Grafting of oligoaniline on CdSe nanocrystals: Spectroscopic, electrochemical and spectroelectrochemical properties of the resulting organic/inorganic hybrid

Article abstract of DOI:10.1039/b413183e

We have synthesised a new organic/inorganic hybrid, in which electroactive aniline tetramer is grafted on the surface of quasi-monodispersed CdSe nanocrystals. The grafting reaction consists of two steps. First initial nanocrystal surface ligands (TOPO) a

Full text of DOI:10.1039/b413183e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Grafting of oligoaniline on CdSe nanocrystals: spectroscopic,
electrochemical and spectroelectrochemical properties of the resulting
organic/inorganic hybrid
Claudia Querner,a Peter Reiss,*a Malgorzata Zagorska,b Olivier Renault,c Renaud Payerne,a
Franc¸oise Genoud,a Patrice Rannoua and Adam Pron*a
Received 26th August 2004, Accepted 7th October 2004
First published as an Advance Article on the web 18th November 2004
DOI: 10.1039/b413183e
We have synthesised a new organic/inorganic hybrid, in which electroactive aniline tetramer is
grafted on the surface of quasi-monodispersed CdSe nanocrystals. The grafting reaction consists
of two steps. First initial nanocrystal surface ligands (TOPO) are exchanged for
4-formyldithiobenzoate linker ligands, which contain an anchor function (the dithioate group)
and an aniline grafting function (the aldehyde group). The condensation reaction between the
primary amine group of aniline tetramer and the aldehyde group of the linker ligand results in the
grafting of the former with simultaneous formation of an azomethine linkage between both
1
reagents. The grafting reaction is nearly quantitative as demonstrated by H NMR, FTIR and
XPS studies. Voltammetric, UV-visible and EPR spectroelectrochemical studies of the hybrid
compound show that after the grafting process the polyconjugated molecule retains its
electrochemical activity i.e. electrochemical switching between the completely reduced state
(leucoemeraldine), the semi-oxidised state (emeraldine) and the completely oxidised state
(pernigraniline) is possible.
However, the possibility of preparing true molecular hybrids,
Introduction
in which the polymer (oligomer) is attached to a nanocrystal
Conjugated polymers have been extensively studied for almost
via a covalent bond, seems even more tempting. Two principal
three decades for two principal reasons. First, in their neutral
types of such hybrids can be envisioned. In the first type
(undoped) state they are semiconductors and can serve as
inorganic nanoparticles are grafted on individual macromole-
active components of so-called ‘‘polymer electronics’’.1
cules of conjugated polymers via an appropriate spacer.
Second, using an appropriate doping reaction, which can be
Alternatively, in hybrids of the second type, conjugated
either of redox or acid–base nature, semiconducting conju-
oligomers or polymers are grafted on the functionalised
nanocrystal surface.7 A(II)B(VI) nanocrystals are especially
gated polymers can be transformed into so-called ‘‘organic
metals’’ which, among others, can be used as electrodes in
organic electronic components and devices.2
well suited for the preparation of the latter type of hybrids
because of a rather facile functionalisation of the nanocrystal
Another aspect of ‘‘electroactive polymers science’’ should
surface. In this case, bi-functional nanocrystal ligands must be
also be pointed out. Since recent generations of conjugated
used, which contain an anchor group for the nanocrystal
polymers are solution processable, they can easily be mixed
surface and a second functional group used for polymer
with inorganic semiconductors or conductors to give new
(oligomer) grafting via an appropriate chemical reaction with
hybrid-type materials whose properties can be altered by the
its end or lateral group.
appropriate selection of the organic and inorganic components
In this paper, we report on basic spectroscopic and
of the blend. Instructive examples of this approach have been
electrochemical properties of the first CdSe nanocrystal/
reported by Huynh et al.3 and more recently by Beek et al.,4
Liu et al.5 and Pientka et al.6 They demonstrated that the use
oligoaniline hybrid. To the best of our knowledge no
semiconductor nanocrystal/poly- or oligoaniline hybrids have
of a composite of an electron-donating organic polymer
ever been prepared. For grafting on nanocrystals we have
(poly(thiophene) derivatives or poly(p-phenylenevinylene)
selected aniline tetramer in the oxidation state of emeraldine
derivatives) with electron-accepting CdSe or ZnO nanocrystals
because it mimics to a large extent the behaviour of poly-
in hybrid-type photovoltaic cells facilitates exciton dissociation
(aniline)—one of the most extensively studied conducting
and charge separation.
polymers. In particular, poly(aniline) shows interesting non-
linear electrical transport properties in its undoped state8 and
Blending of inorganic nanocrystals with conjugated poly-
mers is an interesting and convenient method for the
can be easily transformed into an organic conductor via
protonation with non-oxidizing acids9 i.e. in the chemical
preparation of new materials with improved properties.
environment where semiconductor nanocrystals remain intact.
Thus by grafting of aniline tetramer on CdSe nanocrystals
*preiss@cea.fr (Peter Reiss)
pron@cea.fr (Adam Pron)
two types of molecular junctions can easily be obtained:
554 | J. Mater. Chem., 2005, 15, 554–563
This journal is ß The Royal Society of Chemistry 2005

View Article Online
semiconductor–semiconductor for the base form of the
oligomer and semiconductor–conductor for its protonated
form.
Experimental
Chart 1 Aniline tetramer (4).
Chemicals
Preparation of aniline tetramer in the oxidation state of
emeraldine (4). Aniline tetramer (Chart 1) was prepared
via oxidative coupling of N-phenyl-1,4-phenylenediamine
(C12H12N2?HCl) as described in ref. 14.
1H NMR (d6-DMSO, 200 MHz): d 8.37 (1H, s, N–H), 7.23
(2H, d, J 5 7 Hz), 7.11 (4H, d, J 5 5 Hz), 7.0–6.7 (9H, m), 6.62
(2H, d, J 5 8 Hz), 5.51 (2H, s, N–H2). UV-vis (DMSO):
lmax 5 306 nm, l 5 591 nm. IR (n/cm21): 1597 (s), 1497 (s),
1320 (m), 1166 (w), 839 (m), 745 (m), 692 (w).
Anhydrous solvents and reagents (Aldrich) were of the
highest purity available. Trioctylphosphine (TOP, 90%) and
hexadecylamine (HDA, 90%) were additionally purified by
distillation.
Preparation procedures
CdSe nanocrystals preparation. CdSe nanocrystals were
prepared following the procedure described in detail in
ref. 10. A mixed solvent consisting of 70% of hexadecylamine
(HDA) and 30% of trioctylphosphine oxide (TOPO) was used
as the reaction medium. The basic spectroscopic properties
of the obtained TOPO-capped CdSe nanocrystals are sum-
marised below:
1H NMR (CDCl3, 200 MHz): d 1.7–1.4 (s, 6H), 1.19 (s,
36H), 0.81 (s, 9H). UV-vis (CHCl3): l 5 561 nm (excitonic
peak), 460 nm (higher excited state).
Ligand exchange and aniline tetramer grafting. The ligand
exchange–aniline tetramer grafting sequence is schematically
presented in Scheme 2.
Step i: Bifunctional ligand 39 (12 mg, 0.06 mmol) is dissolved
in 10 mL of ethanol under inert atmosphere, then CdSe
nanocrystals (10 mg) in 10 mL of chloroform are added and
the mixture is stirred for 2 h at room temperature which is
sufficient for essentially quantitative exchange of TOPO for 39.
In the subsequent text nanocrystals containing 4-formyldithio-
benzoate surface ligands will be termed CdSe-39.
Synthesis of 4-formyldithiobenzoic acid (3). 4-Formyl-
dithiobenzoic acid was prepared from 4-bromobenzaldehyde
in a three step procedure, consisting of aldehyde protection,11
followed by a Grignard reaction to introduce the carbodithio-
ate group12 and finally aldehyde deprotection,13 as shown in
Scheme 1, with an overall yield of 45%.
Characterisation of the final product, 39: 1H NMR (d6-
acetone, 200 MHz): d 10.03 (1H, s), 8.32 (2H, d, J 5 8.6 Hz),
7.72 (2H, d, J 5 8.6 Hz). 13C NMR (d6-acetone, 200 MHz): d
128.76 (2C), 129.71 (2C), 137.34, 193.75, 252.80. Elemental
analysis calculated for C8H5OS2Na: C, 47.05; H, 2.47; Na,
11.26; O, 7.83; S, 31.40. Found: C, 47.17; H, 2.65; S, 30.67%.
Step ii:
A solution of aniline tetramer (4) (50 mg,
0.137 mmol) in 10 mL ethanol is added dropwise to the
solution of aldehyde functionalised nanocrystals CdSe-39. The
mixture is then refluxed overnight under inert atmosphere.
The formed black precipitate is filtered under inert atmosphere
and washed repeatedly first with methanol, in order to extract
Scheme 1 Synthetic pathway to obtain the bifunctional ligand 3.
Reagents and conditions: i) neopentyl glycol, CF3COOH, toluene,
90 uC, 15 h; 94% yield; ii) Mg, THF, 60 uC, 2 h; CS2, THF, 0 uC; 15 h;
HCl; 62% yield; iii) CHCl3 : H2O : CF3COOH (2 : 1 : 10), 25 uC, 15 h;
iv) NaHCO3, 25 uC; 78% yield.
Scheme 2 Grafting sequence of aniline tetramer (4) on CdSe
nanocrystals via the bifunctional ligand 39. Reagents and conditions:
i) 39, EtOH–CHCl3, rt, 2 h; ii) 4, EtOH, 80 uC, 15 h; 64% yield.
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 554–563 | 555

View Article Online
non-reacted tetramer, then with chloroform, with the goal to
remove non-grafted aldehyde ligand. The final black solid,
consisting of CdSe nanocrystals with aniline tetramer grafted
on their surface via the bridging ligand, is abbreviated in the
subsequent text as CdSe-4 (7 mg, 64% yield).
Characterisation of the product CdSe-4: 1H NMR (d6-
DMSO, 200 MHz): d 7.99 (6H, s), 7.2–6.6 (17H, m). UV-vis
(DMSO): lmax 5 325 nm, l 5 618 nm. IR (n/cm21): 1697 (w),
1597 (s), 1504 (vs), 1297 (s), 1166 (m), 1012 (w) (Ar–CS22), 819
(m), 750 (m), 696 (w).
spot size 150 6 800 mm) and a concentric hemispherical
analyser.
High-resolution spectra were acquired with steps of 30 meV
at a pass energy of 50 eV. The overall energy resolution of the
measurements (analyser and X-ray line width combined) was
thus 0.8 eV. Elemental quantification was performed by the
analysis of the integral intensity of each core-level line weighed
by its corresponding Scofield sensitivity factor available
from the internal database of the acquisition software.
Deconvolution of the core-level peak profiles was performed
by using a Shirley background subtraction and peaks of mixed
Gaussian–Lorentzian shape.
Synthesis of model compound (5). For comparative reasons
we have synthesised a model compound (5), shown in Chart 2,
which is identical to the organic part of CdSe-4, apart from its
terminal bromine group. Its preparation procedure can be
briefly described as follows. A solution of aniline tetramer (4)
(366 mg, 1 mmol) in 20 mL of anhydrous ethanol is added
dropwise to the solution of p-bromobenzaldehyde (185 mg,
1 mmol) in 10 mL of the same solvent and the mixture is
heated overnight under reflux. The resulting precipitate is
filtered under inert atmosphere and washed with a large excess
of methanol, chloroform and then dried in vacuum to give a
black solid (405 mg, 74% yield).
Cyclic voltammetry. Cyclic voltammetry investigations were
carried out in a one-compartment electrochemical cell using a
platinum counter electrode, an Ag/0.1 M AgNO3 reference
electrode and as the electrolyte a 0.1 M Bu4NBF4 solution in
acetonitrile (1.646 g in 50 mL) to which 0.02 mol (250 mg) of
diphenyl phosphate was added. Thin films of CdSe-4, 4 and 5
were deposited by casting from DMSO solutions on a Pt
working electrode of 5 mm2 surface area. The reference
electrode was calibrated vs. the ferrocene redox couple
measured in the same conditions.
1H NMR (d6-DMSO, 200 MHz): d 8.43 (1H, s), 8.11 (5H, s),
UV-vis spectroelectrochemistry. In spectroelectrochemical
experiments the same electrolytic solution, counter and
reference electrodes were used as in the case of cyclic
voltammetry investigations. Thin layers of CdSe-4 and 5 were
deposited on an ITO working electrode by casting from
DMSO solution. The UV-vis-NIR spectra were measured on
a Perkin Elmer Lambda 2 spectrometer (wavelength range
280–1100 nm).
7.8–7.4 (3H, m), 7.2–6.6 (14H, m). UV-vis (DMSO): lmax
325 nm, l 5 618 nm.
5
Characterisation techniques
The synthesised products were identified by 1H and 13C NMR
on a Bruker AC 200 MHz spectrometer (using deuterated
solvents as specified, according to the solubility of the
material, containing TMS as internal standard), as well as by
elemental analysis, carried out by the Analytical Service of
CNRS Vernaison (France). FT-IR spectra were recorded on a
Perkin-Elmer paragon500 spectrometer (wavenumber range:
4000–400 cm21; resolution 2 cm21) using the ATR technique.
Solution UV-vis spectra were recorded on a HP 8452A
spectrometer (wavelength range 190–820 nm). Powder X-ray
spectra were obtained with a Philips X’Pert MPD diffract-
ometer using a Co X-ray source operated at 50 kV and 35 mA
with a secondary graphite monochromator. AFM measure-
EPR spectroelectrochemistry. EPR spectroelectrochemical
experiments were carried out using an ER X-band Bruker
spectrometer. The spectroelectrochemical cell consisted of an
EPR tube to which three wire-type electrodes were inserted.
The working electrode consisted of a Pt wire with a thin film of
CdSe-4 or 5 deposited on it. Another Pt wire was used as the
counter electrode whereas an Ag wire served as pseudo-
reference. The potential of the pseudo-reference electrode
measured vs. 0.1 M AgNO3 was 336 mV. The spin response to
the applied electrode potential was measured in both increas-
ing and decreasing potential modes in the same electrolyte as
those used for cyclic voltammetry and UV-vis spectroelec-
trochemistry, using 20 mV increments. For each experimental
point the EPR susceptibility was determined by double
integration of the EPR signal.
ments were performed with
Instruments).
a Nanoscope IIIa (Digital
XPS spectroscopy. XPS measurements were performed
under ultra-high vacuum conditions (base pressure in the
analysis chamber: 5 6 1029 mbar) using an S-Probe spectro-
meter from Surface Science Instruments equipped with a
monochromated, micro-focused Al Ka source (hn 5 1486.6 eV,
Results and discussion
Basic spectroscopic and structural properties of the hybrid
CdSe-4
An important feature of the newly developed bi-functional
ligand, sodium 4-formyldithiobenzoate (39), is its ability to
efficiently exchange the initial TOPO ligands on the CdSe
surface under mild conditions. As judged from 1H NMR
spectroscopy and more precisely from the decrease of intensity
of the signal originating from the aliphatic protons of TOPO, a
Chart 2 Model compound (5).
556 | J. Mater. Chem., 2005, 15, 554–563
This journal is ß The Royal Society of Chemistry 2005

View Article Online
1
h room temperature reaction is sufficient to quasi-
quantitatively (.95%) exchange the initial ligands with 39.
This reaction is carried out in chloroform, which is a co-
solvent for TOPO functionalised CdSe nanocrystals as well as
for free TOPO, and an approximately 10-fold excess of the
carbodithioate ligand with respect to the calculated molar
quantity of surface TOPO molecules is used. The resulting
compound CdSe-39 can then be conveniently transformed into
CdSe-4 via a simple condensation reaction between the
aldehyde group of the carbodithioate surface ligand and the
primary amine group of aniline tetramer. The formation of
CdSe-4 is manifested by the disappearance of the 1H NMR
peaks at 10 ppm and 5.5 ppm, which are characteristic of the
reacting groups in both substrates i.e. the aldehyde proton
of 39 and the terminal amine proton of 4, respectively. The
spectrum of the organic/inorganic hybrid CdSe-4 consists of
two sets of broad, poorly resolved lines at ca. 8 ppm and in the
range of 6.6 to 7.2 ppm (Fig. 1). The broadening is most
probably caused by restricted motion of the tetraaniline
grafted on the nanocrystal surface via the phenylene–
azomethine linking group.15 This can be additionally corro-
borated by the comparison of the spectra of CdSe-4 and the
model compound 5. In both cases lines in the same spectral
range are observed, consistent with their chemical constitution.
The model compound, being a ‘‘free’’ molecule, gives a better
resolved spectrum, which is however complicated by the
coexistence of positional isomers together with cis–trans
isomerism.
The IR spectra recorded for CdSe-4 are in agreement with
the 1H NMR data. In particular the peak at 1697 cm21
characteristic of the CLO stretchings in CdSe-39 disappears
upon aniline tetramer grafting, which indicates that the
grafting process is quantitative and consumes all aldehyde
groups present on the nanocrystal surface. The dithioate
Fig. 2 Top: AFM image (tapping mode) of the hybrid CdSe-4 on a
silica substrate (scale bar: 1 mm). Bottom: Height analysis.
anchor function gives rise to a weak band at 1012 cm21 16
,
CdSe-4, after drop casting from a diluted DMSO solution onto
a silica substrate. Mainly isolated particles and to a lesser
extent aggregates comprising a small number of particles are
visible. The mean size of the individual particles determined by
height analysis (3.8 nm) is slightly bigger than the value based
on the documented correlation between the size (from TEM)
and the position of the exciton absorption peak in the UV-vis
spectrum of the original TOPO-capped CdSe nanocrystals
(3.3 nm).17 We attribute this difference to the organic shell,
consisting of the aniline tetramer and the linking ligand. Even
though squeezed together by the AFM tip, it contributes to the
overall size of the hybrid, whereas in TEM it is generally not
visible.
which is absent in the spectra of aniline tetramer (4) and the
model compound 5. Consistent with the grafting via the
primary amine group, its characteristic band at 1380 cm21
,
present in the free tetramer spectrum, is absent in the spectrum
of the hybrid.
Ligand exchange and tetramer grafting do not lead to the
formation of extended aggregates of the functionalised
nanocrystals. Fig. 2 shows an AFM image of the hybrid
Determination of the particle size by powder X-ray
diffraction using the Scherrer formula leads to a value of ca.
4.5 nm. Due to the excessive broadening of the reflections, it is
however difficult to determine the actual peak width, which
explains the slight discrepancy with the AFM results. The
diffractogram of CdSe-4, exhibiting the typical features of
wurtzite CdSe nanocrystals,18 is shown in Fig. 3.
In order to characterise the grafted oligomer–nanocrystal
interface we have performed XPS studies of CdSe-4. In the
analysis of the XPS data one must however be aware of the
fact that the synthesised hybrid is a complex system, which in
Fig. 1 1H NMR spectra of (a) the hybrid CdSe-4, and (b) the model
compound (5).
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 554–563 | 557

View Article Online
Fig. 3 Powder X-ray diffraction spectrum of the hybrid CdSe-4.
the simplest approximation can be considered as consisting of
the stoichiometric core of the nanocrystal, its strongly non-
stoichiometic surface and a layer containing the grafted
organic molecules. This makes quantitative determination
difficult because for all elements determined the elastic mean
free paths are smaller than the size of the object studied. This
means that the intensity of a given photoemission should be
proportional to the number of emitting atoms weighted by a
specific attenuation factor. The latter cannot be easily
calculated, especially in view of the presence of the organic
layer of the grafted oligomers on the surface of the
nanocrystal. In other words the determined contents of the
elements constituting the organic part of the hybrid (C, N, S)
and the inorganic one (Cd, Se) (in atom%) must be treated as
apparent. Moreover, it is expected that the contents of the
organic part elements should be overestimated taking into
account that the grafted oligomers constitute the outer layer.
The XPS spectrum of CdSe-4 deposited on the surface of a
single crystal silicon substrate is shown in Fig. 4. The peaks
corresponding to the photoemission from the core levels of all
expected elements are clearly seen: the 3d peak of Cd at 405 eV
close to the N 1s level at 399 eV, the 3d peak of Se at 54 eV,
the 2p peak of S at 159 eV, the C 1s peak at 285 eV. In
addition, a peak at 531 eV, attributed to O from the oxidised Si
substrate, can be distinguished, along with the 2p and 2s peaks
of Si around the S 2p peak; this reflects a non-continuous
CdSe-4 layer on Si.
Fig. 5 Detailed spectrum of the Cd 3d5/2 level of the hybrid CdSe-4.
is strongly nonstoichiometric, showing a significant excess of
cadmium atoms on the surface with the Cd to Se atomic ratio
equal to 2.7. The Cd 3d5/2 spectral line (Fig. 5) can be
deconvoluted into two components: the principal one at
405.3 eV whose binding energy is close to that measured for
large size CdSe crystals, which probably corresponds to Cd
atoms surrounded by saturated Se coordination spheres. The
second one, at lower binding energy, should be attributed to
those surface cadmium atoms which are complexed with the
oligomer. This signal accounts for ca. 30% of the integrated
total intensity.
Two XPS lines are especially interesting for the character-
isation of the oligomer grafted on the nanocrystal surface,
namely the S 2p and the N 1s peaks. The former should
provide information concerning the chelating process since the
anchor function contains two sulfur atoms, the latter is very
useful in the determination of the oxidation state of the
oligomer grafted on the CdSe surface. Unfortunately, the S 2p
peak interferes with that originating from the Se 3p3/2 core
emission, this makes any detailed analysis difficult. For this
reason in the subsequent text we limit ourselves to the
discussion of the N 1s spectrum (Fig. 6). In the simplest
approach the observed broad peak can be deconvoluted into
two components at 398.5 and 399.6 eV attributed to the amine
and imine nitrogen atoms, respectively.20 The organic part of
the hybrid consists of the aniline tetramer attached to
phenylene dithioate via an azomethine group (see Scheme 2
above). It is therefore not unexpected that the N 1s spectrum
Similarly as previously observed for TOPO-capped CdSe
nanocrystals,19 the surface layer of CdSe in the CdSe-4 hybrid
Fig. 4 XPS survey spectrum of the hybrid CdSe-4.
558 | J. Mater. Chem., 2005, 15, 554–563
Fig. 6 Detailed spectrum of the N 1s level of the hybrid CdSe-4.
This journal is ß The Royal Society of Chemistry 2005

View Article Online
of the hybrid resembles that of free aniline tetramer, not
attached to the nanocrystal (see Chart 1), since in both
compounds only amine and imine nitrogen co-exist.
the fact that upon grafting the conjugation is extended to the
linking molecule via the azomethine group (see Scheme 2).
Similarly to free aniline oligomers and polymers, the organic
part of CdSe-4 undergoes protonation if an appropriate
protonating agent is added to its solution. The protonation
is manifested by a total disappearance of the excitonic band at
618 nm and a significant decrease of the p–p* transition band
intensity. Simultaneously two new bands attributed to the
protonated state appear at ca. 405 nm and 780 nm, the former
corresponding to the polaronic band–p* transition, the latter
being characteristic of the localised polaron-type charge
carriers.22
Taking into account the complexity of CdSe-4, it is difficult
to extract quantitative data from the XPS spectra, however
they provide qualitative information of significant importance.
In particular they show a clear feature of the grafting reaction
which is manifested by the co-existence of two types of Cd
atoms—complexed and non-complexed. Moreover, the com-
parison of the N 1s spectra of the free and the grafted tetramer
indicates that, consistent with NMR data, the grafting process
is a simple condensation reaction.
The chemical nature of the organic part of CdSe-4,
consistent with Scheme 2, is also corroborated by UV-vis
spectroscopy. It must be noted here that because of the high
molar absorption coefficients the bands originating from the
organic part of the hybrid dominate, rendering the bands
originating from the nanocrystals essentially invisible. In the
spectrum of the hybrid (Fig. 7), two bands can be distinguished
at 325 nm and at 618 nm, both being characteristic of the base
form of the grafted ligand. The higher energetic band is
ascribed to the p–p* transition in the aromatic ring whereas
the band at lower energy is attributed to an excitonic-type
transition between the HOMO level of the quinone ring and
the LUMO level of the benzene ring, respectively.21 Thus, the
presence of the band at 618 nm confirms that the grafted
oligomer is in the semi-oxidised state of emeraldine, in which
the benzene and quinone rings coexist. In polyconjugated
systems the positions of the bands originating from electronic
transitions involving p electrons are very sensitive to the
conjugation length. For this reason the p–p* transition band
and the excitonic band in CdSe-4 are bathochromically shifted
by 19 nm and 27 nm with respect to the positions of these
bands in free aniline tetramer. This is a pure consequence of
It is known that protonation of the semi-oxidised form of
poly(aniline) or its oligomers leads, through charge redistribu-
tion, to a so-called ‘‘polaron lattice’’.23 From the chemical
point of view the created polarons are radical cations of the
semi-quinone structure. For this reason it is convenient to
study the protonation process by EPR spectroscopy. The
analysis of the shape of the EPR signal may in this case
provide important information concerning the mobility of
charge carriers introduced into the system via protonation.
However, such investigations have to be carried out with
extreme care, since other than intrinsic properties of the
material may influence the EPR linewidth. In particular the
presence of even minute amounts of fixed paramagnetic
centres, such as for example adsorbed oxygen molecules,
significantly broadens the EPR line attributed to the mobile,
spin carrying polarons.24 Therefore any comparative study of
charge mobility in aniline oligomers requires the use of
vacuum line techniques and extended pumping until the EPR
line parameters stabilise.
As expected neither CdSe-4 nor 4 give any EPR signal in
their semi-oxidised base forms. Protonation with 1 M HCl
gives rise, in both cases, to clear EPR lines which are
Lorentzian in shape and differ in their peak to peak width
(DHpp) (Fig. 8). In particular, the protonated form of 4 is
Fig. 7 UV-visible absorption spectra of the hybrid CdSe-4 in
protonated form (filled up-triangles) and deprotonated form (hollow
circles), registered in DMSO solutions.
Fig. 8 EPR signal of ‘‘free’’ aniline tetramer (4) (hollow circles) and
hybrid CdSe-4 (filled diamonds) protonated by 1 M HCl.
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 554–563 | 559

View Article Online
characterised by a narrow EPR line (DHpp 5 1.2 G) consistent
with its rather high conductivity in the doped state.25 The
grafted ligand shows, after protonation, a significantly broader
line (DHpp 5 4.2 G) indicating reduced charge mobility as
compared to the case of the free tetramer. This is surprising
because spectroscopic studies of the base forms of both
systems seem to indicate a slightly more extended conjugation
in CdSe-4 as compared to 4. However, the charge mobility as
experienced by EPR spectroscopy is a microscopic property
which principally depends on the conformation of the chain
carrying the spin and its immediate vicinity. It can be therefore
postulated that the grafting of the molecule on the nanocrystal
surface induces, in its protonated state, unfavourable con-
formational changes as far as charge mobility properties are
considered. Additionally, an increase in the local disorder
impeding to some extent the intermolecule hopping may
contribute to the line broadening.
This is only one of the possible explanations. The con-
tribution of the spin–orbit coupling between the mobile
organic parts of the hybrid and the nuclei constituting
the nanocrystal cannot be excluded, especially in view of the
conjugation extending essentially to the surface of the
nanocrystal.
Fig. 9 Cyclic voltammograms of thin films of aniline tetramer (4)
(hollow down-triangles), model compound (5) (filled up-triangles) and
the hybrid CdSe-4 (hollow circles), registered in 0.1 M Bu4NBF4
solution in acetonitrile using Ag/Ag+ as reference electrode and a scan
rate of 25 mV s21
.
Electrochemical and spectroelectrochemical properties of the
hybrid CdSe-4
Cyclic voltammograms of 4, CdSe-4 and 5 are shown in
Fig. 9. For 5 the two redox couples discussed above—
leucoemeraldine/emeraldine and emeraldine/pernigraniline—
can clearly be distinguished. The former one gives rise to a
relatively sharp anodic peak with a maximum at 215 mV and
the corresponding cathodic peak at 290 mV. The second
redox process is characterised by a broad anodic peak with its
even broader cathodic counterpart consisting of at least three
strongly overlapping peaks. The immediate effect associated
with the grafting is a change in the slope of the first anodic
peak. Although the oxidation of CdSe-4 starts at the same
potential as in the case of 5, the maximum of the first anodic
peak is slightly shifted towards higher potentials by 35 mV to
20 mV. The second anodic peak registered for CdSe-4 is broad
and poorly defined. In fact, the second oxidation process
occurs at any potential in the potential range from 150 mV to
600 mV. Assuming the same oxidation mechanism in CdSe-4
and 5 we must attribute the observed differences in their
voltammetric behaviour to structural and/or morphological
features. Monodispersed, p-conjugated molecules of 5 should
more easily form ordered structures assuring a more isoener-
getic environment for molecules being oxidised. This leads, in
turn, to narrower anodic peaks. It should be noted here that
the first oxidation process is accompanied by protonation of
the imine nitrogens (acid–base doping) whereas the second
process involves dedoping (deprotonation).28 The doping/
Aniline tetramer, similarly to poly(aniline), is electrochemically
active. Its electrochemical switching from the completely
reduced state through the semi-oxidised state to the completely
oxidised state is possible by potential scanning. In the
completely reduced state (leucoemeraldine state) all nitrogen
atoms are of amine nature, 50% of them are transformed into
the imine form during the first oxidation process to give the
emeraldine oxidation state. The second oxidation process leads
to the transformation of the remaining 50% of amine nitrogens
into imine ones, resulting in the pernigraniline state.
We were tempted to verify to what extent the grafting of an
electrochemically active ligand on the nanocrystal surface
influences its electrochemical behaviour. For this reason, we
have carried out comparative cyclic voltammetry (CV) and
spectroelectrochemical studies of CdSe-4 and 5. Voltammetric
investigations of poly(aniline) or its oligomers are usually
carried out in aqueous acidic electrolytes.26 Unfortunately the
pernigraniline form of the polymer (oligomer) formed at
higher potentials is not stable and in acidic conditions
undergoes hydrolytic degradation.27 This process lowers the
electroactivity of the system and severely perturbs the reverse
scan. For this reason, it is better to perform the CV
investigations in non-aqueous electrolytes such as Bu4NBF4–
acetonitrile. However, cycling of poly(aniline) or its oligomers
in this electrolyte leads to a very poor electrochemical response
unless a good source of protons is added to the electrolytic
solution.28 This is understandable since electrochemical
oxidation or reduction processes involve an exchange of
protons between the polymer (oligomer) deposited on the
electrode and the electrolyte. This process is of course greatly
facilitated in the presence of a good protonating agent such as
diphenyl phosphate, selected in this research.
dedoping sequence of reactions introduces
a significant
disorder to the system which is manifested by consecutive
broadening of the cathodic peaks. For the grafted ligand in
CdSe-4 the translational freedom is blocked and for this
reason its capability of the formation of ordered structures
with isoenergetic environments is severely limited. As a result
both of its anodic peaks are broadened. Note that in its
second redox process CdSe-4 can be more easily oxidised as
560 | J. Mater. Chem., 2005, 15, 554–563
This journal is ß The Royal Society of Chemistry 2005

View Article Online
compared to 5. In contrast, in the cathodic part it is more
difficult to reduce.
Since optical properties of the chromophore groups present
in CdSe-4 and 5 are extremely sensitive to even small changes
in their oxidation and protonation states, it is convenient to
follow their anodic oxidation by spectroelectrochemistry. For
both compounds the spectroelectrochemical investigations
were carried out in the increasing potential mode i.e. the first
spectrum was registered for the most reduced form of the
compound studied and the consecutive spectra were recorded
for potentials increasing in 100 mV increments. The spectro-
electrochemical behaviour of CdSe-4 and 5 is almost identical
showing that grafting of the ligand on the nanocrystal surface
does not influence, to a measurable extent, the mechanism of
its electrochemical oxidation.
Consistent with the cyclic voltammogram, at potentials
below the onset of the first anodic peak, both CdSe-4 and 5 are
in their most reduced base state. The dominant peak,
corresponding to the p–p* transition in the aromatic ring of
the ligand and the model compound being in the most reduced
state, is hypsochromatically shifted by ca. 25 nm (from 325 to
300 nm) as compared to the corresponding peak in the semi-
oxidised form of these compounds (compare Figs. 7 and 10).
However, even at these negative potentials, spectroscopic
signatures of very low intensity characteristic of the partially
oxidised state can be distinguished. These are a small peak at
430 nm and two broader bands around 600 nm and 1000 nm.
The existence of these residual bands can be rationalised either
by the incomplete reduction of both compounds due to kinetic
reasons or by the high reactivity of their reduced form which
locally reacts with traces of oxygen in the electrolyte or any
other source of a suitable oxidizing agent. At E 5 0 mV both
compounds are already partially oxidised. It must be noted
that the oxidation of the completely reduced forms of CdSe-4
and 5 is accompanied by the simultaneous protonation of the
oxidised segments with the protonating agent (diphenyl
phosphate) added to the electrolytic solution. As has already
been mentioned, the protonated state is characterised by a
decrease of intensity of the band originating from the p–p*
transition, a simultaneous growth (and hypsochromic shift) of
a narrow band around 420 nm and an increasing absorption
tail extended towards the near infrared (NIR) spectral region if
the doping induced charge carriers (polarons) are delocalised.
In the case of poly(aniline), the origin of these spectral features
has been explained in terms of protonation induced band
structure modification.22 The same reasoning can be applied
here. It is interesting to compare the spectrum recorded at
E 5 200 mV (Fig. 10b) with that obtained after protonation of
the base form of semi-oxidised CdSe-4 (Fig. 7). Both exhibit
features of the protonated emeraldine, however the former
shows a NIR absorption tail characteristic of delocalised
carriers whereas in the latter a NIR peak characteristic of
localised carriers is observed. This is in perfect agreement with
the analysis presented above and shows additionally that the
removal of the solvent facilitates the delocalisation of polarons
due to more planar conformation in the solid state.
Fig. 10 UV-vis-NIR spectra of (a) the model compound (5) and (b)
the hybrid CdSe-4 and registered for increasing electrode potential
E vs. Ag/Ag+, electrolyte: 0.1 M Bu4NBF4 solution in acetonitrile.
first from 250 mV to 200 mV and the second from 200 mV to
500 mV. The first potential zone corresponds to the complete
transformation of the most reduced forms CdSe-4 and 5 into
their fully protonated semi-oxidised forms. Thus with increas-
ing potential the peak at 425 nm and the absorption tail
extending to the NIR spectral region, which are characteristic
of the semi-oxidised protonated state, grow in intensity. At the
same time the intensity of the p–p* band decreases.
Spectral changes registered in the second potential zone are
consistent with the oxidation of the protonated, semi-oxidised,
forms of CdSe-4 and 5 to the fully oxidised ones. This process
is accompanied by deprotonation of the reaction product.
First we note that above E 5 200 mV the band at 415 nm,
characteristic of the protonated state, starts to decrease.
Second, a new band superimposed on the NIR absorption
In further analysis of the spectroelectrochemical behaviour
of both compounds it is convenient to distinguish two poten-
tials ranges corresponding to two different redox processes; the
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 554–563 | 561

View Article Online
tail appears at ca. 800 nm and starts to dominate the spectrum
at the highest potentials at the expense of the NIR absorption
tail. Third, the band ascribed to the p–p* transition in the
aromatic ring does not decrease any more but undergoes a
bathochromic shift to ca. 320 nm. All these changes unequi-
vocally indicate that the second oxidation results in increasing
contents of deprotonated segments in the completely oxidised
(pernigraniline) state. This, spectroelectrochemically observed,
deprotonation (dedoping) accompanying the second redox
process is also manifested in the cyclic voltammograms of
CdSe-4 and 5. After the first redox process a high capacitance
current is observed characteristic of the penetration of the
dopant anions in the layers deposited on the electrode. After
the second oxidation process this capacitance current strongly
decreases indicating that both CdSe-4 and 5 are present in their
neutral base forms.
Thus cyclic voltammetry studies combined with UV-vis
spectroelectrochemistry clearly show that although the graft-
ing of the ligand on the nanocrystal surface changes the shape
of its voltammogram, its oxidation mechanism remains
basically unchanged as compared to the case of the non-
grafted model compound.
The similarity of the redox and protonation/deprotonation
processes in CdSe-4 and in oligoanilines can be additionally
evidenced by EPR spectroelectrochemistry.
The evolution of the EPR signal with the increase of the
electrode potential for CdSe-4 and the model compound 5 is
presented in Fig. 11. It should be noted here that the measured
potential of the Ag pseudo-reference electrode used in the EPR
spectroelectrochemical experiments vs. Ag/0.1 M AgNO3 refer-
ence, used in other experiments, was 336 mV. This correction
should be taken into account when comparing Figs. 9, 10 and 11.
First, we notice that, in both cases, at potentials characteristic
of the completely reduced state (0 mV vs. Ag) where the spinless
base form of leucoemeraline is expected, a weak but measurable
EPR signal, Lorentzian in shape, is observed. This signal
originates from minute amounts of the non-reduced and
protonated forms of CdSe-4 and 5, present on the electrode.
This is consistent with the UV-vis spectroelectrochemical data
(see Fig. 10) which also show the presence of minute amounts of
the protonated semi-oxidised form of CdSe-4 at the potentials
below the onset of the first oxidation peak. With the increase of
the electrode potential, the intensity of the EPR line increases,
indicating an increasing amount of the semi-oxidised protonated
form of CdSe-4 on the electrode. The maximum of the EPR
susceptibility is obtained for E 5 660 mV vs. Ag (324 mV vs.
Ag/0.1 M AgNO3), then the susceptibility starts to decrease
indicating that further oxidation results in a growing content of
the deprotonated (base) form of pernigraniline (Fig. 12). The
Fig. 12 EPR susceptibility, obtained by double integration of the
EPR signal for varying electrode potentials of CdSe-4 (squares) and 5
(down-triangles) measured for increasing (hollow) and decreasing
(filled) potential E vs. Ag; electrolyte: 0.02 M diphenyl phosphate and
0.1 M Bu4NBF4 solution in acetonitrile.
Fig. 11 EPR signals for varying electrode potentials of (a) CdSe-4
and (b) 5, measured for increasing electrode potential E vs. Ag;
electrolyte: 0.02 M diphenyl phosphate and 0.1 M Bu4NBF4 solution
in acetonitrile.
562 | J. Mater. Chem., 2005, 15, 554–563
This journal is ß The Royal Society of Chemistry 2005

View Article Online
2001, 45, 1, 11–27; H. E. Katz, Z. N. Bao and S. L. Gilat, Acc.
Chem. Res., 2001, 34, 359–369; (d) LASER: N. Tessler, G. J. Denton
and R. H. Friend, Nature, 1996, 382, 695–697; F. Hide,
M. A. DiazGarcia, B. J. Schwartz, M. R. Andersson, Q. B. Pei
and A. J. Heeger, Science, 1996, 273, 1833–1836; F. Hide,
M. A. DiazGarcia, B. J. Schwartz and A. J. Heeger, Acc. Chem.
Res., 1997, 30, 10, 430–436.
2 G. Gustafsson, Y. Cao, G. M. Treacy, F. Klavetter, N. Colaneri
and A. J. Heeger, Nature, 1992, 357, 477–479; C. J. Drury,
C. M. J. Mutsaers, C. M. Hart, M. Matters and D. M. de Leeuw,
Appl. Phys. Lett., 1998, 73, 1, 108–110; G. H. Gelinck,
T. C. T. Geuns and D. M. de Leeuw, Appl. Phys. Lett., 2000, 77,
10, 1487–1489.
proposed sequence of oxidation reactions (spinless leucoemer-
aldine base–protonated, spin carrying emeraldine–spinless
pernigraniline base) should lead to two oxidation peaks in
cyclic voltammetry and one peak in the EPR spectro-
electrochemical response, as observed experimentally. It
should be also noted that the behaviour of CdSe-4 is very
reversible from the electrochemical standpoint since no
hysteresis in the electrochemical spin response is observed
for the reverse scan. In contrast, a clear hysteresis is
observed for 5, consistent with the cyclic voltammetry
curves of both compounds studied (compare Figs. 9 and 12).
3 W. U. Huynh, J. J. Dittmer and A. P. Alivisatos, Science, 2002,
295, 2425–2427.
4 W. J. E. Beek, M. M. Wienk and R. A. J. Janssen, Adv. Mater.,
2004, 16, 12, 1009–1013.
Conclusions
5 J. Liu, T. Tanaka, K. Sivula, A. P. Alivisatos and J. M. J. Fre´chet,
J. Am. Chem. Soc., 2004, 126, 21, 6550–6551.
6 M. Pientka, V. Dyakonov, D. Meissner, A. Rogach, H. Weller,
L. Lutsen and D. Vanderzande, Nanotechnology, 2004, 15, 1,
163–170.
7 S. Roux, G. J. A. A. Soler-Illia, S. Demoustier-Champagne,
P. Audebert and C. Sanchez, Adv. Mater., 2003, 15, 3, 217–221.
8 Y. Kieffel, J. P. Travers and D. Cottevieille, Annu. Rep. Conf.
Electr. Insul. Dielectr. Phenom., 2000, 1, 52–56.
9 E. T. Kang, K. G. Neoh and K. L. Tan, Progr. Polym. Sci., 1998,
23, 2, 277–324.
To summarise, we have synthesised a new organic/inorganic
hybrid consisting of oligoaniline molecules grafted on the
surface of CdSe nanocrystals. The completeness of the grafting
reaction was confirmed by 1H NMR, XPS and IR spectro-
scopies. The grafted polyconjugated ligand retains its electro-
chemical activity as proved by cyclic voltammetry and
spectroelectrochemical UV-vis and EPR investigations. The
developed grafting methods can easily be extended to other
polyconjugated molecules or macromolecules and other
semiconductor nanocrystals.
10 P. Reiss, J. Bleuse and A. Pron, Nano Lett., 2002, 2, 7, 781–784.
11 C. Y. Zhang and J. M. Tour, J. Am. Chem. Soc., 1999, 121, 38,
8783–8790.
12 R. Colorado, R. Villazana and T. R. Lee, Langmuir, 1998, 14, 22,
6337–6340.
13 H. Sashiwa, Y. Shigemasa and R. Roy, Macromolecules, 2000, 33,
19, 6913–6915.
14 B. Dufour, P. Rannou, J. P. Travers, A. Pron, M. Zagorska,
G. Korc, I. Kulszewicz-Bajer, S. Quillard and S. Lefrant,
Macromolecules, 2002, 35, 16, 6112–6120.
Acknowledgements
The authors thank Dr. J. P. Travers for critical review of the
manuscript and valuable discussions, and Dr. B. Divisia-Blohorn
for help with cyclovoltammetric measurements.
15 M. Kuno, J. K. Lee, B. O. Dabbousi, F. V. Mikulec and
M. G. Bawendi, J. Chem. Phys., 1997, 106, 23, 9869–9882.
16 H. Gotthardt, W. Pflaumbaum and P. Gutowski, Chem. Ber.,
1988, 121, 313–322; T. Gade, M. Streek and J. Voß, Chem. Ber.,
1988, 121, 2245–2249.
17 W. W. Yu, L. Qu, W. Guo and X. Peng, Chem. Mater., 2003, 15,
14, 2854–2860.
18 C. B. Murray, D. J. Norris and M. G. Bawendi, J. Am. Chem. Soc.,
1993, 115, 19, 8706–8715.
19 J. Taylor, T. Kippeny and S. J. Rosenthal, J. Cluster Sci., 2001, 12,
4, 571–582.
20 E. T. Kang, K. G. Neoh and K. L. Tan, Adv. Polym. Sci., 1993,
106, 135–190.
Claudia Querner,a Peter Reiss,*a Malgorzata Zagorska,b
Olivier Renault,c Renaud Payerne,a Franc¸oise Genoud,a Patrice Rannoua
and Adam Pron*a
aLaboratoire de Physique des Me´taux Synthe´tiques, UMR5819-SPrAM
(CEA-CNRS-Univ. J. Fourier-Grenoble I), DRFMC,
CEA-GRENOBLE, 17 rue des Martyrs, 38054 Grenoble cedex 9,
France. E-mail: preiss@cea.fr; pron@cea.fr; Fax: +33 4 38 78 51 13
bFaculty of Chemistry, Warsaw University of Technology,
Noakowskiego 3, 00-664 Warsaw, Poland
cSCPIO, DPTS, LETI, CEA-Grenoble, 17 rue des Martyrs, 38054
Grenoble cedex 9, France
21 W. Zheng, M. Angelopulos, A. J. Epstein and A. J. MacDiarmid,
Macromolecules, 1997, 30, 24, 7634–7639.
References
22 Y. Xia, J. M. Wiesinger, A. G. MacDiarmid and A. J. Epstein,
Chem. Mater., 1995, 7, 3, 443–445.
23 A. J. Epstein, J. M. Ginder, F. Zuo, R. W. Bigelow, H. S. Woo,
D. B. Tanner, A. F. Richter, W. S. Huang and A. G. MacDiarmid,
Synth. Met., 1987, 18, 1–3, 303–309.
24 E. Houze, M. Nechtschein and A. Pron, Phys. Rev. B, 1997, 56, 19,
12263–12267.
25 R. Payerne, P. Rannou, D. Djurado, F. Genoud, A. Pron and
J. P. Travers, Synth. Met., 2004, submitted.
26 W.-S. Huang, B. D. Humphrey and A. G. MacDiarmid, J. Chem.
Soc., Faraday Trans. 1, 1986, 82, 8, 2385–2400.
27 A. Pron, F. Genoud, C. Menardo and M. Nechtschein, Synth.
Met., 1988, 24, 3, 193–201.
28 M. Lapkowski, K. Berrada, S. Quillard, G. Louarn, S. Lefrant and
A. Pron, Macromolecules, 1995, 28, 4, 1233–1238.
1 (a) LED: S. R. Forrest, Nature, 2004, 428, 911–918; A. Kraft,
A. C. Grimsdale and A. B. Holmes, Angew. Chem., Int. Ed., 1998,
37, 4, 402–428; 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. Bredas, M. Logdlund and W. R. Salaneck,
Nature, 1999, 397, 121–128; (b) solar cells: G. Yu, J. Gao,
J. C. Hummelen, F. Wudl and A. J. Heeger, Science, 1995, 270,
1789–1791; M. Granstro¨m, K. Petritsch, A. C. Arias, A. Lux,
M. R. Andersson and R. H. Friend, Nature, 1998, 395, 257–260;
C. J. Brabec, N. S. Sariciftci and J. C. Hummelen, Adv. Funct.
Mater., 2001, 11, 1, 15–26; (c) FET: H. Sirringhaus, P. J. Brown,
R. H. Friend, M. M. Nielsen, K. Bechgaard, B. M. W. Langeveld-
Voss, A. J. H. Spiering, R. A. J. Janssen, E. W. Meijer,
P. Herwig and D. M. de Leeuw, Nature, 1999, 401, 685–688;
C. D. Dimitrakopoulos and D. J. Mascaro, IBM J. Res. Dev.,
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 554–563 | 563