Unravelling the effect of anchoring groups on the ground and excited state properties of pyrene using computational and spectroscopic methods

Article abstract of DOI:10.1039/c6cp00571c

Anchoring groups play an important role in dye sensitized solar cells (DSCs). In order to acquire a suitable anchoring group for DSCs, a deeper understanding of the effect of anchoring groups on the ground and excited state properties of the dye is significant. In this context, various anchoring group connected pyrene derivatives are successfully synthesized and well characterized by using ¹H, ¹³C-NMR, FT-IR and EI-MS spectrometry. The anchoring groups employed are carboxylic acid, malonic acid, acrylic acid, malononitrile, cyanoacrylic acid, rhodanine and rhodanine-3-acetic acid. The optimized geometries, HOMO-LUMO energy gap, light harvesting efficiency (LHE) and electronic absorption spectra of these dyes are studied by using density functional theory (DFT) calculations. The results show that pyrene connected with anchoring groups with weak electron pulling strength (PC, PAC and PMC) has a larger HOMO-LUMO energy gap, whereas that connected with anchoring groups with strong electron pulling strength (PCC, PMN, PR and PRA) has a reduced HOMO-LUMO energy gap. These molecules with a reduced energy gap are primarily preferred for DSC applications. Moreover, P, PC, PAC and PMC molecules undergo π → π? transition, whereas PCC, PMN, PR and PRA molecules show significant charge transfer along with π → π? transition. UV-visible absorption spectral studies on these dyes reveal that connecting various anchoring groups with different electron pulling abilities enables the pyrene chromophore to absorb in the longer wavelength region. Notably, an efficient bathochromic shift is observed for PCC, PMN, PR and PRA molecules in both electronic absorption and fluorescence spectral measurements, which suggests that the excitation is delocalized throughout the entire π-system of the molecules. Both theoretical and spectral studies reveal that dyes with an ICT character (PCC, PMN, PR and PRA) are suitable for dye sensitized solar cell applications.

Full text of DOI:10.1039/c6cp00571c

PCCP
View Article Online
View Journal
PAPER
Unravelling the effect of anchoring groups on the
ground and excited state properties of pyrene
using computational and spectroscopic methods†
Cite this:DOI: 10.1039/c6cp00571c
Arunkumar Kathiravan,*a Murugesan Panneerselvam,b Karuppasamy Sundaravel,c
Nagaraj Pavithra,d Venkatesan Srinivasan,a Sambandam Anandand and
Madhavan Jaccob*b
Anchoring groups play an important role in dye sensitized solar cells (DSCs). In order to acquire a suitable
anchoring group for DSCs, a deeper understanding of the effect of anchoring groups on the ground and
excited state properties of the dye is significant. In this context, various anchoring group connected pyrene
derivatives are successfully synthesized and well characterized by using 1H, 13C-NMR, FT-IR and EI-MS
spectrometry. The anchoring groups employed are carboxylic acid, malonic acid, acrylic acid, malononitrile,
cyanoacrylic acid, rhodanine and rhodanine-3-acetic acid. The optimized geometries, HOMO–LUMO energy
gap, light harvesting efficiency (LHE) and electronic absorption spectra of these dyes are studied by using
density functional theory (DFT) calculations. The results show that pyrene connected with anchoring groups
with weak electron pulling strength (PC, PAC and PMC) has a larger HOMO–LUMO energy gap, whereas that
connected with anchoring groups with strong electron pulling strength (PCC, PMN, PR and PRA) has a
reduced HOMO–LUMO energy gap. These molecules with a reduced energy gap are primarily preferred for
DSC applications. Moreover, P, PC, PAC and PMC molecules undergo p - p* transition, whereas PCC, PMN,
PR and PRA molecules show significant charge transfer along with p - p* transition. UV-visible absorption
spectral studies on these dyes reveal that connecting various anchoring groups with different electron pulling
abilities enables the pyrene chromophore to absorb in the longer wavelength region. Notably, an efficient
bathochromic shift is observed for PCC, PMN, PR and PRA molecules in both electronic absorption and
fluorescence spectral measurements, which suggests that the excitation is delocalized throughout the entire
p-system of the molecules. Both theoretical and spectral studies reveal that dyes with an ICT character (PCC,
PMN, PR and PRA) are suitable for dye sensitized solar cell applications.
Received 26th January 2016,
Accepted 8th April 2016
DOI: 10.1039/c6cp00571c
www.rsc.org/pccp
problem due to limited resources and the cumbersome purification
steps needed. In recent years, interest in organic dyes has increased
Introduction
Conversion of solar energy into electricity is generally considered because of their advantages, such as large molar extinction
to be the most promising way to solve the future energy crisis. Over coefficients, easy modification due to relatively short synthetic
the past few decades, a large number of light harvesting devices routes, and low purification cost, when compared with the
have been developed including dye-sensitized solar cells (DSCs).1–3 conventional ruthenium based chromophores.6–8 For instance,
To date, DSC sensitizers based on Ru(II)-polypyridyl complexes porphyrins are a family of organic dyes that have fascinating
have achieved power conversion efficiencies of almost 12% at applications in DSCs owing to their extremely high molar
standard global air mass AM 1.5G.4,5 However, the large scale extinction coefficients and synthetic versatility.9–12 In particular,
¨
application of ruthenium complexes has become a significant Diau and Gratzel et al. have synthesized a push–pull porphyrin
(YD2) with a diarylamino moiety as a strong electron-donating
a National Centre for Ultrafast Processes, University of Madras, Taramani Campus,
Chennai – 600 113, Tamil Nadu, India. E-mail: akathir23@hotmail.com
b Department of Chemistry, Loyola College, Chennai – 600 034, Tamil Nadu, India
c Chemical Laboratory, CSIR-Central Leather Research Institute, Adyar,
Chennai – 600020, Tamil Nadu, India
group and a carboxyphenylethynyl moiety as a strong electron-
withdrawing anchoring group, which exhibited a high Z-value of
11%.13 They further utilized a cobalt(II/III) electrolyte that can
yield a higher open circuit voltage (VOC), with a cocktail of a
YD2 derivative (YD2-o-C8) and a complementary organic dye to
achieve a record Z-value of 12.3%.14 Very recently, a DSC with
d Nanomaterials and Solar Energy Conversion Laboratory, Department of Chemistry,
National Institute of Technology, Tiruchirappalli – 620015, Tamil Nadu, India
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6cp00571c 13% efficiency was achieved through the molecular engineering
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
of SM371 called SM315 (a porphyrin sensitizer).15 SM315 consideration for DSC device functionality.38 Various anchoring
exhibited significant broadening of Soret and Q-band absorbance groups have been successfully demonstrated in DSC devices,39
features compared to SM371 due to the incorporation of the e.g., carboxylate, phosphonate, sulfonate, salicylate, 8-hydroxyl-
proquinoidal benzothiadiazole (BTD) unit and it demonstrated quinoline, acetylacetonate, catechol, hydroxyl, pyridyl, hydroxamate,
an enhancement in green light absorption, resulting in an cyanoacrylic acid and rhodanine-3 acetic acid moieties.40 Carboxylic
improved JSC (18.1 vs. 15.9 mA cmÀ2 for SM315 and SM371, acid, cyanoacrylic acid and rhodanine-3-acetic acid are found to
respectively). Eventually, SM315 achieved 13% power conversion be efficient electron acceptors and have been widely employed
efficiency under full sun illumination without the requirement as anchoring groups for the attachment of the visible wavelength
of a co-sensitizer.
absorbing D–p–A dyes on the TiO2 surface.41,42
Various organic molecules,16 such as merocyanine, coumarin,
Even though dyes with various anchoring groups have been
indoline, squaraine, hemicyanine, phenothiazine, triphenylamine, reported for DSCs, to the best of our knowledge there are no
fluorene, carbazole and tetrahydroquinoline have been reported, systematic studies to experimentally probe the effect of anchoring
thus making their application in DSCs desirable. The most groups on the excited state properties of the dyes. In fact, a minor
efficient metal-free organic dye reported to date has been the change in the dye molecular structure will result in different and
phenyldihexyloxy-substituted TPA (DHO-TPA) dye synthesized by interesting photophysical, electrochemical and other properties.43
¨
Gratzel et al., which exhibits a power conversion efficiency (Z) of To gain a deeper insight into the structure–properties relationship
10.3% in combination with a cobalt redox shuttle.17 Wang et al. of the dye, a series of pyrene derivatives with varying anchoring
have demonstrated Z up to 10.3% for the substituted TPA dye in groups were designed and synthesized. The anchoring groups
combination with an iodine/iodide redox shuttle.18
used were carboxylic acid, malonic acid, acrylic acid, malono-
The pyrene chromophore is another widely studied building nitrile, cyanoacrylic acid, rhodanine and rhodanine-3-acetic
block material due to its durable electronic properties and long lived acid. UV-visible absorption, steady state fluorescence, time
singlet state.19,20 In addition, pyrene has multiple substitution sites resolved fluorescence techniques and computational methods
for molecular engineering.21 Therefore, many pyrene-based dyes have been used to probe the effect of anchoring groups on the
have been developed over the past few decades, and they have been ground and excited state properties of the synthesized dyes (Fig. 1).
used as a potential candidates for various emerging fields such as
organic electronics,22 fluorescent molecular probes,23 monomers,24
Experimental section
triplet sensitizers,25 sensors,26 perovskite solar cells27 and polymer-
based solar cells.28,29 However, DSCs using pyrene-based sensitizers
are less common. In 2006, Galoppini et al. have reported
phenylenethynylene (PE) rigid linkers for TiO2 sensitization30
and found that PE rigid linkers convert absorbed photons into
electrical current with good efficiency. Later, Thomas et al. have
reported a series of pyrene-based organic dyes,31–33 among them
a dye containing a terthiophene moiety in the conjugation
pathway exhibited a solar energy-to-electricity conversion efficiency
of 5.65%.31 Very recently, Song et al. have reported a novel
tetrahydropyrene-based D–p–A organic dye,34,35 which showed a
power conversion efficiency of 6.75%. On the other hand, a pyrene-
coupled zinc porphyrin shows much improved photovoltaic
performance, giving an overall efficiency of 10.06%. This value
is superior to that of a N719-based solar cell fabricated under
similar experimental conditions (under AM1.5 illumination with
an active area of 0.16 cm2).36 Although, the power conversion
efficiencies of pyrene derivatives are scattered in the literature, to
the best of our knowledge there has been no systematic work on
the excited state properties of the pyrene derivatives, for instance,
the effect of anchoring groups, the p-spacer, aggregation behavior,
etc. Here, we are interested in examining the effect of various
anchoring groups and studying how they affect/improve the
characteristics of the pyrene chromophore. Since, dye anchoring
is one of the vital processes to achieve high power conversion
efficiency, anchoring groups immobilize the sensitizer on the TiO2
surface and create an interfacial path by which electrons can be
transferred from the organic dye to the TiO2 semiconductor; this
process initiates the electrical current in a DSC.37 Therefore, the
Pyrene, pyrene-1-carboxylic acid, pyrene-1-carboxaldehyde, cyano-
acetic acid, rhodanine, rhodanine-3-acetic acid and dimethyl sulf-
oxide-d6 (DMSO-d6) were purchased from Sigma-Aldrich. Malonic
acid, glacial acetic acid and piperidine were purchased from Sisco
Research Laboratories (SRL), India. Malononitrile was purchased
nature of the anchoring group to be employed is an important Fig. 1 Structures and photographs of pyrene derivatives.
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2016

View Article Online
PCCP
Paper
from Spectrochem and ammonium acetate and tetrahydrofuran index. Quinine sulphate (fF = 0.5)46 was used as a standard for
were purchased from Qualigens, India and were used without PC, PAC, PMC and PCC, while fluorescein (fF = 0.9)47 was used
further purification unless otherwise noted. The solvents used for as a standard for PMN, PR and PRA.
photophysical studies were purchased from Sigma-Aldrich and were
of HPLC grade. All measurements were performed in an ambient different functionals such as B3LYP,48–50 PBE,50 PBE0,51,52 M0653 and
atmosphere.
o-b97x-D354 with the 6-311+G(d,p) basis set. Among the five func-
Quantum chemical calculations were performed using five
Infrared spectrometer (IR) spectra were recorded on a BRUKER tionals, B3LYP is found to predict the energy gap of the pyrene
VERTEX 70 attenuated total reflection-Fourier transform infrared derivatives reasonably. So, we used the B3LYP functional with the
spectrometer (ATR-FTIR). Nuclear magnetic resonance (NMR) 6-311+G(d,p) basis set for performing the TDDFT55,56 calculations.
spectroscopy was performed on a BRUKER spectrometer operating Harmonic frequency calculations were performed to characterize the
at 400 MHz for 1H and 100 MHz for 13C and collected at ambient stationary points as minima. In order to include the solvent effects,
probe temperature with spectra calibrated using either an internal the self consistent reaction field (SCRF) method was performed using
tetramethylsilane (TMS) standard or a residual protio solvent. polarizable continuum model (PCM) calculations.57,58 In the present
Electron ionization-mass spectra (EI-MS) were acquired using study, a dielectric constant of 7.43 was used to represent the tetra-
a JEOL GC mate-II mass spectrometer photomultiplier tube hydrofuran medium. The ground state dipole moment was calculated
(PMT) detector. The electronic absorption spectra of samples using the B3LYP functional with the 6-311+G(d,p) basis set. All the
were recorded using an Agilent 8453 UV-visible diode array DFT calculations were performed using Gaussian 09 software.59
spectrophotometer. The fluorescence spectral measurements
FTO glass plates were purchased from BHEL, INDIA. The
were carried out using a Fluoromax-4P spectrophotometer (Horiba photoanode60 was prepared by the following procedure: 0.5 g of TiO2
Jobin Yvon). For fluorescence studies, much diluted solutions were (P25) nanoparticles and 0.016 g of polyethylene glycol were added to
used to avoid spectral distortions due to the inner-filter effect and a solution of 1 M HNO3. Then, 0.21 ml of acetyl acetone and 0.04 ml
emission reabsorption. Time resolved picosecond fluorescence of Triton-X were added to the mixture. Later, the mixture was
decays were obtained by the time-correlated single-photon ultrasonicated for 1 h and then stirred for 24 h to get a homogenous
counting (TCSPC) technique using a microchannel plate photo- TiO2 paste. Before the preparation of electrodes, FTO plates were
multiplier tube (Hamamatsu, R3809U) as a detector and a washed by sonicating subsequently in a soap solution, distilled
femtosecond laser as an excitation source. The second harmonic water, acetone, and 2-propanol for 15 min. For photoanode prepara-
(400 nm) output from the mode-locked femtosecond laser tion, the washed FTO plates were pretreated with 40 mM solution of
(Tsunami, Spectra physics) was used as the excitation source. TiCl4 for 30 min at 70 1C. After TiCl4 treatment, the plates were
The instrument response function for the TCSPC system is washed with water and ethanol and then dried at 100 1C on a hot
B50 ps. The data analysis was carried out by the software plate for 10 min. Then, the TiO2 paste was coated on FTO plates
provided by IBH (DAS-6), which is based on the reconvolution using the doctor blade technique. Then, the plates were sintered at
technique using nonlinear least-squares methods. Femtosecond 450 1C for 30 min to remove the binder, solvent and achieve an
fluorescence transients have been collected using the fluorescence electrically-connected network of TiO2 particles by sintering in a
up-conversion technique. In our femtosecond up-conversion tubular furnace. After the sintering process, when the temperature of
setup (FOG 100, CDP, Russia), the sample was excited using the plates decreased to 80 1C, they were immersed into the dye
the second harmonic (400 nm) of a mode-locked Ti-sapphire solution in methanol for 24 h. Excess non-adsorbed dye was washed
laser (Tsunami, Spectra physics). The fundamental beam with anhydrous ethanol. The platinum catalyst counter electrode
(800 nm) was frequency doubled in a nonlinear crystal (1 mm was prepared by the deposition of H2PtCl6Á6H2O solution (0.005 M
BBO, y = 251, f = 901) and used for the excitation. The sample in isopropanol) onto FTO glass and then sintering at 450 1C for
was placed inside a 1 mm-thick rotating quartz cell. The 30 min. The DSC devices were fabricated by the following method:
fluorescence emitted from the sample was up-converted in a the Pt cathode was placed on top of the photoanode and was tightly
nonlinear crystal (0.5 mm BBO, y = 381, f = 901) using the clipped together. Then, liquid electrolyte 0.05 M I2/0.5 M LiI/0.5 M
fundamental beam as a gate pulse. The upconverted light was 4-tertbutyl pyridine (TBP) in 3-methoxypropionitrile was injected in
dispersed in a monochromator and detected using photon between the two electrodes. The photocurrent–voltage characteristics
counting electronics. The instrument response function of the and electrochemical impedance spectra (EIS) of the DSCs were
apparatus was 300 fs. The femtosecond fluorescence decays measured under 85 mW cmÀ2 light illumination by using an
were fitted using a Gaussian shape for the excitation pulse. AUTOLAB12/FRA2 electrochemical analyzer. The impedance spectra
The femtosecond fluorescent decay was analyzed by fixing the were recorded in the frequency range from 100 kHz to 1 Hz at their
longer lifetime component obtained from TCSPC.
The fluorescence quantum yield (fF) of pyrene derivatives44,45
was calculated by using eqn (1)
open circuit potential (OCP).
fF = (AR/AS) (IS/IR) (ZS/ZR)2 fR
(1)
Results and discussion
Molecular design and synthesis
where, subscripts S and R refer to the sample and reference,
respectively. A is the absorbance at the excitation wavelength, A series of systematically tailored pyrene derivatives were synthe-
I is the integrated emission area, and Z is the solvent refraction sized by Knoevenagel condensation of pyrene-1-carboxaldehyde
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
Table 1 Energy gap (in eV) of frontier molecular orbitals of pyrene
derivatives obtained at various levels of theory using the 6-311+G(d,p)
basis set in vacuum
with the corresponding active methylene group containing
compounds in the presence of acetic acid and ammonium
acetate as reported elsewhere.61 The final products were
obtained in good yields and successfully characterized by IR,
NMR and mass spectroscopic techniques. The synthetic route
and characterization data (Fig. S1–S21, ESI†) are described in
the ESI† in detail. Taking pyrene (P) and pyrene-1-carboxylic
acid (PC) as prototypes, a variety of pyrene-based sensitizers
were designed and obtained by incorporating an olefinic bridge
(p-bridge which connects the donor and acceptor) with different
electron acceptor groups, such as acrylic acid, malonic acid,
cyanoacrylic acid, malononitrile, rhodanine and rhodanine-3-
acetic acid to tune the HOMO–LUMO energy gap and thus the
optical properties of pyrene derivatives. The structures and
photographs of the synthesized pyrene-sensitizers are depicted
in Fig. 1, which clearly reveal that upon incorporation of strong
electron dragging moieties to the pyrene chromophore, the
colour of the pyrene derivatives is remarkably intensified, as
expected, which is due to the reduced HOMO–LUMO band gaps.
From Fig. 1, one can easily understand the impact of anchoring
groups on the pyrene chromophore.
Dyes
B3LYP
PBE
PBE0
M06
o-B97X-D3
Expt.a
P
PC
PAC
PMC
PCC
PMN
PR
3.81
3.52
3.27
3.29
3.05
3.01
2.87
2.86
2.61
2.34
2.10
2.06
1.90
1.88
1.73
1.72
4.14
3.84
3.60
3.69
3.37
3.32
3.17
3.16
4.20
3.93
3.69
3.69
3.46
3.42
3.31
3.30
7.23
6.95
6.75
6.79
6.51
6.45
6.32
6.34
3.34
3.24
3.07
3.07
2.78
2.63
2.59
2.55
PRA
a
Measured in THF.
It clearly shows that the anchoring groups play a vital role in
establishing the low energy transition associated with the raised
HOMO and the lowered LUMO. Indeed, it is an essential
prerequisite for a dye molecule with effective power conversion
efficiency and sensitization properties.
From Fig. 2, it is clear that both the HOMO and LUMO are
evenly localized on the entire molecule. However, the anchoring
groups are found to play a little role in stabilizing the HOMO
and LUMO. It is interesting to look into FMOs of PRA where one
can notice that the –CH2–COOH group takes part in neither the
HOMO nor the LUMO. But in PMC, the –CH–(COOH)2 group
plays a vital role in stabilizing both the HOMO and LUMO.
Though the pyrene moiety invariably contributes to the HOMO
and LUMO, its contribution is diminished towards the LUMO.
For instance, in PMN, the participation of the dicyano sub-
stituted vinyl moiety stabilizes the LUMO significantly compared
to the pyrene unit. Similarly in the case of PR and PRA, the
Theoretical investigations
Detailed density functional theory (DFT) and time dependent
DFT (TD-DFT) calculations were performed to investigate the
effect of anchoring groups on the molecular geometry and
electron density distribution of pyrene derivatives.62–64 In order
to find a reliable functional for predicting the energy gap (Es) of
the pyrene derivatives, B3LYP, PBE, PBE0, M06 and o-B97X-D3
functionals were employed. The choice of the best functional
was based on the comparison of experimentally determined
energy gaps. Energy gap values of pyrene derivatives are presented
in Table 1. Different functionals are found to have a greater
influence on the calculated energy gap of the pyrene derivatives.
It is observed that the long-range corrected o-B97X-D3 functional
is found to overestimate the energy gap of the pyrene derivatives
when compared with the experimentally predicted energy gaps.
The energy gap of the pyrene derivatives is slightly overestimated
when the exchange–correlation hybrid functional PBE0 and meta
hybrid generalized gradient approximated (GGA) M06 functionals
were employed. Among the two hybrid functionals (B3LYP and
PBE), it is clearly observed that the PBE functional is found to
underestimate the energy gap values, whereas the B3LYP functional
predicts energy gaps closer to the experimentally predicted values
with an average difference of 0.29 eV. So the entire manuscript
discussion is based on the results obtained at the B3LYP/6-
311+G(d,p) level as well as from TDDFT calculations.
Molecular orbital analysis
The optimized structures, along with HOMO and LUMO levels
are summarized in Fig. 2. It can be clearly seen from Table 2
that connecting various anchoring groups on the pyrene chromo-
phore eventually reduces the energy gap of pyrene from 3.81 eV
to 2.86 eV. The relative order of our computed energy gap (Es)
values is found to have excellent agreement with that of the
Fig. 2 Optimized structures and electron distributions at HOMO/LUMO
experimentally calculated energy gap (Es) values (Table 2). levels of the pyrene derivatives.
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2016

View Article Online
PCCP
Paper
Table 2 Photophysical properties of pyrene derivatives in THF medium (derived from both #experimental and *computational methods)
Pyrene derivatives
Parameters
P
PC
PAC
PMC
PCC
PMN
PR
PRA
#Expt. labs (nm)
*TDDFT labs (nm)
#log e
336
345
4.85
393
377
3.34
—
—
353
381
4.42
406
413
3.24
—
—
—
—
0.3
367
422
4.48
420
421
3.07
—
—
—
—
0.5
367
436
4.18
420
436
3.07
—
—
—
—
0.5
383
471
4.20
479
471
2.78
1.42
À1.36
À0.86
À1.92
0.004
o0.05
À6.04
À2.99
3.05
438
478
4.36
501
478
2.63
1.46
À1.17
À0.67
À1.96
0.003
o0.05
À6.20
À3.19
3.01
442
499
4.56
513
537
2.59
1.30
À1.29
À0.79
À1.80
0.002
o0.05
À5.92
À3.05
2.87
452
503
4.54
517
540
2.55
1.30
À1.25
À0.75
À1.80
0.002
o0.05
À5.89
À3.03
2.86
#lemi (nm)
*lemi (nm)
#Es (eV)
#Eox (V)
#Eox* (V)
#DGinj (eV)
#DGrec (eV)
#fF
—
—
0.75a
10
#t (ns)
7.9
3.0
3.0
*HOMO (eV)
*LUMO (eV)
*Es (eV)
*Transition
*% of transition
*f
À5.67
À1.86
3.81
p - p*
92.5
0.41
0.61
À5.96
À2.44
3.52
p - p*
95.2
0.51
0.69
À5.84
À2.57
3.27
p - p*
98.0
0.61
0.75
À5.90
À2.61
3.29
p - p*
95.2
0.43
0.63
p - p* + ICT
98.0
0.47
p - p* + ICT
98.0
0.48
p - p* + ICT
98.0
0.93
p - p* + ICT
98.0
0.97
*LHE
0.66
0.67
0.88
0.89
a
From ref. 44.
presence of the rhodanine group stabilizes the LUMO effectively feature to enhance the effective electronic coupling of the
through the vinyl bridge which favours significant charge excited state with the wave function of the TiO2 conduction
transfer along with the p - p* transition. The computed band. In the case of PCC, PMN, PR and PRA, there is a
results show that the HOMO and LUMO of all the molecules significant electron density on the acceptor moieties and this
are p and p* orbitals, respectively. This gives some information property shows that these molecules are effective candidates for
on their electronic transitions which mainly originate from DSCs. In addition, the ELUMO values (Scheme 1) of pyrene
the p - p* transition along with the charge-transfer (CT) derivatives were located above the conduction band edge (Ecb)
character. This shows that P, PC, PAC and PMC molecules of TiO2 (À4.0 eV vs. vacuum), while the EHOMO values were
possess the p - p* transition, whereas PCC, PMN, PR and PRA below the energy of the redox species of iodide/triiodide
molecules show p - p* transition along with intramolecular (À4.8 eV vs. vacuum).65 From these values, we have concluded
charge transfer (ICT).
that the relative matching of electronic levels of sensitizers
It is noteworthy to analyze and understand the origin of the would lead to energetically favorable electron injection as well
individual contributions of various segments of the dye mole- as regeneration of oxidized dyes during DSC operation.
cules towards the HOMO and LUMO levels. So, the Chemissian
program was used to compute the individual contributions
of various fragments of the dyes (Table S1, ESI†). The whole
Computed absorption spectra of pyrene derivatives
In order to rationalize the absorption spectra of the pyrene
derivatives in THF continuum, we employed TDDFT calculations
molecule was segmented into two fragments, namely the
pyrene unit and the anchoring group. From Table S1 (ESI†), it
is clear that the majority of the HOMO electron density is
concentrated on the pyrene unit and the contribution of the
acceptor group increases largely in the LUMO level by changing
the anchoring group from diacetic acid to rhodanine acetic acid
thereby significantly reducing the contribution of the pyrene
unit. It is illustrated that changing the specific substituent on
the pyrene ring facilitates the effective intramolecular charge
transfer to the acceptor group. Therefore, HOMO–LUMO transitions
of the above molecules (PCC, PMN, PR and PRA) bear a significant
ICT nature along with the p–p* transition. In the case of PRA, the
contribution of donor and acceptor groups to the HOMO level is
81% and 19%, respectively, whereas the LUMO is localized on
both donor and acceptor groups (55% and 45%, respectively).
It clearly indicates that the rhodanine acetic acid group in PRA
facilitates effective charge transfer.
In all the above cases, apart from P, there is a significant
electronic density on the anchoring groups. This is a favorable Scheme 1 HOMO/LUMO energy levels of the pyrene derivatives.
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
using four different DFT functionals such as B3LYP, PBE, M06 electronic transitions in pyrene derivatives.66 It is calculated
and o-b97x-D3. Table S2 (ESI†) summarizes the absorption based on the transition density matrices and used to represent
maxima obtained from the different functionals along with the transition density between unoccupied electron transition
experimental absorption lmax values. Among the functionals, orbitals and occupied hole transition orbitals. One should not
the long-range and dispersion corrected o-b97x-D3 functional is confuse them with the occupied and unoccupied molecular
found to underestimate the lmax value compared with the orbitals obtained from simple ground state calculations. The
experimental value. A larger deviation from the experimental natural transition orbitals (NTOs) for pyrene derivatives are
lmax value is found using the PBE functional. Both B3LYP and given in Fig. 3. The NTO corresponds to the absorption bands
M06 functionals were found to predict the lmax value of pyrene of P, PC, PAC and PMC shows a characteristic p - p*
derivatives closer to the experimental values. The absorption transition. In the case of PMN, PCC, PR and PRA, NTOs
bands computed using both the functionals are found to have possess an ICT nature along with the p - p* transition. This
the same trend as that of experimental values. In the present effect is much more pronounced in the PR and PRA molecules.
study, the B3LYP functional is used to discuss the absorption The strong ICT nature of PR and PRA was confirmed through
spectra of pyrene derivatives. So, the calculated absorption the higher electronic NTO contribution of the rhodanine part
maxima, level of electronic transitions, excitation energies (eV) in PR and PRA molecules. It is consistent with the results
and oscillator strengths for the most relevant singlet excited obtained from the FMO analysis. Based on the NTO analysis,
states of pyrene derivatives obtained from the B3LYP functional effective intramolecular charge transfer along with the p - p*
are summarized in Table 2. Moreover, upon moving from simple transition can be expected in the case of PR and PRA mole-
pyrene to pyrene derivatives with different anchoring groups, a cules due to the presence of the strong electron withdrawing
promising bathochromic shift was observed. For instance, the rhodanine group.
dominant absorption band of simple pyrene (P) is found at
345 nm (3.59 eV), whereas for PRA it is found at 503 nm (2.46
Dipole moment
eV). The reduction in the energy gap from P to PRA is also In structural chemistry, the variation of the electronic charge
accompanied by an increase in the oscillator strength. This distribution in the molecules is directly inferred from the
clearly suggests that the oscillator strength of the dye molecules ground state dipole moment values. The electronic charge
is easily tuned by varying the electron pulling strength of the distribution is greatly affected even by a subtle change in the
acceptor moieties. From the molecular orbital analysis, the structure of the molecule.
dominant absorption bands for P to PMC are primarily due to
The calculated ground state dipole moments of the pyrene
p - p* transition. The nature of electronic transition is changed derivatives in both gas phase and THF medium at the B3LYP/6-
when we substitute the strong electron withdrawing groups such 311+G(d,p) level are summarized in Table S3 (ESI†). Due to the
as cyano and rhodanine groups. By visualizing the molecular presence of a moderately polar solvent, the dipole moment
orbitals, the dominant absorption bands of PR and PRA mole- computed in solvent is slightly higher than the dipole moment
cules are due to the intramolecular charge transfer (ICT) along computed in the gas phase. However, in both gas and THF
with p - p* transitions. Interestingly, introducing the rhodanine medium, the trend in the computed dipole moment is similar.
moiety in the pyrene ring not only increases the absorption The polarity of the simple pyrene molecule significantly
maxima but also increases the intensity of ICT transition along increases by the introduction of different electron withdrawing
with the p - p* transition. This is exactly consistent with the groups. The charge separation significantly enhances through
conclusions arrived at from FMO analysis. The computed lmax for the introduction of cyano and rhodanine groups in pyrene
pyrene derivatives with different anchoring groups has the follow- (PCC, PMN, PR and PRA). Among all the molecules, PMN is
ing order: P o PC o PAC o PMC o PCC o PMN o PR o PRA. found to have a large dipole moment value due to the presence
The above order is in good agreement with experimentally pre- of two highly polar cyano groups in PMN (9.4 Debye). As
dicted lmax for pyrene derivatives.
we discussed in earlier sections, effective intramolecular charge
In addition, we have measured theoretical emission spectra transfer can be expected in the case of PMN, PCC, PR and
of pyrene derivatives using the B3LYP functional. We found PRA molecules.
that our computed emission spectra are in good agreement
with the experimental emission spectra of pyrene derivatives
Light harvesting efficiency
(Table 2). The observed trends in the absorbance and emission The light harvesting efficiency (LHE)67 is one of the
spectra of pyrene derivatives suggest that the excitation is important parameters related to the intensity of the absorp-
delocalized throughout the entire p-system of the molecules. tion spectra of the dye molecule and is calculated using
From the above results, one can conclude that substitution of eqn (2),
selective electron withdrawing groups on the pyrene ring
strongly influences the nature of electronic transitions.
LHE = 1 À 10Àf
(2)
where, f is the oscillator strength of the dye molecule. For an
efficient photocurrent response, the LHE of the dye molecule
Natural transition orbital analysis
In addition to the above, we have also performed the natural should be high. The calculated LHE values for S0 - S1 bands
transition orbital analysis (NTO) to confirm the nature of of the dye molecules are given in Table 2. It evidently shows
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2016

View Article Online
PCCP
Paper
donor to the electron acceptor and finally to the conduction
band of TiO2.
Absorption characteristics
The light harvesting efficiency of the dye in the photoanode is
the most important and indispensable factor for the high-
efficiency of a DSC, which is mainly related to the molar
extinction coefficient of the sensitizer, dye-loading capacity in
the electrode and the optical path of the incident light in
the electrode. Consequently, the light harvesting efficiency is
primarily governed by the molar extinction coefficient of the
dye.68 Hence, the UV-vis absorption spectra of the pyrene
derivatives were recorded in THF and are shown in Fig. 4. It
can be seen that by varying anchoring/acceptor groups, the
absorption spectra of pyrene derivatives extended to the visible
region compared with that of unsubstituted pyrene. Intriguingly,
PC, PAC and PMC show a moderate bathochromic shift, whereas
PCC, PMN, PR and PRA exhibit a remarkable bathochromic shift.
These results clearly show that cyano acrylic acid, malononitrile,
rhodanine and rhodanine-3-acetic acid have a stronger electron
accepting tendency than carboxylic acid, acrylic acid and
malonic acid.
Typically, pyrene molecules are very prone to form aggregates
in solution.69 Therefore, we intend to investigate the aggregation
behaviour of the pyrene derivatives in THF solvent. Upon increasing
the concentration of pyrene derivatives, the absorbance increased
linearly and no new bands were observed.
At the same time, Beer–Lambert’s law was obeyed in the
concentration range from 10 mM to 50 mM. This shows that
the pyrene derivatives are not significantly aggregated within
the above mentioned concentration range (spectra are not
shown here) and the molar absorption coefficients (e) of pyrene
derivatives are listed in Table 2. The measured e values are
found to be larger than that of Ru-complex photosensitizers,70
suggesting that these molecules can convert the visible light
energy to electrical energy in an effective manner. The absorption
band appearing in the ultraviolet region corresponds to the p - p*
transition, while the other, covering the visible region, can be
assigned to the intramolecular charge transfer (ICT) process.71
Therefore, we assigned the ICT transition to PCC, PMN, PR and
PRA molecules and the p - p* transition to P, PC, PAC and PMC.
To address the ICT transition in a detailed manner, two
structurally similar candidates such as PMC and PCC were
Fig. 3 Natural transition orbitals (NTOs) of pyrene derivatives shown on
occupied (holes) and unoccupied (electrons) NTO pairs.
that PRA is found to have better LHE properties than the rest considered here. PMC shows an absorption band at 367 nm
of the molecules. However, the LUMO in PRA does not extend and PCC shows a maximum absorption band at 383 nm and a
its carboxylate group that links the dye to the TiO2 surface, moderate band at 415 nm. The latter shows a significant
indicating less electronic coupling between the dye LUMO bathochromic shift compared with PMC, due to the intra-
and the acceptor states of TiO2. In the case of other dye molecular charge transfer between the pyrene donor and the
molecules such as PCC, PMN and PR, conjugation is extended cyanide (–CRN) electron acceptor. It is also important to
to the acceptor moiety. Based on our calculations, PCC, PMN, monitor the color differences between these two molecules;
PR and PRA are found to be excellent candidates for DSCs, PMC shows a light yellow color, whereas PCC possesses an
since they show effective charge separation between the orange color. This is an excellent support for the supposition
HOMO and the LUMO induced by light excitation. In spite that PCC is mainly due to ICT. The presence of ICT in PCC
of this, it is expected that the electrons which are generated is further established by the addition of triethylamine (TEA)
upon photoexcitation of the pyrene derivatives anchored onto to the PCC solution. Based on the earlier reports,72,73 TEA
the TiO2 surface can be successively transferred from the partially deprotonates the carboxylic acid in the ground state.
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
Fig. 5 Absorption spectra of PCC (a) with varying TEA concentrations
(b) in 0.5 mM TEA, in DMF and PMC in DMF.
Fig. 4 UV-vis absorption spectra of pyrene derivatives in THF. (a) p - p*
character (b) ICT character.
As a result, it would diminish the electron-accepting strength of (Fig. 5b). The spectral data clearly indicate that the introduction
the acceptor moiety (ÀCRN), which in turn may significantly of a –CRN group induces the ICT character in PCC which is
decrease the donor–acceptor interaction in the dye, which will sensitive to the surroundings. Furthermore, upon introduction
lead to a hypsochromic shift in the absorption spectrum.74 of one more –CRN group in PCC (i.e. PMN), enhancement
Here, we employed the same strategy to establish the ICT of the intensity of color is observed. Obviously, the dicyano-
character of PCC.
methylene group is a strong electron acceptor and shows an
Upon addition of TEA to the solution of PCC, the absorption intense color due to the strong interfacial charge transfer
band at 415 nm decreases with a simultaneous hypsochromic properties.75 Similarly, we also perceived color changes from
shift at 383 nm(Fig. 5a). The decrease of the ICT band is due to orange to deep-orange on moving from PCC to PMN, this
the deprotonation of carboxylic acid by TEA which leads to the supports the ICT.
decrease in the electron accepting strength of the –CRN
Also, it is found in the literature that the rhodanine76 moiety
group. This new absorption band observed is nearly similar exhibits a stronger electron withdrawing ability with concomitant
to the absorption band of PMC (Fig. 5b). However, in the case of widening of the absorption spectral response compared to
PMC, upon addition of TEA, there is no significant change in cyanoacrylic acid. Accordingly, upon replacing the cyanoacrylic
the absorption spectrum (spectra not shown here). Further, the acid in PCC by the rhodanine moiety as in PR, a significant
formation of hydrogen bonding interaction between the solvent bathochromic shift is observed, which clearly shows that the
and the carboxylic acid group of the dye will also reduce the rhodanine moiety has a strong tendency to accept electrons
acceptor strength of the –CRN group, eventually shifting the from the pyrene chromophore. Further, there is no significant
absorption spectrum to the lower-wavelength region. Hence, we change in the absorption spectra of PR and PRA. In both cases,
envisioned to measure the absorption spectrum of PCC in DMF the LUMO is mainly concentrated on the rhodanine unit,
solvent, where the observed spectral behavior of PCC in DMF is especially on the carbonyl and thiocarbonyl groups. However,
quite similar to the spectrum of PCC in 0.5 mM TEA and PMC in the case of PRA, breaking of the conjugation between the
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2016

View Article Online
PCCP
Paper
rhodanine unit and the carboxylic acid group is instigated by
the presence of the methylene group in rhodanine-3-acetic acid.
Perhaps, the most striking result of the UV-visible absorption
spectral studies is that connecting various anchoring groups
with different electron pulling capacities induces the pyrene
chromophore to absorb at longer wavelengths. Interestingly, the
order of magnitude (104) of molar extinction coefficients does
not vary significantly within the pyrene series. Indeed, among
the pyrene-sensitizer scrutinized, PRA is superior to the other
molecules on the basis of light harvesting ability.
Fluorescence characteristics
To explore the effect of anchoring groups on the pyrene excited
state, the fluorescence spectra of pyrene derivatives were
recorded in THF medium and all the samples were excited at
their absorption maximum (Table 2). Pyrene molecules are very
prone to form an excimer in the excited state,77 however, this
can be avoided by using very dilute solutions. Hence, all the
samples were measured under highly dilute conditions (10À6 M).
The normalized fluorescence spectra of pyrene derivatives are
shown in Fig. 6. Similar to absorption spectra, a bathochromic
shift was observed in fluorescence spectral studies. The observed
bathochromic shift in the absorbance and fluorescence spectra of
pyrene derivatives suggest that the excitation is delocalized
throughout the entire p-system of the molecules. Moreover,
besides the p - p* character, the pyrene derivatives with the
ICT character display a weak fluorescence in solution. It is well
known that the ICT is a competitive relaxation process of the
singlet excited state and typically reduces the fluorescence.78
In the p - p* character dyes, an interesting behaviour in the
fluorescence spectra of PAC and PMC was observed. Both
molecules show very similar fluorescence spectra with two
Fig. 6 Fluorescence spectra of pyrene derivatives in THF. (a) p - p*
character (b) ICT character.
pronounced peaks at 420 nm and 438 nm. To determine whether character in the excited state, we have selected two ideal
these candidates have dual emission properties or not, we have candidates such as PMC and PCC. On close examination of
measured the excitation spectra of these dyes. The excitation the fluorescence properties of these molecules (Fig. 7a), PMC
spectra at both emission wavelengths are almost the same, which shows a strong fluorescence and PCC shows a very weak
confirms that these two peaks arise from the same electronic fluorescence in equiabsorbing solutions. Moreover, the fluorescence
excited state. Notably, the obtained excitation spectra are quite quantum yields were dropped from 0.5 (PMC) to 0.001 (PCC).
similar to the absorption spectra (Fig. S22, ESI†). In addition, we Fluorescence quenching of PCC is due to ICT (cf. above). This
have recorded the fluorescence spectra of PAC and PMC in DMSO phenomenon is further verified by using time resolved fluorescence
(high viscous solvent) and observed a single emission peak at spectroscopy. Fig. 7b shows the time resolved fluorescence
438 nm and this indeed indicates that the observed peaks in THF decay of PCC and PMC in THF solution. The fluorescence decay
(low viscous solvent) are not due to dual emission (Fig. S23, ESI†). of PMC is fitted with a single exponential function with a
In addition, the time resolved fluorescence measurements were lifetime of 3 ns, however, the decay of PCC is almost as fast as
also performed to verify the dual emission properties of these our laser instrument response (50 ps).
molecules. The emission monitored at both wavelengths gave
However, the same dye (PCC) measured in DMF shows
similar decays, which evidently preclude the question of dual single exponential fluorescence decay; because the hydrogen
emission properties. Moreover, both the molecules show very bonding of the carboxylic acid group of the dye with the solvent
similar single exponential [F(t) = Aexp(Àt/t)] fluorescence decays reduces the acceptor strength of the ÀCRN group, the ICT
with a lifetime of 3 ns and they exhibit similar fluorescence character of PCC is diminished. It is also interesting to note
quantum yields; it clearly suggest that the additional carboxyl that the fluorescence decay of PCC and PMC in DMF is very
group does not influence the excited state character noticeably similar. Indeed, this indicates that the ICT character of the PCC
relative to that of PAC.
dye completely disappears in DMF solvent. The other ICT dyes
As mentioned earlier, dyes with strong electron withdrawing namely PMN, PR and PRA also exhibit weak fluorescence
groups (PCC, PMN, PR and PRA) show very weak fluorescence in solution. The fluorescence decays of these dyes are also
in solution, due to ICT. Further, to demonstrate the ICT fast (o50 ps), and no fitting could be obtained (Fig. 8a).
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
Fig. 8 (a) Time resolved fluorescence decay of pyrene derivatives in THF.
(b) Femtosecond fluorescence decay of ICT dyes. (lex = 400 nm).
Fig. 7 (a) Fluorescence spectra of PCC and PMC in THF at equiabsorbing
solutions. (b) Time resolved fluorescence decay of PCC and PMC.
(lex = 400 nm).
Table 3 Fitting parameters of pyrene derivatives in THF
Dyes
t1 (ps)
t2 (ps)
A1 (%)
A2 (%)
tav (ps)
PCC
PMN
PR
1.3
2.8
0.96
1.3
22.4
16.63
18.7
28
27
49
48
72
73
51
52
16.4
12.8
10
Hence, a femtosecond-fluorescence setup was used to extract
the lifetime of the ICT dyes. Fig. 8b displays the femtosecond
(fs) fluorescence decays of ICT dyes in THF medium, measured
in a 60 ps time window upon excitation at 400 nm. The
fluorescence decay of these dyes is recorded at their emission
PRA
23.08
12.6
To thermodynamically judge the possibility of electron
maximum. The obtained fluorescence decays were fitted with a transfer from the excited dye to the conduction band of TiO2,
biexponential function and the fitted values are collected in cyclic voltammetry and differential pulse voltammetry (DPV)
Table 3. The ultrafast decay component, t1 is attributed to the measurements were performed to determine the oxidation
solvation relaxation of the excited states and t2 is due to the potentials of the dyes (Fig. S24, ESI†). The measurements were
lifetime of the relaxed charge transfer state. The average lifetime carried out using a typical three-electrode electrochemical cell,
of the ICT dyes is about B10 ps. For an efficient device, the which consists of a glassy carbon electrode which is used as the
requirement of an electron injection process at the photoanode working electrode, while platinum wire and Ag/AgCl were used
(i.e. dye to TiO2) should be very fast (10 fs to 150 fs) relative to as the counter and reference electrodes, respectively. All of the
the excited state decay to the ground state (10 ns). For instance, dyes show an oxidation wave which is attributable to the
the electron injection process of the most successful N3 dye79 removal of an electron from the pyrene segment. The exact
occurred between 10 to 150 fs time scales, which are faster than the oxidation potentials of the pyrene derivatives were calculated
excited state lifetime decay to the ground state (60 ns).70 Hence, using the DPV method and are listed in Table 2. As shown in
the observed excited state lifetime for pyrene derivatives hope- this table, the first oxidation potentials (Eox) correspond to the
fully favours the electron injection process at the TiO2 surface, HOMO levels of the dyes, which are more positive than the
before they relax to the ground state.
redox potential of the (I/I3)À redox couple (0.4 V), indicating
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2016

View Article Online
PCCP
Paper
that reduction of the oxidized dyes with IÀ ions is thermo- we prepare the dye/TiO2 films under identical conditions we
dynamically feasible. On the other hand, the excited state reproduce a similar surface coverage. Now, we turn back to the
oxidation potentials (Eox*), which correspond to the LUMO main objective of this section, i.e., the effect of solvents on the
levels of the dyes can be obtained using eqn (3):
dye/TiO2 interface. Fig. S26 (ESI†) shows the UV-vis absorption
spectra of PRA and PRA/TiO2 films in various solvents. The
absorption spectra of PRA do not vary too much by varying
solvents. However, there is a significant effect on the adsorption
process. The absorbance of the PRA/TiO2 films in methanol
solvent is very high compared with DMF and THF solvents.
Moreover, it is interesting that a large amount of PRA is
adsorbed on the TiO2 surface in methanol solvent. Thus, the
protic solvents strongly influence the adsorption of the PRA on
the TiO2 surface. This result, clearly suggests that the selection
of appropriate solvents plays a key role in improving the
photoanode of a cell. Hence, we have selected methanol as a
suitable solvent for making the photoanode, i.e., dye/TiO2 films.
Moreover, the fluorescence emission band disappears when
the dye is adsorbed onto a TiO2 film. This phenomenon can be
explained by the efficient electron injection from the excited singlet
state of the dye to the conduction band of TiO2 and suggests that
both dyes (PCC and PRA) are good sensitizers on TiO2 films.
Eox* = Eox À E(0–0)
(3)
where, E(0–0) is the excited state energy of the dye, which is
estimated from the intersection between the normalized
absorption and emission spectra. The obtained Eox* of the
dyes are more negative than the conduction band potential of
TiO2 (À0.5 V vs. NHE), indicating that the electron injection
process from the excited dyes to the TiO2 conduction band is
energetically favourable.
Driving forces for electron injection (DGinj) from the dye
excited state to the CB of TiO2 and electron recombination
(DGrec) from the CB of TiO2 to the dye radical cation (Eox) are
determined from the following equations and the values are
shown in Table 2.
DGinj = Eox* À ECB
DGrec = ECB À Eox
(4)
(5)
Photovoltaic performance
The obtained DG values are negative, which indicates that
both of the processes are thermodynamically feasible and most
importantly injection is the dominant process compared to
recombination for these dyes.
The main motivation of this work is to perceive how anchoring
groups influence the ground and excited state properties of the
pyrene chromophore. In addition, it is worth pursuing this molecule
at the application level. Hence, TiO2 sensitization and I–V measure-
ments are carried out. The results are presented below.
In order to investigate the impact of anchoring groups on solar
cell performance, we have preferred PCC and PRA dyes, since
these two candidates adsorb well on the TiO2 surface (Fig. S27,
ESI†). The J–V curves of DSCs based on pyrene sensitized
electrodes under simulated AM 1.5 irradiation (85 mW cmÀ2
)
are shown in Fig. 9. The electrolyte was composed of 0.05 M I2/0.5 M
LiI/0.5 M 4-tert-butylpyridine (TBP) in 3-methoxypropionitrile. The
measured open-circuit photovoltage (VOC), short-circuit photocur-
rent density ( JSC), fill factor (FF) and solar-to-electric conversion
efficiencies (Z) are listed in Table 4. PCC and PRA sensitized solar
Dye adsorption in various solvents
According to the literature, dyes in various solvents exhibit cells give short-circuit photocurrent densities ( JSC) of 2.11 and
diverse interactions with the solvents, which could alter the 0.83 mA cmÀ2, open circuit voltages (VOC) of 0.63 and 0.47 V, and
physical and chemical interactions between dyes and the TiO2 fill factors (FF) 0.53 and 0.50, corresponding to the overall
surface.80 Therefore, optimization of the solvents for TiO2 conversion efficiencies (Z) of 0.84% and 0.23%, respectively.
sensitization is very important for obtaining good conversion Evidently, PCC shows better solar cell performance than PRA,
efficiency (Z) in DSCs.81 For this reason, we have selected especially in terms of JSC and VOC. This can be explained on the
methanol, DMF and THF as solvents. It is noticed that, among basis of electron injection and surface coverage properties. The
the synthesized dyes, PCC and PRA adsorb well on the TiO2 LUMO of the PCC is largely populated on the acceptor moiety,
surface. Although, nitrile and imide groups are reported as however for PRA it is mainly located on the rhodanine moiety and
anchors for TiO2.81 However, in our case PMN and PR does not it does not extend its carboxylate group that links the dye to the
adsorb on the TiO2 surface under ideal conditions. PC, PAC and TiO2 surface, indicating a less strong electronic coupling between
PMC molecules absorb in the same region where TiO2 absorbs, the dye LUMO and the acceptor states of TiO2. However, in the
hence, no sensitization is carried out. In order to optimize the case of the PCC dye, conjugation is extended to the carboxylic
solvents for TiO2 sensitization we have chosen the visible light group that leads to efficient electron injection to the TiO2 surface.
active PRA molecule. While optimizing the solvents for TiO2
Electrochemical impedance
sensitization, one has to prepare identical dye/TiO2 films in a
particular solvent. Keeping this in mind, we have prepared The impact of anchoring groups on the same dyes (PCC vs.
three PRA/TiO2 films under identical conditions, i.e., dye PRA) was further elucidated by electrochemical impedance
concentration (2 Â 10À4 M), sensitization time (1 h) and spectroscopy (EIS). EIS is a useful technique to understand
sensitization solvent (MeOH). The UV-vis absorption spectra the important interfacial charge-transfer processes that occur
of all the PRA/TiO2 films show similar absorption, no broadening in solar cells, such as the charge recombination at the TiO2/dye/
and enhancement of the spectra are observed with respect to the electrolyte interface, electron transport in the TÀiO2 electrode,
prepared films (Fig. S25, ESI†). This clearly shows that whenever electron transfer at the counter electrode and I3 transport in
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
are 6.24 and 3.22 ms, respectively. In principle, a longer electron
lifetime indicates improved suppression of back reactions
between the injected electrons and the electrolyte and leads
to improvement of the VOC value, due to the reduced electron
recombination rate.87 Consequently, the observed longer electron
lifetime in the case of PCC could explain its high VOC and also the
higher efficiency as well as more effective suppression of back
reactions compared to PRA. Hence, the EIS results provide
evidence that the acceptor/anchoring unit plays a pivotal role
in the overall efficiency of DSCs.
New insights from the present investigation
Although the pioneering work on the effect of electron with-
drawing anchoring groups on the pyrene chromophore was
reported by Paramaguru and co-workers,88 the striking differences
between the present and previous reports are discussed below. In
their report, the five different anchoring groups were chosen and
the photophysical properties and TiO2 sensitization were studied.
However, in the present work, we have meticulously engineered
the pyrene chromophore using various anchoring groups. Detailed
investigations of the relationship between the dye structures
and photophysical, electrochemical and photovoltaic properties
are described. The following points will clearly emphasise the
novelty of our work.
We have systematically engineered the pyrene chromophore
using anchoring groups with different electron withdrawing
abilities. This kind of systematic approach will give an idea
about ‘how even a subtle change in the anchoring groups
influences the ground and excited state properties of the pyrene
chromophore’.
Fig. 9 J–V characteristics of the PCC and PRA devices.
Table 4 Photovoltaic properties of PCC and PRA devices
Dyes G (mol cmÀ2
)
JSC (mA cmÀ2
)
VOC (V) FF
Z (%) t (ms)
PCC 18.31 Â 10À7 2.11
0.63
0.47
0.53 0.84
0.50 0.23
6.24
3.22
PRA 3.05 Â 10À7
0.83
the electrolyte.82 The charge recombination dynamics between
À
the injected electrons in TiO2 and I3 in electrolyte competes
with the charge collection process and limits both the VOC and
83–85
JSC
.
Thus, electrochemical impedance spectroscopy has
The effect of anchoring groups on the molecular geometry
and electron density distribution of pyrene derivatives is investigated
by quantum chemical calculations using five different functionals.
B3LYP is found to predict the energy gap of the pyrene derivatives
reasonably.
Both computational and spectroscopic studies show that by
varying the anchoring groups a systematic bathochromic shift
in the lmax and lemi of pyrene derivatives was obtained.
The ICT nature of the pyrene derivatives was incontrovertibly
demonstrated by using time resolved fluorescence spectroscopy.
Choosing an appropriate solvent for preparing the photo-
anode of a cell is of prime importance to obtain better power
conversion efficiency. We have demonstrated that methanol is
a suitable solvent for the sensitization process.
been used to study this effect in the frequency domain.
Fig. S28 (ESI†) shows the EIS Nyquist plots for DSCs based on
PCC and PRA. Generally, the largest semicircle in the lower
frequency range in the Nyquist plot represents the interfacial
charge transfer at the TiO2/dye/electrolyte interface,83,84 which
is the charge recombination between injected electrons and the
electrolyte. Hence, the obtained plot in Fig. S28 (ESI†) looks like
a semicircle which is assigned to the charge transfer resistance
at the TiO2/dye/electrolyte interface.
The diameter of the semicircle is higher for PCC and lower
for PRA, which means that the electron transfer resistance for
the unwanted back reaction is larger for PCC than that of PRA.
This may be responsible for the achieved higher VOC and JSC of
PCC. In Fig. S29 (ESI†), the frequency response plot is in the
range 1–100 Hz, which is indicative of the electron recombination
between the electrolyte and TiO2 and is related to the electron
lifetime in the CB of TiO2. The electron lifetime depends on the
Photovoltaic measurements were performed on PCC and
PRA dyes. PCC shows a better power conversion efficiency than
PRA. We investigated the reason behind the inferior performance
of PRA through FMO analysis.
density of charge traps, which is ultimately related to VOC
.
A better photovoltaic performance of PCC is rationalized
by a high surface coverage, efficient electron injection and a
longer electron lifetime as demonstrated by the electrochemical
impedance spectroscopy (EIS) measurements.
Therefore, Bode phase plots (Fig. S29, ESI†) were used to find
out the effective lifetime (teff) of electrons in the conduction
band of TiO2. teff was calculated86 using eqn (6)
Our results suggest that, to achieve better excited state
and photovoltaic properties of pyrene, an olefinic spacer is
teff = 1/2pf
(6)
where, f is the frequency of the corresponding peak in the Bode not adequate. In an extension of this work, we found that
phase plot. The obtained electron lifetimes for PCC and PRA instead of the olefinic spacer, a rigid p-system improves the
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2016

View Article Online
PCCP
Paper
11 M. J. Griffith, K. Sunahara, P. Wagner, K. Wagner, G. Wallace,
D. L. Officer, A. Furube, R. Katoh, S. Mori and A. J. Mozer,
Chem. Commun., 2012, 48, 4145–4162.
stability of the dye as well as excited state properties, which will
be report in a separate paper.
12 L. L. Li and E. W. G. Diau, Chem. Soc. Rev., 2013, 42,
291–304.
Conclusion
13 T. Bessho, S. M. Zakeeruddin, C. Y. Yeh, E. W. G. Diau and
M. Gratzel, Angew. Chem., Int. Ed., 2010, 49, 6646–6649.
14 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 and
M. Gratzel, Science, 2011, 334, 629–634.
15 S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, B. F. E.
Curchod, N. Astani, I. Tavernelli, U. Rothlisberger, M. K.
Nazeeruddin and M. Gratzel, Nature, 2014, 6, 242–247.
16 A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo and H. Pettersson,
Chem. Rev., 2010, 110, 6595–6663.
17 H. N. Tsao, J. Burschka, C. Yi, F. Kessler, M. K. Nazeeruddin
and M. Gratzel, Energy Environ. Sci., 2011, 4, 4921–4924.
18 W. D. Zeng, Y. M. Cao, Y. Bai, Y. H. Wang, Y. S. Shi,
M. Zhang, F. F. Wang, C. Pan and P. Wang, Chem. Mater.,
2010, 22, 1915–1925.
19 A. Harriman, G. Izzet and R. Ziessel, J. Am. Chem. Soc., 2006,
128, 10868–10875.
20 S. W. Yang, A. Elangovan, K. C. Hwang and T. I. Ho, J. Phys.
Chem., 2005, 109, 16628–16635.
21 J. M. Casas-Solvas, J. D. Howgego and A. P. Davis, Org.
Biomol. Chem., 2014, 12, 212–232.
22 T. M. Figueira-Duarte and K. Muellen, Chem. Rev., 2011,
111, 7260–7314.
The objective of this work was to perform a systematic analysis
of the absorption and emission characteristics and to explore
the effect of anchoring groups. To fulfill this goal, we adopted
the pyrene chromophore with various anchoring groups. Our
results reveal that the absorption and emission properties of
pyrene derivatives follow the sequence P o PC o PAC o PMC
o PCC o PMN o PR o PRA, manifesting a clear anchoring
group effect. For instance, the threshold wavelength of the
absorption spectrum of PRA is 535 nm, while for pyrene it is
375 nm. Furthermore, the lifetimes of the first excited states
(S1) for ICT based dyes are estimated to be of the order of
magnitude of 10À12 s, implying that these dyes can inject
electrons into the semiconductor potentially. Upon dye adsorption,
PCC and PRA molecules are found to be well adsorbed on the
TiO2 surface, this reveals that the dyes devised using cyano-
acrylic acid and rhodanine-3-acetic acid may be promising
sensitizers for DSCs. The overall conversion efficiencies of these
dyes are not very high; however, this issue can be circumvented
by realizing improvements in the synthetic route towards these
dyes, such as the judicious introduction of a rigid ring system
between the pyrene core and the anchoring group, this work is
currently underway in our laboratory.
23 G. Drummen, Molecule, 2012, 17, 14067–14090.
24 A. M. Breul, M. D. Hager and U. S. Schubert, Chem. Soc. Rev.,
2013, 42, 5366–5407.
25 J. Zhao, W. Wu, J. Sun and S. Guo, Chem. Soc. Rev., 2013, 42,
5323–5351.
Acknowledgements
A. K., K. S. and M. J. thank Department of Science and Technology,
India for DST-INSPIRE Faculty Award [IFA12-CH-78, IFA12-CH-35
and IFA13-CH-100]. We thank Mr G. Dhinagaran, NCUFP,
Chennai, for his assistance in synthesis.
26 M. E. Ostergaard and P. J. Hrdlicka, Chem. Soc. Rev., 2011,
40, 5771–5788.
¨
27 N. Joong Jeon, J. Lee, J. H. Noh, M. K. Nazeeruddin, M. Gratzel
and S. Seok, J. Am. Chem. Soc., 2013, 135, 19087–19090.
28 H. U. Kim, J. H. Kim, H. Kang, A. C. Grimsdale, B. J. Kim,
S. C. Yoon and D. H. Hwang, ACS Appl. Mater. Interfaces,
2014, 6, 20776–20785.
References
¨
1 B. O’Regan and M. Gratzel, Nature, 1991, 353, 737–740.
2 M. Gratzel, Nature, 2001, 414, 338–344.
29 S. Y. Liu, W. Q. Liu, J. Q. Xu, C. C. Fan, W. F. Fu, J. Ling,
J. Y. Wu, M. M. Shi, A. K. Jen and H. Z. Chen, ACS Appl.
Mater. Interfaces, 2014, 6, 6765–6775.
3 M. Gratzel, Acc. Chem. Res., 2009, 42, 1788–1798.
4 F. Gao, Y. Wang, J. Zhang, D. Shi, M. Wang, R. Humphry-
¨
Baker, P. Wang, S. M. Zakeeruddin and M. Gratzel, Chem.
30 O. Taratula, J. Rochford, P. Piotrowiak, E. Galoppini,
R. A. Carlisle and G. J. Meyer, J. Phys. Chem. B, 2006, 110,
15734–15741.
Commun., 2008, 2635–2637.
5 J. H. Yum, I. Jung, C. Baik, J. Ko, M. K. Nazeeruddin and
¨
M. Gratzel, Energy Environ. Sci., 2009, 2, 100–102.
31 D. Kumar, K. R. Justin Thomas, C. P. Lee and K. C. Ho, Org.
Lett., 2011, 13, 2622–2625.
6 A. Mishra, M. K. Fischer and P. Bauerle, Angew. Chem., Int.
Ed., 2009, 48, 2474–2499.
32 A. Baheti, C. P. Lee, K. R. Justin Thomas and K. C. Ho, Phys.
Chem. Chem. Phys., 2011, 13, 17210–17211.
7 W. Xu, B. Peng, J. Chen, M. Liang and F. Cai, J. Phys. Chem.
C, 2008, 112, 874–880.
33 K. R. Justin thomas, N. Kapoor, C. P. Lee and K. C. Ho,
Chem. – Asian J., 2012, 7, 738–750.
8 D. Patel, N. Bastianon, P. Tongwa, J. Leger, T. Timofeeva
and G. Bartholomew, J. Mater. Chem., 2011, 21, 4242–4250.
34 J. Huang, K. Jiang, C. C. Yu, S. G. Li, L. M. Yang and
Y. L. Song, RSC Adv., 2014, 4, 22181–22185.
´
´
9 M. V. Martınez-Dıaz, G. de la Torre and T. Torres, Chem.
Commun., 2010, 46, 7090–7108.
35 S. S. Li, K. J. Jiang, C. C. Yu, J. H. Huang, L. M. Yanga and
Y. L. Song, New J. Chem., 2014, 38, 4404–4408.
10 H. Imahori, T. Umeyama, K. Kurotobi and Y. Takano, Chem.
Commun., 2012, 48, 4032–4045.
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
¨
36 C. L. Wang, Y. C. Chang, C. M. Lan, C. F. Lo, E. W. G. Diau 63 H. Oru¨cu¨ and N. Acar, Comput. Theor. Chem., 2015, 1056,
and C. Y. Lin, Energy Environ. Sci., 2011, 4, 1788–1795. 11–18.
37 M. Katono, T. Bessho, M. Wielopolski, M. Marszalek, J. E. 64 X. Feng, J. Y. Hu, X. F. Wei, C. Redshaw and T. Yamato,
¨
Moser, R. H. Baker, S. M. Zakeeruddin and M. Gratzel,
J. Mol. Struct., 2015, 1086, 216–222.
J. Phys. Chem. C, 2012, 116, 16876–16884.
65 B. H. Kim and H. S. Freeman, Dyes Pigm., 2013, 96, 313–318.
38 F. Ambrosio, N. Martsinovich and A. Troisi, J. Phys. Chem. 66 R. L. Martin, J. Chem. Phys., 2003, 118, 4775–4776.
¨
67 J. Feng, Y. Jiao, W. Ma, Md. K. Nazeeruddin, M. Gratzel and
Lett., 2012, 3, 1531–1535.
39 S. Karthikeyan and J. Y. Lee, J. Phys. Chem. A, 2013, 117,
10973–10979.
S. Meng, J. Phys. Chem. C, 2013, 117, 3772–3778.
68 D. Kuang, S. Ito, B. Wenger, C. Klein, J. E. Moser,
¨
40 L. Zhang and J. M. Cole, ACS Appl. Mater. Interfaces, 2015, 7,
3427–3455.
R. H. Baker, S. M. Zakeeruddin and M. Gratzel, J. Am. Chem.
Soc., 2006, 128, 4146–4154.
41 J. Wiberg, T. Marinado, D. P. Hagberg, L. Sun, A. Hagfeldt 69 F. M. Winnik, Chem. Rev., 1993, 93, 587–614.
¨
70 M. Gratzel, J. Photochem. Photobiol., C, 2003, 4, 145–153.
and B. Albinsson, J. Phys. Chem. C, 2009, 113, 3881–3886.
42 G. Wu, F. Kong, J. Li, W. Chen, X. Fang, C. Zhang, Q. Chen, 71 Z. R. Grabowski, K. Rotkiewicz and W. Rettig, Chem. Rev.,
X. Zhang and S. Dai, Dyes Pigm., 2013, 99, 653–660. 2003, 103, 3899–4032.
43 Z. Li, M. Smeu, M. A. Ratner and E. Borguet, J. Phys. Chem. 72 A. Baheti, P. Tyagi, K. R. J. Thomas, Y. C. Hsu and J. T. Lin,
C, 2013, 117, 14890–14898. J. Phys. Chem. C, 2009, 113, 8541–8547.
44 O. Taratula, J. Rochford, P. Piotrowiak, E. Galoppini, R. A. 73 B. Venkatachalapathy and P. Ramamurthy, Phys. Chem.
Carlisle and G. J. Meyer, J. Phys. Chem. B, 2006, 110,
15734–15741.
45 D. F. Eaton, Pure Appl. Chem., 1988, 60, 1107–1114.
Chem. Phys., 1999, 1, 2223–2230.
74 A. Baheti, S. R. Gajjela, P. Balaya and K. R. J. Thomas, Dyes
Pigm., 2015, 113, 78–86.
46 K. Suzuki, A. Kobayashi, S. Kaneko, K. Takehira, T. Yoshihara, 75 R. Jono, J. Fujisawa, H. Segawa and K. Yamashita, Phys.
H. Ishida, Y. Shiina, S. Oishic and S. Tobita, Phys. Chem. Chem.
Phys., 2009, 11, 9850–9860.
47 J. Shen and R. D. Snook, Chem. Phys. Lett., 1989, 155, 583–586.
48 A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652.
49 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter
Mater. Phys., 1988, 37, 785–789.
50 J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett.,
1997, 78, 1396.
51 M. Ernzerhof and G. E. Scuseria, J. Chem. Phys., 1999, 110,
5029–5034.
Chem. Chem. Phys., 2013, 15, 18584–18588.
76 T. Marinado, D. P. Hagberg, M. Hedlund, T. Edvinsson,
E. M. J. Johansson, G. Boschloo, H. Rensmo, T. Brinck,
L. Sun and A. Hagfeldt, Phys. Chem. Chem. Phys., 2009, 11,
133–141.
77 K. Kalyanasundaram and J. K. Thomas, J. Am. Chem. Soc.,
1977, 99, 2039–2044.
78 A. Cidlina, V. Novakova, M. Miletina and P. Zimcik, Dalton
Trans., 2015, 44, 6961–6971.
¨
79 O. Bram, A. Cannizzo and M. Chergui, Phys. Chem. Chem.
52 C. Adamo and V. Barone, J. Chem. Phys., 1999, 110, 6158–6169.
Phys., 2012, 14, 7934–7937.
53 Y. Zhao and D. G. Truhlar, Theor. Chem. Acc., 2008, 120, 80 K. M. Lee, V. Suryanarayanan and K. C. Ho, J. Power Sources,
215–241. 2009, 188, 635–641.
54 J.-D. Chai and M. Head-Gordon, Phys. Chem. Chem. Phys., 81 J. H. Mecchia Ortiz, C. Longo and N. E. Katz, Inorg. Chem.
2008, 10, 6615–6620. Commun., 2015, 55, 69–72.
55 E. Runge and E. K. U. Gross, Phys. Rev. Lett., 1984, 52, 82 Q. Wang, J. E. Moser and M. Gratzel, J. Phys. Chem. B, 2005,
997–1000. 109, 14945–14953.
56 A. Dreuw and M. Head-Gordon, Chem. Rev., 2005, 105, 83 N. Koide, A. Islam, Y. Chiba and L. Y. J. Han, J. Photochem.
4009–4037. Photobiol., A, 2006, 182, 296–305.
57 J. Tomasi and M. Persico, Chem. Rev., 1994, 94, 2027–2094. 84 F. Fabregat-Santiago, G. Garcia-Belmonte, I. Mora-Sero and
¨
58 V. Barone, M. Cossi and J. Tomasi, J. Comput. Chem., 1998,
19, 404–417.
59 M. Frisch, et al., Gaussian 09, Revision D. 01, Gaussian, Inc.,
Wallingford, CT, 2010.
60 N. Pavithra, A. Asiri and S. Anandan, J. Power Sources, 2015,
286, 346–353.
61 W. Xu, B. Peng, J. Chen, M. Liang and F. Cai, J. Phys. Chem.
C, 2008, 112, 874–880.
J. Bisquert, Phys. Chem. Chem. Phys., 2011, 13, 9083–9118.
85 K. Pei, Y. Z. Wu, A. Islam, Q. Zhang, L. Y. Han, H. Tian and
W. H. Zhu, ACS Appl. Mater. Interfaces, 2013, 5, 4986–4995.
86 D. D. Babu, S. R. Gachumale, S. Anandan and A. V. Adhikari,
Dyes Pigm., 2015, 112, 183–191.
87 D. Kuang, S. Uchida, R. Humphry-Baker, S. M. Zakeeruddin
and M. Gratzel, Angew. Chem., Int. Ed., 2008, 47, 1923–1927.
88 G. Paramaguru, R. Vijay Solomon, S. Jagadeeswari,
P. Venuvanalingam and R. Renganathan, J. Photochem.
Photobiol., A, 2013, 271, 31–44.
62 S. K. Pal, V. Sundstrom, E. Galoppini and P. Persson, Dalton
Trans., 2009, 10021–10031.
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2016