Photoelectric conversion from a nitrobenzene dye monolayer modified ITO electrode

Article abstract of DOI:10.1039/a708450a

An amphiphilic nitrobenzene dye (C₁₈H₃₇)₂N-C₆H ₄-CH=N-NH-C₆H₄-NO₂ has been synthesized and deposited on semiconducting transparent indium-tin oxide (ITO) electrodes by

Full text of DOI:10.1039/a708450a

View Article Online / Journal Homepage / Table of Contents for this issue
Photoelectric conversion from a nitrobenzene dye monolayer modiÐed
ITO electrode
Deng-Guo Wu, Yan-Yi Huang, Chun-Hui Huang* and Liang-Bing Gan
State Key L aboratory of Rare Earth Materials Chemistry and Applications, Peking University,
Beijing 100871, China
An amphiphilic nitrobenzene dye (C
H ) NwC H wCHxNwNHwC H wNO has been synthesized and deposited on
18 37 2
6
4
6
4
2
semiconducting transparent indiumÈtin oxide (ITO) electrodes by a LangmuirÈBlodgett (LB) technique. Photocurrent generation
was studied in a conventional photoelectrochemical cell. An action spectrum of the photocurrent generation is coincident with
the absorption spectrum of the LB Ðlm-modiÐed electrode, indicating that the dye aggregate in the LB Ðlm is responsible for the
photocurrent. Some factors which may a†ect the observed photocurrent, such as the presence of O , the concentration of methyl
2
violet (MV2`), and hydroquinone (H Q), pH and the bias voltage have been investigated. Models for the mechanism of photo-
2
current generation under di†erent conditions are proposed.
It is known that some compounds not only show good per-
formance in photoelectric conversion, but also show excellent
behaviour in the Ðeld of non-linear optics. For example,
squaraine, has been used practically in xerographic photo-
receptors.1 Recently, Ashwell has shown that this centro-
symmetric molecule also has good second-order non-linear
optical character.2 Our group have systematically studied the
photoelectric conversion properties of some dye molecules
with second-order non-linear optical character and have
found that the higher the molecular hyperpolarizability, the
better the molecular photoelectric conversion efficiency.3h6
The principle of second-harmonic non-linear optics indicates
that the di†erence in the dipole moment of the molecule
between the ground and excited states is one of the main
factors for molecular hyperpolarizability.7,8 On the other
hand, photocurrent generation and charge dissociation are
connected with the charge separation process,9,10 which also
requires a large dipole moment in the excited state. In order
to obtain a better understanding of the relationship between
molecular structure, second-order non-linear optical proper-
ties and the efficiency of photoelectric conversion, we have
mainly studied hemicyanines, congeners and azopyridinium
compounds with positive charge on the chromophore. As a
part of our systematic study, we have now designed a dye
molecule with electroneutral chromophores in which there is
an electronic donor and an acceptor on each side of the mol-
ecule. This will provide experimental data for developing
improved photoelectric conversion materials.
(CDCl ) d: 0.86 (t, 6H, 2CH ), 1.21 (m, 60H, 30CH ), 1.57 (m,
3
3
2
4H, 2CH ), 3.30 (m, 4H, 2CH N), 6.62 (d, 2H, phenyl-N), 7.05
2
2
(d, 2H, phenyl-N), 7.52 (d, 2H, phenyl-NO ), 7.71 (s, 1H,
2
CH \ N), 7.87 (b, 1H, NH), 8.15 (d, 2H, phenyl-NO ). Mp ca.
45È46 ¡C. The congeneric compound to dye
2
1
(CH ) NwC H wCHxNNHwC H wNO (dye 2) was
3 2
6
4
6
4
2
synthesized by condensing p-nitrophenyl hydrazine with 4-
(dimethylamino)benzaldehyde and puriÐed as above. Mp ca.
179È180 ¡C. The identity of the product was conÐrmed by 1H
NMR and elemental analysis. The spreading solvent was chlo-
roform. Distilled water was deionized water puriÐed with an
EASYpure RF compact ultrapure system (R B 18 M)). The
electrolyte for the electrochemical experiment was KCl (AR).
Hydroquinone (H Q) (AR) was recrystallized from water
2
before use. Methyl viologen diiodide (MV2`) was synthesized
by the reaction of 4,4@-dipyridyl with excess methyl iodide in
reÑuxing ethanol for 6 h. The product was Ðltered and washed
with ethanol at least four times and its identity was conÐrmed
by 1H NMR analysis.
Apparatus
C, H and N analysis of nitrobenzene dyes was carried out
with a Carlo Erba 1106 elemental analyser. 1H NMR spec-
troscopy was carried out using a Bruker ARX 400 BBI
spectrometer. The electronic spectra were measured with a
Shimadzu model 3100 UVÈVISÈNIR recording spectropho-
tometer. The LB Ðlm-modiÐed ITO electrode was fabricated
by using a model 622 NIMA LangmuirÈBlodgett trough. The
light source used for the photoelectrochemical study was a
In this paper, we report an investigation of the photo-
current generation from an electroneutral dye LB Ðlm-
modiÐed ITO electrode. Mechanistic models for photocurrent
generation under di†erent conditions are proposed.
Experimental
Materials
(C
H
) NwC H wCHxNwNHwC H wNO
(dye 1)
18 37 2
6
4
6
4
2
was synthesized by condensing p-nitrophenyl hydrazine
with 4-(dioctadecaneamino)benzaldehyde in absolute ethanol.
Hexadecyltrimethylammonium bromide was used as a phase
transfer catalyst. The synthesis procedure is illustrated in
Scheme 1. The product was puriÐed by column chromatog-
raphy on silica gel with dichloromethane as eluent. Ele-
mental analysis: found: C, 77.41; H, 11.94; N, 7.12%. Calc. for
C
H
N O : C, 77.31; H, 11.12; N, 7.36%. 1H NMR
Scheme 1 Synthesis of nitrobenzene dye
49 84
4 2
J. Chem. Soc., Faraday T rans., 1998, 94(10), 1411È1415
1411

View Article Online
500 W Xe arc lamp, the light beam was passed through a
group of monochromatic Ðlters (ca. 300È800 nm, Toshiba Co.,
Japan, and Schott Co., USA) in order to get a given bandpass
of light. The light intensity at each wavelength was calibrated
with an energy and power meter (Scientech, USA).
LB Ðlm preparation
A solution of dye 1 (0.712 mg ml~1 in chloroform) was spread
dropwise onto the clean water subphase by syringe at a sub-
phase temperature of ca. 20 ¡C. The chloroform was allowed
to evaporate for 15 min and the monolayer was then com-
pressed. The substrates for monolayer deposition were trans-
parent electrodes of ITO-coated borosilicate glass and the
sheet resistance was 50 ) cm~2. The ITO plates were cleaned
using detergent and chloroform, followed by rinsing with pure
water. To ensure the formation of a hydrophilic surface, the
plates were immersed for 2 days in a saturated sodium meth-
anol solution and then thoroughly ultrasonicated with pure
water several times.
Fig. 2 Absorption spectroscopy of (C
H
) NwC H wCHx
18 37 2
6 4
NwNHwC H wNO on ITO substrate (ÈÈ) and in chloroform
6
4
2
solution (. . . .)
For deposition of the monolayer, the ITO slide was Ðrst
immersed in the subphase; a single monolayer was formed
and compressed to 23 mN m~1. The slide was then raised at a
rate of 4 mm min~1. Only Ðlms having transfer ratios of
1.0 ^ 0.1 were used in the experiments.
a
similar compound (C
H ) NwC H wCHxCHw
16 33 2
6
4
C H Nw(CH ) wCOOv reported by Bubeck et al.,11 It can
5
4
2 2
be clearly seen that there is a plateau in the nÈA curve, indi-
cating that the dye molecules reorientate on the water surface
above 20 mN m~1 surface pressure, since the nitrobenzene
Photoelectrochemical and electrochemical measurements
molecule has two hydrophilic radicals (wNO and wNH).
2
A conventional glass three-electrode cell (30 ml capacity)
having a Ñat window for illumination of the working electrode
was used. The electrical connection to the ITO working elec-
trode was obtained by mechanical contact with a clip, the illu-
minated area was 0.38 cm2. The counter-electrode was a
polished Pt wire and the reference was a saturated calomel
electrode (SCE). All photoelectrochemical data were recorded
using a model CH 600 voltammetric analyser controlled by
computer. Cyclic voltammetry (CV) experiments were per-
formed on an EG&GPAR 273 potentiostat/galvanostat with
EG&GPAR 270 electrochemical software. The supporting
electrolyte was an aqueous solution of 0.5 M KCl. Unless
speciÐed, oxygen was removed from the solutions by bubbling
The j
of the LB Ðlm in a UVÈVIS absorption spectrum
'
is at 450 nm while the j
nm (Fig. 2), indicating the presence of J-aggregates.12
of the dye in chloroform is at 430
'
Electrochemical properties of the dye
In order to estimate the redox potentials of the excited-state
dye and to discuss the mechanism of photocurrent generation,
CV studies were carried out. Typical voltammograms are
shown in Fig. 3. Monolayer dyeÈITO was used as a working
electrode in 0.5 M KCl aqueous solution. The pH of the elec-
trolyte solution was adjusted to pH ca. 2 with dilute HCl solu-
tion in the acidic system. The redox potentials of the couples
dye/dye~ and dye/dye` are [0.35 and 0.80 V vs. SCE, respec-
tively, in a neutral medium. It can be seen from Fig. 3 that the
CV curve was shifted towards positive voltage in an acidic
medium compared with that in a neutral medium. The
reduction peak was shifted by ca. 136 mV while the oxidation
peak was only shifted by ca. 32 mV. The positive shift of the
reduction peak indicates that a proton is involved in the redox
reaction and the reduction reaction takes place more easily in
an acidic rather than a neutral medium.13
N before every measurement.
2
Results and Discussion
Characterization of LB Ðlm and dye–ITO electrode
Surface pressure measurements on monolayers were used to
determine the average values of the area per molecule
occupied by the dye at the air/water interface. Fig. 1 shows a
typical surface pressureÈarea (nÈA) isotherm for dye 1. The
limiting area per molecule is 45 Ó2, which agrees with that for
Fig. 3 CV of the dye 1 monolayer on ITO substrate as working elec-
trode in 0.5 M KCl aqueous solution. (ÈÈ) Neutral solution and (. . . .)
pH ca. 2 solution. Sweep rate \ 100 mV s~1 (C reduction peak, B
oxidation peak).
Fig. 1 Surface pressureÈarea isotherm of (C
18 37 2
CHxNwNHwC H wNO at the air/water interface (20 ^ 1 ¡C)
H ) NwC H w
6
4
6
4
2
1412
J. Chem. Soc., Faraday T rans., 1998, V ol. 94

View Article Online
Photocurrent generation from dye–ITO electrode
A steady cathodic photocurrent ranging from 89 to 160 nA
cm~2 was obtained from the dye monolayer-modiÐed elec-
trode when the dyeÈITO electrode was illuminated, by white
light under 110 mW cm~2 light intensity and without any bias
voltage, in 0.5 M KCl electrolyte solution. The photoelectric
response was very stable when switching on and o† many
times, as shown in Fig. 4. Fig. 5 shows the action spectrum of
the cathodic photocurrents. The spectrum responses coincide
with the absorption spectrum (Fig. 2) suggesting that the
aggregate of the dye in the LB Ðlm is responsible for photo-
current generation. A photocurrent of ca. 49.7 nA cm~2 can
be obtained under 1.14 ] 1016 photons cm~2 s~1 irradiation
at 464 nm by white light of 110 mW cm~2 intensity, through a
band-pass Ðlter (KL45 ] GG420) in 0.5 M KCl electrolyte
solution with zero bias voltage. The attainment of such a
photocurrent means that the quantum yield, which is cited per
absorbed photon, is ca. 0.24% for the monolayer dye-modiÐed
electrode (the absorbance ratio of the Ðlm is ca. 1.13% for the
incident light at 464 nm).
Fig. 6 Photocurrent vs. electrode potential for LB Ðlm of dye 1 on
ITO
may result from the applied negative voltage which is in the
same direction and can accelerate the electronic Ñow, forming
a strong electric Ðeld in the space charge region within the LB
Ðlm. Such a strong electric Ðeld may accelerate the photoin-
duced charge separation and decrease the probability of
charge recombination before the target electron acceptors are
reached.
E†ect of bias voltage
In order to determine the polarity of the current Ñow, the
e†ect of bias voltage was investigated. A plot of photocurrent
vs. electrode potential is shown in Fig. 6. The photocurrent is
cathodic in the range from [100 to ca. ]30 mV. The anodic
current can be observed and it increases with increasing posi-
tive bias voltage. When [100 mV bias voltage was applied to
the electrode, ca. 62.1 nA cm~2 photocurrent was obtained at
464 nm. In this case the quantum yield reaches ca. 0.30%. This
E†ect of electron donors and acceptors
Fig. 7 shows the dependence of the photocurrent on MV2`
concentration. It is seen that the photocurrent increases with
increasing MV2` concentration. For comparison with the
e†ect of oxygen on the photocurrent, the photocurrent was
measured both under ambient conditions and in the absence
of the oxygen (N gas was used to remove O from the elec-
2
2
trolyte solution). The results indicate that the quantum yield
for e†ective electron injection is higher in the presence of
oxygen because it acts as an electron acceptor through the
formation of a superoxide anion radical.14 If the electrolyte
solution was saturated with oxygen, by bubbling oxygen, the
photocurrent was noticeably increased by ca. 50%, indicating
that O is a favourable factor in the electron transfer process.
2
When H Q was added to the electrolyte solution the photo-
2
current tended to decrease under ambient conditions and
quickly decreased and became an anodic current in the
absence of oxygen. Fig. 8 shows the dependence of the photo-
current on the H Q concentration. In order to further conÐrm
2
this observation, the e†ect of the electron-donor K [Fe(CN) ]
4
6
on the photocurrent of the LB Ðlm was measured; the photo-
current quickly decreased with increasing K [Fe(CN) ] con-
centration and even became an anodic current under ambient
Fig. 4 Photocurrent generation from the dye 1 LBÈITO electrode
upon irradiation with white light at 110 mW cm~2 (C light switching
on and B light switching o†)
4
6
Fig. 5 Action spectrum of the cathodic photocurrents. (The photo-
currents and the intensities of di†erent wavelength are all normalized.)
Fig. 7 Relationship between MV2` concentration and photocurrent.
(L) Under ambient conditions, (…) in the absence of oxygen.
J. Chem. Soc., Faraday T rans., 1998, V ol. 94
1413

View Article Online
Fig. 10 E†ect of the media on UVÈVIS spectrum, (ÈÈ) ethanol
solution, (. . . .) pH ca. 2 solution, (- - - -) pH ca. 10 solution
Fig. 8 Dependence of the photocurrent on the H Q concentration.
2
(L) Under ambient conditions, (…) in the absence of oxygen.
conditions, indicating that the presence of an electron donor is
unfavourable to the production of a cathodic photocurrent in
the system.
in benzene, 430.0 nm in chloroform and 442.9 nm in ethanol.
The weaker absorption band has a tendency to be blue-shifted
with increasing polarity of the solvent, indicating that the
main absorption band corresponds to a p ] p* transition and
the weaker one to an n ] p* transition.15 Furthermore, the
E†ect of pH
The e†ect of pH on the photocurrent of the monolayer was
investigated in a BrittonÈRobinson bu†er solution containing
0.5 M KCl. It is seen, in Fig. 9, that the photocurrent decreases
quickly with increasing pH. At pH ca. 4, anodic and cathodic
spikes were observed on commencement and interruption of
illumination, respectively. The current changes sign and
becomes anodic at pH ca. 5. This indicates that an acidic
medium is favourable to the production of a cathodic current
while a weak acidic or alkaline medium is not. Under favour-
p ] p* transition absorption band shows
a blue-shift
(*l \ 28.6 nm) in an acidic medium (pH ca. 2), a red-shift
(*l \ 35.7 nm) and becoming broad in an alkaline medium
(pH ca. 10), compared with that in ethanol solution. The rela-
tive height of the n ] p* to p ] p* transition band is less in
an acidic medium than in an alkaline medium (Fig. 10). This
indicates that the conjugation of dye 2, compared with a
neutral medium (ethanol solution), is decreased in an acidic
medium and increased in an alkaline medium. No detectable
spectral change in ethanol solution was observed after 1 h of
irradiation at 464 nm, indicating that the conjugation e†ect of
dye 2 is unchanged on irradiation.
able conditions, e.g. pH ca. 2, in the presence of O and 5
2
mmol l~1 MV2`, with a [100 mV bias voltage, then a photo-
current of ca. 121 nA cm~2 was obtained under 464 nm irra-
diation; the quantum yield was 0.58%. Note that dissolved
oxygen has little e†ect on the photocurrent in the acidic
system. This may be because the oxygen molecule combines
Mechanism of photocurrent generation from the dye–ITO
electrode
with H` ion to form HO ` ion, decreasing the electron-
2
Since there is no detectable spectral change after irradiation of
the dye solution, it is concluded that there is no trans ] cis
photoisomerization occurring. A photoinduced tautomerism,
as shown in Scheme 2, may take place. Upon irradiation A
absorbs energy from the excitation light and undergoes tauto-
merism involving an intramolecular shift of a hydrogen atom
between alternative binding sites to form B. In an acidic
medium, B may accept protons to become C, which is an elec-
accepting ability of the oxygen. Since an alkaline medium has
no e†ect on the electron-accepting ability of oxygen, the e†ect
of dissolved oxygen on the photocurrent was observed in an
alkaline system.
E†ect of the media on the UV–VIS spectrum
Since dye 1 is insoluble in polar solvent, a congener of dye 1
was used for the following experiments. In order to discuss the
e†ect of the solvent on photocurrent generation, the UVÈVIS
spectra of dye 2 were measured in di†erent solvents. The main
absorption band is increasingly red-shifted with increasing
polarity of the solvent. For example, the peak is at 421.4 nm
Fig. 9 E†ect of pH on the photocurrent in BrittonÈRobinson bu†er
solution: (L) under ambient conditions , (…) in the absence of oxygen
Scheme 2
1414
J. Chem. Soc., Faraday T rans., 1998, V ol. 94

View Article Online
Conclusions
The results suggest that the mechanism of generation of a
cathodic photocurrent from the dyeÈITO electrode involves
the formation of excited-state tautomerism. This state accepts
an electron from the conductive band of the ITO electrode
and, meanwhile, transfers an electron to the electrolyte solu-
tion through the LB Ðlm. The mechanism is supported by
studying the e†ect of bias voltage, the addition of donors and
acceptors, and pH, on the photocurrent generation. The
quantum yield is 0.58% under favourable conditions.
Scheme 3 Schematic diagram showing electron transfer processes.
Arrows indicate the electron Ñow. (a) and (b) cathodic photocurrent,
(Dye-H`) represents the state of dye 1 which combines with H` ions
in acidic medium, (c) anodic photocurrent.
This project is Ðnancially supported by a National Funda-
mental Research Key Project of China and the National
Natural Science Foundation of China (No. 29671001)
tron acceptor and thus assists generation of a cathodic photo-
current. This mechanism is also supported by the spectrum
change in an acidic medium. C and D have a resonance struc-
ture which accepts electrons to become E.
References
The direction of the photocurrent generation is shown in
Scheme 3. We can see that it depends not only on the dye
sensitized by light, but also on the nature of the redox couple
in the aqueous phase surrounding the electrode. In the pres-
ence of redox couples favouring electron donation an anodic
photocurrent is generated, while with electron acceptors in the
aqueous phase a cathodic photocurrent is produced.
In order to examine the mechanism of light sensitization for
the production of cathodic and anodic photocurrents, the
energies of the relevant electronic states must be estimated.
1
2
K. Y. Law, Chem. Rev., 1993, 93, 449.
G. J. Ashwell, G. Je†eries, D. G. Hamilton, D. E. Lynch, M. P. S.
Roberts, G. S. Bahra and C. R. Brown, Nature (L ondon), 1995,
375, 385.
3
W. S. Xia, C. H. Huang, L. B. Gan and H. Li, J. Chem. Soc.,
Faraday T rans., 1996, 92, 3131.
W. S. Xia, C. H. Huang and D. J. Zhou, L angmuir, 1997, 13, 80.
W. S. Xia, C. H. Huang, X. Z. Ye, C. P. Luo, L. B. Gan and Z. F.
Liu, J. Phys. Chem., 1996, 100, 2244.
W. S. Xia, C. H. Huang, L. B. Gan, H. Li and X. S. Zhao, J.
Chem. Soc., Faraday T rans., 1996, 92, 769.
Organic Materials for Nonlinear Optics III ed. G. J. Ashwell and
D. Bloor, Royal Society of Chemistry, Cambridge, 1993.
P. N. Prasad and D. J. Williams, Introduction to Nonlinear
Optical E†ects in Molecules and Polymers, Wiley, New York,
1991.
4
5
6
7
8
From the electron affinity the conduction band (E ) and
c
valence band (E ) edges of the ITO electrode surface are esti-
v
mated to be ca. [4.5 and [8.3 eV,16 respectively. The energy
levels of the dye are assumed to be [5.54 eV (0.80 V vs. SCE)
and [2.78 eV on the absolute scale, respectively, with refer-
ence to an oxidation potential of 0.80 V vs. SCE and a band
gap of 2.76 eV (450 nm). The reduction potential of the dye in
an acidic medium is [4.52 eV ([0.22 V vs. SCE), the re-
duction potential of MV2` is [4.51 eV ([0.23 V vs. SCE),17
9
S. Nespurek, Int. J. Electron., 1994, 76, 777.
10 S. Allen, in Molecular Electronics, ed. G. J. Ashwell, Research
Studies, Taunton, 1992, pp. 207È265.
11 C. Bubeck, A. Laschewsky, D. Lupo, D. Neher, P. Ottenbreit, W.
Paulus, W. Prass, H. Ringsdorf and G. Wegner, Adv. Mater.,
1991, 3, 54.
12 A. Haraguchi, Y. Yonezawa and R. Hanawa, Photochem. Photo-
biol., 1990, 52, 307.
the oxidation potential of H Q is [4.61 eV ([0.13 V vs.
2
SCE)17 on the absolute scale.
The cathodic photocurrent probably involves electron
transfer from the excited dye aggregate to the electron accep-
tor, with a subsequent electron transfer from the conduction
band of the ITO electrode to the hole residing in the dye
aggregate17 in the presence of some electron acceptors, such
13 Y. Li and S. Dong, J. Electroanal. Chem., 1993, 348, 181.
14 (a) H. Hada and Y. Yonezawa, Synth. Met., 1987, 18, 791; (b) H.
Hada, Y. Yonezawa and H. Inaba, Ber. Bunsen-Ges. Phys. Chem.,
1981, 85, 425.
15 (a) N. S. Bayliss and E. G. Mcrae, J. Phys. Chem., 1954, 58, 1002;
(b) G. J. Brealey and M. Kasha, J. Am. Chem. Soc., 1955, 77,
4462.
16 L. Sereno, J. J. Silber, L. Otero, M. D. V. Bohorquez, A. L.
Moore, T. A. Moore and D. Gust., J. Phys. Chem. 1996, 100, 814.
17 Y. S. Kim, K. Liang, K. Y. Law and D. G. Whitten, J. Phys.
Chem., 1994, 98, 984.
as O , MV2` and in an acidic medium.
2
Anodic photocurrent generation may occur when an elec-
tron donor injects an electron into the hole of the ground-
state dye aggregate with subsequent electron transfer from the
excited dye aggregate to the conduction band of the ITO elec-
trode,17 in the presence of some electron donors, such as H Q,
2
K [Fe(CN) ] and in an alkaline medium.
Paper 7/08450A; Received 24th November, 1997
4
6
J. Chem. Soc., Faraday T rans., 1998, V ol. 94
1415