Perylene anhydride fused porphyrins as near-infrared sensitizers for dye-sensitized solar cells

Article abstract of DOI:10.1021/ol201303h

Two perylene anhydride fused porphyrins 1 and 2 have been synthesized and employed successfully in dye-sensitized solar cells (DSCs). Both compounds showed broad incident monochromatic photon-to-current conversion efficiency spectra covering the entire vi

Full text of DOI:10.1021/ol201303h

ORGANIC
LETTERS
2011
Vol. 13, No. 14
3652–3655
Perylene Anhydride Fused Porphyrins
as Near-Infrared Sensitizers for
Dye-Sensitized Solar Cells
Chongjun Jiao,† Ningning Zu,‡ Kuo-Wei Huang,§ Peng Wang,*,‡ and Jishan Wu*,†,
Department of Chemistry, National University of Singapore, 3 Science Drive 3,
Singapore, 117543, State Key Laboratory of Polymer Physics and Chemistry,
Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun,
130022, P. R. China, KAUST Catalysis Center and Division of Chemical and Life
Sciences and Engineering, 4700 King Abdullah University of Science and Technology,
Thuwal 23955-6900, Kingdom of Saudi Arabia, and Institute of Materials Research and
Engineering, A*Star, 3 Research Link, Singapore, 117602
peng.wang@ciac.jl.cn; chmwuj@nus.edu.sg
Received May 14, 2011
ABSTRACT
Two perylene anhydride fused porphyrins 1 and 2 have been synthesized and employed successfully in dye-sensitized solar cells (DSCs). Both
compounds showed broad incident monochromatic photon-to-current conversion efficiency spectra covering the entire visible spectral region
and even extending into the near-infrared (NIR) region up to 1000 nm, which is impressive for ruthenium-free dyes in DSCs.
Recent developments in the field of organic solar cells
have boosted interest in development of novel functional
near-infrared (NIR) dyes,1 given that sunlight possesses
50% of its irradiation energy in the infrared spectral
region. Dye-sensitized solar cells (DSCs)2 stand out to
challenge the traditional silicon-based solar cells due to
the cost-effectiveness issue combined with their medium
efficiencies. A rising interest is to use ruthenium-free dyes
as sensitizers for DSCs,3,4 because such dyes hold many
advantages over ruthenium-based dyes such as lower cost,
higher absorption coefficient, and ease of tuning LUMO
and HOMO energy levels. However, how to obtain a single
sensitizer that is capable of effectively capturing light over
the entire spectrum from 400 to 1000 nm and simulta-
neously fulfills all the requirements for an efficient device
remains a long-term challenge.5
† National University of Singapore.
‡ Changchun Institute of Applied Chemistry.
§ King Abdullah University of Science and Technology.
Institute of Materials Research and Engineering.
(1) For selected reviews, see: (a) Fabian, J.; Nakanzumi, H.;
Matsuoka, M. Chem. Rev. 1992, 92, 1197–1226. (b) Qian, G.; Wang,
Z. Chem. Asian J. 2010, 5, 1006–1029. (c) Jiao, C.; Wu, J. Curr. Org.
Chem. 2010, 14, 2145–2168.
€
(2) (a) O’Regan, B.; Gratzel, M. Nature 1991, 353, 737–740.
(4) For refernces concerning high-efficiency DSCs, see: (a) Zeng, W.;
Cao, Y.; Bai, Y.; Wang, Y.; Shi, Y.; Zhang, M.; Wang, F.; Pan, C.;
Wang, P. Chem. Mater. 2010, 22, 1915–1925. (b) Bessho, T.;
Zakeeruddin, S. M.; Yeh, C.-Y.; Diau, E. W.-G.; Gratzel, M. Angew.
Chem., Int. Ed. 2010, 49, 6646–6649.
(5) Yum, J. H.; Baranoff, E.; Wenger, S.; Nazeeruddin, M. K.;
€
(b) Hagfeldt, A.; Gratzel, M. Chem. Rev. 1995, 95, 49–68.
(3) For recent reviews, see: (a) Mishra, A.; Fischer, M. K. R.; Bauerle,
P. Angew. Chem., Int. Ed. 2009, 48, 2474–2499. (b) Li, C.; Liu, M.;
Pschirer, N. G.; Baumgaten, M.; Mullen, K. Chem. Rev. 2010, 110,
6817–6855. (c) Hagfeldt, A.; Boschloo, G.; Sun, L.; Kloo, L.; Pettersson,
€
€
€
€
Gratzel, M. Energy Environ. Sci. 2011, 4, 842–857.
H. Chem. Rev. 2010, 110, 6595–6663.
r
10.1021/ol201303h
Published on Web 06/23/2011
2011 American Chemical Society

Porphyrin has gained recognition as one of the most
versatile building blocks for DSCs due to their unique
optical as well as electrochemical properties and their
excellent photochemical stability.6 Due to the limited
absorption behavior of the porphyrin monomer in the
visible and particularly in the NIR region, efforts have
been made toward the design and synthesis of π-extended
porphyrins, including porphyrin tapes7 and polycyclic
aromatic compounds-fused porphyrins,8 which were pre-
dicted to have potential applications in photovoltaic
devices.8e,h,i Attempts to use these fused dyes in DSCs
were independently made by Yeh and Imahori’s
groups,9,10 and up to 4.1% efficiency was achievable under
optimized conditions. However, DSCs based on fused
porphyrin systems exhibiting a NIR response beyond
900 nm have never been reported. Herein we report two
novel perylene anhydride fused porphyrin dyes 1 and 2
(Figure 1), which have been employed successfully as NIR
sensitizers for DSCs.
Scheme 1
fused porphyrins exhibit intensified NIR absorptions,
large dipole moments, and high photostability as reported
in our recent studies,8h thus providing an extraordinary
molecular platform for applications in DSCs. Neverthe-
less, appropriate structural modifications are necessary for
such a purpose. (2) Anhydride serves as an anchoring
group,11 which can be obtained by saponification of an
imide group in the presence of a strong base. In addition,
the electron-withdrawing anhydride moiety is able to
stabilize the highly conjugated low band gap π-system,8h
leading to stable NIR dyes for practical DSCs application.
(3) In the design of the large-sized, rigid molecules 1 and 2,
bulky 3,5-di-tert-butylphenyl groups were chosen as sub-
stituents because such bulky groups not only surmount the
solubilityproblem but alsoeliminate the aggregation of the
chromophores.8eꢀg (4) Introduction of an electron-donat-
ing 4-(dimethylamino)-phenylethynylene onto the meso-
position of porphyrin in 2 is supposed to result in a more
red-shifted NIR absorption due to the enhanced intramo-
lecular charge transfer, and this is also beneficial to a fast
electron injection from the excited dye to the conduction
band of TiO2 in DSCs.3c
Figure 1. Molecular design and structures of dyes 1 and 2.
As shown in Figure 1, the molecular design is based on
the following considerations: (1) Perylene monoimide-
(6) For selected reviews, see: (a) Imahori, H.; Umeyama, T.; Ito, S.
´ ´
Acc. Chem. Res. 2009, 42, 1809–1818. (b) Martınez-Dıaz, M. V.; de la
Torre, G.; Torres, T. Chem. Commun. 2010, 46, 7090–7108. (c)
Radivojevic, I.; Varotto, A.; Farleya, C.; Drain, C. M. Energy Environ.
Sci. 2011, 3, 1897–1909.
(7) (a) Tsuda, A.; Osuka, A. Science 2001, 293, 79–82. (b) Ikda, T.;
Aratani, N.; Osuka, A. Chem. Asian J. 2009, 4, 1248–1256.
(8) For selected references, see: (a) Gill, H. S.; Marmjanz, M.;
Santamarıa, J.; Finger, I.; Scott, M. J. Angew. Chem., Int. Ed. 2004,
´
43, 485–490. (b) Yamane, O.; Sugiura, K.; Miyasaka, H.; Nakamura, K.;
Fujimoto, T.; Nakamura, K.; Kaneda, T.; Sakata, Y.; Yamashita, M.
Chem. Lett. 2004, 33, 40–41. (c) Kurotobi, K.; Kim, K. S.; Noh, S. B.;
Kim, D.; Osuka, A. Angew. Chem., Int. Ed. 2006, 45, 3944–3947. (d)
Tokuji, S.; Takahashi, Y.; Shinmori, H.; Shinokubo, H.; Osuka, A.
Chem. Commun. 2009, 1028–1030. (e) Davis, N. K. S.; Thompson, A. L.;
Anderson, H. L. Org. Lett. 2010, 12, 2124–2127. (f) Diev, V. V.; Hanson,
K.; Zimmerman, J. D.; Forrest, S. R.; Thompson, M. E. Angew. Chem.,
Int. Ed. 2010, 49, 5523–5526. (g) Jiao, C.; Huang, K.-W.; Guan, Z.; Xu,
Q.-H.; Wu, J. Org. Lett. 2010, 12, 4046–4049. (h) Jiao, C.; Huang, K.-
W.; Chi, C.; Wu, J. J. Org. Chem. 2011, 76, 661–664. (i) Davis, N. K. S.;
Thompson, A. L.; Anderson, H. L. J. Am. Chem. Soc. 2011, 133, 30–31.
(j) Jiao, C.; Zhu, L.; Wu, J. Chem.;Eur. J. 2011, 17, 6610–6614.
(9) Mai, C.-L.; Huang, W.-K.; Lu, H.-P.; Lee, C.-W.; Chiu, C.-L.;
Liang, Y.-R; Diau, E. W.-G.; Yeh, C.-Y. Chem. Commun. 2010, 46, 809–
811.
€
(11) (a) Edvinsson, T.; Li, C.; Pschirer, N.; Schoneboom, J.;
Eickemeyer, F.; Sens, R.; Boschloo, G.; Herrmann, A.; Mullen, K.;
€
Hagfeldt, A. J. Phys. Chem. C 2007, 111, 15137–15140. (b) Li, C.; Yum,
J.-H.; Moon, S.-J.; Herrmann, A.; Eickemeyer, F.; Pschirer, N. G.; Erk,
€
€ €
P.; Schoneboom, J.; Mullen, K.; Gratzel, M.; Nazeeruddin, M. K.
(10) (a) Tanaka, M.; Hayashi, S.; Eu, S.; Umeyama, T.; Matano, Y.;
Imahori, H. Chem. Commun. 2007, 2069–2071. (b) Hayashi, S.; Tanaka,
M.; Hayashi, H.; Eu, S.; Umeyama, T.; Matano, Y.; Araki, Y.; Imahori,
H. J. Phys. Chem. C 2008, 112, 15576–15585.
€
ChemSusChem. 2008, 1, 615–618. (c) Li, C.; Liu, Z.; Schoneboom, J.;
€
Eickemeyer, F.; Pschirer, N. G.; Erk, P.; Herrmann, A.; Mullen, K.
J. Mater. Chem. 2009, 19, 5405–5415.
Org. Lett., Vol. 13, No. 14, 2011
3653

Scheme 1 outlines the synthetic route for compounds 1
and 2 (see Supporting Information for more details). The
synthesis commenced with the preparation of the precur-
sor 5, which was synthesized by Suzuki coupling of por-
phyrin boronic ester 312 and perylene monoimide 413
followed by demetalation and Ni(II)-metalation. Intramo-
lecular cyclization reaction of 5 promoted by FeCl3 then
generated a purple solid as a mixture containing unsepar-
able chlorinated side products, which was directly con-
verted to their corresponding anhydrides without
purification. Separation of 1 from other side products
was successful after saponification of 6 with an overall
yield of 63% for three steps based on 5. Preparation of
electron-donating 4-(dimethylamino)phenylethynylene
substituted 2 followed a slightly revised procedure. The
key intermediate 10 was first prepared by sequential
Miyaura reaction, Suzuki coupling, bromination, and
trans-metalation. In common with the synthesis of 6,
FeCl3-promoted oxidative dehydrogenation of 10 afforded
fused 11 together with some chlorinated side products. The
presence of bromine atom in 11 allowed further functiona-
lization, and the incorporation of an electron-donating
4-(dimethylamino)phenylethynylene group was carried
out at this stage by SonogashiraꢀHagihara coupling reac-
tion to generate “push-pull” type molecule 13, which finally
underwent saponification to give target molecule 2 with an
overall yield of 17% for four steps based on 10.
charge transfer from porphyrin to perylene anhydride
can also decrease the fluorescence quantum yield.
Electrochemical properties of 1 and 2 were investigated
by cyclic voltammetry in deoxygenated DCM solution
containing 0.1 M tetra-n-butylammonium hexafluoropho-
sphate as supporting electrolyte (Table S1 and Figure S1 in
Supporting Information). Compound 1 exhibited four
reversible oxidation waves with half-wave potentials at
0.29, 0.45, 0.73, and 0.83 V (versus Fcþ/Fc), whereas three
reversible or quasireversible oxidative waves were ob-
served with half-wave potentials potentials at 0.15, 0.25,
and 1.02 V for 2; these are distinguishable by differential-
pulse voltammetry (DPV). Furthermore, two reversible
reduction waves were found for both 1 and 2 with half-
wave potentials at ꢀ0.94 and ꢀ1.04 V for 1 and at ꢀ0.94
and ꢀ1.05 V for 2, indicating that they can be reduced into
the corresponding radical anions and dianions, which can
also be stabilized by the delocalized π-system. HOMO
energy levels of ꢀ5.04 and ꢀ4.89 eV and LUMO energy
levels of ꢀ3.92 and ꢀ3.93 eV were estimated for 1 and 2,
respectively.14 Energy gaps of 1.12 eV for 1 and 0.94 eV for
2 were then calculated, and this trend is in agreement with
their optical band gaps (1.42 eV for 1 and 1.31 eV for 2).
To gain better insight into the molecular geometries,
molecular orbitals, and UVꢀvisꢀNIR absorption spectra
of 1 and 2, time-dependent density function theory
(TDDFT at B3LYP/6-31G**) calculations were per-
formed, and the optimized molecular structures and fron-
tier molecular orbital profiles are shown in Figure 3. The
HOMO for both 1 and 2 are delocalized throughout the
The absorption spectra of both 1 and 2 cover the entire
visible and a part of the NIR spectral region (Figure 2).
The absorption maximum of 1 was found at 805 nm
Figure 3. Frontier molecular orbital profiles of molecules 1 and
2 calculated by DFT.
Figure 2. UVꢀvisꢀNIR absorption spectra of 1 and 2 in CHCl3
(8.0 ꢁ 10ꢀ6 M).
highly conjugated backbone as well as the terminal amino
group. In contrast, the LUMO for 1 and 2 are predomi-
nately located at the perylene anhydride motif. Such
orbital partitions are undoubtedly beneficial for electron
injection and overall photovoltaic performance. The cal-
culations also predict that 1 and 2 will show absorption
maximum at 759 and 849 nm (Figures S2, S3 and Tables
(ε = 89,000 Mꢀ1 cmꢀ1), and a further red-shift of 42 nm
was observed for 2 (ε = 81,000 Mꢀ1 cmꢀ1) owing to
enhanced intramolecular charge transfer. Nearly no fluor-
escence was observed for 1 and 2 in chloroform, which is
common for Ni(II)-porphyrins. In addition, other factors
such as aggregation and photoinduced intramolecular
(12) (a) Wang, Q. M.; Bruce, D. W. Synlett 1995, 1267–1268. (b)
Nakamura, Y.; Jang, S. Y.; Tanaka, T.; Aratani, N; Lim, J. M.; Kim,
K. S.; Kim, D.; Osuka, A. Chem.;Eur. J. 2008, 14, 8279–8289.
(13) Baffreau, J.; Ordronneau, L.; Leroy-Lhez, S.; Hudhomme, P.
J. Org. Chem. 2008, 73, 6142–6147.
(14) The HOMO and LUMO energy levels were calculated from the
onset of the first oxidation and reduction waves according to the
following equations: HOMO = ꢀ(4.8 þ Eoxonset) and LUMO =
ꢀ(4.8 þ Eredonset).
3654
Org. Lett., Vol. 13, No. 14, 2011

slightly increases the Jsc but has no positive effect on Voc
and FF. The moderate efficiencies of NIR dyes 1 and 2
perhaps are attributed to the following reasons: (1) The
LUMO energy levels of 1 and 2 are very close to the
conduction band edge of TiO2, thus leading to insufficient
driving force for electron injection.9 (2) Depite the intro-
duction of bulky groups, dyes 1 and 2 still have a strong
tendency to form aggregates, reflected by their broad 1H
NMR spectra recorded at room temperature. It is com-
monly believed that aggregation of dyes reults in non-
radiative decay of dyes from the excited state to the ground
state. Excitons, therefore, are readily trapped or quenched
byaggregate states, which eventuallyreducesphotocurrent
densities.3 (3) The short lifetime of the excited states of the
dyes possibly decreases the device performance, because
short lifetime narrows the time-scale window when elec-
tron injection takes place.15
In summary, a new strategy for designing NIR dyes for
DSCs by fusing an electron-rich porphyrin unit to the
perylene anhydride core was demonstrated. Based on dyes
1 and 2, broader IPCE spectra have been achieved, cover-
ing the entire visible range extending into the NIR region
up to 1000 nm. Although these NIR dyes displayed
moderate efficiencies, the enhanced light harvesting prop-
erties in the NIR region provide possibilities for further
improvement of overall DSC efficiency of ruthenium-free
sensitizers with panchromatic and particularly NIR re-
sponse. Further structural optimization, such as reducing
the aggregation and tuning the energy levels, is very likely
to generate more efficient sensitizers and this work is
currently underway in our laboratories.
Figure 4. (A) IPCE spectra of the devices sensitized by 1 and 2.
(B) JꢀV curves of DSCs based on 1 and 2 at an irradiation of 100
mW cmꢀ2 AM 1.5 G sunlight.16
S2ꢀS5 in Supporting Information), respectively, which are
in good agreement with the experimental data.
The DSC devices applying 1 and 2 as NIR sensitizers
were constructed and characterized (see Supporting In-
formation for experimental details). As shown in Figure
4A, the preliminary investigation of the incident mono-
chromatic photon-to-current conversion efficiency (IPCE)
revealed that IPCE spectra reached the highest values of
31.4% at 772 nm for 1 and 29.7% at 790 nm for 2,
respectively. In common with their absorption spectra,
IPCE curves for both 1 and 2 not only exhibit panchro-
matic responses but also extend the photovoltaic perfor-
mance to the NIR region. Noteworthy is that dye 2
displays an IPCE response as long as 1000 nm, which is
one of the furthest red-shifted responses reported for
ruthenium-free DSCs. The characteristic current densi-
tyꢀvoltage (JꢀV) curve measured under irradiation of
AM 1.5 G full sunlight (Figure 4B) provides a short-circuit
current density (Jsc) of 7.11 mA/cm2, an open-circuit
voltage (Voc) of 329 mV, and a fill factor (FF) of 0.54,
yielding the overall power conversion efficiency (η) of
1.26% for a DSC device based on 1. For comparison,
photovoltaic parameters (Jsc, Voc, FF, and η) of 2 are 7.66
mA/cm2, 333 mV, 0.52, and 1.36%, respectively. The
results demonstrate that the attachment of electron-donat-
ing amino groups onto the π-conjugation system for 2
Acknowledgment. J.W. acknowledges financial support
from Singapore DSTA DIRP Project (DSTA-NUS-
DIRP/2008/03), NRF Competitive Research Program
(R-143-000-360-281), NUS Young Investigator Award
(R-143-000-356-101), A*Star BMRC-NMRC joint grant
(no. 10/1/21/19/642), and IMRE Core Funding (IMRE/
10-1P0509). K.-W.H. acknowledges the financial support
from KAUST.
Supporting Information Available. Experimental de-
tails, characterization data of all new compounds, theo-
retical calculation data, CV data, and the detailed device
fabrication as well as the photovoltaic characterizations.
This material is available free of charge via the Internet at
http://pubs.acs.org.
€
(15) (a) Cherepy, N. J.; Smestad, G. P.; Gratzel, M.; Zhang, J. Z.
J. Phys. Chem. B 1997, 101, 9342–9351. (b) Kira, A; Matsubara, Y.;
Iijima, H.; Umeyama, T.; Matano, Y.; Ito, S.; Niemi, M.; Tkachenko,
N. V.; Lemmetyinen, H.; Imahori, H. J. Phys. Chem. C 2010, 114,
11293–11304.
(16) The spectral distribution of our light resource simulates AM 1.5
G solar emission with a mismatch less than 5%.
Org. Lett., Vol. 13, No. 14, 2011
3655