Effects of meta or para connected organic dyes for dye-sensitized solar cell

Article abstract of DOI:10.1016/j.dyepig.2018.05.037

Two novel phenothiazine-based organic dyes have been designed and synthesized as sensitizers for dye-sensitized solar cell (DSSC). In these dyes, we have employed carbazole as donor, phenothiazine as π-bridge, and rhodanine acetic acid as acceptor. For comparison, a benzene group has been introduced as the π-linker between carbazole donor and phenothiazine π-bridge with meta or para connection mode. Using the dye with the para-connected method, the DSSC devices exhibited higher power conversion efficiency of 5.57% than those based on dye with meta-connection mode (4.56%). The improved performance mainly came from better short-circuit photocurrent current. Photophysical, electrochemical, theoretical calculations and electrochemical impedance spectroscopy have been further investigated to disclose corresponding device mechanisms.

Full text of DOI:10.1016/j.dyepig.2018.05.037

Accepted Manuscript
Effects of meta or para connected organic dyes for dye-sensitized solar cell
Yunfei Jiao, Le Mao, Shuaishuai Liu, Tianwen Tan, Donglin Wang, Dapeng Cao,
Baoxiu Mi, Zhiqiang Gao, Wei Huang
PII:
S0143-7208(18)30292-4
10.1016/j.dyepig.2018.05.037
DYPI 6766
DOI:
Reference:
To appear in: Dyes and Pigments
Received Date: 6 February 2018
Revised Date: 17 May 2018
Accepted Date: 17 May 2018
Please cite this article as: Jiao Y, Mao L, Liu S, Tan T, Wang D, Cao D, Mi B, Gao Z, Huang W, Effects
of meta or para connected organic dyes for dye-sensitized solar cell, Dyes and Pigments (2018), doi:
10.1016/j.dyepig.2018.05.037.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. 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.

ACCEPTED MANUSCRIPT
Graphical Abstract

ACCEPTED MANUSCRIPT
1
Effects of Meta or Para Connected Organic Dyes for
2
3
Dye-sensitized Solar Cell
Yunfei Jiao a, 1, Le Mao b, 1, Shuaishuai Liu a, Tianwen Tan b, Donglin Wang a, Dapeng
4
5
Cao b, Baoxiu Mi a, *, Zhiqiang Gao b, **, Wei Huang a, b, c, ***
a
6
Key Laboratory for Organic Electronics & Informat ion Displays (KLOEID),
7
Jiangsu Key Laboratory for Biosensors (KLB), Institute of Advanced Materials
(IAM), Nanjing University of Posts & Telecommunications (NUPT), 9 Wenyuan
Road, Nanjing 210023, China
8
9
b
10
11
12
13
14
15
16
17
18
19
20
21
Jiangsu Engineering Center for Plate Displays & Solid State Lighting, Jiangsu
National Synergetic Innovation Center for Advanced Materials (SICAM), School of
Material Science and Engineering, Nanjing University of Posts
&
Telecommunications (NUPT), 9 Wenyuan Road, Nanjing 210023, China
c
Shaanxi Institute of Flexible Electronics (SIFE), Northwestern Polytechnical
University (NPU), 127 West Youyi Road, Xi’an 710072, China
1 These authors contributed equally to this work.
* E-mail: iambxmi@njupt.edu.cn
** E-mail: iamzqgao@njupt.edu.cn
*** E-mail: wei-huang@njtech.edu.cn
Declarations of interest: none
1

ACCEPTED MANUSCRIPT
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Abstract
Two novel phenothiazine-based organic dyes have been designed and synthesized as
sensitizers for dye-sensitized solar cell (DSSC). In these dyes, we have employed
carbazole as donor, phenothiazine as π-bridge, and rhodanine acetic acid as acceptor.
For comparison, a benzene group has been introduced as the π-linker between
carbazole donor and phenothiazine π-bridge with meta or para connection mode.
Using the dye with the para-connected method, the DSSC devices exhibited higher
power conversion efficiency of 5.57% than those based on dye with meta-connection
mode (4.56%). The improved performance mainly came from better short-circuit
photocurrent current. Photophysical, electrochemical, theoretical calculations and
electrochemical impedance spectroscopy have been further investigated to disclose
corresponding device mechanisms.
Keywords: Dye-sensitized solar cell; Metal-free organic dyes; D-π-A; Phenothiazine;
Rhodanine acetic acid.
2

ACCEPTED MANUSCRIPT
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
1. Introduction
Confronted with the enormous challenges of energy scarcity and environmental
safety, scientists and engineers are eager to develop new energy resources. Given the
great advantages of cleanliness, abundance and easy availability, solar energy has
been widely recognized as the most promising candidate to solve the problems. In
order to harvest solar energy, solar cells, which can firsthand generate electricity from
sun light, have long been proposed and proved as vital devices, and among which the
dye-sensitized solar cell (DSSC) is one major category.
After the first demonstration in 1991 by Michael Grätzel et al. [1], DSSC has
made tremendous progresses. There are some recent milestones including the highest
power conversion efficiency (PCE, 14.3%) of DSSC based on two co-sensitized dyes
under simulated AM 1.5G irradiation (100 mW/cm2) [2], which is suitable for the
outdoor large module photovoltaic power generation; and the PCE of 28.9% from
device with two dyes under indoor illumination of 1000 lux [3], which is promising
for small module photovoltaic devices under ambient lighting.
Typically, the conventional structure for DSSC consists of TiO2 photoanode, dye
sensitizer, redox coupled electrolyte system and counter electrode [4]. Within the
aforesaid components, the dye sensitizer, which serves as the dominant light
harvesting element, plays an essential role in determining the PCE of a DSSC device.
There are two major classes of dyes, i.e., metal complex dyes and metal-free organic
dyes. Although the former, taking Ru-complexes dye as an example, has a high
performance when applied in DSSC, they are restricted from low natural abundance,
3

ACCEPTED MANUSCRIPT
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
complicated purification process and expensiveness [5]. Alternatively, metal-free
organic dyes have recently been intensively studied, owing to their easy design, facile
synthetic method, high molar extinction coefficient (ε) and low toxicity/cost. Using
metal-free organic dyes as a sensitizer, high PCEs of 11 to 13% have been reported
[6-9], demonstrating their great potential for real applications.
To design pure organic dyes for DSSC, numerous aspects need to be considered,
such as material bandgap, electronic structure, light absorption and the dye stacking
mode on the photoanode [10]. A common strategy for such molecular engineering is
to design molecules with donor-π-acceptor (D-π-A) type structures. It has been well
documented that due to the efficient intramolecular charge transfer (ICT) which
originates from the electron pull-push effect between the D and the A moieties, D-π-A
type of molecules generally exhibit rather high ε in the ICT bands [11-14], beneficial
to the harvesting of solar energy.
In a wide range of D-π-A dyes for DSSC applications, carbazole (as D),
phenothiazine (PTZ, as D or π) moieties have been employed as building block
individually or combinationally [15, 16]. Among them, carbazole is famous for its
good hole-transport and electron donating abilities as well as its high thermal stability,
which is intensively used as building block in functional materials for OLED [17, 18],
OPV [19] and DSSC [20] applications. One of the highest PCEs from a DSSC device
with a carbazole-containing dye is 12.5% [21]. Compared to carbazole, PTZ possesses
two-folder more lone-electron pairs originating from both nitrogen and sulfur atoms,
hence PTZ is also a hole transport group due to the good electron-donating ability.
4

ACCEPTED MANUSCRIPT
88
89
Furthermore, different from the planar molecular conformation, such as carbazole, the
fused non-planar heterocyclic rings of PTZ exhibit butterfly conformation, which can
sufficiently inhibit dye aggregation, beneficial to DSSC device performance [22-24].
On the other hand, within a PTZ scaffold, since the two phenyl groups on both sides
of the central heterocyclic ring are located in a small torsion angle, successive
π-delocalization is possible, which is preferential to carrier transport [8, 25, 26]. When
PTZ was grafted in a dye for DSSC application, good PCEs of 10.4% [27] and 9.98%
[28] have been reported.
90
91
92
93
94
95
96
In terms of electron accepting and photoanode anchoring functionalities in a
D-π-A dye, rhodanine acetic acid moiety has gain much attention [25, 29], though it is
not as widely used as cyanoacrylic acid. It has been reported that rhodanine acetic
acid based organic dyes exhibited high PCE up to 9.5% [30, 31]. A rhodanine acetic
acid dye with PTZ donor gave PCE up to 4.8% [32]. Another reported pure organic
dye with rhodanine acetic acid as electron acceptor and the N-alkyl substituted PTZ as
electron donor exhibited a PCE of 2.26% when assembled in a DSSC device [33].
Although carbazole, PTZ and rhodanine acetic acid all involved dye has been
designed and applied to DSSC devices with PCE of 4.38% [29], besides the not
satisfied PCE, dyes with the combination of these three moieties are less reported.
Further research toward this aspect is needed.
97
98
99
100
101
102
103
104
105
106
107
108
109
In this paper, two metal-free organic dyes, named mCPPR and pCPPR, have
been synthesized, and the chemical structures are shown in Fig. 1. Both dyes comprise
PTZ unit as π-bridge, carbazole as donor, rhodanine acetic acid as acceptor and alkyl
5

ACCEPTED MANUSCRIPT
110
111
112
113
114
115
116
117
118
chains as dye aggregation and charge recombination suppression unit. In addition, we
have introduced a benzene group between the carbazole donor and PTZ π-bridge to
investigate the effects of molecular linking mode. The DSSC devices with pCPPR
dye exhibited high PCE of 5.57% with a short-circuit photocurrent density (JSC) of
15.12 mA/cm2, an open-circuit photovoltage (VOC) of 0.71 V and a fill factor (FF) of
0.52 under simulated AM 1.5G irradiation (100 mW/cm2). pCPPR based devices
have a higher JSC value than mCPPR devices, which comes from broader/stronger
absorption, efficiently charge transfer/electron injection, higher dye loading and lower
recombination rate, thereby enhance DSSC PCE.
119
120
Fig. 1. Chemical structures of mCPPR and pCPPR.
121
122
2. Experimental section
2.1 Materials
6

ACCEPTED MANUSCRIPT
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
All solvents were purified using standard techniques. Commercially available
reagents were used without further purification. All NMR solvents were used as
received.
2.2 Methods and instrumentations
Experimental details for 1H-and 13C-NMR, gas chromatography-mass
spectrometer (GC-MS), matrix-assisted laser desorption/ionization time-of-flight
mass spectrometry (MALDI-TOF), ultraviolet-visible spectra (UV-vis), cyclic
voltammetry analysis (CV) and steady-state photoluminescence spectroscopy (PL)
can be found in our previous paper [18]. High resolution mass spectrometer (HRMS)
data were obtained using a Thermofisher ltq-orbitrap XL. Time-resolved
photoluminescence (TRPL) were performed with an Edinburgh Instruments FLS920
spectrometer. IR data were recorded with PerkinElmer Spectrum Two Fourier
transform infrared spectrometer (FT-IR). The geometry optimization and frontier
molecular orbitals were performed using density functional theory (DFT) with the
Gaussian 09 program under Becke’s 3-parameters employing the Lee-Yang-Perdew
function (B3LYP) with the 6-31G (d) basis set.
2.3 Synthesis
The synthesis route of mCPPR and pCPPR is shown in Scheme 1.
7

ACCEPTED MANUSCRIPT
141
142
Scheme 1. Synthetic route of mCPPR and pCPPR.
8

ACCEPTED MANUSCRIPT
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
2.3.1 10-hexyl-10H-phenothiazine (1)
A mixture of phenothiazine (200 mg, 1 mmol), bromohexane (216.7 mg, 1.3
mmol), KOH (168.3 mg, 3 mmol), tetrabutylammonium bromide (TBAB, 96.6 mg,
0.3 mmol) in dimethyl sulfoxide (DMSO, 1.5 mL) was stirred at room temperature
under a nitrogen atmosphere for 10 h. When the reaction was completed, water
aqueous solution was added to quench the reaction. The crude product was extracted
with dichloromethane (DCM) and saltwater for three times and dried over anhydrous
Na2SO4. The solvent was removed with rotary evaporators, and the crude product was
purified by column chromatography using petroleum ether (PE)/DCM (15:1) as the
1
eluent to give compound 1 as colorless liquid in a yield of 88% (250 mg). H NMR
(CDCl3, 400 Hz, ppm): δ 7.21-7.13 (m, 4H), 6.97-6.85 (m, 4H), 3.86 (t, J = 7.1 Hz,
2H), 1.87-1.77 (m, 2H), 1.51-1.40 (m, 2H), 1.37-1.28 (m, 4H), 0.95-0.86 (m, 3H).
GC-MS: Anal. Calcd. for C18H21NS: 283.43, Found: 283.14.
2.3.2 10-hexyl-10H-phenothiazine-3-carbaldehyde (2)
POCl3 (0.269 mL, 2 mmol) was added into dry N,N-Dimethylformamide
dimethyl acetal (DMF, 0.538 mL, 7 mmol) in an ice water bath, and then the reaction
mixture stirred at room temperature for 1 h. Compound 1 (283 mg, 1 mmol) in DMF
(0.6 mL) was added slowly. Then the mixture was stirred and heated at 80 °C to reflux
under a nitrogen atmosphere for 10 h. When the reaction was completed, water was
added to quench the reaction. The crude product was extracted with DCM and
saltwater for three times and dried over anhydrous Na2SO4. The solvent was removed
with rotary evaporators, and the crude product was purified by column
chromatography using PE/DCM (1:2) as the eluent to give compound 2 as yellow
1
solid in a yield of 81% (253 mg). H NMR (CDCl3, 400 Hz, ppm): δ 9.79 (s, 1H),
7.64 (dd, J = 8.4, 2.0 Hz, 1H), 7.58 (d, J = 1.9 Hz, 1H), 7.19-7.08 (m, 2H), 7.00-6.93
9

ACCEPTED MANUSCRIPT
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
(m, 1H), 6.91-6.85 (m, 2H), 3.91-3.85 (m, 2H), 1.81 (m, J = 15.0, 7.5 Hz, 2H),
1.49-1.39 (m, 2H), 1.34-1.27 (m, 4H), 0.91-0.84 (m, 3H). GC-MS: Anal. Calcd. for
C19H21NOS: 311.44, Found: 311.13.
2.3.3 7-bromo-10-hexyl-10H-phenothiazine-3-carbaldehyde (3)
Compound 2 (311 mg, 1 mmol) was added into CHCl3 (1 mL), and then the
reaction mixture stirred in an ice water bath for 6 h. N-Bromosuccinimide (NBS, 177
mg, 1.2 mmol) was added in batches into mixture in the first two hours. When the
reaction was completed, water was added to quench the reaction. The crude product
was extracted with DCM and saltwater for three times and dried over anhydrous
Na2SO4. The solvent was removed with rotary evaporators, and the crude product was
purified by column chromatography using PE/DCM (1:1) as the eluent to give
1
compound 3 as yellow liquid in a yield of 58% (230 mg). H NMR (CDCl3, 400 Hz,
ppm): δ 9.79 (s, 1H), 7.65 (dd, J = 8.4, 1.9 Hz, 1H), 7.58 (t, J = 2.8 Hz, 1H), 7.27-7.21
(m, 3H), 6.90 (d, J = 8.5 Hz, 1H), 6.72 (d, J = 8.6 Hz, 1H), 3.85 (t, 2H), 1.78 (m, J =
14.8, 7.4 Hz, 2H), 1.41 (m, J = 15.1, 6.9 Hz, 2H), 1.32-1.27 (m, 4H), 0.91-0.83 (t, 3H).
MALDI-TOF (m/z): Anal. Calcd. for C19H20BrNOS: 389.04, 391.04, Found: 389.15,
391.17.
2.3.4
7-(3-(9H-carbazol-9-yl)phenyl)-10-hexyl-10H-phenothiazine-3-carbaldehyde
(4a)
A
mixture
of
compound
3
(389
mg,
1
mmol),
3-(9H-Carbazol-9-yl)phenylboronic acid (574 mg, 2 mmol), K2CO3 (345 mg, 2.5
mmol), Pd[P(C6H5)3]4 (100 mg, 0.1mmol) and TBAB (96.6 mg, 0.3 mmol) in DCM
(15 mL) was stirred and heated to reflux under a nitrogen atmosphere for 18 h. When
the reaction was completed, water was added to quench the reaction. The crude
product was extracted with DCM and saltwater for three times and dried over
10

ACCEPTED MANUSCRIPT
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
anhydrous Na2SO4. The solvent was removed with rotary evaporators, and the crude
product was purified by column chromatography using PE/DCM (2:1) as the eluent to
give compound 4a as yellow solid in a yield of 82% (455 mg). 1H NMR (CDCl3, 400
Hz, ppm): δ 9.80 (s, 1H), 8.17 (d, J = 7.7 Hz, 2H), 7.73 (d, J = 1.6 Hz, 1H), 7.68-7.61
(m, 3H), 7.59 (d, J = 1.9 Hz, 1H), 7.53 (dt, J = 7.2, 1.9 Hz, 1H), 7.48-7.38 (m, 6H),
7.34-7.28 (m, 2H), 6.92 (dd, J = 10.7, 8.5 Hz, 2H), 3.94-3.87 (m, 2H), 1.88-1.79 (m,
2H), 1.45 (dd, J = 15.1, 7.1 Hz, 2H), 1.32 (dt, J = 7.1, 3.4 Hz, 4H), 0.88 (d, J = 7.1 Hz,
3H). MALDI-TOF (m/z): Anal. Calcd. for C37H32N2OS: 552.74, Found: 552.11.
2.3.5
7-(4-(9H-carbazol-9-yl)phenyl)-10-hexyl-10H-phenothiazine-3-carbaldehyde
(4b)
A
mixture
of
compound
3
(389
mg,
1
mmol),
4-(9H-Carbazol-9-yl)phenylboronic acid (574 mg, 2 mmol), K2CO3 (345 mg, 2.5
mmol), Pd[P(C6H5)3]4 (100 mg, 0.1 mmol) and TBAB (96.6 mg, 0.3 mmol) in DCM
(15 mL) was stirred and heated to reflux under a nitrogen atmosphere for 18 h. When
the reaction was completed, water was added to quench the reaction. The crude
product was extracted with DCM and saltwater for three times and dried over
anhydrous Na2SO4. The solvent was removed with rotary evaporators, and the crude
product was purified by column chromatography using PE/DCM (2:1) as the eluent to
give compound 4b as yellow solid in a yield of 80% (443 mg). 1H NMR (DMSO, 400
Hz, ppm): δ 9.77 (s, 1H), 8.23 (d, J = 7.0 Hz, 1H), 7.97-7.88 (m, 1H), 7.77-7.55 (m,
3H), 7.42 (s, 2H), 7.34-7.13 (m, 2H), 4.01-3.96 (m, 1H), 1.77-1.65 (m, 1H), 1.40 (d, J
= 3.3 Hz, 1H), 1.25 (s, 2H), 0.89-0.75 (m, 2H). MALDI-TOF (m/z): Anal. Calcd. for
C37H32N2OS: 552.74, Found: 552.11.
2.3.6 3-(9H-Carbazol-9-y1)Phenyl-7-(N-hexyl-10H-phenothiazine-rhodanine-3-acetic
acid (mCPPR)
11

ACCEPTED MANUSCRIPT
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
A mixture of compound 4a (553 mg, 1 mmol), rhodanine-N-acetic acid (570 mg,
3 mmol) and NH4Ac (77 mg, 1 mmol) in CH3COOH (10 mL) was stirred and heated
to reflux under a nitrogen atmosphere for 24 h. When the reaction was completed,
water was added to quench the reaction. The crude product was extracted with DCM
and saltwater for three times and dried over anhydrous Na2SO4. The solvent was
removed with rotary evaporators, and the crude product was purified by column
chromatography using methanol (MeOH)/DCM (1:20) as the eluent to give mCPPR
as crimson solid in a yield of 95% (691 mg). 1H NMR (DMSO, 400 Hz, ppm): δ 8.24
(d, J = 7.7 Hz, 2H), 7.85 (s, 1H), 7.78 (d, J = 8.0 Hz, 1H), 7.75-7.67 (m, 2H), 7.57 (dd,
J = 14.8, 7.6 Hz, 3H), 7.41 (dt, J = 22.6, 10.3 Hz, 6H), 7.28 (ddd, J = 8.0, 5.3, 2.8 Hz,
2H), 7.11 (dd, J = 12.3, 8.8 Hz, 2H), 4.66 (s, 2H), 3.91 (s, 2H), 1.67 (d, J = 6.6 Hz,
2H), 1.29-1.17 (m, 6H), 0.79 (d, J = 5.1 Hz, 3H). 13C NMR (DMSO, 101 Hz, ppm): δ
193.14, 167.72, 166.84, 147.16, 143.10, 141.14, 140.60, 138.01, 134.69, 133.31,
131.26, 129.95, 127.53, 126.81, 125.82, 124.63, 123.90, 123.30, 120.97, 120.53,
119.26, 117.04, 116.56, 110.20, 47.44, 45.82, 31.26, 26.51, 26.19, 22.53, 14.28. FT-IR
(KBr) v cm-1: 3437, 3052, 2954, 2924, 2853, 1711, 1595, 1574, 1470, 1454, 1403,
1363, 1333, 1275, 1249, 1229, 1196, 1113, 1026, 960, 904, 874, 812, 792, 750, 724,
703, 657, 570. HRMS (ESI, m/z): [M+H]+ Anal. Calcd. for C42H35N3O3S3: 726.1919,
Found: 726.1914.
2.3.7 4-(9H-Carbazol-9-y1)Phenyl-7-(N-hexyl-10H-phenothiazine-rhodanine-3-acetic
acid (pCPPR)
239
240
241
242
A mixture of compound 4b (553 mg, 1 mmol), rhodanine-N-acetic acid (570 mg,
3 mmol) and NH4Ac (77 mg, 1 mmol) in CH3COOH (10 mL) was stirred and heated
to reflux under a nitrogen atmosphere for 24 h. When the reaction was completed,
water was added to quench the reaction. The crude product was extracted with DCM
12

ACCEPTED MANUSCRIPT
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
and saltwater for three times and dried over anhydrous Na2SO4. The solvent was
removed with rotary evaporators, and the crude product was purified by column
chromatography using MeOH/DCM (1:20) as the eluent to give pCPPR as crimson
solid in a yield of 94% (684 mg). 1H NMR (DMSO, 400 Hz, ppm): δ 8.25 (d, J = 7.8
Hz, 2H), 7.87 (s, 1H), 7.80 (d, J = 7.5 Hz, 1H), 7.72 (s, 2H), 7.59 (dd, J = 19.1, 11.0
Hz, 3H), 7.42 (d, J = 17.4 Hz, 5H), 7.28 (s, 2H), 7.23-7.00 (m, 2H), 6.84 (s, 1H), 4.58
13
(s, 2H), 3.93 (s, 2H), 1.69 (s, 2H), 1.24 (s, 6H), 0.81 (s, 3H). C NMR (DMSO, 101
Hz, ppm): δ 193.16, 167.61, 166.92, 147.15, 143.16, 141.18, 140.62, 138.02, 134.71,
131.25, 129.96, 127.61, 126.83, 126.30, 124.66, 123.94, 123.32, 120.97, 120.53,
117.09, 116.59, 110.21, 47.45, 31.26, 29.45, 26.44, 25.74, 22.53, 14.28. FT-IR (KBr) v
cm-1: 3745, 3436, 2594, 2924, 2853, 1711, 1596, 1574, 1488, 1472, 1453, 1405, 1365,
1334, 1277, 1251, 1229, 1197, 1114, 1058, 909, 793, 750, 724, 704, 666, 571, 514.
HRMS (ESI, m/z): [M+H]+ Anal. Calcd. for C42H35N3O3S3: 726.1919, Found:
726.1909.
2.4 DSSC fabrication and photovoltaic measurements
DSSC fabrication details, current-voltage measurement (J-V), incident
photon-to-current efficiencies (IPCEs) and electrochemical impedance spectroscopy
(EIS) can be found in our previous paper [34-36]. EIS was measured under both dark
and illumination conditions.
For the open-circuit voltage decay (OCVD) measurements, the cell was first
illuminated for 20 s to a steady voltage by using the solar simulator, then the
illumination was turned off for 200 s and the OCVD curve was recorded by Bio
analytical Systems CHI660E.
3. Result and discussions
3.1. Photophysical properties
13

ACCEPTED MANUSCRIPT
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
The UV-Vis absorption spectra of mCPPR and pCPPR in DCM solution as well
as those adsorbed on TiO2 films are shown in Fig. 2 (a) and (b). The corresponding
spectroscopic parameters are compiled in Table 1. As shown in Fig. 2 (a), in solution
both dyes exhibit broad absorption with two distinct bands. The higher energy ones
locating at 240-400 nm are assigned to the aromatic π−π* electron transitions of the
dye molecule. The lower energy bands are peaked at 486 nm for mCPPR and 493 nm
for pCPPR, with ε of 12100 and 18900 mol-1 dm3 cm-1, respectively. These low
energy bands are attributed to the ICT between the carbazole donor and rhodanine
acetic acid acceptor [29]. As can be seen, in pCPPR with carbazole donor grafting to
the para position C(4) of benzene, the ICT peak is slightly red-shifted and the
corresponding ε has significantly improved, compared to mCPPR linking at meta
position C(3), indicating that the relevant linear type of pCPPR gives an optimized
framework for better conjugation and light harvesting.
When mCPPR and pCPPR have been adsorbed on TiO2 films, the ICT bands
show broadening and red-shifting, indicating a J-aggregation of the both dyes after
adsorption on the TiO2 films [8, 9, 37]. It can be forecasted that the broadening and
red-shifting in ICT absorption will be beneficial to PCE improvement, due to the
improved light harvesting ability.
14

ACCEPTED MANUSCRIPT
286
287
288
Fig. 2. UV-Vis absorption spectra of mCPPR and pCPPR (a) dissolved in DCM and (b) adsorbed
on a 6 µm TiO2 film.
289
15

ACCEPTED MANUSCRIPT
290
291
Table 1
Photophysical and electrochemical properties of mCPPR and pCPPR.
ε
(mol-1 dm3
cm-1)
Dye
λmax (nm)
Eox a (eV)
Eg b (eV)
Ered c (eV)
486
493
12100
18900
-5.25
-5.13
2.13
2.08
-3.12
-3.05
mCPPR
pCPPR
292
293
294
295
296
a Measured in DCM with 0.1 M TBAPF6 as electrolyte and calibrated with ferrocene/ferrocenium
(Fc/Fc+) as an internal reference of 0.46 eV. Finally, calculated by -(Eox+4.8-E(Fc/Fc+)ox).
b Estimated from the onset wavelength of absorption spectra in DCM solution.
c Calculated by Eox + Eg.
3.2. Electrochemical studies
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
To evaluate the energy level matching-characteristics of mCPPR and pCPPR in
-
a traditional DSSC device with TiO2 photoanode and I-/I3 electrolyte, e.g., the ability
of electron injection from the lowest unoccupied molecular orbitals (LUMO) of a dye
to TiO2 conducting band (CB) and the regeneration of the dye from its oxide state via
the electrolyte, CV analysis was performed. The curves are shown in Fig. 3 and the
corresponding data are collected in Table 1. The first oxidation potentials (Eox), which
is correlated to the highest occupied molecular orbitals (HOMO), occur at 0.90 V and
0.78 V (vs. Ag/AgCl) for mCPPR and pCPPR, respectively. Accordingly, the HOMO
levels of mCPPR and pCPPR are calculated to be -5.25 and -5.13 eV, respectively.
-
Both of them are more negative than the I-/I3 (-4.8 eV vs. vacuum), suggesting
sufficient dye regeneration driving force from the oxidized dye along with the
electrolyte. Based on the HOMO values and the band gaps (Eg) calculated from onset
wavelength of the UV−visible absorption spectra, the LUMO levels of mCPPR and
pCPPR could be calculated as -3.12 and -3.05 eV, respectively, using Eox + Eg [9, 10,
38]. According to these calculation, the schematic energy level diagrams of the
mCPPR and pCPPR based DSSC devices can be drawn as Fig. 4. The LUMOs of the
16

ACCEPTED MANUSCRIPT
313
314
315
316
317
318
319
320
321
two dyes are more positive than the CB of TiO2 (-4.0 eV vs. vacuum), providing
efficient electron injection from the excited dye molecules to the CB of TiO2. It is
found that compared to mCPPR with meta-linked carbazole through the benzene
group, the para grafting of carbazole moiety in pCPPR resulted in higher HOMO and
LUMO levels, as well as narrower bandgap. This is understandable, since
para-linkage can enlarge effective conjugation and render electron more delocalized,
leading to narrow bandgap and raised HOMO level. Since para-linkage conjugation
can increase electron-donating ability [39] and reduce electron affinity, hence the
LUMO level is lifted.
322
323
Fig. 3. Cyclic voltammograms of mCPPR and pCPPR in DCM solution.
17

ACCEPTED MANUSCRIPT
324
325
Fig. 4. Schematic energy levels (vs. vacuum) of TiO2, electrolyte, mCPPR and pCPPR.
326
327
328
329
330
331
332
333
334
335
336
337
338
3.3. Photoluminescence studies
In order to understand the electron dynamics between the dye and the TiO2
photoanode, the steady-state and time-resolved photoluminescence spectroscopies
based on dye-adsorbed TiO2 films were investigated. Blank experiments were carried
out on mesoporous Al2O3 film, since almost no charge dissociation occurs when dyes
are adsorbed on it due to the unmatched electron levels.
As illustrated in Fig. 5 (a), the PL intensity for both dyes with charge extracting
layers (TiO2 films) are much weaker than those with Al2O3 films. The strong exciton
quenching verifies the high-efficiency of injecting electron from the excited dye into
the TiO2 photoanode. PLs of pCPPR exhibit a higher intensity difference between
TiO2 and Al2O3 films than those of mCPPR, implying that the DSSC device based on
pCPPR dye can inject electrons into TiO2 film more efficiently after exciton
generation, which may give rise to high device performance [40, 41].
18

ACCEPTED MANUSCRIPT
339
340
341
342
The TRPL decay curves and the relative fitting curves of TiO2 and Al2O3 films
with the dyes are shown in Fig. 5 (b). Decay times (τ) and corresponding amplitudes
(A) are summarized in Table 2. The measured TRPL signals were fitted by two
exponentials as follows:
ꢆ
ꢆ
ꢀꢁꢂ = ꢃꢄ exp ꢅ− ꢈ + ꢃꢉexp ꢅ−
ꢇꢄ
ꢈ
ꢇꢉ
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
where Ai denotes the proportion of each exponential that meets the relationship of A1
+ A2 = 1, and τ1/τ2 represents the time constant of each exponential component.
For dye-coated Al2O3 films, the PL decay reflects the lifetime of dye itself. The
longer PL decay time for pCPPR/Al2O3 (2.24 ns) than that for mCPPR/Al2O3 (2.05
ns) implies higher excited state stability, which may result from the optimized
molecular structure of pCPPR with para-grafting. Compared to the corresponding
Al2O3 films, the PL lifetime of the dye-coated TiO2 films decreased significantly, both
for mCPPR and pCPPR, because of the efficient electrons injection from the excited
state of molecules into the CB of TiO2. The PL decay is found to be slightly faster for
pCPPR (0.22 ns) than for mCPPR (0.26 ns), though it might be attributed to
experimental errors and can’t definitely conclude that the electron injection efficiency
from pCPPR is more efficient. However, by comparing with the control films based
on Al2O3, the injection efficiency can be calculated [42]. As can be seen, pCPPR
shows relatively higher electron injection efficiency (90.20%) than mCPPR (87.37%),
which is consistent with the steady state PL measurement (see Fig. 5 (a)).
19

ACCEPTED MANUSCRIPT
358
359
360
361
Fig. 5. (a) Steady-state photoluminescence spectra and (b) time-resolved photoluminescence
decay curves (dash lines represent for fitting) of mCPPR and pCPPR on TiO2 and Al2O3 films.
20

ACCEPTED MANUSCRIPT
362
363
Table 2
The fitting decay time (τ) and corresponding amplitude (A) of mCPPR and pCPPR from TRPL.
Dye
Film
TiO2
A1 (%)
88.94
86.13
88.43
85.25
τ1 (ns)
0.14
1.43
0.12
1.74
A2 (%)
11.06
13.87
11.57
14.75
τ2 (ns)
1.21
5.87
0.98
5.13
τav a (ns) ηinj b (%)
0.26
87.37
mCPPR
Al2O3
TiO2
2.05
0.22
90.20
pCPPR
Al2O3
2.24
364
365
a The values of τav were determined with τav = A1τ1 + A2τ2.
ꢎꢏꢐꢑ/ꢒꢓꢔ
b Calculated by ꢊꢋꢌꢍ = 1 −
.
ꢕ
ꢎꢏꢐꢑ/ꢖꢗ
ꢔ
ꢘ
ꢕ
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
3.4. Theoretical calculations
To investigate frontier molecular orbital distributions and ICT pathway, DFT
calculations were performed by the Gaussian 09 package at the B3LYP/6-31G(d)
level [43-45]. Fig. 6 shows the front/top view optimized structure and the
HOMO/LUMO distribution of the dyes. It can be inferred from the optimized
structure that the geometry of PTZ core is non-planar, showing a butterfly shape with
a ~155° angle between two phenyl groups. This framework can reduce π−π* stacking
hence suppress dye-aggregation [24]. The HOMO of both dyes locates over PTZ core
and extends to the nearby benzene group. It is worth noting that when the
peri-carbazole is connected in a para-mode in pCPPR, part of HOMO electron
extends to the carbazole moiety, contrasting to the almost zero-distribution of HOMO
electron in the carbazole part of mCPPR with meta connection. To investigate this
difference, we have tabulated the torsion angles from the Gaussian calculations, as
shown in Table 3. The optimal dihedral angles of carbazole/benzene groups (56° vs.
54°) and benzene/PTZ groups (36° vs. 36°) are similar for mCPPR and pCPPR.
However, large difference is observed in terms of the dihedral angle of
carbazole/benzene, i.e., 18° for pCPPR and 48° for mCPPR. Smaller dihedral angle
means smaller steric hindrance, which can extend the electron delocalization. As a
21

ACCEPTED MANUSCRIPT
384
385
386
387
388
389
390
391
392
result, better conjugation and more efficient ICT for pCPPR were achieved, which is
consistent with the narrower optical bandgap and more efficient ICT absorption as
disclosed previously. The LUMO of mCPPR and pCPPR are mainly distributed on
rhodanine acetic acid acceptor and partly on the neighboring PTZ π-linker. The
electronic distribution for both dyes shows a sufficient overlap of frontier molecular
orbitals on PTZ unit, guaranteeing a facile charge transfer from donor to acceptor
upon irradiation and hence ensuring efficient ICT process. Additionally, considering
hardly any HOMO electron distributes on the alkyl chain-substituted, it can efficiently
prevent electron recombination [46].
22

ACCEPTED MANUSCRIPT
Dye
Optimized structure
HOMO
LUMO
mCPPR
155°
pCPPR
154°
393
Fig. 6. Optimized ground state geometry and the frontier molecular orbital distributions of mCPPR and pCPPR.
23

ACCEPTED MANUSCRIPT
394
395
Table 3
Torsion angle of carbazole/benzene/PTZ of mCPPR and pCPPR.
Groups
Carbazole/Benzene a
Benzene/PTZ a
Carbazole/PTZ b
Schematic
56°
54°
36°
36°
48°
18°
mCPPR
pCPPR
396
397
398
a The dihedral angle is indicated by the four stars.
b Dihedral angle between carbazole unit and closer-phenyl in PTZ core.
3.5. Photovoltaic performances of DSSC
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
The J-V curves of DSSC sensitized by mCPPR and pCPPR are plotted in Fig. 7
(a) and the parameters are listed in Table 4. The DSSC based on mCPPR exhibits a
PCE of 4.56% with a JSC of 12.53 mA/cm2, a VOC of 0.69 V and a FF of 0.53.
Compared to mCPPR, device with pCPPR shows significantly improved JSC (15.12
mA/cm2), generating an overall PCE of 5.57%. The relatively higher JSC from
pCPPR device is mainly ascribed to the follow reasons: 1) broader and stronger
absorption (see Fig. 2 and Table 1), 2) efficient charge injection (see Fig. 4), 3) higher
dye loading amount (see Table 4). The lower part of Fig. 7 (a) presents the dark
currents of the devices, which reflect the dark reactions related to the charge
recombination between the electrons in the TiO2 photoanode and the hole on the redox
couple [47, 48]. Under the same bias voltage, the dark current for pCPPR DSSC is
smaller than mCPPR, indicating weaker recombination, thus leading to the slightly
increased VOC in pCPPR device.
Fig. 7 (b) shows the IPCEs for the DSSC devices. As shown, both dyes show
good ability to convert light to photocurrent in the region of 400-750 nm. The IPCE
values from pCPPR device reaches the highest value of 56.9% at 495 nm, while 50.2%
24

ACCEPTED MANUSCRIPT
415
416
417
at 495 nm for mCPPR device. The relatively higher IPCE for pCPPR leads to larger
photocurrents, which is well consistent with the higher JSC value resulting from the
J-V curves.
418
25

ACCEPTED MANUSCRIPT
419
420
421
422
423
Fig. 7. (a) J-V curves (above part) and dark current curves (lower part) and (b) IPCE spectra for
DSSC devices based on mCPPR and pCPPR.
Table 4
Photovoltaic parameters of DSSC sensitized by mCPPR and pCPPR.
Dye loading
Dye
JSC (mA/cm2)
VOC (V)
FF
PCE (%)
(mol/cm2)
2.92×10-7
4.04×10-7
12.53
15.12
0.69
0.71
0.53
0.52
4.56
5.57
mCPPR
pCPPR
424
425
426
427
428
To better understand charge recombination kinetics, an OCVD measurement,
which monitors the procedure of VOC transient from illumination condition to the dark
equilibrium, was implemented. Fig. 8 (a) presents OCVD curves of mCPPR and
pCPPR devices. The lifetimes of the photogenerated electrons in the device, τn, can
be evaluated according to equation [10]:
ꢣꢄ
429
430
ꢇꢌ = − ꢙ ꢛ ꢝꢞꢟꢔꢠꢢ
ꢚ
ꢜ
ꢞꢡ
where KB is the Boltzmann constant and T is the temperature.
26

ACCEPTED MANUSCRIPT
431
432
433
434
435
The calculated τn values are plotted in Fig. 8 (b). At the potential values of
illuminated quasi equilibrium state, the τn of DSSC devices based on mCPPR and
pCPPR are ~40 ms and ~100 ms, respectively. It can be seen that under the same
potential the DSSC based on pCPPR has longer electron lifetime, which is ascribed
to the slower electron recombination rate and accounts for the relative higher PCE.
436
27

ACCEPTED MANUSCRIPT
437
438
Fig. 8. (a) OCVD for DSSC devices based on mCPPR and pCPPR and (b) electron lifetime
derived from OCVD measurement.
439
440
3.6. Electrochemical impedance spectroscopy analysis
441
442
443
444
445
446
447
448
449
450
451
The EIS measurement was performed under a bias of VOC from 10-1 Hz to 105 Hz
both under the dark and the illumination conditions to get elucidation on the interface
charge-transfer and recombination processes in our DSSC devices. The results are
shown in Fig. 9 and the corresponding data are summarized in Table 5. The EIS
values are fitted through ZSimpWin software and the equivalent circuit is shown in
the inset of Fig. 9 (a). In Nyquist plot (Fig. 9 (a)), the X-intercept value is
corresponding to the lumped series resistance (Rs) for external circuit outside of the
cell. The semicircle at high-frequency region is attributed to charge transfer resistance
at the electrolyte/Pt electrode interface (RPt). The large semicircle in low-frequency is
attributed to the electron diffusion in the TiO2 film and electron back reaction with
oxidized redox species at the TiO2/electrolyte interface (Rct) [49], which has most
28