Functionalization of polypyrroles with acids and β-diketones as complexing groups. Part 1: Electrochemical synthesis and properties

Article abstract of DOI:10.1039/b002460k

The synthesis of pyrroles functionalized by complexing groups such as acetylacetone, dibenzoylmethane or carboxylic acids is described, as well as their electrooxidation into functionalized polypyrroles. We have studied the behavior of poly(functionalized pyrrole) films in the presence of Li⁺, Ni²⁺ and Co²⁺ cations by cyclic voltammetry (CV) and infrared (IR) spectroscopy. The electrochemical response under various conditions differs with the nature of the complexing group. We have also demonstrated the possibility to electrochemically generate copolymers with pyrrole and each functionalized pyrrole. The proportion of functional groups in the copolymers was estimated by following the peak potential dependence in cyclic voltammetry and was confirmed by IR spectroscopy.

Full text of DOI:10.1039/b002460k

View Article Online / Journal Homepage / Table of Contents for this issue
Functionalization of polypyrroles with acids and b-diketones as
complexing groups. Part 1: electrochemical synthesis and properties
Stephane Roux,a,b Pierre Audebert,*c Jacques Pagettia and Maxime Rocheb
a L aboratoire de Corrosion et de T raitements de Surface, Universite de Franche-Comte,
16, Route de Gray, L a Bouloie, 25030 Besancon, France.
E-mail: stephane.roux=univ-Fcomte.fr; jacques.pagetti=univ-Fcomte.fr
b L aboratoire de Chimie et dÏElectrochimie Moleculaire, Universite de Franche-Comte,
16, Route de Gray, L a Bouloie, 25030 Besancon, France.
E-mail: maxime.roche=univ-Fcomte.fr
c L aboratoire de Photophysique et de Photochimie Supramoleculaires et Macromoleculaires
(CNRS UMR 8531), Ecole Normale Superieure de Cachan, 61, Avenue du President W ilson,
94235 Cachan, France. E-mail: audebert=ppsm.ens-cachan.fr
Received (in Montpellier, France) 27th March 2000, Accepted 3rd July 2000
First published as an Advance Article on the web 20th October 2000
The synthesis of pyrroles functionalized by complexing groups such as acetylacetone, dibenzoylmethane or
carboxylic acids is described, as well as their electrooxidation into functionalized polypyrroles. We have studied the
behavior of poly(functionalized pyrrole) Ðlms in the presence of Li`, Ni2` and Co2` cations by cyclic voltammetry
(CV) and infrared (IR) spectroscopy. The electrochemical response under various conditions di†ers with the nature
of the complexing group. We have also demonstrated the possibility to electrochemically generate copolymers with
pyrrole and each functionalized pyrrole. The proportion of functional groups in the copolymers was estimated by
following the peak potential dependence in cyclic voltammetry and was conÐrmed by IR spectroscopy.
In the past few years the preparation of conducting polymers
bearing pendant groups and having tunable properties has
aroused much interest due to their (potential) applications,
especially in the Ðeld of electrochemical sensors, but also for
electrocatalysis, ion extraction or corrosion protection.1h5
Pyrrole polymers are of particular interest because their poly-
merization is easy, either by electrochemical6,7 or chemical8,9
methods. Polypyrroles bearing crown ethers, azamacrocycles,
open chain nitrogen-containing ligands and benzo-15-crown-5
have been described by several groups10h14 and have been
shown to display tunable electrochemical properties in the
presence of, for example, lithium or cobalt ions.
We have prepared pyrrole monomers bearing, as pendant
groups linked to the pyrrole nitrogen, acetylacetone (ACAC),
dibenzoylmethane (DBM) or carboxylic acid (CA) moieties,
which are classically known as good cation complexing
groups in solution. In addition, these groups are known to act
as adhesion promoters on oxide layers.15,16 The classical elec-
tropolymerization of the functionalized pyrroles as well as the
electrochemical properties of the polymers have been studied.
The complexing properties of such polymers towards metal
ions have been investigated. The electrochemical properties of
the CA functionalized polymers have been also found to
change with the protonation state of the Ðlm. Copolymers of
the functionalized monomers with pyrrole have been pre-
pared, and the relative proportion of the functionalized species
in the Ðlm is related to that in the polymerization feed.
samples were obtained by scraping thin polymer Ðlm from an
ITO electrode (2 cm2). Ultraviolet-visible (UV-vis) spectra
were recorded on a Beckman DU 520 UV-vis spectrophotom-
eter. Nuclear magnetic resonance (NMR) spectra were record-
ed at 200 MHz for 1H and 50 MHz for 13C on a Bruker
Spectrospin AC 200 spectrometer.
Synthesis of the monomers
While the preparation of CA pyrroles through another route
has been described,17,18 we have chosen to prepare them by
hydrolysis of the nitrile, which has the advantage of operating
in basic conditions and avoids the use of acetic acid. The
monomers have been prepared by a common strategy, which
is summarized in Scheme 1. The ACAC or DBM functional-
ized pyrroles have been prepared by substitution of the corre-
sponding ACAC or DBM anion on the x-halohexylpyrroles
or N-tosyloxyhexylpyrroles. It should be noted that the sub-
stitution reaction is unfortunately not straightforward in this
latter case, due to the less favorable nucleophilic character/
basic character balance of the diketonates, and that some
elimination took place, which lowers the synthesis yields.
Despite this drawback, appreciable quantities of products
could be obtained by applying this synthetic scheme.
6-(Pyrrol-1-yl)hexanol 2 and N-(x-tosyloxyhexyl)pyrrole 3
were synthesized by the method of Bidan.19,20 The procedure
was repeated without modiÐcation and NMR and IR data
were in accordance with that reported.
N-(6-Bromohexyl)pyrrole (5a) and N-(10-bromodecyl)pyrrole
(5b). 5a: The reaction was carried out under nitrogen and in
dry conditions. Pyrrolylpotassium salt 119 (0.100 mol) was dis-
solved in 20 mL of DMSO. To a three-necked Ñask was intro-
duced 0.120 mol of 1,6-dibromohexane in 50 mL of THF.
Then 1 was slowly added (to permit heat elimination and
avoid disubstitution). After 10 min stirring, the solution was
poured into 500 mL of a saturated aqueous solution of
Experimental
Methods
Melting points were measured with an IA 9200 apparatus.
Infrared (IR) spectra were obtained between 600È4000 cm~1
(resolution 4 cm~1) on a Bruker Spectrospin IFS-45 spectro-
meter on KBr pellets of the crushed samples. The polymer
DOI: 10.1039/b002460k
New J. Chem., 2000, 24, 877È884
877
This journal is ( The Royal Society of Chemistry and the Centre National de la Recherche ScientiÐque 2000

View Article Online
OH); 49.3 (CH ÈN); 31.3, 30.3, 28.9, 27.8, 27.3, 26.4, 26.2, 22.6
2
[(CH ) and CH (enol and diketone)].
2 5
3
6DBM: Same procedure as for 6ACAC using 1,3-diphenyl-
1,3-propanedione instead of 2,4-pentanedione. Yield \ 9%.
IR: 3066 l(C ÈH); 2932 and 2864 l(C ÈH); 1687 cm~1:
sp3
sp3
l(C2O). 1H NMR (CDCl ): d 1.10È1.20 (m, 8H); 3.85 (t,
3
J \ 7.0 Hz, 2H); 5.2 (t, J \ 6.6 Hz, enol); 6.15 (t, J \ 2.1 Hz,
2H); 6.65 (t, J \ 2.0 Hz, 2H); 7.30È8.10 (m, 10H). 13C NMR
(CDCl ): d 198.6 (C2CÈC2O); 136.0 (C2CÈOH); 131.4 (C H );
3
6 5
130.0 (C H ); 122.9 (C ); 110.3 (C ); 68.6 (C2CÈOH); 52.0
6
5
a
b
(CH ÈN) ; 33.9, 31.9, 31.7, 30.6, 28.9 [(CH ) ].
2 5
2
7-(Pyrrol-1-yl)heptanenitrile (6a) and 11-(pyrrol-1-yl)-
undecanenitrile (6b). 6a: Sodium cyanide (1.080 mol) was par-
tially dissolved in 250 mL of DMSO21,22 and the solution
heated to 80 ¡C. 5a was slowly added under vigorous mecha-
nical stirring for 10È15 min. The temperature then increased
rapidly and remained between 135È145 ¡C. Once all of the 5a
had been added, the temperature quickly dropped and below
50 ¡C reaction was complete. The brown crude product was
cooled, poured into water and extracted with diethyl ether.
The yellow organic phase was washed several times with a
saturated aqueous solution of sodium sulfate and dried over
calcium chloride. The crude oil was puriÐed by chromatog-
raphy on silica gel with a solution of 5% diethyl etherÈ95%
petroleum ether as eluent. Yield \ 68È73%. IR: 3110
l(C ÈH); 2931 l(C ÈH); 2246 cm~1 l(C2N). 1H NMR
sp2
sp3
(CDCl ): d 1.30È1.85, (m, 8H); 2.35 (t, J \ 6.8 Hz, 2H); 3.90 (t,
3
J \ 7.0 Hz, 2H); 6.10 (t, J \ 2.1 Hz, 2H); 6.65 (t, J \ 2.1 Hz,
2H).
6b: Some procedure as for 6a using N-(10-bromodecyl)
pyrrole instead of N-(6-bromohexyl)pyrrole. IR: 3105
l(C ÈH); 2925 and 2820 l(C ÈH); 2245 cm~1 l(C2N). 1H
sp2
sp3
NMR (CDCl ): d 1.30È1.85 (m, 16H); 2.35 (t, J \ 6.9 Hz, 2H);
3
Scheme 1 Synthetic route for the functionalization of pyrrole by b-
diketone (6ACAC, 6DBM) and carboxylic acid (2CA, 6CA, 10CA)
groups.
3.90 (t, J \ 7.1 Hz, 2H); 6.10 (t, J \ 2.1 Hz, 2H); 6.65 (t,
J \ 2.1 Hz, 2H).
3-(Pyrrol-1-yl)propanoic acid (2CA), 7-(pyrrol-1-yl)heptanoic
acid (6CA) and 11-(pyrrol-1-yl)undecanoic acid (10CA). 2CA,
6CA and 10CA, were synthesized according to the procedure
described in the literature23 for 2CA.
sodium sulfate and was extracted several times with diethyl
ether. Solvents and pyrrole in the organic phase were removed
under reduced pressure at 80 ¡C. The crude oil was puriÐed by
chromatography on silica gel with a solution of 5% diethyl
etherÈ95% petroleum ether as eluent. Yield \ 40È45%. 1H
6CA: 6a (1.200 mol) and 350 mL of a solution of potassium
hydroxide (6.7 M) were reÑuxed until the organic phase disap-
peared (5È10 h). Water (100 mL) was added and the solution
was acidiÐed to pH 5 with hydrochloric acid (8 M). The crude
product was extracted Ðve times with diethyl ether, which was
removed under reduced pressure to give a beige powder. Re-
crystallization was driven in hot n-heptane. Yield \ 50È55%,
NMR (CDCl ): d 1.35È2.00 (m, 8H); 3.45 (t, J \ 6.7 Hz, 2H);
3
3.90 (t, J \ 7.1 Hz, 2H); 6.15 (t, J \ 2.1 Hz, 2H); 6.65 (t,
J \ 2.1 Hz, 2H).
5b: Same procedure as for 5a, using 1,10-dibromodecane
instead of 1,6-dibromohexane. 1H NMR (CDCl ): d 1.30È1.85
3
(m, 16H); 3.40 (t, J \ 6.8 Hz, 2H); 3.90 (t, J \ 7.1 Hz, 2H);
m.p. \ 60È61 ¡C. IR: 3100È2500 l(OÈH); 2860 l(C ÈH);
sp3
1700 cm~1 l(C2O). 1H NMR (CDCl ): d 1.10È1.90 (m, 8H);
3
6.15 (t, J \ 2.1 Hz, 2H); 6.65 (t, J \ 2.1 Hz, 2H).
2.35 (t, J \ 7.3 Hz, 2H); 3.90 (t, J \ 7.1 Hz, 2H); 6.15 (t,
3-[6-(Pyrrol-1-yl)hexyl]-2,4-pentanedione (6ACAC) and 1,3-
diphenyl-2-[6-(pyrrol-1-yl)hexyl]-1,3-propanedione (6DBM).
6ACAC: To a solution of 0.0287 mol of potassium tert-
butoxide in 5 mL of DMSO was Ðrst added 0.0287 mol of 2,4-
pentanedione and then 0.0143 mol of 3 or 5a. The reaction
was carried out under nitrogen and in dry conditions. The
mixture was heated at 90 ¡C for 2.5 h. The solution was then
allowed to cool before being neutralized with hydrochloric
acid (0.1 M) and the product was extracted with diethyl ether.
The organic phase was washed several times with distilled
water and dried on anhydrous sodium sulfate. The crude
product was puriÐed by chromatography on silica gel (5%
diethyl etherÈ95% petroleum ether as eluent). Yield \ 22%.
IR: 3110 l(C ÈH); 2932 and 2865 l(C ÈH); 1710 cm~1
J \ 2.1 Hz, 2H); 6.65 (t, J \ 2.1 Hz, 2H). 13C NMR (CDCl ):
3
d 179.6 (COOH); 120.3 (C ); 107.7 (C ); 49.3 (CH ÈN); 33.7
a
b
2
(CH ÈCOOH); 31.2, 28.4, 26.3, 24.3 [(CH ) ].
2
2 4
2CA: Same procedure as for 6CA using 3-(pyrrol-1-yl)
propanenitrile (commercially available) instead of 7-(pyrrol-1-
yl)heptanenitrile. m.p. \ 57È59 ¡C. IR: 3200È2400 l(OÈH);
2940 l(C ÈH); 1701 cm~1 l(C2O). 1H NMR (CDCl ): d 2.35
sp3
3
(t, J \ 6.9 Hz, 2H); 4.20 (t, J \ 6.9 Hz, 2H); 6.15 (t, J \ 2.1
Hz, 2H); 6.70 (t, J \ 2.1 Hz, 2H). 13C NMR (CDCl ): d 177.3
3
(COOH); 120.4 (C ); 108.5 (C ); 44.3 (CH ÈN); 36.2
a
b
2
(CH ÈCOOH).
2
10CA: Same procedure as for 6CA using 11-(pyrrol-1-yl)
undecanenitrile instead of 7-(pyrrol-1-yl)heptanenitrile.
m.p. \ 66È68 ¡C. IR: 2500È3000 l(OÈH); 2910 l(C ÈH);
sp3
sp3
sp3
1691 cm~1 l(C2O). 1H NMR (CDCl ): d 1.20È1.45 (m, 12H);
3
l(C2O). 1H NMR (CDCl ): d 1.20È1.90 (m, 8H); 2.1 (s, enol
3
and diketone); 2.2 (s, enol), 3.50 (t, J \ 7.2 Hz, diketone) 3.85
1.50È1.90 (m, 4H); 2.35 (t, J \ 7.2 Hz, 2H); 3.85 (t, J \ 7.2 Hz,
(t, J \ 7.1 Hz, enol); 3.90 (t, J \ 7.1 Hz, 2H); 6.15 (q, J \ 2.7
2H); 6.15 (t, J \ 2.1 Hz, 2H); 6.65 (t, J \ 2.1 Hz, 2H). 13C
NMR (CDCl ): d 180.1 (COOH); 120.3 (C ); 107.6 (C ); 49.5
Hz, 2H), 6.65 (q, J \ 2.7 Hz, 2H). 13C NMR (CDCl ): d 204.2
3
3
a
b
(C2O); 190.7 (C2CÈOH); 120.2 (C ); 107.6 (C ); 68.6 (C2CÈ
(CH ÈN); 33.9 (CH ÈCOOH); 31.4, 29, 26.6, 24.5 [(CH ) ].
a
b
2
2
2 8
878
New J. Chem., 2000, 24, 877È884

View Article Online
Table 1 Oxidation potential values (E ) for each monomer. Oxida-
Electrochemical set-up
ox
tion and reduction peak potential (Epa , E ) values of the correspond-
pc
Analytical experiments were performed in
a
three-
ing homopolymers and their doping level d
compartment cell Ðtted with a saturated calomel reference
electrode (SCE), a platinum electrode (diameter 2 mm) and a
platinum counter electrode. The electrochemical apparatus
was a Tacussel PGP 201 potentiostat (equipped with an inte-
grated pilot for low to moderate scan rates), a Tacussel IG5-N
integrator and a Sefram 164 plotter. The solvent was spectro-
scopic grade acetonitrile, distilled over phosphoric anhydride
(to remove any trace of water or amines that may act as poly-
merization scavengers). Tetraethylammonium perchlorate
(TEAP, Fluka, puriss, recrystallized once in acetonitrileÈ
diethyl ether), or lithium perchlorate (Fluka, puriss) was added
as the supporting electrolyte salt. The concentration of the
monomers was usually between 5 ] 10~3 and 10~2 mol L~1.
For Ðlm studies, two cells were used in parallel, one con-
taining a monomer solution for the Ðlm synthesis and another
containing a clean electrolyte solution, where the Ðlms were
transferred for electrochemical studies. The electrode was
carefully cleaned with acetonitrile between each experiment.
All Ðlms were prepared potentiostatically at 1.08 V (vs. SCE)
by passage of a constant charge of 0.13 C cm~2 through an
Monomer
Polymer
E /mV
E /mV
E /mV
*E /mV
da
ox
pa
pc
p
2CA
1380
1210
1255
1095
1220
655
595
595
580
590
530
555
530
535
570
125
40
65
45
20
0.142
0.136
0.153
0.150
0.108
6CA
10CA
6ACAC
6DBM
a d \ 2Q /(oQ [ Q ) with Q the doping charge (deduced from inte-
d
t
d
d
gration of the anodic zone of the voltammogram), Q the total charge
t
of the electrolysis and
o
the electropolymerization efficiency
(hypothesis: o \ 1).
in Table 1, these are usually in the 10È15% range, which is
quite low when compared to polypyrrole7 but is not too sur-
prising for an N-functionalized pyrrole.1
Behavior of the diketone functionalized pyrroles
acetonitrile solution with either LiClO or TEAP added (the
4
electrolyte salt used for electropolymerization is indicated in
The complexation power of the diketone functionalized pyrro-
les was evaluated by an UV-vis spectrophotometric study
(Table 2). First, spectra were recorded for each compound
used for this study [i.e. M(ClO ) where M represents Co(II)
or Ni(II), diketone functionalized pyrroles (6ACAC, 6DBM)
and triethylamine (Et N) in acetonitrile (0.1 M)]. Then spectra
of mixtures composed of M(ClO ) and b-diketone functional-
ized pyrrole with or without Et N in acetonitrile were
obtained. The important absorption band shift of the metal
ion indicating the reaction of cobalt(II) or nickel(II) with the
pyrrolic derivative was observed only when Et N was added
(see Table 2). It does not result from an interaction between
Et N and Co(II) or Ni(II) since the spectra of a mixture of
brackets in the text or captions). Cycling of the Ðlms was per-
formed in acetonitrile with either LiClO or TEAP added (see
4
text or captions) at a scan rate of 50 mV s~1.
4 2
To study the inÑuence of transition metal ions, a Ðlm was
electrogenerated in acetonitrile solution containing 0.1 mol
L~1 electrolyte salt (Et NClO or LiClO ). Then, CV was
3
4
4
4
4 2
carried out in a monomer-free solution with the same electro-
lyte as used for the formation of the Ðlm. When a stable CV
(i.e. Ðrst scan in Fig. 2) had been obtained (to eliminate pos-
sible memory e†ects in the polymer), the ACAC or DBM
group functionalized Ðlms were immersed or not in tri-
ethylamine, rinsed with acetonitrile and Ðnally cycled into ace-
tonitrile containing 0.1 mol L~1 transition metal perchlorate
as the supporting electrolyte.
3
3
3
3
Et N and M(ClO ) in acetonitrile displays only a weak shift
4 2
of the absorption band of the metal ion towards shorter wave-
length (i.e. in the opposite direction). Furthermore, the simi-
larities of the spectra obtained for the mixture of Co(ClO ) ,
Results and discussion
4 2
6ACAC, Et N and for Co(6ACAC) 25 conÐrm the reaction of
3
2
All the functionalized monomers readily polymerize in aceton-
itrile, although the formation of a polymer Ðlm at a usual rate
requires a concentration of ca. 5 ] 10~3È10~2 mol L~1, in the
range of what is commonly used for classical electrochemical
polythiophene synthesis24 (while polypyrrole can be electro-
polymerized at lower concentrations1,6). Typical voltam-
mograms are represented in Fig. 1. Doping levels (determined
from the integration of cyclic voltammograms) are displayed
Co(II) with 6ACAC. When no base is added, only simple com-
plexation with ACAC occurs and this does not a†ect the
UV-vis spectrum of the metal.
The behavior of the DBM functionalized Ðlms is quite
similar to that of classical N-substituted polypyrroles.26,27
However, its electrochemical behavior is only slightly or not
at all modiÐed when changing the electrolyte salt, even when
the electrolyte is a transition metal salt. This is in contrast to
Table 2 UV-vis spectrophotometric study of Co(II) and Ni(II) com-
plexation by 6ACAC
Systema
Co(II)b
Ni(II)b
M(ClO )
490(0.126)
368(0.061)
595(0.039)
Precipitates
4 2
M(ClO ) ] Et N
469(0.180)
631(0.050)
488(0.154)c
446 shc
580(0.170)
635(0.174)
589(0.608)c
685(0.675)
4 2
3
M(ClO ) ] 6ACAC
595(0.040)c
666(0.044)c
4 2
M(ClO ) ] 6ACAC ] Et N
4 2
3
M(6ACAC)
2
a Concentration of each compound: 0.1 mol L~1 in acetonitrile.
b Data are given in the form wavelength/nm (intensity). c b-Diketone
functionalized pyrroles absorbed strongly at k B 315 nm
(intensity B 3).
Fig. 1 Cyclic voltammetry of Ðlms of (a) poly(6CA) and (b)
poly(6DBM) on a platinum electrode (diameter 2 mm) in 0.1 mol L~1
LiClO ÈCH CN.
4
3
New J. Chem., 2000, 24, 877È884
879

View Article Online
the classical behavior of functionalized polypyrroles, which
are generally sensitive to a modiÐcation of the substituted
moieties, as exempliÐed by the case of biosensors.28 Although
the UV-vis study shows the ability of 6DBM to complex Ni
and Co ions, no change is observed in the case of
poly(6DBM), maybe because of the restricted mobility of the
DBM groups.
In contrast, the ACAC functionalized polypyrroles exhibit
behavior that depends on the complexation of the functional
group. An anodic peak shift is observed upon cycling in an
electrolyte containing a transition metal such as Ni2` or
Co2` (Fig. 2). Treatment with triethylamine (Et N), followed
3
by cycling in a solution containing a transition metal ion
results in the same qualitative e†ect, albeit to a lesser extent.
There is no splitting of the polymer peak, but the large anodic
shift clearly shows the structural change of the polymer. Upon
cycling again in lithium perchlorate, the CV of the polymer
Ðrst remains stable, then changes again, but does not return to
the initial state; this is probably due to mechanical changes in
the Ðlm due to the high level of incorporation of the transition
metal ions.29
This has been conÐrmed by an IR study of the Ðlms under
di†erent conditions, upon comparing the frequencies and the
relative intensities of the carbonyl and perchlorate bands in
the polymers. Table 3 summarizes the IR data on the poly-
mers, according to the electrochemical treatment to which
they were submitted (just after synthesis, after cycling in a
transition metal solution and Ðnally, Ðrst cycled in Ni(ClO )
4 2
and then cycled again in LiClO ). First of all, the frequency of
4
the ACAC carbonyl stretching band is decreased by 10È15
cm~1, compared to the monomer. This results from complex-
ation by the electrolyte cations. This is conÐrmed by the
analysis of the relative intensities of the bands. Immediately
after synthesis in LiClO , the ratio R \ I[l(C2O)]/
4
I[l(ClO)] \ 0.81, the perchlorate present acting as the doping
species of the polymer; there is probably close to one per-
chlorate for 7 to 10 pyrrole rings. Upon cycling in Co(ClO )
4 2
or Ni(ClO ) this ratio decreases to 0.12 or 0.16, respectively,
4 2
indicating thus a large increase of the amount of perchlorate
in the Ðlm. Moreover, a quantitative analysis of the above
Ðgures shows that, relying on 0.1È0.15 perchlorates per
pyrrole ring in the as-grown state (as determined from the
coulometric data), the amount of perchlorate in the Ðlms
cycled with transition metal ions is now close to 6È7 times
more, that is close to 1 perchlorate per pyrrole ring. This
shows not only that the complexation is nearly quantitative,
but also that simple complexation of the metal ions takes
place on the ACAC, without exchange of the labile ACAC
Fig. 2 Cyclic voltammetry of poly(6ACAC) Ðlm in 0.1 mol L~1
LiClO ÈCH CN (Ðrst scan) then (for the other scans) in (a) 0.1 mol
4
3
L~1 Ni(ClO ) ÈCH CN, (b) 0.1 mol L~1 Ni(ClO ) ÈCH CN after
4 2 4 2
immersion in Et N, (c) 0.1 mol L~1 Co(ClO ) ÈCH CN after immer-
4 2
sion in Et N, (d) 0.1 mol L~1 LiClO ÈCH CN. [LiClO ].
3
3
3
3
3
4
3
4
ance with the results of Richter and Bard from electrochemical
studies of europium diketonates.30 In our case, this may also
result from the non-conductive character of the Ðlm at highly
reducing potentials, which suppresses connections between the
redox centers and the electrode.
protons.25 Upon cycling (3 cycles) the Ðlms again in LiClO ,
4
R increases again to 0.31, indicating a decrease in the amount
of perchlorate, associated with the slow incomplete release of
the transition metal. In the case of the DBM modiÐed Ðlm, on
the other hand, no signiÐcant change was found between the
di†erent spectra, as could be expected. All these measurements
conÐrm, and allow us to quantify, the previously discussed
electrochemical results.
Behavior of the carboxylic group functionalized polymers (CA
polypyrroles)
However, we did not observe any reduction of the tran-
sition metal inserted in the Ðlm. This behavior is in accord-
CA polypyrroles exhibit a quite di†erent behavior depending
on the electrolyte medium. Newly electropolymerized Ðlms of
Table 3 E†ect of the metal ion from the electrolytic solution on the IR data of poly(6ACAC) Ðlms, after cycling in several acetonitrile solutions
containing 0.1 mol L~1 of the metal salts
Re-cycled in LiClO
after cycling in
4
As grown
(in LiClO )
Cycled in
Co(ClO )
Cycled in
Ni(ClO )
Ni(ClO )
4
4 2
4 2
4 2
C2O
l /cm~1
1699
18.5
1085
22.8
0.81
1697
6.4
1085
53.2
0.12
1698
8.3
1084
52.0
0.16
1700
14.4
1084
46.6
0.31
1
I
1
ClÈO
l /cm~1
2
I
2
R \ I /I
1
2
880
New J. Chem., 2000, 24, 877È884

View Article Online
Fig. 3 Cyclic voltammetry of poly(6CA) Ðlms in the presence of Li` ions: (a) Ðrst scan in 0.1 mol L~1 TEAPÈCH CN, (b) successive scans in 0.1
3
mol L~1 LiClO È0.1 mol L~1 HClO ÈCH CN; (c) successive scans in 0.1 mol L~1 LiClO ÈCH CN. [TEAP].
4
4
3
4
3
poly(6CA) [Fig. 3(a)] are certainly in the acidic state due to
the excess of protons produced by the electropolymerization
reaction in the vicinity of the carboxylic groups. When the
Ðlms are cycled in an acidic electrolyte (with 0.1 mol L~1 per-
chloric acid added) very stable CVs are obtained as shown in
Fig. 3(b), very similar to the initial CV. However, when
neutral acetonitrile is used as the electrolyte, the CV progres-
sively changes as a result of the deprotonation of the carbox-
ylic acid and exchange with the electrolyte cations. This
process is especially favored when Li` is the electrolyte cation,
which is not unexpected given the acidity of this cation in
acetonitrile. This is attested to by the splitting of the initial
polymer peak and its progressive replacement by another
redox couple at a higher potential [ca. 200 mV, see Fig. 3(c)].
This has again been conÐrmed by an IR study, by compar-
ing the relative intensities of the carbonyl and the perchlorate
bands. The frequencies and relative intensities of the bands are
displayed in Table 4, similarly as in Table 3, for as-grown
Ðlms of poly(6CA) (in TEAP), Ðlms cycled in LiClO ÈHClO
of the carboxylic groups are deprotonated and the resulting
carboxylates participate in the doping process (self-doping).31
The high ratio in the third case shows that the polymer, upon
cycling in pure LiClO , is almost exclusively self-doped.
4
It should be noted that the frequency of the C2O band was
almost constant in the three cases examined, which seems con-
tradictory to the preceding interpretation, because the self-
doped polymer should display the C2O carboxylate band,
which is usually situated ca. 100 cm~1 lower than the corre-
sponding carboxylic acid.32 This can be explained by the fact,
that, in the case of the self-doped polymer, approximately
90% of the carboxylic groups are still in the acidic form,
because the doping level is only about 10% (see Table 1) and
therefore only 10% of carboxylate groups participate in the
self-doping process. Their contribution to the IR spectra is
masked by the broad carboxylic acid band. The determination
of the amount of perchlorate in the Ðlm is therefore, although
indirect, a much better indicator of the dopant nature.
The electrochemical behavior observed for the CA poly-
pyrrole Ðlms is largely independent of the length of the link
between the pyrrole and the carboxylic group, since
poly(10CA) and poly(2CA) display analogous behavior (Fig. 4)
to that of poly(6CA). Nevertheless, the e†ect on poly(2CA) is
4
4
and Ðlms cycled with LiClO without added acid. In as-grown
4
Ðlms in TEAP, the ratio R is 2.0. This ratio comes down to
0.70 when the polymer is cycled in the presence of perchloric
acid, and comes up again to 3.3 when the polymer is cycled in
pure LiClO . It can be assumed that in the presence of per-
less striking. As the alkyl chain of 2CA (2 CH ) is shorter, the
4
2
chloric acid the carboxylic groups in the polymer are totally
approach of the Li` cation is probably disturbed by electro-
protonated, therefore excluding any self-doping by the carbox-
ylate; the low ratio found is due to the large amount of per-
chlorate anions in the polymer (typically one for four pyrrole
rings, according to ref. 7). In the other situations, and espe-
static repulsion between the charge carrier (polaron or
bipolaron) in the oxidized polymer chain and the positive
charge of the Li` cation. In the case of poly(6CA) and
poly(10CA) electrostatic repulsions are likely to be much
cially when the Ðlm is cycled in the presence of LiClO , some
weaker because of the longer alkyl chains of 6CA (6 CH ) and
4
2
10CA (10 CH ) and consequently deprotonation by Li`
2
would be more efficient.
Table 4 E†ect of the metal ion from the electrolytic solution on the
IR data of poly(6CA) Ðlms, after cycling in several acetonitrile solu-
tions containing 0.1 mol L~1 each of the electrolyte salts
The behavior of the CA polypyrroles has also been studied
in water. Under such conditions, only poly(2CA) has been
found to display a signiÐcant electroactivity, since in the case
of the higher homologs, the larger hydrophobicity of the chain
impedes the presence of enough water in the polymer [as
attested by the Ñat cyclic voltammograms obtained after 3È4
cycles with poly(6CA) and poly(10CA)]. The electrochemical
response of poly(2CA) is shown on Fig. 5 at di†erent pH. It is
clear that both the shape of the voltammograms and the peak
potentials change with pH, but, even in the case of the well-
TEAP
LiClO ] HClO
LiClO
4
4
4
C2O
l /cm~1
1706
36.4
1080
18.1
1708
31.2
1083
44.8
0.70
1707
37.7
1083
11.4
1
I
1
ClÈO
l /cm~1
2
I
2
R \ I /I
2.01
3.31
1
2
New J. Chem., 2000, 24, 877È884
881

View Article Online
Fig. 6 Cyclic voltammetry of poly(6CA) Ðlms in the presence of
Ni2` or Co2` ions: Ðrst scan in 0.1 mol L~1 TEAPÈCH CN and
3
Fig. 4 Cyclic voltammetry in 0.1 mol L~1 TEAPÈCH CN (Ðrst
following scans in (a) 0.1 mol L~1 Ni(ClO ) ÈCH CN, (b) 0.1 mol
3
4 2
3
scan) then in 0.1 mol L~1 LiClO ÈCH CN (following scans) of (a)
L~1 Co(ClO ) ÈCH CN. [TEAP].
4
3
4 2
3
poly(10CA) and (b) poly(2CA) Ðlms. [TEAP].
However, CA pyrroles and b-diketone pyrroles display dif-
ferent behavior as a function of x. The potential E¡ increases
only when x is greater than 0.5 and 0.75 for 6ACAC and
deÐned reduction peak, the dependence of the peak potential
vs. pH is not linear, indicating a complex behavior.
Cycling the CA polypyrroles in an electrolyte containing
transition metal ions also a†ects the behavior of the Ðlms, as
shown by Fig. 6. It is worth noting that with CA polypyrroles,
especially in the case of Co2` cations, the shift of the anodic
peak occurs in the direction of more negative (lower) poten-
tials in contrast to what is observed with poly(6ACAC).
6DBM, respectively [Fig. 7(a)]. At lower
x
values
E¡(copolymer) is very close to E¡(polypyrrole); it is clear that
only polypyrrole is made as long as the functionalized
Copolymers
The copolymerization of all monomers with pyrrole (Py) is
feasible and results in a random copolymer, the redox poten-
tial of which varies with the composition of the Ðlms. Fig. 7
displays the calculated potential E¡ shift variation vs. x, the
relative proportion of the functionalized pyrroles (fPy) in the
polymerization feed:
E
pa ] E
n
pc
fPy
] n
E¡ \
and x \
2
n
Py
fPy
It is clear that a similar trend is followed in all cases, the peak
potentials increasing with the proportion of functionalized
pyrroles in the synthetic feed, which in turn should be related
to the proportion of functionalized species in the Ðnal electro-
generated polymer. This shows clearly that there is a modiÐ-
cation of the polymer.
Fig. 7 Electrochemical copolymerization of pyrrole (Py) and func-
tionalized pyrrole (fPy): E¡ vs. relative proportion x of (a) 6ACAC
and 6DBM; (b) 6CA, 10CA and 2CA in the polymerization feed. Elec-
trochemical synthesis of each Ðlm was performed at 1080 mV by
passing 0.13 C cm~2 through a mixture of pyrrole and functionalized
pyrrole ([Py] ] [fPy] \ 10~2 mol L~1) in 0.1 mol L~1
LiClO ÈCH CN.
Fig. 5 Cyclic voltammetry of poly(2CA) Ðlms in 0.1 mol L~1
LiClO ÈH O at pH 2, 4 and 8.
4
2
4
3
882
New J. Chem., 2000, 24, 877È884

View Article Online
monomer is not in excess relative to unsubstituted pyrrole in
the feed. The difficulty of obtaining a copolymer may be due
to the steric hindrance of the b-diketone group; indeed, for the
DBM (bulkier than ACAC) moiety, the synthetic feed must be
more concentrated in 6DBM to obtain a copolymer.
of pyrrole, and can only be ascertained by elemental analysis.
However, in our case this technique would only give poorly
reliable results, due to the very similar compositions of both
monomers.
As shown Fig. 7(b), E¡(copolymer) [ E¡(polypyrrole) for all
Conclusion
x values; the copolymerization of CA pyrrole with pyrrole
We have prepared and polymerized several pyrroles function-
alized with cation complexing or anchoring functions.
Although the electrochemical study as well as the IR data
demonstrate that complexation occurs with transition metal
ions, the efficiency of the process remains average and sensi-
tivity must be improved for use in the sensor Ðeld. The
copolymer synthesis appears to occur analogously to pre-
viously reported examples,34,35 with incorporation of the
functionalities in the copolymer being related to the propor-
tion of functionalized pyrrole in the feed. In addition, Armes
and co-workers used analogous molecules for the preparation
of polypyrrole silica colloidal nanocomposites.23 We have
therefore turned our attention towards the electrochemical
preparation of polypyrroleÈoxide composites using our func-
tionalized pyrroles.
seems
to
be
easier.
For
each
value
of
x,
E¡[poly(2CA)] [ E¡[poly(6CA)] [ E¡[poly(10CA)]; when the
alkyl chain is shorter, E¡ is higher. But for x \ 0.6, the di†er-
ence in E¡ between copolymers obtained at a given value of x
is greater than when x [ 0.6. From this observation we
deduce that, for a given x, the copolymer Ðlm is richer in 2CA
and poorer in 10CA. A plausible explanation lies in the inÑu-
ence of the length of the alkyl chain during the electro-
polymerization; 2CA, which possesses the shortest alkyl chain
of the CA series has lower steric hindrance than 10CA with
the longest alkyl chain. The weak steric hindrance of 2CA
does not prevent its insertion in the growing Ðlm even if the
synthetic feed is richer in pyrrole. Under the same conditions
the bulkiness of 10CA represents a handicap. 6CA shows an
intermediate behavior between 10CA and 2CA. It should be
noted that the contrary would have been expected if one con-
sidered only the relative order of the oxidation potentials of
pyrrole and the CA pyrroles.33
References and notes
1
2
A. Deronzier and J. C. Moutet, Acc. Chem. Res., 1989, 22, 249.
A. Deronzier and J.-C. Moutet, Coord. Chem. Rev., 1996, 147,
339.
However, it is not obvious from this data how the propor-
tion of the functionalized moieties in the Ðlm depends on the
relative amount of the monomers in the feed. An IR spectro-
scopic study allows us to conÐrm the electrochemical obser-
vations. The calculated ratio R \ I /I where I represents
3
4
5
6
7
F. Bedioui, J. Devynck and C. BiedÈCharreton, Acc. Chem. Res.,
1995, 28, 30.
P. Audebert, in T rends in Electrochemistry, ed. S. G. Pandalai,
Trivandrum, India, 1994, vol. 3, p. 459.
1 3
1
the carbonyl peak intensity [characteristic of CA (at ca. 1700
cm~1), ACAC (at 1710 cm~1) and DBM (at 1685 cm~1)
C. A. Ferreira, S. Aeiyach, J. J. Aaron and P. C. Lacaze, Electro-
chim. Acta, 1996, 41, 1801.
pyrroles] and I the pyrrolic ring stretch (at 1540 cm~1) inten-
A. F. Diaz, K. K. Kanazawa and G. P. Gardini, J. Chem. Soc.,
Chem. Commun., 1979, 635.
3
sity, increased with the proportion of functionalized mono-
Handbook of Conducting Polymers, ed. T. A. Skotheim, Marcel
Dekker, New York, 1986.
mers in the synthetic feed. Moreover, the proportion (p) of the
functionalized pyrrole moieties in the copolymer can be esti-
mated by the following relation:
8
9
J. Lei, Z. Cai and C. R. Martin, Synth. Met., 1992, 46, 53.
S. Marina, V. Enkelman, G. Wegner and A. D. Schluter, Synth.
Met., 1992, 51, 299.
I (x) ] I (x \ 1)
I (x \ 1) ] I (x)
1
3
10 B. Fabre, P. Marrec and J. Simonet, J. Electrochem. Soc., 1998,
145, 4110.
p \
1
3
11 H. KouriÈYoussouÐ, M. Hmyene, F. Garnier and D. Delabou-
glise, J. Chem. Soc., Chem. Commun., 1993, 1550.
12 P. N. Bartlett, L.-Y. Chung and P. Moore, Electrochim. Acta,
1990, 35, 1501.
The proportion (p) of functionalized pyrrole in the Ðlm (Fig.
8), estimated through this method, is again smaller than the
one in the synthetic feed as previous studies on pyrrole
copolymers tend to show.34,35 These results are very remi-
niscent of the conclusions of the aforementioned electrochemi-
cal study. This is certainly due to the higher polymerizability
13 P. N. Bartlett, L.-Y. Chung and P. Moore, Electrochim. Acta,
1990, 35, 1273.
14 P. N. Bartlett, A. C. Benniston, L.-Y. Chung, D. H. Dawson and
P. Moore, Electrochim. Acta, 1991, 36, 1377.
15 H. Cattey, P. Audebert and C. Sanchez, New J. Chem., 1996, 20,
1023.
16 H. Cattey, P. Audebert, P. Hapiot and C. Sanchez, J. Phys.
Chem., 1998, 102, 1193.
17 T. J. Bond, R. Jenkins, A. C. Ridley and P. C. Taylor, J. Chem.
Soc., Perkin T rans. 1, 1993, 2241.
18 T. Schalkhammer, E. MannÈBuxbaum, F. Pittner and G. Urban,
Sens. Actuators B, 1991, 24, 273.
19 G. Bidan, T etrahedron L ett., 1985, 26, 735.
20 G. Bidan and M. Guglielmi, Synth. Met., 1986, 15, 49.
21 L. Friedman and H. Shechter, J. Org. Chem., 1960, 25, 877.
22 R. A. Smiley and C. Arnold, J. Org. Chem., 1960, 25, 257.
23 S. Maeda, R. Corrada and S. P. Armes, Macromolecules, 1995, 28,
2905.
24 J. Roncali, Chem. Rev., 1992, 92, 711.
25 This has been conÐrmed by preparing separately the bisacetonate
of 6ACAC and Co2` (prepared by a standard procedure by
reÑuxing overnight a 1 : 2 : 2 stoichiometric mixture of cobalt
dichloride, 6ACAC and ammonium hydroxide), which does not
polymerize, but displays an IR carbonyl band at 1640 cm~1.
26 A. F. Diaz, Chem. Scr., 1981, 17, 145.
27 P. Audebert, G. Bidan and M. Lapkowski, J. Chem. Soc., Chem.
Commun., 1986, 887.
28 H. KorriÈYoussouÐ, F. Garnier, P. Shrivastava, P. Godillot and
A. Yassar, J. Am. Chem. Soc., 1997, 119, 7388.
29 J. Tamm, A. Alumaa, A. Hallik and V. Sammelselg, J. Electro-
anal. Chem., 1998, 448, 25.
Fig. 8 IR spectroscopic study of the copolymers. Composition of the
copolymer Ðlms in functionalized pyrrole vs. x, the relative proportion
of 6CA (dashed line) and 6ACAC (solid line) in the polymerization
feed. Electrochemical conditions as for Fig. 7.
New J. Chem., 2000, 24, 877È884
883

View Article Online
30 M. M. Richter and A. J. Bard, Anal. Chem., 1996, 68, 2641.
31 X. Ren and P. Pickup, J. Electrochem. Soc., 1992, 139, 2097.
32 C. J. Pouchert, T he Aldrich L ibrary of FT -IR Spectra, 2nd edn.,
Aldrich Chemical, Milwaukee, WI, 1997.
34 C. P. Andrieux, P. Audebert and C. Salou, J. Electroanal. Chem.,
1991, 318, 235.
35 P. H. Aubert, A. Neudeck, L. Dunsch, P. Audebert, P. Capdeviel-
le and M. Maumy, J. Electroanal. Chem., 1999, 470, 77.
33 While this is not the only factor that governs the reactivity of
pyrrolic cation radicals (see e.g. P. Audebert, F. Miomandre, S.
Di Magno, V. V. Smirnov and P. Hapiot, Chem. Mater., 2000, 12,
2025), it is clear that the higher the oxidation potential of a
species, the higher its reactivity becomes, provided that other
factors do not exert a larger inÑuence.
884
New J. Chem., 2000, 24, 877È884