Synthesis, computational and electrochemical characterization of a family of functionalized dimercaptothiophenes for potential use as high-energy cathode materials for lithium/lithium-ion batteries

Article abstract of DOI:10.1039/b707235j

We present a family of a novel class of organosulfur compounds based on dimercaptothiophene and its derivatives, with a variety of functional groups (electron-donating or electron-withdrawing groups) and regiochemistries, designed as potential high-energy cathode materials with sufficient charge/discharge cyclability for lithium/lithium-ion rechargeable batteries. This study uses as a point of departure the electrochemical and computational understanding of the electrocatalytic effect of poly(3,4-ethylenedioxythiophene) (PEDOT) towards the redox reactions of 2,5-dimercapto-1,3,4-thiadiazole (DMcT). The effective redox potentials of these materials exhibited good correlation with the highest-occupied molecular orbital (HOMO) levels predicted via computational modeling. Furthermore, the redox reactions of all the compounds studied were electrocatalytically accelerated at PEDOT film-coated glassy carbon electrodes (GCEs), although some materials exhibited higher energy output than others. By using this approach we have identified several compounds that exhibit clear promise as potential cathode materials and have characterized the molecular interactions between the organosulfur compounds and PEDOT film surfaces involved in the electrocatalytic reactions. The Royal Society of Chemistry 2007.

Full text of DOI:10.1039/b707235j

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Synthesis, computational and electrochemical characterization of a family
of functionalized dimercaptothiophenes for potential use as high-energy
cathode materials for lithium/lithium-ion batteries{{
ab
Yasuyuki Kiya, Jay C. Henderson, Geoffrey R. Hutchison and H e´ ctor D. Abru n˜ a*
a
a
a
Received 14th May 2007, Accepted 3rd August 2007
First published as an Advance Article on the web 28th August 2007
DOI: 10.1039/b707235j
We present a family of a novel class of organosulfur compounds based on dimercaptothiophene
and its derivatives, with a variety of functional groups (electron-donating or electron-withdrawing
groups) and regiochemistries, designed as potential high-energy cathode materials with sufficient
charge/discharge cyclability for lithium/lithium-ion rechargeable batteries. This study uses as a
point of departure the electrochemical and computational understanding of the electrocatalytic
effect of poly(3,4-ethylenedioxythiophene) (PEDOT) towards the redox reactions of
2
,5-dimercapto-1,3,4-thiadiazole (DMcT). The effective redox potentials of these materials
exhibited good correlation with the highest-occupied molecular orbital (HOMO) levels predicted
via computational modeling. Furthermore, the redox reactions of all the compounds studied were
electrocatalytically accelerated at PEDOT film-coated glassy carbon electrodes (GCEs), although
some materials exhibited higher energy output than others. By using this approach we have
identified several compounds that exhibit clear promise as potential cathode materials and have
characterized the molecular interactions between the organosulfur compounds and PEDOT film
surfaces involved in the electrocatalytic reactions.
properties of interest. In addition, the capability of this redox
system to capture lithium ions during discharge (due to forma-
tion of the thiolate, which subsequently traps the lithium ions)
allows their easy incorporation into the so-called ‘‘rocking-
chair’’-type system employed in commercial lithium-ion
rechargeable batteries. However, the charge transfer kinetics
of both oxidation and reduction of these materials tend to be
Introduction
For over fifteen years, the investigation of sulfur-based
polymer cathodes for lithium/lithium-ion batteries, with higher
energy density than conventional lithium metal oxide-based
cathodes, has garnered much scientific and technological
1
–18
interest.
A key electrochemical reaction in the field has
2
been the redox chemistry of thiolates (RS ), which can be
too sluggish at room temperature for use in viable lithium/
19,20
?
oxidized to give the corresponding radical (RS ) which can,
lithium-ion batteries.
In order to accelerate these redox
in turn, couple to form disulfides (RSSR). For instance,
reactions, our group has previously investigated the use of
conducting polymers as electrocatalysts and shown that the
redox reactions of DMcT can be dramatically accelerated by
the conducting polymer poly(3,4-ethylenedioxythiophene)
2
,5-dimercapto-1,3,4-thiadiazole (DMcT), which is one of the
most promising organosulfur compounds due to its high
2
1
theoretical capacity (362 A h kg ), forms a disulfide polymer
by coupling reactions of the electrochemically-generated
radical species (Scheme 1).
1
5
5
(
PEDOT) as well as by polyaniline (PANI). However,
composite cathodes, composed of a thiolate compound such as
DMcT and a conducting polymer, often exhibit poor charge/
discharge cyclability due to dissolution of the reduction
products (typically monomer) of the disulfide polymer when
Organic materials also offer the advantage of being
relatively low cost and derived from abundant resources, as
opposed to conventional lithium metal oxides. Moreover,
chemical tunability of organic compounds has made them even
more attractive. That is, organic materials can be modified
2
1,22
an organic liquid electrolyte is employed.
Therefore, in
order for organosulfur compounds to be of practical use (i.e.,
repeated charge/discharge cycles) as high-energy cathode
materials, procedures and/or materials capable of preventing
such dissolution must be developed.
(
designed) to give additional chemical and/or electrochemical
a
Department of Chemistry and Chemical Biology, Baker Laboratory,
Cornell University, Ithaca, New York 14853-1301, USA
In this study, with the aim of designing materials to improve
the charge/discharge cyclability of organosulfur compounds, a
b
Subaru Research and Development Inc., Ann Arbor, Michigan 48108,
USA. E-mail: hda1@cornell.edu
{ The HTML version of this article has been enhanced with colour
images.
{
Electronic supplementary information (ESI) available:
Characterization and synthetic details for the TBT family studied;
CVs obtained from electrochemical characterization of the TBT family
studied; profile for a PEDOT film cast on ITO electrodes; plots of
PEDOT film thickness as a function of charge consumed during the
electrochemical polymerization of EDOT. See DOI: 10.1039/b707235j
Scheme 1 Redox reaction scheme for 2,5-dimercapto-1,3,4-thiadia-
zole (DMcT).
4
366 | J. Mater. Chem., 2007, 17, 4366–4376
This journal is ß The Royal Society of Chemistry 2007

View Article Online
novel class of organosulfur compounds based upon dimercap-
2
behavior of the TBT family, the effect of substituents on the
redox behavior of TBT, and the electrocatalytic effect of
PEDOT towards the TBT family are discussed in detail.
Moreover, chemical interactions between the TBT family and
PEDOT, and the effects on the electrocatalytic activity of
PEDOT, are also discussed. It was observed that, while the redox
reactions of all the compounds were electrocatalytically accel-
erated when oxidized or reduced at a PEDOT film-modified
electrode, as predicted by computational modeling, the electro-
catalytic activity of PEDOT differed substantially depending
on the specific derivative. Furthermore, several compounds
exhibited clear promise as potential energy-storage sites for
polymer cathodes for lithium/lithium-ion rechargeable batteries.
3
tothiophenes
(thiophene-2,5-bis(thiolate), TBT), with a
variety of functional groups (electron-donating or electron-
withdrawing groups) and regiochemistries has been synthe-
sized, and the relationships between the structures and
electrochemical activity, as well as the effects of the structures
on the electrocatalytic activity of PEDOT towards the redox
reactions of these materials, have been investigated. As shown
in Fig. 1, these compounds are synthetically tunable, unlike
DMcT, and thus offer the promise of covalent bonding either
to themselves or to a conducting polymer backbone in order to
prevent the materials from leaching into an electrolyte solution
during charge/discharge cycles. Moreover, TBT and its
derivatives were chosen based on the electrochemical and
computational understanding of the thermodynamics of the
electrocatalytic effects exhibited by the DMcT/PEDOT sys-
Experimental
1
5,21,22
tem.
Therefore, the electrocatalytic activity of PEDOT
Dimercaptothiophene and a number of derivatives were
synthesized (Fig. 1). Characterization and synthetic details
of the dimercaptothiophene family are available in the
Supporting Information{. 2,5-Dimercapto-1,3,4-thiadiazole
dilithium salt (DMcT-2Li) was purchased from Toyo Kasei
Co. (Japan) and used without further purification. 3,4-
Ethylenedioxythiophene (EDOT) was obtained from Bayer
Co. (Germany) and used as received. High-purity HPLC-grade
towards their redox reactions would be anticipated. From the
above-mentioned strategies, this study enabled both the
investigation of a class of synthetically-tunable organosulfur
compounds, the redox reactions of which should be electro-
catalytically accelerated by PEDOT, and further elucidation of
the kinetic, thermodynamic and electrocatalytic aspects of
reactions between organosulfur compounds and conducting
polymers, and thus could lead to future materials that could
realize high capacity (and energy) with sufficient charge/
discharge cyclability.
acetonitrile (AN) was purchased from Burdick and Jackson,
and dried over 3 A˚ molecular sieves. Lithium perchlorate
(LiClO ) (99.99%) and ammonium hydroxide (NH OH)
4
4
The redox behavior of the TBT family and the electro-
catalytic activity of PEDOT towards the compounds have
been characterized in an AN solution via cyclic voltammetry
(99.99+%) were purchased from Aldrich Chemical Co. Inc.,
and used as received.
Cyclic voltammetry (CV), rotating-disk electrode (RDE)
voltammetry, and double potential-step chronoamperometry
(DPSCA) studies were carried out at room temperature using
a potentiostat (Hokuto Denko Co., model HSV-100 and
HABF1510m) with an analog X–Y recorder (GRAPHTEC
Co., model WX4000). In RDE voltammetry studies, linear
(
CV), rotating-disk electrode (RDE) voltammetry, and double
potential-step chronoamperometry (DPSCA). The redox
2
1
sweeps were carried out at 5 mV s at different rotation rates
using a Pine Instrument Co., model AFMSRX rotator.
The experimental voltametric profiles were compared to those
1
obtained via simulation using DigiSim
version 3.0
(
Bioanalytical Systems, Inc. (BAS)). In DPSCA studies, the
first potential step was carried out from 21.50 V to +0.80 V
and the potential was held at +0.80 V for 60 s. The second
potential step was carried out from +0.80 V to 21.50 V and
the potential was held at 21.50 V for 60 s. Measurements were
taken in a three-electrode cell configuration using glassy
carbon electrodes (GCEs) (BAS, 3.0 mm diameter for CV and
DPSCA, and Pine Instrument Co., 5.0 mm diameter for RDE
Fig. 1 Organosulfur compounds synthesized and studied, including a
variety of electron-donating and electron-withdrawing substituents to
the basic thiophene-2,5-bis(thiolate) (TBT) 1 structure: thiophene-3,4-
bis(thiolate) (3,4-TBT) 2, 3-methylthiophene-2,5-bis(thiolate) (methyl-
TBT) 3, 3,4-dimethylthiophene-2,5-bis(thiolate) (3,4-dimethyl-TBT)
voltammetry) as working electrodes, a large area Pt coil
+
counter electrode, and a Ag/Ag (0.05 M AgClO
4
+ 0.1 M
LiClO /AN) reference electrode without regard to the liquid
4
junction potential, and against which all potentials are
reported. The working electrode was polished with 0.3 mm
and 0.05 mm alumina slurries (REFINETEC Ltd), rinsed with
distilled water and acetone, and dried prior to use. Unless
otherwise noted, all experiments were carried out in a 0.1 M
4,
2,5-dimethylthiophene-3,4-bis(thiolate) (2,5-dimethyl-TBT) 5,
3-methoxythiophene-2,5-bis(thiolate) (methoxy-TBT) 6, 3-acetyl-
thiophene-2,5-bis(thiolate) (acetyl-TBT) 7, 3-bromothiophene-2,5-
bis(thiolate) (bromo-TBT) 8, thiazole-2,5-bis(thiolate) (ThiaBT) 9,
thiophene-2-thiolate (TT) 10, and 3-methoxythiophene-2-thiolate
LiClO /AN solution, which was thoroughly purged using pre-
4
purified nitrogen gas.
(methoxy-TT) 11. The acetyl-protected molecules were synthesized to
prevent disulfide formation, and the thiolates were generated in situ
PEDOT films were prepared on GCEs by anodic
electrochemical polymerization of EDOT monomer at a
before electrochemical measurements.
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 4366–4376 | 4367

View Article Online
concentration of 20 mM in a 0.1 M LiClO
4
/AN solution via
in this study to focus solely on the redox behavior of thiolate
species and compare the behavior with those of our new
dimercaptothiophene family shown in Fig. 1, which was
generated in situ from the acetyl-protected materials synthe-
2
potential cycling at 20 mV s over the potential range from
1
+ 24
0.60 and +0.90 V vs. Ag/Ag . After polymerization, the
2
films were thoroughly rinsed with a 0.1 M LiClO /AN solution
4
2
5
and subsequently used for the characterization of the electro-
catalytic activity towards the redox reactions of DMcT-2Li
and the dimercaptothiophenes.
sized. Fig. 2b shows a representative CV (obtained at the
fifth cycle) for 1 mM DMcT-2Li at a bare GCE in a 0.1 M
4
LiClO /AN solution. Oxidations of the thiolates to generate
Deprotection of the thioacetates to generate the correspond-
2
the radical species, resulting in the formation of the corres-
ponding dimers and oligomers (polymers) via the coupling
reactions, were observed over the potential range from 20.60 V
to +0.80 V with current peaks at 20.39 V and 20.27 V, and
with a corresponding reduction current peak at 20.80 V vs.
5
ing thiolates was carried out with NH
4
OH, producing a
combined 25 mL AN solution composed of 50 mM NH OH,
4
1
4
.0 mM of the organothiolate, and 0.1 M LiClO . In order to
complete the deprotection, the solution was left for 3 hours (as
discussed below) under an inert atmosphere, followed by
immediate characterization of the redox behavior by CV, RDE
voltammetry, and DPSCA.
+
Ag/Ag . The very large peak-to-peak separation (.400 mV),
DE
cally irreversible. In addition, the following chemical step
coupling reactions of the radicals) might have an effect on the
p
, clearly indicates that this redox process is electrochemi-
(
electrochemical irreversibility. The oxidation peak current, i ap ,
was proportional to the square root of the scan rate, indicating
that this is a diffusion-limited process. On the other hand, the
Computational methods
All calculations were performed using the Gaussian 03
2
program. All compounds were geometry-optimized using
6
c
fact that the reduction peak current, i , was directly propor-
p
density functional theory (DFT) with the B3LYP hybrid
7,28
tional to the scan rate indicated that the reduction process
involved the reductive stripping of a DMcT oligomer film
that was generated on the GCE surface during the anodic
2
functional
and the 6-31G* basis set, followed by B3LYP
single-point calculations using the 6-31++G** basis set for
greater accuracy in treating anionic compounds. For com-
parison, single-point calculations were also performed using
B3LYP/6-31++G** using a conductor polarizable continuum
21
potential sweep. Furthermore, as anticipated and previously
21,45
reported,
it was observed that the oxidation of DMcT-2Li
(
dithiolate form) was thermodynamically more favorable than
2
9,30
model (C-PCM) for an AN solution.
While HOMO and
that of its dithiol form, DMcT-2H.
LUMO eigenvalues from density functional methods cannot
be formally taken as either the ionization potential or electron
Fig. 2a shows a CV for a PEDOT film-coated GCE in a
0
.1 M LiClO
PEDOT film was obtained by oxidative electropolymerization
of EDOT in a 0.1 M LiClO /AN solution as described in the
4
/AN solution in the absence of DMcT-2Li. The
3
1–34
affinity, respectively,
previous studies have shown that
B3LYP-derived eigenvalues compare favorably with experi-
3
4
5–40
35
ionization potentials, and
mental electron affinities,
4
Experimental section. Over this potential region, p-type
1,42
band gaps.
The differences between the computed
doping and dedoping processes were observed as previously
24,47–49
HOMO/LUMO values and electrochemical formal potentials
are derived from the difference in environments and previous
work has shown that solvent effects such as polarizability and
cavity radius can be used to linearly adjust (via a Kamlet–Taft
reported.
Oxidation of the PEDOT film (doping)
started at 21.08 V and a current peak was observed at
0.33 V followed by a plateau current response due mostly to
22
2
its capacitance. The corresponding reduction (dedoping)
4
3
relationship) computed ionization potentials and electron
4
affinities to compare favorably to electrochemical data.
current peaks were observed at 20.18 V and 20.84 V, respec-
50
tively. The oxidation peak current was directly proportional
4
Results and discussion
We begin with a general discussion of the electrocatalytic cycle
between DMcT and PEDOT using electrochemistry and
computational modeling of the electronic structure of the
species involved. The thermodynamics of the electrocatalytic
cycle suggest that some organodisulfide compounds could be
similarly electrocatalyzed by PEDOT. We discuss the synthesis
of DMcT analogues (i.e., TBT and its derivatives) suggested by
computational modeling, the redox behavior of these com-
pounds, and the electrocatalytic activity of PEDOT films
modified on GCE surfaces towards the redox reactions of the
compounds in solution.
I. Electrocatalytic effect of PEDOT towards the redox reactions
of DMcT
Fig. 2 (a) Representative CV for a PEDOT film-coated GCE in a
4
5,46
Since the redox behavior of thiol compounds at both bare
0
.1 M LiClO
4
/AN solution. Representative CVs for 1 mM DMcT-2Li
2
and PEDOT film-coated GCEs is strongly affected by the
1
at (b) bare and (c) PEDOT film-coated GCEs in 0.1 M LiClO /AN
4
21
solutions. The scan rate in all cases was 20 mV s
2
presence of protons (RSH or RS ), DMcT-2Li was employed
.
4
368 | J. Mater. Chem., 2007, 17, 4366–4376
This journal is ß The Royal Society of Chemistry 2007

View Article Online
to the scan rate, indicating that, as would be anticipated, the
redox reactions of PEDOT are surface processes and charge
propagation is facile.
compound with a HOMO level between the neutral PEDOT
HOMO level and cationic PEDOT singly-occupied molecular
orbital (SOMO) level should exhibit oxidative electrocatalysis
5
1
Fig. 2c shows a representative CV (obtained at the fifth
cycle) for a 1 mM DMcT-2Li solution at a PEDOT film-
by PEDOT (Fig. 3). Similarly, compounds with a LUMO
level approximating the PEDOT HOMO level should show
reductive electrocatalysis by PEDOT (Fig. 4). Since the
thiadiazole ring in DMcT makes synthetic variations impos-
sible, synthetically modifiable thiazole and thiophene analo-
gues with varying electron-donating and electron-withdrawing
substituents were studied in order to establish the applicable
range of the PEDOT electrocatalytic cycle to new aromatic
organosulfur compounds, as illustrated in Fig. 5. With the
exception of dimercaptothiazole 9, all of the DMcT analogues
exhibited very similar orbital energetics to DMcT and thus, the
redox reactions of the analogues were anticipated to be
electrocatalytically accelerated by PEDOT. As Fig. 5 illus-
trates, a polymer film of PEDOT is not one homogeneous
species with well-defined redox levels, but rather it consists of
multiple chain lengths, resulting in energy level broadening. In
addition, PEDOT consists both of neutral (unoxidized) and
cationic (oxidized) forms.
4
coated GCE in 0.1 M LiClO /AN. Two oxidation current
peaks corresponding to formation of the radical species of
DMcT-2Li, resulting in the formation of the corresponding
dimers and oligomers (polymers) via the coupling reactions,
were observed at 20.44 V and 20.28 V, respectively, and the
corresponding reduction peaks were observed at 20.63 V and
2
p
0.37 V, respectively. This dramatic decrease in DE points to
the high electrocatalytic activity of PEDOT towards the redox
reactions of DMcT-2Li as well as those of the doubly
1
5
protonated DMcT-2H. In addition, the reduction peak
shoulder) observed at 20.89 V was assigned to the reduction
(
of PEDOT. Furthermore, a small redox couple was observed
at +0.41 V. This couple is observed reproducibly in 1 mM
DMcT-2Li solutions, but not in 5 mM DMcT-2Li solutions
under otherwise identical experimental conditions. At this
time, we are not certain as to its origin.
II. Computational modeling of dimercaptothiophene and
dimercaptothiazole analogues of DMcT
III. Synthesis of the TBT family
Since it has been reported that dimercaptothiophenes are
5
unstable even for short periods of time at low temperatures
2,53
Based on the electrochemical and computational understand-
ing of the thermodynamics of the DMcT/PEDOT electro-
catalytic cycle, it would appear that any aromatic organosulfur
and handling of thiols can be problematic due to disulfide
Fig. 4 Schematic of the electrocatalytic cycle between DMcT and a
PEDOT film-coated electrode showing the reduction (discharge)
processes: (a) cationic EDOT (monomer unit within PEDOT film)
species is electrochemically reduced, forming neutral EDOT species
followed by (b) reduction of the DMcT oligomers/polymers, regener-
ating cationic EDOT species followed by (c) proton-coupled disulfide-
bond cleavage of DMcT oligomers/polymers into monomers. Note
that step (b) is computed to be slightly endothermic, but this is within
the accuracy of the method used.
Fig. 3 Schematic of the electrocatalytic cycle between DMcT and a
PEDOT film-coated electrode showing the oxidative (charge) pro-
cesses: (a) neutral EDOT (monomer unit within the PEDOT film)
+
species is electrochemically oxidized, forming [EDOT]
n
followed by
(b) oxidation of the DMcT monomer (or oligomer), regenerating
neutral EDOT species followed by (c) proton-coupled polymerization
of DMcT monomers (or oligomers) into oligomers/polymers.
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 4366–4376 | 4369

View Article Online
+
range from 21.50 to +0.80 V vs. Ag/Ag . Protected TBT
showed no significant redox responses over this potential range.
As expected, upon deprotection, redox reactions of deprotected
TBT were clearly observed, as shown in Fig. 6c (the baseline is
also shown in Fig. 6a). Similar results were obtained for all
compounds in the TBT family. Furthermore, the redox current
magnitude did not change significantly between 3–6 hours,
although the current response appeared to degrade after
significantly longer times (e.g. 7–9 hours) (Fig. S1). Thus, it
was evident that the protecting groups were cleanly and
completely cleaved by NH OH within y3 hours, giving rise
4
to the electroactive dithiolates. In addition, the redox behavior
of TT 10, which possesses a single thiolate group, was compared
with that of TBT 1. Relative to TT, TBT exhibited approxi-
mately twice the anodic/cathodic charges, indicating that
both thioacetate groups in TBT were cleaved (Fig. S2).
Fig. 5 Computed B3LYP/6-31++G** orbital energies for protonated
DMcT and EDOT oligomers, showing relative thermodynamics of the
DMcT/EDOT electrocatalyic cycles shown in Fig. 3 and 4.
Fig. 7 shows CVs for all members of the TBT family 1–8 at
4
bare GCEs in 0.1 M LiClO /AN solutions containing 50 mM
NH OH. Since their electrochemical behavior involves
4
5
formation, the protected thioacetate derivatives were synthe-
3
multiple oxidative processes (e.g., formation of dimers and
09
5
4
oligomers), it is difficult to establish a formal potential, E , for
sized. This protecting group has been widely used for thiols,
is known for its facile deprotection, and allows easy isolation,
characterization, and storage for long periods of time without
decomposition. Characterization and synthetic details can be
found in Supporting Information{.
the oxidation/reduction of the isolated monomer. Therefore,
&
we define an effective formal potential, E , as:
&
= (E op x + E rp ed)/2
E
(1)
Previous studies claimed that the N–C–S motifs (e.g., in a
thiazole or thiadiazole ring) exhibit faster charge transfer
2
,3
kinetics than C–C–S motifs (e.g., in a thiophene ring).
Therefore, several attempts were made to synthesize thiazole-
,5-bis(thiolate) 9 and related thiazole derivatives, but the
compounds were never isolated. The thiazole compounds
2
5
5
are expected to prefer the thione tautomer. Therefore, the
electrochemical characterization of the DMcT analogues
below is confined to the family of thiophene-bisthiolates 1–8
and thiophene-thiolates 10 and 11.
IV. Deprotection of thioacetates by ammonium hydroxide and
cyclic voltammograms for the TBT family at bare GCEs
Fig. 6b shows a CV for a 1 mM acetyl-protected TBT 1
4
solution at a bare GCE in 0.1 M LiClO /AN over the potential
Fig. 7 CVs for 1 mM thiophene-bis(thiolate) monomers at bare
GCEs in 0.1 M LiClO /AN solutions containing 50 mM NH OH.
4
4
2
1
Fig. 6 (a) CV for 50 mM NH
AN solution. CVs for 1 mM acetyl-protected TBT 1 at bare GCEs in
.1 M LiClO /AN solutions containing (b) 0 and (c) 50 mM NH OH.
The scan rate in all cases was 200 m s
4
OH at a bare GCE in a 0.1 M LiClO
4
/
The scan rate in all cases was 200 m s . Thiophene-bis(thiolate)
monomers: (a) TBT 1, (b) acetyl-TBT 7, (c) bromo-TBT 8, (d)
methoxy-TBT 6, (e) methyl-TBT 3, (f) 3,4-dimethyl-TBT 4, (g) 2,5-
dimethyl-TBT 5, and (h) 3,4-TBT 2.
0
4
4
2
1
.
4
370 | J. Mater. Chem., 2007, 17, 4366–4376
This journal is ß The Royal Society of Chemistry 2007

View Article Online
Table 1 Electrochemical and computed properties of the TBT family
including peak separation, DE , effective formal potentials, E (as
p
defined in eqn (1)), and computed DFT HOMO and LUMO orbital
energies for dithiolates under the C-PCM solvation model
Next, in order to assess the effects of regiochemistry, the
redox behavior of 2,5-dimethyl-TBT 5 (Fig. 7g) was compared
to that of 3,4-dimethyl-TBT 4 (Fig. 7f) and 3,4-TBT 2 (Fig. 7h)
to TBT 1 (Fig. 7a). Fig. 7g shows a CV for a 1 mM 2,5-
&
Electrochemical Theoretical
dimethyl-TBT 5 solution at a bare GCE in 0.1 M LiClO /AN
4
Compound
number
HOMO/ LUMO/
eV eV
4
containing 50 mM NH OH. With the thiolates at the 3,4-
&
Compound
TBT
3,4-TBT
Methyl-TBT
DE
p
/V
E
/V
positions versus the 2,5-positions the reduction peak potential
did not shift significantly, but the oxidation peak potential did
shift to be more positive, resulting in an increased peak
1
2
3
0.66
—
0.67
0.42
0.78
—
20.84 24.04
24.47
20.89 23.97
20.88 23.88
20.59 24.33
0.07
0.06
0.11
0.01
0.04
—
&
separation and a shift in E . Fig. 7h shows a CV for a 1 mM
3
2
,4-Dimethyl-TBT 4
,5-Dimethyl-TBT 5
3
,4-TBT 2 solution at a bare GCE in 0.1 M LiClO /AN
4
4
containing 50 mM NH OH. Compared to the parent
Methoxy-TBT
Acetyl-TBT
Bromo-TBT
6
7
8
—
24.04
20.49 24.30
20.78 24.21
0.01
21.05
20.30
compound, TBT 1, it was observed that the oxidation peak
potential shifted in the positive direction and the reduction
peak potential shifted in the negative direction, far enough that
the reduction current peak was not observed over this
potential range. From these increases in the peak separations,
therefore, it was deduced that the electron transfer kinetics for
the redox reactions of 3,4-dimethyl-TBT and TBT are faster
than those of 2,5-dimethyl-TBT and 3,4-TBT, respectively,
and indicated that the thiolates at 2,5-positions are more
reactive and thus preferred for fast charge/discharge cycles.
This effect illustrates the importance of the proximity of the
ring heteroatoms to the thiolate (i.e., S–C–S versus C–C–S
bonding patterns), a point previously indicated by Visco and
1.11
0.81
ox
p
where E
red
Ep
is the potential at the oxidative peak current and
is the potential at the reductive peak current. This
&
effective potential E was used to make comparisons between
the redox properties of compounds 1–8 as compiled in Table 1.
While we acknowledge that this analysis is not strictly
thermodynamically rigorous (due to coupled chemical reac-
tions), comparisons among this restricted group of materials is
likely valid if one assumes similar reactions.
In order to elucidate the effects of electron-withdrawing and
electron-donating substituents on the redox behavior of TBT,
TBT derivatives with acetyl, bromo, methoxy, methyl, and
dimethyl substituents were studied. Fig. 7a shows a CV for a
2
,3
co-workers.
Furthermore, mass transport properties for TBT 1, acetyl-
TBT 7, methyl-TBT 3, and dimethyl-TBT 4 were investigated.
Table 2 shows diffusion coefficients for the compounds,
1
mM TBT 1 solution at a bare GCE in 0.1 M LiClO
4
/AN
containing 50 mM NH OH. The oxidation and reduction peak
4
a
p
obtained from plots of the anodic peak current, i , as a
potentials were observed at 20.51 V and 21.17 V, respectively,
and the resulting effective redox potential was 20.84 V vs.
+
Ag/Ag . The peak separation of 660 mV indicated that this
function of the square root of the scan rate for the compounds
in 0.1 M LiClO /AN solutions containing 50 mM NH OH.
4
4
While the values seemed to be reasonable, the value for acetyl-
TBT was significantly greater than the others.
redox system is electrochemically irreversible at the bare GCE,
similar to DMcT-2Li. In the case of an electron-withdrawing
group such as acetyl-TBT 7 (Fig. 7b), the oxidation and
reduction peak potentials were shifted in the positive direction
and the effective redox potential was 20.49 V. In addition, the
oxidation peak current of acetyl-TBT was greater than that
of TBT. We believe that this difference in the magnitude of
the current is due, at least in part, to hydrogen bonding
interactions of the acetyl group in acetyl-TBT with surface
functional groups (such as hydroxyl) known to be present on
the surface of glassy carbon electrodes. Furthermore, in the
case of bromo-TBT 8 (Fig. 7c), although the reduction peak
potential was not shifted significantly, the oxidation peak
potential was shifted in the positive direction (similar to acetyl-
TBT), yielding an effective redox potential of 20.78 V. The
effective redox potential shifts in the positive direction were
consistent with the electron-withdrawing nature of the
substituents, as would be anticipated.
V. Comparison of experimental to theoretical results
Since the use of computational modeling suggested that the
redox reactions of the entire TBT family (Fig. 1) would be
electrocatalytically accelerated at PEDOT film-coated electro-
des, it was important to compare the computed gas-phase and
dielectric continuum model HOMO energy levels with experi-
mental electrochemical data at bare GCEs. In the dielectric
continuum model calculations, the use of computational
predictions offers the ability to rationally target novel
organosulfur compounds and elucidate the redox properties
of the compounds, such as methoxy-TBT 6, which exhibited
little oxidation current at a bare GCE. As illustrated in Fig. 8a,
the computed HOMO energy levels in the gas phase showed
poor correlation while the dielectric continuum model (Fig. 8b)
&
showed a reasonable correlation with experimental E .
For TBT derivatives with electron-donating groups such as
methyl-TBT 3 (Fig. 7e) and dimethyl-TBT 4 (Fig. 7f), although
the oxidation and reduction peak potential shifts were
modest in magnitude, the effective potentials were shifted
in the negative direction (20.89 V and 20.88 V, respectively)
consistent with the electron-donating properties of the
substituents. On the other hand, in the case of methoxy-TBT
Table 2 Diffusion coefficients for TBT 1, acetyl-TBT 7, methyl-TBT
3, and dimethyl-TBT 4
2
Compound number Diffusion coefficient/cm s
21
Compound
26
25
26
26
TBT
1
7
3
4
3.5 6 10
1.5 6 10
4.7 6 10
8.5 6 10
Acetyl-TBT
Methyl-TBT
Dimethyl-TBT
6
(Fig. 7d), no clearly developed oxidation and reduction
waves were observed over this potential range.
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 4366–4376 | 4371

View Article Online
ox
p
trends can be observed for E , suggesting that even though
the electrochemical oxidations are observed to be highly
irreversible, this effect is strongly consistent across the TBT
family. Additionally, while several of the compounds studied
had local minima geometries corresponding to different
conformers, the effect on the computed HOMO energies for
the dithiolate forms using the C-PCM solvation model was
relatively small.
As mentioned above, the main outliers showing relatively
poor correlation between the computed solvation energy levels
and experimental electrochemical measurements were com-
pounds 7 and 8. Since conformational effects did not appear to
cause significant changes in the computed electronic structure,
a likely reason for the differences in computed and experi-
mental values for these species was the distribution of
electrostatic charge density, rationalized by p-resonance forms.
One resonance form for acetyl-TBT 7 illustrates (see below)
how a thiolate group can also exist as a thione resonance form,
resulting in increased anionic character on the carbonyl oxygen
and decreased anionic charge on the sulfur atom.
Fig. 8 Comparison between computed B3LYP/6-31++G** HOMO
&
energy levels and experimental E oxidation energies from Table 1 for
(a) dithiol (i.e., protonated) compounds in gas-phase, showing low
correlation and (b) dithiolates (i.e., deprotonated) using the C-PCM
solvation model for AN. The solid lines indicate the best-fit linear
regressions with the equations indicated. Note that the largest outliers
are the acetyl-TBT 7 species for (a) as well as acetyl-TBT 7 and bromo-
TBT 8 species for (b). For the gas-phase oxidation energies in (a), the
slope of the trend line is physically incorrect since species with higher
computed ionization potentials should have more positive experi-
mental oxidation potentials. On the other hand, the solvation model in
In contrast, although halide substituents exhibit some
electron-withdrawing effects, these are mostly s-inductive,
while the p electron density is electron-donating—suggesting a
net stabilization of the oxidized form of bromo-TBT 8. As
illustrated in Fig. 8b, bromo-TBT 8 is actually easier to oxidize
than was computed (i.e., it lies below the least-squares trend
line), while acetyl-TBT 7 is experimentally harder to oxidize
than was computed (i.e., it lies above the least-squares trend
line). Since the first-principles methods such as DFT implicitly
include simple valence-bond resonance pictures as described
above, the error is likely due to problems in the treatment of
polar solvents interacting with anions in the C-PCM dielectric
model.
(b) indicates a qualitatively correct trend line.
The gas-phase calculations exhibited poor correlation in
Fig. 8a because the protonated dithiols showed different
conformations of the thiol protons across the series (i.e.,
coplanar with the thiophene ring versus nonplanar), which had
a pronounced effect on the computed electronic structure and
resulting energies. Consequently, not only were the gas-phase
&
HOMO levels poorly correlated with the experimental E
values, but the trend was qualitatively (physically) incorrect,
implying that a more negative computed HOMO energy (i.e.,
large ionization potential) would correlate to a more negative
experimental oxidation potential (i.e., easier to oxidize). On
the other hand, deprotonated dithiolate compounds exhibited
poor correlation between gas-phase HOMO energy levels and
experimental electrochemical data because the dianion com-
pounds were poorly described by the gas-phase calculations
VI. Characterization of the electrocatalytic activity of PEDOT
films towards the redox reactions of the TBT family
The electrocatalytic activity of PEDOT films towards the
redox reactions of the TBT family was also characterized by
CV, RDE voltammetry, and DPSCA. The potential range
+
employed for CV was from 21.50 V to +0.80 V vs. Ag/Ag ,
since the electrocatalytic reactions are expected to occur over
the potential region where a PEDOT film is conductive (i.e., in
the p-doped state). PEDOT film-modified electrodes exhibited
electrocatalytic activity towards both the oxidation and
reduction of all the compounds characterized in this study.
In this section, we exemplify the electrocatalytic activity of
PEDOT films by focusing on the TBT 1 (Fig. 9) and methyl-
TBT 3 (Fig. 10) systems, while the rest can be found in the
Supporting Information (Fig. S3–S8){.
(i.e., the computed HOMO energies were .0 eV, implying
oxidation of the dithiolate in the gas phase was favorable).
In contrast, using the acetonitrile C-PCM solvation model,
the calculated HOMO energies were well described and Fig. 8b
showed good correlation between the computed HOMO
&
energies and experimental E values. However, it is also
worth noting that the correlation includes some error in the
&
experimental E (i.e., an error bar for the y-coordinates),
which is due to overpotentials involved in the irreversible
redox processes at a bare GCE. The largest outliers were
acetyl-substituted 7 and bromo-substituted 8 species. Similar
Fig. 9b and 9c present representative CVs for a 1 mM TBT 1
solution at a bare GCE (Fig. 9b) and a PEDOT film-coated
GCE (obtained at the fifth cycle, Fig. 9c) in 0.1 M LiClO
4
/AN
4
372 | J. Mater. Chem., 2007, 17, 4366–4376
This journal is ß The Royal Society of Chemistry 2007

View Article Online
exchange between TBT and PEDOT. Therefore the use of a
composite cathode in practical lithium/lithium-ion batteries
would retain high electrocatalytic activity of PEDOT and thus
higher TBT redox currents.
Fig. 10b and 10c present CVs for a 1 mM methyl-TBT 3
solution at a bare GCE (Fig. 10b) and a PEDOT film-coated
GCE (obtained at the fifth cycle, Fig. 10c) in 0.1 M LiClO
4
/
AN containing 50 mM NH OH. As was the case for TBT 1,
4
both the oxidation and reduction reactions were electrocata-
lyzed at the PEDOT film-modified GCE. For the oxidation
(
oligomerization process), the current response obtained due
to the oxidation of methyl-TBT clearly increased at the
PEDOT film-modified GCE. Moreover, the reduction process
was greatly electrocatalyzed. This is most evident by noticing
the shift in the reduction peak potential from 21.15 V to
2
0.89 V (DE = 260 mV) and the increased current magnitude
Fig. 9 (a) CV for a PEDOT film-coated GCE in a 0.1 M LiClO
solution. Representative CVs for 1 mM TBT 1 at (b) bare and (c)
PEDOT film-coated GCEs in a 0.1 M LiClO /AN solution containing
0 mM NH OH. The scan rate in all cases was 20 m s
4
/AN
at 20.89 V.
4
Next, in order to gain some insight on the charge transfer
kinetics of the redox reactions of the dimercaptothiophene
compounds at bare and PEDOT film-modified GCEs, the
dimerization process of TBT was characterized quantitatively
and compared with that of DMcT-2Li using Koutecky–Levich
2
1
.
5
4
(
K–L) plots obtained via RDE voltammetry. In these
electrochemical kinetics studies, the PEDOT film was treated
2
4
as an extension of the electrode. We feel that this is a
reasonable assumption since in the potential region of interest,
the PEDOT film is present in the conducting (p-doped) form.
TT 10, which possesses a single thiolate group, was employed
in this study to avoid any complications caused by poly-
merization process present in TBT. As discussed in Section IV,
unlike DMcT-2Li, it is difficult to establish a formal potential
for the dimerization of TBT because of the following
polymerization process as well as the sluggish charge transfer
kinetics.
Fig. 10 (a) CV for a PEDOT film-coated GCE in a 0.1 M LiClO
4
/AN
Fig. 11 shows the K–L plot for the oxidation of TT 10 at a
solution. Representative CVs for 1 mM methyl-TBT 3 at (b) bare and
bare GCE in a 0.1 M LiClO /AN solution containing 50 mM
4
(
c) PEDOT film-coated GCEs in a 0.1 M LiClO
4
/AN solution
NH
4 L
OH. In this experiment the limiting current, i , was
21
containing 50 mM NH
4
OH. The scan rate in all cases was 20 m s
.
+
measured at 20.25 V vs. Ag/Ag at different rotation rates. A
1
/2
Levich plot of i
L
versus v (not shown) exhibited curvature,
containing 50 mM NH
4
OH. Fig. 9a shows a CV for a PEDOT
/AN solution for
indicating that the reaction is kinetically limited. A plot of 1/i
21/2
L
film-coated GCE in a 0.1 M LiClO
4
versus v
(K–L plot) was linear, and from the intercept the
comparison. For the oxidation of TBT at a PEDOT film-
modified GCE, the onset potential shifted towards negative
values and the current response increased relative to a bare
GCE. On the other hand, for the reduction, the peak potential
obtained at a bare GCE shifted with the increase in the current
magnitude from 21.06 V to 20.85 V (DE = 210 mV) at a
PEDOT film-modified GCE. Thus, PEDOT is capable of
electrocatalyzing both the oxidation and reduction processes
of TBT as well as DMcT-2Li, as shown previously (Fig. 2).
Furthermore, the amount of the increase in the TBT redox
current responses is likely smaller than would be found in
practical lithium/lithium-ion batteries. In such applications, an
organosulfur compound (e.g., TBT) would be incorporated
into a composite cathode with PEDOT, rather than freely
diffusing towards a PEDOT film from the electrolyte solution
as studied in this work. Consequently, these composite
electrodes limit the deposition of an insulating layer of TBT
polymer over a PEDOT film, which diminishes electron
Fig. 11 Koutecky–Levich plot for the oxidation (dimerization) of TT
10 at a bare GCE in a 0.1 M LiClO /AN solution containing 50 mM
4
+
OH. The limiting currents were measured at 20.25 V vs. Ag/Ag .
NH
4
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 4366–4376 | 4373

View Article Online
kinetically limited current, i
2
k
, was estimated to be 8.1 6
A. The i value includes the rate constant k(E), which is a
4
1
0
k
function of applied potentials. The k(E) value for the dimeriza-
2
2
21
tion at 20.25 V was determined to be 4.2 6 10 cm s from
which the standard rate constant, k , was estimated to be
0
2
8
21
about 2.9 6 10 cm s . A simulated RDE voltammogram
2
8
21
with the same rate constant (2.9 6 10 cm s ), obtained
via DigiSim , exhibited behavior quite similar to the experi-
1
mental data.
Table 3 summarizes the rate constants, determined as
described above, for the oxidation (dimerization) of TT and
DMcT-2Li at bare and PEDOT film-modified GCEs. The rate
constant for the oxidation of DMcT-2Li at a bare GCE was
2
6
21
Fig. 12 Anson plots for (a) anodic and (b) cathodic reactions of a
determined to be 6.6 6 10 cm s . Therefore, at a bare
GCE, the dimerization process of DMcT-2Li was found to be
y200 times faster than that of TT. The difference in charge
transfer kinetics points to the importance of molecular
structure (i.e., thiadiazole versus thiophene structures) as well
as regiochemistry (dithiolates at 2,5-positions versus 3,4-
positions) as discussed earlier (see Section IV). At a PEDOT
PEDOT film-coated GCE in a 0.1 M LiClO
plots for (c) anodic and (d) cathodic reactions of 1 mM DMcT-2Li at a
PEDOT film-coated GCE in a 0.1 M LiClO /AN solution. Theta in
4
/AN solution. Anson
4
1/2
1/2
1/2
the x-axis legend is equal to t + (t 2 t) 2 t (t = 60 s in this
experiment).
PEDOT film-coated GCE in a 0.1 M LiClO /AN solution as
4
0
film-modified GCE, k for the dimerization of DMcT-2Li was
4
background responses. The anodic and cathodic charges due
to the redox reactions of DMcT-2Li can be estimated by
subtracting the redox charge of PEDOT (Fig. 12a and 12b)
from the redox charge of the DMcT-2Li/PEDOT system
(Fig. 12c and 12d). The anodic and cathodic charges due to the
redox reactions of DMcT-2Li were determined to be 1.03 mC
and 1.02 mC, respectively, indicating high coulomb efficiency
2
21
determined to be 2.9 6 10 cm s , representing a 44-fold
acceleration. On the other hand, k for the dimerization of TT
0
2
6
21
was determined to be 2.7 6 10 cm s , which represents a
3-fold acceleration. It appears that, while the dimerization
9
process of DMcT-2Li at a PEDOT film-modified GCE is
y100 times faster than that of TT, the electrocatalytic effect of
PEDOT is greater (ca. 2-fold) towards TT than towards
DMcT-2Li.
(y99%) at a PEDOT film-modified GCE.
Table 4 summarizes the mean normalized charge ratios for
the oxidation and reduction of TBT, methyl-TBT, dimethyl-
TBT, and acetyl-TBT, compared to the charge obtained for the
DMcT-2Li/PEDOT system under identical experimental con-
ditions. The electrocatalytic activity of PEDOT towards the
oxidations of the compounds (i.e., the anodic charge) in this
family was, in general, greater than that towards DMcT-2Li.
On the other hand, the activity towards the reductions (i.e., the
cathodic charge) was, in general smaller than DMcT-2Li in
most of the trials. Furthermore, the deviations obtained for the
anodic charges for the compounds were wider than those for
the cathodic charges. For the oxidation process, in which the
charge derives from oxidation of the compounds in solution
Finally, in order to compare the electrocatalytic activity of
PEDOT towards the redox reactions of TBT, methyl-TBT,
dimethyl-TBT, and acetyl-TBT with that towards DMcT-2Li,
the anodic and cathodic charges due to redox reactions of the
compounds at PEDOT film-modified GCEs were charac-
terized using Anson plots obtained via DPSCA. The obtained
2
2
charges were normalized to the surface coverage (mC cm ) of
PEDOT since it was found that the catalytic activity was
proportional to the charge consumed during the electro-
chemical polymerization of EDOT over the range of charges
2
2
employed in this study (y180–215 mC cm ). As a com-
parison, the normalized charges for the DMcT-2Li/PEDOT
system were set to 1 for the anodic and cathodic charges,
respectively.
(i.e., a diffusional process), the charge appeared to be more
dependent on the morphology of the PEDOT film (the
morphology of the electrode surface) rather than its coverage.
Indeed, as illustrated in Fig. S9, the PEDOT film is very rough
and exhibits significant changes in film morphology from film
to film. Consequently, the measured anodic charge was highly
variable from trial to trial, and the mean measured anodic
charges of the compounds studied were relatively similar.
Upon oxidation, a fraction of the oxidized dimercaptothio-
phenes is incorporated into the PEDOT matrix, so that upon
reduction, the oligomers incorporated into the PEDOT film as
well as the ones in solution can be reduced. Thus, the measured
cathodic charge can be strongly dependent on the PEDOT
coverage. Since the normalization process includes the
PEDOT coverage as described above, the cathodic charge
shown in Table 4 was highly consistent across trials for a given
compound. However, while compounds 1, 3, and 4 exhibited
normalized cathodic charge y0.6 times that measured for
Fig. 12c and 12d present Anson plots for (c) anodic and (d)
cathodic reactions of DMcT-2Li at a PEDOT film-coated
4
GCE in a 0.1 M LiClO /AN solution. Fig. 12a and 12b show
Anson plots for (a) anodic and (b) cathodic reactions of the
Table 3 Standard rate constants for the oxidation (dimerization) of
TT 10 and DMcT-2Li at bare and PEDOT film-coated GCEs. The
transfer coefficient, a, for both reactions was assumed to be 0.5, and an
&
effective formal potential, E , was employed (20.98 V and 20.61 V
for TT and DMcT-2Li, respectively) in order to determine the
standard rate constants
0
Standard rate constant, k /cm s
21
Acceleration
&
Compound
TT
E /V At bare GCEs At PEDOT films factor
28
26
24
20.98 2.9 6 102
DMcT-2Li 20.61 6.6 6 10
2.7 6 10
2.9 6 10
93
44
6
4
374 | J. Mater. Chem., 2007, 17, 4366–4376
This journal is ß The Royal Society of Chemistry 2007

View Article Online
Table 4 Mean normalized charge ratios (with standard deviations indicated) for the oxidation and reduction of TBT 1, methyl-TBT 3, dimethyl-
TBT 4, and acetyl-TBT 7, compared to the charges obtained in the DMcT-2Li/PEDOT system under identical experimental conditions
Compound
Compound number
Mean normalized anodic charge ratio
Mean normalized cathodic charge ratio
No. of trials
TBT
1
3
4
7
—
1.51 ¡ 0.54
1.72 ¡ 0.92
2.06 ¡ 0.40
1.93 ¡ 1.05
1
0.66 ¡ 0.08
0.69 ¡ 0.08
0.59 ¡ 0.05
0.31 ¡ 0.05
1
4
3
3
3
—
Methyl-TBT
Dimethyl-TBT
Acetyl-TBT
DMcT-2Li
DMcT-2Li, compound 7 exhibited a significantly lower value
i.e., y0.3 times that measured for DMcT-2Li). As described
rechargeable batteries. Therefore, it was found that the
rational electrocatalytic cycle predicted via computational
modeling helped in the design of organosulfur compounds
(thiolates) the redox reactions of which can be electrocataly-
tically accelerated by conducting polymers. Furthermore, the
synthetically-tailorable materials prepared in this study would
also enable the rational design of additional promising
electroactive materials by further chemical modification.
Information about molecular interactions between the TBT
family and PEDOT film surfaces, involved in the electro-
catalytic reactions, was also obtained in this study. The
measured anodic charge, normalized to the PEDOT coverage,
was found to be rather variable (from trial to trial) due to the
solution measurements employed, which are highly dependent
on the morphology of the PEDOT film (the morphology of the
electrode surface) rather than its coverage. On the other hand,
the measured cathodic charge, normalized to the PEDOT film
coverage, was highly consistent for a given compound, since
it largely reflects the amount of an oligomeric disulfide
compound incorporated into the PEDOT film after oxidation
of the thiolate monomers. The electron-withdrawing acetyl
substituent exhibits a markedly smaller reduction charge,
reflecting an unfavorable interaction between the positively-
charged PEDOT film and the acetyl-TBT substrate. This
suggests that for p-type conducting polymer electrocatalysts
such as PEDOT, electron-donating substituents should be
employed to give rise to higher partitioning into the polymer
film matrix, and thus provide greater energy output and better
charge/discharge cycle performance as cathode materials for
lithium/lithium-ion rechargeable batteries.
(
above, the measured cathodic charge can be highly dependent
on the amount of reduced oligomers incorporated into the
PEDOT film-modified GCE rather than in solution.
Therefore, this measurement is, at least in part, indicative of
the degree of interaction (whether favorable or unfavorable)
between the PEDOT matrix and the oligomeric disulfide
compounds.
Since acetyl-TBT 7 exhibited a significantly smaller cathodic
charge, it was clear that it experienced a less favorable
interaction with the PEDOT film-coated GCE and after
oxidation, less of the oligomeric disulfide material appeared
to be incorporated into the PEDOT film. As described above
(
see Section V), an acetyl group has a p-resonance withdrawing
effect on the dimercaptothiophene structure, which results in a
non-aromatic resonance structure (involving an exo-thione
group). Similarly, this resonance structure should have a
higher positive partial charge on the thiophene ring as electron
density migrates into the acetyl group. The consequence of
both effects is an increase in the unfavorable interactions
between the acetyl-TBT oligomers and the positively-charged
PEDOT film. Higher positive partial charges on the thiophene
ring result in unfavorable electrostatic interactions with the
positively-charged PEDOT film, and the decreased aromaticity
of the thiophene ring decreases favorable p–p interactions.
Therefore, it appears that for electrocatalysis between
p-type conducting polymers and organosulfur compounds,
p electron-withdrawing substituents such as an acetyl group
are undesirable.
Conclusions
Acknowledgements
A family of a novel class of organosulfur compounds based on
TBT and its derivatives, with a variety of functional groups
This work was supported in part by the Cornell Center for
Materials Research (CCMR), a Materials Research Science
and Engineering Center of the National Science Foundation
(DMR-0079992) and Fuji Heavy Industries, Ltd (FHI). The
authors are grateful to Professor Noboru Oyama (Tokyo
University of Agriculture and Technology), Dr Osamu
(
electron-donating or electron-withdrawing groups) and regio-
chemistries, has been synthesized, and its redox behavior as
well as the electrocatalytic activity of PEDOT towards the
family has been investigated in detail. As anticipated, the
&
–
effective redox potential (E ) shifts, determined experimen-
Hatozaki (FHI), Professor J o´ n T. Njardarson (Cornell
¨
University), and Burak Ulg u¨ t (Cornell University) for helpful
tally for the compounds, were consistent with the electron-
donating and electron-withdrawing nature of the substituents.
The effective redox potentials of the TBT family exhibited
good correlation with prediction via computational modeling.
Moreover, the redox reactions of all the compounds
synthesized were found to be electrocatalytically accelerated
by PEDOT film-modified GCEs, as predicted, based on the
understanding of the electrocatalytic cycle exhibited in the
DMcT/PEDOT system. Several compounds exhibited clear
promise as potential cathode materials for lithium/lithium-ion
discussions. JCH is grateful to Maria Morar (Cornell
University) for translation assistance. We are grateful to
Professor Dotsevi Y. Sogah and FHI for resources.
References
1
M. Liu, S. J. Visco and L. C. De Jonghe, J. Electrochem. Soc.,
991, 138, 1891–1895.
1
2
M. Liu, S. J. Visco and L. C. De Jonghe, J. Electrochem. Soc.,
1991, 138, 1896–1901.
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 4366–4376 | 4375

View Article Online
3
4
5
M. M. Doeff, S. J. Visco and L. C. De Jonghe, J. Electrochem.
Soc., 1992, 139, 1808–1812.
M. M. Doeff, M. M. Lerner, S. J. Visco and L. C. De Jonghe,
J. Electrochem. Soc., 1992, 139, 2077–2081.
N. Oyama, T. Tatsuma, T. Sato and T. Sotomura, Nature, 1995,
M. W. Wong, C. Gonzalez and J. A. Pople, GAUSSIAN 03
(Revision B.04), Gaussian, Inc., Wallingford, CT, 2004.
27 A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652.
28 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter
Mater. Phys., 1988, 37, 785–789.
3
74, 196–196.
29 M. Cossi, N. Rega, G. Scalmani and V. Barone, J. Comput. Chem.,
2003, 24, 669–681.
30 V. Barone and M. Cossi, J. Phys. Chem. A, 1998, 102, 1995–2001.
31 J. P. Perdew and M. Levy, Phys. Rev. B: Condens. Matter Mater.
Phys., 1997, 56, 16021–16028.
32 A. Seidl, A. Gorling, P. Vogl, J. A. Majewski and M. Levy, Phys.
Rev. B: Condens. Matter Mater. Phys., 1996, 53, 3764–3774.
33 M. Levy, Phys. Rev. A: At., Mol., Opt. Phys., 1995, 52,
R4313–R4315.
6
7
E. M. Genies and S. Picart, Synth. Met., 1995, 69, 165–166.
K. Naoi, Y. Oura, Y. Iwamizu and N. Oyama, J. Electrochem.
Soc., 1995, 142, 354–360.
P. Novak, K. Muller, K. S. V. Santhanam and O. Haas, Chem.
Rev., 1997, 97, 207–281.
K. Naoi, K. Kawase, M. Mori and M. Komiyama, J. Electrochem.
Soc., 1997, 144, L173–L175.
0 L. Yu, X. H. Wang, J. Li, X. B. Jing and F. S. Wang, J. Power
8
9
1
1
1
1
1
1
1
1
Sources, 1998, 73, 261–265.
34 R. W. Godby, M. Schluter and L. J. Sham, Phys. Rev. B: Condens.
Matter Mater. Phys., 1988, 37, 10159–10175.
35 C. G. Zhan, J. A. Nichols and D. A. Dixon, J. Phys. Chem. A,
2003, 107, 4184–4195.
36 J. C. Rienstra-Kiracofe, G. S. Tschumper, H. F. Schaefer, S. Nandi
and G. B. Ellison, Chem. Rev., 2002, 102, 231–282.
37 J. C. Rienstra-Kiracofe, C. J. Barden, S. T. Brown and
H. F. Schaefer, J. Phys. Chem. A, 2001, 105, 524–528.
38 G. de Oliveira, J. M. L. Martin, F. de Proft and P. Geerlings, Phys.
Rev. A: At., Mol., Opt. Phys., 1999, 60, 1034–1045.
39 L. A. Curtiss, P. C. Redfern, K. Raghavachari and J. A. Pople,
J. Chem. Phys., 1998, 109, 42–55.
1 H. Tsutsumi, Y. Oyari, K. Onimura and T. Oishi, J. Power
Sources, 2001, 92, 228–233.
2 J. S. Cho, S. Sato, S. Takeoka and E. Tsuchida, Macromolecules,
2
001, 34, 2751–2756.
3 H. Uemachi, Y. Iwasa and T. Mitani, Electrochim. Acta, 2001, 46,
305–2312.
2
4 L. J. Xue, J. X. Li, S. Q. Hu, M. X. Zhang, Y. H. Zhou and
C. M. Zhan, Electrochem. Commun., 2003, 5, 903–906.
5 N. Oyama, Y. Kiya, O. Hatozaki, S. Morioka and H. D. Abru n˜ a,
Electrochem. Solid-State Lett., 2003, 6, A286–A289.
6 X. G. Yu, J. Y. Xie, J. Yang, H. J. Huang, K. Wang and Z. S. Wen,
J. Electroanal. Chem., 2004, 573, 121–128.
7 S. R. Deng, L. B. Kong, G. Q. Hu, T. Wu, D. Li, Y. H. Zhou and
Z. Y. Li, Electrochim. Acta, 2006, 51, 2589–2593.
40 F. DeProft and P. Geerlings, J. Chem. Phys., 1997, 106, 3270–3279.
41 J. Muscat, A. Wander and N. M. Harrison, Chem. Phys. Lett.,
2001, 342, 397–401.
1
1
8 M. Amaike and T. Iihama, Synth. Met., 2006, 156, 239–243.
9 M. Liu, S. J. Visco and L. C. De Jonghe, J. Electrochem. Soc.,
42 G. R. Hutchison, Ph.D. Dissertation, Northwestern University,
Evanston, Illinois, 2004.
1
990, 137, 750–759.
43 M. J. Kamlet, J. L. M. Abboud, M. H. Abraham and R. W. Taft,
J. Org. Chem., 1983, 48, 2877–2887.
44 M. Jonsson, A. Houmam, G. Jocys and D. D. M. Wayner, J. Chem.
Soc., Perkin Trans. 2, 1999, 425–429.
2
2
0 S. Picart and E. Genies, J. Electroanal. Chem., 1996, 408, 53–60.
1 Y. Kiya, G. R. Hutchison, J. C. Henderson, T. Sarukawa,
O. Hatozaki, N. Oyama and H. D. Abru n˜ a, Langmuir, 2006, 22,
1
0554–10563.
45 E. Shouji, H. Matsui and N. Oyama, J. Electroanal. Chem., 1996,
417, 17–24.
46 E. Shouji and D. A. Buttry, J. Phys. Chem. B, 1999, 103,
2239–2247.
2
2 Y. Kiya, A. Iwata, T. Sarukawa, J. C. Henderson and H. D.
Abru n˜ a, J. Power Sources, 2007, DOI: 10.1016/j.jpowsour.
2007.04.086.
2
2
2
3 O. Y. Posudievsky, S. A. Biskulova and V. D. Pokhodenko,
Electrochem. Commun., 2005, 7, 477–482.
4 Y. Kiya, O. Hatozaki, N. Oyama and H. D. Abru n˜ a, J. Phys.
Chem. C, 2007, DOI: 10.1021/jp073486f.
5 J. M. Tour, L. Jones, D. L. Pearson, J. J. S. Lamba, T. P. Burgin,
G. M. Whitesides, D. L. Allara, A. N. Parikh and S. V. Atre, J. Am.
Chem. Soc., 1995, 117, 9529–9534.
47 C. Kvarnstrom, H. Neugebauer, S. Blomquist, H. J. Ahonen,
J. Kankare and A. Ivaska, Electrochim. Acta, 1999, 44, 2739–2750.
48 G. Zotti, S. Zecchin and G. Schiavon, Chem. Mater., 2000, 12,
2996–3005.
49 Q. Pei, G. Zuccarello, M. Ahlskog and O. Inganas, Polymer, 1994,
35, 1347–1351.
50 X. W. Chen and O. Inganas, J. Phys. Chem., 1996, 100,
15202–15206.
2
6 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi,
V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
J. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador,
J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels,
M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui,
A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu,
A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox,
T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara,
M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen,
51 It should be mentioned that whereas in the electrochemical studies
we employed the deprotonated forms of the thiophenes, in the
DFT calculations we employed the protonated forms. This was
deliberate from our part in an effort to minimize complications in
the observed redox response in the former, and the computational
difficulties of dealing with anions in the latter. However, we
feel that the results from the computations can be compared to
the electrochemical ones after taking into account the differences
in pH.
52 S. Gronowitz and P. Moses, Acta Chem. Scand., 1962, 16, 105–110.
53 G. S. Ponticello, C. N. Habecker, S. L. Varga and
S. M. Pitzenberger, J. Org. Chem., 1989, 54, 3223–3224.
54 B. de Boer, H. Meng, D. F. Perepichka, J. Zheng, M. M. Frank,
J. Y. Chabal and Z. Bao, Langmuir, 2003, 19, 4272–4284.
55 T. Yamamoto, M. Itoh, N. U. Saitoh, M. Muraoka and
T. Takeshima, J. Chem. Soc., Perkin Trans. 1, 1990, 2459–2463.
4
376 | J. Mater. Chem., 2007, 17, 4366–4376
This journal is ß The Royal Society of Chemistry 2007