Unravelling the miniscule size effect of auxiliary donor in D-D-π-A type dipolar photosensitizers on quasi-solid state DSSC performance

Article abstract of DOI:10.1016/j.dyepig.2020.108959

Three new dipolar organic dyes of the type D-D-π-A were constructed on fluorenothiophene backbone by varying the terminal donor auxiliary such as -NMe₂, -NEt₂ and -NPh₂. The effect of such structural variation on the photophysical and electrochemical properties has been investigated. Computational studies obtained from DFT/TD-DFT calculations were found to be in good correlation with the experimental optoelectronic results. TiO₂ based n-type quasi-solid state DSSCs constructed on the synthesized dyes revealed a gradual increase in photovoltaic performance with increase in donor size. The trend in power conversion efficiency (η) was found to be -NPh₂ (4.10%) > -NEt₂ (3.44%) > -NMe₂ (2.57%). Electrochemical impedence analysis also confirmed the longer electron life time for dye possessing bulky -NPh₂ groups in comparision to -NEt₂ and -NMe₂ groups, thus demonstrating the impact of increased donor size on the DSSC performance. Overall, this paper validates the impact of donor size on the DSSC performance and could inspire the selection of appropriate auxiliary donors to enhance the efficiency in futuristic endeavors.

Full text of DOI:10.1016/j.dyepig.2020.108959

Journal Pre-proof
Unravelling the miniscule size effect of auxiliary donor in D-D-π-A type dipolar
photosensitizers on quasi-solid state DSSC performance
Chitra Kumar, Abbasriyaludeen Abdul Raheem, Ramesh Mohan, Ramasamy
Shanmugam, Chandrasekar Praveen
PII:
S0143-7208(20)31656-9
DOI:
https://doi.org/10.1016/j.dyepig.2020.108959
DYPI 108959
Reference:
To appear in: Dyes and Pigments
Received Date: 22 September 2020
Revised Date: 16 October 2020
Accepted Date: 20 October 2020
Please cite this article as: Kumar C, Raheem AA, Mohan R, Shanmugam R, Praveen C, Unravelling
the miniscule size effect of auxiliary donor in D-D-π-A type dipolar photosensitizers on quasi-solid state
DSSC performance, Dyes and Pigments, https://doi.org/10.1016/j.dyepig.2020.108959.
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 Elsevier Ltd. All rights reserved.

Author statement
Respected Editor,
At the outset, I would like to thank yourself as well as reviewers for the critical insights and
fruitful suggestions to shape up the submitted manuscript (Manuscript ID: DYPI-D-20-
0
0722) in good quality. As per suggestion, I have made all corrections in the revised
manuscript and highlighted them in yellow color along with a point-by-point response to the
reviewer comments. Herewith, I am submitting the revised manuscript titled “Unravelling
the miniscule size effect of auxiliary donor in D-D-π-A type dipolar photosensitizers on
quasi-solid state DSSC performance’’ for your kind perusal. I sincerely hope that the
revision of the manuscript makes the article a fitting one to the standards of Dyes and
Pigments.
Thanking You!
Yours sincerely
C. PRAVEEN

Unravelling the miniscule size effect of auxiliary donor in D-D-π-
A type dipolar photosensitizers on quasi-solid state DSSC
performance
[
a]
[a,b]
[c]
[d]
Chitra Kumar, Abbasriyaludeen Abdul Raheem,
Ramesh Mohan, Ramasamy Shanmugam, and
[
a,b]
Chandrasekar Praveen*
[
a]
Electrochemical Power Sources (ECPS) Division, Central Electrochemical Research Institute (CSIR
Laboratory), Alagappapuram, Karaikudi-630003, Sivagangai District, Tamil Nadu, India.
[
b]
Academy of Scientific and Innovative Research (AcSIR), Alagappapuram, Karaikudi-630003,
Sivagangai District, Tamil Nadu, India.
[
c]
Microsystem Packaging Group, Central Electronics Engineering Research Institute (CSIR Laboratory),
Pilani-333031, Jhunjhunu District, Rajasthan, India.
[
d]
Department of Chemistry, Thiagarajar College, Madurai-625009, Madurai District, Tamil Nadu, India
Email: chandrasekar.praveen@gmail.com

Abstract: Three new dipolar organic dyes of the type D-D-π-A were constructed on fluorenothiophene
backbone by varying the terminal donor auxiliary such as -NMe , -NEt and -NPh . The effect of such
2
2
2
structural variation on the photophysical and electrochemical properties has been investigated.
Computational studies obtained from DFT/TD-DFT calculations were found to be in good correlation
with the experimental optoelectronic results. TiO based n-type quasi-solid state DSSCs constructed on
2
the synthesized dyes revealed a gradual increase in photovoltaic performance with increase in donor size.
The trend in power conversion efficiency (η) was found to be -NPh (4.10 %) > -NEt (3.44%) > -NMe
2
2
2
(2.57%). Electrochemical impedence analysis also confirmed the longer electron life time for dye
possessing bulky -NPh groups in comparision to -NEt and -NMe groups, thus demonstrating the impact
2
2
2
of increased donor size on the DSSC performance. Overall, this paper validates the impact of donor size
on the DSSC performance and could inspire the selection of appropriate auxiliary donors to enhance the
efficiency in futuristic endeavors.
Keywords:
Dipolar push-pull dyes; Donor size; Optoelectronics; DFT calculations; Quasi solid-state DSSC
2

Introduction
Photosensitizers play a critical role in dye sensitized solar cells (DSSCs) by absorbing solar photons and
injecting the photoexcited electrons onto the conduction band of a n-type semiconductor (typically
[1]
TiO2).
Though inorganic photosensitizers are well-established for this purpose, their organic
counterparts possess distinct advantages such as high molar absorptivity, inexpensive synthetic strategies
[
2]
and modifiable optoelectronic properties. In terms of efficiency (η), some of the organic sensitizers are
indeed on par with inorganic analogues. For instance, Hanaya et al. disclosed a combination of organic
[
3]
dyes to induce collaborative sensitization to achieve an efficiency >14%. Within the landscape of
organic sensitizers, dipolar push-pull dyes have received significant attention in DSSC owing to their
unique structural characteristics. These molecular chromophores which are otherwise known as donor-π-
acceptor (D-π-A) dyes possess an electron-donor (D), π-conjugated bridge (π) and electron acceptor
[
4]
(
A). Each of these sub-units play a significant role in defining the physical properties of the molecule to
enable a clear understanding on the structure-property relationship for appropriate applications in
.
[5]
optoelectronics For example, the function of donor is to control the position of Highest Occupied
Molecular Orbital (HOMO) and retarding the charge recombination. On the other hand, the acceptor
moiety controls the Lowest Unoccupied Molecular Orbital (LUMO) level through electronic coupling
between dye and conduction band of TiO . The π-bridge permits the unidirectional electron flow from the
2
dye to TiO , thereby diminishes the back-electron transfer. Hence, a combination of these structural
2
parameters is important for the realization of a suitable photosensitizer for DSSC. Particularly, the
emphasis on structural modification of donor moiety has led to the understanding of steric and
electrostatic effects on the device performance. This is because of the fact that bulkiness of the donor
-
substituent could significantly inhibit the approach of electrolyte (I ions) to the TiO surface.
3
2
Consequently, the charge recombination is suppressed leading to an increased electron life time. To name
few works pertaining to this, Wang et al. showed an increase in efficiency of 10.2% by incorporating -
[
6]
OC H group on the donor fragment of dinaphthopentacene based dyes. The same group also
6
13
3

witnessed an increased efficiency of 12.5% by fusing the donor backbone of indenoperylene dyes by a
[
7]
rigid diaryl group. Likewise, Grätzel et al. showed that a thienothiophene ring linked on the donor unit
[
8]
exhibited higher efficiency (12.6%) compared to the non-linked dye (11.1%). Moreover, the impact of
different aromatic/heteroaromatic cores on the donor support of metal-free dyes has been a subject of
[9]
frequent investigation in DSSC. Hence, it is essential to unravel the impact of donor groups on the
molecular properties at a fundamental level while designing an appropriate push-pull dye for DSSCs.
With this aforementioned literature background coupled with our on-going research program on
[
10]
optoelectronic push-pull chromophores, we herein disclose our investigation on the structure-property
relationship of three new dipolar dyes possessing closely related donor units (-NPh , -NEt and -NMe )
2
2
2
and the effect of such structural variation on the optoelectronic properties as well as DSSC performance.
As per the logical speculation, the dye with relatively large donor size (-NPh ) emerged as the best in
2
terms of efficiency and electron life time and all these results were disclosed in this paper.
Experimental Section
Materials, methods and measurements
All commercially available reagents were used without further purification unless otherwise stated.
Organic reagents and catalysts were purchased from Sigma-Aldrich and TCI Chemicals. Solutions in organic
solvents were dried with anhydrous sodium sulphate or magnesium sulphate. Solvents were evaporated
under reduced pressure. Column chromatography was performed using thick-walled glass columns along
with a mixture of petroleum ether and ethyl acetate on silica gel (100-200 mesh, SRL, India). The relative
proportions of solvents in chromatography solvent mixtures refer to the volume to volume ratio.
Analytical TLC was performed on precoated plastic sheets of silica gel G/UV-254 of 0.2 mm thickness
(Macherey-Nagel, Germany) using analytical grade solvents and visualized with iodine spray (10% w/w
I in silica gel), UV light (λ = 254 and 365 nm) and alkaline KMnO solution. Melting points were
2
4
recorded on micro controller-based melting point apparatus using open capillaries. Infrared (IR) spectra
4

for all solid compounds were recorded on a Bruker Optik GmbH, TENSOR 27 FT-IR spectrophotometer
1
13
as KBr pellets. H and C NMR spectra were obtained in acetone-d and CDCl on a BRUKER
6
3
spectrometer at 400, 500 and 100, 125 MHz respectively. Proton chemical shifts (δ) are relative to
tetramethylsilane (TMS, δ = 0.00) as internal standard and expressed in parts per million. The number of
protons (n) for a given resonance was indicated as nH. Spin multiplicities are given as s (singlet), d
(doublet), t (triplet), q (quartet) and m (multiplet). Coupling constants (J) are given in hertz. GC-MS
spectra were obtained using a PerkinElmer instrument. High-resolution mass spectra (HRMS) were obtained
using a Q-TOF and JEOL-GCMate II mass spectrometer. GC-MS spectra were obtained using a
PerkinElmer instrument. UV-Visible spectra in solution mode
were recorded in chloroform solution on
SHIMADZU UV-3600 plus UV-VIS-NIR spectrometer. For solid films, the UV-visible absorption spectra
were measured in UV-visible spectrophotometer (Cary 5000 scan) at room temperature with diffuse reflectance
mode. Photoluminescence spectra were recorded in Jasco spectrofluorometer FP-8500. Cyclic
voltammetry (CV) experiments were performed with Autolab potentiostat galvanostat (Model PGSTAT-
3
0). All CV measurements were carried out at room temperature with a conventional three-electrode
configuration employing a glassy carbon electrode as the working electrode, silver wire as the pseudo-
reference electrode and a Pt wire as the counter electrode. Tetrabutylammonium hexafluorophosphate (0.1
-1
M) as supporting electrolyte was used in argon-purged chloroform with a scan rate 50 mV s . Density
functional theory (DFT) calculations were utilized for optimization at the B3LYP/6-31G**(d,p) level
using the G09 program. The photocurrent-voltage signal (J–V curve) measurements were performed in a
Sciencetech (SS 1.6 K, AAA) solar simulator attached with Keithley Instrument source meter (Model
No.: 2400) under the light illumination of 100 mW cm−2 with the optical filter of AM 1.5 G (USA) and
electrochemical properties of the fabricated DSSCs were studied using Bio-Logic SP-150 instrument.
Experimental synthetic procedures and characterization
Experimental procedure for 2,7-Dibromo-9H-fluorene (II)
5

3
.3638 g of N-bromosuccinimide I (0.0189 mol, 2.1 eq.) was added to 40 mL of dichloromethane
containing 1.4960 g of fluorene (0.0090 mol, 1.0 eq.) precursor. Then, 266 mg of FeBr (0.0009 mmol, 10
3
mol %) was added, refluxed for 5 h. Later on, the reaction mixture was cooled to room temperature and
then the excess bromine was neutralized with sodium thiosulphate solution (10%). The crude mixture was
worked up with ethyl acetate (3 x 30 mL) and purified over silica column afforded the corresponding
1
dibromo compound II as yellow solid (2.8577 g, 98 %). HNMR (500 MHz, CDCl δ in ppm): 3.82 (s,
3,
1
3
1
H), 7.47 (d, 2H, J = 5.0 Hz), 7.55 (d, 2H, J = 5.0 Hz), 7.63 (s, 2H); C NMR (125 MHz, CDCl δ in
3,
ppm): 36.5, 120.9, 121.1, 128.2, 130.1, 139.6, 144.7.
Experimental procedure for 2,7-Dibromo-9,9-dioctyl-9H-fluorene (III)
9
72 mg of 2,7-dibromofluorene (0.0030 mmol, 1.0 eq.) and 1.6831 g of potassium hydroxide (0.0300
mmol, 10.0 eq.) in dimethyl sulphoxide (10 mL) was stirred at room temperature for 10 minutes. Then,
o
1
.4484 g of 1-bromooctane was added and continued the reaction at 60 C for 5 days. The reaction
mixture was transferred to water (100 mL) and then extracted with ethyl acetate (3 X 20 mL). The organic
portion was dried over anhydrous Na SO and concentrated under reduced pressure. The obtained crude
2
4
product was chromatographed on a silica column afforded compound III as colourless semisolid (1.6453
g, 76 %). R = 0.70 (5 % EA/PE); FT-IR (KBr): 3735, 2925, 2855, 2354, 1878, 1745, 1573, 1455, 1254,
f
-1 1
1
124, 1061, 1061, 1004, 879, 808, 735 cm ; H NMR (500 MHz, CDCl δ in ppm): 0.86 (t, 6H, J = 5.0
3,
13
Hz), 1.08-1.31 (m, 24 H), 1.92-1.95 (m, 4 H), 7.47 (d, 4H, J = 5.0 Hz), 7.53 (d, 2H, J = 10.0 Hz); C
NMR (125 MHz, CDCl , δ in ppm): 14.0, 22.6, 23.6, 29.15, 29.17, 29.8, 31.7, 40.1, 55.7, 121.1, 121.4,
3
1
26.2, 130.1, 139.0, 152.5.
Experimental procedure for 2-(7-Bromo-9-heptyl-9-octyl-9H-fluoren-2-yl)-3-hexylthiophene (IV)
1
.4259 g of 2,7-dibromo-9,9-dioctyl-9H-fluorene (0.0026 mol, 1.0 eq.) and 559 mg of 2-(3-
hexylthiophen-2-yl)-4,4,5,5-tetramethyl-1,3,2-dioxaborolane (0.0019 mmol, 0.75 eq.) in toluene (40 mL)
6

taken under argon atmosphere. Subsequently, 2M potassium carbonate solution (20 mL) followed by
31.1 mg of tetrakistriphenylphosphine palladium(0) (0.0002 mmol, 10 mol %) were added and refluxed
2
for overnight. The crude mixture was worked up with ethyl acetate / water (3 x 30 mL) and purified by
silica column chromatography afforded compound IV as colourless semisolid (909.2 mg, 55%). R = 0.57
f
-
1 1
(
(
(
PE); FT-IR (KBr): 2926, 2857, 1605, 1457, 1248, 1063, 1006, 884, 816, 724, 662 cm ; H NMR
500MHz, CDCl δ in ppm): 0.86 (t, 7H, J = 5.0 Hz), 1.09-1.16 (m, 18H), 1.21-1.33 (m, 14H), 1.61-1.67
3,
m, 2H),1.95-2.00 (m, 4H), 2.69 (t, 2H, J = 5.0 Hz), 7.02 (d, 1H, J = 5.0 Hz), 7.27 (d, 1H, J = 5.0 Hz),
.39 (d, 1H, J = 5.0 Hz), 7.42-7.44 (m, 1H), 7.48 (t, 2H, J = 5.0 Hz), 7.58-7.60 (m,1H), 7.71 (d, 1H, J =
7
5
3
1
6
1
3
.0 Hz); C NMR (125 MHz, CDCl , δ in ppm), 14.0, 22.6, 23.8, 28.9, 29.2, 29.2, 29.7, 30.0, 31.0, 31.6,
3
1.7, 40.3, 55.4, 119.7, 121.0, 121.1, 123.5, 123.7, 126.1, 128.3, 129.6, 130.0, 133.9, 138.3, 138.7, 139.2,
+
Br
+
39.7, 150.5, 153.2. HRMS: m/z calcd for C H BrS [M] : 620.3051, found: 621.3067 [ M +H] ;
3
8
53
79
Br
+
23.3071 [ M +H] .
8
1
Experimental
procedure
for
5-(7-Bromo-9,9-dioctyl-9H-fluoren-2-yl)-4-hexylthiophene-2-
carbaldehyde (V)
To the 264.4 mg of N, N-dimethylformamide (3.5385 mmol, 2.5 eq.) in 1,2-dichloroethane was added
54.7 mg of phosphoryl trichloride (3.5385 mmol, 2.5 eq.) under ice cooled condition and stirred for 20
5
minutes. Then, 900 mg of 2-(7-bromo-9,9-dioctyl-9H-fluoren-2-yl)-3-hexylthiophene (1.4154 mmol, 1.0
eq.) was added to the reaction mixture and allowed to reflux for 5 h. Then the reaction mixture was
cooled to room temperature and then neutralized with sodium acetate solution (10 %). The requisite
organic portion was separated using ethyl acetate and water (3 x 25 mL) followed by the silica column
purification afforded compound V as yellow semisolid (501.8 mg, 54 %). R = 0.57 (5 % EA/PE); FT-IR
f
-1 1
(
KBr): 2926, 2856, 1735, 1665, 1446, 1250, 1154, 1063, 823, 675, 518 cm ; H NMR (500 MHz, CDCl
3,
δ in ppm): 0.75 (t, 6H, J = 5.0 Hz), 0.98-1.06 (m, 17H), 1.12-1.24 (m, 17H), 1.53-1.59 (m, 2H), 1.87 (t,
H, J = 5.0 Hz), 2.63 (t, 2H, J = 5.0 Hz),7.19 (s,1H), 7.32 (s,1H), 7.35-7.37 (m, 1H), 7.39 (t, 1H, J = 10.0
3
1
3
Hz), 7.50 (d, 1H, J = 5.0 Hz), 7.59 (s, 1H), 7.64 (d, 1H, J = 5.0 Hz), 9.80 (s, 1H); C NMR (125 MHz,
7

CDCl , δ in ppm), 14.0, 14.05, 22.5, 22.59, 23.8, 28.8, 29.1, 29.2, 29.7, 29.9, 30.8, 31.6, 31.7, 40.2,
3
5
1
6
5.5, 120.0, 121.3, 121.7, 123.5, 126.2, 128.2, 130.2, 132.5, 138.6, 139.2, 140.2, 140.7, 141.0, 149.2,
+
Br
+
50.8, 153.2,1 82.7. HRMS: m/z calcd for C H BrOS [M] : 662.3157, found: 663.3251 [ M +H] ;
4
0
55
79
Br
+
65.3239 [ M +H] .
8
1
Experimental
procedure
for
5-(7-(4-(Dimethylamino)phenyl)-9,9-dioctyl-9H-fluoren-2-yl)-4-
hexylthiophene-2-carbaldehyde (VI)
00 mg of 5-(7-bromo-9,9-dioctyl-9H-fluoren-2-yl)-4-hexylthiophene-2-carbaldehyde (0.6025 mmol, 1.0
4
eq.) and 228.6 mg of 4-(dimethylamino)phenylboronic acid (1.3857 mmol, 2.3 eq.) were taken in toluene
under argon atmosphere. Then, 2M potassium carbonate solution (20 mL) and 11.1 mg of
tetrabutylammonium iodide (0.0301 mmol,
5
mol%) followed by 69.5 mg of
tetrakistriphenylphosphinepalladium(0) (0.0602 mmol, 10 mol %) were added and refluxed for overnight.
The crude mixture was worked up with ethyl acetate/water (3 X 30 mL) and purified by silica column
chromatography afforded compound VI as greenish yellow semisolid (254.5 mg, 60%). R = 0.24 (2%
f
-1
EA/PE); FT-IR (KBr): 2924, 2855, 1667, 1607, 1528, 1443, 1352, 1242, 1152, 947, 812, 738, 671 cm ;
1
H NMR (500 MHz, CDCl δ in ppm): 0.78 (t, 6H, J = 5.0 Hz), 0.85 (t, 4H, J = 5.0 Hz), 1.06-1.19 (m,
3,
1
6
1
3
1
9H), 1.26-1.30 (m, 11H), 1.61-1.66 (m, 3H), 1.98-2.03 (m, 3H), 2.71 (t, 2H, J = 5.0 Hz), 3.01 (s, 5H),
.83 (d, 2H, J = 10.0 Hz), 7.40-7.44 (m, 2H), 7.53-7.59 (m, 4H), 7.67 (s, 1H), 7.72-7.75 (m, 2H), 9.87 (s,
1
3
H); C NMR (125 MHz, CDCl δ in ppm): 14.0, 14.07, 22.5, 22.6, 23.9, 28.9, 29.1, 29.2, 29.7, 30.0,
3,
0.8, 31.6, 31.7, 40.4, 40.6, 40.64, 55.2, 112.8, 119.7, 120.2, 120.6, 123.4, 125.2, 126.0, 126.3, 127.7,
28.0, 128.6, 129.5, 131.5, 138.2, 138.7, 140.1, 140.7, 141.8, 149.8, 150.0, 151.3, 151.7, 182.8. HRMS:
+
+
m/z calcd for C H NOS [M] : 703.4787, found: 704.4888 [M+H] .
4
8
65
Experimental
procedure
for
5-(7-(4-(Diethylamino)phenyl)-9,9-dioctyl-9H-fluoren-2-yl)-4-
hexylthiophene-2-carbaldehyde (VII)
8

2
50 mg of 5-(7-bromo-9,9-dioctyl-9H-fluoren-2-yl)-4-hexylthiophene-2-carbaldehyde (0.3765 mmol, 1.0
eq.) and 167.1 mg of 4-(diethylamino)phenylboronic acid (0.8659 mmol, 2.3 eq.) were taken in toluene
under argon atmosphere. Then, 2 molar potassium carbonate solution (20 mL) and 6.9 mg of
tetrabutylammonium iodide (0.0188 mmol, 5 mol%) followed by 43.4 mg of tetrakistriphenylphosphine
palladium(0) (mmol, 10 mol %) were added and refluxed for overnight. The crude mixture was worked
up with ethyl acetate / water (3 x 30 mL) and purified by silica column chromatography afforded
compound VII as yellow solid (206.7 mg, 75 %). R = 0.35 (5% EA/PE); FT-IR (KBr): 3935, 3769, 2923,
f
-1 1
2
855, 1665, 1607, 1524, 1445, 1260, 1200, 1081, 809, 741,670 cm ; H NMR (500 MHz, CDCl δ in
3,
ppm): 0.74 (t, 6H, J = 5.0 Hz), 0.78(t, 3H, J = 5.0 Hz), 0.98-1.05 (m, 17H), 1.09-1.15 (m, 12H), 1.18-1.21
m, 6H), 1.54-1.60 (m,3H),1.90-1.95(m, 4H), 2.64 (t, 2H, J = 5.0 Hz), 3.33 (m, 4H), 6.71 (d, 2H, J = 5.0
Hz), 7.18 (s, 1H), 7.33-7.36 (m, 2H), 7.45 (s, 1H), 7.49 (t, 3H, J = 5 Hz), 7.60 (s, 1H,), 7.65-7.67 (m, 2H),
(
1
3
9
2
1
.80 (s, 1H); C NMR (125 MHz, CDCl , δ in ppm): 12.6, 14.0, 14.06, 22.5, 22.6, 23.8, 28.9, 29.1, 29.2,
3
9.7, 30.0, 30.8, 31.6, 31.7, 40.4, 44.4, 55.2, 112.0, 119.6, 120.2, 120.5, 123.4, 125.0, 127.9, 128.0, 128.2,
31.5, 138.0, 138.7, 140.1, 140.7, 140.9, 141.9, 147.2, 149.8, 151.3, 151.6, 182.8. HRMS: m/z calcd for
+
+
C H NOS [M] : 731.5100, found: 732.5195 [M+H]
50
69
Experimental
procedure
for
5-(7-(4-(Diphenylamino)phenyl)-9,9-dioctyl-9H-fluoren-2-yl)-4-
hexylthiophene-2-carbaldehyde (VIII)
2
50 mg of 5-(7-bromo-9,9-dioctyl-9H-fluoren-2-yl)-4-hexylthiophene-2-carbaldehyde (0.3931 mmol, 1.0
eq.) and 261.4 mg of 4-(diphenylamino)phenylboronic acid (0.9041 mmol, 2.3 eq.) were taken in toluene
under argon atmosphere. Then, 2 molar potassium carbonate solution (20 mL) and 7.2 mg of
tetrabutylammonium iodide (0.0196 mmol, 5 mol%) followed by 45 mg of tetrakistriphenylphosphine
palladium (0) (0.0393 mmol, 10 mol %) were added and refluxed for overnight. The crude mixture was
worked up with ethyl acetate / water (3 x 30 mL) and purified by silica column chromatography afforded
compound VIII as yellow solid (283.1 mg, 87%). R = 0.35 (5 % EA/PE); FT-IR (KBr): 3039, 2925,
f
-1 1
2
855, 1903, 1668, 1592, 1487, 1277, 1158, 1820, 820, 750, 690, 510 cm ; H NMR (500 MHz, CDCl δ
3,
9

in ppm): 0.72 (t, 6H, J = 5.0 Hz), 0.77 (t, 3H, J = 5.0 Hz), 0.98-1.01 (m, 17H), 1.07-1.12 (m, 4H), 1.18-
.26 (m, 8H),1.54-1.60 (m, 3H), 1.91-1.96 (m, 4H), 2.66 (t, 2H, J = 5.0 Hz), 6.95 (t, 2H, J = 5.0 Hz), 7.07
t, 6H, J = 5.0 Hz), 7.17-7.22 (m, 4H), 7.35 (t, 2H, J = 5.0 Hz), 7.48 (t, 4H, J = 5.0 Hz), 7.60 (s, 1H), 7.67
1
(
(
1
3
d, 2H, J = 5.0 Hz), 9.8 (s, 1H); C NMR (125 MHz, CDCl , δ in ppm): 14.0, 14.06, 22.5, 22.6, 23.9,
3
2
1
1
8.9, 29.1, 29.2, 29.25, 30.0, 30.8, 31.6, 31.7, 40.4, 55.3, 119.9, 120.3, 121.9, 123.0, 123.5, 123.9, 124.4,
25.7, 127.8, 128.1, 129.3, 131.9, 135.3, 138.7, 139.1, 140.1, 140.18, 140.8, 141.6, 147.2, 147.6, 149.6,
+
+
51.4, 151.7, 182.8. HRMS: m/z calcd for C H NOS [M] : 827.5100, found: 828.5193 [M+H] .
5
8
69
Experimental procedure for (E)-2-Cyano-3-(5-(7-(4-(dimethylamino)phenyl)-9,9-dioctyl-9H-fluoren-2-
yl)-4-hexylthiophen-2-yl)acrylic acid (FMD)
1
00
mg
of
5-(7-(4-(diphenylamino)phenyl)-9,9-dioctyl-9H-fluoren-2-yl)-4-hexylthiophene-2-
carbaldehyde (0.1420 mmol, 1.0 eq.) and 60.3 mg of cyanoacetic acid (0.7100 mmol, 5.0 eq.) were
dissolved in a mixture of acetonitrile (6 mL) and chloroform (4 mL). To this mixture, catalytic amount of
piperidine was added and refluxed for 12h. Afterward, the reaction mixture was cooled to room
temperature and acidified with HCl solution (1.5 M). Then it was extracted with chloroform/water (3 X
3
0 mL) and the organic potion was dried over anhydrous sodium sulphate followed by concentrated under
reduced pressure. The crude product was purified by silica column using DCM/MeOH/AcOH as an eluent
afforded the expected FMD dye as red semisolid (60.2 mg, 55%). R = 0.64 (DCM/MeOH/AcOH
f
-1
9
:0.5:0.5); FT-IR (KBr): 3406, 2923, 2854, 2216, 1710, 1587, 1459, 1255, 1170, 901, 817, 728, 649 cm ;
1
H NMR (500 MHz, CDCl δ in ppm): 0.80 (t, 9H, J = 5.0 Hz), 0.83-0.89 (m, 7H), 1.04-1.17 (m, 11H),
3,
1
.21-1.30 (m, 15H), 1.59-1.65 (m, 2H), 1.95-2.06 (m, 2H), 2.70 (t, 1H, J = 5.0 Hz), 3.02 (s, 6H), 6.76
(d, 1H, J = 5.0 Hz), 6.88 (t, 2H, J = 10.0 Hz), 7.51-7.55 (m, 3H), 7.59 (d, 1H, J = 5.0 Hz), 7.68 – 7.71 (m,
1
3
3
2
1
1
H), 8.33 (s, 1H); C NMR (125 MHz ,CDCl δ in ppm): 14.0, 14.05, 22.5, 22.7, 23.9, 28.8, 29.2,
3,
9.25, 29.7, 30.8, 31.6, 31.7, 40.4, 40.9, 55.2, 113.1, 113.3, 119.8, 120.3, 120.7, 122.4, 123.2, 125.3,
27.8, 128.0, 128.3, 130.2, 131.3, 133.5, 138.3, 140.2, 140.6, 140.7, 141.9, 146.8, 149.7, 149.8, 151.3,
+
+
51.7. HRMS: m/z calcd for C H N O S [M] : 770.4845, found: 771.4944 [M+H] .
5
1
66
2
2
1
0

Experimental procedure for (E)-2-cyano-3-(5-(7-(4-(Diethylamino)phenyl)-9,9-dioctyl-9H-fluoren-2-
yl)-4-hexylthiophen-2-yl)acrylic acid (FED)
1
00 mg of 5-(7-(4-(diethylamino)phenyl)-9,9-dioctyl-9H-fluoren-2-yl)-4-hexylthiophene-2-carbaldehyde
(0.1365 mmol, 1.0 eq.) and 58 mg of cyanoacetic acid (0.6825 mmol, 5.0 eq.) were dissolved in a
mixture of acetonitrile (6 mL) and chloroform (4 mL). To this mixture, catalytic amount of piperidine was
added and refluxed for 12 hours. Afterward, the reaction mixture was cooled to room temperature and
acidified with HCl solution (1.5 M). Then it was extracted with chloroform / water (3 x 30 mL) and the
organic potion was dried over anhydrous sodium sulphate followed by concentrated under reduced
pressure. The crude product was purified by silica column using DCM/MeOH/AcOH (9:0.5:0.5) as an
eluent afforded the expected FPD dye as red semisolid (67.6 mg, 62%). R = 0.64 (DCM/MeOH/AcOH
f
9
4
:0.5:0.5); FT-IR (KBr): 3418, 2924, 2857, 2214, 1731, 1595, 1524, 1457, 1379, 1260, 1086, 806, 654,
-1 1
67 cm . H NMR (500 MHz, CDCl δ in ppm): 0.79 (t, 7H, J = 5.0 Hz), 0.83-0.89 (m, 9H), 1.06-1.11
3
,
(m, 15H), 1.16-1.20 (m, 5H), 1.30-1.33 (m, 7H), 1.61-1.67 (m, 3H), 1.99-2.03 (m, 5H), 2.73 (t, 2H, J =
1
0.0 Hz), 3.48 (q, 4H, J = 10.0 Hz), 6.77 (d, 1H, J = 10.0 Hz), 7.14 (s, 1H), 7.43 (s, 1H), 7.45 (d, 1H, J =
1
3
1
0.0 Hz), 7.51-7.58 (m, 3H), 7.66 (d, 1H, J = 5.0 Hz), 7.73-7.76 (m, 3H), 8.33 (s, 1H); C NMR (125
MHz, CDCl , δ in ppm): 14.04, 14.7, 22.5, 22.7, 23.9, 28.7, 29.1, 29.2, 29.7, 30.0, 30.8, 31.6, 31.7, 31.9,
3
4
0.4, 55.3, 113.7, 114.5, 116.3, 119.9, 120.4, 120.8, 123.3, 128.0, 128.33, 128.36, 131.4, 133.5, 140.4,
+
1
40.8, 141.8, 147.1, 151.4, 151.8; HRMS: m/z calcd for C H N O S [M] : 798.5158, found: 799.5251
5
3
70
2
2
+
[
M+H] .
Experimental procedure for (E)-2-Cyano-3-(5-(7-(4-(diphenylamino)phenyl)-9,9-dioctyl-9H-fluoren-2-
yl)-4-hexylthiophen-2-yl)acrylic acid (FPD)
1
00
mg
of
5-(7-(4-(diphenylamino)phenyl)-9,9-dioctyl-9H-fluoren-2-yl)-4-hexylthiophene-2-
carbaldehyde (0.1207 mmol, 1.0 eq.) and 51.3 mg of cyanoacetic acid (0.6035 mmol, 5.0 eq.) were
dissolved in a mixture of acetonitrile (6 mL) and chloroform (4 mL). To this mixture, catalytic amount of
1
1

piperidine was added and refluxed for 12 hours. Afterward, the reaction mixture was cooled to room
temperature and acidified with HCl solution (1.5 M). Then it was extracted with chloroform/water (3 x 30
mL) and the organic portion was dried over anhydrous sodium sulphate followed by concentrated under
reduced pressure. The crude product was purified by silica column using DCM/MeOH/AcOH (9:0.5:0.5)
as an eluent afforded the expected FPD dye as red semisolid (85.3 mg, 79 %). R = 0.64
f
(DCM/MeOH/AcOH 9:0.5:0.5); FT-IR (KBr): 3198, 2924, 2856, 2345, 2214, 1724, 1605, 1424, 1382,
-
1 1
1
1
2
278, 1092, 703, 617 cm ; H NMR (500 MHz, CDCl δ in ppm): 0.77-0.80 (m, 4H), 0.86-0.89 (m, 5H),
3,
.01-1.07 (m, 4H), 1.22-1.28 (m, 28H), 1.62-1.66 (m, 1H), 2.00-2.04 (m, 2H), 2.35 (t, 2H, J = 5.0 Hz),
.72 (t, 1H, J = 5.0 Hz), 7.01-7.06 (m, 2H), 7.14-7.18 (m, 6H), 7.26-7.29 (m, 4H), 7.43 (d, 1H, J = 1.7.0
1
3
Hz), 7.46-7.47 (m,1H), 7.56-7.59 (m, 3H), 7.75-7.77 (m, 3H), 8.32(s, 1H); C NMR (125 MHz, CDCl , δ
3
in ppm): 14.0, 14.1, 22.5, 22.7, 23.8, 24.7, 28.77, 29.6, 29.7, 30.7, 31.7, 31.9, 34.0, 40.4, 55.3, 114.0,
1
1
15.8, 120.0, 120.4, 120.9, 122.9, 123.2, 123.9, 124.4, 125.7, 127.8, 128.3, 129.3, 131.3, 133.3, 135.2,
39.0, 140.2, 140.9, 141.9, 147.2, 147.6, 151.5, 151.8, 168.0, 179.8. HRMS: m/z calcd for C H N O S
6
1
70
2
2
+
+
[
M] : 894.5158, found: 895.5255 [M+H] .
Experimental procedure for DSSC fabrication
2
FTO glasses were cut in to 2.0 X 2.5 cm dimension and cleaned in detergent solution, distilled water,
acetone and isopropyl alcohol in an ultrasonic bath for 30 min and it subjected to UV/O 3 treatment. The
TiO photoanodes was prepared by taking 600 mg of TiO powder mixed with PEO (100 mg) and PEG
2
2
(100 mg) in a mortar and grinded well for 5 min by adding 2–3 mL of water to form pastes. Then, 150 µL
of acetyl acetone and one drop of Triton-X-100 have been added and it ensured that the mixtures were
stirred overnight. The TiO photoanodes pastes were coated on a conducting side of FTO glasses plates
2
by doctor-blade technique with the aid of polyimide tape. To obtain a stable photoanodes, the films were
o
dried at room temperature, sintered at 500 C for 30 min in a muffle furnace and then allowed to cool
naturally to room temperature. The prepared TiO photoanodes were immersed in ethanolic solution of
2
0
.3 mM FPD, FED, FMD dye for 24 h at room temperature under dark condition. The dye-adsorbed
1
2

photoanodes were withdrawn from the solution and immediately rinsed using ethanol. The platinum
deposited FTO counter electrodes were prepared by drop casting of platinic acid over FTO glass plates
o
and heated at 435 C. The gel electrolyte was prepared 0.032 g of iodine crystal was dropped in 240 µL of
ionic liquid (1-Methyl 3-propylimidizolium iodide). Then, 0.183 mL of TBP (4-tert butylpyridine), 1 mL
of acetonitrile and 0.5 g of PEO were introduced and stirred well. To this 1 mL of acetonitrile and 0.035 g
of PEG were added and stirred well thoroughly. Finally, 0.5 mL acetonitrile and 0.033 g of lithium iodide
were added and the resultant mixture was stirred overnight. The dye adsorbed TiO photoanode
2
containing a drop of gel electrolyte was placed over the platinized counter electrode and clamped firmly
2
together. An active area of 0.125 cm was used to measure the cell performance. The photocurrent-
voltage signal (J–V curve) measurements were performed in a Sciencetech (SS 1.6 K, AAA) solar
-2
simulator under the light illumination of 100 mW cm with the optical filter of AM 1.5 G (USA) attached
with Keithley Instrument source meter (Model No.: 2400) and electrochemical properties of the
fabricated DSSCs were studied using Bio-Logic SP-150 Instrument.
Results and Discussion
Chemical Synthesis
Based on their good optoelectronic properties and application in photovoltaics, we intended to choose
[
11]
fluorenothienyl cores as π-spacers for our study. As far as the auxiliary donor is concerned, structurally
comparable groups such as -NPh , -NEt and -NMe groups were chosen. As depicted in Scheme 1, the
2
2
2
chemical synthesis started from the electrophilic bromination of commercially available 9H-fluorene I to
obtain 2,7-dibromo-fluorene II. The later upon nucleophilic substitution with 2.5 equivalence of n-
octylbromide affords 2,7-dibromo-9,9-dioctyl-9H-fluorene III. Suzuki-Miyaura coupling between III and
3
-hexylthiophene-2-boronic acid pinacol ester (1.0 eq) results in cross-coupled intermediate IV. This was
followed by Vilsmeier-Haack formylation to obtain the carbaldehyde intermediate V. Suzuki coupling of
V with corresponding 4-aminophenylboronic acid partners lead to the formation of penultimate
intermediates VI, VII and VIII. Knoevenagel condensation of these intermediates with cyanoacetic acid
1
3

affords the corresponding dye FMD, FED and FPD. All dyes were obtained in synthetically acceptable
1
13
yields and their structural characterization was performed by spectroscopic analysis (FT-IR, H NMR, C
-1
NMR and MS). As an illustrative example, the FT-IR spectrum of FMD displayed bands at 3406 cm
-1
and 1710 cm corresponding to the hydroxyl and carbonyl stretching of -COOH group respectively. The
-1
1
other distinctive peak at 2216 cm corresponds to -CN functionality. The H NMR spectrum of FMD
1
3
recorded in CDCl showed a singlet at δ 8.33 ppm was assigned to the olefinic proton. In C NMR
3
H
spectrum, a peak at δ 151.7 ppm indicates the presence of carboxyl carbon. Finally, HRMS data
C
+
exhibited a molecular ion peak at m/z = 771.4944 [M+H] unambiguously confirmed the product.
1
4

O
N
Br
2.1 eq)
(
Br
O
(2.5 eq)
Br
Br
Br
Br
FeBr3 (10 mol%)
KOH (10 eq), DMSO, reflux, 5
days
CH Cl , reflux, 12 h
I
2
2
II
7
6% yield
III
9
8% yield
O
B
S
POCl3 (2.5 eq)
DMF (2.5 eq)
O
(1.0 eq)
Br
Br
(
Ph P) Pd (10 mol%), TBAI (5 mol%)
1,2-DCE, reflux, 12 h
54% yield
S
O
3
4
S
K CO , toluene, reflux, 12 h
2
3
V
5
5% yield
IV
NC
COOH
5 eq)
(
Me
COOH
CN
Me
OH
N
O
CH CN / CHCl (6:4)
piperidine, reflux, 12 h
55% yield
N
S
3
3
S
B
Me
Me
OH
Me
VI
N
FMD
Me
OH
B
OH
Et
NC
COOH
(5 eq)
N
(
1.1 eq)
Et
Et
V
N
O
Et
COOH
CN
S
(
Ph3P)4Pd (10 mol%), TBAI (5 mol%)
K CO , toluene, reflux, 12 h
Et
N
CH CN / CHCl (6:4)
S
3
3
VII
2
3
Ph
N
piperidine, reflux, 12 h
62% yield
Et
FED
FPD
Ph
OH
B
OH
NC
COOH
(5 eq)
Ph
N
O
Ph
COOH
CN
S
N
Ph
CH CN / CHCl (6:4)
S
3
3
VIII
Ph
piperidine, reflux, 12 h
9% yield
7
Scheme 1. Chemical synthesis of FMD, FED and FPD dyes
Photophysics
-
The UV-visible absorption spectra of FMD, FED and FPD dyes recorded as chloroform solution (c = 10
4
M) are shown in Figure S1 and the related optical data are summarized in Table 1. All dyes exhibited
more or less identical peak pattern (Figure 1). The absorption maxima (λmax) of the dyes FMD, FED,
FPD (411, 416, 426 nm) were attributed to the intramolecular charge-transfer (ICT) between the HOMO
of the donor and LUMO of the acceptor, while that in the range of 300-350 nm is assigned to π-π*
1
5

-1
-1
transition. The molar absorptivity (ε) of FPD was found to be higher (120650 M cm ) than FMD (13650
-1
-1
-1
-1
M cm ) and FED (22490 M cm ), which can be ascribed to the extensive π-delocalization offered by
-4
the triphenylamine group. Photoluminescence of the dyes measured in chloroform (C = 10 M) revealed
a red shift in the emission band (λ ) for all dyes (FMD = 556, FED = 540, FPD = 630 nm), indicating
em
the facile development of excited species leading to low energy emissions (Figure S1). As depicted in
Table S1, the emission quantum yield (φ) of dyes calibrated against the reference quinine sulphate (φ =
[
12]
5
5%) in H SO (0.05 M) was found to be 30.7% (FMD), 19.8% (FED) and 22.7% (FPD). From the
2 4
intersection of normalized absorption and photoluminescence spectra (λ ), the zeroth-zeroth energy (E
0-
0
-0
0)
of all dyes were calculated, which corresponds to 2.53, 2.58 and 2.48 eV for FMD, FED and FPD
respectively. When sensitized on TiO surface, all dyes exhibited an extended absorption wavelength
2
opt
(
λonset) of 520 nm (FMD), 532 nm (FED) and 540 nm (FPD) relating to a reduced optical band gap (Eg )
of 2.38, 2.33 and 2.29 eV respectively (Figure 1b). This decrease in solid state band gap compared to
solution mode suggests the J-aggregation of dyes on the semiconductor surface. Notably, this reduction in
optical band gap or otherwise extended light absorption towards higher wavelength is crucial for the
enhancement of spectral overlap with the solar photon flux spectrum and should be advantageous for
attaining a higher short circuit current density (J ) when applied in DSSCs.
SC
Figure 1. UV-vis absorption of dyes a) in chloroform; b) on TiO surface
2
1
6

Table 1. Optical data of FMD, FED and FPD
a
-1
-1
b
c
opt
g
d
Dyes
λ
max /
λ
0-0 (nm)
ε (M cm )
E0-0 (eV)
λ
max /
λonset (nm)
E
(eV)
FMD
FED
FPD
411 / 490
416 / 486
426 / 498
13650
22490
120650
2.53
2.55
2.48
420 / 520
404 / 532
407 / 540
2.38
2.33
2.29
a
-4
b
Absorption recorded in chloroform (C = 10 M). E energy was calculated using
0
-0
the factor E0-0 = 1240/λ , where λ is the intersection wavelength between the
0-0
0-0
normalized absorption and fluorescence emission of the lowest energy band.
c
d
Absorption of dyes recorded on TiO surface. Optical band gap calculated from
2
opt
λonset of TiO -anchored dyes using E = 1240/λonset
2
g
Electrochemistry
The electrochemical characteristics of the dyes were investigated by performing cyclic voltammetry (CV)
-3
in CHCl (2.0 × 10 M) utilizing nBu NPF (0.1 M) as supporting electrolyte. Platinum foil, glassy
3
4
6
carbon and Ag/AgCl (3M KCl) were used as counter, working and reference electrodes respectively. As
+
depicted in Figure S3, a standard ferrocenium/ferrocene (Fc /Fc) redox couple showing a half wave
potential (E ) of 0.575 V vs. Ag/AgCl was used for calibration. All dyes exhibited an irreversible or
1
/2
quasi-reversible oxidation potential with no cathodic current, which could be attributed to the subsequent
[
13]
complex reactions spurred by electron transfer of the initially formed radical-cationic species. Analysis
of CV results revealed that the oxidation potential (E ) of dyes shifts gradually with increase in the donor
ox
size (FMD = 0.80 V; FED = 0.85 V; FPD = 0.96 V) and the obtained values are consistent with electron-
[14]
[15]
rich amine oxidation. Using Pommerehne empirical relation, the HOMO energy (EHOMO) of the dyes
was calculated from the respective oxidation potential (FMD = -5.02 eV; FED = -5.07 eV; FPD = -5.18
eV). Because of the irreversibility, the LUMO energies (ELUMO) of all dyes were calculated by using the
factor ELUMO = EHOMO + E . According to which, E
of FPD (-2.70 eV) was found to be lower in
LUMO
0-0
energy compared to FMD (-2.49 eV) and FED (-2.52 eV). Nevertheless, ELUMO of all dyes are located
[
16]
sufficiently higher than the conduction band of TiO (E = -4.0 eV)
which is suggestive of the
feasibility of electron injection from the excited state of dye. Also, the appropriately large energy
difference between the E of TiO and E of dye indicates the need to use 4-tert-butylpyridine (TBP)
2
cb
cb
2
LUMO
1
7

for the treatment of iodolyte during device fabrication. By doing so, the E of TiO shifts negatively to
cb
2
enhance the open-circuit voltage (V ) which in turn improves the power conversion efficiency of the
oc
[
17]
cell. On the other hand, the electron injection efficiency (φ ) of a dye is influenced by the free energy
inj
of electron injection (ΔG ) and is therefore a crucial parameter to predict the short circuit current density
inj
[
18]
(
J ). As the ΔG values of the excited state dyes decrease in the order of FPD > FED > FMD, it
SC inj
suggests the same order of driving force for electron injection.
-3
Figure 2. CV of dyes in CHCl (2.0 × 10 M) using nBu NPF (0.1 M) as supporting electrolyte at a scan
3
4
6
-1
rate of 50 mV s
Table 2. Electrochemical data of FMD, FED and FPD
a
b
c
d
e
Dyes
E
ox (V)
EHOMO (eV)
E
LUMO (eV)
Eox* (V)
ΔGinj (V)
FMD
FED
FPD
0.80
0.85
0.96
-5.02
-5.07
-2.49
-2.52
-2.70
-1.73
-1.70
-1.52
2.27
2.30
2.48
-5.18
a
b
First oxidation potential. E
was calculated using the empirical relation,
HOMO
EHOMO = -[(E - E (Fc vs Ag/AgCl (3M KCl)) + 4.8]; E1/2 (Fc vs Ag/AgCl
ox
1/2
c
d
(
=
3M KCl) = 0.575 V; E
E -E ; ΔG = E *-E
ox 0-0 inj ox cb
was calculated using ELUMO = EHOMO + E ; E *
LUMO 0-0 ox
e
1
8

Computational chemistry
Density functional theory (DFT) calculations were utilized for the structural optimization of all dyes at
[
19]
the B3LYP/6-31G**(d,p) level basis set.
The resulting energy minimized structure of all dyes in
chloroform are shown in Figure 3. The selected carbon-carbon (C-C) bond distance between the various
inter-linking units are designated as D1, D2 and D3. For all dyes, the bond length was calculated to be in
the range of 1.43 to 1.48 Ǻ which is expected for C-C with partial double bond character. This shortened
C-C bond length in all dyes is an indicative of the operation of strong intramolecular charge transfer
[
20]
(
ICT) as characteristic of any dipolar dye. Likewise, the torsion angle denoted as φ and φ between the
1 2
aromatic linkers was computed to glean information on the co-planarity of the π-conjugated framework.
o
The dihedral angle (φ ) between the donor aryl ring and the π-spacer was found to be around 35 (+syn-
1
~
o
clinal) and the angle between the π-spacer and acceptor thienyl ring was observed to be ~ -40 (-syn-
clinal). The estimated torsional potential between the aryl linkers infer significant planarity, which in turn
[
21]
is crucial for effective π-electron delocalization to improvise the optoelectronic properties.
Figure 3. Energy minimized geometry, selected bond length and torsional angle of FMD, FED and FPD
dyes (from left to right)
The electronic structure of frontier π-molecular orbitals of all dyes as depicted in Figure 4, suggests the
delocalization of HOMO in the donor fragment comprising disubstituted aniline and fluorothiophene
cores. On the other hand, the LUMO is distributed mostly on the electron-withdrawing cyanoacrylic acid
1
9

and thienyl scaffold. This observation is typical of an asymmetric push-pull transition revealing a facile
[
22]
intramolecular charge transfer from HOMO of donor to LUMO of acceptor. The computed HOMO-
LUMO energies and band gap of all dyes were found to be in fair correlation with the results obtained
from photophysical and electrochemical studies. The deep location of HOMO of all dyes (-4.99 eV;
-
-
FMD, -4.92; FED and -5.01 eV; FPD) compared to the I /I redox potential (-4.6 eV) ensures dye-
3
regeneration and hole transport. On the other hand, the energy level of LUMO of all dyes were found to
be higher than the conduction band of TiO (E = -4.0 eV) which, therefore could favour the injection of
2
cb
photoexcited electrons. As depicted in Table 3, we also gain insights on other parameters such as major
vertical transitions, oscillator strength (f), light harvesting efficiency (LHE) by time-dependent density
functional theory (TD-DFT). All dyes exhibited two bands in the visible region of the electronic spectrum
due to π-π* and ICT with predominant S →S transitions (Figure S4 to S6). It was found that the f values
0
1
corresponding to π-π* were significantly higher than that to ICT. This is an indication that the photon
harvesting is majorly contributed by π-π* transition (HOMO-1→LUMO) compared to ICT (HOMO →
LUMO). The resulting LHE was found to be in the order FPD > FED > FMD which is consistent with
the increasing size of the donor substituents. Since the life time of photoexcited electron correlates
directly to the electron injection efficiency to semiconductor, we also calculated the radiative life time (τ)
2
from the computed excitation energies using the formula, τ = 1.499/f E , where f is the oscillator strength
-1
of the electronic state and E is the excitation energy in cm . Analysis of τ values suggests that FMD
3.48 ns) has slightly higher life time than FED (3.35 ns) and FPD (3.35 ns). This minor increase in
(
lifetime for FMD is expected to exhibit relatively better retardation of the charge recombination. In other
words, FMD dye will be more stabilized in its cationic form for a long time to realize long-term stability.
Moreover, the somewhat closer τ values illustrates that the donor strength of the investigated dyes does
not have substantial control over electron transport properties.
2
0

Figure 4. HOMO-LUMO electron distributions of FMD, FED and FPD dyes at B3LYP/6-31G** level
Table 3. Computed electronic transition of FMD, FED and FPD
Dyes Electronic
transition (nm)
→S 582 2.12 0.3428
S0→S1 428 2.89 0.792
→S 598 2.07 0.3311
S0→S1
→S
S0→S1
λ
eV
f
Major contribution
Assignment
LHE
τ (ns)
max
FMD
FED
FPD
S
0
1
HOMO → LUMO (99%)
HOMO-1 → LUMO (97%)
HOMO → LUMO (99%)
HOMO-1 → LUMO (96%)
HOMO → LUMO (99%)
ICT
π-π*
ICT
π-π*
ICT
π-π*
0.5458
0.8385
0.5334
0.8517
0.4592
0.8668
3.48
3.35
3.35
S
0
1
431 2.87
574 2.15
0.829
0.267
S
0
1
442 2.80 0.8757 HOMO-1 → LUMO (97%)
At this point of our study, we became interested in calculating the secondary quantum chemical
parameters in order to compare the molecular structure, stability and chemical reactivity of the dyes
(Table 4). As the polarizability of different sensitizers can alter the electronic properties of TiO2
[
23]
semiconductor and could influence the nature of dye-acceptor interactions, we compared the computed
2
1

dipole moment (ρ) of the dyes. A high dipole moment for all dyes was observed which is typical of a
[
24]
strong intermolecular interaction in dipolar push-pull molecules. Especially, FED with highest ρ value
of 14.95 D could induce an increase in the local concentrations of the acceptor species at the TiO2
surface. As a consequence, it favours the penetration of acceptor species into the adsorbed dye layer for a
fast electron recombination. Based on the computed chemical potential (μ), the electron escaping
tendency of FMD and FPD (-3.80 eV) was found to be slightly greater than FED (-3.76 eV). This
suggests that the electrons in FED are less probable to escape from a system in equilibrium compared to
FMD and FPD. On the other hand, the chemical hardness (η) follows the order of FPD (1.23 eV) > FMD
(1.18 eV) > FED (1.15 eV). The higher value of FPD can be understood as a resistance to the charge
transfer of the system, while the lower value of FED suggests that its electron density could be readily
-1
influenced and would be more reactive. In other words, the lower softness of FPD (σ = 0.81 eV )
-1
-1
compared to FMD (σ = 0.84 eV ) and FED (σ = 0.86 eV ) further indicates the impede of electron
redistribution during chemical reaction. However, the computed electronegativity (χ) for all dyes were
found to be almost identical; FMD (3.80 eV), FED (3.76 eV) and FPD (3.80 eV), which is confirmatory
of no significant electronegativity change by altering the substituent on the donor fragment.
Table 4 Quantum chemical parameters of FMD, FED and FPD
a
b
c
d
Dyes
E
HOMO (eV)
E
LUMO (eV) ρ (D) μ (eV)
η (eV)
σ (eV)
χ (eV)
FMD
FED
FPD
-4.99
-4.92
-5.03
-2.62
-2.61
-2.57
14.78
14.95
12.30
-3.80
-3.76
-3.80
1.18
1.15
1.23
0.84
0.86
0.81
3.80
3.76
3.80
μ = (EHOMO + ELUMO)/2; η = (ELUMO-EHOMO)/2; σ = 1/ η; χ = -(EHOMO +ELUMO)/2
Photoelectrochemistry
To study the photovoltaic performance of the dyes, DSSC based on gel electrolyte were fabricated on
TiO films sensitized with each dye and evaluated under simulated solar AM 1.5G irradiation of 100
2
2
mW/cm . The respective open-circuit photovoltage (V ), fill factor (FF) short circuit photocurrent
OC
density (J ) and power conversion efficiency (η) were derived from Figure S7 and listed in Table 5. At
SC
-2
the outset, FPD (JSC = 7.42 mA cm ; V = 0.785 V, FF = 0.597) displayed better current–voltage (J-V)
OC
2
2

-2
characteristics compared to FED (J = 5.26 mA cm ; V = 0.697 V, FF = 0.571) and FMD (J = 5.26
SC
OC
SC
-2
mA cm ; V = 0.678 V, FF = 0.598). On the basis of the aforementioned parameters, the solar energy-
OC
to-electricity conversion efficiency (η) follows the order: FPD (3.48%) > FED (2.09%) > FMD (2.10%).
The improved efficiency of FPD over FMD and FED could be attributed to the combination of steric and
electronic factors brought by the two phenyl groups of the donor moiety, thereby impedes charge
[
25]
recombination.
The reasonable efficiency of the dyes prompted us to investigate the effect of co-
[
26]
adsorbent in the form of chenodeoxycholic acid (CDCA), which is known to enhance η by diminishing
the aggregation of dyes on the TiO surface (Figure 5a). The result of which clearly showed a significant
2
-2
-2
increase in J for all dyes (FMD: JSC = 6.10 mA cm ; FED: J = 9.53 mA cm FPD: J = 8.98 mA
SC
SC
SC
-2
cm ), which is indicative of favourable electron transfer through an event of CDCA facilitated reduced
aggregation. Overall, the η of cells fabricated with CDCA noticeably increased from 2.10 to 2.57% for
FMD, 2.09 to 3.44% for FED and 3.48 to 4.1% for FPD. Moreover, the trend in efficiency with and
without CDCA treatment was retained in good agreement. Particularly, the efficiency of FPD dye was
very much closer to the values of standard N719 dye, which exhibited an efficiency of 4.14% under
optimized conditions (Figure S9). Though, the efficiencies reported are only moderate, it should be
remembered that the cells are constructed on gel electrolyte (not on liquid electrolyte) for stability
reasons. Moreover, our efficiencies are based only on using TiO as photoanodes and there is scope to
2
improve the efficiency by using alternative transparent electrodes like TiO -modified conductive carbon
2
[
27]
nanotubes (CNTs). Nevertheless, our photovoltaic results established the correlation of donor bulkiness
to the DSSC performance, which is the actual focus of our study. To unravel the electronic processes at
the device interface, the electrochemical impedance spectroscopy (EIS) was used as an analytical tool. To
analyse the interfacial charge transfer process that takes place in the cell, EIS was recorded in a frequency
range between 0.01 Hz and 100 kHz. In the Nyquist plots (Figure 5b), the first semicircle describes the
charge-transfer process taking place at the counter electrode/electrolyte interface and the second
semicircle reveals charge-recombination process at the TiO /dye/electrolyte interface of the device. The
2
2
3

components such as R , R , R in the equivalent circuit corresponds to the total series resistance. The
s
ct1
ct2
first semicircle is due to charge transfer resistance (R ) at the counter electrode/electrolyte interface,
ct1
while the second larger semicircle is due to the charge transfer resistance (R ) at
ct2
dye/photoanode/electrolyte interface and R being the sheet resistance of the transparent conductive
s
oxide. As expected, the R values of dyes saturated with CDCA was found to be lesser (FMD = 38.34;
s
FED = 31.58; FPD = 28.75) than those without CDCA treatment (FMD = 41.3; FED = 35.82; FPD =
29.1), because of the better power conversion efficiency of the latter series. In Bode phase plots (Figure
5c), the maximum frequency peak (Log f) for CDCA treated dyes (FMD = 1.641; FED = 1.554; FPD =
1.214) was shifted to lower frequency compared to untreated dyes (FMD = 1.641; FED = 1.554; FPD =
1.384). This is suggestive of a longer electron life time enabled by the presence of CDCA. The frequency
shift towards the lower region corresponding to increased electron life-time (τ ) was calculated from the
n
formula τ = 1/(2πf). Analysis of the τ values further revealed that dyes in combination with CDCA
n
n
exhibited higher electron life time (FMD = 97 ms; FED = 102 ms; FPD = 131 ms) over dyes without
CDCA saturation (FMD = 92 ms; FED = 102 ms; FPD = 115 ms). Nevertheless, FPD in both
circumstances showed a high τ value and could be attributed to the favourable electron stabilization by
n
the presence of two phenyl groups in the donor segment.
2
4

Figure 5. (a) J-V curves of fabricated DSSC; (b) EIS Nyquist plots and (c) EIS Bode phase plots
a
Table 5. Photovoltaic data of DSSC based on FMD, FED and FPD
a
-2
Dyes
V
OC (V)
J
SC (mA cm )
FF
η (%)
R
s
(Ω)
R
CT1 (Ω)
R
CT2 (Ω) Log f (Hz)
τ
n
(ms)
92
102
115
97
b
FMD
0.678
0.697
0.785
0.703
0.704
0.784
5.26
5.26
7.42
6.10
9.53
8.98
0.590
0.571
0.597
0.60
0.514
0.578
2.10 (2.09 ± 0.01)
2.09 (2.07 ± 0.02)
3.48 (3.45 ± 0.03)
2.57 (2.55 ± 0.02)
3.44 (3.41 ± 0.03)
41.30
35.82
29.10
38.34
31.58
28.75
8.76
38.8
33.0
18.11
40.25
34.86
23.24
1.725
1.554
1.384
1.641
1.554
1.214
b
FED
FPD
10.78
4.49
6.26
6.85
b
c
FMD
c
FEDc
102
131
FPD
4.10 (4.09 ± 0.01)
4.17
a
-2
2
a
b
Irradiated light: AM 1.5G illumination (100 mW cm ) on a working area of 0.125 cm ; Dyes without CDCA; Dyes with CDCA
2
5

Conclusions
In summary, three new organic push-pull dyes based on D-D-π-A type architecture (FMD,
FED and FPD) were synthesized. The donor size of these dyes was systematically varied by
groups such as -NMe , -NEt and -NPh to investigate their effect the on the optoelectronic
2
2
2
properties and quasi-state DSSC performance. This minute variation of donor size suggests
that the dye possessing a bulky -NPh group exhibits better optical (molar absorptivity and
2
band gap) and electrochemical (HOMO-LUMO) properties to match the requirements for
DSSC. The experimental photophysical and electrochemical results correlated well with DFT
calculations. Computed quantum chemical properties and other molecular parameters were
found to be in favour of FPD as better dye candidate. In fact, the better physical and chemical
properties of FPD was reflected in the photovoltaic performance of prototype DSSC devices,
which revealed a power conversion efficiency in the order of FPD (4.10%) > FED (3.44%) >
FMD (2.57%). Electrochemical impedance spectroscopy also confirmed the trend by
exhibiting higher life time for FPD followed by FED and FMD. Our overall study showed
that the steric and electronic factors of donor moieties are crucial in controlling the
optoelectronic properties of the push-pull dyes, which in turn governs the DSSC
performance. We are currently working on improving the device efficiency by a logical
structural randomization of the dyes and these results will be communicated shortly.
Declaration of competing interest
There are no conflicts to declare.
2
6

Acknowledgements
Financial support by Department of Science and Technology (DST), India to carry out this
work through INSPIRE faculty award (IFA14-MS27) is greatly acknowledged. One of the
authors C. K. would like to thank Mr. P. Santhosh Kumar and Ms. M. R. Anagha, the MSc
project students for their assistance during chemical synthesis. We thank Ms. K. B. Bhojanaa
for fruitful discussions on photoelectrochemistry part. We also acknowledge the Central
Instrumentation Facility (CIF), CSIR-CECRI for providing the required analytical facilities.
Notes and references
[
1] (a) Carell, A, Borbone F, Centore R. Research Progress on Photosensitizers for
DSSC. Front Chem 2018; 3: 481. https://doi.org/10.3389/fchem.2018.00481; (b) Hug
H, Bader M, Mair P, Glatzel T. Biophotovoltaics: Natural pigments in dye-sensitized
solar
cells.
Appl
Energ
2014;
115:
216-225.
https://doi.org/10.1016/j.apenergy.2013.10.055.
[
[
2] Mishra A, Fischer MKR, Bäuerle P. Metal-Free Organic Dyes for Dye-Sensitized
Solar Cells: From Structure: Property Relationships to Design Rules. Angew Chem
Int Ed 2009; 48: 2474-2499. https://doi.org/10.1002/anie.200804709.
3] Kakiage K, Aoyama Y, Yano T, Oya K, Fujisawa Ji, Hanaya M. Highly-efficient dye-
sensitized solar cells with collaborative sensitization by silyl-anchor and carboxy-
anchor
dyes.
Chem
Commun
2015;
51:
15894-15897.
https://doi.org/10.1039/c5cc06759f.
[
4] (a) Ooyama Y, Harima Y. Photophysical and electrochemical properties, and
molecular structures of organic dyes for dye-sensitized solar cells. ChemPhysChem
2
012; 13: 4032-4080. https://doi.org/10.1002/cphc.201200218. (b) Chaurasia S, Liang
CJ, Yen YS, Lin JT, Sensitizers with rigidified-aromatics as the conjugated spacers
2
7

for dye-sensitized solar cells. J Mater Chem C 2015; 3: 9765-9780.
https://doi.org/10.1039/C5TC02356D.
[
5] (a) Gadenne V, Grenier B, Praveen C, Marsal P, Valmalette JC, Patrone L, Raimundo
JM, Combined SERS/DFT studies of push–pull chromophore self-assembled
monolayers: insights into their surface orientation. Phys Chem Chem Phys 2019; 21:
2
5865-25871. https://doi.org/10.1039/C9CP04008K. (b) Brisse R, Praveen C, Maffeis
V, Bourgeteau T, Tondelier D, Berthelot T, Geffroy B, Gustavsson T, Raimundo JM,
Jousselme B, Sustainable Energy Fuels 2018; 2: 648-654.
https://doi.org/10.1039/C7SE00474E. (c) Aboubakr H, Praveen C, Malytskyi V,
Sawadogo R, Sotiropoulos JM, Belec L, Brisset H, Raimundo JM, D/A cruciform
bithiophene chromophores as potential molecular scaffolds for optoelectronic
applications.
Tetrahedron
2016;
72:
1381-1386.
https://doi.org/10.1016/j.tet.2016.01.038.
[
[
6] Ren Y, Liu J, Zheng A, Dong X, Wang P. 2H-Dinaphthopentacene: A Polycyclic
Aromatic Hydrocarbon Core for Metal-Free Organic Sensitizers in Efficient Dye-
Sensitized
Solar
Cells.
Adv
Sci
2017;
4:
1700099.
https://doi.org/10.1002/advs.201700099.
7] Ren Y, Sun D, Cao Y, Tsao HN, Yuan Y, Zakeeruddin SM, Wang P, Grätzel M. A
Stable Blue Photosensitizer for Color Palette of Dye-Sensitized Solar Cells Reaching
1
2.6%
Efficiency.
J
Am
Chem
Soc
2018;
140:
2405-2408.
https://doi.org/10.1021/jacs.7b12348.
[
[
8] Yao Z, Zhang M, Wu H, Yang L, Li R, Wang P. Donor/Acceptor Indenoperylene Dye
for Highly Efficient Organic Dye-Sensitized Solar Cells. J. Am. Chem. Soc. 2015,
1
37, 3799-3802. https://doi.org/10.1021/jacs.5b01537.
9] (a) Murakami TN, Koumura N, Kimura M, Mori S. Structural Effect of Donor in
2
8