Efficient charge injection from the S2 photoexcited state of special-pair mimic porphyrin assemblies anchored on a titanium-modified ITO anode

Article abstract of DOI:10.1002/chem.200600304

A novel surface fabrication methodology has been accomplished, aimed at efficient anodic photocurrent generation by a photoexcited porphyrin on an ITO (indium-tin oxide) electrode. The ITO electrode was submitted to a surface sol-gel process with titanium

Full text of DOI:10.1002/chem.200600304

FULL PAPER
DOI: 10.1002/chem.200600304
Efficient Charge Injection from the S2 Photoexcited State of Special-Pair
Mimic Porphyrin Assemblies Anchored on a Titanium-Modified ITO Anode
Mitsuhiko Morisue,[a, b] Noriko Haruta,[a] Dipak Kalita,[a] and Yoshiaki Kobuke*[a]
Abstract: Anovel surface fabrication
methodology has been accomplished,
aimed at efficient anodic photocurrent
generation by a photoexcited porphyrin
on an ITO (indium–tin oxide) elec-
trode. The ITO electrode was submit-
ted to a surface sol–gel process with ti-
tanium n-butoxide in order to deposit a
titanium monolayer. Subsequently, por-
phyrins were assembled as monolayers
on the titanium-treated ITO surface
via phosphonate, isophthalate, and thi-
olate groups. Slipped-cofacial porphyr-
in dimers, the so-called artificial special
pair at the photoreaction center, were
organized through imidazolyl-to-zinc
complementary coordination of imida-
zolylporphyrinatozinc(ii) units, which
were covalently immobilized by ring-
closing olefin metathesis of allyl side
chains. The modified surfaces were an-
alyzed by means of X-ray photoelec-
tron spectroscopy. Photoirradiation of
the porphyrin dimer generated a large
anodic photocurrent in aqueous elec-
trolyte solution containing hydroqui-
none as an electron sacrificer, due to
the small reorganization energy of the
dimer. The use of different linker
groups led to significant differences in
the efficiencies of anodic photocurrent
generation. The apparent flat-band po-
tentials evaluated from the photocur-
rent properties at various pH values
and under biased conditions imply that
the band structure of the ITO elec-
trode is modified by the anchoring spe-
cies. The quantum yield for the anodic
photocurrent generation by photoexci-
tation at the Soret band is increased to
15%, a surprisingly high value without
a redox cascade structure on the ITO
electrode surface, while excitation at
the Q band is not so significant. Exten-
sive exploration of the photocurrent
properties has revealed that hot injec-
tion of the photoexcited electron from
the S2 level into the conduction band
of the ITO electrode takes place
before internal conversion to the S1*
state, through the strong electronic
communication of the phosphonyl
anchor with the sol–gel-modified ITO
surface.
Keywords: ITO electrode
· por-
phyrinoids · self-assembly · sol–gel
processes · special pair
Introduction
as organic light-emitting diodes, thin-film transistors, and
dye-sensitized solar cells.[1] The principal strategy requires a
cascade alignment of the energy levels of functional mole-
cules leading smoothly to the Fermi levels of the contact
electrode. At the same time, an interfacial dipole at the mol-
ecule/electrode junctions alters the charge-injection barri-
ers.[2–4] Manipulation of the surface dipole is thus another
important device parameter for tuning the charge-injection
efficiency.[5] Also, the electron-transfer rate depends critical-
ly on the nature of the chemical linkage.[6,7] The surface
chemical modification must offer various manipulable pa-
rameters in order to regulate the electron transfer across the
molecule/electrode junctions. Surface modification of opti-
cally transparent indium–tin oxide (ITO) has thus received
much attention due to its practical relevance to organic elec-
tronic devices.[3,5]
Charge injection at an electrode surface is a critical step in
the operation of organic molecular electronic devices, such
[a] Dr. M. Morisue, N. Haruta, Dr. D. Kalita, Prof. Y. Kobuke
Graduate School of Materials Science
Nara Institute of Science and Technology
8916-5 Takayama, Ikoma, Nara 630-0192 (Japan)
Fax : (+81)743–72-6119
E-mail: kobuke@ms.naist.jp
[b] Dr. M. Morisue
Present address: Graduate School of Science and Technology
Department of Biomolecular Engineering
Kyoto Institute of Technology
Matsugasaki, Sakyo-ku, Kyoto 606-8585 (Japan)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author. It contains details
of blocking experiments, XPS survey spectra, and peak-deconvoluted
profiles in the Sn(3d) and Zn(2p) regions, as well as further photocur-
rent properties.
Molecular engineering in relation to photocurrent genera-
tion has provoked a great deal of effort in developing pho-
tovoltaics and in emulating natural photosynthetic systems.
Chem. Eur. J. 2006, 12, 8123 – 8135
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8123

Aprecise architecture of the
redox cascade on the molecu-
lar scale is one of the key fac-
tors for achieving an efficient
sequential electron-transfer re-
action. Therefore, molecular
layered systems, such as Lang-
muir–Blodgett films, self-as-
sembling monolayers (SAMs),
and layer-by-layer assemblies,
incorporating porphyrins or
fullerene, have been em-
ployed.[8–11] For the efficient
generation of a photocurrent,
an ITO electrode is advanta-
geous compared with a gold
electrode, since in the latter
electron transfer competes
with energy-transfer quenching
of the excited singlet state by
the surface plasmon in the vi-
cinity of the gold surface.[9a,b]
The efficiency of photocurrent
generation can, in principle, be
expected to be enhanced by
the ITO system, whereas the
quantum yield for the anodic
photocurrent on an ITO elec-
trode has been limited to
around 20% with even the
Scheme 1. a)–c) Potential diagrams of the surface states of an ITO electrode and redox potentials of the at-
tached porphyrin at various potentials; d) schematic illustration of the “special pair”-mimic porphyrin dimer
immobilized on an ITO electrode; e) chemical formulae of imidazolylporphyrins.
most elaborate porphyrin/fullerene/ITO systems.[9c,11b] In
contrast, quantum yields for cathodic photocurrent genera-
tion on gold electrodes modified with fullerene-terminated
SAMs of 50–51% were claimed in the reports of Imahori
et al.[8b] and Jen et al.[8c] The molecular systems attached to
ITO electrodes for anodic photocurrent generation still
remain a limiting factor with regard to improving charge-in-
jection efficiencies.
Despite its high carrier density, ITO is classified as an n-
type semiconductor.[12–14] The anodic photocurrent genera-
tion system should involve an electron injection from the
HOMO of the sensitizer or electron-mediator molecules
into the ITO electrode. In this step, the conduction band of
the ITO electrode will contribute to the anodic photocurrent
generation. The present study aims to evaluate the effect of
anchoring group(s) of porphyrin adsorbates on the band
structure of an ITO electrode through observing efficient
anodic photocurrent generation. Photocurrent efficiencies
may reflect differences in charge-injection barriers. On ap-
plying high cathodic potentials to ITO, the space charge
vanishes and the potential reaches a threshold, at which
point the anodic photocurrent falls to zero (Scheme 1a,b).
Above the threshold potential, defined as the flat-band po-
tential (Vfb), the photocurrent flow is inverted from the
anodic to the undesirable cathodic direction (Scheme
1c).[15,16] Therefore, we have explored the photocurrent
properties of porphyrin-bearing ITO electrodes in order to
observe possible modification of the charge-injection barri-
ers.
The SAM technique is a facile method for fabricating mo-
lecular systems.[17] SAM formation on ITO surfaces using
thiol, carboxylate, and phosphonate derivatives is well estab-
lished.[18] However, the effects of these anchoring groups on
electron-transfer processes across the molecule/electrode
junctions have not been fully examined, in spite of their
utmost importance with regard to the rational design of or-
ganic molecular devices. Herein, anodic photocurrent gener-
ation efficiencies are compared in order to elucidate the
effect of anchoring group(s) on apparent Vfb values of a flat
ITO electrode under various bulk pH conditions.
Inspired by natural photosynthetic systems, the use of por-
phyrin SAMs represents a promising basic strategy for the
conversion of light energy into an electric current.[8–11] The
primary electron donor in natural photosynthesis is consti-
tuted by a slipped-cofacial chlorophyll dimer, the so-called
“special pair” (SP), at the reaction center.[19] Analogously,
two imidazolyl–porphyrinatozinc units, Zn(ImP)s, mutually
A
coordinate through imidazolyl-to-zinc bonds to form a slip-
ped-cofacial dimer as the SP-mimic structure.[20] It has been
demonstrated that, due to the delocalization of the radical
cation over the two porphyrin rings, the SP-mimic cofacial
dimer plays a crucial role in photoinduced electron transfer,
by both accelerating charge separation and decelerating
charge recombination.[21] Successive self-organization/ring-
8124
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8123 – 8135

Porphyrin Assemblies on ITO Anodes
FULL PAPER
closing olefin metathesis processes of Zn(ImP)s on the elec-
N
trode surface can provide surface-grafted porphyrin assem-
blies with structural persistence of the organized array.[22] In
the present study, a titanium monolayer is deposited by
means of a surface sol–gel process. Subsequently, the SP-
mimic porphyrin dimer is organized to exhibit efficient pho-
tosensitization.
Results and Discussion
Deposition of a titanium monolayer on an ITO surface: A n
ITO surface was modified as depicted in Scheme 2. Details
of the steps are given in the Experimental Section. Reaction
of the surface terminal OH groups with metal alkoxides rep-
resents a facile method for modifying the nature of the ITO
surface.[23–25] Schwartz and co-workers attempted the modifi-
cation by treating ITO with tetra(tert-butoxy)tin and tetra-
(tert-butoxy)zirconium under ultra-high-vacuum condi-
tions.[23] Similarly, the surface sol–gel fabrication process can
provide a molecular metal oxide layer by treatment under
mild and ambient conditions.[24] To improve the surface nu-
Figure 1. Noncontact mode AFM images of the bare ITO surface (a) and
titanium-treated ITO (b).
AHCTREUNG
AHCTREUNG
AHCTREUNG
AHCTREUNG
À
cleophilicity, the In OH groups of the ITO surface were
À
converted into Ti OH groups by sol–gel treatment with tita-
ACHTREUNG
nium tetra-n-butoxide (Ti
A
of the O(1s) binding energy toward higher values suggests
that the electron-withdrawing effect of the metal on the
oxygen is decreased by the sol–gel process. The increased
core binding energy of the oxygen proves that the nucleo-
The sol–gel-modified ITO was characterized by using X-
ray photoelectron spectroscopy (XPS), and atomic force mi-
croscopy (AFM) images were obtained (Figure 1). The Ti-
Scheme 2. Procedure for the preparation of the SP-mimic porphyrin dimer on the ITO electrode surface. i) Abare ITO electrode was immersed in
100mm Ti(OnBu)4 in toluene/ethanol for 3 min. ii) After rinsing thoroughly with ethanol, the ITO was hydrolyzed in water and cleaned by UV/ozone ex-
posure. iii) The titanium-treated ITO was soaked in H2(ImPR) solution under appropriate conditions. iv) Zinc was introduced into the porphyrin in the
SAM by immersion in CHCl3 containing an aliquot of saturated Zn(OAc)2 in MeOH at 508C for 2 h. v) The SAM-modified ITO was soaked in 1mm Zn-
(ImPPh) in CH2Cl2 containing 20mm MeOH at room temperature for 24 h, and then rinsed with CH2Cl2 to form the coordination dimer. vi) The coordi-
nation dimer was covalently immobilized by soaking in a dilute solution of the Grubbs catalyst at room temperature for 10 min.
ACHTREUNG
AHCTREUNG
AHCTREUNG
ACHTREUNG
Chem. Eur. J. 2006, 12, 8123 – 8135
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8125

Y. Kobuke et al.
the UV/visible region (330–850 nm). From this, it can be
concluded that the titanium monolayer on the ITO does not
possess a band structure with a small energy gap.
The electrochemical behavior of the titanium-treated ITO
electrode was characterized by using K3[Fe(CN)6] as a bulk
redox probe.[29] No substantial change of the redox couple of
bulk K3[Fe(CN)6] was found in the voltammograms by using
the titanium-modified ITO working electrodes (see the Sup-
porting Information). From this it can be concluded that the
titanium layer does not insulate the redox process between
the ITO electrode and the outer redox site.
Organization of the porphyrin dimer as a “special pair”
mimic
Zn
ACHTREUNG
mersed in a 1.0mm solution of H
AHCTREUNG
room temperature for 40 h.[30] Thereafter, the substrate was
washed with ethanol and dichloromethane, and dried under
a stream of nitrogen. The porphyrin incorporated into the
SAM was metalated by gently refluxing at 508C for 2 h in
CHCl3 (4 mL) containing 0.2 mL of a saturated solution of
zinc acetate in MeOH.
Figure 2. XPS-deconvoluted profiles of bare ITO, sol–gel-modified ITO,
and H2(ImPPO3)-attached ITO (bottom to top) in the In(3d) region (a)
and the O(1s) region (b) observed at a take-off angle of 608.
ACHTREUNG
philicity of the surface terminal OH groups had been im-
proved. The angular dependence of the photoelectron inten-
sity (Figure 3) is strongly supportive of the location of the
SP-mimic dimer formation on Zn
The Zn(ImP)-derivatized ITO electrodes were immersed in
1.0mm Zn(ImPPh) in CH2Cl2/MeOH (9:1, v/v) for 20 h. Zn-
(ImP)-derivatized ITO:
AHCTREUNG
AHCTREUNG
(ImP)s form slipped-cofacial dimers with extremely high as-
sociation constants (3.3”1011 mÀ1) in noncoordinating sol-
vents through complementary imidazolyl-to-zinc coordina-
tion. For organization into a heterodimeric porphyrin with a
complementary Zn
(ImPPh) unit to the monomer is required. In 1.0mm Zn-
(ImPPh) solution in CH2Cl2, the coordination dimer is parti-
A
AHCTREUNG
AHCTREUNG
ally dissociated by the addition of 20mm MeOH as a coordi-
nating solvent. During soaking of the SAM-modified ITO
electrode for 24 h, the dissociated Zn
A
into the dimer on the Zn(ImP) SAM as a result of the com-
AHCTREUNG
Figure 3. Dependence on take-off angle of XPS spectra for the O(1s)
region of the sol–gel-modified ITO.
plementary coordination. The ITO substrate was then rinsed
with CH2Cl2 to decrease the MeOH ratio so as to drive the
formation of the SP-mimic porphyrin heterodimer on the
ITO electrode. The organized structure was covalently fixed
through ring-closing metathesis of the allyl side chains by
immersion in a dilute solution of the Grubbs catalyst in
CH2Cl2 for 10 min. Thorough washing with MeOH and
nucleophilic OH groups at the outer terminal positions on
the titanium-modified ITO electrode surface. At the same
time, inelastic mean free paths cannot completely eliminate
the background peak of In(3d3/2), suggesting that the thick-
G
ness of the titanium layer is sufficiently thin as expected
(Figure 2a). The disappearance of the peak at 531.8 eV after
porphyrin immobilization strongly suggests that the surface
nucleophilic terminal OH groups react with the porphyrin
anchoring groups (see below). Therefore, the surface sol–gel
process led to successful derivatization of the terminal OH
groups as required.
CH2Cl2 removed Zn(ImPPh) molecules adsorbed on the
A
substrate. The SP-mimic porphyrin dimer was thus con-
structed on the ITO electrode by the organization/metathe-
sis protocol. This methodology has already been established
for the construction of structurally robust, surface-grafted
multiporphyrin arrays.[22]
Each free-base porphyrin on the ITO electrode showed
the Soret band at 421–422 nm, which shifted to 424–426 nm
after the incorporation of zinc (Figure 4; see also Figures 7
and 12). The Q band at 516 nm, attributed to the free-base
porphyrin, was shifted to 550 nm, indicating that zinc had
been incorporated as the central metal ion in the monolayer
Cyclic voltammetry in
a buffered aqueous solution
showed no charge/discharge process characteristic of a TiO2
semiconductor[28] for the titanium-modified ITO. The ab-
sorption spectrum of the titanium-modified ITO showed no
detectable change compared with that of the bare ITO in
8126
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8123 – 8135

Porphyrin Assemblies on ITO Anodes
FULL PAPER
monolayer assembly showed characteristic peaks of the free-
base porphyrin at 397.6 and 399.3 eV in the N(1s) region
(Figure 5a).[31] The split N(1s) peaks were unified to a single
peak at 398.1 eV by the introduction of zinc due to structur-
al equivalence.[32] From peak deconvolution for the N(1s)
region, zinc atoms were estimated to be inserted into 80%
of the porphyrin centers on the ITO surface. The peak of
the nitrogen of the imidazolyl residue is somewhat unclear,
since the N(1s) peak may involve shake-up satellite in the
higher binding energy region. In the C(1s) region, the char-
À
acteristic peaks attributable to C=C and C N bonding ap-
peared at 286.0 and 288.5 eV, respectively, in addition to
that of the residual carbon at 284.4 eV (Figure 5b).[33] The
Zn(2p) peak was not available for the estimation of zinc in-
sertion, because immersion of the ITO electrode in a solu-
tion of zinc acetate led to adsorption of excess zinc ions on
its surface, judging from the large Zn(2p) peak.[32] The ratio
of P(2p) to N(1s) indicates that the SP-mimic dimer was
Figure 4. Absorption spectra of H2
A
G
Zn(ImPPh)/Zn(ImPPO3)/ITO. These spectra were obtained by back-
A
ACHTREUNG
ground subtraction.
formed by organization/metathesis of Zn
of the Zn(ImPPO3) layer. Consequently, H2
(20%), Zn(ImPPO3) in a tilted orientation (40%), and Zn-
(ImPPh)/Zn(ImPPO3) as the SP-mimic structure (40%)
ACHTREUNG
A
ACHTREUNG
AHCTREUNG
A
ACHTREUNG
were immobilized on the sol–gel-modified ITO surface.
Anodic photocurrent generation
Effect of the SP-mimic dimer structure: The SP-mimic modi-
fied ITO was mounted at a round cell window (6 mm diame-
ter) and exposed to a buffered aqueous solution containing
1.0m NaClO4 as electrolyte and 50mm hydroquinone (H2Q)
as a sacrificial electron donor. Steep rise and depletion were
observed as soon as the light irradiation was turned on and
off, respectively. Typical photoresponse patterns are shown
in Figure 6. The magnitude of the anodic photocurrent
showed a dependence on the applied potential and the exci-
tation wavelength. Prompt photoresponse persisted over at
least several hours during the experimental operations.
Figure 7 depicts the photocurrent action spectra and their
corresponding absorption spectra on the ITO electrode. The
maximum photocurrent at each ITO electrode was obtained
by excitation at 400–410 nm. Close similarity between the
Figure 5. XPS deconvoluted profiles before and after zinc incorporation
into H2
A
of Zn(ImPPO3)/ITO in the C(1s) region (b) and the P(2p) region (c).
ACHTREUNG
assemblies. For the monolayer assemblies on the ITO elec-
trode, the Soret band was found to show a bathochromic
shift, suggesting that the porphyrin planes are in a mutually
tilted orientation. Organization of the SP-mimic structure
resulted in a slight broadening of the Soret band, even
though 40% of the porphyrin afforded the SP-mimic dimer
as described below. Similar spectral changes in the course of
the organization/metathesis processes were observed for
each porphyrin anchor. Therefore, the organized structures
are identical, irrespective of the anchoring group, even
though the surface concentrations vary considerably.
XPS measurements were adopted for further structural
elucidation of Zn
N
N
ACHTREUNG
decreased drastically (Figure 2b). In the P(2p) region, obser-
vation of a unimodal peak indicates that the phosphonate
group exists as a single chemical species (Figure 5c). The
phosphonate group is assumed to react with the surface ter-
Figure 6. Photocurrent response of Zn
C
G
minal OH groups in linking to the ITO surface. H2(ImPPO3)
A
Chem. Eur. J. 2006, 12, 8123 – 8135
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8127

Y. Kobuke et al.
similar manner, photoexcitation at the Q band (S1) of Zn-
A
G
A
=
8.2–10.0%) at 100 mV. Combining these observations, it is
assumed that the photo-liberated electron hardly exceeds
the activation barrier for charge injection from the H2–por-
phyrin or the S1* state of the Zn–porphyrin. In contrast, the
higher potential of the S2* state is appropriate for charge in-
jection. This enhancement may stem from the shift of the
redox potentials. Furthermore, the SP-mimic dimer forma-
tion augments the anodic photocurrent, even though the ab-
sorption stays at a similar level, whereby F410 is improved
from 8.2% to 10%. The enhanced photocurrent generation
proves the advantage of the slipped-cofacial dimer in pro-
viding photoinduced charge separation.
Figure 7. Photocurrent action spectra of H2
A
A
N
ACHTREUNG
of 50mm hydroquinone in aqueous solution with 1.0m NaClO4 as sup-
porting electrolyte (pH 6.2) at 100 mV versus Ag/AgCl.
The potential diagram depicts the mechanism of the
anodic photocurrent generation (Scheme 3). It is known that
the conduction band (CB) for an ITO electrode is located at
Figure 8. Potential dependence of the anodic photocurrent of Zn-
A
N
action spectra and the corresponding absorption spectra
proves that the observed photocurrent was attributable to
photoexcitation of the surface-confined porphyrin. Zinc in-
corporation and dimer formation enhanced the anodic pho-
tocurrent. Simple comparison can be made on the basis of
the photocurrent quantum yield, F, which excludes the dif-
ference of the absorbances of the ITO surfaces.
Comparison of the photocurrent action spectra helps to
elucidate the charge-injection mechanisms (Figure 7 and
Table 1). Relatively low and similar quantum yields for
anodic photocurrent generation from both the S1* and S2*
Scheme 3. Potential diagram for anodic photocurrent generation of Zn-
A
N
absorption spectra. The S2 state of H2
the sake of clarity.
A
around À1.3 V versus Ag/AgCl.[10c] This value is consistent
with the reduction potential of the ITO electrode itself,
which is the negative limit of the potential window. In
CH2Cl2, the H2–porphyrin, the SP-mimic dimer, and the dis-
sociated monomer of the Zn–porphyrin are reduced at À1.2,
À1.55, and À1.54 V versus Ag/AgCl, respectively.[34] There-
fore, the S1 levels of the H2– and Zn–porphyrins are not suf-
ficient for electron transfer to the CB level. In contrast, the
S2 level of the Zn–porphyrin at À2.3 V is considerably
higher and exceeds the driving force for electron transfer to
the CB level. However, the S2 level of the H2–porphyrin at
À1.9 V may be less effective in generating an anodic photo-
current, considering that similar quantum yields are ob-
served for photoexcitation at the Soret and Q-band regions.
The high anodic photocurrent efficiency observed upon pho-
toexcitation at the Soret band of the Zn–porphyrin may be
attributed to hot injection from the S2 level directly into the
CB before internal conversion (IC) to the S1 level. On the
states for H2(ImPO3)/ITO are suggestive of low charge-in-
G
jection efficiency and/or high charge recombination. In a
Table 1. Quantum yield and IPCE (incident photon-to-current efficiency)
for the anodic photocurrent obtained by excitation at 410 nm (Soret
band) and 550 nm (Q band) in aqueous electrolyte containing H2Q
(50mm) at 100 mV versus Ag/AgCl (pH 6.2).
F410/F550 [%]
IPCE410/IPCE550 [%]
H
G
4.7/3.8
8.2/2.3
10.0/4.3
4.8/2.9
0.28/0.01
0.46/0.015
0.55/0.03
0.16/0.005
0.02/0.0006
Zn
Zn
Zn
Zn
0.9/0.02
8128
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8123 – 8135

Porphyrin Assemblies on ITO Anodes
FULL PAPER
other hand, the photocurrent from S1* states involves charge
injection into the Fermi level. The charge separation in the
S2* state points to strong coupling[35] between the band
levels of the porphyrin and ITO. Direct anchoring on the
ITO surface via a phosphonate group without any insulating
bonds is thus concluded to be satisfactory for strong cou-
pling between the S2* state and the CB level. For instance,
no steady-state fluorescence was detectable, supporting this
conclusion.
photocurrent is therefore limited by the redox process of
the H2Q sacrificial electron donor. These F410 values are
comparable to the values of 11–19.5% obtained for a por-
phyrin–fullerene dyad system on ITO electrodes.[9] There-
fore, the above values are surprisingly high considering the
fact that no redox cascade structure incorporating electron
donors or acceptors is employed.
On applying more anodic potentials, the anodic photocur-
rent increased. The concomitant shift of the CB level with
the Fermi level results in band bending, even though the
band-edge potential remains at the same level under certain
pH conditions (Scheme 1a–c). Therefore, a large anodic po-
tential can bend the CB structure to form a Schottky barrier
to suppress the backward electron-transfer process from the
CB to the vacant S0 level. On the other hand, the Fermi
level, without a band structure, is not effective for such a
unidirectional electron transfer. The phosphonate anchor,
promoting the hot injection from the S2* state to the CB,
may modify the CB structure at the ITO surface.
The slipped-cofacial porphyrin dimer, which accelerates
charge separation and retards charge recombination,[21] is
considered to contribute to efficient sequential electron
transfer across the junction Zn
the presence of H2Q. The H2(ImPPO3) site, coexisting with
Zn(ImPPh)/Zn(ImPPO3), may trap excitation energy as an
energy sink.[36] On the other hand, the monomer site of Zn-
(ImPPO3) possesses almost similar HOMO–LUMO levels
to those of Zn(ImPPh)/Zn(ImPPO3). The enhanced anodic
photocurrent from the S2* state of photoexcited Zn-
(ImPPh)/Zn(ImPPO3) as the SP-mimic points to effective
A
G
AHCTREUNG
A
ACHTREUNG
ACHTREUNG
A
ACHTREUNG
A
ACHTREUNG
electron transfer to the CB level. Therefore, the lateral
energy-transfer (EnT) process may be negligible in Zn-
pH variation and anchoring group effect: Variation of the
pH revealed the important influence on the surface state of
the ITO electrode in manipulating the surface dipole and
the band-edge potential. The pH effects of isophthalate and
thiolate anchoring groups on the photocurrent properties
were compared with that of the phosphonate anchor (Fig-
ures 9 and 10). Although the anodic photocurrent of the Zn-
A
N
S2 photoexcited Zn–porphyrin into the ITO levels with very
high quantum yield is the dominant process in the anodic
photocurrent generation.
The enlarged ionic radius of the porphyrin dimer is effec-
tive in providing a long-lived charge-separated state.[21] The
anodic photocurrent may be enhanced by the effect of the
cofacial dimer together with the effect of organized water
molecules in the vicinity of the ITO surface, assuming that
backward electron transfer in the aqueous solution occurs in
the deep “inverted” regime of the Marcus parabola. Besides,
the aqueous environment may affect the ITO surface, be-
cause to some extent the ITO surface remains “naked” after
immobilization of the SP-mimic porphyrin dimers on the
surface (see the blocking experiments in the Supporting In-
formation). The Gouy–Chapman–Stern model proposes that
hydrating water molecules in the vicinity of the metal oxide
electrode surface constitute a Helmholtz layer. This is sup-
ported by the observation of distinct effects of structured
water or water clusters in facilitating electron transfer at
metal oxide/water interfaces on the sub-femtosecond time-
scale, the so-called “wet-electron” states.[37,38] This assump-
tion may account for more effective sensitization at the
Soret band than that in the Q-band region in the photocur-
rent action spectra. Therefore, the S2 photoexcited singlet
state facilitates efficient charge injection directly into the
conduction band. Such an effect may be relevant to the im-
provement of anodic photocurrent through modification of
the pH conditions as discussed below.
A
N
Figure 9. Potential dependence of the anodic photocurrent of Zn-
(ImPPh)/Zn(ImPAc2)/ITO excited at 410 nm in the presence of 50mm hy-
G
ACHTREUNG
droquinone with 1.0m NaClO4 as supporting electrolyte at various pH
values.
cally on increasing the pH, the photocurrents of the Zn-
(ImPPh)/Zn
R
N
systems showed only a small improvement on varying the
pH. Fromherz and Arden have established the pH depend-
ence of Vfb.[13] Depending on the pH, the surface states of n-
type semiconductor electrodes are strongly affected by the
potential drop across the bulk medium in the Helmholtz
layer.[13,39] Comparison of the pH dependences for the an-
choring groups implies that the surface state strongly de-
pends on the surface anchoring group.
Potential variation: The anodic photocurrent quantum yield
(F410) of the Zn
N
N
on applying more anodic potentials and saturated at around
13–15% at +100–200 mV versus Ag/AgCl, which is close to
the oxidation potential of H2Q in the bulk (Figure 8). The
Chem. Eur. J. 2006, 12, 8123 – 8135
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8129

Y. Kobuke et al.
mechanism underpinning this phenomenon requires further
study.
The apparent Vfb value could be tuned by varying the an-
choring porphyrin. The apparent Vfb values for Zn
Zn(ImPS2)/ITO were more anodic than those for Zn-
(ImPPh)/Zn(ImPPO3)/ITO and Zn(ImPPh)/Zn(ImPAc2)/
ITO. As a consequence, larger anodic potentials must be ap-
plied to generate an anodic photocurrent in the Zn(ImPPh)/
Zn(ImPS2)/ITO system. The higher Schottky barrier, created
(ImPPh)/
AHCTREUNG
A
G
A
ACHTREUNG
AHCTREUNG
AHCTREUNG
by surface modification with phosphonate and isophthalate
groups, is therefore advantageous for anodic photocurrent
generation. With the aim of fabricating dye-sensitized solar
cells, this anchoring group effect has been verified on a TiO2
surface, although the results are not yet conclusive.[42,43] The
Figure 10. Potential dependence of the anodic photocurrent of Zn-
(ImPPh)/Zn(ImPS2)/ITO excited at 410 nm in the presence of 50mm hy-
droquinone with 1.0m NaClO4 as supporting electrolyte at various pH
A
ACHTREUNG
present study has demonstrated that a Zn(ImPPO3)-based
A
values.
SP-mimic can modify the band-edge potential, and so en-
hances the n-type semiconductor characteristics of the ITO
electrode. This allows an efficient conversion of the photo-
excited state and leads to increased electric current genera-
tion. Phosphonate-modified ITO thus promotes hot injec-
tion to afford a larger F410 for the anodic photocurrent at
100 mV versus Ag/AgCl (pH 6.2) by a factor of two and by
a factor of ten compared with those of isophthalate-modi-
fied and thiolate-modified ITO, respectively (Figure 12, and
The apparent Vfb for the porphyrin-modified ITO was
evaluated from the zero-current intercept in the I–V plots
(Figures 8, 9, and 10). Apparent Vfb is plotted as a function
of pH in Figure 11. At alkaline pH, the apparent Vfb is shift-
ed to more cathodic potentials. As the pH increases, the
Figure 11. Plot of the apparent flat-band potential (Vfb) as a function of
*
^
pH conditions for Zn
G
(ImPR)/ITO (R = ÀPO3 ( ), -Ac2 ( ),
~
and -S2 ( )).
Figure 12. Photocurrent action spectra of Zn
A
ACHTREUNG
(a), Zn(ImPPh)/Zn(ImPAc2)/ITO (b), and Zn
A
G
A
ACHTREUNG
(c) in the presence of 50mm hydroquinone in aqueous solution contain-
ing 1.0m NaClO4 as supporting electrolyte (pH 6.2) at 100 mV versus Ag/
AgCl.
band-edge potentials are therefore assumed to rise and pro-
vide higher Schottky barriers, suppressing backward electron
transfer from the CB to the porphyrin radical cation (Sche-
me 1a–c). Since the slope in the Vfb–pH plot exceeds the
Nernstian slope of H2Q (À59 mV(pH)À1),[40] the efficiency
of anodic photocurrent generation is determined not only
by the redox potential of H2Q, but also by the ITO surface
state. The Vfb value of intrinsic n-type semiconductors nor-
mally changes linearly as a function of pH.[39] Thus, a rela-
Figure S6 in the Supporting Information). The effect of the
different anchoring groups on the efficiency of the photocur-
rent generation should be related to the nature of the chem-
ical linkage and the surface state, and therefore to the
charge-injection step. The rate-determining step is therefore
electron injection into the ITO anode, that is, the hot-injec-
tion process from the S2 photoexcited state.
tively linear pH dependence of Vfb for Zn
A
ACHTREUNG
band of the ITO electrode and porphyrin are strongly cou-
pled via the phosphonate group (Figure 11). On the other
hand, the existence of inflection points may be attributed to
a protonation–deprotonation equilibrium of the physisorbed
species. This may imply that the surface charge of ITO af-
fects the band-edge potentials.[41] However, the detailed
Unusual concentration effect: It is important to note that
measurements of Soret band absorbance on ITO indicated
that Zn
Zn(ImPAc2) or Zn
current action spectra and their corresponding absorption
ACHTREUNG
A
ACHTREUNG
8130
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8123 – 8135

Porphyrin Assemblies on ITO Anodes
FULL PAPER
on the flat ITO electrode. The effect of pH suggests that the
surface state of the ITO is dependent on the anchoring spe-
cies. Thus, the anchoring species has an impact on the mag-
nitude of the anodic photocurrent.
We consider that the high efficiency observed for photo-
current generation may stem not only from the phospho-
nate-modulated ITO surface, but also from the increased
radius of the radical cation.[21] The especially high quantum
yield for the anodic photocurrent from the S2* state com-
pared with that from the S1* state would suggest that suffi-
ciently fast charge injection occurs before internal conver-
sion to the S1* state. Comparison of the photocurrent action
spectra supports the view that the difference in driving
forces from the S2* and S1* states to the ITO conduction
band level is critical for the anodic photocurrent generation.
We assume that the effect of the cofacial dimer is enhanced
in aqueous media, providing a deep “inverted” Marcus
regime for the SP-mimic dimer and a “wet-electron” effect
to stabilize the liberated electron on the sol–gel-treated ITO
surface.
Figure 13. Experimental setup for photocurrent generation.[10c]
The phenylphosphonyl group is especially effective in fa-
cilitating strong electronic communication between the S2*
state of the Zn–porphyrin and the ITO conduction band, as
evidenced by XPS observation, the photocurrent action
spectra, and the pH variation of the apparent flat-band po-
spectra of the SP-mimic on the ITO electrode. This compari-
son clearly shows the difference in the efficiencies of photo-
current generation (Table 1). It is generally observed that F
values increase on dilution of the porphyrin surface concen-
tration in conventional monolayer assemblies.[44] This is
probably due to the decrease in the number of deactivation
pathways.[44] However, the SP-mimic porphyrin assemblies
are ordered in a slipped-cofacial orientation and these sys-
tems appear to be insensitive to such processes, even when
the surface concentration is increased. Phosphonate-modi-
fied ITO serves as a good electron acceptor and so should
enhance the effect of the SP-mimic dimer in the charge-sep-
aration step, and this should in turn lead to amplification of
the anodic photocurrent from the S2 photoexcited state. As
a consequence of its high surface concentration and excep-
tentials. The Zn(ImPPO3)/ITO scaffold can provide facile
A
immobilization of multiporphyrin arrays for extensive pho-
tovoltaic applications.[22] The present insights are important
for understanding the essential principle underlying interfa-
cial electron transfer at the ITO electrode. This approach
allows the rational design of molecule/ITO junctions for fur-
ther elaborate self-assembling systems. An artificial photo-
synthetic reaction center based on the “special pair”-mimic
structure on the ITO electrode is currently under construc-
tion with a view to achieving more efficient anodic photo-
current generation based on redox cascade alignment for se-
quential electron transfer.
tionally high efficiency, the photoexcitation of Zn
Zn(ImPPO3)/ITO generates remarkably large photocur-
rents.
A
ACHTREUNG
Experimental Section
General: 1H NMR spectra were recorded on JEOL JNMEX 270
(270 MHz) or JEOL ECP 600 (600 MHz) spectrometers; chemical shifts
are referenced to TMS as an internal standard. 31P NMR spectra were ac-
quired under off-spinning conditions and are referenced to 85% aqueous
phosphoric acid inside a sealed capillary as an external standard. The
Conclusion
Anovel surface fabrication approach utilizing a surface sol–
gel process and molecular organization has been performed
with the aim of obtaining efficient photoexcited charge in-
jection. The nucleophilicity of the terminal OH groups of
ITO has been successfully improved by sol–gel modification
procedure for the synthesis of Zn
A
where.[45]
Diethyl 4-formylbenzphosphonate (1): Asolution of 4-bromobenzalde-
hyde (1.5 g, 8 mmol) in toluene (3 mL) was concentrated to two-thirds of
its original volume for the azeotropic removal of water. Diethyl phos-
phite (1.2 g, 1.1 equiv) and triethylamine (1.2 mL) were then added. The
mixture was transferred to a Schlenk flask, degassed by repeated freeze–
pump–thaw cycles, and purged with argon. Tetrakis(triphenylphosphine)-
palladium(0) (600 mg) was then added and the mixture was stirred at
908C for 7 h.[46] In the course of the coupling reaction, the reaction mix-
ture changed from brown to pale greenish-brown. Diethyl ether was then
added to the reaction mixture and the precipitate that formed was fil-
tered off. Diethyl 4-formylbenzphosphonate was purified by Kugelrohr
distillation and was obtained as a slightly yellowish liquid. The coupling
with Ti(OnBu)4, such that monolayer assemblies could be
U
readily immobilized with appropriate anchoring groups.
Exploration of the photocurrent properties has allowed
elucidation of the role of the conduction band in the photo-
excited charge-injection process. The increase in band-edge
potential due to the introduction of Zn(ImPPO3) led to an
increase in the anodic photocurrent to 15% at F410. This
G
was achieved even without incorporating a redox gradient
Chem. Eur. J. 2006, 12, 8123 – 8135
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8131

Y. Kobuke et al.
reaction was successfully accomplished without any protection of the al-
dehyde residue, giving the product in a fair yield (0.79 g, 41%). H NMR
(270 MHz, CDCl3): d = 1.34 (t, J = 7.0 Hz, 6H; POCH2CH3), 4.07–4.24
(m, 4H; POCH2CH3), 7.96–8.00 (m,
The synthesis of 4-{5-(1-methylimidazol-2-yl)-10,20-bis(3-allyloxypropyl)-
1
porphyrin-15-yl}benzphosphonic acid, H2
Scheme 4.
A
4H; Ph), 10.09 ppm (s, 1H; -CHO);
31P NMR (242 MHz, CDCl3):
16.89 ppm.
d =
5-(1-Methylimidazol-2-yl)-10,20-
bis(3-allyloxypropyl)-15-(4-diethoxy-
phosphonylphenyl)porphyrin (2): 2-
Formyl-1-methylimidazole
(0.18 g,
1 equiv), aldehyde 1 (0.41 g, 1 equiv),
and meso-(3-allyloxypropyl)dipyrro-
methane (0.82 g, 2 equiv) were dis-
solved in CHCl3 (850 mL) and stirred
at room temperature in the presence
of
trifluoroacetic
acid
(TFA)
(3 equiv) for 3 h. Chloranil (4 equiv)
and triethylamine (1 equiv) were
then added to the reaction mixture to
oxidize the porphyrinogen intermedi-
ate, and the mixture was stirred over-
night. The crude mixture was subject-
ed to column chromatography on
Scheme 4. Synthesis of H2(ImPPO3) (3).
A
silica gel (eluents: CHCl3 and CHCl3/acetone, 2:1, v/v) three times. The
desired porphyrin 2 was isolated as a purple solid (86 mg, 6%). H NMR
5-(1-Methylimidazolyl)-10,20-bis(3-allyloxypropyl)-15-(3,5-diethoxycarbo-
nylphenyl)porphyrinatozinc (5): Porphyrin 5 was synthesized by TFA-cat-
alyzed cyclization. TFA(3 equiv) was added to a mixture of 2-formyl-1-
methylimidazole (0.18 g, 1 equiv), 4 (0.42 g, 1 equiv), and meso-(3-allyloxy-
propyl)dipyrromethane (0.82 g, 2 equiv) in CHCl3 (850 mL). The reaction
mixture was stirred at room temperature for 4 h, chloranil (4 equiv) was
added, and stirring was continued for a further 10 h. The reaction mixture
was then neutralized by washing with saturated aqueous sodium hydro-
gencarbonate solution and brine. The crude free-base porphyrin was ob-
tained by using column chromatography on silica gel eluting with CHCl3/
acetone (9:1, v/v) (yield 133 mg, 9%). The crude porphyrin was metalat-
ed by treating it with an aliquot of a saturated solution of zinc acetate in
MeOH (3 mL). The title zinc porphyrin 5 was obtained by using column
chromatography on silica gel, eluting with CHCl3/acetone (9:1, v/v). 1H
NMR (270 MHz, CDCl3): d = 1.38 (t, J = 7.0 Hz, 3H; -COOCH2CH3),
1.63 (t, J = 7.0 Hz, 3H; -COOCH2CH3), 1.69 (s, 3H; imidazolyl-CH3),
2.14 (d, J = 1.6 Hz, 1H; 4-imidazolyl), 3.09 (m, 4H; -CH2CH2O-allyl),
1
(270 MHz, CDCl3): d = À2.69 (s, 2H; inner pyrrole), 1.34 (t, J = 5.0 Hz,
6H; POCH2CH3), 2.78 (t, J = 5.0 Hz, 4H; -CH2CH2O-allyl), 3.39 (s, 3H;
imidazolyl-CH3), 3.65 (t, J
5.0 Hz, 4H; -CH2(CH2)2O-allyl), 4.38–4.44 (m, 4H; POCH2CH3), 5.13 (t,
7.2 Hz, 4H; -OCH2CH=CH2), 5.26 (dd, J 13.5, 1.9 Hz, 2H;
= 5.0 Hz, 4H; -CH2O-allyl), 4.07 (t, J =
ACHTREUNG
J
=
=
-CH2CH=CHH), 5.26 (dd, J = 17.6, 1.9 Hz, 2H; -CH2CH=CHH), 6.02–
6.07 (m, 2H; -OCH2CH=CH2), 7.47 (d, J = 1.4 Hz, 1H; 5-imidazolyl),
7.68 (d, J = 1.4 Hz, 1H; 4-imidazolyl), 8.1–8.4 (m, 4H; Ph), 8.80 (d, J =
4.9 Hz, 4H; b-pyrrole), 9.52 ppm (dd, J = 11.1, 5.1 Hz, 4H; b-pyrrole);
MALDI-TOF MS: m/z calcd for 2: 798.37; found: 799.54.
H2(ImPPO3) (3): Aflask charged with 2 (85 mg) and potassium bromide
T
(1.7 g) was purged with argon. Dry triethylamine (2.8 mL) and dry DMF
(16 mL) were then added, followed by chlorotrimethylsilane (TMSCl,
2.6 mL), and the mixture was stirred at 60–658C for 24 h.[47] Awhite pre-
cipitate that formed in the course of the deprotection was filtered off.
After evaporation of the solvent, the mixture was dispersed in water
(60 mL) and concentrated HCl (0.5 mL) and stirred at room temperature
for 3 h. The aqueous phase was neutralized with sodium hydrogencarbon-
ate, the desired porphyrin was extracted with CHCl3, and the combined
extracts were dried over anhydrous sodium sulfate. The porphyrin was
extracted with MeOH and purified by reprecipitation from diethyl ether.
The desired porphyrin was obtained almost quantitatively. 1H NMR
(270 MHz, CD3OD): d = 2.59 (t, J = 6.5 Hz, 4H; -CH2CH2O-allyl), 3.41
(s, 3H; imidazolyl-CH3), 3.58 (t, J = 5.7 Hz, 4H; -CH2CH2O-allyl), 3.96
(d, J = 5.7 Hz, 4H; -OCH2CH=CH2), 4.91 (t, J = 7.3 Hz, 4H; -CH2-
3.94 (t, J = 7.3 Hz, 4H; -CH2
-OCH2CH=CH2), 4.48, 4.68 (q, J = 7.0 Hz, 4H; -COOCH2CH3), 5.25 (t,
7.3 Hz, 4H; -CH2(CH2)2O-allyl), 5.34 (d, 10.3 Hz, 2H;
(CH2)2O-allyl), 4.24 (d, J = 5.4 Hz, 4H;
J
=
A
J =
-CH2CH=CHH), 5.44 (d, J = 4.86 Hz, 2H; -CH2CH=CHH), 5.44 (d, J=
10.5 Hz, 2H; -OCH2CH=CHH), 5.45 (s, 1H; 5-imidazolyl), 5.55 (d, J =
1.6 Hz, 2H; b-pyrrole), 6.13–6.24 (m, 2H; -OCH2CH=CH2), 8.89 (d, J =
1.6 Hz, 2H; b-pyrrole), 8.94 (s, 1H; 4-Ph), 8.98 (d, J
= 4.6 Hz, 2H;
b-pyrrole), 9.21, 9.54 (s, 2H; 2,6-Ph), 9.66 ppm (d, J = 4.6 Hz, 2H; b-pyr-
role); MALDI-TOF MS: m/z calcd for 5: 868.29; found: 867.996.
H2(ImPAc2) (6): Diethyl Zn–porphyrinyl isophthalate 5 (100 mg) was dis-
U
solved in THF/ethanol (13:7, v/v) at room temperature and stirred in the
presence of 8n NaOH for 3 h. The crude mixture was then washed with
brine and extracted with chloroform. The organic layer was treated with
p-toluenesulfonic acid monohydrate (p-TsOH) to displace the central
zinc. The bright-blue organic layer was washed with a large excess of
brine, which resulted in a color change to red due to neutralization, and
dried over anhydrous sodium sulfate. The desired porphyrin was obtained
as a brownish-red solid (35%). 1H NMR (270 MHz, [D6]DMSO, 608C):
d = À2.76 (s, 2H; inner pyrrole), 2.08 (s, 3H; imidazolyl-CH3), 3.5 (br,
(CH2)2O-allyl), 5.11 (d, J = 12.2 Hz, 2H; -CH2CH=CHH), 5.27 (dd, J =
15.4 Hz, 2H; -CH2CH=CHH), 5.89–5.95 (m, 2H; -OCH2CH=CH2), 7.56
(s, 1H; 5-imidazolyl), 7.67 (d, J = 0.8 Hz, 1H; 4-imidazolyl), 8.07–8.14
(m, 4H; Ph), 8.10 (d, J = 30.5 Hz, 4H; b-pyrrole), 9.49 ppm (d, J =
30.5 Hz, 4H; b-pyrrole); MALDI-TOF MS: m/z calcd for 3: 742.30;
found: 742.26.
3,5-Diethoxycarbonylbenzaldehyde (4): Diethyl 5-(hydroxymethyl)isoph-
thalate was obtained from benzene-1,3,5-triethoxycarbonate by reduction
with LiBH4 in dry THF. Subsequent oxidation of 3,5-triethoxycarbonyl-
benzyl alcohol with PCC (pyridinium chlorochromate) in dry chloroform
afforded 3,5-diethoxycarbonylbenzaldehyde. In the literature proce-
dure,[48] oxidation to the aldehyde was achieved with CAN (cerium(IV)
ammonium nitrate) as the oxidant. In our trial, we found it difficult to
prevent over-oxidation to the corresponding carboxylic acid with CAN.
Instead, oxidation with PCC was found to smoothly afford the desired
product (94%).
4H; -CH2CH2O-allyl), 4.0 (br, 4H; -CH2
-OCH2CH=CH2), 4.8 (br, 4H; -COOCH2CH3), 5.2 (br, 4H; -CH2-
(CH2)2O-allyl), 5.3 (br, 2H; -CH2CH=CHH), 6.5 (br, 2H; -CH2CH=
U
AHCTREUNG
CHH), 7.54 (s, 1H; 4-imidazolyl), 7.77 (s, J = 1.6 Hz, 1H; 5-imidazolyl),
8.68 (br, 3H; Ph), 8.8 (br, 4H; b-pyrrole), 9.6 ppm (br, 4H; b-pyrrole);
MALDI-TOF MS: m/z calcd for 6: 750.32; found: 752.592.
The synthesis of 5-{5-(1-Methylimidazolyl)-10,20-bis(3-allyloxypropyl)-
porphyrin-15-yl}isophthalic acid, H2(ImPAc2) (6), is shown in Scheme 5.
A
8132
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8123 – 8135

Porphyrin Assemblies on ITO Anodes
FULL PAPER
1H; CHO); 13C NMR (67 MHz,
CDCl3): d = 31.5, 129.7, 135.0, 137.2,
139.5, 190.7 ppm.
3,5-Di(acetylthiomethyl)benzalde-
hyde (10): Asolution of 3,5-di(bro-
momethyl)benzaldehyde (9) (1.0 g,
3.4 mmol) and potassium thioacetate
(0.94 g, 8.2 mmol, 2.4 equiv) in ace-
tone (20 mL) was stirred at room
temperature and then refluxed for
3 h. The reaction was then quenched
by the addition of water (40 mL), and
the mixture was extracted with ethyl
acetate (2”100 mL). The combined
organic layers were dried over anhy-
drous sodium sulfate. The crude mix-
ture was purified by using column
chromatography on silica gel using
hexane/ethyl acetate (4:1, v/v) as
eluent to furnish the title compound
Scheme 5. Synthesis of H2(ImPAc2) (6).
G
Diethyl 5-(5,5-dimethyl-[1,3]dioxan-2-yl)isophthalate (7): Amixture of
3,5-diethoxycarbonylbenzaldehyde (4) (3.1 g, 12 mmol), 2,2-dimethyl-1,3-
propanediol (1.6 g, 15 mmol), and p-TsOH (240 mg, 1.3 mmol) was re-
fluxed in benzene (50 mL) in a two-necked flask equipped with a Dean–
Stark trap and water condenser under an argon atmosphere. After reflux-
ing at 1008C for 4 h, the water that had collected in the Dean–Stark trap
was separated. The mixture was washed with saturated aqueous sodium
hydrogencarbonate and water, and the organic layer was dried over an-
hydrous sodium sulfate. The crude product was purified by using column
chromatography on silica gel (eluent: hexane/ethyl acetate, 2:1, v/v) to
as a pale-yellow solid (yield 0.82 g, 85%). TLC: Rf
= 0.37 (hexane/
EtOAc, 4:1, v/v); 1H NMR (270 MHz, CDCl3): d = 2.37 (s, 6H; OAc),
4.14 (s, 4H; CH2-ester), 7.49 (d, J = 1.62 Hz, 1H; Ph), 7.68 (d, J =
1.62 Hz, 2H; Ph), 9.96 ppm (s, 1H; CHO); 13C NMR (67 MHz, CDCl3):
d = 30.4, 32.7, 128.7, 135.0, 136.9, 139.3, 191.4, 194.3 ppm.
H2(ImPS2) (11): Amixture of 10 (0.56 g, 2.0 mmol, 1.0 equiv), meso-(3-al-
U
lyloxypropyl)dipyrromethane (0.98 g, 4.0 mmol, 2.0 equiv), and 2-formyl-
1-methylimidazole (0.22 g, 1.0 mmol, 1.0 equiv) was dissolved in chloro-
form (400 mL). The mixture was degassed by bubbling nitrogen through
it for 10 min and then TFA(0.15 mL, 2.0 mmol, 2.0 equiv) was added by
means of a syringe over 30 s. The mixture was stirred at room tempera-
ture under a nitrogen atmosphere and then neutralized by adding tri-
ethylamine (1.1 mL, 2.0 mmol) and p-chloranil (0.74 g, 3.0 mmol) in THF
(20 mL). The resulting mixture was then stirred at room temperature for
6 h. One-third of the solvent was removed in a rotary evaporator and the
crude purple residue was chromatographed on silica gel using CHCl3 as
an eluent to yield the title compound 11 in almost 98% purity. The com-
pound was further chromatographed on silica gel using CHCl3/acetone
(10:1, v/v) as eluent. Finally, the material was passed through a column
of basic alumina to afford the pure title porphyrin (yield 0.19 g, 12%).
1
furnish the title compound 7 as a white solid (yield 4.1 g, 98%). H NMR
(270 MHz in CDCl3): d = 1.30 (s, 6H; CH3-acetal), 1.41 (t, J = 7.3 Hz,
6H; CH3-ester), 3.79 (dd, J = 35.37, 10.26 Hz, 4H; -CH2), 4.42 (q, 4H;
CH2-ester), 5.41 (s, 1H; -CH), 8.36 (dd, J = 1.62, 0.54 Hz, 2H; Ph),
8.67 ppm (t, J = 1.62 Hz, 1H; Ph).
1-(5,5-Dimethyl-[1,3]dioxan-2-yl)-3,5-di(hydroxymethyl)benzene (8):
A
solution of diethyl 5-(5,5-dimethyl-[1,3]dioxan-2-yl)isophthalate (7)
(4.0 g, 12 mmol) in dry diethyl ether (300 mL) was added dropwise to
lithium aluminum hydride (0.90 g, 24 mmol) under a nitrogen atmos-
phere. The mixture was stirred at room temperature for 5 h, then cooled
to 08C, and quenched by the addition of water (10 mL). It was then treat-
ed with 5% sulfuric acid until a clear solution was obtained. The ethereal
layer was separated and the aqueous layer was extracted with ethyl ace-
tate (2”100 mL). The combined organic layers were washed with water
(100 mL) and dried over anhydrous sodium sulfate. The crude material
obtained was used for the next step without any further purification
1
TLC: Rf = 0.41 (CHCl3/acetone, 10:1, v/v); H NMR (270 MHz, CDCl3):
d = À2.70 (s, 2H; inner pyrrole), 2.41 (s, 6H; -SCOCH3), 2.792 (t, J =
6.5 Hz, 4H; -CH2CH2O-allyl), 3.41 (s, 3H; imidazolyl-CH3), 3.66 (t, J =
5.7 Hz, 4H; -CH2CH2O-allyl), 4.08 (dd, J = 5.4, 1.4 Hz, 4H; -OCH2CH=
CH2), 4.39 (d, J = 3.0 Hz, 4H; PhCH2SAc), 5.11 (t, J = 7.3 Hz, 4H;
-CH2
5.42 (dd, J
N
= 10.3, 1.4 Hz, 2H; -CH2CH=CHH),
=
1
(yield 3.0 g, 99%). H NMR (270 MHz, CDCl3): d = 0.79 (s, 3H; -CH3),
-OCH2CH=CH2), 7.47 (d, J = 0.8 Hz, 1H; 5-imidazolyl), 7.65 (s, 1H; 4-
phenyl), 7.68 (d, J = 0.8 Hz, 1H; 4-imidazolyl), 7.93, 8.03 (s, 2H; 2,5-
phenyl), 8.81 (dd, J = 10.0, 4.6 Hz, 4H; b-pyrrole), 9.52 ppm (dd, J =
15.1, 5.1 Hz, 4H; b-pyrrole); 13C NMR (67 MHz, CDCl3): d = 29.75,
30.44, 31.36, 33.43, 34.52, 37.85, 69.14, 71.95, 104.31, 116.71, 119.29,
119.59, 121.23, 127.77, 128.20, 128.35, 129.10, 131.87, 133.73, 134.91,
136.11, 136.20, 143.00, 148.71, 194.64; MALDI-TOF MS: m/z calcd.:
1.28 (s, 3H; CH3), 2.79 (br, 2H; -OH), 3.73 (dd, J = 35.37, 10.26 Hz,
4H; -CH2), 5.35 (s, 1H; -CH), 4.57 (s, 4H; CH2-ester), 7.24 (s, 1H; Ph),
7.34 ppm (d, J = 0.54 Hz, 2H; Ph).
3,5-Di(bromomethyl)benzaldehyde (9): Asolution of 1-(5,5-dimethyl-
[1,3]dioxan-2-yl)-3,5-di(hydroxymethyl)benzene (8) (0.75 g, 3.0 mmol) in
toluene (20 mL) was refluxed with 48% aqueous HBr (18.8 mL) for
3.5 h. The reaction mixture was then diluted with water (20 mL). The or-
ganic layer was separated and the aqueous layer was extracted with ethyl
acetate (2”100 mL). The combined organic layers were washed with 5%
aqueous sodium hydrogensulfite and water, and then dried over anhy-
drous sodium sulfate. The crude material was dissolved in chloroform
(10 mL) and stirred with TFA(2.0 mL) and water (1.0 mL) for 20 h at
room temperature. Aqueous sodium hydrogencarbonate (5%, 50 mL)
was added to the reaction mixture, and the organic phase was separated
and dried over anhydrous sodium sulfate. The crude material was puri-
fied by using column chromatography on silica gel using n-hexane/ethyl
acetate (4:1, v/v) as eluent to furnish the title compound 9[49] as a light-
yellow solid (yield 0.85 g, 98%). TLC: Rf = 0.51 (hexane/ethyl acetate,
4:1, v/v); 1H NMR (270 MHz, CDCl3): d = 4.53 (s, 4H; CH2-ester), 7.70
(d, J = 1.89 Hz, 1H; Ph), 7.84 (d, J = 1.62 Hz, 2H; Ph), 10.01 ppm (s,
839.08; found: 839.9; UV/Vis: lmax (CH2Cl2)
= 418, 516, 551, 592,
649 nm; fluorescence: lem (CHCl3) = 654, 720 nm (lex = 516 nm).
The synthesis of 5-(1-methylimidazol-2-yl)-10,20-bis(3-allyloxypropyl)-15-
(3,5-diacetylthiomethylphenyl)porphyrin, H2
Scheme 6.
A
Procedures for ITO modification: The ITO substrate (commercially
available from Geomatic Co., 2200 Š thick with a sheet resistance of
10 W per square) was cleaned by successive sonication in acetone, etha-
nol, and Milli-Q water. The metal species were estimated to be present
in a ratio of 93.1% In to 6.9% Sn from XPS measurements (In:Sn
ꢀ 12:1). The surface chemical composition showed a good agreement
with the analysis reported by Dunphy, Armstrong et al.[27b]
The ITO substrate was soaked in a solution of Ti
A
chased from Nacalai Tesque Co.) in toluene/ethanol (1:1, v/v) for
Chem. Eur. J. 2006, 12, 8123 – 8135
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8133

Y. Kobuke et al.
light irradiation from a 150 W xenon
arc lamp (Hamamatsu Photonics,
L2274 light source equipped with a
C7535 power supply and
a C4251
starter). The ITO working electrode
was irradiated from behind with exci-
tation light passed through a mono-
chromator (Shimadzu, SPG-120S)
(Figure 13). The reproducibility of all
photocurrent properties was con-
firmed by performing at least dupli-
cate measurements.
X-ray photoelectron spectroscopy
(XPS): X-ray photoelectron spectra
were obtained by using an X-ray pho-
toelectron
AXIS 165) equipped with
spectrometer
(Kratos,
mono-
Scheme 6. Synthesis of H2(ImPS2) (11).
A
a
chromatic AlKa source at 1486.6 eV.
Take-off angles of 608 and 158 were adopted to observe the ITO surface
species. The ITO samples were electrically contacted with the spectrome-
ter and the charge was externally neutralized using a charge neutralizer
(applied potential: 1.85 V) to avoid peak broadening and distortion. The
binding energies were normalized to the residual C(1s) peak at 284.4 eV
3 min.[24] It was then thoroughly rinsed with a large excess of ethanol and
then with water to hydrolyze the n-butoxy groups on the surface. Prior to
SAM formation, the ITO surface was cleaned by exposure to UV/ozone
for 15 min.
The morphologies of the ITO surface before and after Ti(OnBu)4/hydrol-
G
À
and the In(3d) peak at 443.9 eV attributable to In O bonding. The peak
ysis treatment and subsequent hydrolysis were observed by means of
noncontact atomic force microscopy (AFM, Seiko Instruments SPA400)
(Figure 1). The roughness factor was estimated to be 1.06 from the mean
area per apparent area. This value is identical to that estimated before
the titanium treatment, indicating that the titanium monolayer formation
was not accompanied by undesirable domain formation and that the flat-
ness of the surface of the ITO was maintained.
profiles were deconvoluted by a Gaussian–Lorenztian (70:30) function.
In the In(3d) and Sn(3d) regions, spin multiplicity was taken into consid-
eration to fit multiplet splitting with a 3:2 population ratio for the 3d5/2
and 3d3/2 orbitals. In a similar way, a 2:1 population ratio for the Zn
and Zn(2p1/2) orbitals was applied for peak deconvolution.
A
AHCTREUNG
Zn
1.0mm solution of H2
v/v) for 120 h. The central zinc was introduced by the same procedure as
described for the Zn(ImPPO3) monolayer. The Zn(ImPAc2) monolayer
on the titanium-treated ITO was shown to be robust during the immobili-
zation procedures and photocurrent measurements. On the other hand,
ACHTREUNG
ACHTREUNG
Acknowledgements
A
ACHTREUNG
The authors thank Mr. Yasuo Okajima (Nara Institute of Science and
Technology) for carrying out the XPS measurements. This work was sup-
ported by a Grant in Aid of Scientific Reasearch (A) (No. 15205020)
from the Ministry of Education, Culture, Sports, Science and Technology,
Japan (Monbu Kagakusho).
Zn(ImPAc2) adsorbed on the bare ITO surface was completely cleaved
U
in aqueous solution. For the conventional immobilization of carboxylates,
DCC (dicyclohexylcarbodiimide as a condensing agent) condensation has
been employed.[50] Successful covalent immobilization was thus achieved
by the titanium treatment due to improved surface nucleophilicity.
[1] D. M. Adams, L. Brus, C. E. D. Chidsey, S. Creager, C. Creutz, C. R.
Kagan, P. V. Kamat, M. Lieberman, S. Lindsey, R. A. Marcus, R. M.
Metzger, M. E. Michel-Beyerle, J. R. Miller, M. D. Newton, D. R.
Rolison, O. Sankey, K. S. Schanze, J. Yardley, X. Zhu, J. Phys.
Chem. B 2003, 107, 6668.
[2] H. Ishii, K. Sugiyama, E. Ito, K. Seki, Adv. Mater. 1999, 11, 605.
[3] a) F. Nüesch, F. Rotzinger, L. Si-Ahmed, L. Zuppiroli, Chem. Phys.
Lett. 1998, 288, 861; b) M. Carrara, N. Nüesch, L. Zuppiroli, Synth.
Met. 2001, 121, 1633.
[4] X. Crispin, V. Geskin, A. Crispin, J. Comil, R. Lazzaroni, W. R.
Salaneck, J.-L. BrØdas, J. Am. Chem. Soc. 2002, 124, 8131.
[5] a) J. Guo, N. Koch, J. Schwartz, S. L. Bernasek, J. Phys. Chem. B
2005, 109, 3966; b) E. L. Hanson, J. Guo, N. Koch, J. Schwartz, S.
Bernasek, J. Am. Chem. Soc. 2005, 127, 10058.
[6] a) S. Creager, C. J. Yu, C. Bamdad, S. OꢁConnor, T. MacLean, E.
Lam, Y. Chong, G. T. Olsen, J. Luo, M. Gozin, J. F. Kayyem, J. Am.
Chem. Soc. 1999, 121, 1059; b) H. D. Sikes, J. F. Smalley, S. P.
Dudek, A. R. Cook, M. D. Newton, C. E. D. Chidsey, S. W. Feldberg,
Science 2001, 291, 1519.
Zn(ImPS2) monolayer formation: The ITO glass was modified with thiol-
N
20 h in a 0.1mm solution of H2(ImPS2) in methanol containing 5 vol% of
G
28% NH4OH.[51] Thioacetyl groups were cleaved by soaking in this alka-
line solution, allowing attachment to the titanium-modified ITO surfaces.
Strong absorbance at the Soret band was observed after rinsing with
methanol and CH2Cl2.
Absorption spectra: Absorption spectra were measured by using a Shi-
madzu UV-3100PC spectrometer. Each spectrum was corrected for back-
ground absorption of the ITO electrode by subtraction.
Electrochemical and photocurrent measurements: The electrochemical
investigation of the surface-confined porphyrin layers was carried out in
an aqueous system under continuous nitrogen flow. The modified ITO
working electrodes (WEs), mounted on 0.28 cm2 of a round cell window,
were exposed to aqueous buffered solution containing 1.0m NaClO4 as a
supporting electrolyte. The pH conditions were adjusted by means of
citric acid/Na2HPO4 buffer (pH 2.9–5.6) and phosphate buffer (Na2HPO4/
KH2PO4, pH 6.2–7.0). Platinum wire and Ag/AgCl(sat. KCl) were used
G
[7] A. A. Yasseri, D. Syomin, R. S. Loewe, J. S. Lindsey, F. Zaera, D. F.
Bocian, J. Am. Chem. Soc. 2004, 126, 15603.
as an auxiliary (CE) and a reference electrode (RE), respectively. The
conventional three-electrode configuration was connected to a potentio-
stat (BAS, CV-50W).
[8] SAMs on Au electrodes: a) K. Uosaki, T. Kondo, X.-Q. Zhang, M.
Yanagida, J. Am. Chem. Soc. 1997, 119, 8367; b) H. Imahori, H. Nor-
ieda, H. Yamada, Y. Nisimura, I. Yamazaki, Y. Sakata, S. Fukuzumi,
J. Am. Chem. Soc. 2001, 123, 100; c) K.-S. Kim, M.-S. Kang, H. Ma,
A. K.-Y. Jen, Chem. Mater. 2004, 16, 5058.
For anodic photocurrent generation, 50mm hydroquinone (H2Q), recrys-
tallized from acetone, was dissolved in the buffered electrolyte solution
as a sacrificial electron donor. The electrochemical cell was placed at a
distance of 31 cm from the lamp house in the dark. The current was
monitored under potentiostatic conditions during the on/off cycles of
[9] SAMs on ITO electrodes: a) H. Yamada, H. Imahori, Y. Nishimura,
I. Yamazaki, T. K. Ahn, S. K. Kim, S. Fukuzumi, J. Am. Chem. Soc.
8134
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8123 – 8135

Porphyrin Assemblies on ITO Anodes
FULL PAPER
2003, 125, 9129; b) H. Imahori, M. Kimura, K. Hosomizu, T. Sato,
T. K. Ahn, S. K. Kim, D. Kim, Y. Nishimura, I. Yamazaki, Y. Araki,
O. Ito, S. Fukuzumi, Chem. Eur. J. 2004, 10, 5111; c) Y.-J. Cho, T. K.
Ahn, H. Song, K. S. Kim, C. Y. Lee, W. S. Seo, K. Lee, S. K. Kim, D.
Kim, J. T. Park, J. Am. Chem. Soc. 2005, 127, 2380.
[31] a) A. Ghosh, Acc. Chem. Res. 1998, 31, 189; b) D. H. Karweuk, N.
Winograd, Inorg. Chem. 1976, 15, 2336.
[32] N. Nishimura, M. Ooi, K. Shimazu, H. Fujii, K. Uosaki, J. Electroa-
nal. Chem. 1999, 473, 75.
[33] F. Da Cruz, K. Driaf, C. Berthier, J.-M. Lameille, F. Armand, Thin
Solid Films 1999, 349, 155.
[34] D. Kalita, M. Morisue, Y. Kobuke, New J. Chem. 2006, 39, 77.
[35] a) N. Mataga, H. Chosrowjan, Y. Shibata, N. Yoshida, A. Osuka, T.
Kikuzawa, T. Okada, J. Am. Chem. Soc. 2001, 123, 12422; N.
Mataga, S. Taniguchi, H. Chosrowjan, A. Osuka, K. Kurotobi,
Chem. Phys. Lett. 2005, 403, 163; b) R. H. Hayes, C. J. Walsh, M. R.
Wasielewski, J. Phys. Chem. A 2004, 108, 2375.
[36] a) H. S. Cho, D. H. Jeong, M.-C. Yoon, Y. H. Kim, Y.-R. Kim, D.
Kim, S. C. Jeoung, S. K. Kim, N. Aratani, H. Shinmori, A. Osuka, J.
Phys. Chem. A 2001, 105, 4200; b) S. Faure, C. Stern, R. Guilard,
P. D. Harvey, J. Am. Chem. Soc. 2004, 126, 1253; c) T. Komatsu, M.
Moritake, E. Tsuchida, Chem. Eur. J. 2003, 9, 4626.
[10] Langmuir–Blodgett films: a) M. Fujihira, K. Nishiyama, H. Yamada,
Thin Solid Films 1985, 132, 77; b) M. Fujihira, M. Sakomura, T.
Kamei, Thin Solid Films 1989, 180, 43; c) A. Aoki, Y. Abe, T. Miya-
shita, Langmuir 1999, 15, 1463.
[11] Layer-by-layer membranes: a) C. Luo, D. M. Guldi, M. Maggini, E.
Menna, S. Mondini, N. A. Kotov, M. Prato, Angew. Chem. 2000, 112,
4052; Angew. Chem. Int. Ed. 2000, 39, 3905; b) A. Ikeda, T. Hatano,
S. Shinkai, T. Akiyama, S. Yamada, J. Am. Chem. Soc. 2001, 123,
4855; c) F. B. Abdelrazzaq, R. C. Kwong, M. E. Thompson, J. Am.
Chem. Soc. 2002, 124, 4796; d) D. M. Guldi, I. Zilbermann, G. An-
derson, A. Li, D. Balbinot, N. Jux, M. Hatzimarinaki, A. Hirsch, M.
Prato, Chem. Commun. 2004, 726; e) H. Furukawa, N. Inoue, T. Wa-
tanabe, K. Kuroda, Langmuir 2005, 21, 3992.
[12] S. M. Sze, Physics of Semiconductor Devices, 2nd ed., Wiley, New
York, 1981.
[13] P. Fromhelz, W. Arden, J. Am. Chem. Soc. 1980, 102, 6211.
[14] L. Sereno, J. J. Silber, L. Otero, M. del Valle Bohorquez, A. L.
Moore, T. A. Moore, D. Gust, J. Phys. Chem. 1996, 100, 814.
[15] R. H. Wilson, J. Appl. Phys. 1977, 48, 4292.
[37] a) K. S. Kim, I. Park, S. Lee, K. Cho, J. Y. Lee, J. Kim, J. D. Joanno-
poulos, Phys. Rev. Lett. 1996, 76, 956; b) L. Kevan, Acc. Chem. Res.
1981, 14, 138; c) J. R. R. Verlet, A. E. Bragg, A. Kammrath, O.
Cheshnovsky, D. M. Neumark, Science 2005, 307, 93.
[38] a) C. Zimmermann, F. Willig, S. Ramakrishna, B. Burfeindt, B. Pet-
tinger, R. Eichberger, W. Storck, J. Phys. Chem. B 2001, 105, 9245;
b) J. Schnadt, P. A. Brühwiler, L. Patthey, J. N. OꢁShea, S. Sçdergen,
M. Odelius, R. Ahuja, O. Karis, M. Bässler, P. Persson, H. Siegbahn,
S. Lunell, N. Mårtensson, Nature 2002, 418, 620; c) K. Onda, B. Li, J.
Zhao, K. D. Jordan, J. Yang, H. Petek, Science 2005, 308, 1154.
[39] J. M. Bolts, M. S. Wrighton, J. Phys. Chem. 1976, 80, 2641.
[40] a) Y. Sato, M. Fujita, F. Mizutani, K. Uosaki, J. Electroanal. Chem.
1996, 409, 145; b) M. Y. Okamura, M. L. Paddock, M. S. Graige, G.
Feher, Biochim. Biophys. Acta 2000, 1458, 148; c) H.-G. Hong, W.
Park, Langmuir 2001, 17, 2485.
[41] F. Nüesch, L. J. Rothberg, F. W. Forsythe, Q. T. Le, Y. Gao, Appl.
Phys. Lett. 1999, 74, 880.
[42] D. F. Watson, A. Marton, A. M. Stux, G. J. Meyer, J. Phys. Chem. B
2004, 108, 11680.
[43] a) F. Odobel, E. Blart, M. Lagree, M. Villieras, H. Boujtita, N.
El Murr, S. Caramori, C. A. Bignozzi, J. Mater. Chem. 2003, 13, 502;
b) W. M. Campbell, A. K. Burrell, D. L. Officer, K. W. Jolley, Coord.
Chem. Rev. 2004, 248, 1363.
[44] a) B. Choudhury, A. C. Weedon, J. R. Bolton, Langmuir 1998, 14,
6199; b) H. Imahori, T. Hasobe, H. Yamada, Y. Nishimura, I. Yama-
zaki, S. Fukuzumi, Langmuir 2001, 17, 4925.
[45] A. Ohashi, A. Satake, Y. Kobuke, Bull. Chem. Soc. Jpn. 2004, 77,
365.
[46] T. Hirao, T. Masunaga, Y. Ohshiro, T. Agawa, Synthesis 1981, 56.
[47] a) R. Rabinowitz, J. Org. Chem. 1963, 28, 2975; b) D. Deniaud, B.
Schçllorn, D. Mansuy, J. Rouxel, P. Battioni, B. Bujoli, Chem. Mater.
1995, 7, 995.
[48] W. Fudickar, J. Zimmermann, L. Ruhlmann, J. Schneider, B. Rçder,
U. Siggel, J.-H. Furhop, J. Am. Chem. Soc. 1999, 121, 9539.
[49] E. Díez-Barra, J. C. García-Martínez, S. Merino, R. del Rey, J. Ro-
dríguez-López, P. Sµnchez-Verdffl, J. Tejeda, J. Org. Chem. 2001, 66,
5664.
[50] T. J. Meyer, G. J. Meyer, B. W. Pfennig, J. R. Schoonover, C. J. Timp-
son, J. F. Wall, C. Kobusch, X. Chen, B. M. Peek, C. G. Wall, W. Ou,
B. W. Erickson, C. A. Bignozzi, Inorg. Chem. 1994, 33, 3952.
[51] J. M. Tour, L. Jones, II, D. L. Pearson, J. J. S. Lamba, T. P. Burgin,
G. M. Whitesides, D. L. Allara, A. N. Parikh, S. V. Atre, J. Am.
Chem. Soc. 1995, 117, 9529.
[16] T. Miyasaka, T. Watanabe, A. Fujishima, K. Honda, J. Am. Chem.
Soc. 1978, 100, 6657.
[17] A. Ulman, Chem. Rev. 1996, 96, 1533.
[18] a) T. J. Gardner, C. D. Frisbie, M. S. Wrighton, J. Am. Chem. Soc.
1995, 117, 6927; b) C. Yan, M. Zharnikov, A. Gçlzhäuse, M. Grunze,
Langmuir 2000, 16, 6208; c) B. Vercelli, G. Zotti, G. Schiavon, S.
Zecchin, Langmuir 2003, 19, 9351.
[19] a) J. Deisenhofer, O. Epp, K. Miki, R. Huber, H. Michel, J. Mol.
Biol. 1984, 180, 385; b) J. Deisenhofer, O. Epp, K. Miki, R. Huber,
H. Michel, Nature 1985, 318, 618; c) J. Deisenhofer, H. Michel, Sci-
ence 1989, 245, 1463; d) P. Allen, G. Feher, T. O. Yeates, D. C. Rees,
J. Deisenhofer, H. Michel, R. Huber, Proc. Natl. Acad. Sci. USA
1986, 83, 8589.
[20] a) Y. Kobuke, H. Miyaji, J. Am. Chem. Soc. 1994, 116, 4111; b) Y.
Kobuke, K. Ogawa, Bull. Chem. Soc. Jpn. 2003, 76, 689.
[21] H. Ozeki, A. Nomoto, K. Ogawa, Y. Kobuke, M. Murakami, K.
Hosoda, M. Ohtani, S. Nakashima, H. Miyasaka, T. Okada, Chem.
Eur. J. 2004, 10, 6393.
[22] M. Morisue, S. Yamatsu, N. Haruta, Y. Kobuke, Chem. Eur. J. 2005,
11, 5563.
[23] K. L. Purvis, G. Lu, J. Schwartz, S. L. Bernasek, J. Am. Chem. Soc.
2000, 122, 1808.
[24] a) I. Ichinose, H. Senzu, T. Kunitake, Chem. Mater. 1997, 9, 1296;
b) I. Ichinose, T. Kawakami, T. Kunitake, Adv. Mater. 1998, 10, 535.
[25] a) J. B. Miller, J. Schwartz, S. L. Bernasek, Langmuir 1994, 10, 2629;
b) S. K. Van der Kam, G. Lu, S. L. Bernasek, J. Schwartz, J. Am.
Chem. Soc. 1997, 119, 11639; c) G. Lu, J. Schwartz, S. L. Bernasek,
Langmuir 1998, 14, 1532.
[26] a) W. Gçpel, J. A. Anderson, D. Frankel, M. Jaehnig, K. Phillips,
J. A. Schäfer, G. Pocker, Surface Sci. 1984, 139, 333; b) S. Cherian,
C. C. Wamser, J. Phys. Chem. B 2000, 104, 3624.
[27] a) T. Ishida, H. Kobayashi, Y. Nakano, J. Appl. Phys. 1993, 73, 4344;
b) C. Donley, D. Dunphy, D. Paine, C. Carter, K. Nebesny, P. Lee,
D. Alloway, N. R. Armstrong, Langmuir 2002, 18, 450.
[28] a) F. Cao, G. Oskam, P. C. Searson, J. M. Stipkala, T. A. Heimer, F.
Farzad, G. J. Meyer, J. Phys. Chem. 1995, 99, 11974; b) R. L. Willis,
C. Olson, B. OꢁRegan, T. Lutz, J. Nelson, J. R. Durrant, J. Phys.
Chem. B 2002, 106, 7605.
Received: March 4, 2006
Revised: June 9, 2006
Published online: September 15, 2006
[29] M. D. Porter, T. B. Bright, D. L. Allara, C. E. D. Chidsey, J. Am.
Chem. Soc. 1987, 109, 3559.
[30] B. Vercelli, G. Zotti, G. Schiavon, S. Zecchin, Langmuir 2003, 19,
9351.
Please note: Minor changes have been made to this manuscript since its
publication in Chemistry—A European Journal Early View. The Editor.
Chem. Eur. J. 2006, 12, 8123 – 8135
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8135