Benzoporphyrins: Selective Co-sensitization in Dye-Sensitized Solar Cells

Article abstract of DOI:10.1002/chem.201600802

A novel class of dyes, namely benzoporphyrins, was synthesized and implemented into dye-sensitized solar cells. They feature complementary absorptions compared to N719, which renders them promising candidates for co-sensitization in DSSCs. Notably, metallated benzoporphyrins reveal a TiO₂-nanoparticle attachment that is size and aggregation dependent. Therefore, unproductive energy-transfer events between the selectively attached dyes can be prevented. In light of the latter, an efficiency improvement of 39 % has been achieved upon selective adsorption of benzoporphyrins and N719 onto different layers of TiO₂ photoelectrode.

Full text of DOI:10.1002/chem.201600802

DOI: 10.1002/chem.201600802
Full Paper
&
Photophysics
Benzoporphyrins: Selective Co-sensitization in Dye-Sensitized
Solar Cells
Fabian Lodermeyer, RubØn D. Costa,* Jenny Malig, Norbert Jux, and Dirk M. Guldi*[a]
Abstract: A novel class of dyes, namely benzoporphyrins,
was synthesized and implemented into dye-sensitized solar
cells. They feature complementary absorptions compared to
N719, which renders them promising candidates for co-sen-
sitization in DSSCs. Notably, metallated benzoporphyrins
reveal a TiO2–nanoparticle attachment that is size and aggre-
gation dependent. Therefore, unproductive energy-transfer
events between the selectively attached dyes can be pre-
vented. In light of the latter, an efficiency improvement of
39% has been achieved upon selective adsorption of benzo-
porphyrins and N719 onto different layers of TiO2 photoelec-
trode.
Introduction
region. To circumvent this limitation, the co-sensitization with
another dye, whose absorption spectrum complements that of
N719, has emerged as a powerful strategy to enhance the
light-harvesting efficiency (LHE) of the photoelectrode and, in
turn, the overall device efficiency.[10]
Incentives, such as low material costs and easy-to-fabricate
processes, render dye-sensitized solar cells (DSSCs) a promising
and competitive solar energy conversion technology.[1] After
several decades of optimization, state-of-the-art efficiencies
range from 10 to 13%.[2–4]
The major asset of co-sensitization is higher short-circuit
current densities (Jsc). The different approaches towards co-sen-
sitization include dye cocktail,[10,11] stepwise co-sensitiza-
tion,[12,13] selective co-sensitization,[6,14,15] and the use of a thin
top Al2O3 layer,[16] which is selectively sensitized with a second
dye.
Nevertheless, DSSCs bear several drawbacks, among which
electron-hole pair recombinations stand out. Here, the electron
back transfer either to the electrolyte or to the oxidized dye,
reduces the open-circuit voltage (Voc).[5] Several strategies to
suppress such loss processes have been established. One of
them is the use of dyes featuring bulky groups that protect
the interface between the photoanode and the oxidized form
of the electrolyte.[6] Alternatively, the addition of chenodeoxy-
cholic acid either to the dye solution or to the electrolyte has
been shown to prevent fast back electron transfer, which, for
example, dominates in ferrocene-based electrolytes.[7]
The benefits that stem from co-sensitization are offset by
the challenges to integrate two or more different dyes into the
mesoporous photoelectrode. Hardin et al. reported that in
most cases co-sensitization results in lower efficiencies.[17] The
latter relates to decreased injection efficiencies, owing to inter-
molecular interactions between co-adsorbed dyes.[18] Therefore,
suppression of electronic interactions between dyes adsorbed
onto the photoelectrode is crucial to afford efficient DSSCs.
Here, the spatial separation throughout the semiconductor
network was one of the first challenges dealt with in the field
of co-sensitization.
Another aspect of significant importance in DSSCs is the ab-
sorption cross-section of the dye. The latter should match as
much as possible the solar spectrum. To this end, the dye
should reveal high molar extinction coefficients throughout
the visible region and absorptions that extend into the near-in-
frared region. Considering, for example, the standard cis-diiso-
thiocyanato-bis(2,2’-bipyridyl-4,4’-dicarboxylato)ruthenium(II)
bis(tetrabutylammonium) complex (N719), maximum efficien-
cies of around 11% have been realized.[8,9] Nevertheless, the
use of N719 is restricted by its lack of absorption in the far-red
In this regard, only a few examples are known from the liter-
ature to date. In chronologically order, Lee and Park et al. dem-
onstrated an elegant technique to place several dyes onto
a single TiO2 layer by a concept that mimics the basics of
column chromatography.[19] Although an improvement of
around 28% was achieved featuring top efficiencies of 4.8%,
this selective co-sensitization involved up to eight steps that
render it rather time consuming. It is, nevertheless, very flexi-
ble and, indeed, it has been followed by other groups showing
similar results.[6,20]
[a] F. Lodermeyer, Dr. R. D. Costa, Dr. J. Malig, Prof. Dr. N. Jux,
Prof. Dr. D. M. Guldi
Department of Chemistry and Pharmacy and Interdisciplinary Center for
Molecular Materials (ICMM), Friedrich-Alexander-Universität
Erlangen-Nürnberg, Egerlandstrasse 3, Erlangen 91058 (Germany)
E-mail: ruben.costa@fau.de
Soon after this contribution, an impressive improvement in
the overall efficiency by means of a film transfer technique
was reported by Miao et al.[21] Hence, an overall efficiency im-
provement of 37.6% was achieved, although this technique
bears the risk of spatially intermittent layers.
dirk.guldi@fau.de
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201600802.
Chem. Eur. J. 2016, 22, 7851 – 7855
7851
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Taking all of the aforementioned into account, the grand
challenge is to explore novel approaches that facilitate the se-
lective but simple implementation of different dyes into spa-
tially separated parts of the photoelectrode without compro-
mising the integrity of the corresponding photoelectrodes. To-
wards this aim, Fan et al. have specifically designed dyes with
and without bulky substituents to gain control over the ad-
sorption kinetics and to enhance the efficiency by 18%.[14] This
approach is, indeed, very powerful, owing to the myriad of
possibilities to properly modify dyes.
H2cTPBP after a dehydrogenation procedure (Schemes S1 and
S2 in the Supporting Information). The zinc complex ZncTPBP
was obtained by a simple metalation reaction.
Figure 1 shows the absorption spectra of H2cTPBP and
ZncTPBP in ethanol. Here, the benefit of the p-extended
framework is clearly documented in terms of their absorption
features in comparison to standard porphyrins such as
5,10,15,20-tetraphenylporphyrin (TPP) or zinc(II)-5,10,15,20-tet-
raphenylporphyrin (ZnTPP; Figure 1 and Figure S1 in the Sup-
porting Information). In particular, a 40 nm bathochromic shift
is noted for the Soret band of H2cTPBP and ZncTPBP, which
appear at 460 and 462 nm, respectively. Likewise, the Q bands
are red-shifted and evolve between 600 and 750 nm. As
a matter of fact, both exhibit complementary absorption spec-
tra to that of N719 (Figure 1). Next, the singlet excited state
features of H2cTPBP and ZncTPBP were investigated by means
of steady-state fluorescence spectroscopy. As shown in Fig-
ure S2 in the Supporting Information, upon exciting H2cTPBP
at 460 nm (i.e., the Soret band), 699 and 780 nm fluorescence
bands evolved. As expected, the fluorescence of ZncTPBP is
blue shifted with respect to that of H2cTPBP, showing fluores-
cence maxima at 660 and 721 nm (Figure S2). For H2cTPBP
and ZncTPBP, the fluorescence quantum yields are around
0.029. Additionally, their singlet excited-state lifetimes were de-
termined by means of time-correlated single photon counting
(TCSPC; Figure S3 in the Supporting Information). In dilute eth-
anol solutions of H2cTPBP and ZncTPBP (2”10À6 m), which
were utilized to avoid aggregation and/or inner filter effects,
lifetimes of 2.98 and 0.37 ns were determined, respectively.
The short ZncTPBP lifetime relates to heavy atom effect, which
enhances intersystem crossing from the singlet to the triplet
excited states.[24]
In the current work, the major thrust is to combine the com-
plementary absorption features of two carboxy-substituted tet-
raphenylbenzoporphyrins (H2cTPBP and ZncTPBP) relative to
N719 and their selective adsorption behavior towards a novel
co-sensitization concept in DSSCs. Two key aspects should be
highlighted. Firstly, to the best of our knowledge, TPBP have
not yet been explored as sensitizers in DSSCs, although they
complementary absorb relative to N719 (Figure 1). Secondly,
zinc-containing TPBP (ZncTPBP) shows an unforeseen adsorp-
tion behavior, which depends on the size of the TiO2 nanopar-
ticles and the nature of the solvent. Notably, the last feature is
a breakthrough towards straightforward selective co-sensitiza-
tion in a simple device architecture, in which ZncTPBP covers
only the transparent part of the photoelectrode, while the
light-scattering part is only covered by N719. As a matter of
fact, the success of our co-sensitization approach is document-
ed by a 39% enhancement of the overall device efficiency
when compared to a non-co-sensitized device with N719.
To implement both TPBPs into TiO2-based DSSCs, we utilized
a standard double-layer architecture that consists of a transpar-
ent layer (TL TiO2) at the bottom and a light-scattering layer
(LS TiO2) on the top. The latter was stepwise firstly sensitized
with both TPBPs using different soaking times and secondly
with N719 for 18 h. Remarkably, ZncTPBP attached only to TL
TiO2 leaving LS TiO2 for adsorption of N719 (Figure 2), while
H2cTPBP attached equally to TL TiO2 and LS TiO2. Hence, the
selective adsorption behavior of ZncTPBP enables straightfor-
ward co-sensitization by sequential dipping of photoelectrodes
into different dye solutions. The integrity of the TiO2 layer is
not compromised, which circumvents drawbacks related to
spatially intermittent electrical contacts between different
layers of the electrode. As a matter of fact, our finding might
contribute to tackle the challenge in co-sensitization, if the
device performance gives rise to improvement.
Figure 1. Steady-state absorption spectra of H2cTPBP (red spectrum) and
ZncTPBP (black spectrum) in ethanol as well as N719 (grey spectrum) in
ethanol/acetonitrile (1:1 v/v). Inset: molecular structure of H2cTPBP for
M=H2 and ZncTPBP for M=Zn.
To corroborate our hypothesis, we firstly focused on docu-
menting a complete TL TiO2 coverage with ZncTPBP by means
of adsorption kinetic assays (Figure S4 in the Supporting Infor-
mation). Saturation of the photoelectrode was noted after
12 h. From the latter we conclude that TL TiO2 will not at all or
only negligibly be covered with N719 after more than 12 h ad-
sorption times. Hence, it is safe to compare the stepwise co-
sensitized TL TiO2/LS TiO2 photoanodes with a LS TiO2 N719
photoanode as reference.
Results and Discussion
Details regarding the synthesis of asymmetrical carboxy-substi-
tuted cTPBP are provided in the Supporting Information. In
brief, applying literature-known protocols, a tetrahydroisoindole
precursor was prepared,[22] which was reacted in a mixed-
aldehyde approach[23] to yield the desired benzoporphyrin
Chem. Eur. J. 2016, 22, 7851 – 7855
www.chemeurj.org
7852
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 2. Upper part: schematic representation of different device architec-
tures (left: TL+LS TiO2, middle: LS+TL TiO2, right: MS+LS TiO2). Lower part:
the corresponding photographs of the different TiO2 electrode configura-
tions after having been immersed into a ZnTPBP solution for 12 h are
shown.
Figure 3 depicts the corresponding J–V curves and the inci-
dent photon-to-electron conversion (IPCE) spectra of the co-
sensitized and the non-co-sensitized N719 reference DSSCs.
The latter features Voc of 645 mV, Jsc of 7.20 mAcmÀ2, and FF of
0.72, which result in an overall efficiency of 3.46% (Table 1).
Co-sensitized DSSCs exhibit very different trends. For instance,
although H2cTPBP/N719 DSSCs reveal a significant contribu-
tion of the Soret and Q band absorptions in the IPCE spectrum,
Jsc values are lower than in the non-co-sensitized N719 refer-
ence DSSCs (Figure 3). This is further corroborated in the IPCE
spectrum by a maximum value of only 31.7% at the Soret
band maximum at 470 nm. In addition, time-dependent ad-
sorption assays show that the efficiencies decrease upon
longer adsorption times with H2cTPBP (Figure 3). This trend is
caused by diminished Voc, FF, and, especially, Jsc, which is as-
cribed to unproductive energy transfer events, due to the non-
selective adsorption of H2cTPBP and N719, vide supra. As
a consequence, the most efficient solar cells are found upon
the shortest H2cTPBP adsorption time of 1 h, which resulted in
a Voc of 567 mV, Jsc of 1.92 mAcmÀ2, FF of 0.72, and, in turn, an
efficiency of 0.78%. In stark contrast, the results from the co-
sensitized devices with ZncTPBP/N719 show a significant en-
hancement compared to non-co-sensitized reference devices.
As a matter of fact, the efficiencies increase upon longer ad-
sorption times peaking at 12 h (Figure 3). Here, the device fea-
tures (Voc of 732 mV, Jsc of 9.56 mAcmÀ2, FF of 0.69, and effi-
ciencies of 4.83%) outperform the non-co-sensitized N719 ref-
erence devices. Under optimized ZncTPBP adsorption times,
that is, 12 h, the overall efficiency increased by 39% when
compared to the reference DSSCs. A better light-harvesting, es-
pecially around 470 and 650 nm, which stems from the Soret
and Q band absorptions, respectively, is noted in the IPCE
spectrum (Figure 3). In fact, after 12 h adsorption time the TL
TiO2 layer is fully covered with ZncTPBP, avoiding any unpro-
ductive energy transfers, vide supra. Longer ZncTPBP adsorp-
tion times fail to result in higher currents and, in turn, in
higher efficiencies (Figure 3). This is likely to relate to ZncTPBP
aggregation at the surface, hampering the dye regeneration. It
is interesting to note that after 12 h adsorption of ZncTPBP,
the non-co-sensitized N719 reference and the ZncTPBP/N719
co-sensitized devices exhibit identical contributions in the
parts of the IPCE spectra, where N719 absorbs. In a stark con-
Figure 3. Upper part: J–V curves of best performing reference N719 (grey),
H2cTPBP/N719 (red), and ZncTPBP/N719 (black) DSSCs. Middle part: corre-
sponding IPCE spectra of the best performing N719 (grey), H2cTPBP/N719
(red), and ZncTPBP/N719 (black) DSSCs. Lower part: efficiency of H2cTPBP/
N719 (red) and ZncTPBP/N719 (black) DSSCs versus adsorption time of
TPBP. The efficiency of N719 reference DSSCs is given by the dashed line.
trast, co-sensitized DSSCs built with H2cTPBP/N719 give rise to
decreased contributions of N719 due to the competitive ad-
sorption of H2cTPBP and N719.
A reasonable rationale for the distinct adsorption behavior
could imply the different aggregation behaviors of both ben-
zoporphyrins when dissolved in ethanol. In particular, H2cTPBP
does not exhibit any aggregation in solution even at high con-
centrations of 1”10À4 m, since the full width at half maximum
(FWHM) of the Soret band exhibits a constant value of around
13.5 nm (Figure S5 in the Supporting Information). In stark con-
trast, pronounced aggregation of ZncTPBP in ethanol at con-
Chem. Eur. J. 2016, 22, 7851 – 7855
www.chemeurj.org
7853
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(top layer) but not to LS TiO2 (bottom layer) as shown in
Figure 2. We exclude, at this point, a device architecture de-
pendent selective adsorption behavior. Notably, the device per-
formances of inverted layer DSSCs are comparable to those of
standard layer architecture DSSCs (Table 1).
Table 1. Device parameters obtained under AM 1.5 conditions and 1 sun
illumination for the best performing devices.
Compound
Adsorption time of Voc
TPBP+N719 [h]
Jsc
FF
h
[mV] [mAcmÀ2
]
[%]
H2cTPBP+N719[a]
1+18
567
732
737
645
1.92
9.56
10.17
7.20
0.72 0.78
0.69 4.83
0.63 4.74
0.72 3.46
ZncTPBP+N719[a] 12+18
N719+ZncTPBP[b] 18+12
Conclusions
N719[c]
18
In summary, we have successfully implemented for the first
time a novel class of dyes, namely TPBPs, into TiO2-based
DSSCs. Our work underpins, firstly, the complementary absorp-
tion features of TPBPs with respect to N719 and, secondly, the
selective adsorption of ZncTPBP into specific layers of the pho-
toelectrodes in ethanol solutions. This phenomenon depends,
on one hand, on the TiO2 nanoparticle size but not on the
layer architecture, and, on the other hand, on the different ag-
gregation behavior in solution. In fact, we took this finding fur-
ther and fabricated co-sensitized DSSCs by integrating
ZncTPBP and N719 into spatially separated regions of the
photoanodes in a two-step procedure. By means of eliminating
energy wastage and unproductive energy-transfer processes
between the two dyes we realized 39% more efficient DSSCs
relative to a non-co-sensitized N719 reference device. It is im-
portant to point out that the rationale design of ZncTPBPs in
terms of linkers, bulky groups, etc., is a valuable strategy to
reach new heights in overall DSSC efficiencies.
[a] Regular electrodes: a 7 mm TL TiO2 layer and 7 mm LS TiO2 layer. [b] In-
verted electrodes: a 7 mm LS TiO2 layer and 7 mm TL TiO2 layer. [c] Refer-
ence DSSCs with 7 mm LS TiO2 electrodes.
centrations higher than 5”10À7 m is observed. Here, the FWHM
decreases from 9.5 to 7 nm, which implies a reduced broad-
ness of the Soret band and, as a consequence, decreased ag-
gregation. Based on these results, we deduce that adsorption
from an aggregated solution onto LS TiO2 layers is effectively
hampered. Strikingly, LS TiO2 photoanodes remain even after
72 h of adsorption from ethanol solutions of ZncTPBP uncol-
ored. In stark contrast, ZncTPBP lacks aggregation in THF as
the FWHMs taken at the high and low wavelength side of the
Soret band is more or less with 10.5 nm (Figure S5). The lower
tendency towards aggregation is explained by the coordinat-
ing character of THF as it binds to the metal center of the ben-
zoporphyrins. As a consequence, the different aggregation be-
havior of ZncTPBP in THF compared to ethanol is likely to sup-
press the dye adsorption onto TiO2 films even after 72 h or
more of adsorption times (Figure S6 in the Supporting Infor-
mation). As such it is safe to conclude that selective adsorption
onto TiO2 films is limited to aggregates as they are present in
solutions.
Experimental Section
Experimental details are provided in the Supporting Information.
To further provide a mature study of the unique adsorption
behavior of ZncTPBP, two additional experiments were per-
formed. On one hand, another light-scattering layer containing
medium-size TiO2 (MS TiO2) was utilized. This was meant to elu-
cidate the adsorption selectivity of ZncTPBP in dependence on
the TiO2 size. The average crystallite sizes of TiO2 in the films
were investigated by XRD measurements using the Scherrer
equation (Figure S7 in the Supporting Information). The parti-
cle sizes of TL, MS, and LS TiO2 were 12.49, 20.81, and
34.25 nm, respectively. Diffuse reflectance measurements fur-
ther corroborate the increased particle size, since the light-
scattering features follow the later trend (Figure S5).[25] Owing
to the fact that ZncTPBP also immobilizes onto MS TiO2
(Figure 2), the adsorption selectivity is only seen for LS TiO2. In
other words, the size of the TiO2 nanoparticles matters in
terms of ZncTPBP adsorption. H2cTPBP attaches to all of the
aforementioned TiO2 layers, a finding that further corroborates
our assumption.
Acknowledgements
F.L., D.M.G. and R.D.C. acknowledge the EAM cluster in the
frame of the DFG excellence programs for their support.
D.M.G. acknowledges the DFG, ICMM, and the ZMP for finan-
cial and intellectual support. N.J. and J.M. acknowledge sup-
port by SFB953 “Synthetic Carbon Allotropes”. J.M. also was
supported by the Graduate School Molecular Science (GSMS).
Keywords: benzoporphyrins · donor–acceptor systems · dye-
sensitized solar cells
sensitization
· electron transfer · selective co-
[1] B. O’Regan, M. Grätzel, Nature 1991, 353, 737–740.
[2] A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin,
E. W.-G. Diau, C.-Y. Yeh, S. M. Zakeeruddin, M. Grätzel, Science 2011, 334,
629–634.
[3] A. El-Shafei, M. Hussain, A. Atiq, A. Islam, L. Han, J. Mater. Chem. 2012,
22, 24048–24056.
On the other hand, an inverted device, in which the LS layer
is at the bottom of the FTO and the TL layer is placed on top
(FTO/(LS+TL TiO2)), was assembled. This assisted in ruling out
that the layer distribution is decisive for the adsorption phe-
nomenon. With the LS layer at the bottom of the FTO, it is ex-
posed to the same conditions as the TL layer in the former ex-
periments, vide supra. Again, ZncTPBP attaches only to TL TiO2
[4] R. D. Costa, F. Lodermeyer, R. Casillas, D. M. Guldi, Energy Environ. Sci.
2014, 7, 1281–1296.
[5] B. A. Gregg, F. Pichot, S. Ferrere, C. L. Fields, J. Phys. Chem. A 2001, 105,
1422–1429.
[6] N. C. Jeong, H.-J. Son, C. Prasittichai, C. Y. Lee, R. A. Jensen, O. K. Farha,
J. T. Hupp, J. Am. Chem. Soc. 2012, 134, 19820–19827.
[7] T. Daeneke, A. J. Mozer, T.-H. Kwon, N. W. Duffy, A. B. Holmes, U. Bach, L.
Spiccia, Energy Environ. Sci. 2012, 5, 7090–7099.
Chem. Eur. J. 2016, 22, 7851 – 7855
www.chemeurj.org
7854
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[8] Y. Chiba, A. Islam, Y. Watanabe, R. Komiya, N. Koide, L. Han, Jpn. J. Appl.
Phys. 2006, 45, L638–L640.
[9] R. D. Costa, S. Feihl, A. Kahnt, S. Gambhir, D. L. Officer, G. G. Wallace,
M. I. Lucio, M. A. Herrero, E. Vµzquez, Z. Syrgiannis, M. Prato, D. M.
Guldi, Adv. Mater. 2013, 25, 6513–6518.
[17] B. Hardin, A. Sellinger, J. Am. Chem. Soc. 2011, 133, 10662–10667.
[18] B. E. Hardin, H. J. Snaith, M. D. McGehee, Nat. Photonics 2012, 6, 162–
169.
[19] K. Lee, S. W. Park, M. J. Ko, K. Kim, N.-G. Park, Nat. Mater. 2009, 8, 665–
671.
[10] P. J. Holliman, M. L. Davies, A. Connell, B. Vaca Velasco, T. M. Watson,
Chem. Commun. 2010, 46, 7256–7258.
[20] S. W. Park, K. Lee, D.-K. Lee, M. J. Ko, N.-G. Park, K. Kim, Nanotechnology
2011, 22, 045201.
[11] J.-J. Cid, J.-H. Yum, S.-R. Jang, M. K. Nazeeruddin, E. Martínez-Ferrero, E.
Palomares, J. Ko, M. Grätzel, T. Torres, Angew. Chem. Int. Ed. 2007, 46,
8358–8362; Angew. Chem. 2007, 119, 8510–8514.
[12] C.-M. Lan, H.-P. Wu, T.-Y. Pan, C.-W. Chang, W.-S. Chao, C.-T. Chen, C.-L.
Wang, C.-Y. Lin, E. W.-G. Diau, Energy Environ. Sci. 2012, 5, 6460–6464.
[13] M. Kimura, H. Nomoto, N. Masaki, S. Mori, Angew. Chem. Int. Ed. 2012,
51, 4371–4374; Angew. Chem. 2012, 124, 4447–4450.
[14] S. Fan, C. Kim, B. Fang, K. Liao, J. Phys. Chem. A 2011, 115, 7747–7754.
[15] F. Huang, D. Chen, L. Cao, R. A. Caruso, Y.-B. Cheng, Energy Environ. Sci.
2011, 4, 2803–2806.
[21] Q. Miao, L. Wu, J. Cui, M. Huang, T. Ma, Adv. Mater. 2011, 23, 2764–
2768.
[22] P. Hopkins, P. Fuchs, J. Org. Chem. 1978, 43, 1208–1217.
[23] J. Lindsey, H. Hsu, I. Schreiman, Tetrahedron Lett. 1986, 27, 4969–4970.
[24] J. AndrØasson, G. Kodis, S. Lin, A. L. Moore, T. A. Moore, D. Gust, J. Mår-
tensson, B. Albinsson, Photochem. Photobiol. 2002, 76, 47–50.
[25] S. Hore, P. Nitz, C. Vetter, C. Prahl, M. Niggemann, R. Kern, Chem.
Commun. 2005, 2011–2013.
[16] H. Choi, S. Kim, S. O. Kang, J. Ko, M.-S. Kang, J. N. Clifford, A. Forneli, E.
Palomares, M. K. Nazeeruddin, M. Grätzel, Angew. Chem. Int. Ed. 2008,
47, 8259–8263; Angew. Chem. 2008, 120, 8383–8387.
Received: February 19, 2016
Published online on April 23, 2016
Chem. Eur. J. 2016, 22, 7851 – 7855
www.chemeurj.org
7855
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim