Highly asymmetrical porphyrins with enhanced push-pull character for dye-sensitized solar cells

Article abstract of DOI:10.1002/chem.201303460

A porphyrin π-system has been modulated by enhancing the push-pull character with highly asymmetrical substitution for dye-sensitized solar cells for the first time. Namely, both two diarylamino moieties as a strong electron-donating group and one

Full text of DOI:10.1002/chem.201303460

FULL PAPER
DOI: 10.1002/chem.201303460
Highly Asymmetrical Porphyrins with Enhanced Push–Pull Character for
Dye-Sensitized Solar Cells
Kei Kurotobi,[a] Yuuki Toude,[b] Kyosuke Kawamoto,[b] Yamato Fujimori,[b] Seigo Ito,[c]
Pavel Chabera,[d] Villy Sundstrçm,*[d] and Hiroshi Imahori*[a, b]
Abstract: A porphyrin p-system has
been modulated by enhancing the
push–pull character with highly asym-
metrical substitution for dye-sensitized
solar cells for the first time. Namely,
both two diarylamino moieties as
a strong electron-donating group and
one carboxyphenylethynyl moiety as
a strong electron-withdrawing, anchor-
ing group were introduced into the
meso-positions of the porphyrin core in
a
lower symmetrical manner. As
than 10% power conversion efficiency,
which exceeded that of a representative
highly efficient porphyrin (i.e., YD2)-
sensitized solar cell under optimized
conditions. The rational molecular
design concept based on highly asym-
metric, push–pull substitution will open
the possibilities of further improving
cell performance in organic solar cells.
a result of the improved light-harvest-
ing property as well as high electron
distribution in the anchoring group of
LUMO, a push–pull-enhanced, porphy-
rin-sensitized solar cell exhibited more
Keywords: absorption spectrosco-
py · electron transfer · porphyri-
noids · solar cells · titanium oxide
Introduction
400 nm (ca. 105 mcmÀ1) and moderate Q bands at 500–
600 nm (ca. 104 mcmÀ1)] and synthetic versability.[7–11]
A
Dye-sensitized solar cells (DSSCs) based on mesoporous,
nanocrystalline oxides are a promising low-cost alternative
to conventional silicon-based solar cells for the conversion
of solar energy to electric power.[1] Because of the critical
role of sensitizers in DSSC performance, optimizing molecu-
lar design is essential to maximize power conversion effi-
ciency (h). Current high-efficiency DSSCs (h=10–12%)
have been based on Ru complexes as sensitizers.[1–4] Howev-
er, the limited supply and potential environmental impact of
Ru have led to intensive search for green, organic-based
dyes.[5,6] Porphyrins are a family of organic dyes that are fas-
cinating for application in DSSCs owing to their extremely
high molar extinction coefficients [i.e., intense Soret band at
drawback of typical porphyrins is poor light-harvesting
property at wavelengths around 450 nm and >600 nm. Nev-
ertheless, the integrated absorption molar extinction coeffi-
cient of a typical porphyrin with respect to wavelength is
several times larger than that of Ru complexes, demonstrat-
ing the potential high utility of porphyrins in DSSCs.
Indeed, with elongation of p-conjugation as well as enhance-
ment of push–pull character in porphyrins, the absorption
can be shifted and broadened, making it possible to increase
the light-harvesting property and the resultant h-value. Spe-
cifically, Diau and Grꢀtzel et al. synthesized a push–pull por-
phyrin YD2 with a diarylamino moiety as a strong electron-
donating group and a carboxyphenylethynyl moiety as
a strong electron-withdrawing, anchoring group (Figure 1);
YD2 exhibited a high h-value of 11%.[12] They further uti-
[a] Prof. Dr. K. Kurotobi, Prof. Dr. H. Imahori
Institute for Integrated Cell-Material Sciences (WPI-iCeMS)
Kyoto University, Nishikyo-ku, Kyoto 615-8510 (Japan)
Fax : (+81)75-383-2571
lized a cobaltACTHNUGTRNEUNG(II/III) electrolyte that can yield a higher open
circuit voltage (VOC), with a cocktail of a YD2 derivative,
YD2-o-C8 (Figure 1), and a complementary organic dye to
achieve a record h-value of 12.3%.[13] However, there is still
room to improve the h-value by modulating HOMO and
LUMO levels as well as other critical factors affecting
DSSC performance.
E-mail: imahori@scl.kyoto-u.ac.jp
[b] Y. Toude, K. Kawamoto, Y. Fujimori, Prof. Dr. H. Imahori
Department of Molecular Engineering
Graduate School of Engineering
Kyoto University, Nishikyo-ku, Kyoto 615-8510 (Japan)
Herein we report the achievement of an h-value of 10.1%
by using an asymmetrically enhanced push–pull porphyrin,
ZnPBAT (Figure 2), that is superior to YD2 (h=9.1%)
under our optimized conditions. We have previously exam-
ined the effects of the number and position of strongly elec-
tron-donating diarylamino groups at meso-positions of
5-(4-carboxyphenyl)porphyrins on DSSC performance
(Figure 1).[14] With increasing the number (n) of diarylamino
groups introduced into the meso-positions from n=0 to 2,
[c] Prof. Dr. S. Ito
Department of Electrical Engineering and Computer Sciences
Graduate School of Engineering, University of Hyogo
2167 Shosha, Himeji, Hyogo 671-2280 (Japan)
[d] Dr. P. Chabera, Prof. Dr. V. Sundstrçm
Department of Chemical Physics, Lund University
Box 124, 22100 Lund (Sweden)
E-mail: villy.sundstrom@chemphys.lu.se
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201303460.
Chem. Eur. J. 2013, 19, 17075 – 17081
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
17075

6.5%)]. This trend can be rationalized by the lower electron
density of the anchoring groups (i.e., carboxyphenyl group)
in the LUMO of the zinc–porphyrin complexes (ZnP) with
the two diarylamino moieties (n=2; cis-ZnP and trans-ZnP)
than with the one diarylamino moiety (mono-ZnP), which
would lower the electronic coupling between the excited ad-
sorbed dye and 3d orbital of TiO2, as well as the anchoring
ability onto TiO2, leading to a decrease in DSSC perfor-
mance.[15] To increase the electron density of the anchoring
group in the LUMO as well as the push–pull character of
the porphyrin, an anchoring group with more electron-with-
drawing character is desirable. More importantly, introduc-
tion of the two diarylamino moieties into the meso-positions
in a low-symmetry manner, as in the case of cis-ZnP, rather
than a high-symmetry manner, as in the case of trans-ZnP,[14]
would further enhance the push–pull effects, as inspired by
the high electronic asymmetry of photosynthetic chloro-
phylls that lead to broadening and redshift of the absorption
and enhanced Q bands relative to Soret band.[7] In this
regard, there are still challenges in synthesizing asymmetric
porphyrins with different substituents at four meso-posi-
tions.[16,17] With these synthetic requirements in hand, we de-
cided to introduce one carboxylphenylethynyl moiety as
a strong electron-withdrawing, anchoring group[12,13,18,19] and
two diarylamino moieties as strong electron-donating groups
directly into the meso-positions in an asymmetric manner
(Figure 2). One mesityl moiety was also tethered to the re-
maining meso-position to retain the highly asymmetrical
meso-substitution. ZnPBA was utilized as a reference to ad-
dress the effect of the carboxyphenylethynyl moiety
(Figure 2), whereas YD2 was used to assess the asymmetric
double substitution by the diarylamino moieties.
Figure 1. Molecular structures of porphyrins in previous studies.[12–14]
Figure 2. Molecular structures of porphyrins for DSSCs.
Scheme 1. Synthetic route to ZnPBAT. 1) trifluoroacetic acid, CHCl3,
96%; 2) a) NBS (0.9 equiv), CH2Cl2/pyridine; b) Zn
MeOH, 27% (2 steps); 3) triisopropylsilylacetylene, [PdCl
THF/Et3N, 87%; 4) NBS (2 equiv), CH2Cl2/pyridine, 50%; 5) bis(4-meth-
ylphenyl)amine, nBuLi, Pd(OAc)2, DPEphos, toluene, 61%; 6) a) TBAF,
ACHTUNGTRENNUNG
the light-harvesting property of the porphyrins was im-
proved in the order ZnP-ref (n=0)<mono-ZnP (n=1)<
cis-ZnP (n=2)ꢀtrans-ZnP (n=2), but the solar cell reached
a maximum h-value in n=1 [trans-ZnP (h=3.8%)<ZnP-
ref (h=4.4%)<cis-ZnP (h=5.5%)<mono-ZnP (h=
CTHUNGTRENNUNG
AHCTUNGTRENNUNG
THF; b) 4-iodobenzoic acid, [Pd
(2 steps).
2ACTHNUGTREN(NUNG dba)3], AsPh3, THF/Et3N, 79%
17076
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 17075 – 17081

Highly Asymmetrical Porphyrins
FULL PAPER
Results and Discussion
The synthesis of ZnPBAT is shown in Scheme 1. The key
feature involves highly asymmetric successive introduction
of three different substituents (i.e., two diarylamino, one
carboxylphenylethynyl, and one trimethylphenyl moieties)
into the four meso-positions of a porphyrin core. Monoaryl-
Figure 3. UV/Vis absorption spectra of ZnPBAT (solid line), YD2
(dotted line), and ZnPBA (dashed line) in EtOH.
porting Information. In accordance with the trend of the ab-
sorption properties, the emission maximum is shifted toward
longer wavelength in the order of ZnPBA, YD2, and
ZnPBAT. From the intersection of the normalized absorp-
tion and fluorescence spectra, the zero-zero excitation ener-
gies (E0–0) are determined to be 1.83, 1.89, and 1.92 eV for
ZnPBAT, YD2, and ZnPBA, respectively (Table S1 in the
Supporting Information).
Scheme 2. Synthetic route to ZnPBA. 1) a) BF3·OEt2, CHCl3; b) p-chlor-
anil, CHCl3, 2.0% (2 steps); 2) Zn
ACHTUGNTREN(NUNG OAc)2·2H2O, CHCl3/MeOH, 85%; 3)
a) bis(4-methylphenyl)amine, iodobenzenediacetate, NaAuCl4·2H2O,
CH2Cl2; b) NaOH, THF; c) Zn
(3 steps).
ACHTUGNTREN(NUNG OAc)2·2H2O, CHCl3/MeOH, 15%
To determine the first oxidation potential (Eox) and the
first reduction potential (Ered) of the porphyrins in solution,
differential pulse voltammetry measurements were per-
formed in THF containing 0.1m Bu4NPF6 as a supporting
electrolyte. The Eox values of ZnPBAT (0.85 V vs. NHE)
and ZnPBA (0.85 V vs. NHE) are shifted in a negative di-
rection relative to that of YD2 (0.93 V vs. NHE), whereas
Ered of ZnPBAT (À1.15 V vs. NHE) and YD2 (À1.12 V vs.
NHE) are similar and shifted in a positive direction relative
to ZnPBA (À1.29 V vs. NHE) (Table S1 in the Supporting
Information). These results imply that the introduction of
the diarylamino and carboxyphenylethynyl moieties at the
meso-positions has desirable effect on the redox properties,
leading to the smaller bandgap in ZnPBAT. Indeed, the
electrochemical HOMO–LUMO gaps are determined to be
2.00, 2.05, and 2.14 eV for ZnPBAT, YD2, and ZnPBA, re-
spectively, which agrees with the trend of the optical
HOMO–LUMO gaps. From the spectroscopic and electro-
chemical measurements, driving forces for electron transfer
(ET) (DGinj) from the ZnP excited singlet state (1ZnP*) to
the conduction band (CB) of TiO2 (À0.5 V vs. NHE),[13] and
for the regeneration of the ZnP radical cation (ZnP+·) by
substituted free-base porphyrin 1 was synthesized from the
corresponding dipyrromethanes.[20] Monobromination of
1 yielded a mixture of intermediate cis- and trans-isomers,
of which the cis-intermediate was separated and subsequent-
ly treated with zinc acetate to afford the cis-isomer 2.[21]
SonoACHTUNGTRENNUNGgashira coupling of 2 and subsequent dibromination of
3 gave 4.[22] Diarylamino-substituted porphyrin 5 was ob-
tained in 61% yield by the substitution reaction of 4 with
bis(4-methylphenyl)amine.[23] Deprotection of
5
with
tetrabutACHTUNGTRENNUNGylACHTUNGTRENNUNGammonium fluoride (TBAF) followed by Sonoga-
shira coupling to 4-iodobenzoic acid produced ZnPBAT.[24]
ZnPBA was prepared according to similar procedures,[25,26]
as shown in Scheme 2. YD2 was also synthesized by follow-
ing the previously reported method.[27] Synthetic procedures
and optical and electrochemical properties are provided in
Supporting Information (Figure S1–S3).
Figure 3 displays the UV/Vis absorption spectra of the
ZnPs in EtOH. The peak positions and molar absorption co-
efficients (e) of Soret and Q bands are listed in Table S1 in
the Supporting Information. The absorption bands of
ZnPBAT and ZnPBA are broadened compared with those
of YD2 due to the introduction of the two diarylamino moi-
eties. Moreover, the absorption band of ZnPBAT is red-
shifted toward longer wavelength relative to those of
ZnPBA and YD2, illustrating the improved light-harvesting
property of ZnPBAT in the visible spectral region. The
steady-state fluorescence spectra of the ZnPs were mea-
sured in EtOH (Figure S4 in the Supporting Information)
and the emission maxima are listed in Table S1 in the Sup-
À
IÀ/I3 redox couple (+0.4 V vs. NHE)[13] (DGreg) have been
evaluated (Table S1 in the Supporting Information). Both of
the processes are thermodynamically feasible and the driv-
ing forces are sufficient for efficient ET.
DFT calculations were employed to gain insight into the
equilibrium geometry and electronic structures of the fron-
tier orbitals of the ZnPs.[14] They are found to possess planar
porphyrin rings, to which the diarylamino and trimethylphe-
nyl moieties are almost perpendicular, avoiding the steric
congestion around the meso-positions (Figure S5). Porphy-
AHCTUNGTERNGUNrN ins with D4h symmetry generally reveal two energetically
Chem. Eur. J. 2013, 19, 17075 – 17081
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
17077

V. Sundstrçm, H. Imahori et al.
degenerate LUMOs (LUMO+1, LUMO) and two nearly
degenerate HOMOs (HOMO, HOMOÀ1).[14] All the stud-
ied porphyrins largely retain these characteristics (Table S2
in the Supporting Information). The HOMO–LUMO gaps
of the ZnPs are as follows: ZnPBAT (2.15 eV)<YD2
(2.25 eV)<ZnPBA (2.41 eV). This trend agrees well with
those of the optical and electrochemical HOMO–LUMO
gaps. The electron density distributions of the ZnPs in their
respective HOMOs and LUMOs are visualized in Figure S6
in the Supporting Information. As expected, the electron
densities of the anchoring group in the LUMOs of ZnPBAT
and YD2 are comparable and significantly larger than that
of ZnPBA. Accordingly, we can anticipate a high electron
TiO2/ZnPBA. The FTIR spectra exhibit a significant in-
crease in the symmetric carboxylate band, n
around 1400 cmÀ1.[14] The disappearance of n
(C=O) and the
increased intensity of n
(COOsÀ) corroborate that a proton is
(COOsÀ), at
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
detached from the carboxylic acid group during the porphy-
rin adsorption on the TiO2 surface, leading to the bidentate
binding of the carboxylate group to TiO2. This is consistent
with the previous assignment that a carboxylic acid of analo-
gous porphyrins is bound to a TiO2 surface by means of
a bridging bidentate mode.[14]
To optimize the DSSC performance, we first examined
the effects of immersion time of the TiO2/ZnPBAT, TiO2/
YD2, and TiO2/ZnPBA electrodes (Figure S8 in the Sup-
porting Information).[28] The h-value is derived from the
equation: h=JSC ꢂVOC ꢂff, in which JSC is the short circuit
current and ff is the fill factor. The change in the h-value of
the TiO2/YD2 and TiO2/ZnPBA cells as a function of im-
mersion time matches the adsorption behavior of YD2 and
ZnPBA on TiO2. On the other hand, the h-value of the
TiO2/ZnPBAT cell is increased rapidly with increasing the
immersion time to reach a maximum in 1 h and then de-
creased to level off in 2 h. This suggests a stronger aggrega-
tion tendency of ZnPBAT than YD2 and ZnPBA, as seen in
the adsorption behavior of analogous porphyrins on TiO2.[14]
Note that the maximum h values are 7.1% after 1 h sensiti-
zation for ZnPBAT, 6.7% after 3 h for YD2, and 4.3% after
3 h for ZnPBA. To further optimize the cell performances,
each porphyrin was sensitized onto the TiO2 surface under
the best immersion time conditions in which coadsorbant
chenodeoxycholic acid (CDCA) was added to the porphyrin
solution to reduce the porphyrin aggregation on the TiO2
(Figure S9 in the Supporting Information). Maximum h
values are obtained at different CDCA concentrations for
the various ZnPs (ZnPBAT: 8.6% with 10 equiv CDCA;
YD2: 8.1% with 2 equiv CDCA; ZnPBA: 6.2% with
5 equiv CDCA), reflecting the difference in aggregation be-
havior of the porphyrins. Finally, TiO2/ZnPBAT, TiO2/YD2,
and TiO2/ZnPBA cells were kept under dark conditions to
evaluate the aging effect under dark conditions on the
DSSC performance (Figure 5).[29] All the cells reach the
highest h-value after several days aging. Figure 6 depicts the
photocurrent–voltage characteristics of the TiO2/ZnPBAT,
TiO2/YD2, and TiO2/ZnPBA cells under the respective max-
imal h-conditions (Table 1).
1
injection efficiency (finj) of ZnPBAT from the ZnP* to the
CB of TiO2, similar to YD2 (vide infra).[12,13]
A TiO2 electrode was immersed into a solution of the por-
phyrin in EtOH to give a porphyrin-stained TiO2 electrode.
The porphyrin surface coverage (G) adsorbed on the TiO2
film was determined by measuring the absorbance of the
porphyrin that was dissolved from the porphyrin-stained
TiO2 film into an alkaline aqueous solution containing THF.
During the immersion, all the porphyrins reach saturated
surface coverage on the TiO2 films in 3 h (Figure 4). The
Figure 4. Plots of porphyrin surface coverage (G) as a function of immer-
sion time for ZnPBAT (solid line with circles), YD2 (dotted line with tri-
angles), and ZnPBA (dashed line with squares) that adsorb on the TiO2
films without the scattering layer.
sat
ACHTUNGTRENNUNG
G=9.2ꢂ10À11
,
,
ZnPBAT, YD2, and ZnPBA, respectively. Given a projected
area of one porphyrin molecule on a flat TiO2 surface with
a perpendicular orientation, the saturated G values are cal-
culated to be 9.6ꢂ10À11 molcmÀ2 for ZnPBAT and ZnPBA
and 1.0ꢂ10À10 molcmÀ2 for YD2. Taking into account the
comparable calculated and experimental G values together
with the saturated adsorption behavior on the TiO2 surface,
all the porphyrins are concluded to form well-packed por-
phyrin monolayers on TiO2. The FTIR spectra of ZnPBAT,
YD2, and ZnPBA that were obtained from solid samples
Table 1. Photovoltaic performance under the optimized conditions.[a]
Aging JSC
[days]
[mAcmÀ2
VOC
[V]
ff
h
N
G
]
[%]
ZnPBAT 1.5 19.33
0.719
0.724
10.1
(18.61Æ0.7) (0.712Æ0.007) (0.734Æ0.01)
17.05 0.742 0.718
(16.75Æ0.3) (0.742Æ0.006) (0.715Æ0.02)
16.26 0.713 0.719
(15.69Æ0.6) (0.713Æ0.002) (0.731Æ0.01)
G
YD2
2
8
9.1
ZnPBA
reveal the characteristic band of n
acid group at around 1700 cmÀ1 (Figure S7 in the Supporting
Information).[14] This diagnostic peak for n
(C=O) disappears
for the FTIR spectra of TiO2/ZnPBAT, TiO2/YD2, and
ACHTUNGERTN(NUNG C=O) of the carboxylic
[a] Values correspond to photovoltaic parameters exhibiting the highest h
value. Values in parenthesis also denote average values from five inde-
pendent experiments.
AHCTUNGTRENNUNG
17078
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 17075 – 17081

Highly Asymmetrical Porphyrins
FULL PAPER
Figure 5. Plots of the power conversion efficiency (h) as a function of
aging time for TiO2/ZnPBAT cell (solid line with circles), TiO2/YD2 cell
(dotted line with triangles), and TiO2/ZnPBA cell (dashed line with
squares). The solar cells were kept under dark condition except the
period of the photovoltaic measurements.[29] The porphyrin-stained TiO2
electrodes were prepared under the optimized conditions (immersion
time: 1 h for ZnPBAT and 3 h for YD2 and ZnPBA; CDCA concentra-
tion: 10 equiv CDCA for ZnPBAT, 2 equiv CDCA for YD2, and 5 equiv
CDCA for ZnPBA).
Figure 7. Photocurrent action spectra of the TiO2/ZnPBAT cell (solid
line), the TiO2/YD2 cell (dotted line), and the TiO2/ZnPBA cell (dashed
line).
Figure 8. UV/Vis/NIR absorption spectra of the TiO2/ZnPBAT electrode
(solid line), the TiO2/YD2 electrode (dotted line), and the TiO2/ZnPBA
electrode (dashed line). The scattering TiO2 layers were not applied to
the TiO2 electrodes to measure absorbance accurately. The weak absorp-
tion at 800 nm can be attributed to the formation of ZnP+· on the
TiO2.[27]
Figure 6. Photocurrent–voltage characteristics of the TiO2/ZnPBAT cell
(solid line), the TiO2/YD2 cell (dotted line), and the TiO2/ZnPBA cell
(dashed line).
substitution in ZnPBAT in which the two diarylamino
groups are introduced at the meso-positions with a cis-con-
figuration.
The highest DSSC performance increases in the order
TiO2/ZnPBA (JSC =16.26 mAcmÀ2, VOC =0.713 V, ff=0.719,
h=8.3%)<TiO2/YD2 (JSC =17.05 mAcmÀ2, VOC =0.742 V,
ff=0.718, h=9.1%)<TiO2/ZnPBAT (JSC =19.33 mAcmÀ2,
Taking into account the rather similar VOC and ff values,
the difference in the h values primarily results from differ-
ences in the JSC values. Each of the photocurrent action
spectra (Figure 7) follows the absorption feature of the cor-
responding porphyrins on TiO2 (Figure 8). In fact, the order
of the improvement in the incident photon-to-current effi-
ciency (IPCE) values at 650–800 nm matches that in the
light-harvesting ability of the porphyrins (Figure 2), rational-
izing the higher JSC and h values of the TiO2/ZnPBAT cell
than the TiO2/YD2 cell. As a result of the enhanced push–
pull character achieved with the highly asymmetric substitu-
tion, the TiO2/ZnPBAT cell exhibits efficient photocurrent
generation extending to 800 nm.
VOC =0.719 V, ff=0.724, h=10.1%). Large improvement of
the TiO2/ZnPBA cell over the TiO2/trans-ZnP and cis-ZnP
cells[14] derives from the difference in the device fabrication
(i.e., immersion solvent, electrolyte solution, and TiO2 film)
and the aging effect under dark conditions.
It should be emphasized here that the h-value under our
experimental conditions of the TiO2/ZnPBAT cell is higher
than that of the TiO2/YD2 cell that has been reported to ex-
hibit one of the highest h values (11%) of DSSCs.[12,13] Diau
and Yeh et al. reported that YD14 (h=6.8%) with two di-
The IPCE is divided into three components in the follow-
ing equation: IPCE=LHEꢂfinj ꢂhcol, in which LHE (light-
harvesting efficiency) is the number of absorbed photons
per the number of incident photons and hcol is the charge
collection efficiency. The relative integrated IPCE value in-
ACHTUNGTRENNUNGarylACHTUNGTRENNUNGamino groups at the meso-positions with trans-configura-
tion (Figure 1) is comparable to YD2 (h=7.1%) in terms of
photovoltaic performances under their optimized condi-
tions,[30] supporting the advantage of the highly asymmetrical
Chem. Eur. J. 2013, 19, 17075 – 17081
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
17079

V. Sundstrçm, H. Imahori et al.
creases in the order TiO2/ZnPBA (1)<TiO2/YD2 (1.11)<
TiO2/ZnPBAT (1.28), which parallels the trend of the JSC
values (Table 1). The maximum adsorbed photon-to-current
efficiency (APCE) value at 400–500 nm is close to a unity
and increases slightly in the order TiO2/YD2 (88%)<TiO2/
ZnPBA (92%)<TiO2/ZnPBAT (95%). This implies that
the finj and/or hcol are somewhat different for the three
ZnPs, resulting in a switched order between TiO2/YD2 and
TiO2/ZnPBA for the APCE values.
Finally, the long-term stability of the TiO2/ZnPBAT cell
was assessed under the standard AM 1.5 condition (Fig-
ure S13 in the Supporting Information). Continued exposure
of the TiO2/ZnPBAT cell for 300 h to the full sunlight at
258C showed only a 10% decrease of the overall efficiency
over this extended light-soaking period. Note that the initial
increase in the h value is related to the aging effect on the
photovoltaic performance, as shown in Figure 5. Overall, the
cell is robust and that the small decline is probably caused
by losses of the volatile acetonitrile solvent and desorption
of ZnPBAT from the TiO2 electrode.
Electron injection and charge recombination (CR) of the
ZnP-sensitized TiO2 films were examined by time-resolved
transient absorption (TA) spectroscopy, similarly to what
was performed previously for other ZnPs on TiO2 (Fig-
ure S10 in the Supporting Information).[31–33] TA spectra
were measured in the wavelength range 475 to 750 nm for
the dyes in ethanol (Figure S10A–S10C in the Supporting
Information), as well as attached on the nanostructured
TiO2 electrodes (Figure S10D–S10F in the Supporting Infor-
mation). From these spectra, kinetics can be obtained at key
wavelengths to illustrate the dynamics of excited state
decay, electron injection and CR. Thus, kinetics at about
Conclusion
We have successfully designed and synthesized highly asym-
metric porphyrin ZnPBAT with enhanced push–pull charac-
ter for DSSCs for the first time. ZnPBAT improved the
light-harvesting property compared with references in visi-
ble region. More importantly, the ZnPBAT-sensitized solar
cell exhibited more than h=10%, which exceeds that of the
widely known YD2-sensitized solar cell under our optimized
conditions. The molecular design concept based on the
push–pull enhancement by the asymmetric substitution will
be useful for further improving cell performance in DSSCs.
1
700 nm probes both decay of ZnP* stimulated emission and
+
C
ZnP formation and decay, and can therefore be used to
monitor both electron injection and CR (Figure S10E and
S10F in the Supporting Information). As discussed in more
detail in Supporting Information, electron injection is very
fast, occurring on the few ps timescale and competing with
relaxation from higher excited states and solvation dynam-
ics. CR, on the other hand, is very slow, @100 ns for all the
sensitizers (Figure S10G in the Supporting Information).
Compared to previously studied ZnPs on TiO2,[31–33] electron
injection is faster and CR much slower, conditions ideal for
high IPCE values. The similar electron injection and CR for
all three ZnPs in addition rationalizes the small difference
in APCEs.
To understand the slight difference in the VOC values, we
also applied electrical impedance spectroscopy (EIS) to
compare the electron flow in the solar cells. The EIS Ny-
quist plots for DSSCs based on ZnPBAT, YD2, and ZnPBA
were obtained under standard AM 1.5 illumination at open-
circuit conditions (Figure S11 in the Supporting Informa-
tion).[34,35] The small semicircle at the left-hand side, large
semicircle in the middle, and the small semicircle at the
right-hand side of Figure S11 correspond to the ET process-
es at the Pt electrode, at the TiO2-dye-electrolyte interface
(Rp), and in the electrolyte, respectively. A small Rp should
reflect faster ET between the TiO2 and the electrolyte, con-
tributing to a decrease in the VOC values. In fact, the trend
in the VOC values is largely consistent with that in the Rp
values [ZnPBAT (15.28 W)<ZnPBA (18.41 W)<YD2
(20.06 W)]. We also measured the current–voltage character-
istics under dark conditions (Figure S12 in the Supporting
Information). As a more positive voltage is applied, the
onset appears in the order ZnPBAT<ZnPBA<YD2, which
is in good agreement with the trend of the Rp values. Over-
all, the difference in the JSC and VOC values of the three
TiO2 cells matches that in the h values.
Experimental Section
Photovoltaic measurements: Preparation of TiO2 electrodes and the fab-
rication of sealed cells for photovoltaic measurements were performed
following the previously reported method.[28] Nanocrystalline TiO2 parti-
cles (d=20 nm, CCIC:PST18NR, JGC-CCIC, and d=30 nm,
CCIC:PST30NRD, JGC-CCIC) were used as the transparent layer of the
photoanode, whereas submicrocrystalline TiO2 particles (d=400 nm,
CCIC:PST400C, JGC-CCIC) were used as the light-scattering layers of
the photoanode. A TiO2 film with a three-layer structure composed of
PST18NR (8 mm thickness), PST30NRD (4 mm thickness), and PST400C
(4 mm thickness) was fabricated on a FTO glass (Solar 4 mm thickness,
10 W&À1, Nippon Sheet Glass). These thickness conditions were initially
optimized by using YD2. The TiO2 electrode was immersed into a solu-
tion of the porphyrin in ethanol (0.20mm; 10mL) containing CDCA (0~
7.85mg). The sandwich cell was prepared by using the dye-anchored
TiO2 film as a working electrode and a counter Pt electrode. An electro-
lyte solution consisting of 1,3-dimethylimidazolium iodide (1.0m), I2
(0.03m), LiI (0.05m), guanidinium thiocyanate (0.1m), and 4-tert-butylpyr-
idine (0.50m) in an 85:15 acetonitrile/valeronitrile mixture was used for
the cell.[12] Incident photon-to-current efficiency (IPCE) and photocur-
rent–voltage (J–V) performance were measured on an action spectrum
measurement setup (CEP-2000RR, BUNKOUKEIKI) and a solar simu-
lator (PEC-L10, Peccell Technologies) with
a
simulated sunlight of
(Winꢂl),
AM 1.5 (100 mW cmÀ2), respectively: IPCE (%)=100ꢂ1240ꢂi/
ACHTUNGTRENNUNG
in which i is the photocurrent density (AcmÀ2), Win is the incident light
intensity (WcmÀ2), and l is the excitation wavelength (nm). During the
photovoltaic measurements, a black plastic mask was attached on the
back of the TiO2 electrode except for the TiO2 film region to reduce scat-
tering light.
17080
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 17075 – 17081

Highly Asymmetrical Porphyrins
FULL PAPER
[18] J. R. Stromberg, A. Marton, H. L. Kee, C. Kirmaier, J. R. Diers, C.
Muthiah, M. Taniguchi, J. S. Lindsey, D. F. Bocian, G. J. Meyer, D.
Holten, J. Phys. Chem. C 2007, 111, 15464–15478.
[19] C. Y. Lee, J. T. Hupp, Langmuir 2010, 26, 3760–3765.
[20] M. Taniguchi, A. Balakumar, D. Fan, B. E. McDowell, J. S. Lindsey,
J. Porphyrins Phthalocyanines 2005, 9, 554–574.
Acknowledgements
This work was financially supported by Advanced Low Carbon Technolo-
gy Research and Development Program (ALCA, JST) and WPI Initia-
tive (MEXT, Japan).
[21] L. Yu, K. Muthukumaran, I. V. Sazanovich, C. Kirmaier, E. Hindin,
J. R. Diers, P. D. Boyle, D. F. Bocian, D. Holten, J. S. Lindsey, Inorg.
Chem. 2003, 42, 6629–6647.
[22] R. W. Boyle, C. K. Johnson, D. Dolphin, J. Chem. Soc. Chem.
Commun. 1995, 527–528.
[23] Y. Chen, X. P. Zhang, J. Org. Chem. 2003, 68, 4432–4448.
[24] C. Monnereau, E. Blart, V. Montembault, L. Fontaine, F. Odobel,
Tetrahedron 2005, 61, 10113–10125.
[25] S. Taniguchi, H. Hasegawa, S. Yanagiya, Y. Tabeta, Y. Nakano, M.
Takahashi, Tetrahedron 2001, 57, 2103–2108.
[26] D.-M. Shen, C. Liu, X.-G. Chen, Q.-Y. Chen, J. Org. Chem. 2009, 74,
206–211.
[27] C.-P. Hsieh, H.-P. Lu, C.-L. Chiu, C.-W. Lee, S.-H. Chuang, C.-L.
Mai, W.-N. Yen, S.-J. Hsu, E. W.-G. Diau, C.-Y. Yeh, J. Mater. Chem.
2010, 20, 1127–1134.
[1] M. Grꢀtzel, Acc. Chem. Res. 2009, 42, 1788–1798.
[2] F. Gao, Y. Wang, D. Shi, J. Zhang, M. Wang, X. Jing, R. Humphry-
Baker, P. Wang, S. M. Zakeeruddin, M. Grꢀtzel, J. Am. Chem. Soc.
2008, 130, 10720–10728.
[3] C. Y. Chen, M. Wang, J. Y. Li, N. Pootrakulchote, L. Alibabaei, C.
Ngoc-le, J. D. Decoppet, J. H. Tsai, C. Grꢀtzel, C. G. Wu, S. M. Za-
keeruddin, M. Grꢀtzel, ACS Nano 2009, 3, 3103–3109.
[4] L. Han, A. Islam, H. Chen, C. Malapaka, B. Chiranjeevi, S. Zhang,
X. Yang, M. Yanagida, Energy Environ. Sci. 2012, 5, 6057–6060.
[5] A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo, H. Pettersson, Chem.
Rev. 2010, 110, 6595–6663.
[6] W. Zeng, Y. Cao, Y. Bai, Y. Wang, Y. Shi, M. Zhang, F. Wang, C.
Pan, P. Wang, Chem. Mater. 2010, 22, 1915–1925.
[7] H. Imahori, T. Umeyama, S. Ito, Acc. Chem. Res. 2009, 42, 1809–
1818.
[8] M. V. Martꢃnez-Dꢃaz, G. de la Torre, T. Torres, Chem. Commun.
2010, 46, 7090–7108.
[9] H. Imahori, T. Umeyama, K. Kurotobi, Y. Takano, Chem. Commun.
2012, 48, 4032–4045.
[10] M. J. Griffith, K. Sunahara, P. Wagner, K. Wagner, G. Wallace, D. L.
Officer, A. Furube, R. Katoh, S. Mori, A. J. Mozer, Chem.
Commun. 2012, 48, 4145–4162.
[11] L.-L. Li, E. W.-G. Diau, Chem. Soc. Rev. 2013, 42, 291–304.
[12] T. Bessho, S. M. Zakeeruddin, C.-Y. Yeh, E. W.-G. Diau, M. Grꢀtzel,
Angew. Chem. 2010, 122, 6796–6799; Angew. Chem. Int. Ed. 2010,
49, 6646–6649.
[28] S. Ito, T. Murakami, P. Comte, P. Liska, C. Grꢀtzel, M. K. Nazeerud-
din, M. Grꢀtzel, Thin Solid Films 2008, 516, 4613–4619.
[29] K. Wagner, M. J. Griffith, M. James, A. J. Mozer, P. Wagner, G.
Triani, D. L. Officer, G. G. Wallace, J. Phys. Chem. C 2011, 115,
317–326.
[30] S.-L. Wu, H.-P. Lu, H.-T. Yu, S.-H. Chuang, C.-L. Chiu, C.-W. Lee,
E. W.-G. Diau, C.-Y. Yeh, Energy Environ. Sci. 2010, 3, 949–955.
[31] H. Imahori, S. Kang, H. Hayashi, M. Haruta, H. Kurata, S. Isoda,
S. E. Canton, Y. Infahsaeng, A. Kathiaravan, T. Pascher, P. Chꢄbera,
A. P. Yartsev, V. Sundstrçm, J. Phys. Chem. A 2011, 115, 3679–3690.
[32] S. Ye, A. Kathiravan, H. Hayashi, Y. Tong, Y. Infahsaeng, P.
Chꢄbera, T. Pascher, A. P. Yartsev, S. Isoda, H. Imahori, V. Sund-
strçm, J. Phys. Chem. C 2013, 117, 6066–6080.
[13] A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Na-
zeeruddin, E. W.-G. Diau, C.-Y. Yeh, S. M. Zakeeruddin, M. Grꢀtzel,
Science 2011, 334, 629–634.
[14] H. Imahori, Y. Matsubara, H. Iijima, T. Umeyama, Y. Matano, S.
Ito, M. Niemi, N. V. Tkachenko, H. Lemmetyinen, J. Phys. Chem. C
2010, 114, 10656–10665.
[33] H. Nemec, E. Galoppini, H. Imahori, V. Sundstrçm, Comprehensive
Nanoscience and Technology, Vol 2. (Eds.: D. Andrews, G. Scholes,
G. Wiederrecht) 2011, Elsevier, Amsterdam, pp.325–359.
[34] X.-F. Wang, H. Tamiaki, L. Wang, N. Tamai, O. Kitao, H. Zhou, S.-i.
Sasaki, Langmuir 2010, 26, 6320–6327.
[35] Impedance Spectroscopy—Theory, Experiment, and Applications
(Eds.: E. Barsoukov, J. R. Macdonald), Wiley, Hoboken, New
Jersey, 2005.
[15] D. P. Hagberg, T. Edvinsson, T. Marinado, G. Boschloo, A. Hagfeldt,
L. Sun, Chem. Commun. 2006, 2245–2247.
[16] J. S. Lindsey, Acc. Chem. Res. 2010, 43, 300–311.
[17] M. O. Senge, Chem. Commun. 2011, 47, 1943–1960.
Received: September 3, 2013
Published online: November 13, 2013
Chem. Eur. J. 2013, 19, 17075 – 17081
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
17081