Strategies towards functionalised electronically conducting organic copolymers: Part 2. Copolymerisation

Article abstract of DOI:10.1039/B001279N

Here we discuss the application of X-ray photoelectron spectroscopy and absorbance-reflectance FT-IR spectroscopy to establish and quantify reactivity relationships between a range of thiophene and pyrrole monomers. In particular we investigate the application of these techniques to the characterisation of conducting polymer materials grown potentiostatically from solutions containing a binary mixture of monomers. Our data have shown that XPS is especially effective in determining polymer composition and the linear correlation between this and solution composition has enabled prescriptive synthesis of copolymer materials from the different combinations of monomers described here. This technique is much more convenient and more reliable than elemental analysis. In contrast we show that FT-IR studies, whilst providing a useful qualitative guide to the functional group content of the material, do not facilitate detailed quantitative analysis because of large intrinsic errors.

Full text of DOI:10.1039/B001279N

View Article Online / Journal Homepage / Table of Contents for this issue
J O U R N A L O F
Strategies towards functionalised electronically conducting organic
copolymers: Part 2. Copolymerisation
Karl S. Ryder,*a Lutz F. Schweiger,b Andrew Glidlec and Jon. M. Cooperc
aDepartment of Chemistry, Hawthorn Building, De Montfort University, The Gateway,
Leicester, UK LE1 9BH
C H E M I S T R Y
bDepartment of Chemistry, University of Aberdeen, Meston Walk, Old Aberdeen, UK AB24
3UE
cDepartment of Electronics & Electrical Engineering, Rankine Building, University of
Glasgow, Glasgow, UK G12 8LT
Received 16th February 2000, Accepted 12th May 2000
Published on the Web 26th June 2000
Here we discuss the application of X-ray photoelectron spectroscopy and absorbance±re¯ectance FT-IR
spectroscopy to establish and quantify reactivity relationships between a range of thiophene and pyrrole
monomers. In particular we investigate the application of these techniques to the characterisation of conducting
polymer materials grown potentiostatically from solutions containing a binary mixture of monomers. Our data
have shown that XPS is especially effective in determining polymer composition and the linear correlation
between this and solution composition has enabled prescriptive synthesis of copolymer materials from the
different combinations of monomers described here. This technique is much more convenient and more reliable
than elemental analysis. In contrast we show that FT-IR studies, whilst providing a useful qualitative guide to
the functional group content of the material, do not facilitate detailed quantitative analysis because of large
intrinsic errors.
oligomer solubilities of pyrrole and thiophene monomers and
is problematic because characterisation of the insoluble,
infusible polymer materials is dif®cult.{ For example the
composition of copolymers of pyrrole and N-arylpyrroles were
studied using elemental analysis.9±11 This is typical of many
subsequent studies but it was concluded at an early stage that
these methods, whilst both tedious and expensive, are also
often unreliable. Nevertheless, many research groups have
adopted this approach and their efforts have been recently
summarised.12 In cases where the component reactivities and
solubilities are similar and there is no steric hindrance, e.g.
pyrrole and N-methylpyrrole, the composition of the feedstock
solution does mirror the make up of the copolymer; however
this situation is unusual.4,5 Other methodologies to control the
composition of a copolymer grown from a particular solution
mixture have been adopted, including manipulation of
diffusion layer concentrations using pulsed polarography.13
However, this latter technique still requires an implicit
empirical knowledge of the reactivities of the component
monomers.
Here we concentrate on the study of copolymer materials
prepared electrochemically from solutions containing a binary
mixture of different monomer species. The monomer species
chosen for this study are good examples of functional utility
where subsequent chemistry at these groups imparts useful and
speci®c properties to the material. These studies have involved
quantitative determination of polymer composition using X-
ray photoelectron spectroscopy and qualitative comparisons of
intensity data using absorbance±re¯ectance FT-IR spectro-
scopy. These complementary techniques have yielded an
insight into the relative ability of a series of heterocyclic
conducting polymer precursor compounds to form adherent
copolymer ®lms. Ultimately we seek to determine the
appropriate calibration data and conditions for each monomer
mixture to enable prescriptive synthesis of copolymer materials
with a given composition from a particular set of monomer
components. We also describe our attempts to determine the
relative reactivities of several functionalised pyrrole and
1.0 Introduction
In our research we have sought routes to functionalised organic
conductors as materials with potentially good electronic
properties e.g. high conductivity, low bias voltage, and
functional utility. Both academic and industrial interest in
such materials continues to expand and important application
growth areas include ®lms for light emitting devices and
polymer transistors for active displays and data storage,
biosensors and arti®cial olfaction.1,2 Our efforts have focused
on the fabrication of polymer materials based on copolymers of
functionalised pyrrole and thiophene units. In this approach we
seek to minimise the steric interactions between adjacent
functionalised repeat units in the polymer that are known to be
detrimental to the electronic properties of the material e.g.
conjugation length, conductivity.3±7 We have adopted a two-
fold strategy:
(1) Synthesis of mixed oligomeric precursors containing
pyrrole and functionalised thiophene.
(2) Electrochemical copolymerisation from solutions con-
taining a mixture of thiophene and pyrrole monomers.
The former strategy has certain advantages associated with
control resulting from synthesis of discrete precursor units but
is made dif®cult by the rigorous nature of the synthetic
chemistry. This is described in an earlier paper in this journal.8
The second strategy is attractive from a different perspective
because electrochemical synthesis of copolymers from solution
mixtures is experimentally facile. Additionally, the desired
functionality can be incorporated into the ®lm as either a
pyrrole or thiophene derivative. The main disadvantage of this
procedure is that the composition of copolymer materials
produced by this technique seldom corresponds to the
composition of the solution from which the material was
polymerised. This re¯ects the disparate reactivities and
{Another more subtle problem lies in the general assumption that such
copolymer materials are homogeneous rather than phase separated
mixtures, however, we will not address this issue here.
DOI: 10.1039/b001279n
J. Mater. Chem., 2000, 10, 1785±1793
This journal is # The Royal Society of Chemistry 2000
1785

View Article Online
thiophene species by comparing the composition of polymer
material, as determined using XPS, with the composition of the
binary feed solution from which the polymers were electro-
chemically grown. In addition we have examined the
absorbance±re¯ectance FT-IR spectra of the polymer ®lms
to determine the relative intensity of the nitrile and ester
carbonyl bands of the functionalised heterocyclic species.
previously described.15,16 These four stages consist of (i)
Friedel±Crafts acylation with succinic anhydride to give 4-
[(N-phenylsulfonyl)pyrrol-3-yl]-4-oxobutyric acid, (ii) Clem-
mensen reduction of the ketone group, (iii) cleavage of the N-
protecting phenylsulfonyl group to give 1H-pyrrole-3-butyric
acid and (iv) coupling of the free pyrrole acid with
penta¯uorophenol using dicyclohexylcarbodiimide. These pro-
cedures are described sequentially below.
2.0 Experimental
Preparation of 4-[(N-phenylsulfonyl)pyrrol-3-yl]-4-oxobuty-
ric acid, (i). To a suspension of AlCl3 (14.67 g; 110 mmol) in
175 ml of CH2Cl2, succinic anhydride (5.5 g; 55 mmol) was
added. This was stirred for 20 min at RT after which time all
material was in solution. A solution of N-phenylsulfonylpyr-
role (10.35 g; 50 mmol) in 25 ml of CH2Cl2 was then added
dropwise. The resultant orange solution was stirred at RT for a
further 18 h. The reaction mixture was quenched with ice and
H2O, the organics were extracted with CH2Cl2 (3650 ml); the
extracts were washed with brine, and dried over MgSO4 and the
solvent was removed under partial vacuum. The target, (i), was
obtained as a white crystalline solid, 6.68 g (21.7 mmol; 43%);
mp 124±126 ³C, after recrystallisation.
2.1 General methods
Commercial reagents were used without further puri®cation.
Solvents were puri®ed and degassed according to standard
1
procedures. H and 13C NMR data were obtained on Varian
Unity Inova 400 MHz and Bruker AC-F 250 MHz spectro-
meters. FT-IR spectra were recorded using a Nicolet 205 FT-
IR spectrometer; polymer ®lm spectra were recorded ex situ
using a specular re¯ectance accessory at an incident angle of
45³. In order to prevent spectral baseline distortion caused by
conduction band effects the polymer samples were prepared in
their fully reduced form. The FT-IR spectra were analysed and
integrated using the ATI Mattson WinFIRST software. Mass
spectra were recorded on a Finnigan MassLab Navigator. The
composition of the polymer ®lms was examined using the
Scienta ESCA 300 X-ray photoelectron spectrometer (typical
slit width~0.8 mm, take off angle~90³) at the CCLRC
RUSTI facility at Daresbury.
Preparation of 4-[(N-phenylsulfonyl)pyrrol-3-yl)butyric
acid, (ii). To a solution of (i) (6.68 g; 21.74 mmol) in toluene
(100 ml) and H2O (20 ml), freshly prepared amalgamated zinc
was added. [Amalgamated zinc was prepared by stirring a
mixture of mercuric chloride (5.02 g; 18.49 mmol) and zinc
powder (48.40 g; 740 mmol) in HCl conc. (20 ml) and H2O
(100 ml) for 20 min.] The mixture was stirred for 20 min, after
which more HCl (40 ml) was added. This mixture was then held
at re¯ux for 20 h, during which 3 additional portions of HCl
conc. (3625 ml) were added. After cooling to RT the aqueous
fraction was extracted with diethyl ether (3650 ml), after
which the combined organic fractions were washed with brine,
then with H2O, dried over MgSO4, ®ltered and concentrated.
The remaining orange solid was recrystallised from toluene
which gave (ii) as a white solid, 2.99 g (10.22 mmol; 47%); mp
91±93 ³C.
2.2 Monomer synthesis
Monomers 1 and 8 were obtained commercially (Aldrich) and
used as supplied. Synthesis and characterisation of monomers
2, 3, 4, and 5 (Fig. 1) have been described by us elsewhere.8,14
To our best knowledge compounds 6 and 7 are new compounds
and their preparation is described below.
2.2.1 Preparation of penta¯uorophenyl 1H-pyrrole-3-buty-
rate, 6. Monomer 6 was prepared in four stages from N-
phenylsulfonylpyrrole, the synthesis of which has been
Fig. 1 Peak anodic potentials for the irreversible oxidation of species above, EpOx, were recorded relative to the ferrocinium/ferrocene couple as an
internal calibrant.
1786
J. Mater. Chem., 2000, 10, 1785±1793

View Article Online
Preparation of 1H-pyrrole-3-butyric acid, (iii). A solution of
(ii) (1.8 g; 6.1 mmol) in 25 ml of methanol was cooled to
0 ³C. Then 5 M NaOH (17 ml) solution was added dropwise
after which the yellow reaction mixture was held at re¯ux for
5 h. The methanol was removed and the green aqueous solution
was cooled in an ice±H2O bath. This solution was acidi®ed with
HCl conc. to pH 2 and the product extracted with diethyl ether
(3625 ml). Drying of the combined organic phases over
Na2SO4, ®ltration and concentration gave a white solid.
Recrystallisation from toluene gave (iii) as white crystals,
0.783 g (5.18 mmol; 85%); mp 93±94 ³C.
CHCl3 (3650 ml). After drying over MgSO4 and concentrat-
ing, an orange oil remained which was then distilled. 3-
Aminobutyronitrile was obtained as a colourless oil, 2.25 g
(0.0267 mol; 46%).
IR(®lm): n~2936±2871 (C±H aliph.), 2245 (CMN) cm21. 1H-
NMR (250 MHz; CDCl3): d~1.72 (quintet, 2H, -CH2-CH2-
CH2-), 2.41 (t, 2H, Br-CH2-), 2.81 (t, 2H, -CH2-CN).
Preparation of N-(3-cyanopropyl)-2-pyrrol-3-ylacetamide,
7. In the ®nal step of this synthesis, penta¯uorophenyl 1H-
pyrrole-3-acetate (0.8 g; 2.75 mmol) and 3-aminobutyronitrile,
(vi) (0.271 g; 3 mmol) were dissolved in THF (30 ml). This
solution was then held at re¯ux for 18 h after which it was
allowed to cool to RT. A white precipitate appeared, which was
®ltered off. The solvent was subsequently removed to yield a
light yellow oil. Column chromatography (SiO2, ethyl acetate±
hexane (1 : 1)) gave N-(3-cyanopropyl)-2-pyrrol-3-ylacetamide,
7, as colourless crystals, 0.514 g (2.69 mmol; 98%); mp 49±
50 ³C.
Preparation of penta¯uorophenyl 1H-pyrrole-3-butyrate,
6. Monomer (iii) (1.12 g; 7.39 mmol), dicyclohexylcarbodii-
mide (DCC) (1.51 g; 7.39 mmol) and penta¯uorophenol
(1.36 g; 7.39 mmol) were dissolved in 35 ml of MeCN. After
stirring for several minutes a dense white precipitate of
dicyclohexylurea (DCU) appeared. The yellow mixture was
stirred at RT for a further 18 h. The precipitate was ®ltered off
and the solvent removed under vacuum to yield a yellow oil.
Column chromatography (SiO2, CH2Cl2± hexane (1 : 1) gave 6
as colourless crystals, 2.12 g (6.64 mmol; 90%); mp 38 ³C.
IR(KBr): n~3368 (N±H), 3140±3110 (C±H arom.), 2965±
IR(KBr): n~3584±3240 (N±H), 3084 (C±H arom.), 2935±
1
2882 (C±H aliph.), 2247 (CMN), 1651 (CLO) cm21. H-NMR
(250 MHz; CDCl3): d~1.81 (quintet, J~13.7 Hz, 2H, CH2-
CH2-CH2), 2.32 (t, J~7.3 Hz, 2H,-CH2-CMN), 3.32 (q,
J~12.8 Hz, J~6.4 Hz, 2H, NH-CH2), 3.47 (s, 2H, CH2-
CO), 6.09 (br s, 1H, NH), 6.11 (dd, J4,2~2.4 Hz, J4,5~4.0 Hz,
1H, H-4), 6,76 (dd, J5,4~4.0 Hz, J5.2~4.9 Hz, 1H, H-5), 6.81
(dd, J2,4~2.4 Hz, J2,5~4.9 Hz, 1H, H-2), 8.32 (br s, 1H, N-H).
13C-NMR (250 MHz; CDCl3): d~14.77 (CH2-CH2-CH2),
25.58 (CH2-CH2-CMN), 35.14 (CH2-CO), 38.25 (NH-CH2),
109.28 (CMN-), 116.13, 117.34 (C-2,5), 119.17, 119.29 (C-3,4),
172.94 (CLO).
2909 (C±H aliph.), 1789 (CLO) cm21
.
1H-NMR (250 MHz;
CDCl3): d~2.06 (quintet, J~14.9 Hz, 2H, -CH2-CH2-CH2),
2.60, 2.72 (2t, J~7.3, 7.6 Hz, 4H, -CH2-CH2-CH2-), 6.11 (dd,
J
4,5~4.0 Hz, J4,2~2.4 Hz, 1H, H-4), 6.61 (dd, J2,4~2.4 Hz,
2,5~5.2 Hz, 1H, H-2), 6.75 (dd, J5,4~4.0 Hz, J5,2~5.2 Hz,
J
1H, H-5), 8.16 (br s, 1H, NH).
2.2.2 Preparation of N-(3-cyanopropyl)-2-pyrrol-3-ylaceta-
mide, 7. Monomer 7 was prepared by nucleophilic substitution
of penta¯uorophenyl 1H-pyrrole-3-acetate (the preparation of
which is described elsewhere17,18) with 3-amino-n-butyronitrile.
The latter was prepared by reductive cleavage of 3-phthalimi-
dobutyronitrile.
2.3 Electrochemical methods
Copolymer materials were polymerised under potentiostatic
control from binary monomer mixtures on Au coated glass
electrodes (ca. 1 cm2) from acetonitrile solutions using 0.1 M
[NBu4][ClO4]or 0.1 M [NBu4][BF4] as a supporting electrolyte.
The concentration of the minority component of the mixture
(typically the pyrrole) was in the range 5±20 mM whilst that of
the majority component was scaled according to the speci®c
mol ratio (see Results and discussion). The polymerisation
potential was set in the range 200±250 mV positive of the foot
of the wave for the monomer with the most anodic process.
Thus it was ensured that the polymerisation reaction proceeded
under diffusion control. All solutions were thoroughly
degassed with Ar prior to polymerisation. Subsequent to
polymer growth, samples were examined using absorbance±
re¯ectance FT-IR and cyclic voltammetry. All electrochemical
procedures were carried out using an E.G. & G. M263A
potentiostat/galvanostat driven by the M270 software package.
Preparation of 3-phthalimidobutyronitrile, (v). A solution of
potassium phthalimide (30 g; 0.16 mol) in 3-bromobutyroni-
trile (14.78 ml; 0.148 mol) and dry ethanol (100 ml) was heated
to a gentle re¯ux for 18 h. The ethanol (ca. 50 ml) was distilled
off and the residual liquid was poured into dist. H2O (400 ml).
An oil separated and solidi®ed leaving a white solid which was
collected and recrystallised from dry methanol. Subsequently,
(v) was obtained as white crystals, 26.3 g (0.123 mol; 83%); mp
64±65 ³C (lit.,19 65±66 ³C).
IR(KBr): n~3060 (C±H arom.), 2947±2922 (C±H aliph.),
.
2243 (CMN), 1695±1615 (CLO) cm21 1H-NMR (250 MHz;
CDCl3): d~2.06 (quintet, 2H, -CH2-CH2-CN), 2.42 (t, 2H,
CH2-CN), 3.81 (t, 2H, N-CH2), 7.69±7.76 (m, 2H, H-5, H-6),
7.81±7.89 (m, 2H, H-4, H-7,).
Preparation of 3-aminobutyronitrile, (vi). A mixture of (v)
(12.4 g; 0.0579 mol) and 100 ml of dry ethanol was heated to a
gentle re¯ux until all was dissolved. Hydrazine hydrate
(5.95 ml; 0.123 mol) and water (5.95 ml) were added and the
mixture was stirred at RT for 15 min until a white precipitate
appeared. After adding a further 50 ml of dist. H2O and 50 ml
of ethanol the white precipitate dissolved and conc. HCl was
then added dropwise until pH 3.5. This mixture was then held
at re¯ux for 3 h. A white precipitate was ®ltered off and the
residue of the solution was concentrated down to a small
volume. This was cooled to 0 ³C and 50 ml of 5 M NaOH was
added dropwise. The white slurry was extracted with CHCl3
(3675 ml). The combined organic fractions were dried over
MgSO4 and concentrated. A yellow oil was obtained and this
was dissolved in 75 ml of Et2O. Anhydrous HCl gas was then
admitted and a white precipitate appeared ,which was isolated
and dried. Finally 40 ml of a 10 M NaOH solution was added
to the dry solid and the aqueous layer was extracted with
3.0 Results and discussion
The experiments described in this manuscript comprise ®rstly
anodic electrochemical (potentiostatic) growth of copolymer
(on a gold coated glass slide in 0.1 M [NBu4][ClO4] or
[NBu4][BF4] in MeCN) from a solution containing a binary
mixture of monomers, A and B, followed by analysis of the
copolymer ®lms using absorbance±re¯ectance FT-IR and X-
ray photoelectron spectroscopies. A number of polymers were
grown for each binary mixture covering a wide range of
solution compositions. The polymer materials were charac-
terised in terms of the ratio of A and B units within the material
by integrating the relative intensities of the S(2p) and F(1s)
regions of the X-ray photoelectron spectrum unique to each
monomer. It is an important feature of these experiments that
the oxidation potential for all the copolymerisation reactions
was chosen so that both monomers of the mixture were reacting
at the electrode under diffusion controlled conditions. This was
J. Mater. Chem., 2000, 10, 1785±1793
1787

View Article Online
achieved by poising the applied potential at a value at least
200 mV positive of the most anodic peak potential. Conse-
quently the composition of the copolymer materials was
determined by the relative reactivities of the oxidised monomer
species rather than by the difference in their redox potentials.
For each set of experiments, homopolymers of both species, A
and B, were separately prepared using the value of applied
potential corresponding to that employed for the preparation
of the copolymer samples. In each case facile homopolymer
growth was observed indicating that the conditions for
copolymer formation were not suf®cient to cause overoxida-
tion of the species with the least anodic redox potential. In
these experiments it is certainly the case that the intrinsic
reactivity of the radical species is not the only in¯uence that
governs how rapidly that species is incorporated into a
copolymer ®lm. For example solubility of small oligomers or
larger species will affect the composition of the solid. In this
work we have not sought to characterise the soluble products
of oxidation, rather we have applied the term reactivity to a
solution species in respect of the propensity for that species to
be incorporated into a solid adherent polymer ®lm.
The thiophene and pyrrole species selected for this study are
shown in Fig. 1, and represent a sample of novel tricyclic mixed
derivatives, 3 and 4, functionalised pyrrole monomers, 5, 6, 7, 8
and commercially available thiophene systems 1 and 2. The
peak potentials for the irreversible anodic oxidations are also
presented for completeness. In particular, the penta¯uorophe-
noxy group of 5 and 6 whilst facilitating subsequent functional
group chemistry at the carbonyl group of these molecules,14,18
also acts as a spectroscopic label for XPS determination. This is
because F has a high sensitivity for XPS measurements.20
Additionally, the nitrile and carbonyl stretching bands of
species 3±8 provide scope for qualitative comparison by
observation of the relative intensity of these bands in
copolymers fabricated from solution mixtures at different
compositions.
Fig. 2 XPS, F(1s) region of series of copolymers grown from solution
mixtures containing A~1 and B~5 in ratios (A : B); 100 : 1, 75 : 1,
50 : 1, 25 : 1, 15 : 1, and 5 : 1. Thin ®lm sample (ca. 0.5 mm) supported on
a Au coated SiO2 electrode recorded using a slit width of 0.8 mm and a
take off angle of 90³.
3.1 Analysis of copolymers using XPS
Fig. 2 shows a series of X-ray photoelectron spectra in the F(1s)
region for a range of copolymers grown from solution mixtures
of monomers A~1 and B~5, Table 1. This series of spectra
shows two important features. Firstly, the presence of the F(1s)
line in the spectrum is evidence of the presence of B in the
polymer ®lm, as is the presence of a N(1s) signal in the lower
binding energy region. Second, the relative amount of this unit
in the copolymer, evidenced by the intensity of the F(1s) signal,
decreases as the proportion of the other component, A
(monomer 1, 2,2'-bithiophene) in the feed solution is
increased.{ This behaviour is mirrored by the intensities of
the corresponding signals in the S(2p) spectral region for the
same polymer samples, Fig. 3, i.e. the S(2p) intensity increases
with the proportion of bithiophene, A~1, in the ®lm. Since B
(monomer 5) contains no sulfur atoms and, similarly, A
(monomer 1, 2,2'-bithiophene) contains no ¯uorine atoms,
quantitative elemental analysis of the copolymer ®lms is
facilitated by the weighted ratio of the two spectral regions
(taking the appropriate elemental spectral sensitivity factors
into account). The composition of these copolymers, in terms
of their A and B components, calculated thus are presented in
Table 1. From these data it is clear that the bithiophene, 1, is
incorporated into the ®lm during electropolymerisation, much
more slowly than is the pyrrole ester, 5. For example the
polymer that was produced from a solution containing 25 mol
of bithiophene, 1, to 1 mol pyrrole ester, 5, contains only 1.8
units of bithiophene for every pyrrole moiety, Table 1. This is
not surprising since, to our knowledge, there are very few
examples of copolymers grown from pyrrole±thiophene
mixtures that proportionately re¯ect the composition of the
feedstock solution. In one such case the powder material was
prepared but only using FeCl3 as a chemical oxidant.21
The data presented in Table 1, and in subsequent Fig. 4 and
5, are characterised by a reasonably good linear correlation
between the solution composition and the resultant polymer
composition. Consequently the composition of a given polymer
sample can be determined from a knowledge of the solution
composition and the linear scaling factor that is related to the
relative reactivities of the two monomers i.e. the gradient. For
example if the two components of the mixture were of similar
reactivity then the composition of the polymer would be the
Table 1 Correlation of solution and copolymer composition expressed
as mol ratio
Mol ratio, A : B, of monomers
in feed solution A~1, B~5
Mol ratio, A : B, of components
in the copolymer
100 : 1
75 : 1
50 : 1
25 : 1
15 : 1
5 : 1
7.1
4.1
2.3
1.8
1.1
0.5
{Best polymer growth for this experiment was achieved using
[NBu4][BF4] as supporting electrolyte. Consequently the F(1s) spectra
in Fig. 2 show a low binding energy feature corresponding to small
amounts of BF4 ion. These features were excluded from our
quantitative analysis by deconvolution from the main F(1s) signal
using the Scienta WinEsca spectral manipulation software.
2
1788
J. Mater. Chem., 2000, 10, 1785±1793

View Article Online
Fig. 3 XPS, S(2p) region of series of copolymers (same samples as
Fig. 2) grown from solution mixtures containing A~1 and B~5 in
ratios (A : B); 100 : 1, 75 : 1, 50 : 1, 25 : 1, 15 : 1, and 5 : 1. Thin ®lm
sample (ca. 0.5 mm) supported on a Au coated SiO2 electrode recorded
using a slit width of 0.8 mm and a take off angle of 90³.
Fig. 4 Composition of a range copolymers (determined from integra-
tion of the F(1s) and S(2p) regions of the XPS spectra), in terms of
component ratio (A~1, 2, or 3: B~6), versus composition of
polymerisation solution. The gradients of the linear regressions are 1
same as that of the solution from which the polymer was
formed and so the corresponding gradient would be unity.
However, the gradients of the plots in Fig. 4 and 5 show very
disparate reactivities for the species investigated here, e.g.
bithiophene 1 is very much less reactive than the pyrrole ester 5,
under these reaction conditions.
$~13.261023, 2 ~22661023 and 3 +~11561023
.
presented in Fig. 5. In these experiments the pyrrole compo-
nent of the mixture was monomer 5, in combination with
species 1, 2, 3 or 4. The linear correlations for this data set
whilst less good than those in Fig. 4, nevertheless show a clear
trend of reactivities. For the experiment where copolymers
were grown from a solution containing a mixture of 2 and 5
[Fig. 5, (&)] correlation has resulted in a line of linear
regression (dotted line) that deviates markedly from the real
origin. An intercept in these plots has no physical signi®cance,
as a solution ratio, A : B, of zero represents pure B, where in
this example B~5. Consequently the observed intercept in
Fig. 5 is likely to be due to experimental error in the polymer
analysis.§
The gradients from Fig. 4 and Fig. 5 are collected together in
Table 2. From these gradients the order of relative reactivity is
mutually consistent and is summarised as follows; 2w4w3w1,
where terthiophene is the most reactive and bithiophene is the
least reactive. This is consistent with the observations of other
groups in that terthiophenes are generally regarded as more
reactive than bithiophenes which are, in turn, more reactive
The results from a series of experiments using feed solutions
with different binary combinations are presented in Fig. 4. In
these experiments the pyrrole component of the mixture was
monomer 6, in combination with three different thiophene
species 1, 2, or 3. The gradients for the linear correlations of
these data are presented in Table 2. Apparent from this plot is
that the thiophene component is in all cases very much less
reactive than the pyrrole component. For example, the
copolymer grown from a solution containing a 100 : 1 mol
ratio mixture of molecules 1 and 6, Fig. 4 ($), consists of a
mixture of the same units in a 2 : 1 ratio. In other words there is
proportionately 30 times more pyrrole in the copolymer than
there was in solution. However, the copolymers fabricated
from solution mixtures containing molecules 2 and 6, Fig. 4
(&), show a much greater proportion of thiophene units in the
copolymer at similar feed solution ratios indicating that
terthiophene, 2, is more reactive than bithiophene, 1, under
these conditions. Intermediate to the behaviour of the species 1
and 2 is that of species 3, Fig. 4 (+). This novel functionalised
terthiophene is evidently less reactive than unsubstituted
terthiophene presumably because of steric effects resulting
from the cyanoalkyl substituent or because its oligomers are
more soluble.
§Constraining the linear regression of these data to a real origin
nevertheless results in a signi®cantly increased gradient and so the value
that appears on Fig. 6, and in Table 2 represents a minimum estimate.
The additional error introduced by this constraint, however, does not
alter the order of reactivity of the molecules in this series.
The results from another series of similar experiments are
J. Mater. Chem., 2000, 10, 1785±1793
1789

View Article Online
Table 2 Summary of gradient data from the linear correlations in
Fig. 4 and Fig. 5
Monomer A
Monomer B~5
Gradient 6103 (Fig. 5)
Monomer B~6
Gradient 6103 (Fig. 4)
1
2
3
4
63
322
150
216
13.2
226
115
Ð
another dissimilar species. For example we can consider an
imaginary binary mixture of species A and B, in equimolar
proportions where the species have oxidation potentials of
z0.5 V and z1.0 V respectively. If the copolymerisation
experiments were carried out at an applied potential of
z1.05 V then we might reasonably expect to obtain a
copolymer that contains more A than B. This is not necessarily
because of the relative intrinsic reactivities of the A and B
radical cations (about which we know nothing), but simply
because the overpotential for the generation of Az
?
is much
larger than that for generation of Bz , i.e. the concentration
?
[Az ] at the electrode surface is much larger than the
?
concentration [Bz ]. Consequently, the extent to which each
?
monomer, of the binary mixture, is incorporated into a
copolymer is unlikely to correlate with the oxidation potential
unless one or both of the reactions is under kinetic (potential)
control. In this case the composition of copolymer is entirely
dependent on the arbitrary choice of applied electrode
potential. Our experiments speci®cally exclude this condition
because, by imposing mass transport control, the reactivity of
both solution species with the electrode surface is very rapid
hence the composition of polymer is independent of applied
potential.
These arguments are illustrated in this study by considera-
tion of the combinations of 3 or 4 with 5, Fig. 5. In the ®rst
case, (3z5), the oxidation potentials are very similar, whilst in
the latter case, (4z5), the thiophene system has a redox
potential 0.351 V less anodic than the pyrrole species, 5, with
which it is combined. In both cases the pyrrole component is
shown to be more reactive, where, presumably, the solubility of
oligomeric species is very similar.
Fig. 5 Composition of a range of copolymers (determined from
integration of the F(1s) and S(2p) regions of the XPS spectra), in
terms of component ratio (A~1, 2, 3, or 4: B~5), versus composition
of polymerisation solution. The gradients of the linear regressions are 1
Finally, in this analysis, we can make quantitative compar-
isons of the relative reactivities of the thiophene systems in the
two experimental situations shown in Fig. 4 and Fig. 5. For
example the ratio of gradients for the bithiophene, 1, and
terthiophene, 2, in combination with b-pyrrole ester 6, Fig. 4,
reveals that terthiophene is w17 times more reactive than
bithiophene. Similarly terthiophene, 2, is just under twice as
reactive as the substituted terthiophene, 3, under the same
conditions. In contrast, terthiophene is only ®ve times more
reactive than bithiophene when in combination with the N-
pyrrole ester, 5, Fig. 5. Consequently, although the b-pyrrole
ester, 6, is more likely to react with itself when in combination
with any of the thiophenes of the binary mixtures in this study,
nevertheless, its reactions with these thiophenes are more
selective with respect to ®lm formation, than are those of the N-
substituted pyrrole ester 5. It is also interesting to note the
effect of substituting the S atom of 3 for O in the dithienylfuran
derivative, 4. Presumably the greater reactivity of 4 compared
to that of 3 is entirely electronic in origin since the steric
constraints and solubility for both molecules are very similar.
$~6361023
,
2
&~32261023
,
3
+~150(80)61023 and
4
,~21661023. [N.B. The data set labeled $ corresponds to that
presented in Table 1].
than thiophene monomers towards ®lm formation.2 Further-
more, the effect of the cyanoalkyl substituent in 3 and 4 is likely
to be steric as is observed in other pyrrole and thiophene
systems.7,12 In addition, comparison of like data sets from
Fig. 4 and Fig. 5, for example; {A~1zB~6}, Fig. 4 ($), with
{A~1zB~5}, Fig. 5 ($), or, alternatively, {A~2zB~6},
Fig. 4 (&), with {A~2zB~5}, Fig. 5 (&), reveal that the
gradients are consistently and considerably larger in the
experiments where the pyrrole component, B, is the N-
substituted pyrrole ester 5. Since a larger gradient is indicative
of a larger thiophene content we can conclude that the b-
substituted pyrrole ester, 6, is more reactive in combination
with these thiophene systems than is the N-substituted pyrrole
ester 5. Consequently the overall reactivity order for this series
of compounds with respect to the formation of coherent,
adhesive ®lms is as follows; 6w5&2w4w3w1. Interestingly
this reactivity order bears little resemblance to the order of
peak redox potentials, presented in Fig. 1, for these com-
pounds. This is because the redox potential of the molecular
species provides information only about how that species reacts
with the electrode surface. It does not provide information
about the rate of reaction between one radical cation and
3.2 Analysis of copolymers using re¯ectance FT-IR
Copolymer materials grown from binary solution mixtures, in
the manner described above, were also examined using
absorbance±re¯ectance FT-IR spectroscopy. In this experi-
ment the incident IR beam was oriented at 45³ from the plane
of the polymer coated electrode. The polymers were grown on
gold coated glass slides such that the IR beam path length, l,
1790
J. Mater. Chem., 2000, 10, 1785±1793

View Article Online
Table 3 Summary of FT-IR intensity data and solution composition
Integrated intensity ratio, n(CMN)/n(C~O)ester, from polymer spectra
Mol ratio, (A : B), of monomers
in solution
A~7, B~5 (CMN)/
(C~O)ester
A~7, B~6 (CMN)/
(C~O)ester
A~8, B~5 (CMN)/
(C~O)ester
A~8, B~6 (CMN)/
(C~O)ester
30 : 1
25 : 1
20 : 1
18 : 1
16 : 1
15 : 1
14 : 1
12 : 1
10 : 1
8 : 1
5.36
2.38
1.43
1.61
0.90
Ð
1.21
0.91
Ð
0.90
Ð
0.23
0.13
0.11
0.67
0.48
0.24
0.26
Ð
0.17
Ð
0.17
0.10
Ð
Ð
Ð
10.0
10.0
Ð
5.00
Ð
Ð
Ð
Ð
1.11
Ð
0.67
0.29
Ð
0.11
Ð
0.08
Ð
0.04
Ð
Ð
5 : 1
4 : 1
2 : 1
1 : 1
0.09
Ð
Ð
0.53
0.15
0.06
Ð
Ð
Ð
Ð
through the polymer is de®ned as l~2d/sin(45³)~2.83d where
d is the ®lm thickness.
at various solution compositions are presented in Table 3 and
Fig. 6. Although these data do not provide a quantitative
determination of the copolymer composition they nevertheless
clearly show that under these experimental conditions the b-
pyrrole ester, 6, is incorporated into the copolymer more
rapidly than is the N-substituted pyrrole derivative, 5. Whilst
this result is in agreement with our ®ndings from the XPS
experiments the linear regressions for these data are relatively
poor because of intrinsic error.
Monomers, 5, 6, 7 and 8, were selected for these experiments
because they possess clearly de®ned, well spaced, stretching
bands ascribed to the carbonyl and nitrile functionalities. For
each copolymer sample the integrated intensities of the relevant
carbonyl and nitrile stretching bands were ratioed and
compared with the composition of the solution from which
the sample was prepared. The data for a series of experiments
In an attempt to quantify these errors a second series of
experiments was undertaken using monomers 6 and 7. A series
of copolymers prepared from these two monomers was grown
Fig. 7 Absorbance±re¯ectance FT-IR spectra for copolymer ®lms
grown on a Au coated glass slide from solution mixtures of monomers
A~7 and B~6 in the ratios 10 : 1, 15 : 1 and 25 : 1 respectively. The
spectra were recorded at a resolution of 4 cm21 and have been baseline
corrected, normalised and offset to facilitate overlay comparison.
Fig. 6 Integrated intensity ratio of the (CO)ester and (CN) stretching
bands from the FT-IR spectra of the copolymer materials plotted
against the composition of the polymerisation solution, where A~8
and B~5 or 6.
J. Mater. Chem., 2000, 10, 1785±1793
1791

View Article Online
and three representative copolymer spectra are shown in Fig. 7.
Intrinsic to these spectra are the nitrile and carbonyl (amide)
stretches of 7 at 2246 cm21 and 1642 cm21 respectively, and the
carbonyl (ester) stretch of 6, at 1782 cm21. Furthermore, as the
proportion of monomer 7 in the solution is increased from
10 : 1 to 25 : 1, the concomitant increase in the relative intensity
of the nitrile stretch compared with the carbonyl (ester) stretch
in the copolymer spectra is clearly visible. The intensity ratio
data (with appropriate error bars) and solution compositions
for the full series are plotted in Fig. 8. The error bars presented
in Fig. 8 represent the variation in the repeat determination of
the same ratio from a given spectrum. The large error
associated with the data point at the solution ratio 30 : 1 is
probably caused by high noise levels in this spectrum.
The dual functionality of 7 allows internal calibration of
these data since the intensity ratio of the nitrile stretch and the
carbonyl (amide) stretch should be constant regardless of the
polymer composition. This calibration ratio was determined
from the spectrum of each copolymer in this series and the data
are plotted in the insert to Fig. 8. Whilst the trend of the main
graph in Fig. 8 is intuitive the insert shows that the calibration
ratio varies by almost a factor of three. This result, whilst
precluding any further detailed analysis of the FT-IR data
gathered here, may re¯ect the presence of some degree of order
within the copolymer ®lms such that the discrete IR
chromophores in the polymer lie within regions at different
orientations with respect to each other and with respect to the
incident IR radiation. Alternatively, variation in levels or
aerobic oxidation in samples and between samples might
contribute to small changes of relative intensity between
adjacent bands as has been observed in other systems.22
4.0 Conclusions
In this study we have used FT-IR and XPS techniques to
establish qualitative and quantitative data regarding the
relative ®lm forming reactivities of thiophene and pyrrole
species in binary combinations. Our observations are largely
consistent with those of others in that the thiophene species are
generally incorporated into copolymers in lower proportions
than are the pyrrole species (under diffusion controlled
electrochemical synthesis) and also the terthiophene derivatives
are more reactive than their lower order homologues. In
particular the XPS data have allowed the elucidation of a
quantitative reactivity order of the monomers in this study and
the linear correlation between solution composition and
copolymer composition enables prescriptive synthesis of
copolymer materials of a given composition from a given
solution mixture. This is important because prescriptive
determination of function, composition and properties such
as conductivity are essential for device applications. In
contrast, the FT-IR data whilst providing qualitative corro-
boration of the XPS analysis and conformation of functional
group content, do not facilitate a detailed analysis of these
materials. This is because internal calibration of the polymer
spectra shows the intensity ratio data to be highly variable and
therefore unreliable. Although the errors associated with
integrated intensity measurements from noisy spectra were
quite substantial the latter problem may be more closely
associated with the internal microstructure of the copolymer
materials implying regions or localised order or phase
separation. In addition, variations in aerobic dopant levels
might also contribute to these errors.
Acknowledgements
The authors are gratefully indebted to Dr D. G. Morris
(University of Glasgow) and to Dr G. Beamson (RUSTI,
Daresbury Lab.) for insight and helpful discussion, also to the
University of Aberdeen, the Royal Society (RSRG17906), and
the EPSRC (GR/L10185), EPSRC (CCLRC RUSTI RG168,
RG213) for ®nancial support.
References
1
R. H. Friend, R. W. Gymer, A. B. Holmes, J. H. Burroughes,
R. N. Marks, C. Taliani, D. D. C. Bradley, D. A. Dos Santos,
J. L. BreÂdas, M. LoÈglund and W. R. Salaneck, Nature (London),
1999, 397, 121.
2
Handbook of Conducting Polymers, 2nd edn., ed. T. A. Skotheim,
R. L. Elsenbaumer and J. R. Reynolds, Marcel Dekker, New
York, 1998.
3
4
5
6
G. Zotti, S. Martina, G. Wegner and A. D. Schluter, Adv. Mater.,
1992, 4, 798.
K. K. Kanazawa, A. F. Diaz, M. F. Diaz, M. T. Krounbi and
G. B. Street, Synth. Met., 1981, 4, 119.
O. Inganas, B. Liedberg, W. R. Wu and H. Wynberg, Synth. Met.,
1985, 11, 239.
Y. Chen, C. T. Imrie, J. M. Cooper, A. Glidle, D. G. Morris and
K. S. Ryder, Polymer Int., 1998, 47, 43.
Fig. 8 Main graph: Integrated intensity ratio of the (CO)ester and (CN)
stretching bands from the FT-IR spectra of the copolymer materials
plotted against the composition of the polymerisation solution, where
A~7 and B~6. Insert: Integrated intensity ratio of the (CO)amide and
(CN) stretching bands (both due to component 7 from the FT-IR
spectra of the copolymer materials plotted against the composition of
the polymerisation solution as above.
7
8
J. P. Ferraris and M. D. Newton, Polymer, 1992, 32, 391.
L. Schweiger, A. Glidle, D. G. Morris, J. M. Cooper and
K. S. Ryder, J. Mater. Chem., 2000, 10(1), 107.
M. V. Rosenthal, T. A. Skotheim, C. Linkous and M. I. Florit,
Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.), 1984, 25,
258.
9
1792
J. Mater. Chem., 2000, 10, 1785±1793

View Article Online
10 J. R. Reynolds, P. A. Poropatic and R. L. Toyooka, Synth. Met.,
1987, 18, 95.
16 M. Kakushima, P. Hamel, R. Frenette and J. Rokache, J. Org.
Chem., 1983, 48, 3214.
11 J. R. Reynolds, P. A. Poropatic and R. L. Toyooka, Macro-
molecules, 1987, 20, 958.
17 K. S. Ryder, D. G. Morris and J. M. Cooper, Langmuir, 1996,
12(23), 5681.
12 J. P. Ferraris and D. J. Guerrero, in Handbook of Conducting
Polymers, 2nd edn., ed. T. A. Skotheim, R. L. Elsenbaumer and
J. R. Reynolds, Marcel Dekker, New York, 1998.
13 W. Schuhmann, C. Kranz, H. Wohlschlager and J. Strohmeier,
Biosens. Bioelectron., 1997, 12(12), 1157.
14 C. J. Pickett and K. S. Ryder, J. Chem. Soc., Dalton Trans., 1994,
2128.
15 J. Rokach, P. Hamel and M. Kakushima, Tetrahedron Lett., 1981,
22(49), 4901.
18 K. S. Ryder, D. G. Morris and J. M. Cooper, J. Chem. Soc., Chem.
Commun., 1995, 1471.
19 A. A. Goldberg and W. Kelly, J. Chem. Soc, 1947, 1369.
20 G. Beamson and D. Briggs, High Resolution XPS of Organic
Polymers. The SCIENTA 300 Database, Wiley, New York,
1992.
21 S.-A. Chen and J.-M. Ni, Macromolecules, 1993, 26, 3230.
22 P. A. Christensen, A. Hamnett, A. R. Hillman, M. F. Swann and
S. J. Higgins, J. Chem. Soc., Faraday Trans., 1992, 38, 595.
J. Mater. Chem., 2000, 10, 1785±1793
1793