Design, Synthesis and Photophysical Analysis of New Unsymmetrical Carbazole-Based Dyes for Dye-Sensitized Solar Cells

Article abstract of DOI:10.1016/j.jphotochem.2020.112521

The molecular design, synthesis, and characterization of di-anchoring carbazole-based dyes (Car-Cy, Car-Amin, and Car-Mal) with A-π-D-π-A-π-A structure as materials for dye-sensitized solar cells applications (DSSCs) are reported. The electron-rich carbazole moiety in the structure of synthesized dyes is connected with acceptor/ or anchoring groups (cyanoacetic acid, 4-aminobenzoic acid, and malonic acid) and π-spacers (vinylene and cynovinyl thiophene). Electronic characteristics and molecular geometry of the sensitizers were optimized using Density Functional Theory (DFT), and the influence of dye structure on their photovoltaic performances was studied. Among the synthesized dyes in this research, Car-Amin dye presents a better photovoltaic performance as a sensitizer in the constructed device, with a power conversion efficiency (PCE) of 2.27%, J_SC of 5.95 mAcm^?2, V_OC of 0.54 V and FF of 71%. The enhanced performance of this dye could be related to the powerful electron-withdrawing characteristic of the 4-aminobenzoic acid as an acceptor group in the dye structure. Consequently, the impact of various acceptor groups on the constructed DSSC devices was examined, and the results indicated that the improvement of DSSC performance is due to the presence of 4-aminobenzoic acid as acceptor on the carbazole-based dyes.

Full text of DOI:10.1016/j.jphotochem.2020.112521

Journal Pre-proof
Design, Synthesis and Photophysical Analysis of New Unsymmetrical
Carbazole-Based Dyes for Dye-Sensitized Solar Cells
Faegheh Ghasempour Nesheli, Mahmood Tajbakhsh, Behzad
Hosseinzadeh, Rahman Hosseinzadeh
PII:
S1010-6030(20)30320-8
DOI:
https://doi.org/10.1016/j.jphotochem.2020.112521
JPC 112521
Reference:
To appear in:
Journal of Photochemistry & Photobiology, A: Chemistry
Received Date:
Revised Date:
Accepted Date:
19 January 2020
12 March 2020
26 March 2020
Please cite this article as: Nesheli FG, Tajbakhsh M, Hosseinzadeh B, Hosseinzadeh R,
Design, Synthesis and Photophysical Analysis of New Unsymmetrical Carbazole-Based Dyes
for Dye-Sensitized Solar Cells, Journal of Photochemistry and amp; Photobiology, A:
Chemistry (2020), doi: https://doi.org/10.1016/j.jphotochem.2020.112521
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
© 2020 Published by Elsevier.

Design, Synthesis and Photophysical Analysis of New Unsymmetrical Carbazole-Based Dyes
for Dye-Sensitized Solar Cells
a
a
Faegheh Ghasempour Nesheli , Mahmood Tajbakhsh a,*, Behzad Hosseinzadeh , Rahman
Hosseinzadeha
a Department of Organic Chemistry, Faculty of Chemistry, University of Mazandaran, Babolsar
47416-95447, Iran
*Corresponding author E-mail: Tajbaksh@umz.ac.ir
Graphical abstract
Highlights
 Three new carbazole-based di-anchoring unsymmetrical organic dyes were
designed and synthesized for the DSSC application.
1

 The effect of different acceptor groups on the performance of the prepared
DSSCs was studied.
 Car-Amin dye with 4-aminobenzoic acid as an acceptor unit exhibited
better photovoltaic performance compared to the other dyes.
Abstract
The molecular design, synthesis, and characterization of di-anchoring carbazole-based dyes (Car-
Cy, Car-Amin, and Car-Mal) with A-π-D-π-A-π-A structure as materials for dye-sensitized solar
cells applications (DSSCs) are reported. The electron-rich carbazole moiety in the structure of
synthesized dyes is connected with acceptor/ or anchoring groups (cyanoacetic acid, 4-
aminobenzoic acid, and malonic acid) and π-spacers (vinylene and cynovinyl thiophene).
Electronic characteristics and molecular geometry of the sensitizers were optimized using Density
Functional Theory (DFT), and the influence of dye structure on their photovoltaic performances
was studied. Among the synthesized dyes in this research, Car-Amin dye presents a better
photovoltaic performance as a sensitizer in the constructed device, with a power conversion
efficiency (PCE) of 2.27%, JSC of 5.95 mAcm−2, VOC of 0.54 V and FF of 71%. The enhanced
performance of this dye could be related to the powerful electron-withdrawing characteristic of the
4-aminobenzoic acid as an acceptor group in the dye structure. Consequently, the impact of various
acceptor groups on the constructed DSSC devices was examined, and the results indicated that the
2

improvement of DSSC performance is due to the presence of 4-aminobenzoic acid as acceptor on
the carbazole-based dyes.
Abbreviations
Dye-sensitized solar cells (DSSCs)
Density Functional Theory (DFT)
Cyclic Voltammetry (CV)
Power conversion efficiency (PCE)
Open-circuit photovoltage (Voc)
Fill factor (FF)
Short-circuit photocurrent density (JSC)
Electrochemical impedance spectroscopy (EIS)
Incident photon-to-current efficiency (IPCE)
Current–voltage (J –V)
Intramolecular charge transfer (ICT)
Highest occupied molecular orbital (HOMO)
Lowest unoccupied molecular orbital (LUMO)
First oxidation potentials (Eox)
Frontier molecular orbital (FMO)
3

Charge recombination resistance (Rct)
Electron lifetime (τ)
Chemical capacitances (Cµ)
Dimethylformamide (DMF)
Dimethyl sulfoxide (DMSO)
Phosphorus oxychloride (POCl3)
Ethanol (EtOH)
Methanol (MeOH)
Acetic acid (AcOH)
Ammonium acetate (NH4OAc)
Keywords: Dye-sensitized solar cell (DSSC), Carbazole, 4-aminobenzoic acid, Cyanoacetic acid,
Malonic acid.
Introduction
The necessity of using non-expensive renewable energy has attracted many researchers to focus
on organic photovoltaic material. Many researchers have concentrated on the DSSCs based on
mesoporous TiO2 thin films because of their low-cost and suitable PCE [1]. Photosensitizers as
the most vital and major components of DSSCs have a considerable task in the light harvesting
4

process for electricity generation. The Ruthenium complexes like N3, N719, and NCSU-10 are
excellent light harvesting compounds in DSSCs because of their good optical and electrochemical
behavior [2]. Although the maximum PCE for such DSSCs is reported to be more than 10%, some
disadvantages including the utilization of expensive noble metals, unavailability, high cost of
preparation, being detrimental to the environment and having problematic separation have limited
their further applications [3-9]. Moreover, the sensitizers without metal in their structures have
become more popular compared to Ruthenium complexes because of their design versatility, easy
availability, convenient synthesis, low cost, high capability of light harvesting, easy purification
and being environment-friendly [10-13]. Organic sensitizers usually employ donor–π–acceptor
(D–π–A) architecture, which includes a donor (D) that is connected to an acceptor unit (A) via a
π-linker (π). Recently, series of indolo[3,2,1-jk]carbazole- based organic dyes with the D-π-A
architecture synthesized and the effect of different π-bridges on photovoltaic performance have
been studied [14, 15].
The donor groups of the dyes are important components in DSSCs because of their effects on the
absorption properties and energy levels [16, 17]. Typically, various compounds including
coumarin [18], squaraine [19], indoline [20], phenothiazine (PTZ) [21], triphenylamine [22],
fluorene [20], carbazole [23] and tetrahydroquinoline [24] have been used as electron-donating
moieties in most of the organic dyes. Among the mentioned donors, as can be observed in Figure
1, carbazole is an aromatic tricyclic compound, containing pyrrole and benzene rings. The donor
and hole-transporting properties of carbazole are mostly due to the existence of the pyrrole ring in
its structure. Up to 240°C, the carbazole system is thermally stable, and this stability can be
improved even more by attaching it to other molecules. Carbazole is cheap, easily available, and
if necessary its solubility in organic solvents can be improved by functionalizing its nitrogen atom
5

with alkyl groups [25]. The organic dyes on the basis of carbazole were first developed in 2006.
After that, many carbazole-based dyes were prepared and employed in DSSCs as donor or π-spacer
[26, 27]. Further, many carbazole-based hole-transporting materials (HTMs) were synthesized and
employed in perovskite solar cells as donor or π-spacer [28, 29].
For improving the performance of DSSCs, the π-conjugated units have a critical task in the
architecture of metal-free organic dyes. It is known that thiophene derivatives are employed as π-
conjugated groups because of their unique electrochemical and polarizability properties as well as
their good charge transporting ability [13]. However, the synthetic procedure of connecting
thiophene derivatives to the carbazole ring system is usually complicated and needs costly metal
catalysts under anhydrous and inert conditions [30, 31]. To overcome these problems,
Knoevenagel condensation reaction of thiophene-2-acetonitrile with aldehydes can be easily used
to introduce the thiophene derivatives onto the carbazole ring system. The advantage of this
synthetic route is the introduction of thiophene moiety and cyanovinyl group to the carbazole ring
system, which leads to the extension of π-conjugations as well as the reduction of the gap between
LUMO and HOMO energy levels.
Recently, the advantages of using 4-aminobenzoic acid, cyanoacetic acid, and malonic acid as
effective electron acceptors for fabricating DSSCs including their powerful electron-withdrawing
properties as well as having the capability of connecting onto the TiO2 surface via carboxylic acid
have drawn much attention from the researchers [32]. Furthermore, di-anchoring organic
sensitizers reported in the literature exhibit several advantages including multi binding abilities to
TiO2 nanoparticles, the red shift in UV-absorption, and a more desirable electron injection which
improves the generated photocurrent [33-37]. Studies have indicated that employing di-anchoring
organic dyes instead of single D–π–A dyes in DSSCs results in a better PCE [9, 38-42].
6

Di-anchoring sensitizers provide lower open-circuit photovoltage (Voc) in comparison with
similar mono-anchoring sensitizers. This is assumed to be related to the undesirable improved
charge recombination process in these sensitizers [37]. Thus, the structure of di-anchoring
sensitizers can be optimized to achieve DSSCs with higher efficiencies. There are many
symmetrical di-anchoring sensitizers that have been reported in the literature, however, there are
not many reports on the asymmetric di-anchoring sensitizers.
In this regard, this research has aimed to design three novel A- π -D-π-A- π-A configurated dyes
(Car-Cy, Car-Amin, Car-Mal) containing two acceptor/or anchoring units and one donor group
(Figure 1). The new asymmetrical di-anchoring sensitizers have a carbazole moiety, which plays
the role of the donor, and a π-conjugated fragment (cyano vinylene and thiophene groups) that is
attached to acceptor group on one side, whereas an acceptor unit is connected to carbazole on the
other side. Cyanoacetic acid, 4-aminobenzoic acid, and malonic acid in the structure of the
synthesized dyes act as electron acceptor/anchoring units.
Figure 1 shows the structure of the synthesized dyes. These dyes were analyzed using various
spectral methods such as 1H and 13C NMR, mass spectroscopy, CHN examination, and FT-IR. The
electrochemical and optical behavior of these synthesized dyes were also examined to evaluate the
effects of the anchoring groups in their structure. The constructed devices using the synthesized
dyes were investigated for photovoltaic performance and electrochemical impedance spectroscopy
(EIS) behavior.
7

Figure. 1 The chemical structures of Car-Cy, Car-Amin and Car-Mal dyes.
2 The Experimental
2.1 Equipment and materials
The chemicals and solvents were bought from Aldrich and Merck companies. These materials
were utilized without further purification. The samples were examined using an FT-IR Bruker
Vector 22 spectrometer. A Bruker 400 MHz Ultrashield Avance DRX spectrometer was used for
analyzing 1H and 13C NMR spectra, using DMSO-d6 and CDCl3 as solvents. Electrothermal 9100
apparatus (UK) was used to measure melting points of samples. A Finnigan-Matt 8430 mass
spectrometer was employed for recording the mass spectra at the ionization potential of 70 eV. A
LECO-600 CHN Elemental Analyzer was used in order to perform elemental analysis.
The UV/VIS absorption spectra were examined using Jasco V-750 spectrophotometer. Further, the
fluorescence emission spectra were recorded using Jasco FP-8300 fluorescence
spectrophotometer. Cyclic Voltammetry (CV) experiments were conducted by VersaSTAT 4
Potentiostat Galvanostat (Princeton Applied Research), connected to a regular 3-electrode system
which included a glassy carbon working electrode, a platinum counter electrode, and an Ag/AgCl
reference electrode in a DMF solution with room temperature. The potentials were recorded with
a scan rate of 0.05 Vs-1, and the solution of the samples was prepared using the 10-4 M sample and
also 0.1 M tetrabutylammonium tetrafluoroborate as the supporting electrolyte. An AM 1.5G solar
8

simulator (Sharif Solar, SIM-1000, light intensity: 100 mW/cm2) was used for photovoltaic
evaluation of the prepared cells. A Bio-Logic SAS SP-300 was utilized for measuring the current–
voltage (J –V). Also, an IPCE system with a 100 W halogen lamp (Sharif Solar, IPCE-015) was
employed to determine the incident photon-to-current efficiency (IPCE). The electrochemical
impedance spectroscopy (EIS) were measured on an impedance analyzer (EIS-25H, Sharif Solar)
at their open-circuit photovoltage (Voc).
2.2. The synthetic process and characterization
The synthesis of the novel carbazole-based dyes was performed as it has been shown in scheme 1.
The synthesizing process involved a series of synthetic procedures such as N-alkylation,
Vilsmeier-Haack, and Knoevenagel condensation reaction. The synthesizing procedure of the
intermediates and dyes, as well as their spectral data, have been reported in the following sections.
9

Scheme 1. The pathway of synthesizing for the novel carbazole-based dyes. a 1-bromopentane,
NaOH, DMSO, 110 °C; b POCl3, DMF, 100 °C; c thiophene-2-acetonitrile, EtOH, reflux; d
POCl3, DMF, 100 °C; e cyanoacetic acid, AcOH, NH4OAc, reflux; f 4-aminobenzoic acid,
MeOH, reflux; g malonic acid, piperidine, pyridine, reflux.
2.2.1 The procedure of synthesizing 9- pentyl-9H-carbazole (1)
In order to synthesize compound 1, a previously reported technique was utilized with some
adjustments [43]. A mixture of carbazole (2.0 g, 0.01196 mol), 1-bromopentane (3.071 g, 0.02033
mol), dimethylsulfoxide (DMSO) (30 mL), and sodium hydroxide (4.0 g, 0.10046 mol) underwent
10

stirring at 110 °C for 12 hours. Subsequently, the mixture temperature was reduced to 25 °C by
leaving it in the room. Then, hexane and water were employed for extracting the product. A rotary
evaporator was utilized in order to evaporate the organic solvent under a vacuum condition. The
remaining solid material underwent a purification process by silica gel column chromatography
with hexane as an eluent, and consequently a white solid was generated (92%, 2.61 g). m.p; 51-52
°C, 1H NMR (300 MHz, CDCl3): δ 8.18 (d, 2H, J = 8 Hz), 7.55–7.51 (m, 2H), 7.47–7.45 (m, 2H),
7.29 (t, 2H, J = 7.4 Hz), 4.34 (t, 2H, J = 7.2 Hz), 1.93 (q, 2H, J = 7.2 Hz), 1.44–1.41 (m, 4H), 0.94
(t, 3H, J = 6.8 Hz), 13C NMR (100 MHz, CDCl3): δ 140.47, 125.60, 122.84, 120.38, 118.72, 108.69,
43.08, 29.47, 28.73, 22.55, 14.02.
2.2.2. The procedure of synthesizing 9- pentyl-9H-carbazole-3-carbaldehyde (2)
Initially, a cold DMF solution (1.5 mL, 20 mmol, 0 0C) was prepared, then phosphorus oxychloride
(1.6 mL, 17 mmol) was added to the cold solution dropwise. The mixture underwent stirring for 1
hour at 0 0C. Afterward, the 9- pentyl-9H-carbazole (1, 3.56 g, 15 mmol) that was produced in the
previous section of the study, was added to this mixture, and the mixture was heated at 95 0C for
7 hours. Subsequently, it was gradually cooled to 25 0C, and was poured into a mixture of ice and
water, also, NaOH (4 M) was employed for basifying the mixture. CH2Cl2 and brine were used for
extracting the product, and Na2SO4 was utilized in order to dry the organic phase. After that, in
order to remove the Na2SO4, the mixture went through filtration, and the filtrate was evaporated
under a vacuum condition. Then, the crude product underwent a purification process using silica
gel column chromatography (ethyl acetate/ hexane, 1/10, v/v) to obtain a white powder (70%, 2.78
g), m.p; 63-64 0C, FT-IR (KBr pellet, cm-1): 2956 (C-H stretching), 1688 (C=O stretching), 1592
1
(C=C stretching), H NMR (300 MHz,CDCl3): δ 10.10 (s, 1H), 8.60 (s, 1 H), 8.16 (d, J=7.6 Hz,
1H) , 8.01 (d, J= 8.4 Hz, 1H), 7.55 (t, J= 8.4 Hz, 1H), 7.47-7.45 (m, 2H), 7.36-7.32 (m, 1H), 4.31
11

(t, J= 7.2 Hz, 2H) , 1.90 (q, 2H, J = 7.2 Hz), 1.39-1.37 (m, 4H), 0.90 (t, J=7.2 Hz, 3H), 13C NMR
(100 MHz, CDCl3): δ 191.79, 144.05, 141.15, 128.45, 127.13, 126.71, 123.98, 123.03, 122.97,
120.73, 120.28, 109.40, 108.93, 43.38, 29.35, 28.64, 22.45, 13.95, Mass: m/z: 265.1, Anal. Calcd.
for C18H19NO: C, 81.47; H, 7.22; N, 5.28. Found: C, 81.30; H, 7.12; N, 5.32.
2.2.3. The procedure of synthesizing (Z)-3-(9-pentyl-9H-carbazol-3-yl)-2-(thiophen-2-yl)
acrylonitrile (3)
A mixture of thiophene-2-acetonitrile (1.44 mL, 13.6 mmol), ethanol (30 mL), compound 2 (3 g,
11.3 mmol), and NaOH (0.1 mL, 6 M) was refluxed for 2 hours. The obtained yellow residue went
through filtration, and it was rinsed using methanol. Finally, the recrystallization in chloroform
gave a yellow solid as the product with a yield of 72 %, m.p; 104-105 0C, FT-IR (KBr pellet, cm-
1
1): 2949 (C-H stretching), 2215 (-CN stretching), 1587 (C=C stretching), H NMR (400 MHz,
CDCl3): δ 8.59 (s, 1H), 8.18-8.10 (m, 2H), 7.58 (s, 1H), 7.53 (t, 1H), 7.47-7.44 (m, 2H), 7.39 (d,
1H, J= 4 Hz), 7.33-7.28 (m, 2H), 7.10 (d, 1H), 4.32 (t, 2H, J= 7.2 Hz), 1.91 (q, 2H, J = 7.2 Hz),
1.41-1.37 (m, 4H) 0.91 (t, 3H, J= 7.2 Hz), 13C NMR (100 MHz, CDCl3): δ 141.61, 141.23, 141.01,
140.17, 128.02, 126.82, 126.44, 125.99, 125.14, 124.39, 123.25, 122.75, 122.45, 120.74, 119.82,
117.95, 109.18, 109.15, 102.05, 43.31, 29.38, 28.70, 22.46, 13.96, Mass: m/z 370.2, Anal. Calcd.
for C24H22N2S: C, 77.80; H, 5.98; N, 7.56. Found: C, 77.80; H, 5.92; N, 7.52.
2.2.4. The procedure of synthesizing (Z)-3-(3-formyl-9-pentyl-9H-carbazol-6-yl)-2-(5-
formylthiophen-2-yl) acrylonitrile (4)
Phosphorus oxychloride (4.4 ml, 25eq) was added to a cold DMF solution (3.4 ml, 23eq) dropwise
under an inert atmosphere at the temperature of 0° C. The mixture underwent stirring for 1 hour.
Then, compound 3 (0.74 g, 2 mmol) was added to the mixture, and they underwent stirring again
12

at 100°C for 4 hours. After the reaction was ended (identified by TLC), the mixture temperature
was decreased to 25 °C and it was poured into a mixture of water and ice. Afterward this mixture
was basified using 4M NaOH. CH2Cl2 and brine were used for extracting the product, and Na2SO4
was utilized in order to dry the organic phase. Then, a rotary evaporator was used for evaporating
the organic solvent under a vacuum condition and the resulting crude product underwent a
purification process by the means of silica gel column chromatography (ethyl acetate/ hexane, 1/5,
v/v), to obtain an orange colored solid (32%; 0.27 g), m.p; 118 0C, FT-IR (KBr pellet, cm-1): 2924
(C-H stretching), 2212 (-CN stretching), 1671 (C=O stretching), 1583 (C=C stretching), 1H NMR
(400 MHz, CDCl3): δ 10.15 (s, 1H), 9.93 (s, 1H), 8.74 (s, 1H), 8.68 (s, 1H), 8.20-8.09 (m, 2H),
7.78-7.76 (m, 2H), 7.56 (d, 2H, J=8.4 Hz), 7.51 (d, 1H, J=4 Hz), 4.40 (t, 2H, J=7.2 Hz), 1.94 (q,
2H, J = 7.2 Hz), 1.42-1.39 (m, 4H), 0.90 (t, 3H, J=7.2 Hz), 13C NMR (100 MHz, CDCl3) δ 191.54,
182.64, 167.83, 149.07, 144.67, 143.86, 142.51, 137.02, 129.61, 128.47, 127.74, 126.86, 125.08,
123.70, 123.34, 122.89, 119.19, 116.85, 110.14, 109.74, 102.26, 43.87, 31.93, 29.69, 22.39, 14.10,
Mass: m/z; 426.2, Anal. Calcd. for C26H22N2O2S: C, 73.21; H, 5.20; N, 6.57. Found: C, 73.19; H,
5.17; N, 6.50.
2.2.5. The procedure of synthesizing (2Z)-3-[5-((E)-2-{6-[(E)-2-carboxy-2-cyanovinyl]-9-
pentyl-9H-carbazol-3-yl}-1-cyanovinyl)-2-thienyl]-2-cyanoacrylic acid (Car-Cy)
A mixture of cyanoacetic acid (0.038 g, 0.45 mmol), compound 4 (0.064 g, 0.15 mmol), glacial
acetic acid (15 mL) and ammonium acetate (0.057 g, 0.74 mmol) was refluxed for 12 hours. Then
the mixture temperature was decreased to 25 °C and it was poured into a mixture of ice and water,
to generate a red colored solid. The mixture was then filtered to collect the generated solid. Finally,
the crude product underwent a purification process with silica gel column chromatography
(CHCl3:CH3OH, 10:1, v/v) to result in a red colored solid. Yield 58%, m.p; 290 0C, FT-IR (KBr
13

pellet, cm-1): 3252 (O-H stretching), 2924 (C-H stretching), 2219 (-CN stretching), 1730 (C=O
stretching), 1H NMR (400 MHz DMSO-d6,): δ 8.89 (s, 1H), 8.76 (s, 1H), 8.49 (s, 2H), 8.28 (d, 2H,
J=9.2 Hz), 8.24 (s, 1H), 8.01 (d, 1H, J=4 Hz), 7.9 (d, 2H, J=8.8 Hz), 7.65 (d, 1H, J=4 Hz), 4.51 (t,
2H, J=7.6 Hz), 1.82 (q, 2H), 1.29-1.23 (m, 4H), 0.82 (t, 3H, J=6.8 Hz), 13C NMR (125 MHz
DMSO-d6): δ 164.42, 163.82, 155.23, 148.24, 146.33, 144.90, 143.70, 142.85, 141.10, 139.12,
137.77, 135.62, 129.50, 127.98, 126.65, 125.58, 125.30, 124.37, 123.95, 122.85, 122.80, 117.57,
116.84, 111.48, 100.88, 100.06, 99.97, 43.39, 28.96, 28.71, 22.32, 14.33, Mass: m/z 562. Anal.
Calcd. for C32H24N4O4S: C, 68.56; H, 4.31; N, 9.99. Found C, 68.50; H, 4.29; N, 9.95.
2.2.6. The procedure of synthesizing 4-[((1E)-{5-[(E)-2-(6-{(Z)-[(4-carboxyphenyl)imino]
methyl}-9-pentyl-9H-carbazol-3-yl)-1-cyanovinyl]-2-thienyl}methylene)amino]benzoic acid
(Car-Amin)
A mixture of 4-aminobenzoic acid (0.06 g, 0.45 mmol), dry methanol, and compound 4 (0.064 g,
0.15 mmol) was refluxed for 3 hours. After the filtration and recrystallization of the precipitated
product from methanol, a bright orange colored solid was acquired. Yield 77%, m.p; 325 0C, FT-
IR (KBr pellet, cm-1): 3437 (O-H stretching), 2927 (C-H stretching), 2214 (-CN stretching), 1686
(C=O stretching), 1H NMR (400 MHz DMSO-d6): δ 8.88 (s, 1H), 8.82 (d, 2H, J=12 Hz), 8.72 (s,
1H), 8.21-8.17 (m, 2H), 8.15 (s, 1H), 8.01-7.98 (m, 4H), 7.87-7.82 (m, 2H), 7.75 (d, 1H, J=4 Hz),
7.50 (d, 1H, J=4 Hz), 7.38-7.36 (m, 4H), 4.50 (t, 2H, J=6.8 Hz), 1.83 (q, 2H), 1.30-1.28 (m, 4H),
0.82 (t, 3H, J=6.8 Hz), 13C NMR (400 MHz DMSO-d6): δ 167.99, 167.45, 156.23, 155.24, 154.57,
153.59, 144.59, 144.43, 143.49, 143.26, 142.71, 142.54, 142.29, 142.17, 135.62, 131.68, 131.09,
129.46, 128.72, 128.53, 128.26, 127.84, 126.89, 125.59, 122.93, 122.60, 122.51, 121.72, 121.52,
117.62, 117.31, 113.03, 111.22, 101.62, 101.14, 43.22, 28.96, 28.65, 22.30, 14.30 , Mass: m/z
664.2. Anal Calcd. for C40H32N4O4S: C, 72.27; H, 4.85; N, 8.43. Found: C, 72.20; H, 4.81; N, 8.41.
14

2.2.7. The procedure of synthesizing (2E)-3-[5-((E)-2-{6-[(E)-2-carboxyvinyl]-9-pentyl-9H-
carbazol-3-yl}-1-cyanovinyl)-2-thienyl] acrylic acid (Car-Mal)
Initially, a mixture of piperidine (0.05 mL, 0.53 mmol), malonic acid (0.15 g, 1.45 mmol), pyridine
(3 mL), and compound 4 (0.20 g, 0.48 mmol), was refluxed for 5 hours. Afterward, the mixture
temperature was reduced to 25 0C, and it was poured into a mixture of water and ice. Then, an HCl
solution (2 M) was used in order to acidify the mixture until a red solid is generated. After the
filtration and rinsing this solid material with water, a vacuum oven was employed for drying it for
12 hours. Subsequently, in order to purify the precipitate, it was recrystallized from ethanol to
result in a dark red colored solid with a yield of 75 %, m.p; 240 0C, FT-IR (KBr pellet, cm-1): 3428
(O-H stretching), 2929 (C-H stretching), 2213 (-CN stretching), 1683 (C=O stretching), 1H NMR
(400 MHz, DMSO-d6): δ 12.38 (s, 2H), 8.74 (s, 1H), 8.51 (s, 1H), 8.19 (d, 1H, J=8.8 Hz), 8.00 (s,
1H), 7.90-7.88 (m, 2H), 7.84-7.82 (m, 1H), 7.78 (d, 1H, J=3.2 Hz), 7.74-7.72 (m, 1H), 7.57 (d,
1H, J=3.6 Hz), 7.45 (d, 1H, J=3.6 Hz), 6.58 (d, 1H, J=16 Hz), 6.22 (d, 1H, J=15.6 Hz), 4.48 (t,
2H), 1.80 (q, 2H), 1.29 (m, 4H), 0.81 (t, 3H, J=6.4 Hz), 13C NMR (100 MHz, DMSO-d6): δ 168.38,
167.56, 145.38, 143.27, 142.39, 142.26, 142.18, 139.32, 138.49, 136.60, 133.37, 132.23, 127.20,
127.01, 126.80, 125.04, 123.79, 122.91, 122.80, 121.73, 118.85, 117.78, 111.02, 104.53, 101.28,
43.50, 29.00, 28.74, 22.34, 14.32, Mass: m/z 509.5. Anal. Calcd. For C30H26N2O4S: C, 70.57; H,
5.13; N, 5.49. Found: C, 70.51; H, 5.10; N, 5.44.
2.3. The fabrication of titanium dioxide electrode
The titanium dioxide electrode was constructed using the reported method in the literature [44]. A
mixture of acetic acid (12 g, 0.2 mol) and titanium isopropoxide (58.6 mg, 0.2 mol) underwent
stirring at 25 °C for 15 minutes. Afterward, it was added into water (290 mL) at once and while
stirring (600 rpm). The stirring of the mixture was continued for another 1 hour to complete the
15

hydrolysis. Then concentrated nitric acid (4 mL) was added to the mixture, and they underwent
stirring at 80 °C for about 40 minutes and were peptized for 75 minutes.The mixture was cooled
by adding water to obtain a volume of 370 mL, then it was heated in a titanium autoclave (570
mL) at the temperature of 250 °C for 12 hours.
Subsequently, nitric acid (65%, 2.4 mL) was added to the contents of autoclave at 25 °C and was
ultrasonicated with the power of 200 W at the frequency of 20 pulses in every 3 seconds. The
obtained solution was evaporated to a concentration of 13 wt% TiO2. The residue was centrifuged
and rinsed using ethanol, to generate a white solid.
2.4. Preparing the DSSC devices
A blocking layer was prepared by treating 40 mM TiCl4 aqueous solution with fluorine-doped tin
oxide (FTO) coated glass substrate for half an hour at the temperature of 70 °C. Titania pastes
(Sharif Solar Co.) and a doctor blade were used to prepare the photoanodes. The photoanodes were
made of transparent and scattering layers of TiO2 nanoparticles. The scattering layer with a
thickness of 5 μm contained TiO2 nanoparticles in the range of 150–300 nm, while the transparent
layer with a thickness of 7 μm contained TiO2 nanoparticles with a size of 20 nm. Thickness of the
TiO2 layers were measured with a profilometer and confirmed with the cross-sectional scanning
electron microscope (SEM) image (Figure S 28).
The FTO films with titanium dioxide coatings were heated for 10 minutes at the temperature of
375 °C, 15 minutes at the temperature of 450 °C, and 30 minutes at the temperature of 500 °C.
Then, their temperature was reduced to 25 °C. The obtained films underwent another treatment
with a 40 mM TiCl4 aqueous solution at the temperature of 70 °C for half an hour. They were
washed using water and heated at a temperature of 500 °C for 30 minutes. Afterward, dye solutions
(0.4 mM in DMF) were employed for immersing the titanium dioxide electrodes for 20 hours
16

without the presence of light, and the electrodes were rinsed with water. The FTO glass was coated
by the H2PtCl6 solution (one drop) to set the counter electrode. This electrode was kept at 500 °C
for half an hour. An iodine-based liquid electrolyte (100mM lithium iodide (LiI), 100 mM iodine
(I2), 500 mM 4-tert-butylpyridine (TBP), 600mM tetrabutylammonium iodide (TBAI) in
acetonitrile solution) was utilized while assembling the two electrodes into a sandwich-type cell.
2.5. Computational detail
Density functional theory (DFT) calculations were employed using the hybrid B3LYP functional
and 6-31+ G (d,p) basis set to compare the electronic properties and geometries of Car-Cy, Car-
Mal and Car-Amin [45]. All structural optimization and energy calculations were performed using
the Gaussian 09 program package [46]. The vibrational frequency calculations were performed to
confirm that all optimized geometries were local minima. The time-dependent DFT (TD-DFT)
was carried out to find the lowest vertical excitation energies of the synthesized compounds. The
electronic structure and TD-DFT calculations were performed in DMF solution, using the
conductor-like polarizable continuum model (CPCM) to characterize the UV-Vis spectrum and
frontier molecular orbital energies, at B3LYP/6-31+G (d,p) levels of theory [47].
17

3 Results and discussion
3.1 The optical characteristics
The UV–visible spectra of Car-Cy, Car-Amin, and Car-Mal in DMF solution (10−5 M) have been
depicted in Figure 2. The recorded spectra for Car-Cy and Car-Mal dyes display two distinct peaks:
the first peak at 296 - 309 nm is related to the π–π* electron transition, whereas the second peak at
392–397 nm is attributed to the intramolecular charge transfer (ICT) from carbazole to the acceptor
group [48]. However, the dye Car-Amin exhibits three distinct bands: two bands in the UV region
(~298 nm and ~336 nm) corresponding to the π-π* electron transition. Whereas the third peak in
the visible region (426 nm) is possibly attributed to the intramolecular charge transfer (ICT) from
carbazole to the acceptor group [49]. The coefficient of molar extinction (ε) at the maximum
absorption (λmax) of ICT for Car-Cy, Car-Amin and Car-Mal were determined to be 52,457, 69,265
and 45,260 M−1 cm−1, respectively. These data cleary demonstrate the excellent light harvesting
capacity of the synthesized dyes (Table 1).
The ICT absorption peak of a dye on the TiO2 film exhibits a red or blue shift compared with that
in solution. This is because of the interaction between the dye and TiO2 which leads to the
accumulation of dye on the surface of titanium dioxide, resulting in a J- and H-aggregation for a
red and blue shift, respectively. As shown in Table 1 and Figure 3, the absorption spectrum of dyes
on the TiO2 films were expanded, and the maximum absorption peaks for Car-Cy, Car-Amin, and
Car-Mal were observed at 352, 424, 364 nm, demonstrating a blue shift of 45, 2, and 28 nm in
comparison with those in DMF solution, respectively. H-aggregation or deprotonation of the dyes
on the titanium dioxide surface can be good reasons for this blue shift [50]. Interestingly, the dye
Car-Amin exhibits a much less blue shift with no remarkable differences as compared with that in
DMF solution, which can imply that there is less aggregation of this dye on the TiO2 surface [51].
18

This phenomenon may indicate that the dye Car-Amin can generate the highest open- circuit photo
voltage (Voc) value among other two dyes, which is in favor of a better photovoltaic performance.
The broader absorption of dye Car-Amin on the TiO2 film is beneficial to improve the light-
harvesting capability, resulting in the higher photocurrent density and IPCE value [52].
Since dyes tend to accumulate at the solid-liquid interface, the blue shift often occurs for the
organic dyes [53]. Furthermore, the adsorption of dyes on TiO2 film makes the carboxylic acid
group deprotonated, thus the capability of the TiO2–carboxylate group as the acceptor will be less
than the carboxylic acid itself [54]. So, all of the discussed issues can be effective in the observed
blue shift. The interaction between titanium dioxide film and the dye broadens the absorption
spectra, which is absolutely favorable in the production of light harvesting devices since it can
possibly improve the photocurrent [55].
Fluorescence emission spectra of Car-Cy, Car-Amin, and Car-Mal were recorded in DMF solution
(10−5 M) upon their excitation at their absorption wavelengths (λabs). Figure 4 exhibits the emission
spectra of dyes and their outcomes are summarized in Table 1. As can be seen, the dyes show
emission maxima in wavelength range of 524–560 nm which is in the same trend as observed in
the absorbance spectra. Furthermore, Stokes shift values of dyes were calculated from their
normalized UV-Vis absorption and fluorescence emission spectral data (Table 1)[56]. The Stokes
shift values for the sensitizers were estimated to be in the range of 132-134 nm, exhibiting the
efficient electron shift from donor unit (carbazole) to the acceptor / anchoring groups. In addition,
optical band gaps (E0-0) of the synthesized dyes were determined from the intersection of
absorption and emission spectra and their calculated values are summarized in Table 1.
19

Table 1. The electrochemical and optical characteristics of the examined dyes.
Dyes
Car-Cy
-5.98/-3.40
2.57
Car-Amin
-5.68/-2.98
Car- Mal
-5.72/-2.94
2.78
HOMO/LUMO[a](ev)
Band gap[a]
2.70
λabs [b] (nm)/(ε M−1 cm−1)
λem (nm)
397/52,457
426/69,265
392/45,260
530
133
352
560
134
424
524
132
364
Stokes shift (nm)
λmax[c] (nm)
HOMO[d] (eV)
-5.95
2.68
-5.99
2.61
-5.6
2.75
-2.85
E0-0 [e](eV)
LUMO[f] (eV)
-3.27
-3.38
[a]DFT computed energy values; [b]Absorption wavelengths of charge-transfer transition were evaluated in
DMF; [c]Maximum absorption wavelength on TiO2 films; [d]HOMO values were obtained using CV in DMF
[e]
with 0.1 M tetrabutylammonium tetrafluoroborate;
computed using HOMO+E0–0.
E
= 1240/λ intersection; [f]LUMO values were
0−0
20

Figure 2. The recorded absorption spectra for the examined dyes in the DMF solution.
Figure 3. The recorded absorption spectra for the examined dyes absorbed on TiO2 films.
21

Figure 4. The recorded fluorescence emission spectra for the examined dyes in the DMF
solution.
3.2 Electrochemical properties
The highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital
(LUMO) values of the prepared sensitizers were assessed by carrying out CV experiments. The
obtained values can evaluate the thermodynamic possibility of electron injection from the LUMO
of the dye molecule into the conduction band edge of the titanium dioxide semiconductor, and they
can assess the capability of the redox electrolyte for regenerating the dye through refilling the
generated hole in the HOMO. Figure 5 presents the results of the CV experiment. Also, the data
and a graphical presentation of the measured HOMO and LUMO levels have been presented in
Table 1, and Figure 6, respectively.
22

The first oxidation potentials (Eox) of Car-Cy, Car-Amin, and Car-Mal were found to be 1.25, 1.29,
and 0.9 V (vs. NHE), respectively. HOMO levels of dyes were determined from their first
oxidation potentials (Eox) using equation (1) [56].
HOMO = -[Eox + 4.7 eV]
(1)
. Since these values for all of the examined dyes are more positive in comparison with the redox
-
potential of the I3 / I- electrolyte system (-5.2 eV), it can be concluded that there is sufficient power
for the regeneration of the dyes. Moreover, the energy difference between the excited and ground
states of the dyes (E0–0) was calculated with E0–0 = 1240/ λint, in which, λint is the intersection of
the emission and absorption spectra. The λint values for the Car-Cy, Car-Amin, and Car-Mal were
found to be 461, 474 and 450 nm, respectively. The values of E0–0 have been presented in Table
1. The LUMO levels were determined using equation (2).
LUMO = HOMO+E0–0
(2)
These values for the Car-Cy, Car-Amin, and Car-Mal were found to be -3.27, -3.38 and -2.85 eV,
respectively, and since they are more negative than the conduction band of titanium dioxide (− 4.2
eV), the electron injection from the dyes into TiO2 is possible. Consequently, it can be assumed
that the synthesized dyes in this study are potential sensitizers in DSSCs.
23

Figure 5. The recorded CV results for the examined dyes in the DMF solutions.
24

Figure 6. The schematic of the energy levels in the examined dyes based on absorption and
electrochemical characteristics.
3.3 Theoretical investigations
The DFT computations are known as one of the important computational investigations and are
broadly applied in the DSSC systems to get a better look at the electronic properties and molecular
structure of the sensitizers [57, 58]. Figure 7 illustrates the optimized molecular structures as well
as the distribution of the electron density in their related frontier molecular orbital (FMO) energy
levels.
The results of natural bond orbital (NBO) studies indicate that the HOMO electron density are
frequently delocalized on donor (carbazole ring) and π-linker (vinyl thiophene) segments, whereas,
the LUMO electrons are obviously transferred from carbazole to the acceptors (cyanoacetic acid,
4-aminobenzoic acid, and malonic acid). The extent of this transformation depends on the nature
25

of electron-withdrawing groups of the acceptor [56]. In addition, it is worth to mention that the
electron cloud of LUMO energy level in the examined dyes was only shifted toward the
acceptor/anchoring unit connected to the vinyl thiophene π-spacer and not the anchoring group
connected to vinylene unit [59]. This observation can be due to the higher abundance of electrons
in the π-conjugation section, which results in more interactions between the acceptor and donor
[60]. The theoretical data show that the HOMO levels obtained for dyes Car-Cy (-5.98 eV), Car-
Mal (-5.72 eV) and Car-Amin (-5.68 eV) are sufficiently lower than the electrode potential of
redox couple (-5.2 eV) facilitating the regeneration process. Likewise, obtained LUMO levels of
Car-Cy (-3.40 eV), Car-Mal (-2.94 eV) and Car-Amin (-2.98 eV) clearly show that, electron
injection is also facilitated as the values are higher than the CB of TiO2 (-4.2 eV). Therefore, from
theoretical point of view, these dyes are also useable in DSSCs. The theoretical HOMO and LUMO
levels are close to the experimental data obtained from cyclic voltammograms (Table 1).
Furthermore, TD-DFT calculations were performed to elucidate the electronic transitions and also
to interpret of UV-Vis spectrum of all three compound in solution. The maximum absorption
wavelengths, oscillator strength (f) and corresponding transition assignment are listed in Table 2,
and the absorption spectra are shown in Figure 8. The data show that the lowest energy transition
of the dyes corresponds to HOMO→LUMO excitation. Therefore, the bands seen in the visible
region of the experimental absorption spectra include the HOMO → LUMO transitions. The red
shift of λmax from theory to experiment is associated to the self-interaction error in TD-DFT arising
through the electron transfer in the extended charge-transfer state (Figure 8) [61]. The other
transitions involve HOMO-1 → LUMO, HOMO → LUMO +1 and HOMO → LUMO +2
corresponding to the π-π* transitions. In addition, the f calculated for the aforementioned transition
is 1.56 for Car-Cy, 1.45 for Car-Mal and 1.78 for Car-Amin, respectively, which is consistent with
26

the trend of molar extinction coefficients (Table 2). Therefore, from these observations, it can be
mentioned that the theoretical absorption spectra are in accordance with the experimentally ones.
However, Car-Cy dye shows slightly more red-shift comparing to the other dyes, which could be
related to the effect of functional and basis set on the energy levels [62].
Figure 7. The FMOs of the HOMO and LUMO from the DFT computations.
27

Figure 8. The TD-DFT absorbance spectra of dyes calculated with B3LYP/6-31+G (d, p) level.
Table 2. Maximum absorption wavelengths, orbital energies, oscillator strength (f) and assignment
of synthesized dyes in DMF solution optimized at B3LYP/61+G (d, p) level.
Dyes
λmax
(Experimental)
397
λmax
(B3LYP)
552.6
448.1
438.5
Composition
(%)
H→L (70.0)
H-1→L (63.0)
H→L+2 (63.0)
E (eV)
f
Car-Cy
2.24
2.76
2.83
1.56
0.15
0.13
309
Car-Amin
Car-Mal
426
298,336
533.1
431.0
419.3
H→L (70.0)
H-1→L (70.0)
H→L+1 (69.0)
2.32
2.87
2.95
1.78
0.25
0.17
392
296
511.3
423.7
H→L (70.0)
H-1→L (69.0)
2.42
2.92
1.45
0.21
28

3.4 Photovoltaic performance
To establish the applicability of the prepared dyes as sensitizers, DSSC devices were constructed
using the synthesized dyes and their performance was studied. The photovoltaic parameters
including PCE, fill factor (FF), VOC and short-circuit photocurrent density (JSC) were measured for
the constructed devices and the obtained results have been summarized in Table 3. Figure 9
presents the Current-Voltage (J–V) curves for sensitizer-based DSSCs.
As is clear in Table 3, the maximum PCE, JSC, and VOC, with the values of 2.27%, 5.95 mA/cm2,
and 0.54 V, respectively, were observed for the Car-Amin dye. The JSC value of Car-Amin is also
bigger than that in the other dyes which could be related to its greater molar extinction coefficient
(Table 1) [63]. Reports have also shown that a higher value of Jsc can be effective in obtaining
greater PCE percentages [43]. According to the results, the photovoltaic performance of the
examined dyes follows this order: Car-Amin > Car-Cy > Car-Mal. The higher performance of the
Car-Amin dye could be due to the good electron-withdrawing ability from carbazole donor
followed by good efficiency of 4-aminobenzoic acid anchoring group in injecting electrons into
the conduction band edge of titanium dioxide. These observations are well compatible with the
previous reports in similar studies [64]. The data of Figure 9 further confirms the influence of
acceptor/anchoring groups on the DSSC device performance. As can be observed, when malonic
acid was used as the anchoring group, the corresponding dye that is, Car-Mal presented lower Jsc
and PCE (4.99 mA cm−2 and 1.69%, respectively) in comparison with the other investigated dyes.
The results of this research revealed that the efficiency of anchoring/acceptor groups in constructed
devices follows this order: 4-aminobenzoic acid > cyanoacetic acid > malonic acid. Therefore,
among the used acceptors, 4-aminobenzoic acid showed to be a better electron acceptor that can
be attached to TiO2 more efficiently. Also, we have compared the solar cell efficiency of the
29