Fluorene functionalized porphyrins as broadband absorbers for TiO 2 nanocrystalline solar cells

Article abstract of DOI:10.1039/c4ta01954g

Three 9,9-dihexyl-9H-fluorene (DHF) functionalized zinc porphyrin dyes (coded as ZZX-N3, ZZX-N4, and ZZX-N5) were designed and synthesized for dye-sensitized solar cells. Then, DHF and benzoic acid were conjugated to the porphyrin ring through triple bonds to act as a spacer to elongate the π-conjugation and as an acceptor for an efficient electron injection, respectively. A bis(9,9-dihexyl-9H-fluorene-7-yl)amine (BFA) and a bis(4-hexylphenyl)amine (BPA) were further linked to DHF to act as electron donors in ZZX-N3 and ZZX-N4, respectively. ZZX-N5 did not have any electron donor and served as a reference. Moreover, ZZX-N3- and ZZX-N4-sensitized cells exhibited broader sunlight absorption than ZZX-N5, and as a result, higher photon-to-electricity efficiency (PCE) (ZZX-N3, 3.83%; ZZX-N4, 4.2%; ZZX-N5, 3.70%) was observed. The results are consistent with well-separated HOMO (highest occupied molecular orbital) and LUMO (lowest unoccupied molecular orbital) in ZZX-N3 and ZZX-N4 than in ZZX-N5. However, the overall conversion efficiency of ZZX-N3- and ZZX-N4-sensitized cells was low, which is due to significant dye aggregation induced by the extra long alkyl-chains on the donor groups. This was evidenced by blue and red shifts of the absorption spectra of dye-coated TiO₂ films. In addition, the extra long-chains also did not offer better shielding to prevent electron recombination of injected electrons with I₃^- in electrolyte as revealed by electrochemical impedance spectroscopy. When a co-sensitizer (coded as PBS) was used, a new peak corresponding to the absorption of PBS at 560 nm was observed on the incident photon to charge carrier efficiency (IPCE) spectra; however, the overall photovoltaic performance was not improved due to the significant decrease of dye-loading density of porphyrin dyes, indicating a need to break off the trade-off between dye-loading and light-harvesting. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4ta01954g

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
Fluorene functionalized porphyrins as broadband
absorbers for TiO2 nanocrystalline solar cells†
Cite this: J. Mater. Chem. A, 2014, 2,
Wenhui Li,‡a Liping Si,‡bc Zonghao Liu, Zhixin Zhao, Hongshan He,
Kai Zhu,d
13667
a
a
bc
*
*
Brian Mooree and Yi-Bing Chengaf
Three 9,9-dihexyl-9H-fluorene (DHF) functionalized zinc porphyrin dyes (coded as ZZX-N3, ZZX-N4, and
ZZX-N5) were designed and synthesized for dye-sensitized solar cells. Then, DHF and benzoic acid were
conjugated to the porphyrin ring through triple bonds to act as a spacer to elongate the p-conjugation
and as an acceptor for an efficient electron injection, respectively. A bis(9,9-dihexyl-9H-fluorene-7-yl)-
amine (BFA) and a bis(4-hexylphenyl)amine (BPA) were further linked to DHF to act as electron donors in
ZZX-N3 and ZZX-N4, respectively. ZZX-N5 did not have any electron donor and served as a reference.
Moreover, ZZX-N3- and ZZX-N4-sensitized cells exhibited broader sunlight absorption than ZZX-N5, and
as a result, higher photon-to-electricity efficiency (PCE) (ZZX-N3, 3.83%; ZZX-N4, 4.2%; ZZX-N5, 3.70%)
was observed. The results are consistent with well-separated HOMO (highest occupied molecular
orbital) and LUMO (lowest unoccupied molecular orbital) in ZZX-N3 and ZZX-N4 than in ZZX-N5.
However, the overall conversion efficiency of ZZX-N3- and ZZX-N4-sensitized cells was low, which is
due to significant dye aggregation induced by the extra long alkyl-chains on the donor groups. This was
evidenced by blue and red shifts of the absorption spectra of dye-coated TiO2 films. In addition, the
extra long-chains also did not offer better shielding to prevent electron recombination of injected
ꢀ
electrons with I3 in electrolyte as revealed by electrochemical impedance spectroscopy. When a co-
Received 19th April 2014
Accepted 19th June 2014
sensitizer (coded as PBS) was used, a new peak corresponding to the absorption of PBS at 560 nm was
observed on the incident photon to charge carrier efficiency (IPCE) spectra; however, the overall
photovoltaic performance was not improved due to the significant decrease of dye-loading density of
porphyrin dyes, indicating a need to break off the trade-off between dye-loading and light-harvesting.
DOI: 10.1039/c4ta01954g
www.rsc.org/MaterialsA
However, the energy conversion efficiency was less than 2%.4 In
Introduction
¨
2007, Gratzel et al. conjugated a malonic acid group to
b-pyrrolic positions of the porphyrin ring.5 The photoelectrode
exhibited a green color, and the resulting device produced an
energy conversion of 7.1% under AM 1.5 G conditions. Since
then, a variety of porphyrins have been synthesized and their
photovoltaic performance has been evaluated. The results
revealed that higher efficiency can be achieved by adding donor
groups,1,6–16 attaching bulky substituents,11 or by elongating the
p-conjugation through double17 or triple bonds18 between the
porphyrin ring and an anchoring carboxylic acid group. Upto
now, the donor-p-acceptor structure has emerged as a premium
model for constructing high performance dyes.19 In this model,
a diphenylamine and a 4-ethynylbenzoic acid in the 5 and 15
positions of the porphyrin ring are usually used as an electron
donor and an electron acceptor, respectively. The insertion of
phenylethyne, perylene, uorene, and anthracene units
between the donor and the porphyrin ring has been found to be
very effective for spectral broadening.1 Recently, the highest
Porphyrin dyes as excellent “natural” pigments are promising
for dye-sensitized solar cells (DSCs).1–3 Early studies used
5,10,15,20-tetraphenylporphyrin zinc derivatives as dyes, and
the resulting photoelectrodes showed a deep purple color.
a
¨
Michael Gratzel Center for Mesoscopic Solar Cells, Wuhan National Laboratory for
Optoelectronics Department, School of Optical and Electronic Information,
Huazhong University of Science and Technology, 1037 Luoyu Road, Wuhan 430074,
P. R. China. E-mail: zhixin-zhao@hust.edu.cn
bDepartment of Chemistry, Eastern Illinois University, Charleston, IL 61920, USA.
E-mail: hhe@eiu.edu
cCenter for Advanced Photovoltaics, South Dakota State University, Brookings, SD
57007, USA
dChemical and Materials Science Center, National Renewable Energy Laboratory,
Golden, CO 84041, USA
eUniversity Networking Research & Services, South Dakota State University, Brookings,
SD 57007, USA
fDepartment of Materials Engineering, Monash University, Melbourne, Victoria 3800,
¨
Australia
energy conversion efficiency of 13% was reported by Gratzel
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c4ta01954g
et al. using a functionalized porphyrin dye and a cobalt-based
redox couple.20 This signicant achievement pushes DSC
‡ W. Li and L. Si contributed equally.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 13667–13674 | 13667

View Article Online
Journal of Materials Chemistry A
Paper
research toward a new paradigm with commercialization not far
away.
Porphyrin has a high tendency to form H-type and J-type
aggregates21 because of its planar structural characteristics. In
aggregates, the energy of excited dye molecules are more prone
to transfer to neighbouring molecules, competing with the
process of electron injection to the conduction band of TiO2
nanoparticles. Long alkyl chains are oen introduced into
porphyrin rings to prevent the formation of aggregates. Diau
et al. found that long alkoxyl chains (OC12H25) on the phenyl
groups in the two meso positions of the porphyrin ring effec-
tively enveloped the porphyrin ring with signicantly lessened
aggregates.11 However, it is unclear if the aggregation can be
further reduced when additional alkyl chains are introduced
into the electron donor groups.
A complementary dye is also used in high performance
porphyrin-sensitized solar cells. This is due to the weak sunlight
absorption of porphyrin dyes between 500 and 650 nm.
Complementary dyes with an optimal physical footprint and
photophysical properties are needed to cosensitize the devices
for better efficiency.14,19,20 Several organic dyes have been
successfully used in combination with porphyrin dyes. These
dyes exhibited a large physical size, which reduces the dye
loading density of porphyrin and limits the potential for effi-
ciency enhancement. In a recent report,9 we found a small
BODIPY sensitizer (BET) with a maximal absorption at ꢁ510 nm
matched the deciency of the porphyrin absorption very well.
When the BET is paired with the porphyrin dye LD12, the energy
efficiency of the resulting devices increased from 6.5% to 7.2%.
Given the structural exibility of this type of dye to tune their
absorption spectra, this type of complementary dye is prom-
ising for co-sensitized solar cells. It is unclear though if this type
of dye can also act as an effective complementary dye in other
porphyrin systems.
Thus, to this end, three new porphyrin dyes were designed
and synthesized to address the aforementioned questions.
Their structures are shown in Scheme 1. The design is
based on a [10,20-bis(2,6-di-octoxyphenyl)porphinato]zinc(II)
molecule, due to its excellent photovoltaic performance.1 A
9,9-dihexyl-9H-uorenyl group (DHF) as a spacer was intro-
duced between the donor and porphyrin ring to elongate the
p-conjugation and thus to broaden the absorption spectra.
We focus on two different electron donors: bis(9,9-dihexyl-
9H-uoren-7-yl)amine (BFA) and bis(4-hexylphenyl)amine
(BPA), both having four and two n-hexyl chains, respectively,
Scheme 1 Structure of ZZX-N3, ZZX-N4, ZZX-N5 and PBS.
Experimental section
Detailed synthetic procedures of all dyes and their character-
ization are described in the ESI.†
Theoretical calculations
Theoretical calculations were performed at the density func-
tional theory level. The calculations were performed in a
vacuum as implemented in the Gaussian 09 (ref. 22) program
package. The structures were optimized at the B3LYP/6-31g(d)
levels. No negative frequency was observed in the optimized
structures. The molecular orbitals were visualized by GaussView
3.0 soware. This method has been widely used for geometry
optimization and electronic calculations of porphyrins deriva-
tives.23–25 It should be pointed out that linear conformation of
the alkyl chains from initial optimization was adopted for all
calculations.
Fabrication of solar cells
and are thus expected to provide better shielding to reduce Nanocrystalline TiO2 paste (DSL 18NR-T, ꢁ20 nm, Dyesol,
electron recombination at the TiO2/electrolyte interface. A Australia) was coated onto commercial uorine-tin-oxide
dye without an electron donor was also designed to serve as a (FTO) (7 U cmꢀ2, Hartford Glass, USA) using a doctor-blade
reference for the convenient evaluation of the other two dyes. method, and 3M magic tape was used as a spacer. Aer the
A small organic dye (PBS) with a maximal absorption at 560 lms were air dried for about 30 minutes, the samples were
nm was prepared to complement the porphyrin dyes. We put into an oven for sintering using a three-stage ramp-up
expect that the photophysical, photovoltaic, and electro- scheme (8 ꢂC minꢀ1, 200 ꢂC for 15 minutes, 350 ꢂC for 15
chemical studies of these dyes will provide evidence to minutes, and 550 ꢂC for 30 minutes) in ambient oxygen. Aer
address the aforementioned questions as well as to provide the samples were cooled to room temperature slowly, they
fundamental information for the design of next generation of were then coated with a scattering layer (WER2-O, 150–400 nm
porphyrin dyes.
Dyesol, Australia) and sintered again using a same three-stage
ramp scheme. The samples were then cooled to room
13668 | J. Mater. Chem. A, 2014, 2, 13667–13674
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
temperature and immersed in a freshly prepared TiCl4
aqueous solution (20 mM) for one hour. The lms were air-
dried and sintered again at 550 ꢂC for 30 minutes. Further-
more, the lms were put into the dye solution (ꢁ0.3 mM) at
room temperature and kept for four hours. The lms were
then taken out, ushed with methanol, and vacuum dried for
one hour. The thickness of the lm was ꢁ14 mm, determined
from a prolometer. The counter-electrode was prepared by
sputtering a 16 nm-thick lm of Pt onto the FTO glass. A mask
made from the Paralm was used as a spacer for the two
electrodes and sealing material. The electrolyte solution was
injected from two pre-cut channels in the mask. The outside
of the nal cell was sealed by hot glue. The electrolyte is an
acetonitrile/valeronitrile solution with 0.6 M 1-propyl-2,3-
dimethylimidazolium iodide, 0.05 M I2, 0.1 M LiI, 0.1 M
guanidine thiocyanate, and 0.5 M tert-butylpyridine.
Results and discussion
ZZX-N3, ZZX-N4, and ZZX-N5 were synthesized and puried as
described in the ESI.† The synthesis procedure is outlined in
Scheme 2. Iodouorene derivatives 1a and 1b were prepared by
the reaction of BFA (or BPA) and 2,7-diiodo-9,9-dihexyl-9H-u-
orene (DIDHF) under Ullmann cross-coupling reaction condi-
tions and the yield was ꢁ40%. A slight excess of BFA and DPA to
DIDHF were added to maximize the yield of target products. The
intermediates 2a–2c were obtained from the reaction between 1
and (triisopropylsilyl)acetylene under Sonogashira cross-
coupling conditions with excellent yields. 2a–2c were then
subjected to a two-step reaction to give intermediates 4a–4c with
15–25% yield. The nal products: ZZX-N3, ZZX-N4, and ZZX-N5
were obtained in 60–85% yields under the same reaction
conditions by reacting 4a–4c with 4-iodobenzoic acid. All three
dyes exhibited good solubility in chloroform, dichloromethane,
and toluene, and were slightly soluble in acetonitrile and
ethanol. PBS was prepared by three consecutive reactions as
shown in Scheme 3. Further, PBH was obtained from the reac-
tion of 4-formylbenzoic acid with 2,4-dimethylpyrrole in
CH2Cl2, followed by oxidation with DDQ and coordination of
BF2 in the presence of TEA, giving a 23% yield. PBI was
synthesized by the reaction of PBH with I2/HIO3 in ethanol. PBI
was then treated with ethynyltrimethylsilane under Sonogashira
coupling conditions to produce PBS with ꢁ35% yield. PBI was
purple with no uorescence, while PBH and PBS were orange
and emitted very strongly under UV and visible light
illumination.
Characterization of devices
The applied potential and cell current were measured on an
Agilent 4155C semiconductor parameter analyser with a 450 W
Xenon light source. The light intensity at the surface of cell was
calibrated to 100 mW cmꢀ2, equivalent to one sun at air mass
1.5 G condition. The efficiency (h) and ll factor (FF) were
calculated by h (%) ¼ Pmax ꢃ 100/(Pin ꢃ A) and FF ¼ Pmax/(Isc
ꢃ
Voc), where Pmax is the maximum output power of the cells
(mW), Pin is the power density of the light source (mW cmꢀ2), Isc
is the short-circuit current (mA), Voc is the open-circuit voltage
(V), and A is the active area (0.16 cm2) of the cell. Incident
photon to charge carrier efficiency (IPCE) experiments were
performed on an Agilent semiconductor parameter analyzer
with an Xe arc lamp and a Newport monochromator. Mono-
chromatic light was incident on the sample through focusing
lenses. The National Renewable Energy Laboratory (NREL)
calibrated photodetector was used as a reference. The samples
were scanned from 350–850 nm, and the voltage was recorded.
IPCE was calculated using the equation IPCE ¼ (sample voltage
ꢃ reference voltage)/reference IPCE. The reference IPCE was
supplied by the NREL-calibrated DSSC under an AM 1.5 G
photon ux. Charge transport and recombination were studied
by intensity-modulated photocurrent spectroscopy (IMPS) and
photovoltage (IMVS) spectroscopy, respectively, as detailed
previously.26–28 In brief, the cells were probed with a small
sinusoidally-modulated beam of 660 nm LED light super-
imposed on a relatively large background (bias) illumination
also at 660 nm. The probe and bias light entered the cell from
the substrate side. Neutral density lters were used to vary the
illumination intensity. The amplitude of the modulated
photocurrent density was kept at 10% or lower compared to the
steady-state photocurrent density. In addition, electrochemical
impedance spectroscopy (EIS) measurements were performed
with a potentiostat/frequency analyzer (PARSTAT 2273) using a
two-electrode conguration, with the modulation frequencies
ranging from 10ꢀ2 to 105 Hz and a modulation amplitude of 10
mV. Z-view 2.9c (Scribner Associates) was used to t the EIS
spectra to the equivalent circuit, based on the transmission line
model.29
The absorption and uorescence spectra of ZZX-N3, ZZX-N4
and ZZX-N5 in ethanol solution are shown in Fig. 1. Three
porphyrins exhibit strong absorption around 450 nm (Soret
Scheme 2 Synthesis of fluorene-based porphyrin dyes. Reaction
conditions: (i) copper bronze, K2CO3, 18-crown-6,1,2-dichlorobenze,
reflux, (ii) ethynyltriisopropylsilane, Pd(PPh3)2Cl2, CuI, THF, TEA, (iii) (a)
TBAF, THF; (b) Pd(PPh3)2Cl2, CuI, THF, TEA, (iv) (a) TBAF, THF; (b)
Pd2(dba)3, AsPh3, 4-iodobenzoic acid, THF, TEA.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 13667–13674 | 13669

View Article Online
Journal of Materials Chemistry A
Paper
solution. The blue-shi was more pronounced in ZZX-N3. A new
red-shied shoulder was also observed for ZZX-N3, which was
much weaker in ZZX-N4 and not observed in ZZX-N5. As the
blue-shi is usually indicative of H-type aggregates of porphyrin
dyes and a red-shi of J-type aggregates,11 the observed spectral
changes indicated that H-type aggregates existed in the three
dyes, which was more severe in ZZX-N3 than in the other two.
J-type aggregation was also more obvious in ZZX-N3 than in
ZZX-N4, whereas no sign of a J-type aggregation was observed in
ZZX-5. The observations indicate that long alkyl chains on the
uorene groups induced H-type aggregation, which was
consistent with a report from Diau et al.11 In that report, both
the H- and J-type aggregates were observed in a porphyrin dye
with two C8H17 substituents on the donor group. This suggested
that additional long alkyl chains facilitated the formation of
aggregates.
Scheme 3 Synthesis of complementary dye PBS reaction conditions:
(i) TFA, CH2Cl2, overnight, N2, (ii) DDQ, 10 minutes, (iii) TEA, BF3$OEt2,
(iv) I2, HIO3, EtOH, 60 ꢂC,
Pd(PPh3)2Cl2, CuI, THF, TEA.
1 h and (v) ethynyltrimethylsilane,
The photovoltaic performance of ZZX-N3, ZZX-N4, and ZZX-
N5 were examined in solar cells using a conventional “sand-
wiched” conguration. The Pt-coated FTO glass was used as a
counter electrode. The thickness of the lm (compact layer +
nanocrystalline layer + scattering layer) was 14 mm. The
measurements were performed under AM 1.5 G conditions.
Their J-V curves and IPCE spectra are shown in Fig. 2 and Fig. 3
with the detailed photovoltaic parameters in Table 2. The
overall energy conversion efficiency changed in the order of
ZZX-N3 (3.83%) < ZZX-N4 (4.20%) > ZZX-N5 (3.70%). The ZZX-
N3- and ZZX-N4-sensitized cells exhibited slightly higher Voc
than ZZX-N5-sensitized cells, but this was only ꢁ2% larger. A
much bigger difference was observed in Jsc. ZZX-N3- and ZZX-N4
sensitized cells exhibited ꢁ13% larger Jsc than that of ZZX-N5
sensitized solar cells. The trend in Jsc was consistent with their
IPCE, as shown in Fig. 3. The three dyes exhibited photon-to-
electron responses from 350 nm to 725 nm. Strong IPCE values
at 472 and 670 nm corresponded to Soret and Q bands in their
absorption spectra. The IPCE maximums of ZZX-N3 and ZZX-N4
were similar to that of ZZX-N5, but the two bands were broader
than the latter, which was consistent with their absorption
behaviour on the lms. The measured amounts of dye mole-
cules on TiO2 lms were nearly the same for all three dyes (10.37
ꢃ 10ꢀ9, 10.45 ꢃ 10ꢀ9, and 10.14 ꢃ 10ꢀ9 mol for ZZX-N3, ZZX-
N4, and ZZX-N5, respectively), indicating that the better
Fig. 1 Absorption spectra of ZZX-N3 (A), ZZX-N4 (B) and ZZX-N5 (C) in
EtOH and on TiO2 films, and fluorescence spectra of three dyes in
EtOH (D).
band). Two medium absorption peaks (Q bands) between 550
and 700 nm were also observed. The detailed spectroscopic data
were listed in Table 1. The peak positions of the three porphy-
rins were very similar to each other except a slight red-shi of
ZZX-N3 and ZZX-N4 compared to ZZX-N5. In addition, the Soret
bands and Q bands of ZZX-N3 and ZZX-N4 were broader than
ZZX-5, which was more pronounced on the right side of each
band. The broadening, according to the Gouterman four-orbital
theory,24,30 is due to the lower symmetry of ZZX-N3 and ZZX-N4
aer adding an electron donor BFA/BPA to the DHF group. No
signicant difference of their emission spectra was observed.
The absorption spectra of porphyrin dyes on TiO2 nanoparticle
lms were also shown in Fig. 1. These spectra were obtained by
dipping dye-coated lms in EtOH solution. All peaks became
broader and blue-shied compared to their counterparts in
Table 1 Spectral and electrochemical data of ZZX-N3, ZZX-N4 and
ZZX-N5
Solvent
ZZX-N3
ZZX-N4
ZZX-N5
Absorption (nm)
In EtOH
On TiO2
In EtOH
459, 666
434, 465, 652
672, 738
460, 666
445, 657
672, 738
456, 662
445, 652
668, 732
Fig. 2 J–V curves of ZZX-N3-, ZZX-N4- and ZZX-N5-sensitized solar
cells.
Fluorescence (nm)
13670 | J. Mater. Chem. A, 2014, 2, 13667–13674
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 4 Geometry-optimized structure of ZZX-N3, ZZX-N4 and ZZX-
N5. The red sphere at the bottom of each image represents O atoms of
benzoic acid. The images were produced from a view with the
porphyrin ring perpendicular to the paper. The donor is on the top and
the acceptor is at the bottom. The phenyl groups on two meso
positions fall into the lines of the long alkyl chains.
Fig. 3 IPCE spectra of ZZX-N3-, ZZX-N4- and ZZX-N5-sensitized
solar cells.
Table 2 Photovoltaic parameters of ZZX-N3-, ZZX-N4- and ZZX-N5-
sensitized solar cells
from the geometry-optimized structures due to their high ex-
ibility; however, these results provide a snapshot of the possible
relaxation states on the surface of the TiO2 nanoparticles.
The electron density distribution proles of the frontier
orbitals of the three dyes are shown in Fig. 5, and the calculated
energy levels are summarized in Table 3. The LUMO energy
levels of all three dyes were higher than the conduction band of
TiO2 NP (ꢀ4.67 eV), indicating that these dyes were energetically
favourable for DSC operation. A distinct separation of HOMO
and LUMO was observed in ZZX-N3 and ZZX-N4. This will be
benecial to the charge separation during the cell operation.
ZZX-N5 did not show such a separation because of the lack of a
donor group in its structure.
Electrochemical impedance spectroscopy (EIS) was carried
out to determine the electron recombination resistance at the
interface between the TiO2/dye/electrolyte medium. Fig. 6
shows the typical Nyquist plots of the impedance responses for
a ZZX-N3-sensitized solar cell measured at three different bias
voltages under open circuit under illumination. The model used
for the impedance analysis has been discussed in detail previ-
ously.32–34 From these plots, two features of the impedance
response can be identied: (1) a large semicircle at low
frequencies, which corresponds to the back charge-transfer
process at the TiO2/redox electrolyte interface with a recombi-
nation resistance; and (2) a small semicircle at high frequen-
cies, which is associated with the charge-transfer process at the
Jsc (mA cmꢀ2
)
Voc (mV) FF
ha (%)
ZZX-N3
ZZX-N4
ZZX-N5
9.93 ꢄ 0.06
9.97 ꢄ 0.05
8.79 ꢄ 0.09
659 ꢄ 3
662 ꢄ 2
649 ꢄ 1
632 ꢄ 1
643 ꢄ 3
653 ꢄ 2
0.586 ꢄ 0.02 3.83 ꢄ 0.02
0.637 ꢄ 0.02 4.20 ꢄ 0.01
0.649 ꢄ 0.05 3.70 ꢄ 0.01
0.662 ꢄ 0.03 3.24 ꢄ 0.03
0.658 ꢄ 0.03 3.75 ꢄ 0.05
0.669 ꢄ 0.03 2.76 ꢄ 0.02
ZZX-N3/PBS 7.73 ꢄ 0.05
ZZX-N4/PBS 8.88 ꢄ 0.06
ZZX-N5/PBS 6.32 ꢄ 0.04
a
Raw data, see Table S1 in the ESI.
photovoltaic performance of ZZX-N3 and ZZX-N4 was caused by
the existence of donor groups. These results conrmed that the
addition of an electron donor to the uorenyl group enhances
the cell's light-harvesting capacity, leading to a higher Jsc.
To shield some light on the different photovoltaic perfor-
mance of the three dyes, we performed density functional
theory (DFT) calculations for the three molecules at the B3LYP/
6-31g(d) level. This method has previously been successfully
used for the geometry optimization and frontier molecular
orbital calculations in zinc porphyrin compounds.10,24,31 In our
calculations, linear congurations of long alkyl or alkoxyl
chains were adopted in the initial structures to simplify the
calculations. The geometry-optimized structures are shown in
Fig. 4. The uorene core of the spacer DHF group was aligned in
the plane of the porphyrin ring with a torsion angle less than 1ꢂ.
Two n-hexyl chains were aligned on the two sides of uorenyl
unit and stretched away from each other. Furthermore, when
the donor BFA was attached to DHF, the four n-hexyl chains on
BFA aligned in a cross-over conguration with an angle ꢁ86ꢂ;
whereas, when BPA was used, two n-hexyl chains stretched out
with an angle ꢁ148ꢂ. Two 2,4-bis(n-octoxyl)phenyl groups on
the meso positions of the porphyrin ring were tilted in two
different directions. As a result, the four alkoxyl groups also
formed a cross-over shape with angles of ꢁ13ꢂ, 21ꢂ, and 28ꢂ, for
ZZX-N3, ZZX-N4, and ZZX-N5, respectively. The four alkoxyl
chains became more parallel aer the BFA and BPA groups were
introduced into the system. It should be pointed out that the
Fig. 5 Electron density distribution profiles of LUMO (top) and HOMO
actual congurations of these alkyl chains may be different (bottom) of ZZX-N3, ZZX-N4 and ZZX-N5.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 13667–13674 | 13671

View Article Online
Journal of Materials Chemistry A
Paper
Table 3 Calculated frontier molecular orbital energy levelsa
are very similar to ZZX-N5 under open circuit voltage, which is
consistent with their recombination lifetimes of electrons on
the TiO2 surface. These results indicate that the long alkyl
groups on the electron donor of ZZX-N3 and ZZX-N4 did not
offer an extra shielding effect to reduce the charge recombina-
tion than ZZX-N5 at the TiO2/redox electrolyte interface. The
four alkoxyl chains (OC8H17) are sufficient enough to reduce the
charge recombination.
ZZX-N3
ZZX-N4
ZZX-N5
HOMO (eV)
LUMO (eV)
LUMO–HOMO(eV)
ꢀ4.5772
ꢀ2.4373
2.1399
ꢀ4.5590
ꢀ2.3851
2.1739
ꢀ4.6817
ꢀ2.4169
2.2648
a
In vacuum.
The EIS results combined with the absorption spectra of the
dye-coated TiO2 lms also indicated that the different photo-
voltaic performances of ZZX-N3 and ZZX-N4 was mainly caused
by the dye aggregation. The long alkyl chains on the donor led to
increased dye aggregates.11 In ZZX-N3, the two phenyl groups in
the two meso positions are almost vertical to the porphyrin ring
(torsion angle 86.52ꢂ). The two phenyl groups from the neigh-
bouring molecules are more parallel, making them much easier
to stack over each other than those from ZZX-N4 and ZZX-N5
(torsion angles are 84.57ꢂ and 79.91ꢂ, respectively).
In an effort to enhance the light-harvesting capacity in the
region between the Soret band and Q bands, a co-sensitizer PBS
was synthesized. PBS is a BODIPY-based dye with strong
absorption at 560 nm, which is complementary to the absorp-
tion of ZZX-N3, ZZX-N4 and ZZX-N5. The solutions containing
PBS and porphyrin dyes were used for the cell preparation. The
J–V curves of the co-sensitized solar cells are shown in Fig. 8.
The molar ratio of PBS to porphyrin dye was xed at 1 : 1, and
the dye-loading time was 4.5 h. The overall energy conversion
efficiency of these cells changed in the order of ZZX-N3/PBS
(3.24%) < ZZX-N4/PBS (3.75%) > ZZX-N5/PBS (2.74%), which is
lower than their counterparts without co-sensitization, owing to
the reduced Jsc. The results are consistent with the decreased
IPCE as shown in Fig. 3 and 9.
The decreased IPCE of co-sensitized solar cells is due to the
decreased dye-loading density of the porphyrin dyes on the TiO2
surface. When PBS was co-sensitized with porphyrin, the cell
indeed picked up the photons in the region around 560 nm, as
shown in Fig. 9; however, the IPCE of the co-sensitized cells was
generally lower than those of the cells without co-sensitization
and a typical comparison was shown in Fig. 10. As we previously
reported in similar systems,15 this reduction in IPCE magnitude
is ascribed to the decreased loading density of porphyrin, as
Fig.
6 Typical Nyquist plots of the electrochemical impedance
responses of a solar cell using the ZZX-N3 dye with three different bias
voltages.
counter electrode/electrolyte interface with a charge-transfer
resistance. Normally, there is a linear region with 45ꢂ slope at
medium frequencies, which is characteristic for the trans-
mission line model and can be used to derive the transport
resistance of the TiO2 electrode. However, this feature is not
clear in the current study. However, this does not affect our
understanding on the charge recombination resistance at the
TiO2/redox electrolyte interface.
Fig. 7 shows the electron recombination resistance and
electron lifetime measured at different biases. It was expected
that long alkyl groups on the donor could provide extra
shielding to minimize the charge recombination on the inter-
face between TiO2 and the electrolyte. However, it was found
that the charge recombination resistance of ZZX-N3 and ZZX-N4
Fig. 7 Electron recombination and recombination lifetimes of ZZX-
N3, ZZX-N4, and ZZX-N5 obtained from electrochemical impedance Fig. 8 J–V curves of ZZX-N3/PBS, ZZX-N4/PBS, and ZZX-N5/PBS co-
spectroscopy.
sensitized solar cells.
13672 | J. Mater. Chem. A, 2014, 2, 13667–13674
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
Acknowledgements
This work was partially supported by the National Science
Foundation/EPSCoR, grant no. 0554609 (H.H), Eastern Illinois
University The Council on Faculty Research Grant (H.H.), the
National Basic Research Program of China (973 program), grant
no. 2011CBA00703, the Fundamental Research Funds for the
Central Universities, grant no. HUST: 2012QN108 (Z.Z.) and the
Division of Chemical Sciences, Geosciences, and Biosciences,
Office of Basic Energy Sciences, U.S. Department of Energy,
under contract no. DE-AC36-08GO28308 with the National
Renewable Energy Laboratory (K.Z.).
Fig. 9 IPCE spectra of ZZX-N3/PBS, ZZX-N4/PBS, and ZZX-N5/PBS
sensitized solar cells.
Notes and references
1 L. L. Li and E. W. Diau, Chem. Soc. Rev., 2013, 42, 291–304.
¨
2 A. Kay and M. Gratzel, J. Phys. Chem., 1993, 97, 6272–6277.
3 H. Imahori, T. Umeyama and S. Ito, Acc. Chem. Res., 2009, 42,
1809–1818.
4 W. M. Campbell, A. K. Burrell, D. L. Officer and K. W. Jolley,
Coord. Chem. Rev., 2004, 248, 1363–1379.
5 W. M. Campbell, K. W. Jolley, P. Wagner, K. Wagner,
P. J. Walsh, K. C. Gordon, L. Schmidt-Mende,
¨
M. K. Nazeeruddin, Q. Wang, M. Gratzel and D. L. Officer,
J. Phys. Chem. C, 2007, 111, 11760–11762.
6 C.-F. Lo, S.-J. Hsu, C.-L. Wang, Y.-H. Cheng, H.-P. Lu,
E. W.-G. Diau and C.-Y. Lin, J. Phys. Chem. C, 2010, 114,
12018–12023.
7 C.-W. Chang, C. J. Bhongale, C.-S. Lee, W.-K. Huang,
C.-S. Hsu and E. W.-G. Diau, J. Phys. Chem. C, 2013, 116,
15146–15154.
Fig. 10 Absorption spectrum of ZZX-N3/PBS in ethanol (solid line) and
the IPCE of ZZX-N3- and ZZX-N3/PBS-sensitized solar cells (dotted
lines).
8 J. Luo, M. Xu, R. Li, K.-W. Huang, C. Jiang, Q. Qi, W. Zeng,
J. Zhang, C. Chi, P. Wang and J. Wu, J. Am. Chem. Soc.,
2013, 136, 265–272.
9 M. Shrestha, L. Si, C.-W. Chang, H. He, A. Sykes, C.-Y. Lin
and E. W.-G. Diau, J. Phys. Chem. C, 2012, 116, 10451–10460.
10 R. Ambre, K.-B. Chen, C.-F. Yao, L. Luo, E. W.-G. Diau and
C.-H. Hung, J. Phys. Chem. C, 2012, 116, 11907–11916.
11 C.-L. Wang, C.-M. Lan, S.-H. Hong, Y.-F. Wang, T.-Y. Pan,
C.-W. Chang, H.-H. Kuo, M.-Y. Kuo, E. W.-G. Diau and
C.-Y. Lin, Energy Environ. Sci., 2012, 5, 6933–6940.
porphyrin and PBS compete for binding to TiO2 NP with a xed
surface area. This trade-off could be maximized by choosing the
appropriate: molar ratio of PBS and porphyrin dyes, dye loading
time, and thickness of the lm, because these factors will affect
the thermal dynamics and kinetics of the dye adsorption
process.35–38
Conclusions
In an effort to broaden the photo-to-current response of dye- 12 C.-H. Wu, T.-Y. Pan, S.-H. Hong, C.-L. Wang, H.-H. Kuo,
sensitized solar cells, new porphyrin dyes, ZZX-N3, ZZX-N4,
ZZX-N5, and a complementary dye, PBS, were developed. It was
Y.-Y. Chu, E. W.-G. Diau and C.-Y. Lin, Chem. Commun.,
2012, 48, 4329–4331.
found that further linkage of bis(9,9-dihexyl-9H-uoren-7-yl)- 13 W.-S. Chao, K.-H. Liao, C.-T. Chen, W.-K. Huang, C.-M. Lan
amine (BFA) or bis(4-hexylphenyl)amine (BPA) to DHF as a and E. W.-G. Diau, Chem. Commun., 2012, 48, 4884–4886.
donor led to broader Soret and Q bands, because of the loss of 14 C.-M. Lan, H.-P. Wu, T.-Y. Pan, C.-W. Chang, W.-S. Chao,
symmetry, resulting in a higher energy conversion efficiency of
BFA and BPA-functionalized porphyrin dyes than the porphyrin
C.-T. Chen, C.-L. Wang, C.-Y. Lin and E. W.-G. Diau, Energy
Environ. Sci., 2012, 5, 6460–6464.
dye without a donor. Comparison of the dye absorption on TiO2 15 H.-P. Wu, Z.-W. Ou, T.-Y. Pan, C.-M. Lan, W.-K. Huang,
nanoparticle lms revealed that extra long-alkyl-chains on the
H.-W. Lee, N. M. Reddy, C.-T. Chen, W.-S. Chao, C.-Y. Yeh
donor promoted the formation of aggregates. The application of
and E. W.-G. Diau, Energy Environ. Sci., 2012, 5, 9843–9848.
complementary dye PBS indicated the photon-to-electron 16 L.-L. Li, C.-W. Chang, H.-H. Wu, J.-W. Shiu, P.-T. Wu and
response around 560 nm; however, the overall light-harvesting E. W.-G. Diau, J. Mater. Chem., 2012, 22, 6267–6273.
efficiency decreased due to the trade-off between decreased dye- 17 H. He, A. Gurung, L. Si and A. G. Sykes, Chem. Commun.,
loading density and broadened absorption spectra.
2012, 48, 7619–7621.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 13667–13674 | 13673

View Article Online
Journal of Materials Chemistry A
Paper
18 Y.-C. Chang, C.-L. Wang, T.-Y. Pan, S.-H. Hong, C.-M. Lan, 25 Q. Wang, W. M. Campbell, E. E. Bonfantani, K. W. Jolley,
H.-H. Kuo, C.-F. Lo, H.-Y. Hsu, C.-Y. Lin and E. W.-G. Diau,
Chem. Commun., 2011, 47, 8910–8912.
D. L. Officer, P. J. Walsh, K. Gordon, R. Humphry-Baker,
¨
M. K. Nazeeruddin and M. Gratzel, J. Phys. Chem. B, 2005,
19 A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran,
109, 15397–15409.
¨
M. K. Nazeeruddin, E. W.-G. Diau, C.-Y. Yeh, 26 G. Schlichthorl, S. Y. Huang, J. Sprague and A. J. Frank,
¨
S. M. Zakeeruddin and M. Gratzel, Science, 2011, 334, 629–
J. Phys.Chem. B, 1997, 11, 8141–8155.
634.
27 G. K. Boschloo, Ph.D. Thesis, Del University of Technology,
20 S. Mathew, A. Yella, P. Gao, R. Humphry-Baker,
1996.
B. F. E. Curchod, N. Ashari-Astani, I. Tavernelli, 28 L. M. Peter, Chem. Rev., 1990, 90, 753–769.
¨
U. Rothlisberger, M. K. Nazeeruddin and M. Gratzel, Nat. 29 F. Fabregat-Santiago, J. Bisquert, G. Garcia-Belmonte,
Chem., 2014, 6, 242–247.
21 C.-F. Lo, L. Luo, E. W.-G. Diau, I.-J. Chang and C.-Y. Lin,
Chem. Commun., 2006, 1430–1432.
22 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
G. Boschloo and A. Hagfeldt, Sol. Energy Mater. Sol. Cells,
2005, 87, 117–131.
30 J. C. Earles, K. C. Gordon, A. W. I. Stephenson, A. C. Partridge
and D. L. Officer, Phys. Chem. Chem. Phys., 2011, 13, 1597–
1605.
¨
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, 31 J. Feng, Y. Jiao, W. Ma, M. K. Nazeeruddin, M. Gratzel and
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, S. Meng, J. Phys. Chem. C, 2013, 117, 3772–3778.
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, 32 F. Fabregat-Santiago, G. Garcia-Belmonte, J. Bisquert,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr,
A. Zaban and P. Salvador, J. Phys. Chem. B, 2002, 106, 334–
339.
J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, 33 I. Mora-Sero, T. Dittrich, A. Belaidi, G. Garcia-Belmonte and
E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Bisquert, J. Phys. Chem. B, 2005, 109, 14932–14938.
J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, 34 K. Zhu, S.-R. Jang and A. J. Frank, J. Phys. Chem. Lett., 2011, 2,
S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, 1070–1076.
M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, 35 J.-H. Yum, B. E. Hardin, E. T. Hoke, E. Baranoff,
J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski,
R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth,
S. M. Zakeeruddin, M. K. Nazeeruddin, T. Torres,
¨
M. D. McGehee and M. Gratzel, ChemPhysChem, 2011, 12,
657–661.
P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, 36 B. E. Hardin, J.-H. Yum, E. T. Hoke, Y. C. Jun, P. Pechy,
O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and
D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
T. Torres, M. L. Brongersma, M. K. Nazeeruddin,
¨
M. Gratzel and M. D. McGehee, Nano Lett., 2010, 10, 3077–
23 T. M. Koh, K. Nonomura, N. Mathews, A. Hagfeldt,
3083.
¨
M. Gratzel, S. G. Mhaisalkar and A. C. Grimsdale, J. Phys. 37 M. D. Brown, P. Parkinson, T. Torres, H. Miura, L. M. Herz
Chem. C, 2013, 117, 15515–15522. and H. J. Snaith, J. Phys. Chem. C, 2011, 115, 23204–23208.
24 M. P. Balanay and D. H. Kim, Phys. Chem. Chem. Phys., 2008, 38 A. Harriman, L. J. Mallon, G. Ulrich and R. Ziessel,
10, 5121–5127.
ChemPhysChem, 2007, 8, 1207–1214.
13674 | J. Mater. Chem. A, 2014, 2, 13667–13674
This journal is © The Royal Society of Chemistry 2014