Modulating dye E(S+/S*) with efficient heterocyclic nitrogen containing acceptors for DSCs

Article abstract of DOI:10.1039/c2cc17142b

Acceptor motifs based on nitrogen containing heterocycles have been synthesized for use in dye-sensitized solar cells (DSCs). Through the selective addition of nitrogen atoms and increased conjugation of the nitrogen containing heterocycles the excited-state oxidation potential, E_(S+/S*), may be conveniently tuned with minimal effect on the ground-state oxidation potential, E_(S+/S), of the dye. The Royal Society of Chemistry.

Full text of DOI:10.1039/c2cc17142b

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 2295–2297
www.rsc.org/chemcomm
COMMUNICATION
Modulating dye E(S+/S*) with efficient heterocyclic nitrogen containing
acceptors for DSCsw
¨
Jared H. Delcamp,* Aswani Yella, Mohammad K. Nazeeruddin and Michael Gratzel
Received 17th November 2011, Accepted 9th January 2012
DOI: 10.1039/c2cc17142b
Acceptor motifs based on nitrogen containing heterocycles have been
synthesized for use in dye-sensitized solar cells (DSCs). Through the
selective addition of nitrogen atoms and increased conjugation
of the nitrogen containing heterocycles the excited-state oxidation
potential, E(S+/S*), may be conveniently tuned with minimal effect on
the ground-state oxidation potential, E(S+/S), of the dye.
Scheme 1 Heterocyclic nitrogen containing acceptor based D-p-A
dyes for DSCs.
Dye sensitized solar cells (DSCs) have undergone continuous
improvement since their introduction in 1991,1 which is culminating
with DSCs being a strong contributor toward the solar energy
conversion market.2 DSCs currently exhibit solar-to-electric power
conversion efficiencies (PCEs) ranging from 9–12%.2,3 Among
DSCs, metal-free organic sensitizers are attractive because they
do not require the use of expensive metals and demonstrate strong
extinction coefficients.4 Donor-p-acceptor (D-p-A) dyes are proving
to be a good, strategic dye design scaffolding for metal-free dyes
and this class of dyes is at the forefront of organic sensitizers
displaying PCEs reproducibly near5 and greater than 10%.6
Currently, cyanoacrylate is commonly employed as the acceptor
and anchor for D-p-A dyes presumably due to an increase in
spectral response through intramolecular charge transfer (ICT) and
good electron injection properties.7 We envisioned a new class of
carboxylate anchored heterocyclic nitrogen containing acceptors
which may increase spectral response through both ICT and
lengthening p-conjugation while maintaining good charge
injection.8 In this manuscript we examine the effect of the
number of nitrogen atoms in heterocyclic aryl acceptors with
one or two fused rings for metal-free dyes.
1 (TPA)10 and stannylated CPDT 211 fragments (Scheme S1, ESIw).
The TPA-CPDT molecule, 3, was stannylated and Stille
coupled to the desired brominated heterocyclic nitrogen
containing acceptor ester. The final D-p-A dyes were synthesized
after THF/LiOH hydrolysis to generate the carboxylic acid
anchor. JD8 was synthesized in a similar manner; however, a
brominated quinoxaline benzyl alcohol was coupled in place of
the corresponding ester (Scheme S2, ESIw). The alcohol was then
subsequently oxidized to the desired JD8 dye.
The optical and electrochemical properties of the completed
D-p-A dyes were examined to ensure good device compatibility
ꢀ
with TiO2 semiconductor and Iꢀ/I3 redox couple. In order to
estimate the excited-state oxidation potential (E(S+/S*)) according
to the equation E(S+/S*) = E(S+/S) ꢀ Eogpt, both the ground-state
oxidation potential E(S+/S) and the E(S+/S*) ꢀ E(S+/S) gap (Egopt
)
were measured. Eogpt was determined from the absorption
spectrum at lonset in CH2Cl2 (Fig. S1, ESIw).12 The Egopt values
are shrinking in heterocyclic nitrogen containing series
from JD2 through JD8. Pyridine dye, JD2, with the largest
Eogpt (2.27 eV) has a lmax of 461 nm (Table 1, Fig. S1, ESIw).
The selected DSC target dyes keep the donor and p-bridge
regions consistent throughout the series examined and are based on
a triarylamine donor and the good p-bridge, di-n-hexyl substituted
cyclopentyldithiophene (CPDT).4,5c,9 A range of electron deficient
heterocyclic nitrogen containing acceptor (pyridine, quinoline,
pyrazine, and quinoxaline) based dyes were synthesized
(Scheme 1). The synthetic route to dyes JD2, JD4, JD7, and
JD8 begins with Stille coupling of the brominated triphenylamine
Table 1 Optical and electrochemical data
Dye
lmaxa/nm
Eogpt b/eV
E(S+/S)c/V
E(S+/S*)d/V
JD2
JD4
JD7
JD8
C218e
461
507
508
579
550
2.27
2.11
2.05
1.79
1.95
0.81
0.82
0.82
0.79
0.89
ꢀ1.46
ꢀ1.29
ꢀ1.23
ꢀ1.00
ꢀ1.06
a
Measured in CH2Cl2. b Calculated from UV-vis spectrum at lonset
.
Laboratory for Photonics and Interfaces, Institution of Chemical
Sciences and Engineering, School of Basic Sciences, Swiss Federal
Institute of Technology, CH-1015 Lausanne, Switzerland.
E-mail: jared.delcamp@epfl.ch
w Electronic supplementary information (ESI) available: Synthetic
routes, intermediate characterization, and additional device studies.
See DOI: 10.1039/c2cc17142b
c Measured in a 0.1 M Bu4NPF6 in CH2Cl2 solution with glassy carbon
working electrode, Pt reference electrode, and Pt counter electrode with
ferrocene as an internal standard. Values are reported versus NHE.
d Calculated from E(S+/S*) = E(S+/S) ꢀ Egopt
.
e C218 was synthesized
through a modified synthetic route compared to the original report (see ESI).
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 2295–2297 2295

View Online
Both the quinoline acceptor (JD4) and pyrazine acceptor
(JD7) are stronger electron withdrawing groups than pyridine
which results in bathochromic shifts at a lmax of 46 nm and
47 nm, respectively. The quinoxaline acceptor (JD8) is the
most red shifted with a lmax of 579 nm and strong absorbance
into the near-infrared (NIR) with a value of 710 nm at 5% the
lmax height. All of the dyes reported have significant absorbance
(>20% lmax peak height) from 400 nm to the lmax value of the
dye. The absorption onset at low energy for JD4, JD7 and JD8 is
at wavelengths greater than 600 nm. The heterocyclic nitrogen
containing acceptors display a strong effect on Eogpt ranging from
2.27 eV to 1.79 eV.
Cyclic voltammetry was performed in CH2Cl2 with ferrocene
as an internal standard (Table 1). The oxidation potentials,
E(S+/S), of all the dyes synthesized were measured to be within
30 mV and are reported versus normal hydrogen electrode
(NHE). The close E(S+/S) values indicate that E(S+/S) is only
slightly affected based on the heterocyclic nitrogen containing
structure.13 No reversible reduction potentials were observed
during cyclic voltammetry experiments (Table S1, ESIw). The
E(S+/S*) values were calculated from the measured values for
E(S+/S) and Egopt. A large effect is seen for the range of E(S+/S*)
values when compared to E(S+/S) (460 mV vs. 30 mV). The
heterocyclic acceptors modulate E(S+/S*) with little effect on
E(S+/S) presumably due to the added electron density through
increased conjugation introduced with each of the heterocycles.
The increase in p-conjugation negates any stabilization to E(S+/S)
which is commonly observed with the addition of an electron
acceptor. Effectively, these heterocyclic nitrogen containing
acceptors provide a simple, efficient way to tune the E(S+/S*) of
Fig. 1 IPCE spectrum of JD2, JD4, JD7 and JD8.
prepared with each dye and TiO2 semiconductor as described
in the ESI.w IPCEs for the series are shown in Fig. 1 with
optimized electrolyte additives for each device. Of the four dyes,
JD7 was found to be the most efficient over its absorption
window with an IPCE near 90% from 440–550 nm. The
remaining dyes required the addition of chenodeoxycholic acid
(CDCA) for optimal IPCEs performance. The loss in IPCEs
without CDCA is due to dye aggregation which is known to be
disrupted by CDCA (Fig. S2, ESIw). Dyes JD2, JD4 and JD8
also exhibit lower dye loadings leading to less TiO2 surface
coverage when compared to JD7 which favors diminished
IPCEs (Table S2, ESIw). A strikingly pronounced effect on the
IPCE was observed with JD2. This dye shows a strong IPCE
maximum (85%) when the dye is co-adsorbed with CDCA;
however, without CDCA a weak IPCE spectrum was obtained
with a maximum of o50% (Fig. S2, ESIw). Similar behaviors
were noted for the IPCE spectrum of dyes JD4 and JD8. These
two dyes generally suffer from lower dye loading possibly due to
the increased steric hindrance near the carboxylate anchor.
The photovoltaic parameters of the DSCs prepared with the
dyes JD2, JD4, JD7 and JD8 under standard AM 1.5
illumination, 100 mW cmꢀ2, are summarized in Table 2. The
best power conversion efficiency (PCE) of 6.55% was obtained
with dye JD7 as derived from the equation Z = JscVocFF/I(sun).
An optimal JD7-based cell shows a short-circuit photocurrent
density (Jsc) of 11.5 mA cmꢀ2, an open-circuit voltage (Voc)
of 743 mV, and a 0.76 FF resulting in a Z of 6.55% without
any co-adsorbent in the dye solution (Fig. 2). A 40 mV loss
D-p-A dyes with marginal influence on E(S+/S)
.
Due to the donor and p-bridge regions remaining the same
with only the acceptor groups changing for dyes JD2, JD4,
JD7, JD8 and C128, E(S+/S*) values are used to determine
acceptor strength (Table 1). E(S+/S*) becomes progressively less
negative as the electron withdrawing strength of the acceptor
group is increased.14 The cyanoacrylate electron withdrawing
group of C218 is less withdrawn than the quinoxaline acceptor
and the series progresses as follows for increasing electron
withdrawing strength: JD2 o JD4 o JD7 o C218 o JD8.
In order to efficiently inject electrons into the TiO2 conduction
band, the excited-state oxidation potential of the sensitizer should
be more negative than ꢀ0.50 V versus normal hydrogen electrode
(NHE). The dyes reported herein are all thermodynamically
suitable for good electron injection with E(S+/S*) ranging from
ꢀ1.46 V to ꢀ1.00 V (Table 1). For efficient dye regeneration
E(S+/S) should be more positive than +0.35 V based on the
potential of Iꢀ/I3ꢀ. With E(S+/S) ranging from 0.79 to 0.82 eV all
Table 2 Photovoltaic parameters measured under AM 1.5. CDCA
is chenodeoxycholic acid. JD2 optimal electrolyte 1.0 M DMII
(1,3-dimethylimidazolium iodide), 0.03 M iodine, 0.025 M LiI, 0.5
TBP, 0.1 M guanidinium thiocyanide, acetonitrile solvent. JD4 same
as JD2 except 0.6 M DMII and 0.05 M LiI. JD7 same as JD4 except
no LiI was added. JD8 same as JD2 except 0.05 M LiI added
ꢀ
dyes are capable of being regenerated from the Iꢀ/I3 redox
couple. Therefore the heterocyclic nitrogen containing dyes
studied are all suitable for TiO2/iodine based DSCs.
Dye
Additive
Jsc/mA cmꢀ2
Voc/mV
FF
Z (%)
A range of electron withdrawing strengths from efficient
acceptors for D-p-A dyes are desirable for tuning E(S+/S*)
according to the demands from the rest of the dye. The
heterocyclic nitrogen containing acceptors offer a range of
electron withdrawing strengths; however, for widespread use
they must demonstrate efficient incident photon-to-current
conversion efficiencies (IPCEs) over the absorbance range of
the dye. Liquid electrolyte based photovoltaic devices were
JD2
JD2
JD4
JD4
JD7
JD7
JD8
JD8
None
CDCA
None
CDCA
None
CDCA
Lil
5.62
7.87
7.80
7.74
11.5
11.1
11.8
9.89
665
691
680
693
743
703
607
696
0.71
0.75
0.71
0.75
0.76
0.74
0.73
0.77
2.63
4.05
3.77
4.02
6.55
5.78
5.20
5.25
CDCA
c
2296 Chem. Commun., 2012, 48, 2295–2297
This journal is The Royal Society of Chemistry 2012

View Online
heterocyclic nitrogen containing acceptor for metal-free D-p-A
DSC dyes. The acceptors presented in this work offer a range of
withdrawing strengths for maximizing color and injection efficiency
based on E(S+/S*). These efficient acceptors include two hetero-
cycles, pyridine and pyrazine, with less withdrawing strength than
cyanoacrylate and one heterocycle, quinoxaline, with more with-
drawing strength. The carboxylate anchored pyrazine acceptor
used in JD7 and the pyridine acceptor used in dye JD2 demon-
strate excellent charge separation and efficient charge injection.
These acceptors will likely find future application for dye design
involving p-bridge chromophores where E(S+/S*) is already too low
for the addition of a strongly withdrawn cyanoacrylate. In this
case, both pyridine and pyrazine acceptors would provide a dipole
for photon induced charge separation and a better driving force for
injection by generating a smaller shift with respect to the E(S+/S*)
level than cyanoacrylate. Furthermore, dye JD7 is a promising
dye for co-sensitized DSC devices due to a strong IPCE from
400–550 nm (near 90%) and no need for the addition of
CDCA to diminish aggregation.
Fig. 2 Current versus potential plot for dyes JD2, JD4, JD7 and JD8.
in Voc is observed for devices with JD7 and CDCA. All the
devices tested (with or without CDCA) give substantially
lower voltage than JD7 which can in part be attributed to
the difference in dye loading since higher loadings aid in
blocking potential recombination sites with the electrolyte and
semiconductor surface (see ESIw). Comparatively, JD2 has
significantly less absorbance in the visible region which leads
to a lower Jsc of 7.87 mA cmꢀ2 and an optimized PCE of 4.05%
in the presence of CDCA (Table 2). JD2 is the most strongly
affected dye with respect to the benefits of CDCA and displays
an increase in both current and voltage when compared to
solutions with no co-adsorbent. JD8 shows an increase in
voltage with the addition of CDCA and a loss of current.
JD8 suffers from a decreased driving force for injection, lower
dye loading and aggregation (see ESIw). In all cases except
devices with JD7, lack of CDCA addition to the dye solutions
led to diminished photovoltaic performance. Dye aggregation
likely causes intermolecular energy transfer and subsequently
results in excited-state quenching of the dyes.
We would like to acknowledge ERC-ARG and the
FP7-Energy-2010-FET project Molesol (Contract No. 256617)
for financial support.
Notes and references
1 B. O’Regan and M. Gratzel, Nature, 1991, 353, 737.
¨
2 (a) A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo and H. Pettersson,
Chem. Rev., 2010, 110, 6595; (b) Q. Schiermeier, J. Tollefson, T. Scully,
A. Witze and O. Morton, Nature, 2008, 454, 816; (c) D. Butler, Nature,
2008, 454, 558; (d) L. M. Goncalves, V. Z. Bermudez, H. A. Ribeiro
and A. M. Mendes, Energy Environ. Sci., 2008, 1, 655.
3 (a) T. Bessho, S. M. Zakeeruddin, C.-Y. Yeh, E. W.-G. Diau and
M. Gratzel, Angew. Chem., Int. Ed., 2010, 49, 6646; (b) Y. Cao,
¨
Y. Bai, Q. Yu, Y. Cheng, S. Liu, D. Shi, F. Gao and P. Wang,
J. Phys. Chem. C, 2009, 113, 6290; (c) S. Ito, H. Miura, S. Uchida,
chy and
M. Takata, K. Sumioka, P. Liska, P. Comte, P. Pe
´
M. Gratzel, Chem. Commun., 2008, 5194.
¨
4 A. Mishra, M. K. R. Fischer and P. Ba
¨
Ed., 2009, 48, 2474.
uerle, Angew. Chem., Int.
5 (a) Y. Bai, J. Zhang, D. Zhou, Y. Wang, M. Zhang and P. Wang,
J. Am. Chem. Soc., 2011, 133, 11442; (b) H. N. Tsao, C. Yi,
T. Moehl, J.-H. Yum, S. M. Zakeeruddin, M. K. Nazeeruddin and
To further understand the role of CDCA as a co-adsorbent for
these new acceptor based dyes, photovoltage decay transients
were studied at various white light bias intensities (Fig. S3, ESIw).
The charge recombination rates are plotted versus the open circuit
potential (Voc) of the device adjusted by varying the white light
bias. The recombination rate is known to increase exponentially
with increasing white light bias due to a filling of intraband trap
states in the TiO2, which allows a faster detrapping of electrons to
the conduction band and subsequent recombination with the
electrolyte. A large decrease in the recombination rates for devices
made with CDCA and JD2 (2-fold decrease) and JD8 (16-fold
decrease) was measured. This increase in lifetime is clearly
represented by the gain in Voc for both JD2 (26 mV) and JD8
(89 mV) when CDCA is included in the electrolyte. No obvious
change in the recombination rate was observed for JD4 with the
addition of CDCA. Interestingly, in the case of JD7, addition of
CDCA enhances charge recombination, which is also visible as
the loss of the open circuit potential resulting in a lower PCE.
Additionally, the higher charge recombination rate reduces the
charge collection efficiency and consequently the Jsc and IPCE.
Several new nitrogen acceptor-based dyes have been synthesized
and characterized for use in liquid electrolyte DSC devices. E(S+/S*)
can be modulated by 460 mV through proper selection of a
M. Gratzel, ChemSusChem, 2011, 4, 591; (c) R. Li, J. Liu, N. Cai,
¨
M. Zhang and P. Wang, J. Phys. Chem. B, 2010, 114, 4461.
6 W. Zeng, Y. Cao, Y. Bai, Y. Wang, Y. Shi, M. Zhang, F. Wang,
C. Pan and P. Wang, Chem. Mater., 2010, 22, 1915.
7 J. Wiberg, T. Marinado, D. P. Hagberg, L. Sun, A. Hagfeldt and
B. Albinsson, J. Phys. Chem. C, 2009, 113, 3881.
8 For pyridine as an acceptor and anchor see: Y. Ooyama, S. Inoue,
T. Nagano, K. Kushimoto, J. Ohshita, I. Imae, K. Komaguchi and
Y. Harima, Angew. Chem., Int. Ed., 2011, 50, 7429. For nitrogen
heterocycles as efficient p-spacers see: L. Y. Lin, C.-H. Tsai,
K.-T. Wong, T.-W. Huang, C.-C. Wu, S.-H. Chou, F. Lin,
S.-H. Chen and A.-I. Tsai, J. Mater. Chem., 2011, 21, 5950.
9 B. P. Karsten, J. C. Bijleveld, L. Viani, J. Cornil, R. A. Gierschner
and J. Janssen, J. Mater. Chem., 2009, 19, 5343.
10 G. Qian, B. Dai and M. Luo, Chem. Mater., 2008, 20, 6208.
11 4H-cyclopenta[2,1-b:3,4-b0]dithiophene is commercially available
from Astar Pharmaceuticals.
12 Although E(0-0) is desirable for determining E(S+/S*), Eogpt is used for
consistency due to the lack of fluorescence emission from dye JD8.
Emission spectra and data for the remaining dyes are reported in
Fig. S1 (ESIw) and Table S1 (ESIw) along with E(0-0) values.
13 (a) Assuming E1/2([FeCp2]+/[FeCp2]) = +0.70 V vs. NHE based on
values of +0.24 V for SCE vs. NHE; (b) N. G. Connelly and
W. E. Geiger, Chem. Rev., 1996, 96, 877; (c) Handbook of Physics and
Chemistry, ed. R. C. Weast, CRC, Boca Raton, FL, 63rd edn, 1982.
14 C218 was synthesized through a modified route compared to the
reported procedure in ref. 5c (see ESIw).
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 2295–2297 2297