past the half wave potential of the redox
couple to give a diffusion-limited current. The potential of the
electrode is kept constant for a specific length of time, and the
resultant charge is measured.
3.5.2 Procedure
Chronocoulometry studies were carried out in a one compartment
cell, identical to that used for DPV studies. The procedure for the
experiment is the same as for DPV (section 3.3.2).
3.5.3 Results
The diffusion coefficient of the polymer was determined from
Chronocoulometry experiments. The integrated Cottrell equation,
which describes the amount of charge passed as a function of time,
was used to determine the diffusion coefficient of the polymer. The
potentials utilized for the experiment were obtained after consulting
the DPV response of the polymer. The initial starting potential of
the experiment, was a potential where no redox response was observed

in the DPV of the silicon polymer. The final potential, was a value

where the DPV response indicated that no further oxidation of the

93

polymer was occurring. In the equation below, all of the parameters
were known, except for the diffusion coefficient of the polymer.
Typical values for a chronocoulometry experiment of the silicon

polymer are shown below.

Qd-ZnFACO 001/2 J” ’1/2 (5)
where
Co - concentration of the polymer, mol cm'3 (6.08 x 10-8)
Do - diffusion coefficient, cm2 sec"1 ( 7 )
A - area of the electrode, cm2 (.018)
n - number of electrons (25)
F - Faraday's constant, C mol'1 (95435)
t - time of the experiment, 5 (2)
Qd - charge, 0 (2.1 x 10")

A diffusion coefficient of 2.47 x 10'7 cm2 sec'1 was calculated from
the experimental data. The net charge was obtained by subtracting
the charge obtained from a blank solution, from the charge obtained
from the polymer solution.

The diffusion coefficient of bis(trimetylsilyloxy)(tetra-t-
butylphthalocyaninato)silicon, the capped monomer [(t-
BuhPcSiOZ)(Si(CH3)3)2], was also determined by the chronocoulometry
technique. The potential utilized for the experiment was obtained
after consulting the CV response of the capped monomer. The initial
starting potential of the experiment, was a potential where no redox
response was observed in the CV. The final potential of the

chronocoulometry experiment was at a potential positive of the

94

potential required for the first one electron oxidation. The same
equation used to calculate the diffusion coefficient of the polymer
is used for the capped monomer. Typical values for a

chronocoulometry experiment of the capped monomer are listed below.

Qd - 2 n F A Co 001/2 c1/2 '1/2 (6)

C - concentration of

o [(t-BuaPcSiOZ)(Si(CH3)3)2], mol cm'3 (1.25 x 10-6)
Do - diffusion coefficient, cm2 sec'1 ( ? )
A - area of the electrode, cm2 (.018)
n — number of electrons (1)
F - Faraday's constant, C mol"1 (96485)
t — time of the experiment, 5 (2)
0d - charge, 0 (5.97 x 10'5)

A diffusion coefficient of 2.97 X 10'6 cm2 sec.1 was calculated from
the experimental data. This calculated diffusion coefficient is very
similar to diffusion coefficients of unsubstituted phthalocyanines
reported in the literature.19c

The relationship between molecular weight and diffusion
coefficient for a monomer and the corresponding polymer has been
established from both an experimental and a theoretical

viewpoint26c’28.

The diffusion coefficients of the capped monomer
and polymer, along with the molecular weights of the capped monomer
and repeat unit of the polymer can be used to determine the degree of

polymerization. From previous electrochemical studies of the small

95

length silicon phthalocyanines oligomers (monomer thru tetramer), it
is known that significant interaction occurs between the

phthalocyanine rings.17

This suggests that the cofacial arrangement
is maintained in solution. Therefore, assuming that the silicon
polymer also maintains a cofacial, rod-like structure in solution,

the equation that relates diffusion coefficients to molecular weights

for the monomer and polymer is:29b

Dp/Dm - (Hm/up)°-31 (7)
where:
Dp - diffusion coefficient of the polymer 2.47 x 10'7 cm2 3‘1

Dm - diffusion coefficient of the monomer 2.97 x 10'6 cm2 s'1

Mm — molecular weight of the monomer - 943.4 g mol"1

Mp - molecular weight of the polymer - ?

The formula of the monomer is CS4H66N881302 resulting in a molecular

weight of 943.4 g mol'l. The repeat unit of the silicon polymer,

C48H48N88i0, gives a molecular weight of 781.1 g mol'l. Solving the

above equation for the molecular weight of the polymer, one obtains a

value of 2.038 x 104 grams mole‘l. By dividing the molecular weight

of the polymer by the molecular weight of the repeat unit, the degree
of polymerization can be calculated. The degree of polymerization
obtained is 26.1, this shows excellent agreement with the ‘value

reported in the literature and the one used in the digital
17b

simulations.

4 . Conclusions

Electrochemical methods can be utilized in the study of soluble
electrically conductive materials. The electrochemical response of a
soluble electrically conductive material is very different from that
of a common non-interacting redox couple. The interaction between
adjacent molecules causes widespread changes in the electrochemical
response of the electrically conductive material. The oxidation of
the silicon polymer occurs over a 1700 millivolt range as seen in the
RDE, CPC, and DPV results. A non-kinetically limited conventional
redox couple or polymer comprised of non-interacting centers

typically has a several hundred millivolt response.26c

The typical
response of a conventional redox couple, for example the capped
monomer, is not the same as observed for a conductive material.
Conductive materials have a wide range of energetically closely
spaced redox states from which charge can either be added or removed.

The partial oxidation for the soluble silicon polymer can be
accurately and systematically varied to obtain any value from O to
100%, by simply changing the potential of the working electrode. The
(t-Bul‘l’cSiO)n polymer is chemically stable at all degrees of partial
oxidation, occurs over a 1700 mV range, and undergoes a simple charge
transfer reaction at the electrode. The smooth doping profile along
with this broad oxidation process, is indicative of a material with
interactive redox sites. The polymer is also stable to repetitive

oxidation/re-reduction cycles. The polymer can be cycled to the 1001

level and back to the 0% level (neutral polymer), the polymer can

96

97

then be oxidized to any previous percent oxidation with no loss or
gain in polymer redox activity. Repeat measurements at the same
potential provide the same percent oxidation vs. potential values
obtained initially. The solubility of the silicon polymer in common
non-aqueous solvents, permits the doping process to be unaffected by
structural changes which occur in the insoluble unsubstituted

species.15

The small difference in charge between the oxidation and
subsequent re-reduction for the CPC results suggests, that there is
no inherent stability at any particular oxidation state. The RDE,
CPC, and digital simulation (based on an experimentally obtained DPV)
experiments all provide remarkably similar results. As seen in
Figure 25, the three fundamentally different techniques provide
essentially the same percent oxidation vs. potential curve. The
excellent agreement between these three techniques, suggests that the
potentials for specific levels-of partial oxidation are accurately
known. These experiments have effectively mapped out the
thermodynamic potentials for the oxidation of the polymer.

From an analysis of the rotating disk voltammograms, the
oxidation of the silicon polymer was determined to be limited by the
mass transport of the material to the electrode and not to a kinetic
limitation. This was determined from a Levich plot, where the
current was plotted as a function of the square root of angular

velocity. The RDE voltammograms also provide information similar to

that obtained by CPC results. The CPC results took approximately 24

98

Figure 25. Degree of oxidation vs. potential for the silicon
polymer, (t-BuaPcSiO)n, from the CPC, RDE, digital simulation
results.

 

 

99

mum—~59...—
fiomm .9, «=60 2:228

 

 

 

   

  

8J8 FWD 8.. Pm... 88 o
-o.

-8

-8

A...

-8

r8

6.8 08 * .8

28 8.. o -8
co:o_:E_m I [00
w -8.

Lmezom COU:_M 0...: 1.0% _O_.—C0*Om .0) COZOUCAOR

 

 

NOIlVClIXO%

100

hours to obtain a percent oxidation vs. potential profile of the
polymer, while the RDE results were obtained in less than 1 hours.

The differential pulse voltammogram of the silicon polymer is
very broad and unlike most voltammograms of normal redox couples, it
is not very smooth. It is obvious that the redox states are too
close together to resolve with this technique. A digital simulation
utilizing the finite differences method, and oxidizing the polymer
based on a sequential electron transfer model, provides information
on the redox potential and distribution of states in the polymer.
The simulation also provides information on the literature reported
degree of polymerization, because it requires 25 sites to accurately
model both the current as well as the potential response of the
experimental DPV. The degree of polymerization calculated from the
diffusion coefficient data also supports the degree of polymerization
used for the digital simulations.

Potential applications of these materials will depend on the
ability to shuttle charge in and out of the polymer. The
determination of these rates of electron transfer over the entire
range of band-filling could have a large effect on their uses in a
number of fields. Preliminary results for reductive doping of the
silicon phthalocyanine polymer have also been obtained. The
reductive doping appears to be similar in most respects to the
oxidative data. The percent reduction-potential curve occurs over a
wide potential range very similar to that for the percent oxidation-
potential curve. With both reduction and oxidation possible, a wide
range of oxidation states is possible, and will provide numerous

opportunities to study the physical and chemical properties over a

101

unprecedented range of band-filling. A great deal of information can
be obtained on this particular system. as a function of polymer
oxidation or reduction. One example is the current dip at 1.00 V in
the DPV voltammogram, this may be due to a reoganization in the
polymer’s redox states. Bulk samples can be prepared at specific
levels of partial oxidation (or reduction), and spectroscopic
measurements can be performed, to characterize the physical
properties of the polymer at that degree of partial oxidation or
reduction. Bulk samples can be characterized by a number of
techniques, such as FTIR, conductivity, and EPR. These methods will
provide information concerning the properties of the silicon polymer

as a function of the degree of partial oxidation or reduction.

APPENDIX

 

APPENDIX A

The model utilized for the digital simulation of the DPV and CV
current-potential responses is based on. the assumption. that each
phthalocyanine ring is a redox active site, with its own unique
standard potential (E01). The oxidation of the polymer can be viewed

as a system undergoing sequential electron transfer

Rn fie Rn-101 + e' ' E01 A.l

Rn-101 4.- R“.202 + e' E°2 4.2

Ru-202 :- R16.303 + 8‘ E°3 A.3
etc.

where R - reduced species, 0 - oxidized species and n - the length of
the polymer. The polymer was taken to be 25 redox states (reference
1a, Vol. 1, chapter 5). The fractional concentration of any species
at a given potential can be calculated from the potential dependant
equilibrium constant (Kj) for that particular electron transfer
event.

The fractional concentration of any species, an - Rn/CT (where CT is
the concentration of the total number of redox active sites) is

calculated from:

no - ------------------------------------------- A.4

102

103

Each individual equilibrium constant K1 is calculated from the Nernst
equation: Kj - exp[(E-E°j)RT/njF]. This model is exactly analogous
to that used for the dissociation of a polyfunctional acid or base.
This model works for both the oxidation as well as for the reduction

of a multi-redox site polymer.

REFERENCES

(a) Skotheim, T.A., Ed. ”Handbook of Conducting Polymers”

(Dekker, New York 1986) Vols . 1-2 . (b) Ferraro , J . R. ;
Williams , J .M. , "Introduction to Synthetic Electrical
Conductors" , (Academic Press: New York, 1987) . (c) Becker ,

J.Y.; Bernstein, J.; Bittner, S., Eds., Israel 1. Chem., 1986,
271 2. (d) Skotheim, T. A., Ed. ”Electroresponsive molecular
and polymeric systems” (Dekker, New York, 1988), vol. 1” (e)
Proceedings of the International Conference on Science and
Technology of Synthetic Metals (ICSM '86), mg}. Met” 1987,
11. (f) Jerome, D.; Caron, L.G., Eds. "Low-Dimensional
Conductors and Superconductors", Plenum: New York, 1987. (g)
Proceedings of the International Conference on Science and
Technology of Synthetic Metals (ICSM '88), m. ML, 1988-
1989, 27-29 (h) A.J. Epstein, E.M. Conwell, Eds. ”Proceeding of
the International Conference on. Low-Dimensional Conductors";
Boulder, Colorado, August 9-14, 1981. flgl. grygg. Lig. Crygt.
77, 79, 83, 85, 86 Parts A-F (1981-1982). (1) Hanack, M.;
Mitulla, K.; Pawlowski, C.; Subramanian, L. R. J. grggggmgtgllig
Chem., 1981, 294 315-325.

(a) Chiang, C. K.; Fincher, C. R. Jr.; Park, Y. W.; Heeger, A.
J.; Shirakawa, M.; Louis, E. J.; Gau, S. C.; MacDiarmid, A. G.
Phys. Reg. Lg£§., 1977, 32, 1098. (b) Chiang, C. K.; Druy, M.
A.; Gau, S. C.; Heeger, A. J.; Louis, E. J.; MacDiarmid, A. C.;
Park, Y. W. J. Am. Chem. §g§., 1978, 190, 1013. (c)
Ehrenfreund, E.; Rybaczewski, E. F.; Garito, A. F.; Meeger, A.
J. Phys. Rev. Lg;;., 1972, 28, 873. (d) Ferraris, J.; Cowan, D.
0.; Walatka, V.; Perlstein, J. H. J. Am. Chem. £29., 1973, 25,
948.

(a) Duke, C. B. m. net” 1987, 2],, 5-12. (b) Inganas, 0.;
Lundstrom, I. 3181211- m., 1987, 2],, 13-19. (c) MacDiarmid, A.
C. m. 118.9. 1987, 21, 79-83. (d) Akhtar, M.; Weakliem, H.
A.; Paiste, R. M.; Gaughan, K. m. Met” 1988, 26, 203-208.

(a) Cowan, D. 0.; Wiygul, F. M. Chem. Eng. News, 1986, 64(29),
28-45. (b) Ward, M.W. "Electroanalytical Chemistry", Bard,
A.J., Ed. (Dekker, New York, 1989) Vol. 16, 181-312. (c) Bryce,

M.; Murphy, L. Nature, 1984, 302, 109. (d) Wudl,.F. 59;. Chem

104

10.

11.

105

Reg... 1984, 12(6), 227-232. (E) Williams, J.M. "Progress in
Inorganic Chemistry”, Lippard, S. J ., Ed. (Wiley, New York,
1985), Vol 33, 183-220.

(a) Scrosati, B. 2mg. Selig Sgate ghee” 1988, 18(1), 1-77.
(b) Wegner, G. Angew. Q)”. 1113;. ESL Eng, 1981, 29(4), 361-381.
(c) Reynolds, J.R. (Ebeneeeh, 1988, 18(2), 440-447. (d) Miller,
J .S., Ed. ”Extended Linear Chain Compounds” (Plenum Press, New
York, 1982), Vols. 1-3.

(a) Greene, R.L.; Street, G.B. fielenee, 1984, 22g, 651-656.

(a) Hoffmann, R. m. 911m. 13;. new Eng” 1987, 2.6.. 846.
(b) Miller, J. 8.; Epstein, A. J. "Progress in Inorganic
Chemistry", Lippard, S. J., Ed. (Wiley, New York, 1985), Vol.
20, 1-151. (c) Duke, B. J.; O'Leary, B. ,1. Chem. fie” 1988,
5.211;).- 319-321. (d) Duke, B. J.; O'Leary, B. ,1. m. E...
1988, 65(5), 379-383.

(a) Marks, T.J. Science, 1985, 221, 881-889.

(a) Diaz, A. F.; Lacroix, J. C., Neg ,1. (3153., 1988, 12(4), 171-
180. (b) Wudl, F. 1mg. Qhem” 1986, 22(4), 289-292.

(a) Diaz, A. F.; Logan, A. J. ,1. Eleetroanal. glee” 1980, 111,
111-114. (b) Diaz, A. F.; Clark, T. C. ,1. ElectroanaL. ghee”
1980, 111, 115-117. (c) Diaz, A. F.; Castillo, J. I.; Logan, J.

A.; Lee, W.-Y. ,1. W. my” 1981, 122, 115-132. (d)
Diaz, A. F.; Martinez, A.; Kanazawa, K. K. g. e as

ghee” 1981, 110, 181-187. (e) Bull, R. A.; Fan, F.-R. F.;
Bard, A. J. ,1. fleeemhem mu 1982, 122, 1009-1015. (f)
Diaz, A. J.; Castillo, J.; Kanazawa, K. K.; Logan, J. A. ,1.
W. Chen” 1982, 111, 233-239. (g) Tourillon, C.;
Garnier, F. ,1. W. Qhem., 1982, 135, 173-178. (h)
Genies, E. M.; Bidan, C.; Diaz, A. F. ,1. W. ghee”
1983, 152, 101-113. (1) Waltman, R. J.; Bargon, J.; Diaz, A.
J. ,1. mg. m” 1983, 81, 1459-1463.

(a) Feldberg, S. W. ,1. Am. ghee. See” 1984, mg, 4671-4674,
and references therein. (b) Murray, R. W., "Electroanalytical
Chemistry", Bard, A. J., Ed. (Dekker, New York, 1984) vol 13.

12.

13.

14.

15.

106

(a) Miller, J. 8.; Epstein, A. J. m. Qhen. 1mg. 31., Eng.,
1987, 21(3), 287-293. (b) Conwell, E.M.; Howard, I. A. 51111213-
Net” 1986, l_3_, 71-85. (c) Epstein, A. J.; Kaufer, J. W.;
Rommelmann, H.; Howard, 1. A.; Conwell, E.M.; Miller, J. S.;
Pouget, J. P.; Comes, R. flue. Rex Leggy, 1982, 52(14), 1037-
1041. (d) Conwell, E.M.; Howard, 1. A. Nel. 9:11;. 1.19. $1155.,
1985, l2_Q, 51-58. (e) Epstein, A. J.; Bigelow, R. W.; Miller,
J. S.; McCall, R. P.; Tanner, D. B. Nel. m. 143. meg,
1985, 129, 43-49. (f) Mazumdar, 8.; Dixit, S. N.; Bloch, A. N.
Nel. m. ng. 91m” 1985, 129, 35-42.

(a) Diel, B. D.; Inabe, T.; Lyding, J. W.; Schoch, Jr, K. F.;
Kannewurf, c. R.; Marks, '1'. J. ,1. m. m. 53,-,” 1983, 192,
1551-1567. (b) Petersen, J. L.; Schranml, C. 8.; Stojakovic, D.
R.; Hoffman, B. M.; Marks, T. J. ,1. M. Chem. 535., 1977, 22,
286-288. (c) Schramm, C. 8.; Scaringe, R. P.; Stojakovic, D.
R.; Hoffman, B. M.; Ibers, J. A.; Marks, T. J. ,1. Am. Chem.
See, 1980, 192, 6702-6713. (d) Die1,B. D.; Inabe, T.; Jaggi,
N. K.; Lyding, J. W.; Schneider, 0.; Hanack, M.; Kannewurf, C.
R.; Marks, T. J.; Schwartz, L. H. ,L,_ Am. Chem. 595., 1984, l9_6_,
3207-3214. (e) Inabe, T.; Gaudiello, J. C.; Moguel, M. K.;
Lyding, J. W.; Burton, R. L.; McCarthy, W. J.; Kannewurf, C. R.;
Marks, T. J. ,1. Am. ghee. See, 1986, 198, 7595-7608. (f)
Marks, T. J.; Schoch, K. F. Jr.; Kundalkar, B. R. m. Mega,
1979-1980, l, 337. (g) Brant, F.; Nohr, R. 8.; Wynne, K. J.;
Weber, D. C. Mel. Cryst. ng. Cryst., 1982, §_l, 255. (h) Dirk,
C. W.; Inabe, T.; Schoch, Jr, K. F.; Marks, T. J. ,1. M. M.
5353., 1983, 115.. 1539-1550. (1) Nohr, R. 8.; Kuznesof, P. M.;
Wynne, K. J.; Kenney, M. E.; Siebenmann, P. G. ,1. Amer. Chem.
See” 1981, 191, 4371-4377.

(a) Ofer, D.; Wrighton, M. S. ,1. Am. ghee). See” 1988, llQ,
4467-4468.

(a) Gaudiello, J. C.; Marcy, H. 0.; McCarthy, W. J.; Moguel, M.
K.; Kannewurf, C. R.; Marks, T. J. m. Ne_t., 1986, l1, 115-
128. (b) Gaudiello, J. C.; Almeida, M.; Marks, T. J.; McCarthy,
W. J.; Butler, J. C.; Kannewurf, C. R. ,1. me. Qhem., 1986, 2Q,
4917-4920. (c) Almeida, M.; Gaudiello, J. 6.; Butler, J. C.;
Marcy, H. 0.; Kannewurf, C. R.; Marks, T. J. m. Net., 1988,
21, 261-266. (d) Kellogg, G. E.; Gaudiello, J. C.; Schlueter,
J. A.; Tetrick, S. M.; Marks, T. J.; Marcy, H. 0.; McCarthy, W.

16.

17.

18.

19.

20.

107

J.; Kannewurf, c. R. m. 8232.. 1989, 22, PIS-F24. (e)
Gaudiello, J. C.; Kellogg, G. E.; Tetrick, S. M.; Marks, T. J.
,1. An. m. 539., 1989, 11],, 5259-5271. (f) Almeida, M.;
Gaudiello, J. C.; Kellogg, G. E.; Tetrick, 8. M.; Marcy, H. 0.;
McCarthy, W. J.; Butler, J. C.; Kannewurf, C. R.; Marks, T. J.
,1. Au. thg. 895... 1989, 111, 5271-5284.

(a) Hanack, M. ; Lange, A. ; Rein, M. ; Behnisch, R. ; Renz, C.;
Leverenz, A. m- MAL, 1989, 22, Fl-F8. (b) Hanack, M.
Israel ,L Chem” 1985, 25, 205-209.

(a) Metz, J.; Pawlowowski, C.; Hanack, M. Z- flaturfiorsgh., 1983,
m, 378-382. (b) Hanack, M.; Datz, A.; Fay, R.; Fishcher,
K.; Keppeler, H.; Koch, J.; Metz, J.; Mezger, M.; Schneider, 0.;
Schulze, H., "Handbook of Conducting Polymers”, Skotheim, T. A.,
Ed. (Dekker, New York, 1986) vol. 1, and references therein.
(c) Hanack, M.; Metz, J.; Pawlowski, G. Chem. 331., 1982, 112,
2836-2853.

(a) Hanack, M.; Leverenz, A. m. u_e_§., 1987, 22, 9-14.

(a) Simic-Glavaski, B.; Tanaka, A. A.; Kenney, M. E.; Yeager, E.
,1. Elegggoanal. Chem” 1987, 222, 285-296. (b) Mezza, T. M.;
Armstrong, N. R.; Ritter, G. W.; Iafalice, J. P.; Kenney, M. E.
,1. Electroanal. Chem” 1982, 117,, 227. (c) Wheeler, B. L.;
Nagasubramanian, C.; Bard, A. J.; Schechtman, L. A.; Dininny, D.
R.; Kenney, M. E. ,1. Amer. M. 895., 1984, 126, 7404. (d)
DeWulf, D. W.; Leland, J. K.; Wheeler, B. L.; Bard, A. J.;
Batzel, D. A.; Dininny, D. R.; Kenney, M. E. 1391;. Chem” 1987,
26, 266. (e) Ciliberto, E.; Doris, K. A.; Pietro, W. J.;
Reisner, G. M.; Ellis, D. E.; Fragala, 1.; Herbstein, F. H.;

Ratner, M. A.; Marks, T. J. ,1. Amen. m. §_Q§., 1984, 1516,
7748.

(a) Elsenbaumer, R. L.; Jen, K. Y.; Ododi, R. m. 1113;” 1986,
15,, 169-174. (b) Patil, A. 0.; Ikenoue, Y.; Wudl, F.; Heeger,
A. J. ,1 Am. Chem. Soc... 1987, 102, 1858-1859. (c) Patil, A.
0.; Ikenoue, Y.; Wudl, F.; Heeger, A. J. m net“ 1987, 20,
151, and references therein. (d) Patil, A. 0.; Ikenove, Y.;
Basecu, N.; Colaneri, H.; Chen, J.; Wudl, P.; Heeger, A. J.
mm. 1195., 1987, 29, 151-159. (e) Rughooputh, S. D. D. V.;
Nowak, M.; Hotta, 8.; Heeger, A. J.; Wudl, F. 531511. 1195., 1987,

21.

22.

23.

24.

25.

26.

27.

28.

29.

108

21, 41-50. (f) Sato, M-A.; Tanaka, 8.; Kaeriyama. K- $111511-
m., 1987, 18, 229-232. (g) Hotta, 8.; Rughooputh, S. D. D.
V.; Heeger, A. J.; Wudl, F. W, 1987, 29, 212-215.

(a) Schriver, D. E. ”The Manipulation of Air-Sensitive
Compounds”; (McGraw-Hill, New York, N.Y. 1969).

(a) Smith, W. H.; Bard, A. J. ,1. Ag. Chem. ,Sg_., 1975, 22, 5203-
5210.

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

(a) Bard, A. J.; Faulkner, L. R., ”Electrochemical Methods"
(Wiley, New York, 1980). (b) Rieger, P. H., "Electrochemistry"
(Prentice-Hall, Englewood Cliffs, N. J. 1987). (c) Kissenger, P.
T.; Heineman, W. R., Eds. ”Laboratory Techniques in
Electroanalytical Chemistry”, (Marcel Dekker, New York, 1984).
(d) Harrar, J. E. "Electroanalytical Chemistry", Bard, A.J., Ed.
(Dekker, New York, 1975) Vol. 8, 1-167.

(a) Couves, J. W.; Wright, J. D. Smth. Met., 1989, 22, F77-F82.

(a) Brown, C. H.; Meyer, T. J.; Cowan, D. 0.; LeVanda, C.;
Kaufman, F.; Roling, P. V.; Rausch, M. D. Inorg. gh_e_m., 1975,
15;, 506-511. (b) Brennam, D. E.; Geiger, W. E. ,1. M- M.
399, 1979, 101, 3399-3400. (c) Flanagan, J. B.; Margel, 8.;
Bard, A. J.; Anson, F. C. ,1. Am. Chem. Soc” 1979, 199, 4248-
4253.

(a) Su, W. P.; Schrieffer, J. R.; Heeger, A. J. M. m.
ngg., 1979, 32, 1698. (b) Rice, M. J. 21113-1. Lgtt., 1979, A-_Z_1,

152. (c) Schrieffer, J. R. M91. C_r1s_t. .L_i_q. Crys_t., 1981, 11,
201.

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

(a) Feldberg, 8. W. "Electroanalytical Chemistry"; Bard, A. J.
Ed.,(Marce1 Dekker, New York, 1964) Vol 3, 271. (b) Prater, K.
B. , "Computers in Chemistry and Instrumentation"; Mattson, J.
8.; Mark, H. B.; MacDonald, H. C. Eds., (Marcel Dekker, New

109

York, 1972) Vol 2. (c) Malloy, J. T. ”Laboratory Techniques in
Electroanalytical Chemistry", Kissenger, P. T.; Heineman, W. R.
Eds.,(Marcel Dekker, New York, 1984). (d) Britz, D. "Digital
Simulations in Electrochemistry, Lecture notes in Chemistry"
(Springer-Verlag, Heidleberg, 1981).

I!

HICH

1111111111“