Effect of mono- and di-anchoring dyes based on o,m-difluoro substituted phenylene spacer in liquid and solid state dye sensitized solar cells

Article abstract of DOI:10.1016/j.dyepig.2019.108021

Novel mono- and di-anchoring organic dyes have been designed and synthesized with o,m-difluoro substituted phenylene spacer and were tested for DSSCs in presence of solid-state (SJE-4) as well as liquid (BMII) electrolytes. The new and simple structures o

Full text of DOI:10.1016/j.dyepig.2019.108021

Dyes and Pigments xxx (xxxx) xxx
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: http://www.elsevier.com/locate/dyepig
Effect of mono- and di-anchoring dyes based on o,m-difluoro substituted
phenylene spacer in liquid and solid state dye sensitized solar cells
,
,
c
,
,
,**
Telugu Bhim Rajua 1, Jayraj V. Vaghasiyab 1, Mohammad Adil Afroza 1, Saurabh S. Sonib
,
Parameswar Krishnan Iyera
,d,*
a Department of Chemistry, Indian Institute of Technology Guwahati, Guwahati, 781039, Assam, India
b Department of Chemistry, Sardar Patel University, Vallabh Vidyanagar, 388 120, Gujarat, India
c Department of Materials Science and Engineering, National University of Singapore, 9 Engineering Drive 1, 117575, Singapore
d Center for Nanotechnology, Indian Institute of Technology Guwahati, Guwahati, 781039, Assam, India
A R T I C L E I N F O
A B S T R A C T
Keywords:
Novel mono- and di-anchoring organic dyes have been designed and synthesized with o,m-difluoro substituted
phenylene spacer and were tested for DSSCs in presence of solid-state (SJE-4) as well as liquid (BMII) electro-
lytes. The new and simple structures of Cz-D1 and Cz-D2 dyes have same carbazole donor unit with either one or
Mono- and di-anchor
o,m-difluoro
Solid state electrolyte
Dye sensitized solar cell
Stability
two side substitution of o,m-difluoro substituted phenylene π-spacer and cyanoacrylic acid acceptor. The pho-
tophysical and electrochemical properties of photosensitizes are investigated in detail and correlated with the
solar cell performance. The Cz-D2 dye have ~20% higher device efficiency than Cz-D1 due to its lower LUMO
level, and presence of two acceptor groups which provide efficient electron extraction from carbazole donor,
lesser aggregation, high molar extinction coefficient and better charge transfer. Without using any additive, Cz-
D2 exhibited an attractive power conversion efficiency (PCE) of 5.35% (Jsc ¼ 10.38 mA/cm2, Voc ¼ 0.75 V and
FF ¼ 0.60) in presence of iodide redox electrolyte. These dyes exhibit comparatively less efficiency when used to
fabricate a solid state dye sensitized solar cell in presence of SJE-4 electrolyte due to aggregation between
fluorine substituted phenylene spacer and electrolyte through strong H-bonding that might cause more electron
recombination/back electron transfer. Good stability was also observed for both the dyes.
1. Introduction
dye molecules being used as a sensitizer, capped with a counter elec-
trode and filled with electrolyte system in between the electrodes
To resolve the energy crisis and global environment issues alternate
renewable energy sources are required to be developed [1–3].
Dye-sensitized solar cells (DSSCs) are an attractive alternative to con-
ventional solar cell, because of their reasonable solar to electricity
conversion efficiency. Due to the high production cost of silicon solar
cells, scientists are focusing on organic-materials based solar cells to
replace the traditional inorganic semiconductor technologies [4–23].
Since its discovery in 1991, DSSC is fast growing technology because of
low material cost, low production cost, facile fabrication process,
reasonable power conversion efficiency (PCE) and less environmental
issues [4,5,8–20]. Typically, DSSC is an electrochemical cell consisting
of a transparent conductive glass sheet (ITO/FTO) coated with a thin
film of mesoporous nano-crystalline TiO2 as an anode and monolayer of
[8–20]. Extensive efforts to improve every component of DSSC to ach-
ieve higher PCE have been made over the years. To date, highest PCE of
>10% has been achieved with porphyrin and Ru based dyes in presence
of liquid/solid electrolyte [9–13]. However, for the large scale appli-
cation, these dyes are not favorable because of the high cost of “Ru”
metal, tedious separation process, low molar-extinction coefficients and
environmental issues [22]. Recently, scientists have paid attention on
metal-free organic dyes due to their low cost, simple preparation, easy
purification, high molar-extinction coefficients and environmental
friendliness. The advantage of structural diversity in metal-free organic
dyes allows us to easily tune optical and electrochemical properties.
Although a decent PCE of ~14% has been achieved with metal-free dyes
[24], investigation of novel organic dyes is still challenging and
* Corresponding author. Department of Chemistry, Indian Institute of Technology Guwahati, Guwahati 781039, Assam, India.
** Corresponding author. Department of Chemistry, Sardar Patel University, Vallabh Vidyanagar, 388 120, Gujarat, India.
E-mail addresses: soni_b21@yahoo.co.in (S.S. Soni), pki@iitg.ac.in (P.K. Iyer).
1
Equal Contribution.
https://doi.org/10.1016/j.dyepig.2019.108021
Received 31 August 2019; Received in revised form 24 October 2019; Accepted 4 November 2019
Available online 9 November 2019
0143-7208/© 2019 Elsevier Ltd. All rights reserved.
Please cite this article as: Telugu Bhim Raju, Dyes and Pigments, https://doi.org/10.1016/j.dyepig.2019.108021

T.B. Raju et al.
Dyes and Pigments xxx (xxxx) xxx
desirable to overcome the energy crisis.
charge transfer (ICT) from D to A. Compare to Cz-D1, di-anchor dye Cz-
D2 exhibited strong ICT peak as both acceptor units help in efficient
electron extraction from carbazole donor and reduces H-aggregation in
solution and solid state. Cz-D1 has strong aggregation in CHCl3 and TiO2
surface due to the close packing, generally observed in mono-anchoring
dyes. Cz-D1 dye has maximum absorption peaks at 239, 288 nm and Cz-
D2 dye has maximum absorption peaks at 253, 290 nm. The absorption
spectrum of Cz-D2 dye is slightly red shifted in solution as well as in
The PCE and stability of a DSSC can be enhanced by controlling the
light harvesting, charge recombination, dye regeneration, and electron
injection properties of the dye [25] and these factors are strongly
associated with the dye structure. Planar units with large conjugation
were utilized in the dye structure to improve its light harvesting ability
and achieve high absorption [26,27]. Charge recombination with the
electrolyte was curtailed by incorporating bulky group or alkyl chains on
the dye molecule which prevents the direct contact of electrolyte with
the TiO2 surface [28]. Co-adsorbents such as deoxycholic acid (DCA)
and chenodeoxycholic acid (CDCA) are being used as anti-aggregation
agents to suppress dye aggregation over the surface of TiO2 [29,30].
Considering these points, new metal-free organic sensitizers can be
designed to achieve high efficiency and stability in DSSCs.
solid state due to increased effective π-conjugation length and reduced
aggregation. The molar absorption coefficient at λmax for Cz-D1 and Cz-
D2 were 22,000 MÀ 1cmÀ 1 (398 nm) and 30,000 MÀ 1cmÀ 1 (401 nm)
respectively. The red shift in the absorption and superior molar ab-
sorption coefficient resulted in better efficiency of Cz-D2.
Herein, two new o,m-di fluoro substituted phenylene spacer dyes
with mono- and di-anchoring groups have been synthesized. The new
and simple structure of Cz-D1 and Cz-D2 dyes have same carbazole
2.2. Electrochemical properties
Energy levels (HOMO-LUMO) of dyes were investigated by cyclic
voltammetry (CV) and the results are shown in Fig. 3 and relevant data
are listed in Table 1. HOMO level of the photosensitizers were calculated
from onset oxidation potential of dyes with Fc/Fcþ as internal standards
(4.8 eV reference energy levels) using EHOMO ¼ À [(Eox- E1/2(ferrocene)) þ
4.8] eV formula. Notably, the HOMO level of Cz-D1 and Cz-D2 corre-
sponding to their first onset oxidation potential (Eox ¼ 1.13 V) in CV, is
at same energy level of À 5.82 eV as carbazole donor unit is present in
both the dyes. The oxidation potential of all photosensitizers versus
normal hydrogen electrode (V vs NHE) is more positive than the redox
potential of electrolytes, indicating that thermodynamically favorable
oxidized photosensitizers can regenerate effectively. The LUMO level of
photosensitizers were calculated from the onset absorption spectra of
dyes (ELUMO ¼ Eg – EHOMO), and are À 3.13, and À 3.27 eV for Cz-D1 and
Cz-D2 dyes, respectively. The LUMO level of Cz-D2 dye is lower than Cz-
D1 dye due to the increased acceptor ability. These values are above the
conduction band (CB) of the TiO2 electrode, which is necessary for
sufficient driving force to inject the electrons from light-oxidized dye to
CB of nanocrystaline TiO2 surface. Based on CV results, Fig. 3b illustrates
the energy level diagram of dyes. It is clear that the LUMO of Cz-D2 dye
matches well with the conduction band of TiO2 and HOMO matches well
with the potential of redox electrolyte (Liquid and SJE-4 electrolyte);
this improves Jsc and results in high PCE.
donor unit, o,m-difluoro substituted phenylene π-spacer and cyanoa-
crylic acid (CAA) as acceptor, and only variation is presence of mono-
and di-anchors, respectively (Fig. 1). Fluorine (F) atom was incorporated
into π–bridge due to its capability to tunes the molecular energy levels
(lowering the LUMO level), and improve the electron mobility [7,
15–17]. Linear octyl chain on carbazole moiety was used for the better
suppression of dye aggregation on metal oxide surface (TiO2). It is clear
from our previous reports that carbazole is the best donor in simple
D-π-A dyes with ortho/meta-F substituted phenylene spacer [22,23].
The planarity of backbone in carbazole based dyes improves charge
transfer from donor(D) to acceptor(A), results in high molar extension
coefficients (ε) and long life time values. The molecular properties of the
newly synthesized dyes were extensively studied and correlated with the
photovoltaic performance for both liquid as well as solid state DSSCs
(ss-DSSCs). Further, electrochemical impedance spectroscopy (EIS)
measurements were used to estimate carrier transport and interfacial
charge recombination in the solar cell devices.
2. Results and discussion
2.1. Optical characterization
Fig. 2 displays the photophysical properties of Cz-D1 and Cz-D2 dyes
in chloroform solution as well as on TiO2 film and the corresponding
characteristic data are summarized in Table 1. Both the dyes exhibited
two strong distinct absorption bands in CHCl3 solution in the range of
2.3. Theoretical calculations of molecular orbital
The electronic distribution of frontier molecular orbitals (HOMO,
LUMO) of dyes were calculated by density functional theory (DFT) using
B3LYP/6-31G (d,p) basis set for optimization of dye molecules. Elec-
tronic transitions of dyes from HOMO to LUMO and other transitions
200–350 nm. These peaks correspond to the localized aromatic π-π*
transitions of conjugated back bone. Cz-D1 exhibited a weak absorption
band in the range of 350–500 nm which is related to the intramolecular
Fig. 1. Synthetic route of organic sensitizers (i) NBS, THF, 12 h, (ii) 2,3-Difluoro-4-formylphenylboronic acid, Pd(PPh3)4, K2CO3, THF:H2O ¼ 2:1, Aliquat, 85 �C, 18 h
(iii) Cyanoacetic acid, Piperidine, CH3CN, 85 �C, 8 h.
2

T.B. Raju et al.
Dyes and Pigments xxx (xxxx) xxx
Fig. 2. Absorption spectra of photosensitizers Cz-D1 and Cz-D2 in CHCl3 solution (a) and on nanocrystaline TiO2 film (b).
photo excited state of dye.
Table 1
Photophysical, electrochemical and life time properties of carbazole based dyes.
2.4. Preparation of electrolytes
Dye
Solution λmax
Film λmax
Ɛmax [MÀ 1
HOMO
LUMO
[nm]a
[nm]b
cmÀ 1
]
[eV]c
[eV]c
The electrolyte contains solid organic ionic conductor (SJE-4, Fig. 5),
iodide source (1-butyl-3-methylimidazolium iodide, BMII), iodine and
TBP (tert-butyl pyridine). The molar proportion of composition was
usually 0.03(SJE-4): 0.04 (BMII): 0.05 (I2): 0.5 (TBP). All components
were mixed in acetonitrile solvent.
Cz-
239, 288, 398
417
22,000
À 5.82
À 5.82
À 3.13
À 3.27
D1
Cz-
253,290, 401
453
30,000
D2
a
b
Maximum absorption wavelength of dyes in chloroform solution.
Maximum absorption wavelength of dyes that were absorbed onto the TiO2
films.
2.5. Fabrication of DSSCs
c
Onset potentials from CV measurements of thin films in 0.1 M Bu4NPF6/
ox
CH3CN solution, estimated from EHOMO ¼ -{(E -E1/2(ferrocene)) þ 4.8} eV and
on
The preparation of TiO2 photoelectrode was done as per our previ-
ously reported method [22,23]. The dye solar cells were assembled by
placing a Pt-coated FTO (CEs) on the (Cz-D1 & Cz-D2) dyes sensitized
TiO2 photo-electrode using a hot-melt Surlyn film. A solution of elec-
trolytes in acetonitrile was injected into the interior space of the device
from the predrilled holes on the Pt-based CEs and sealed using epoxy
resin. In solid state devices the same procedure was followed and after
filling the electrolyte, devices were kept in a vacuum oven at 50 �C for at
least 6 h to remove the organic solvent. The same process was repeated
three times to ensure that the TiO2 porous film was filled with a compact
layer of the SJE-4. The active area of all DSSC devices was 0.20 cm2.
ELUMO ¼ Eg calculated using onset of UV (TiO2 film)–EHOMO
.
were also ascertained by TD-DFT calculations performed by CAM-
B3LYP/6-31G (d,p) basis set. The results of computational analysis are
presented in Fig. 4 and Table 2. This analysis reveals that increasing
π
-spacer units and anchoring groups (mono and di anchors) results in red
shift of absorption spectra and modifies the energy levels. The perfor-
mance of dyes depends on planarity of back bone and charge transfer
from D to A. Moreover, the acceptor moiety and length of alkyl chain
also affects the dye performance. In Cz-D1 and Cz-D2 dyes HOMO levels
are mainly located on donor segment and strongly distributed into
spacer moiety whereas the LUMO levels are mainly located on acceptor
and spacer unit. These results reveal that the better performance of Cz-
D2 dye might be due to the well separated HOMO and LUMO orbitals on
the donor and acceptor parts, respectively. Proper separation of HOMO
and LUMO orbitals is an advantage for charge migration from D to A in
2.6. Photovoltaic characterization of DSSCs
The DSSCs were fabricated using Cz-D1 and Cz-D2 sensitizers with
nanocrystalline TiO2 anatase semiconductor as the photoanode. The
DSSCs device performance parameters, such as short-circuit current
Fig. 3. (a) Cyclic voltammograms and, (b) Energy level diagram of Photosensitizers compared to the energy levels of TiO2 and electrolytes.
3

T.B. Raju et al.
Dyes and Pigments xxx (xxxx) xxx
Fig. 5. Structure of solid organic ionic conductor, SJE-4 (where carbon in
green, nitrogen in blue, iodide in purple, hydrogen in grey and oxygen in red)
[33]. (For interpretation of the references to colour in this figure legend, the
reader is referred to the Web version of this article.)
Table 3
Photovoltaic parameters of DSSCs with solid and liquid electrolytes.
Photovoltaic
parameters
Cz-D1
Cz-D2
A
B
A
B
J
sc (mA cm¡2
)
7.78
6.22 �(0.20)
10.2 � (0.18)
8.74 � (0.25)
�(0.12)
0.690
�(0.05)
58.0
V
oc (V)
0.658
0.707 � (0.05)
59.2 � (1.0)
0.691�
�(0.06)
55.0 �(1.2)
(0.10)
FF (%)
55.0 � (1.5)
�(1.0)
3.95
PCE (%)
2.60 � (0.22)
5.20 � (0.15)
4.02 � (0.30)
�(0.20)
aPerformances of DSSCs were measured with 0.20 cm2 working area (measure-
ments were performed under AM 1.5, 100 mW/cm2 irradiation). Jsc (mA/cm2):
short-circuit current, Voc (V): open circuit voltage, FF (%): fill factor and PCE
(%): Photo conversion efficiency, A and B indicates liquid-state and solid-state
electrolyte based DSSCs respectively.
group, has a lower Jsc value of 7.78 mA/cm2 (Fig. 6a). As per literature,
the IPCE value is closely related to photocurrent density (Jsc) at a specific
wavenumber (λmax) and power input (Pin) [31].
Fig. 4. Frontier molecular orbital of the designed dyes.
Table 2
Selected transition energies wavelengths, their orbital contribution and oscil-
lator strengths (f) for the dyes calculated using TD-DFT, CAM-B3LYP/6-31G (d,
p) basis set.
Jsc
0ðλÞλðnmÞ
IPCEð%Þ ¼ 1240 �
� 100
I
To confirm the different Jsc results between Cz-D1 and Cz-D2, the
IPCE values which usually depend on charge collection, light harvesting,
and electron injection efficiencies were measured. Fig. 6b shows the
IPCE intensity curves as a function of the incident light wavelength for
the DSSCs based on the new organic sensitizers on nanocrystaline TiO2
photoanode. Both devices can efficiently convert the light into current in
the range of 350 nm–550 nm. The DSSCs containing Cz-D2 sensitizer
based device exhibited a broad and higher IPCE of 66% in the range of
350–650 nm, while the Cz-D1 sensitizer based device shown blue-shift
and lower IPCE of 54% in the range of 350–630 nm. The higher per-
centage of IPCE was observed for the sensitizer Cz-D2, which is due to its
better light harvesting capacity and broad absorption in the visible re-
gion. The changes of IPCE curves are consistent with the trend of the
absorption spectra on the nanocrystaline TiO2 photoanode (Fig. 2b). In
Cz-D2 dye based device the Jsc and Voc are improved upon substitution
of the second o,m-difluoro substituted phenylene spacer and acceptor
Dye
λ/nm
f
Wavefunctions
Cz-D1 Dye
347
279
259
233
219
445
424
383
381
325
318
279
1.3211
0.2239
0.1721
0.1650
0.3614
0.5915
0.1425
0.1333
0.1975
0.1995
0.3377
0.5303
H → L (82%)
H → Lþ1 (37%) H-1 → L (21%)
H-3 → L (42%)
H → Lþ2 (43%)
H → Lþ3 (38%)
Cz-D2 Dye
H → L (98%)
H → Lþ1 (98%)
H-1 → L (92%)
H-1 → Lþ1 (94%)
H-2 → L (63%) H → L (30%)
H-2 → Lþ1 (89%)
H-3 → L (65%)
density (Jsc), fill factor (FF), open-circuit photovoltage (Voc), and power
conversion efficiency (PCE), were measured under 100 mW/cm2 (AM
1.5) illumination, and are depicted in Table 3. The DSSCs based on Cz-
D2, which has bis o,m-difluoro substituted Phenylene spacer and two
anchoring units (-CNCH2COOH), features Jsc and Voc values of 10.2 mA/
cm2 and 0.707 V respectively. The PCE was calculated to be 5.20% using
the following equation: PCE ¼ Pmax/Pin � 100. In difference, the cell
based on Cz-D1, which has single phenylene spacer and –CNCH2COOH
groups. It has a higher PCE because the D-(
based dye has a positive effect on IPCE curve and dye loading on TiO2
photoanode surface ( ) compared with that of Cz-D1 having one phe-
π-A)2 system of carbazole
ε
nylene spacer and acceptor unit. The
ε value of Cz-D1 and Cz-D2 were
0.74 � 10À 6 and 1.23 � 10À 6 molÀ 1 cmÀ 1 respectively. The reason for
higher loading of Cz-D2 is its better electronic coupling with the TiO2
4

T.B. Raju et al.
Dyes and Pigments xxx (xxxx) xxx
trend was also observed in IPCE spectra (ESI, Fig. S1). The enhanced Voc
value (in liquid electrolyte) may be explained in terms of the reduction
in the back electron transfer from the nanocrystalline TiO2 conduction
band to the triiodide ions in the electrolyte [32].
Moreover, liquid electrolyte has easy penetration in porous TiO2
matrix and had better interfacial contact compared to solid electrolyte.
In ss-DSSCs, reduced Jsc was obtained because of low ion transfer via
hopping or grotthuss mechanism rate is slightly lower than liquid based
electrolytes [14,33,34]. Moreover, pore filling and interface connectiv-
ity of solid electrolyte might also be low with TiO2 and counter elec-
trode, leading to a decrease in fill factors and loss of photocurrent. But
the long term stability of solid state electrolytes is superior compared to
liquid electrolyte based DSSCs. Liquid electrolytes have many draw-
backs such as; (i) it contains highly volatile organic solvents, (ii) TiO2
surface bound organic sensitizer after coming in contact with liquid
electrolytes start desorbing, (iii) liquid electrolyte is corrosive in nature,
and (iv) leakage problem, etc. [11,35,36] Therefore, in Fig. 7b, we have
performed the stability analysis of liquid and solid electrolyte based
DSSCs. To show the stability of the SOICs electrolyte, as mentioned in
the introduction (i.e. solvent evaporation or leakage), compared to the
liquid electrolyte, a comparison is carried out for four devices (two de-
vices for each dye) fabricated and sealed in the same procedure, except
one was filled with solid electrolyte and the other with liquid electrolyte.
The PCEs were measured at an interval of 5 days, and after 25 days the
liquid electrolyte based device displayed that most of its electrolyte in
the device was evaporated, whereas the solid electrolytes based device
retained its electrolyte and the PCE was pretty much unchanged. In
Fig. 7b the PCE as a function of time are illustrated for a period of 25
days, it is found that the ss-DSSCs show excellent stability.
The EIS technique was used to study the internal resistance, inter-
facial properties and charge transfer kinetics of nanocrystalline TiO2
layers in DSSCs [37,38]. EIS spectra were measured in dark condition
using electrochemical workstation (CHI660E, CH instruments, USA).
Fig. 8 shows the Nyquist plots of solid and liquid electrolytes based
DSSCs. The EIS spectra can be interpreted by fitting it using equivalent
circuit (Table 4), with components as explained below. Each equivalent
circuit consisted of several components as shown in Fig. 8 (inset):
Interfacial resistance (Rpt) of electron transfer at Pt-CEs/electrolyte
interface, recombination resistance (Rk) in TiO2 matrix, diffusion resis-
tance (RD) in electrolyte and constant phase element (CPE-P) the value
of CPE-P is associated with the surface roughness and porosity of the
photoelectrode [23,39,40]. Nyquist spectra of the DSSCs showed three
semicircles: 1st high frequency region, represents Rpt; 2nd mid fre-
quency region, represent Rk and 3rd lower frequency region, represent
Fig. 6. (a) J-V characteristics of DSSCs based on the liquid electrolyte (inset; J-
V measured in dark condition) and (b) IPCE spectra and integral current density
were measured under 1 sun (100 mW/cm2, AM 1.5 irradiation) light.
matrix due to the presence of two carboxylic acceptor groups.
Under similar condition we have also fabricated ss-DSSCs with Cz-D1
and Cz-D2 dyes and solid organic ionic conductors (SOICs) SJE-4 and
the measurement was performed at a light intensity of 100 mW/cm2.
The results of ss-DSSCs are depicted in Fig. 7a and the corresponding
photovoltaic parameters are presented in Table 3. A maximum PCE of
4.02% with Jsc of 8.74 mA/cm2 and Voc of 0.691 V were obtained for the
Cz-D2 sensitizer based ss-DSSCs. It is noted that reduced Jsc and Voc
values compared with liquid based electrolyte were obtained, and same
Fig. 7. (a) J-V characteristics of DSSCs based on the solid electrolyte (inset; J-V measured in dark condition) and (b) devices stability curve: PCE as a function of time.
5

T.B. Raju et al.
Dyes and Pigments xxx (xxxx) xxx
3. Conclusion
In conclusion, we have developed o,m-difluoro substituted Phenyl-
ene spacer and carbazole based dyes Cz-D1 and Cz-D2, consisting mono-
and di-anchoring groups. Both dyes exhibited interesting optical, elec-
trochemical and photovoltaic properties. The absorption spectra of
planar Cz-D1 dye show strong H-aggregation in solution and dye
absorbed on TiO2. Cz-D2 contain two electron acceptor groups and non-
planarity ensuing less aggregation, provide efficient electron extraction
from carbazole donor, high molar extinction coefficient, better charge
transfer and lower LUMO level resulted in better photovoltaic perfor-
mance than Cz-D1, in presence of liquid (BMII) as well as solid SJE-4
electrolyte systems. Without using any additives, Cz-D2 and Cz-D1
exhibited attractive power conversion efficiency of 5.35% and 4.15%,
respectively in presence of Iodide redox electrolyte. Both o,m-difluoro
substituted dyes exhibited less efficiency in presence of SJE-4 solid
electrolyte system due to the aggregation between fluorine substituted
phenylene spacer and electrolyte through strong H-bonding and resulted
in more electron recombination/back electron transfer as confirmed by
TD-DFT and EIS measurements. PCE as a function of time for a period of
25 days were also recorded, the mono anchoring Cz-D1 device exhibits
better stability than di anchoring Cz-D2 in presence of both electrolyte
systems.
Fig. 8. Nyquist plots for the fabricated DSSCs based on solid and liquid elec-
trolytes with Cz-D1 and Cz-D2 sensitizers in the dark at Voc (0.65 V) in a fre-
quency range from 0.1 Hz to 100 kHz. (Inset: equivalent circuit model used for
fitting and higher frequency zooming).
RD. As shown in Table 4, Rpt of the DSSCs with liquid electrolyte is lower
compared to solid electrolytes. However, Rk was higher for Cz-D2
sensitizer with solid and liquid electrolytes based devices, compared to
Cz-D1 sensitizer based devices. The electron lifetime (τe) of Cz-D2 is also
higher in both solid and liquid electrolytes, while a shift of low fre-
quency (mid) corresponds to a longer τe and is subjected to a higher
value of Voc.
4. Experimental
4.1. Reagents, materials and methods
Tafel polarization measurement was used to study the interfacial
charge transfer properties of the electrolytes on the nanocrystaline TiO2
matrix [41]. As discussed in our previous papers [23,42,43], the rates of
anodic (βa) and cathodic (βc) reactions at the TiO2 photoanode/redox
mediator interface can be described by the following equation (the
Buttler Volmer Equation) [44–46].
All required chemicals used for the synthesis of dyes were purchased
from sigma Aldrich, TCI and were used without any further purification
unless otherwise explained. Nano-crystalline TiO2 (<20 nm, 99.8%,
anatase), alpha-terpineol, ethyl cellulose, titania (IV) propoxide, TiCl4,
BMII and tert-butyl pyridine were purchased from Sigma-Aldrich (India)
and used as received. Acetonitrile and ethanol were purified using the
standard process. Fluorine tin oxide glass (FTO, 12Ω/sq, Solaronix, SA,
À
�
À
�
�
À
��
CnF
anF
j ¼ À j0 expβ
RT E À Eeq À expβ
RT E À Eeq
Switzerland) and sealing materials (Meltronix tape, thickness 60 μm)
E
eq is the equilibrium potential of IÀ /I3À electrolyte, J0 is the current
exchange density and E is the applied voltage. In dark condition the
equation simplifies to j ¼ jo as E becomes Eeq, also J0 depends on the
concentrations of redox couple in the electrolyte and reaction area on
the TiO2 surface. In Fig. 9, tafel analysis is illustrated which gives an
insight into the electrode kinetics of DSSCs based on the solid and liquid
electrolytes. In liquid electrolyte system, the Jo for the IÀ /I3À redox
couple at the TiO2/electrolyte interface was about 0.69 nA/cm2 (Cz-D1)
and 0.82 nA/cm2 (Cz-D2). It is much higher than that of the DSSCs based
on the solid state electrolyte (Table 4).
The solid electrolyte resulted in a decrease of J0 which increased the
charge transfer resistance at electrolyte/TiO2 surface. Lower βc slope
value was found in the liquid electrolyte based DSSCs as shown in Fig. 9
and Table 4. These βc slope value (lower than solid electrolyte), close to
235 mV/decade (Cz-D1) and 220 mV/decade (Cz-D2), corresponds to
the recombination reaction at the sensitized TiO2 photoanode/IÀ /I3À
electrolyte interface between electrons from the CB of titania and
triiodide in the liquid electrolyte. The reduced βc slope indicates an
enhanced reaction of the triiodide at the sensitized TiO2 photoanode/IÀ /
IÀ3 electrolyte interface.
Fig. 9. Tafel polarization study of the solid and liquid electrolytes based DSSCs
fabricated using new dyes.
Table 4
Electrochemical parameters of DSSC with liquid and solid electrolyte.
Device
State of the Electrolyte
Rpt (Ω)a
Rk (Ω)a
CPE-P (%)a
τ
(ms)a
βa (mV/decade)b
βc (mV/decade)b
Jo (nA/cm2)b
Cz-D1
Cz-D2
Cz-D1
Cz-D2
Liquid
Liquid
Solid
20.9
21.9
25.1
24.6
229
323
210
265
65
69
70
70
0.83
1.68
0.80
1.15
145.5
115.6
122.4
115.4
235.4
220.0
259.5
308.3
0.69
0.82
0.21
0.32
Solid
a
b
Electrochemical Impedance spectroscopy.
Data obtained from Tafel polarization curve.
6

T.B. Raju et al.
Dyes and Pigments xxx (xxxx) xxx
were used for the fabrication of the DSSC device. The alkylation and
bromination of carbazole were done according to the published litera-
ture procedure [22,23]. The synthetic procedures of two dyes (Cz-D1,
Cz-D2) are shown in Fig. 1. By using Suzuki coupling reaction,
CZ-PhF2-CHO and Cz-2PhF2-CHO were synthesized. The final step to
synthesize new dyes (Cz-D1, Cz-D2) was performed by using Knoeve-
nagel condensation reaction method. All the necessary products were
purified by column chromatography, characterized by NMR (1H, 13C)
and HRMS spectrometry. Instruments used for different characterization
purpose and the characterization data are described in supporting in-
formation (ESI Page S2–S9).
7.49–7.44 (m, 3H), 7.30–7.27 (d, 1H), 4.33 (t, 2H) 1.91–1.89 (t, 2H),
1.42–1.32 (m, 5H), 1.30–1.24 (m, 9H), 0.89–0.86 (t, 3H). 13C NMR
(100 MHz, CDCl3): δ (ppm) 186.14, 154.73, 152.88, 149.15, 146.89,
141.12, 140.93, 138.34, 126.65, 126.48, 125.65, 121.33, 120.70,
119.64, 109.27, 109.17, 43.47, 31.99, 29.56, 29.37, 29.19, 27.51,
22.80, 14.26. HRMS (ESI) m/z: [MþNa]þ calcd for C27H28F2NO
420.2139, found 420.2123.
4.3.4. 4,4’-(9-octyl-9H-carbazole-3,6-diyl)bis(2,3-difluoro benzaldehyde)
[Cz-2PhF2-CHO]
Yield ¼ 68%, 1H NMR (400 MHz, CDCl3): δ (ppm) 10.37 (s, 1H), 8.37
(s, 1H), 7.75–7.64 (m, 4H), 7.56–7.53 (d, 2H), 7.48–7.46 (m, 4H), 4.37
(t, 2H) 1.94–1.92 (t, 2H), 1.56 (t, 1H), 1.44–1.25 (m, 9H), 0.89–0.84 (t,
3H). 13C NMR (100 MHz, CDCl3): δ (ppm) 186.13, 141.39, 137.31,
132.32, 128.79, 128.71, 127.34, 125.66, 123.31, 123.06, 121.51,
114.13, 109.61, 43.69, 31.96, 29.89, 29.54, 29.35, 27.49, 22.78, 14.24,
14.26. HRMS (ESI) m/z: [M]þ calcd for C34H29F4NO2 559.2134, found
559.2141.
4.2. General synthetic procedure of aldehyde
4.2.1. Synthesis of Cz-PhF2-CHO
A 2 M K2CO3 solution in THF: H2O (2:1) was added to the mixture of
3-bromo-9-octyl-9H-carbazole (0.2 mmol) and 2,3-Difluoro-4-formyl-
phenylboronic acid (0.3 mmol) then degassed properly. Finally, in
presence of argon, Pd(PPh3)4 (2 mol%) was added and heated to 85 �C
for 18 h. Then the solution was evaporated under reduced pressure. The
extracted compound was washed with chloroform and aqueous brine
solution. Finally, organic part was dried with anhydrous MgSO4. The
desired product was purified by column chromatography (silica gel,
EtOAc-hexane 1:2 as eluent).
4.3.5. (Z)-2-Cyano-3-(2,3-difluoro-4-(9-octyl-9H-carbazol-3-yl)phenyl)
acrylic acid [Cz-D1]
Yield ¼ 73%, 1H NMR (400 MHz, DMSO‑d6): δ (ppm) 8.86–8.42 (d,
2H), 8.30 (s, 1H), 8.24–8.22 (d, 2H), 8.10 (S, 2H), 8.02 (s, 1H),
7.71–7.70 (m, 4H), 7.62.7.61 (m, 6H), 7.56–7.48 (m, 7H), 7.23 (m, 3H),
4.41 (t, 2H), 1.57 (m, 5H), 1.33–1.15 (t, 6H), 0.84–0.78 (m, 4H),
1.30–1.24 (m, 9H), 0.89–0.86 (t, 3H). 13C NMR (150 MHz, DMSO‑d6): δ
(ppm) 149.54, 148.03, 146.33, 140.51, 140.06, 136.79, 133.09, 126.47,
126.22, 125.59, 123.84, 122.94, 122.44, 122.04, 121.42, 120.95,
120.63, 119.19, 118.64, 109.58, 50.91, 30.99, 28.51, 27.22, 26.14,
24.16, 22.03, 13.85. HRMS (ESI) m/z: [M À H]- calcd for C30H28F2N2O2
485.2041, found 485.2031.
4.2.2. Synthesis of Cz-2PhF2-CHO
In
a
clean reaction flask 3,6-dibromo-9-octyl-9H-carbazole
(0.2 mmol) and 2,3-Difluoro-4-formylphenylboronic acid (0.5 mmol)
taken. Then, 2 M K2CO3 solution in THF: H2O (2:1) was added to the
reaction mixture. Finally, in presence of argon, Pd(PPh3)4 (2 mol%) was
added and heated to 85 �C for 18 h. Then the solution was evaporated
under reduced pressure. The extracted compound was washed with
chloroform and aqueous brine solution. Finally, organic part was dried
with anhydrous MgSO4. The desired product was purified by column
chromatography (silica gel, EtOAc-hexane 1:2 as eluent).
4.3.6. (2Z,20E)-3,3’-((9-octyl-9H-carbazole-3,6-diyl)bis(2,3-difluoro-4,1-
phenylene))bis(2-cyanoacrylic acid) [Cz-D2]
Yield ¼ 68%, 1H NMR (400 MHz, CDCl3): δ (ppm) 8.32–8.18 (m, 2H),
7.92 (s, 1H), 7.75–7.68 (m, 3H), 7.56–7.40 (m, 3H), 7.17–7.10 (d, 2H).
13C NMR (100 MHz, CDCl3): δ (ppm) 149.10, 146.82, 140.39, 133.99,
132.19, 128.92, 126.29, 122.99, 122.73, 121.16, 120.68, 117.88,
117.06, 114.29, 109.39, 108.64, 51.17, 31.91, 29.88, 29.15, 27.32,
26.57, 22.74, 14.23. MALDI-TOF m/z: [M À 2H]þ calcd for
4.3. General synthetic procedure of dyes
4.3.1. Synthesis of Cz-D1
The solution of acetonitrile: chloroform (2:1) was added to the
mixture of Cz-PhF2-CHO (0.2 mmol) and cyanoacetic acid (0.3 mmol)
and degassed it thoroughly. Finally, in presence of argon atmosphere the
piperidine (catalytic amount) was added into the solution. Then the
solution was refluxed for 18 h. After cooling to RT, the solution was
evaporated under reduced pressure. Extracted compound was washed
with chloroform and 0.1 M aq. HCl. Then the organic part was dried with
MgSO4. The product was purified by conventional column chromatog-
raphy (silica gel, EtOAc -hexane 1:2 as eluent) and it was obtained as a
yellow-orange solid.
C40H31F4N3O4 691.6696, found 691.6610.
Declaration of competing interest
None.
Acknowledgment
This work was supported by the Department of Science and Tech-
nology, New Delhi, India through the projects DST/TSG/PT/2009/23,
IGSTC/MPG/PG (PKI)/2011A/48 and DST/SB/S1/PC-020/2014. Cen-
tral Instruments Facility, IIT Guwahati, is acknowledged for providing
various instrumentation facilities. S. S. Soni and J. V. Vaghasiya thank
DST, CERI (DST/TM/CERI/C285) for financial support for fabrication
facilities.
4.3.2. Synthesis of Cz-D2
The solution of acetonitrile: chloroform (2:1) was added to the
mixture of Cz-2PhF2-CHO (0.2 mmol) and cyanoacetic acid (0.5 mmol)
and degassed it thoroughly. Finally, in presence of argon atmosphere the
piperidine (catalytic amount) was added into the solution. Then the
solution was refluxed for 18 h. After cooling to RT, the solution was
evaporated under reduced pressure. Extracted compound was washed
with chloroform and 0.1 M aq. HCl. Then the organic part was dried with
MgSO4. The product was purified by conventional column chromatog-
raphy (silica gel, EtOAc -hexane 1:2 as eluent) and it was obtained as a
yellow-orange solid.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.dyepig.2019.108021.
References
4.3.3. 2,3-Difluoro-4-(9-octyl-9H-carbazol-3-yl)benzaldehyde [Cz-PhF2-
CHO]
[1] Fathabadi H. Sol Energy 2017;146:30–43.
[2] Liu J, Zhao H, Wu M, Van der Schueren B, Li Y, Deparis O, Ye JH, Ozin GA,
Hasan T, Su BL. Adv Mater 2017;29:1605349.
Yield ¼ 70%, 1H NMR (400 MHz, CDCl3): δ (ppm) 10.37 (s, 1H), 8.33
(s, 1H), 8.15–8.13 (d, 1H), 7.71–7.69 (t, 3H), 7.53–7.50 (t, 2H),
[3] Sellami MH, Loudiyi K. Renew Sustain Energy Rev 2017;70:1331–5.
7

T.B. Raju et al.
Dyes and Pigments xxx (xxxx) xxx
[4] Hwang H, Sin DH, Kulshreshtha C, Moon B, Son J, Lee J, et al. J Mater Chem A
2017;5:10269–79.
[24] Kakiage K, Aoyama Y, Yano T, Oya K, Fujisawa J, Hanaya M. Chem Commun 2015;
51:15894–7.
[5] Zhang H, Zhang SQ, Gao K, Liu F, Yao HF, Yang B, et al. J Mater Chem A 2017;5:
10416–23.
[25] Lu XJ, Mou XL, Wu JJ, Zhang DW, Zhang LL, Huang FQ, et al. Adv Funct Mater
2010;20:509–15.
[6] Xu Z, Fan Q, Meng X, Guo X, Su W, Ma W, et al. Chem Mater 2017;29:4811–8.
[26] Ren YM, Liu J, Zheng AB, Dong XD, Wang P. Adv Sci 2017;4:1700099.
[27] Huang ZS, Zang XF, Hua T, Wang LY, Meier H, Cao DR. ACS Appl Mater Interfaces
2015;7:20418–29.
�
�
�
[7] Roy C, Bura T, Beaupre S, Legare M-A, Sun J-P, Hill IG, et al. Macromolecules 2017;
50:4658–67.
[8] Oregan B, Gratzel M. Nature 1991;353:737–40.
[9] Chen WC, Kong FT, Li ZQ, Pan JH, Liu XP, Guo FL, et al. ACS Appl Mater Interfaces
2016;8:19410–7.
[28] Kimura M, Masuo J, Tohata Y, Obuchi K, Masaki N, Murakami TN, et al. Chem Eur
J 2013;19:1028–34.
[29] Jia HL, Huang MZ, Peng ZJ, Wang DM, Zhang GH, Guan MY. Dalton Trans 2017;
46:15124–9.
[10] Li CM, Luo L, Wu D, Jiang RY, Lan JB, Wang RL, et al. J Mater Chem A 2016;4:
11829–34.
[30] Han LY, Islam A, Chen H, Malapaka C, Chiranjeevi B, Zhang SF, Yang XD,
Yanagida M. Energy Environ Sci 2012;5:6057–606.
[11] Mathew S, Yella A, Gao P, Humphry-Baker R, Curchod BFE, Ashari-Astani N, et al.
Nat Chem 2014;6:242–7.
[31] Mao M, Zhang X, Cao L, Tong Y, Wu G. Dyes Pigments 2015;117:28–36.
[32] Park SS, Won YS, Choi YC, Kim JH. Energy Fuel 2009;23:3732–6.
[33] Vaghasiya JV, Sonigara KK, Prasad J, Beuvier T, Gibaud A, Soni SS. J Mater Chem
A 2017;5:5373–82.
[12] Xie YS, Tang YY, Wu WJ, Wang YQ, Liu JC, Li X, et al. J Am Chem Soc 2015;137:
14055–8.
[13] She ZJ, Cheng YG, Zhang LQ, Li XY, Wu D, Guo Q, et al. ACS Appl Mater Interfaces
2015;7:27831–7.
[34] Song D, Cho W, Lee JH, Kang YS. J Phys Chem Lett 2014;5:1249–58.
[35] Yella H, Lee W, Tsao HN, Yi C, Chandiran AK, Nazeeruddin MK, Diau EW-G,
Yeh CY, Zakeeruddin SM, Gratzel M. Science 2011;334:629.
[36] Li, Wang L, Kang B, Wang P, Qiu Y. Sol Energy Mater Sol Cells 2006;90:549–73.
[37] Bisquert J. J Phys Chem B 2002;106:325–33.
[14] Wu JH, Lan Z, Lin JM, Huang ML, Huang YF, Fan LQ, et al. Chem Rev 2015;115:
2136–73.
[15] Cho N, Han J, Song K, Kang MS, Jun MJ, Kang Y, et al. Tetrahedron 2014;70:
427–33.
[16] Paek S, Cho N, Song K, Jun MJ, Lee JK, Ko J. J Phys Chem C 2012;116:23205–13.
[17] Chen DY, Hsu YY, Hsu HC, Chen BS, Lee YT, Fu H, et al. Chem Commun 2010;46:
5256–8.
[38] Huang SY, Schlichthorl G, Nozik AJ, Gra1tzel M, Frank AJ. J Phys Chem B 1997;
101:2576–82.
[39] Adachi M, Sakamoto M, Jiu J, Ogata Y, Isoda S. J Phys Chem B 2006;110:13872.
[40] Vaghasiya JV, Sonigara KK, Fadadu KB, Soni SS. Perspect Sci 2016;8:46–9.
[41] Huo Z, Zhang C, Fang X, Cai M, Wang K. J Power Sources 2010;195:4384–90.
[42] Fadadu KB, Soni SS. Electrochim Acta 2013;88:270–7.
[43] Vekariya RL, Sonigara KK, Fadadu KB, Vaghasiya JV, Soni SS. ACS Omega 2016;1:
14–8.
[18] Afroz MA, Sonigara KK, Raju TB, Soni SS, Iyer PK. Phys Chem Chem Phys 2017;19:
28579–87.
[19] Hao L, Wang J, Bai FQ, Xie M, Zhang HX. Comput Theor Chem 2015;1067:119–28.
[20] Scrascia, De Marco L, Laricchia S, Picca RA, Carlucci C, Fabiano E, et al. J Mater
Chem A 2013;1:11909–21.
[21] Lin YD, Chow TJ. J Photochem Photobiol, A 2012;230:47–54.
[22] Raju TB, Vaghasiya JV, Afroz MA, Soni SS, Iyer PK. Org Electron 2016;39:371–9.
[23] Raju TB, Vaghasiya JV, Afroz MA, Soni SS, Iyer PK. Phys Chem Chem Phys 2016;
18:28485–91.
[44] Soderberg JN, Co AC, Sirk AHC, Birss VI. J Phys Chem B 2006;110:10401–10.
[45] Petrii OA, Nazmutdinov RR, Bronshtein MD, Tsirlina GA. Electrochim Acta 2007;
52:3493–504.
[46] Vaghasiya JV, Sonigara KK, Suresh L, Fard MP, Soni SS, Tan SC. Nano Energy 2019;
60:457–66.
8