Design and Characterization of Novel Porphyrins with Oligo(phenylethylnyl) Links of
Varied Length for Dye-Sensitized Solar Cells: Synthesis and Optical, Electrochemical, and
Photovoltaic Investigation
Ching-Yao Lin,*,†Chen-Fu Lo,† Liyang Luo,‡,§Hsueh-Pei Lu,‡Chen-Shiung Hung,§ and Eric Wei-Guang Diau*,‡
Department of Applied Chemistry, National Chi Nan UniVersity, Puli, Nantou 545, Taiwan, Department of Applied Chemistry and Institute of Molecular Science, National Chiao Tung UniVersity, Hsinchu 300, Taiwan, and Institute of Chemistry, Academia Sinica, Taipei 115, Taiwan
ReceiVed: July 30, 2008; ReVised Manuscript ReceiVed: September 29, 2008
Novel zinc porphyrins with 1-4 π-conjugated phenylethylnyl (PE) units (labeled PE1-PE4) as a link of controlled length were synthesized for fundamental tests and applications as a dye-sensitized solar cell (DSSC). The UV-visible spectra of the solution samples show clear absorption patterns of the PE groups in a region 300-400 nm, consistent with results calculated with density-functional theory. Cyclic voltammograms of PE1-PE4 in tetrahydrofuran show similar electrochemical potentials for each compound. Femtosecond fluorescence up-conversion of solution samples and of porphyrin-sensitized TiO2films was measured with
excitation at 420 or 430 nm and emission at 460, 470, 620, and 680 nm. When these porphyrins were fabricated into DSSC devices, the efficiency of power conversion of these devices decreased systematically with increasing length of the link: 2.5 ( 0.2% (PE1), 2.0 ( 0.1% (PE2), 0.78 ( 0.09% (PE3), 0.25 ( 0.02% (PE4). This great photovoltaic degradation from PE1 to PE4 is not interpretable according to the rate of electron injection independent of length; other factors, including electron transfer from the semiconductor back to the porphyrin cation or the electrolyte, must be considered to account for the observed dependence of photovoltaic performance on length.
1. Introduction
The adsorption of selected photosensitizers on nanocrystalline films, having a large surface area, of semiconductors with a wide band gap, such as titanium dioxide (TiO2), has been an
active area of research on efficient dye-sensitized solar cells (DSSC).1 The covalent attachment of electrochemically and
photochemically active chromophores to semiconductor surfaces is an important step toward the development of such devices. Because of their efficacy in this respect, RuII polypyridyl
complexes adsorbed on TiO2nanocrystalline films have proved
promising systems. Among RuIIpolypyridyl complexes, the
so-called N3 dye, RuII(dcbpyH
2)2(NCS)2, is the most efficient
photosensitizer known.2This N3 dye exhibits an efficiency of
incident conversion of photons to electrons (IPCE) as great as 85% in the region 400-800 nm and an overall efficiency (η) 11% of power conversion.3Porphyrins have become prospective
photosensitizers4because of the vital roles of porphyrin
deriva-tives in photosynthesis, their strong absorption in the visible region, and the ease of adjusting their chemical structures (hence their electrochemical and photochemical properties) for light harvesting. Adsorbing porphyrins with modified structures onto TiO2 nanocrystalline films thus provides an opportunity to
improve DSSC applications. The best performance of a por-phyrin-sensitized SC has attainedη ) 7.1%.4d
Several factors are considered important when seeking or designing an efficient photosensitizer. The appropriate sen-sitizers should have these properties: the ability to adsorb
strongly on the surface of a semiconductor, absorption of light in the visible and NIR region for efficient light harvesting, an excited-state reduction potential negative enough for efficient electron injection, a ground-state reduc-tion potential positive enough for efficient regenerareduc-tion of the dye, and a small reorganizational energy for efficient excited- and ground-state electrochemical processes.5 The
relation between the orientation of attached porphyrins with respect to TiO2nanocrystalline surfaces and the cell
perfor-mance has been investigated.4b,6For porphyrins,4,6an efficient
injection of electrons from the photosensitizer into a semi-conductor is considered a crucial factor in developing an efficient DSSC. Apart from the effect of cosensitization,7the
performance of a cell can be improved through appropriate modification of a dye structure with a selected link.
We reported8the fundamental properties of a novel porphyrin
in solution and adsorbed on TiO2nanocrystalline films. In our
design, 10,20-biphenylporphinato zinc(II) (ZnBPP) served as a light-harvesting center and a carboxylic acid as an anchoring group. To control the distance between the porphyrin and the anchoring group, we utilized a phenylethylnyl (PE) unit as a bridging moiety to connect a meso position of ZnBPP and the carboxylic acid end. We chose ZnBPP because of its ease of synthesis, the stability of zinc porphyrins against irradiation, and the efficiency of zinc porphyrins that is superior to that of copper and free-base porphyrins.4b 4-Carboxylic acid groups
were used to ensure efficient adsorption on TiO2surfaces and
to promote electronic coupling between the donor levels of the excited porphyrins and the acceptor levels of TiO2.6PE units
were employed to control the distance because of their rigidity, their linear structure, and the possibleπ interactions between the porphyrin core and the anchoring end.9Porphyrins with an
* To whom correspondence should be addressed. E-mail: cyl@ ncnu.edu.tw (C.-Y.L.); [email protected] (E.W.-G.D.).
†National Chi Nan University. ‡National Chiao Tung University. §Academia Sinica.
10.1021/jp806777r CCC: $40.75 2009 American Chemical Society Published on Web 12/22/2008
ethynyl motif were employed for photonic and electrochemical tests on multichromophore arrays.10
PE units as spacers have been reported for other systems.10b,11
For example, seeking to use porphyrins as molecular photonic materials, Lindsey and co-workers investigated several multi-porphyrin arrays to understand their energy-transfer phenom-ena;12 in their design, monomeric chromophores were linked
with phenyl-ethynyl-phenyl bridging groups to create weak interactions between the chromophores and to preserve the characteristics of each chromophore.10b,c Piotrowiak et al.
reported “tripod” RuIIcomplexes with PE groups as rigid links
to connect the sensitizers and TiO2;11these sensitizers consisted
of a tripod-shaped base with three COOR groups to attach to TiO2nanocrystalline surfaces, a rigid (4-ethynyl)phenylethyne
spacer, and a RuIIcomplex as a chromophoric center. The base
design provides a stable and well-defined orientation of the sensitizers with respect to the nanoparticulate surfaces, whereas the rigidity of the (4-ethynyl)phenylethyne spacer ensures control of the distance between the light-harvesting centers and the TiO2surfaces.
To investigate the interfacial electron transfer and for DSSC applications with a similar design, we have prepared novel porphyrins: 5-(4-carboxy-phenylethynyl)-10,20-biphenylporphinato zinc(II), 5-[4-(4-carboxy-phenylethynyl)-phenylethynyl]-10,20-bi-phenylporphinato zinc(II), 5-{4-[4-(4-carboxy-phenylethynyl)-phe-nylethynyl]-phenylethynyl}-10,20-biphenylporphinato zinc(II), and 5-(4-{4-[4-(4-carboxy-phenylethynyl)-phenylethynyl]-phenylethy-nyl}-phenylethynyl)-10,20-biphenylporphinato zinc(II), abbreviated PE1, PE2, PE3, and PE4, respectively. As shown in Chart 1, these porphyrins share the same light-harvesting unit (ZnBPP) with rigid, linear, edgewise, and fully conjugated substituents of varied length. When we fabricated DSSC devices using PE1-PE4 as photosensitizers, the cell performance of the devices exhibited a systematic dependence on length according to which the DSSC with a long link (PE4) performed much less effectively than the DSSC with a short link (PE1). To understand the interfacial electron-transfer dynamics that govern the photovoltaic perfor-mance of the system dependent on length, we conducted femtosecond measurements, and we report here the spectral and electrochemical properties and the fluorescence decay of PE1-PE4 in solution and on TiO2nanocrystalline films.
2. Results and Discussion
2.1. Synthesis. To prepare PE1-PE4, we reacted porphyrin precursors with carboxyphenylethynyl reagents according to the Sonogashira cross-coupling method.13This synthetic strategy,
which is well developed for preparing porphyrins with varied ethyne-linked functional groups,14,15was readily adaptable to
the synthesis of PE1-PE4. Among a few points worth noting, cross-coupling ZnBPPBr with a suitable substituent precursor effectively yielded PE1 and PE2, but reacting 5-(4-ethynyl-phenylethynyl)-ZnBPP with 4-(4-iodo-phenylethynyl)-benzoic acid seemed to be the most efficient approach to prepare PE3. Our attempts to synthesize PE3 by other routes were unsuc-cessful because of small yields and the difficulty of recognizing and separating the reaction products. For PE4, the strategy was simply to add a further phenyl group to the porphyrin precursor from the PE3 synthesis. Because of the carboxylic group, these porphyrins all tend to adsorb slightly on the silica gel used in chromatographic purification; the number of chromatographic separations should thus be kept to a minimum to ensure acceptable yields of the porphyrins.
2.2. Absorption and Emission Spectra of Porphyrins in Solution. Parts a and b of Figure 1 show absorption and emission spectra of ZnBPP and PE1-PE4 in tetrahydrofuran (THF), respectively. The absorption maxima and absorption coefficients of these porphyrins are listed in Table 1; the emission maxima and lifetimes of excited states are listed in Table 2. As shown in the figures, these molecules exhibit characteristic porphyrin spectra:16strong Soret (B) bands near
440 nm and weaker Q bands in the region of 550-650 nm. The fluorescence spectra show mirror images of the Q bands. Absorption bands of PE1-PE4 are red-shifted from those of CHART 1: Molecular Structures of ZnBPP and
PE1-PE4 with the Distance between the Porphyrin Core and the Anchoring Group Calculated with Density Functional Theory (DFT)
Figure 1. (a) UV-visible absorption and (b) fluorescence spectra of
ZnBPP (gray), PE1 (black), PE2 (green), PE3 (red), and PE4 (blue) in THF withλex) 430 nm.
ZnBPP; this shift is due to the extended π-conjugation, consistent with reports of other ethyne-linked porphyrins.15Most
importantly, the features in the absorption and emission spectra of PE1-PE4 have similar shapes and intensities, indicating that the energy gaps between the ground and the excited states of those porphyrins are similar despite the varied lengths of the PE links.
Further weak absorptions appear in the near-UV region for PE1-PE4; their wavelengths increased systematically as the conjugation length of the PE bridge increased from PE1 to PE4. From PE2 to PE4 the intensities of these near UV absorptions increased, whereas the intensities of the Soret bands decreased, indicating the existence of strong coupling between the PE bridge and the porphyrin core, so that the intensity of the former becomes “borrowed” from the intensity of the latter. As there is no such absorption in ZnBPP, we infer that these near UV absorptions arise from the PE groups of the PE1-PE4 porphy-rins. Nielsen et al. reported a similar phenomenon for a zinc porphyrin linked with poly(phenylenethynylene).17In the case
of PE1-PE4, the coupling and interaction between the bridging moiety and the porphyrin core are effective because similar emission spectra, as shown in Figure 1b, were obtained when those porphyrins were excited in the near -UV region at 350 nm.
2.3. Relaxation Dynamics of Porphyrins in Solutions. Parts a-d of Figure 2 show fluorescence transients of PE1-PE4 in THF, respectively. Those transients were recorded with excita-tion in the Soret band atλex ) 420 or 430 nm and emission
observed at eitherλem) 460 or 620 nm. The transients observed
atλem) 460 nm (circles) are satisfactorily fitted with a single
exponential decay convoluted with the instrument-response function (fwhm ) 220 fs), which yields decay coefficients 910, 680, 610, and 600 fs for PE1-PE4, respectively. The transients observed at λem ) 620 nm (squares) differ remarkably from
the transients observed atλem) 460 nm (circles): the former
show a rising feature with a rise coefficient comparable to the decay coefficient of the latter at short times, but the transient signals of the former increase gradually to an asymptotic level at longer times in the same range. According to femtosecond investigations reported elsewhere,8athe transients observed at
λem) 460 and 620 nm reflect the dynamical behavior occurring
in electronic states S2 and S1 of PE1, respectively. We thus
conclude that internal conversion (IC) S2fS1of PE1-PE4 in
solution occurs within the observed interval τ ) 0.6-0.9 ps and that vibrationally hot S1species are produced on that time
scale.τ decreased significantly from PE1, τ ) 910 fs, to PE2, τ ) 680 fs, but only slightly further to ∼600 fs for PE3 and PE4. This dynamical feature is consistent with the trend of the spectral shift of the absorption and emission spectra shown in Figure 1b in which the spectra are almost identical for PE2-PE4, but the spectrum of PE1 is somewhat blue-shifted with respect to the others. The systematic decrease of S2
relaxation periods with increasing length of the PE bridge of PE1-PE4 implies that the electronic coupling of the extended π-conjugation significantly affects the IC rate for S2fS1, i.e., a longer bridge produces stronger coupling between the por-phyrin core and the PE link that leads to a greater rate of internal conversion.
Because of the rising character of the transients shown in Figure 2, the transients observed atλem) 620 nm are described
satisfactorily according to a consecutive kinetic model, A f B fC, with the signals convoluted with the laser pulse. For PE1, the first component (B) rises withτ1) 910 fs and decays with
τ2 ) 8.5 ps, whereas the second component (C) appears
following the decay of the first component but persists on a scale withτ3≈ 2 ns (Table 2); for PE2-PE4, τ1) 680, 610,
and 600 fs andτ2) 8.4, 8.2, and 8.4 ps, respectively. As the
fluorescence transients observed at λem ) 620 nm reflect the
relaxation dynamics of PE1-PE4 in the S1state, the observed
two consecutive components (B and C) might be assigned to hot and cold S1species, respectively; component A represents
the S2species not observed atλem) 620 nm. Accordingly, the
hot S1species were produced from the S2state in 910-600 fs.
The excess vibrational energies in the hot S1species become
released to the solvent molecules via vibrational relaxation, and this process yields the cold S1 species produced in∼8.5 ps. TABLE 1: Absorption Maxima and Absorption
Coefficientsaof Zinc Porphyrins in THF
dye near-UV bands λS2(log )/nm λS1(log )/nm
ZnBPP not observed 412 (5.71) 543 (4.27), 579 (3.34) PE1 not observed 439 (5.64) 567 (4.21), 616 (4.41) PE2 304, 353 443 (5.67) 569 (4.27), 618 (4.60) PE3 322, 359 443 (5.58) 569 (4.21), 619 (4.56) PE4 338, 363 444 (5.56) 569 (4.17), 619 (4.53)
aThe absorption coefficient () has a unit of cm-1M-1.
TABLE 2: Fluorescence Maxima and Lifetimesaof the
Excited State of Zinc Porphyrins in THF
dye emission /nm τS2/ps τS1/ns ZnBPP 636, 699 PE1 621, 674 0.91 1.90 PE2 624, 678 0.68 2.10 PE3 624, 677 0.61 2.10 PE4 624, 677 0.60 2.10 a
S2 and S1 lifetimes were monitored at 460 and 620 nm,
respectively.
Figure 2. Femtosecond fluorescence transients of (a) PE1, (b) PE2,
(c) PE3, and (d) PE4 in THF solution (c ) 1× 10-5M). Open symbols denote raw data: circles, points obtained atλex) 420 nm and λem)
460 nm; squares, points obtained atλex) 430 nm and λem) 620 nm.
The solid curves represent theoretical fits with convolution of the laser pulse.
The cold S1species endure because S1fT1intersystem crossing
of PE1-PE4 in solution occurs in∼2 ns, as confirmed by our picosecond measurements. Our direct observation on the PE1-PE4 system is consistent with femtosecond results for the ZnTPP system, for which the lifetime of the S2state of ZnTPP
was reported in a range 1.45-2.35 ps, and vibrational cooling from the nonrelaxed vibronic state of the hot S1species occurs
on a time scale∼10 ps.18-20
2.4. Electrochemistry, DFT Calculations, and Energy Levels. Figure 3 displays cyclic voltammograms (CV) for PE1-PE4 reductions in THF, and Table 3 presents their electrochemical potentials. The first reduction of ZnBPP gave a quasireversible couple at -1.40 V vs SCE, consistent with the formation of the tetraphenylporphinato zinc(II) anion radical at -1.31 V.21For PE1-PE4, reduction of the first porphyrin
ring occurs at about -1.20 V.22 The positive shift of the
reduction potentials is attributed to the extendedπ-conjugation. In contrast, the first oxidations of PE1-PE4 are all irreversible reactions (not shown), likely due to THF being capable of attacking the oxidized products. Nevertheless, the oxidation potentials of all these complexes are similar: Epa) +1.07 V vs
SCE, Table 3.
To assist our qualitative understanding of the reduction behaviors of these complexes, we performed calculations with DFT on ZnBPP and PE1-PE4 at the B3LYP/LanL2DZ level. Figure 4 depicts their MO patterns from HOMO-1 to LUMO+2. These MO patterns of PE1-PE4 are consistent with those predicted by Gouterman’s four-orbital model,16i.e., HOMO-1
and HOMO resemble those of the a1u and a2u orbitals,
respectively, whereas LUMO and LUMO+1 are similar to those of the eg orbitals. The HOMO and LUMO of PE1-PE4 are,
however, partially delocalized from the porphyrin core to the
substituents; this minor deviation from Gouterman’s four-orbital model is rationalized through the decreased symmetry of the complexes and the extendedπ-conjugation of the substituents. LUMO+2 was found to be concentrated at theπ-conjugated substituents, consistent with the result of Wang et al. based on a time-dependent DFT approach for a series of β-substituted zinc porphyrins.4c
Figure 5 compares the energy levels of PE1-PE4 and TiO2.
This illustration was prepared on the basis of the electrochemical results and the absorption spectra according to literature methods.23,24Figure 5 shows that the conducting band (CB) of
TiO2 is located between the LUMO and the HOMO of
PE1-PE4, and the MO energy levels between PE1-PE4 are similar, consistent with the DFT results. These results indicate that electron injection from LUMO of the porphyrin to the CB of TiO2is feasible and that the rate of electron injection for
such a process might be similar for PE1-PE4/TiO2 films
because of their similar energy levels.
2.5. Absorption Spectra of Porphyrins on TiO2
Nanoc-rystalline Films. Figure 6 displays UV-visible absorption spectra of PE1-PE4/TiO2 nanocrystalline films in air (black
curves) and PE1-PE4 in THF (gray curves). Comparison of the film spectra with the solution spectra reveals that the Q bands of the PE1-PE4/TiO2 films are only slightly shifted and
broadened from those of the solution spectra, whereas the B bands became blue-shifted with a red shoulder. These shoulders of the Soret bands in the film spectra represent a small proportion of monomeric porphyrins adsorbed on TiO2films;
their wavelengths are the same as those of the B bands in solution.25In spectra of PE1-PE4 films the blue-shifted B bands
occur at 413, 415, 420, and 426 nm, respectively. These bands are not caused by effects of reflection26because the quality of
TiO2films (e.g., thickness) seems to have only a slight effect
on the intensities. We suggested previously an attribution of these blue-shifted B bands to the H-aggregation of the porphyrin assembly on the surfaces of TiO2nanocrystalline films.8bThe
stacked, face-to-face porphyrinπ-aggregation (H-type) is well documented to produce a blue shift of the absorption, whereas the side-by-side porphyrinπ-aggregation (J-type) results in a red shift.27 This phenomenon has been observed for not only
porphyrins28but also other planar dyes.29Measurements8bwith
atomic-force microscopy show little variation of morphology between bare TiO2 nanocrystalline film surfaces and those
modified with porphyrin, indicating that the film surfaces are covered with a porphyrin monolayer (thickness only 1-3 nm), supporting the possibility of porphyrins being H-aggregated on the surfaces.
2.6. Relaxation Dynamics of Porphyrins on TiO2
Nanoc-rystalline Films. To investigate the dependence of the relaxation dynamics of porphyrin/TiO2films on the length of the PE link,
we sensitized the four porphyrins on identical TiO2
nanocrys-talline films. Figure 7 shows a scanning electron microscopy image of a typical TiO2film (thickness∼5 µm) with
nanopar-ticle size ∼20 nm. Figure 8 shows the absorption spectra of the four porphyrin-sensitized films with the absorbance of the Q(0,0) band adjusted to be equal for PE1-PE4. Parts a-d of Figure 9 show the fluorescence transients of PE1-PE4 on TiO2
nanoscrystalline films, respectively, obtained with excitation at λex) 420 nm and emission observed at λem) 470 nm. The
transients observed atλem) 470 nm (circles) are satisfactorily
fitted with a single exponential decay convoluted with the instrument response function (fwhm ) 220 fs), which yields decay coefficients smaller than 100 fs for all porphyrins. According to our previous work,8athe S
2dynamics of the PE1/ Figure 3. Cyclic voltammograms of ZnBPP and PE1-PE4 reductions
in THF.
TABLE 3: First Electrochemical Potentials/V of Zinc Porphyrins in THFa
dye E1/2(ox 1) E1/2(red 2)
ZnBPP 0.99 (170 mV) -1.40 (160 mV) PE1 ∼1.07 (Epa) -1.23 (130 mV) PE2 ∼1.07 (Epa) -1.18 (125 mV) PE3 ∼1.07 (Epa) -1.19 (130 mV) PE4 ∼1.07 (Epa) -1.20 (130 mV) a
Experimental conditions: zinc porphyrin (0.5 mM) in freshly distilled and degassed THF/TBAP, scan rate ) 100 mV/s, Pt working and counter electrodes, SCE reference electrode. For ferrocene/forrocenium (Fc/Fc+) in the same conditions, E1/2 )
+0.49 V vs SCE. Peak-to-peak separations (mV) are shown in parentheses for quasireversible reactions. For irreversible reactions, only the anodic peak potentials, denoted Epa, are shown.
TiO2films reflect not only the aggregate-induced energy transfer
among pophyrins but also an interfacial electron transfer from porphyrin to TiO2. Because the fluorescence transients were
observed also at the S1state (see below), we conclude that the
S2 f S1 IC is ultrarapid, <100 fs. In the present work we
extended the investigation of the S2dynamics to porphyrins with
an extended PE bridge. The length of the PE group did not affect the observed S2dynamics, which are due mainly to the
ultrarapid aggregate-induced IC relaxation for all our porphyrin/ TiO2films. The hot S1species were thus produced in <100 fs,
and the relaxation dynamics occurring at the porphyrin/TiO2
interface were probed as described in the following.
Figures 10 and 11 show the S1 relaxation dynamics of
porphyrin/TiO2 films probed at λem ) 620 and 680 nm,
respectively, with PE1-PE4 for parts a-d in each figure (λex
) 430 nm). All fluorescence transients were fitted with two decay components and a tiny offset. Atλem) 620 nm, the
rapid-decay coefficient τ1 was observed to be 180 fs for PE1,
remaining almost constant from PE2 (τ1) 200 fs), PE3 (τ1)
190 fs), and PE4 (τ1) 170 fs); the slow-decay coefficient τ2
increased slightly from 1.2 ps for PE1 to 1.6 ps for PE4, but
the variation was within experimental error. Atλem) 680 nm,
τ1was in a range 390-540 fs for PE1-PE4 andτ2in a range
2.7-3.7 ps for PE1-PE4. The relaxation dynamics observed at 620 nm are much more rapid than those observed at 680 nm because the detection window at the former wavelength probed more “hot” species than that at the latter wavelength.30
Two contributions are responsible for the observed S1
dynamics: interfacial electron transfer from dye to TiO2(electron
injection) and aggregate-induced energy transfer among dye molecules.8aBecause the TiO
2films were prepared identically
and the absorbances of the dye were similar for PE1-PE4 (Figure 8), we expect that the amounts of dye loading on TiO2
films were also similar for PE1-PE4. By assumption that the extents of aggregation inside the nanoporous environment of the TiO2films were similar for PE1-PE4, our observations thus
indicate that electron injection occurring in the S1state might
be independent of the length of the link between porphyrin and
Figure 4. Molecular orbitals of PE1-PE4 with geometries of each molecule optimized at the B3LYP/LanL2DZ level of theory.
Figure 5. Energy-level diagram of PE1-PE4 showing the HOMO
and the LUMO of each porphyrin and the valence band (VB) and conducting band (CB) of TiO2.
Figure 6. Comparison of absorption spectra of PE1-PE4 on TiO2
nanocrystalline films (black) and in THF (gray) for (a) PE1, (b) PE2, (c) PE3, and (d) PE4.
TiO2. DFT calculations (Figure 4) indicate, however, that the
electron densities of the HOMO and LUMO are localized in the porphyrin ring and extend only to the first unit of the PE link. The detection window of the time-resolved emission occurring from the S1state was therefore limited mainly to the
porphyrin core. When the electron transfer occurred from the porphyrin moiety to the anchoring group through the PE link, the emission decay might reflect only the initial propagation of electrons in the excited state and thus lack sensitivity for the detection of electron transfer toward longer links.
From the measurement of transient absorption, Galoppini and co-workers11efound that the biexponential decays of a Ru-bpy/
TiO2 system occur on a scale of 10-100 ps, and the time
coefficients decreased with increasing length of the rigid π-conjugated PE links. The discrepancy between our results (porphyrin/PE/TiO2) and their results (Ru-bpy/PE/TiO2) might
be due to the detection window, which is much broader using a transient absorption method than using an emission decay method. As a result, the length-dependent trend of electron transfer through PE links of varied length was observed from the former method reported by Galoppini.11e Alternatively, if
the electron transfer though the PE link is not a kinetic bottleneck, the discrepancy might reflect the varied number of anchoring groups in the dyes and the structure of the links on the TiO2surface, which might affect also the electron injection
kinetics. For perylene/TiO2films Gundlach et al.31 found that
the rate of electron injection of a rigid-rod system is much greater than that of a tripod system, providing evidence for the variation of electron injection kinetics with structure of the anchoring groups. The dependence of the performance of the porphyrin-sensitized solar cells on the length of the link is discussed in the following section.
2.7. Photovoltaic Performance of Porphyrin-Sensitized Solar Cells. By use of PE1-PE4 as sensitizers, we fabricated DSSC devices according to a standard procedure described in the experimental section. To demonstrate the reproducibility of the data, three independent measurements were performed for each dye using TiO2films fabricated with an identical procedure.
The measured photovoltaic performances of the DSSC devices with PE1-PE4 are shown in parts a-d of Figure 12; the corresponding parameters are summarized in Table 4. As a reference, the average value ofη of the Zn-3-sensitized DSSC was 4.0 ( 0.2%, which is smaller than the literature value (η ) 5.2%)4c because of thinner TiO
2films without addition of
the scattering layer in our case. The efficiencies of power conversion of the devices decreased greatly, from 2.5 ( 0.2 (PE1), 2.0 ( 0.1 (PE2), 0.78 ( 0.09 (PE3), to 0.25 ( 0.02% (PE4). This systematic variation of the cell performance reflects the photocurrent density, which decreased from 6.6 ( 0.6 (PE1), 5.5 ( 0.1 (PE2), 2.0 ( 0.3 (PE3), to 0.70 ( 0.03 mA cm-2 (PE4) as the length of the PE link increased from d /nm ) 1.2 (PE1), 1.9 (PE2), 2.6 (PE3), to 3.3 (PE4). Our results unam-biguously demonstrate that the cell performance depends directly
Figure 7. SEM image of TiO2nanoparticles spin-coated on a glass
plate showing the average particle size∼20 nm. The films (10 layers with thickness∼5 µm) were prepared using the same procedure for sensitization of the dye molecules (PE1-PE4) on the films with equal quality from film to film for fs fluorescence measurements.
Figure 8. Visible absorption spectra of TiO2 nanocrystalline films
sensitized with PE1-PE4 as indicated.
Figure 9. Femtosecond fluorescence transients of (a) PE1, (b) PE2,
(c) PE3, and (d) PE4 on TiO2nanocrystalline films. Open circles denote
raw data obtained atλex) 420 nm and λem) 470 nm. The solid curves
on the length of a rigidπ-conjugated link: the longer the link, the poorer the cell performance becomes. The observed length-dependent photovoltaic performance of the porphyrin-based DSSC is similar to the trend reported for a rigid planar meta-substituted porphyrin/TiO2system,11ffor which a short link with
enhanced photovoltaic efficiency was attributed to increased efficiency of electron injection from porphyrin to TiO2.
We measured the conversion efficiency of incident photons to current (IPCE) for the devices with PE1 and PE4 in the wavelength range 400-800 nm; the results appear in Figure 13. The photocurrent action spectra of the PE1 and PE4 devices mimic their corresponding absorption spectra, but the IPCE values of PE1 are ten times those of PE4, consistent with the difference in JSC and η for these two devices. The IPCE is
expressible in terms of the light-harvesting efficiency (LHE), the quantum yield of electron injection (φinj), and the efficiency
of charge collection at the counter electrode (ηc) according to2
IPCE(λ) ) LHE(λ)φinjηc (1)
As the absorbances of all sensitized films are similar, the values of LHE of all porphyrin-sensitized films are similar. If the rates of electron injection are similar for PE1-PE4, the φinj
values for all porphyrin-sensitized films are likewise similar. The factor affecting the IPCE values and other photovoltaic parameters (JSCandη) between PE1 and PE4 might be due to
ηc, which is dependent upon both the rate of charge
recombina-tion from the reduced TiO2 nanoparticles to the oxidized
porphyrin and the rate of dye regeneration by the oxidation of electrolyte.11fFor the dependence on distance of the open-circuit
voltage in a tripodal sensitizer system, Clark et al. found that VOCwas much greater for a device with a longer PE link.32In
general, VOCis larger if the dye regeneration (dye + + e
-f
dye) due to oxidation (3I-fI3-+ 2e-) occurs over a greater distance from the TiO2 surface. The charge recombination
between TiO2(e
-) and I3
-would hence be much slower over a greater distance so as to increase the efficiency of dye regenera-tion for the sensitizer with a longer link. We found, however, that VOC in porphyrin-sensitized SC was even smaller in the
porphyrin with the longest link (PE4), in contrast with the results reported for the Ru-bpy/TiO2system.32The smaller JSCobserved
for the device with a longer PE link might reflect much more
Figure 10. Femtosecond fluorescence transients of (a) PE1, (b) PE2,
(c) PE3, and (d) PE4 on TiO2nanocrystalline films. Open circles denote
raw data obtained atλex) 430 nm and λem) 620 nm. The solid and
dashed curves represent theoretical fits with convolution of the laser pulse.
Figure 11. Femtosecond fluorescence transients of (a) PE1, (b) PE2,
(c) PE3, and (d) PE4 on TiO2nanocrystalline films. Open circles denote
raw data obtained atλex) 430 nm and λem) 680 nm. The solid and
dashed curves represent theoretical fits with convolution of the laser pulse.
Figure 12. IV curves of porphyrin-sensitized SC for (a) PE1, (b) PE2,
(c) PE3, and (d) PE4. Three independent measurements were conducted with the same TiO2films (labeled a-c, Table 4); devices were fabricated
with the same procedure. All data were collected under one solar AM1.5 illumination (100 mW cm-2) with active area 0.25 cm2.
rapid electron transfer from TiO2back to the dye cation. The
kinetics of charge recombination for the system reported here are thus of great interest; this work is in progress.
3. Conclusion
Novel porphyrins (PE1-PE4) were designed and synthesized for the purposes of investigating the interfacial electron-transfer and DSSC applications. The porphyrins involve a ZnBPP moiety as a light-harvesting center and a carboxylic acid as an anchoring group with theπ-conjugated PE units serving as a link to control the distance between the porphyrin core and the TiO2surface.
When these porphyrins were fabricated into DSSC devices, the efficiencies of power conversion of these devices decreased systematically from 2.5 ( 0.2% for PE1 to 0.25 ( 0.02% for PE4. Measurements of femtosecond fluorescence of the por-phyrin-sensitized TiO2films showed that the emission decays
of the system are similar for PE1-PE4, indicating that electron injection from the porphyrin core to the surface of TiO2might
be equally rapid for PE1-PE4. DFT calculations indicate that electron densities are localized on the porphyrin core and the first PE link, which might indicate that our observations reflect a narrow detection window that lacks sensitivity to probe electron transfer through links of varied length. Although a slow electron injection in a long link might occur at the dye/TiO2
interface, as Galoppini and co-workers11eand Gundlach et al.31
reported, other factors must be considered to account for the great degradation of performance in the devices with a long PE link.
4. Experiments
4.1. Materials. Solvents CH2Cl2, CHCl3, and ethyl acetate
(ACS grade, Mallinckrodt Baker, Kentucky, USA), hexanes (Haltermann, Hamburg, Germany), and THF (Merck, Darmstadt, Germany) in the syntheses were used as received unless otherwise stated. THF and tetrabutylammonium perchlorate (TBAP) were purified according to literature methods.33
Pd-(PPh3)4 catalyst (Strem Chemical Inc., MA, USA), silica gel
60 for chromatographic purification (230-400 mesh, Merck), and other chemicals (Acros Organics, NJ, USA) were obtained from the indicated suppliers.
4.2. General Syntheses. The Sonogashira cross-coupling method was used to perform most reactions:13
5-bromo-10,20-biphenylporphinato zinc(II) (ZnBPPBr, 100 mg) under dinitro-gen was typically mixed with ethynyl substituents (2-4 equiv), Pd(PPh3)4(20 mol%), CuI (10 mol%), and Et2NH (5 mL) in
freshly distilled THF (75-85 mL). The degassed reaction mixtures were stirred under dinitrogen at 40°C and monitored with thin-layer chromatography and UV-visible spectra. Upon completion, the reactions were quenched with NH4Cl(aq)washes
followed by chromatographic separation on silica gel with CH3OH/CH2Cl2eluents and crystallization from THF/hexanes.
PE1 was synthesized on cross-coupling ZnBPPBr and 4-ethy-nylbenzoic acid in 80% yield. 4-Ethy4-ethy-nylbenzoic acid was obtained from cross-coupling trimethylsilylacetylene with 4-io-dobenzoic acid (78%) with deprotection from tetrabutylammo-nium fluoride (TBAF/ THF) in 82% yield.1H NMR (d
6-DMSO
at 2.50 ppm)δ/ppm 10.36 (s, 1H), 9.84 (d, J 5 Hz, 2H), 9.46 (d, J 5 Hz, 2H), 8.92 (d, J 5 Hz, 2H), 8.84 (d, J 5 Hz, 2H), 8.25 (m, 8H), 7.86 (m, 6H). Anal. C41H24N4O2Zn · C4H8O: calcd. C
72.83%, H 4.35%, N 7.55%; found C 72.35%, H 4.60%, N 7.27%. Visible absorption/nm: 439 (log /cm-1M-1) 5.64), 567 (4.21), 616 (4.41).
PE2 was prepared from ZnBPPBr and 4-(4-ethynyl-phenyl-ethynyl)benzoic acid in 32% yield. 4-(4-Ethynyl-phenylethy-nyl)benzoic acid was generated on cross-coupling 4-ethynyl-benzoic acid with 1,4-diethynylbenzene (65%). Cross-coupling diiodobenzene with trimethylsilylacetylene yielded 1,4-bis(trimethylsilylethynyl)benzene in nearly 100% yield. The deprotection reaction in TBAF/THF gave 1,4-diethynylbenzene (81%) after chromatographic separation.1H NMR (d
6-DMSO at 2.50 ppm)δ/ppm 10.34 (s, 1H), 9.38 (d, J 5 Hz, 2H), 9.45 (d, J 5 Hz, 2H), 8.90 (d, J 5 Hz, 2H), 8.83 (d, J 5 Hz, 2H), 8.20 (t, J 4 Hz, 6H), 8.01 (d, J 8 Hz, 2H), 7.86 (m, 8H), 7.74 (d, J 8 Hz, 2H). Anal. C49H28N4O2Zn · C4H8O: calcd. C 75.58%, H 4.31%, N 6.65%; found C 75.05%, H 4.27%, N 6.68%. Visible absorption/nm: 443 (log ) 5.67), 569 (4.27), 618 (4.60).
PE3 was obtained on reacting 5-(4-ethynyl-phenylethynyl)-ZnBPP with 4-(4-iodo-phenylethynyl)benzoic acid in 30% yield. 5-(4-Ethynyl-phenylethynyl)-ZnBPP was generated on cross-coupling ZnBPPBr with 1,4-diethynylbenzene at 41% yield. 4-(4-Iodo-phenylethynyl)benzoic acid was produced from 1,4-diiodobenzene and 4-ethynylbenzoic acid in 19% yield. 1H
NMR(d6-DMSO at 2.50 ppm)δ/ppm 10.33 (s, 1H), 9.83 (d, J
5 Hz, 2H), 9.44 (d, J 5 Hz, 2H), 8.90 (d, J 4 Hz, 2H), 8.93 (d,
TABLE 4: Photovoltaic Parameters of Porphyrin-Based DSSCs under AM1.5 Illumination (Power 100 mW cm-2) and Active
Area 0.25 cm2
dye films JSC/mA cm-2 VOC/V FF η/%
Zn3 a 9.4 0.65 0.62 3.78 b 10.5 0.63 0.63 4.17 c 10.1 0.65 0.62 4.07 average 10.0 ( 0.6 0.64 ( 0.01 0.62 ( 0.01 4.0 ( 0.2 PE1 a 7.3 0.57 0.65 2.70 b 6.4 0.55 0.68 2.39 c 6.2 0.56 0.67 2.33 average 6.6 ( 0.7 0.56 ( 0.01 0.67 ( 0.02 2.5 ( 0.2 PE2 a 5.6 0.58 0.66 2.14 b 5.5 0.57 0.65 2.04 c 5.6 0.55 0.62 1.91 average 5.5 ( 0.1 0.57 ( 0.02 0.64 ( 0.02 2.0 ( 0.1 PE3 a 2.1 0.57 0.70 0.84 b 2.3 0.56 0.70 0.82 c 1.7 0.56 0.72 0.68 average 2.0 ( 0.3 0.56 ( 0.01 0.70 ( 0.01 0.78 ( 0.09 PE4 a 0.67 0.51 0.69 0.24 b 0.70 0.49 0.66 0.23 c 0.72 0.52 0.73 0.27 average 0.70 ( 0.03 0.51 ( 0.02 0.69 ( 0.04 0.25 ( 0.02
J 4 Hz, 2H), 8.21 (t, J 8 Hz, 6H), 7.99 (d, J 8 Hz, 2H), 7.85 (overlapped, 8H), 7.69 (overleapped, 6H). Anal: C57H32N4O2Zn ·
CH3OH · 3H2O: calcd. C 72.82%, H 4.43%, N 5.86%; found C
72.59%, H 4.46%, N 5.26%. Visible absorption/nm: 443 (log ) 5.58), 569 (4.21), 619 (4.56).
PE4 was prepared from 5-[4-(4-iodo-phenylethynyl)-phenyl-ethynyl]-ZnBPP and 4-(4-ethynyl-phenylethynyl)benzoic acid in 38% yield. 5-[4-(4-Iodo-phenylethynyl)-phenylethynyl]-Zn-BPP was synthesized on cross-coupling 5-(4-ethynyl-phenyl-ethynyl)-ZnBPP with 1,4-diiodobenzene (yield 39%). 4-(4-Ethynyl-phenylethynyl)benzoic acid was generated on cross-coupling 1,4-diethynylbenzene with 4-iodobenzoic acid (yield 65%). 1H NMR d
6-DMSO at 2.50 ppmδ/ppm 10.34 (s, 1H),
9.83 (d, J 4 Hz, 2H), 9.45 (d, J 4 Hz, 2H), 8.90 (d, J 4 Hz, 2H), 8.82 (d, J 4 Hz, 2H), 8.20 (t, J 4 Hz, 6H), 7.95 (d, J 8 Hz, 2H), 7.86 (overlapped, 8H), 7.66 (overlapped, 10H). Anal. C65H36N4O2Zn · 2H2O: calcd. C 77.57%, H 4.01%, N 5.57%;
found C 77.61%, H 4.00%, N 5.36%. Visible absorption/nm: 444 (log ) 5.56), 569 (4.17), 619 (4.53).
4.3. Steady-State Spectral Measurements. Absorption spec-tra (Agilent 8453 UV-visible spectrophotometer), fluoresence spectra (Cary Eclipse fluorescence spectrophotometer), NMR spectra (Varian Unity Inova 300WB NMR spectrometer), and elemental analyses (Elementar Vario EL III) were measured with the indicated instruments. Cyclic voltammograms (CH Instru-ments Electrochemical Workstation 660A) were recorded with a standard three-electrode cell: a Pt working electrode, a Pt auxiliary electrode isolated from the main compartment by a layer of fine glass frit, and a reference electrode (SCE) isolated from the main compartment by a junction tipped with a platinum wire. A glovebox (M. Braun Unilab), a vacuum line, and standard Schlenk glassware were employed to process all air-sensitive materials.
4.4. Time-Resolved Fluorescence Measurements. Fluores-cence decays were recorded with an optically gated (up-conversion) system (FOG100, CDP), described elsewhere.34
Briefly, the femtosecond laser system (Mira 900D, Coherent) generated output pulses at 860 nm (or 840 nm) of duration∼120 fs at a repetition rate 76 MHz. The frequency of the laser pulse was doubled for excitation (λex) 430 or 420 nm). The residual
fundamental pulse served as an optical gate; a dichroic beam splitter separated excitation and gate beams. The intensity of the excitation beam was appropriately attenuated and focused onto a rotating cell (optical path of length 1 mm) containing either the solution or the solid thin-film samples. The emission
was collected with two parabolic mirrors and focused onto a crystal (BBO type-I); the gate pulse was also focused on that BBO crystal for sum-frequency generation. The latter signal was collected with a lens, separated from interference light with an iris, a band-pass filter, and a double monochromator (DH10, Jobin Yvon) in combination, and then detected with a photo-multiplier (R1527P, Hamamatsu) connected to a computer-controlled photon-counting system. By variation of the temporal delay between gate and excitation pulses via a stepping-motor translational stage, we obtained a temporal profile (transient). The polarization between pump and probe pulses was fixed at the magic-angle condition, 54.7°.
4.5. Cell Fabrication and Performance Characterization. Three key components are essential in the construction of a sandwich-type DSSC device: a light-harvesting layer (dye/TiO2)
is deposited on a transparent conducting oxide (TCO) surface to serve as a working electrode (anode); a Pt-coated layer is deposited on a TCO surface to serve as a counter electrode (cathode), and an diiodine-based electrolyte fills the space between the anode and the cathode to serve as a redox couple of the cell. For a working electrode, a TiO2nanoparticulate film
was produced on a fluoride-doped tin oxide (FTO, 30 Ω/0, Sinonar, Taiwan) glass via screen printing. Crystallization of TiO2films (thickness ∼6 µm and active area 0.25 cm2) was
performed on annealing in two stages: heating at 450°C for 5 min followed by heating at 500°C for 30 min. The TiO2film
was then immersed in a aqueous solution of TiCl4(40 mM, 70
°C, 30 min) followed by the same two-stage thermal treatment for final annealing of the electrode. The electrode was then immersed in the dye/THF solution for dye loading onto the TiO2
film. The Pt counter electrodes were prepared on spin-coating drops of H2PtCl6solution onto FTO glass and heating at 400
°C for 15 min. To prevent a short circuit, the two electrodes were assembled into a cell of sandwich type and sealed with a hot-melt film (SX1170, Solaronix, thickness 25 µm). The electrolyte solution containing LiI (0.1 M), I2(0.05 M), BMII
(0.6 M), and 4-tert-butylpyridine (0.5 M) in a mixture of acetonitrile and valeronitrile (volume ratio 1:1) was introduced into the space between the two electrodes, so completing the fabrication of these DSSC devices.
The performance of a DSSC device was assessed on measurement of an IV curve with an AM-1.5 solar simulator (Newport-Oriel 91160). The solar simulator employs filters and other optical components to mimic solar radiation with an air mass 1.5 spectrum; the output intensity is evenly distributed for illumination of a large area. When the device is irradiated with the solar simulator, the source meter (Keithley 2400, computer-controlled) sends a voltage (V) to the device, and the photocurrent (I) is read at each step controlled by a computer via a GPIB interface. The solar simulator was calibrated with a Si-based reference cell (S1133, Hamamatsu) and an IR-cut filter (KG5) to correct the spectral mismatch of the lamp.35 The
actively illuminated area was 0.25 cm2for all measurements.
Acknowledgment. National Science Council of Republic of China (Project Contracts 2628-M-009-018-MY2 and 96-2627-M-009-005 for E.W.-G.D. and 95-2113-M-260-006-MY2 and 96-2627-M-260-001 for C.-Y.L.) provided financial support. References and Notes
(1) O’Regan, B.; Gra¨tzel, M. Nature 1991, 353, 737.
(2) Nazeeruddin, M. K.; Kay, A.; Rodicio, I.; Humphry-Baker, R.; Muller, E.; Liska, P.; Vlachopoulos, N.; Gra¨tzel, M. J. Am. Chem. Soc.
1993, 115, 6382.
(3) (a) Wang, Q.; Ito, S.; Gra¨tzel, M.; Fabregat-Santiago, F.; Mora-Sero´, I.; Bisquert, J.; Bessho, T.; Imai, H. J. Phys. Chem. B 2006, 110,
Figure 13. Photocurrent action spectra of PE1- and PE4-sensitized
25210. (b) Wei, M.; Konishi, Y.; Zhou, H.; Yanagida, M.; Sugihara, H.; Arakawa, H. J. Mater. Chem. 2006, 16, 1287. (c) Koide, N.; Islam, A.; Chiba, Y.; Han, L. J. Photochem. Photobiol. A 2006, 182, 296.
(4) (a) Nazeeruddin, M. K.; Humphry-Baker, R.; Officer, D. L.; Campbell, W. M.; Burrell, A. K.; Gra¨tzel, M. Langmuir 2004, 20, 6514. (b) Campbell, W. M.; Burrell, A. K.; Officer, D. L.; Jolley, K. W. Coord.
Chem. ReV. 2004, 248, 1363. (c) Wang, Q.; Campbell, W. M.; Bonfantani,
E. E.; Jolley, K. W.; Officer, D. L.; Walsh, P. J.; Gordon, K.; Humphry-Baker, R.; Nazeeruddin, M. K.; Gra¨tzel, M. J. Phys. Chem. B 2005, 109, 15397. (d) Campbell, W. M.; Jolley, K. W.; Wagner, P.; Wagner, K.; Walsh, P. J.; Gordon, K.; Schmidt-Mende, L.; Nazeeruddin, M. K.; Wang, Q.; Pechy, P.; Renouard, T.; Zakeeruddin, S. M.; Gra¨tzel, M.; Officer, D. J.
Phys. Chem. C 2007, 111, 11760.
(5) Odobel, F.; Blart, E.; Lagree, M.; Villieras, M.; Boujtita, H.; Murr, N. E.; Caramori, S.; Bignozzi, C. A. J. Mater. Chem. 2003, 13, 502.
(6) Bignozzi, C. A.; Argazzi, R.; Kleverlaan, C. J. Chem. Soc. ReV.
2000, 29, 87.
(7) Cid, J.-J.; Yum, J.-H.; Jang, S.-R.; Nazeeruddin, M. K.; Martı´nez-Ferrero, E.; Palomares, E.; Ko, J.; Gra¨tzel, M.; Torres, T. Angew. Chem.,
Int. Ed. 2007, 46, 8358.
(8) (a) Luo, L.; Lo, C.-F.; Lin, C.-Y.; Chang, I.-J.; Diau, E. W.-G. J.
Phys. Chem. B 2006, 110, 410. (b) Lo, C.-F.; Luo, L.; Diau, E. W.-G.;
Chang, I.-J.; Lin, C.-Y. Chem. Commun. 2006, 1430.
(9) (a) Siemsen, P.; Livingston, R. C.; Diederich, F. Angew. Chem.,
Int. Ed. 2000, 39, 2632. (b) Bunz, U. H. F. Chem. ReV. 2000, 100, 1605.
(c) Yang, S. I.; Prathapan, S.; Miller, M. A.; Seth, J.; Bocian, D. F.; Lindsey, J. S.; Holten, D. J. Phys. Chem. B 2001, 105, 8249.
(10) (a) Kim, D.; Osuka, A. J. Phys. Chem. A 2003, 107, 8791. (b) Holten, D.; Bocian, D. F.; Lindsey, J. S. Acc. Chem. Res. 2002, 35, 57, and references therein. (c) Loewe, R. S.; Lammi, R. K.; Diers, J. R.; Kirmaier, C.; Bocian, D. F.; Holten, D.; Lindsey, J. S. J. Mater. Chem.
2002, 12, 1530. (d) Wojaczynski, J.; Latos-Grazynski, L. Coord. Chem. ReV. 2000, 204, 113. (e) Rubtsov, I. V.; Susumu, K.; Rubtsov, G. I.; Therien,
M. J. J. Am. Chem. Soc. 2003, 125, 2687. (f) Susumu, K.; Therien, M. J.
J. Am. Chem. Soc. 2002, 124, 8550. (g) Screen, T. E. O.; Lawton, K. B.;
Wilson, G. S.; Dolney, N.; Ispasoiu, R.; Goodson, T., III.; Martin, S. J.; Bradley, D. D. C.; Anderson, H. L. J. Mater. Chem. 2001, 11, 312.
(11) (a) Piotrowiak, P.; Galoppini, E.; Wei, Q.; Meyer, G. J.; Wiewior, P. J. Am. Chem. Soc. 2003, 125, 5278. (b) Hoertz, P. G.; Carlisle, R. A.; Meyer, G. J.; Wang, D.; Piotrowiak, P.; Galoppini, E. Nano Lett. 2003, 3, 325. (c) Galoppini, E.; Guo, W.; Zhang, W.; Hoertz, P. G.; Qu, P.; Meyer, G. J. J. Am. Chem. Soc. 2002, 124, 7801. (d) Galoppini, E. Coord. Chem.
ReV. 2004, 248, 1283. (e) Myahkostupov, M.; Piotrowiak, P.; Wang, D.;
Galoppini, E. J. Phys. Chem. C 2007, 111, 2827. (f) Rochford, J.; Chu, D.; Hagfeldt, A.; Galoppini, E. J. Am. Chem. Soc. 2007, 129, 4655. (g) Rochford, J.; Galoppini, E. Langmuir 2008, 24, 5366.
(12) Sazanovich, I. V.; Kirmaier, C.; Hindin, E.; Yu, L.; Bocian, D. F.; Lindsey, J. S.; Holten, D. J. Am. Chem. Soc. 2004, 126, 2664.
(13) (a) Sonogashira, K.; Tohda, Y.; Hagihara, N. Tet. Lett. 1975, 4467. (b) Takahashi, S.; Kuroyama, Y.; Sonogashira, K. Synthesis 1980, 627.
(14) (a) Zhang, T.-G.; Zhao, Y.; Asselberghs, I.; Persoons, A.; Clays, K.; Therien, M. J. J. Am. Chem. Soc. 2005, 127, 9710. (b) Susumu, K.; Duncan, T. V.; Therien, M. J. J. Am. Chem. Soc. 2005, 127, 5186. (c) Liu, Z.; Schmidt, I.; Thamyongkit, P.; Loewe, R. S.; Syomin, D.; Diers, J. R.; Zhao, Q.; Misra, V.; Lindsey, J. S.; Bocian, D. F. Chem. Mater. 2005, 17, 3728. (d) Yasseri, A. A.; Syomin, D.; Malinovskii, V. L.; Loewe, R. S.; Lindsey, J. S.; Zaera, F.; Bocian, D. F. J. Am. Chem. Soc. 2004, 126, 11944. (e) Screen, T. E. O.; Thorne, J. R. G.; Denning, R. G.; Bucknall, D. G.; Anderson, H. L. J. Am. Chem. Soc. 2002, 124, 9712. (f) Blake, I. M.; Rees, L. H.; Claridge, T. D. W.; Anderson, H. L. Angew. Chem., Int. Ed. 2000,
39, 1818. (g) Wilson, G. J.; Arnold, D. P. J. Phys. Chem. A 2005, 109,
6104. (h) Webb, S. J.; Sanders, J. K. M. Inorg. Chem. 2000, 39, 5912. (15) (a) Lin, C.-Y.; Chuang, L.-C.; Lin, Y.-C.; Lin, C.-L. Dalton Trans.
2004, 4006. (b) Lin, C.-Y.; Chuang, L.-C.; Yang, Y.-F.; Lin, C.-L.; Kao,
H.-C.; Wang, W.-J. Dalton Trans. 2004, 456. (c) Blake, I. M.; Rees, L. H.; Claridge, T. D. W.; Anderson, H. L. Angew. Chem., Int. Ed. 2000, 39, 1818. (d) Taylor, P. N.; Wylie, A. P.; Huuskonen, J.; Anderson, H. L. Angew.
Chem., Int. Ed. 1998, 37, 986. (e) LeCours, S. M.; Philips, C. M.; de Paula,
J. C.; Therien, M. J. J. Am. Chem. Soc. 1997, 119, 12578. (f) LeCours, S. M.; Guan, H. W.; DiMagno, S. G.; Wang, C. H.; Therien, M. J. J. Am.
Chem. Soc. 1996, 118, 1497.
(16) Gouterman, M. J. Mol. Spectrosc. 1961, 6, 138.
(17) Nielsen, K. T.; Spanggaard, H.; Krebs, F. C. Macromolecules 2005,
38, 1180.
(18) Gurzadyan, G. G.; Tran-Thi, T.-H.; Gustavsson, T. J. Chem. Phys.
1998, 108, 385.
(19) Mataga, N.; Shibata, Y.; Chosrowjan, H.; Yoshida, N.; Osuka, A.
J. Phys. Chem. B 2000, 104, 4001.
(20) Yu, H.; Baskin, J. S.; Zewail, A. H. J. Phys. Chem. A 2002, 106, 9845.
(21) (a) Kadish, K. M.; Caemelbecke, E. V.; Royal, G. In The Porphyrin
Handbook; Kadish, K. M.; Smith, K. M.; Guilard, G., Eds.;Academic Press:
New York, 2000; Vol. 8, pp 1-97. (b) Kadish, K. M.; Royal, G.; Caemelbecke, E. V.; Gueletti, L. In The Porphyrin Handbook; Kadish, K. M.; Smith, K. M.; Guilard, G., Eds.; Academic Press: New York, 2000; Vol. 9.
(22) We observed ill-defined reduction waves before the porphyrin first ring-reductions; these reduction waves are similar to the reduction waves of the substituents under the same experimental conditions, e.g., 4-iodo-benzoic acid, 4-ethynyl4-iodo-benzoic acid, 4-(trimethylsilylethynyl)4-iodo-benzoic acid, and 4-(4-ethynylphenylethynyl)benzoic acid.
(23) (a) Thomas, K. R. J.; Lin, J. T.; Tao, Y.-T.; Ko, C.-W. J. Am.
Chem. Soc. 2001, 123, 9404. (b) Koene, B. E.; Loy, D. E.; Thompson,
M. E. Chem. Mater. 1998, 10, 2235. (c) Thelakkat, M.; Schmidt, H.-W.
AdV. Mater. 1998, 10, 219. (d) Janietz, S.; Bradley, D. D. C.; Grell, M.;
Giebeler, C.; Inbasekaran, M.; Woo, E. P. Appl. Phys. Lett. 1998, 73, 2453. (24) (a) Hagfeldt, A.; Gra¨tzel, M. Chem. ReV. 1995, 95, 49. (b) Pommerehne, J.; Vestweber, H.; Guss, W.; Mahrt, R. F.; Bassler, H.; Porsch, M.; Daub, J. AdV. Mater. 1995, 7, 551.
(25) To avoid saturation of the intensity of the film absorption band, we limited the amount of PE1-PE4 adsorbed onto the TiO2plates. Some
porphyrins are expected to adsorb on TiO2surfaces in monomeric form
with a less extent of aggregation.
(26) (a) Donker, H.; Koehorst, R. B. M.; Schaafsma, T. J. J. Phys. Chem.
B 2005, 109, 17031. (b) Donker, H.; Hoek, A.; Schaik, W.; Koehorst,
R. B. M.; Yatskou, M. M.; Schaafsma, T. J. J. Phys. Chem. B 2005, 109, 17038. (c) Schaafsma, T.; Dag, I.; Sitters, R.; Glasbeek, M.; Lifshitz, E. J.
Phys. Chem. B 2005, 109, 17047.
(27) Kasha, M. Radiat. Res. 1963, 20, 55.
(28) Maiti, N. C.; Mazumdar, S.; Periasamy, N. J. Phys. Chem. B 1998,
102, 1528.
(29) (a) Chowdhury, A.; Wachsmann-Hogiu, S.; Bangal, P. R.; Raheen, I.; Peteanu, L. A. J. Phys. Chem. B 2001, 105, 12196. (b) Wang, M.; Silva, G. L.; Armitage, B. A. J. Am. Chem. Soc. 2000, 122, 9977.
(30) Lu, Y.-C.; Chang, C.-W.; Diau, E. W.-G. J. Chin. Chem. Soc. 2002,
49, 693.
(31) Gundlach, L.; Ernstorfer, R.; Willig, F. J. Phy. Chem. C 2007, 111, 13586.
(32) Clark, C. C.; Meyer, G. J.; Wei, Q.; Galoppini, E. J. Phy. Chem.
B 2006, 110, 11044.
(33) Perrin, D. D.; Armarego, W. L. F. Purification of laboratory
chemicals, 3rd ed.; Pergamon Press: Oxford, 1988.
(34) Lu, Y. C.; Diau, E. W.-G.; Rau, H. J. Phys. Chem. A 2005, 109, 2090.
(35) Ito, S.; Matsui, H.; Okada, K.; Kusano, S.; Kitamura, T.; Wada, Y.; Yanagida, S. Sol. Energy Mater. Sol. Cells 2004, 82, 421.
