Dihydrophenazine-based double-anchoring dye for dye-sensitized solar cells

Article abstract of DOI:10.1002/jccs.201900410

A novel dihydrophenazine-based organic di-anchoring dye DK-11 was synthesized by utilizing a simple synthetic protocol. The dye was characterized by optical and electrochemical studies and used as a sensitizer for dye-sensitized solar cell. The proposed butterfly structure was supported by IR experiments which ensured the binding of both carboxylic acid units on the semiconductor surface. Using the dye DK-11, the device generated an efficiency of 5.07% with J_SC, V_OC, and FF values of 10.65 mA/cm², 0.67 V, and 0.71, respectively.

Full text of DOI:10.1002/jccs.201900410

Received: 24 September 2019
DOI: 10.1002/jccs.201900410
Revised: 19 November 2019
Accepted: 20 November 2019
S P E C I A L I S S U E P A P E R
Dihydrophenazine-based double-anchoring dye for
dye-sensitized solar cells
Dhirendra Kumar1 | Yu-Lin Chen2 | Chih-Hung Tsai2 | Ken-Tsung Wong1,3
1Department of Chemistry, National
Abstract
Taiwan University, Taipei, Taiwan
2Department of Opto-Electronic
Engineering, National Dong Hwa
University, Hualien, Taiwan
A novel dihydrophenazine-based organic di-anchoring dye DK-11 was synthe-
sized by utilizing a simple synthetic protocol. The dye was characterized by
optical and electrochemical studies and used as a sensitizer for dye-sensitized
solar cell. The proposed butterfly structure was supported by IR experiments
which ensured the binding of both carboxylic acid units on the semiconductor
surface. Using the dye DK-11, the device generated an efficiency of 5.07% with
JSC, VOC, and FF values of 10.65 mA/cm2, 0.67 V, and 0.71, respectively.
3Institute of Atomic and Molecular
Science, Academia Sinica, Taipei, Taiwan
Correspondence
Chih-Hung Tsai, Department of Opto-
Electronic Engineering, National Dong
Hwa University, Hualien 97401, Taiwan.
Email: cht@mail.ndhu.edu.tw
K E Y W O R D S
di-anchoring dyes, donor-acceptor dyes, dye-sensitized solar cells, dye-TiO2 binding, molecular
geometry
Ken-Tsung Wong, Department of
Chemistry, National Taiwan University,
Taipei 10617, Taiwan.
Email: kenwong@ntu.edu.tw
Funding information
Ministry of Science and Technology
Taiwan, Grant/Award Number: MOST
107-2221-E-259-028
1 | INTRODUCTION
because of their higher molar extinction coefficients for
visible absorption; accessibility of functional tuning by
The demand for renewable, green, and clean energy
sources continues because of increasing energy require-
ments and concerns about global warming.[1] After Grä-
tzel invented dye-sensitized solar cells (DSSCs), DSSCs
have become one of the most promising low-cost alterna-
tive energy sources.[2] Using a dye as a sensitizer plays an
important role in harvesting light energy in such devices.
Several organometallic dyes, particularly ruthenium com-
plexes, served as efficient sensitizers and achieved excel-
lent power conversion efficiencies.[3] However, such
complexes are practically less commercialized because of
the expenses associated with their use and refinement.
As an alternative to the expensive ruthenium complexes,
interests in cheap and relatively environmentally friendly
metal-free organic dyes have grown exponentially
varying the donor (D), acceptor (A), and π-conjugated
linker (π) via simple chemical modifications; and low
cost.[4] Among various organic dyes developed for DSSSs,
di-anchoring organic dyes have emerged as the new
design to be employed as sensitizers for solar cells in the
last few years.[5] These organic dyes result in better
charge collection on the semiconductor surface because
these dyes have more binding sites and hence better per-
formance than that of mono-anchoring dyes. However,
some di-anchoring dyes have inferior performance
because of lower dye loading, which is due to their bulk-
ier size and/or dye aggregation.[6] For instance, (D-π-A)2-
configured dyes typically have better performance than
that of mono-anchoring analogues because the individual
D-A unit contributes not only equal electron transfer but
© 2019 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
J Chin Chem Soc. 2019;1–9.
http://www.jccs.wiley-vch.de
1

2
KUMAR ET AL.
also net dye-loading because of effective binding on the
semiconductor surface.[7–9] However, the longer π- or
σ-linker connecting the two D-A units will reduce the
dye loading and thus result in inferior performance.[10]
Previously, we have already investigated and demon-
strated that directly connected (D-π-A)2 performed better
than that having longer connecting spacer between the
individual D-A units.[11] A single donor-containing dye,
D(π-A)n, suffers from competitive electron delocalization
between the donor and acceptor and hence has lower
performance. In addition, Meng and coworkers synthe-
sized single triphenylamine-cored dyes (S1-S3) with 1–3
cyanoacrylic acid units, whose efficiency decreased as the
number of acceptor units increased.[12] Several fused
electron-rich systems have been designed to overcome
this problem; however, some of them had inferior perfor-
mance for two possible reasons: (i) both anchoring units
were not readily available to bind on TiO2 because of
their opposite locations in such linear geometry; or (ii) if
they can bind simultaneously, the dye acquired a larger
amount of space and therefore there was a lower amount
of dye loading on semiconductor surface.[13] As a result,
there is still demand for a di-anchoring design that
ensures efficient charge transfer between the D-A units
as well as effective binding of both the anchoring units
on the TiO2 surface.
anchoring dyes that do not contain a thiophene
chromophore.
2 | RESULTS AND DISCUSSION
2.1 | Synthesis
Scheme 1 depicts the synthesis of the new di-anchoring
dye (DK-11). The synthesis was initiated by the reaction
of catechol (1) and o-phenylenediamine to yield the DHP
(2). However, the resulting product is highly sensitive to
air that was converted to the oxidized product phenazine
during the work-up. This conversion was confirmed
when the product was treated directly with alkyl bromide
in the presence of a base, in which the alkylated product
was not formed exclusively and the starting material
remained intact. Therefore, the initially formed com-
pound was reduced by Na2S2O4 and followed by the
immediate alkylation of using 2-ethylhexyl bromide to
afford 3. The N-alkyl dihydrophenazine 3 was formylated
using the Vilsmeier-Haack reaction[21] to get dialdehyde
4. Finally, the dialdehyde 4 was converted to dye DK-11
by the Knoevenagel reaction.[22]
Dihydrophenazine (DHP) and its derivatives have
been exploited extensively for biological and chemical
studies over the last decade.[14] However, there are few
reports about the use of DHP in electronic applications.
Because of DHP's high electron-richness and redox-
active properties,[15] it has been utilized as a hole-
transporting material,[16] a hole-injection material[17]
for OLEDs, as well as a temperature-dependent fluores-
cence[18,19] and photo-redox catalyst.[20] These unique
electronic properties inspired us to construct a DHP-
based sensitizer for the DSSC. To the best of our
knowledge, we report herein the first organic sensitizer
based on DHP as electron-donating core. To achieve di-
anchor design, cyanoacrylic acids was incorporated on
the second and seventh positions of the DHP core,
respectively. In addition, 2-ethylhexyl group that can
ensure good solubility and also increase the dye's
hydrophobic nature to control dye-electrolyte interac-
tions was introduced on the nitrogen centers of DHP.
We postulate that the new DHP-based dye would adopt
a butterfly structure as both the carboxylic acid units
anchoring on the TiO2 surface, which would be useful
for inhibiting dye-aggregation without interfering with
the donor-acceptor conjugation. Further, the structure
is intriguing as it is a “thiophene-free” dye that gener-
ates a good power conversion efficiency (PCE) of
5.07%, which makes it unique among other (D-A)2 di-
2.2 | Photophysical properties
The absorption spectrum of DK-11 was recorded in tetra-
hydrofuran (THF) solution (Figure 1a). DK-11 exhibits
an absorption maximum centered at 508 nm, which can
be attributed to the charge transfer transition, whereas
the higher-energy bands correspond to π ! π* transi-
tions. The data are summarized in Table 1. The absorp-
tion exhibits pronounced changes after the addition of a
base such as triethylamine (TEA). There is a 50 nm blue-
shift in the absorption maximum accompanying with a
color change in the dye solution that was visible to the
naked eye as shown in the inset of Figure 1a. This color
change occurred because of the deprotonation of the car-
boxylic acid units of the dye, which inhibit the strong
intramolecular donor-acceptor interactions.[23] The
results also revealed the major involvement of charge-
transfer transitions in the longer wavelength absorption
peak, which is a key factor to determine the light-
harvesting ability of the dye. However, there was not
much change after the addition of an acid such as
trifluoroacetic acid (TFA). The protonation of the carbox-
ylic acid unit and nitrogen center of the DHP core might
occur simultaneously, which would cancel out the rela-
tive effect.
The absorption spectrum for the dye DK-11 upon
adsorbing on the TiO2 film was also recorded (Figure 1b).

KUMAR ET AL.
3
SCHEME 1 Synthesis of dye
DK-11
FIGURE 1 (a) Absorption spectra of dye DK-11 recorded in THF. (b) Absorption spectrum of dye DK-11-TiO2 film. THF,
tetrahydrofuran
TABLE 1 Photophysical properties of dye DK-11
Dye
λabs, nm (ε, ×104 M−1 cm−1 a
508 (2.85)
)
λabs on TiO2, nmb
E0-0, eVc
Eox, Vd
Ered, Vd
Eox*, Ve
−1.30
DK-11
500
2.02
0.72
−1.88, −2.09
aAbsorption spectra recorded in THF (10−5 M).
bAbsorption spectra of the dye anchored on 7 μm porous TiO2 nanoparticle films.
cE0-0 was determined from the onset of the absorption spectra in THF.
dOxidation potential was measured by CV in THF; reduction potential was measured by DPV in THF; potentials were measured versus FC/FC+ were converted
to normal hydrogen electrode (NHE) by addition of 630 mV.
eExcited state oxidation potential was calculated by Eox − E0-0
.
An unusual sharp absorption peak with a slight hyp-
sochromic shift (ca. 8 nm) was observed when the dye
was anchored on the TiO2 film. Usually, dyes give broad
absorption peaks when they are anchored on TiO2
because of aggregation. We speculate that the dye adopts
a butterfly-like structure on the TiO2 surface, leading to
the suppression of the typical dye aggregation that will be
beneficial for the reduction of charge recombination
process.
The FT-IR spectrum of DK-11 dye-anchored TiO2
films was scrutinized (Figure 2). The experimental results
indicate that the CN band (2,218 cm−1 ! 2,223 cm−1
)
remains almost intact while the -COOH band
(1,720 cm−1) of the neat dye vanishes. The participation
of both the carboxylic acid units upon anchoring to TiO2
was further confirmed by the presence of the characteris-
tic stretching absorption band of COO− observed at 1,590
and 1,384 cm−1 for asymmetric and symmetric stretching

4
KUMAR ET AL.
vibrations, respectively.[24,25] The IR results support our
hypothesis that the dye adopts a butterfly-like structure
as only this structure can render the binding of both the
carboxylic acid units on the TiO2 surface. The di-anchor
binding thereby can promote higher electron injection
and hence higher cell performance.
ground-state and excited-state oxidation potentials were
calculated (Table 1) and found to be thermodynamically
favorable for the dye-regeneration and charge-injection
processes, respectively.
2.4 | Photovoltaic properties
2.3 | Electrochemical properties
The photovoltaic properties of DK-11 as the sensitizer for
the DSSC were estimated with a sandwich DSSC cell
using 0.6 M 1-butyl-3-methylimidazolium iodide (BMII),
0.05 M LiI, 0.03 M I2, 0.5 M 4-tert-butylpyridine, and
The electrochemical characteristics were investigated by
cyclic voltammetry (CV) and differential pulse
voltammetry (DPV) in THF (Figure 3). The dye showed a
quasi-reversible oxidation peak with an oxidation poten-
tial of 0.72 V (vs. NHE), which revealed its electron rich-
ness. The CV also showed two reduction potentials as
determined by DPV experiments. The calculated ground-
state and excited-state oxidation potentials were found to
be suitable for the DSSC's electron processes. The
0.1
M guanidinium thiocyanate in a mixture of
acetonitrile–valeronitrile (5:1, vol/vol) as the redox elec-
trolyte (details of the device preparation and characteri-
zation are described in the Section 3).
Figure 4a shows the incident monochromatic photon-
to-current conversion efficiency (IPCE) spectrum of the
DSSC based on DK-11 as a sensitizer. The IPCE of the
device using the DK-11 showed a maximum IPCE of 78%
at 475 nm. Figure 4b shows the current density-voltage
(J-V) characteristics of the DSSC under standard global AM
1.5G solar irradiation. The DK-11-based device exhibited a
decent power conversion efficiency (PCE) of 5.07% with the
photovoltaic parameters as: JSC = 10.65 mA cm−2,
VOC = 0.67 V, and FF = 0.71. Although the FF is found to
be higher and VOC is almost the same when compared to
that of structurally similar phenothiazine dyes (PR6C1: (E)-
2-cyano-3-(10-[2-ethylhexyl]-10H-phenothiazin-3-yl)acrylic
acid; PR6C2: (2E,20E)-3,30-(10-[2-ethylhexyl]-10H-phenothi-
azine-3,7-diyl)bis(2-cyanoacrylic acid)), the PCE still
remains lower because of inferior JSC value.[26]
The electrochemical impedance spectrum (EIS) char-
acteristics of the DK-11-based solar cells were also exam-
ined to reveal the charge transport features of the several
interfaces in the device. Figure 5a shows the EIS Nyquist
plot of the device. The smaller semicircle in the lowest
frequency indicates the ion diffusion impedance in the
electrolyte, and the larger semicircle in the intermediate
FIGURE 2 FT-IR spectrum of pristine dye (top) and dye DK-
11 adsorbed on the TiO2 film (bottom)
FIGURE 3 Cyclic voltammogram
and differential pulse voltammogram of
DK-11

KUMAR ET AL.
5
FIGURE 4 (a) IPCE spectrum of DSSC based on DK-11, (b) current density-voltage curve for DK-11-based DSSC under AM 1.5G
simulated solar light (100 mW/cm2). DSSC, dye-sensitized solar cell; IPCE, incident monochromatic photon-to-current conversion efficiency
FIGURE 5 (a) EIS Nyquist plot for DK-11-based DSSC, (b) EIS Bode plot for DK-11-based DSSC. DSSC, dye-sensitized solar cell; EIS,
electrochemical impedance spectrum
frequency corresponds to the charge transfer imped-
ance in the TiO2/dye/electrolyte interface. The charge-
transfer resistance value in the TiO2/dye/electrolyte
interface was calculated and found to be 233.2 Ω.
Figure 5b shows the EIS Bode plots (i.e., the phase of
the impedance vs. the frequency) of the DK-11-based
DSSC. The representative frequency of the middle peak
in the Bode plot is related to the charge recombination
rate, and its reciprocal is associated with the electron
lifetime, which was 2.83 ms for the DK-11-based DSSC.
The EIS results suggest that the di-anchoring feature of
DK-11 on TiO2 can suppress the dye aggregation and
thus reduce the charge recombination/dark current at
the TiO2/dye/electrolyte interface and thus exhibit a
proper electron lifetime.
The device using DK-11 exhibited decent value of
charge-transfer resistance (under light) and electron life-
time. The former is responsible for the efficient charge
injection on TiO2 while later evident the ability of the
dye to retard the undesirable charge recombination pro-
cess. Usually, the planar dyes have tendency of aggrega-
tion which facilitate the charge recombination process
faster and therefore exhibit much lower lifetime and
hence very low open-circuit voltage and efficiency. The
FT-IR spectrum of DK-11 clearly indicates that both car-
boxylic acid groups are anchored on semiconductor sur-
face, which is only possible if the dye adopt butterfly
structure on TiO2 surface. Such nonplanar structure
inhibits the dye-aggregation and exhibits decent lifetime
and open-circuit voltage of 0.67 V.

6
KUMAR ET AL.
3 | EXPERIMENTAL
1.0 mM DK-11 organic dye in THF for 24 hr at room
temperature. A spectrophotometer was then employed
for the analysis of the UV–vis absorption spectra of the
TiO2 films loaded with dye.
3.1 | General methods and
instrumentation
Reagent grade chemicals and solvents were purchased
commercially and used without further purification
unless otherwise stated. Solvents for syntheses and ana-
lyses were freshly distilled prior to use. All reactions were
carried out under argon and were magnetically stirred
unless otherwise specified. The reactions were monitored
by TLC using Macherey-Nagel pre-coated aluminum foil
sheets (0.20 mm with fluorescent indicator UV254) and
visualized with UV light at 254 and 365 nm. Column
chromatography was carried out using flash silica gel
from Merck (230–400 mesh). Infrared (IR) spectra were
recorded within KBr on a Varian-610 FT-IR spectrome-
ter. 1H NMR and 13C NMR were recorded in benzene-d6,
CDCl3, and DMSO-d6 using a Varian (Unity Plus 400)
spectrometer at 400 MHz and 100 MHz, respectively.
Low- and high-resolution mass spectra were recorded
using a Thermo Orbitrap QE Plus mass spectrometer in
ESI mode. Melting points were determined on a Fargo
MP-1D melting point apparatus.
Absorption spectra (in THF) were recorded on a Jasco
V-670 spectrophotometer; five different concentrations
ranging from 5 × 10−5 M to 5 × 10−7 M were performed in
order to obtain the extinction coefficient. Electrochemical
properties were performed with a CH Instruments, Inc.
CH1619B in a three-electrode electrochemical cell with a
glassy carbon working electrode, a platinum wire counter
electrode, and an Ag/AgCl reference electrode. The cyclic
voltammetry and differential pulse voltammetry measure-
ments were carried out in anhydrous THF containing 0.1 M
tetrabutylammonium hexafluorophosphate (for oxidation)
or tetrabutylammonium perchlorate (for reduction) as a
supporting electrolyte, purging with argon prior to conduct
the experiment. All potentials reported were calibrated with
the ferrocene/ferrocenium redox couple (Fc/Fc+) as internal
standard, and all potentials mentioned herein were
converted to normal hydrogen electrode (NHE) by addition
of 630 mV.
3.1.2 | Fabrication of dye-sensitized solar
cells
Fabrication of the working electrodes (WEs) for the
DSSCs used the methodology described below. Initially
cleaning was undertaken of a fluorine-doped tin oxide
(FTO) substrate, which then had a 3 M tape with 4 mm
diameter holes in it attached. These holes were punched
beforehand for definition of the device coating area. The
doctor blade technique was used to give the area a uni-
form coating of 20 nm TiO2 nanoparticle paste. These
coated samples were heated for 10 min at 150ꢀC; subse-
quently, another coating was added to result in 12 μm
thick TiO2 WEs. An identical methodology was then used
for coating a 200 nm TiO2 nanoparticle layer for employ-
ment as a scattering layer. These coated samples then
underwent sintering for 30 min at 500ꢀC. Once they had
cooled to 80ꢀC, the TiO2 electrodes underwent 24-hr
immersion in a DK-11 dye solution. This was a THF
solution containing 0.5 mM chenodeoxychloic acid
(CDCA) and 1.0 mM DK-11 dye. The following method
was used to fabricate the DSSC counter electrode (CE).
The substrate comprised a piece of FTO conductive glass
that had two holes already drilled in it for the injection of
the electrolyte solution. This substrate then underwent
washing with deionized water and alcohol using an ultra-
sonic cleaner for 10 min. Nitrogen gas was then used to
dry the substrate surface. The FTO substrate was then
uniformly coated with a Pt nanoparticle paste employing
the doctor blade coating technique. This substrate then
underwent sintering in a furnace for 30 min at 450ꢀC.
Fabrication of the Pt counter electrode took place when
the substrate had naturally cooled to room temperature.
Sixty micrometers of encapsulation films were then cut
with an outer dimension of 2.5 × 2.5 cm and an inner
dimension of 0.8 × 0.8 cm for the assembly of the Ws and
CEs. Sealing of the upper and lower electrodes was
achieved by the application of 3 min of constant pressure
at 3 kg/cm2 at 130ꢀC. Once the electrodes were cool, each
device was injected with 5 μl of an electrolyte solution,
employing a micropipette. The electrolyte solution com-
prised 0.6 M 1-butyl-3-methylimidazolium iodide (BMII),
0.05 M LiI, 0.03 M I2, 0.5 M 4-tert-butyloyridine, 0.1 M
guanidinium thiocyanate (GuSCN), and a 5:1 mixture
(by volume) of acetonitrile and valeronitrile. The
0.8 × 0.8 cm encapsulation films that remained were
employed with glass plates for sealing the CE holes
3.1.1 | Measurement of absorption
spectra of dye-loaded nanoporous TiO2
films
A glass substrate was coated with a 7-μm-thick transpar-
ent porous TiO2 nanoparticle layer (using 20-nm anatase
TiO2 nanoparticles), employing the doctor-blade method.
Once this had been sintered at 500ꢀC for 30 min, the
TiO2 film underwent immersion in a dye solution of

KUMAR ET AL.
7
through application of 3 kg/cm2 of pressure for 1 min at
130ꢀC to stop the electrolyte leakage or evaporation. To
complete the fabrication of the DSSC device, the FTO
glass plate surface undergoes cleaning using alcohol. On
a testing device, an aperture area of 0.125 cm2 was
masked off. The device properties then underwent analy-
sis using current density-voltage (J-V), incident photo-to-
current conversion efficiency (IPCE), and electrochemi-
cal impedance spectroscopy (EIS) analysis.
application of a bias at the cell's Voc (i.e., with no DC
electric current applied), employing an AC amplitude
of 10 mV.
3.2 | Synthesis and characterization
3.2.1 | Synthesis of 5,10-bis(2-ethylhexyl)-
5,10-dihydrophenazine (3)
o-phenylenediamine (1.08 g, 10 mmol) and catechol
(1.65 g, 15 mmol) were ground finely and transferred to a
sealed tube. The tube was closed and heated at 180ꢀC for
4 days. The product was cooled and digested with hot
water. The solid was further washed thoroughly with hot
water and dried under vacuum to collect green-colored
5,10-dihydrophenazine (2). The compound was trans-
ferred to a flask and 50 ml of ethanol was added. The
mixture was refluxed under inert atmosphere and then
Na2S2O4 (6.96 g, 40 mmol) dissolved in degassed water
was added. The reaction was continued overnight and
then quenched by addition of water. The precipitated
solid was filtered, washed with water, and wet solid was
transferred to flask immediately and evacuated and kept
under argon. Dimethyl sulfoxide (30 ml), distilled water
(2 ml), KOH (1.68 g, 30 mmol), and tetabutylammonium
bromide (0.13 g, 0.5 mmol) were added successively and
the mixture was stirred under inert atmosphere. After
½ hr, 2-ethylhexyl bromide (3.6 ml, 20 mmol) was added
slowly and the reaction was refluxed for overnight. The
reaction mixture was cooled and poured into water and
extracted with n-hexane. The solvent was removed and
residue was purified by column chromatography on neu-
tral alumina using n-hexane as eluent to afford light-
3.1.3 | Photocurrent density-voltage
measurement
The measurement of the DSSCs' current density-voltage
(J-V) characteristics took place using a 550 W Xenon
solar simulator lamp providing a simulated AM1.5G solar
light. A reference cell was used to calibrate the incident
light intensity to 100 mW/cm2. Certification for the refer-
ence cell came from Bunkoh-Keiki Co. Ltd., Japan. An
external bias voltage was applied to the cell and the gen-
erated photocurrent was measured to obtain the
photocurrent-voltage curves. The power conversion effi-
ciency, fill factor (FF), open-circuit voltage (VOC), and
short-circuit current density (JSC) were extracted from
the J-V characteristics of the DSSCs.
3.1.4 | Incident monochromatic photon-
to-current conversion efficiency
measurement
A 150 W Xenon arc lamp, coupled with a monochroma-
tor, was employed as a light source to measure the inci-
dent monochromatic photon-to-current conversion
efficiency (IPCE) spectra. The monochromatic light of
the solar cells was illuminated at wavelength sampling
intervals of 5 nm between 400 and 650 nm, and the
short-circuit current for the solar cells was then mea-
sured to supply the IPCE data. This measurement was
fully controlled by a computer.
1
green oil. Yield 1.26 g (31%, 3 steps); H NMR (400 MHz,
C6D6) δ 6.75–6.77 (m, 4H), 6.45–6.49 (m, 4H), 3.25–3.28
(m, 4H), 2.02–2.06 (m, 2H), 1.23–1.44 (m, 16H), 0.83–0.90
(m, 12H); 13C NMR (100 MHz, C6D6) δ 140.3, 121.7,
113.2, 48.3, 36.1, 31.5, 29.4, 24.7, 23.9, 14.7, 11.1; HRMS
(ESI) m/z calcd. For C28H42N2 [M]+ 406.3343, found
406.3342.
3.1.5 | Electrochemical impedance
spectroscopy measurement
3.2.2 | Synthesis of 5,10-bis(2-ethylhexyl)-
5,10-dihydrophenazine-
2,7-dicarbaldehyde (4)
An impedance analyzer with a frequency range between
0.1 Hz and 100 kHz was employed to measure cell
impedance for electrochemical impedance spectroscopy
(EIS). When the impedance was being measured,
the cell was kept under constant illumination of
AM 1.5G 100 mW/cm2. Measurement of the cell imped-
ance (0.1 Hz–100 kHz) was undertaken through the
To a solution of 3 (0.41 g, 1 mmol) in 5 ml of dry DMF,
POCl3 (0.21 ml, 2.2 mmol) was added dropwise at 0ꢀC
under inert atmosphere. The mixture was stirred at the
same temperature for 2 hr and then promoted at 90ꢀC for
24 hr. The reaction mixture was cooled, poured into ice
water, and neutralized by aqueous sodium hydroxide. It

8
KUMAR ET AL.
was extracted with dichloromethane; the organic layer
was collected and evaporated. The crude product was
purified by silica gel column chromatography using dic-
hloromethane/hexane as eluent. After purification, the
sticky solid was precipitated after treatment of n-pentane
to collect orange solid. Yield 0.28 g (60%); mp 87ꢀC; IR
aggregation and thus reduces the propensity of charge
recombination. The solar cell using DK-11 had a power
conversion efficiency of 5.07%, which is decent among
non-thiophene dyes. The EIS results also suggested that
the suppressed dye aggregation reduced the charge
recombination/dark current at the TiO2/dye/electrolyte
interface and exhibited a adequate electron lifetime.
1
(KBr, cm−1) 1656 (νC=O); H NMR (400 MHz, DMSO-d6)
δ 9.62 (s, 2H), 7.27 (d, J = 8.0 Hz, 2H), 6.92 (s, 2 H), 6.71
(d, J = 8.4 Hz, 2H), 3.56 (d, J = 7.2 Hz, 4H), 1.80–1.84 (m,
2H), 1.20–1.43 (m, 16H), 0.79–0.88 (m, 12H); 13C NMR
(100 MHz, CDCl3) δ 189.5, 144.2, 136.8, 129.9, 128.8,
111.4, 110.1, 47.8, 35.0, 30.2, 28.3, 23.6, 22.7, 13.7, 10.4;
HRMS (ESI) m/z calcd. For C30H42N2O2 [M + H]+
463.3319, found 463.3319.
ACKNOWLEDGMENTS
The financial support from the Ministry of Science and
Technology Taiwan (MOST 107-2221-E-259-028) was
acknowledged.
REFERENCES
[1] N. Armaroli, V. Balzani, Angew. Chem. Int. Ed. 2007, 46, 52.
[2] B. O'Regan, M. Grätzel, Nature 1991, 353, 737.
[3] H. Ozawa, T. Sugiura, T. Kuroda, K. Nozawa, H. Arakawa,
J. Mater. Chem. A 2016, 4, 1762.
3.2.3 | Synthesis of (2E,20E)-3,30-(5,10-bis
(2-ethylhexyl)-5,10-dihydrophenazine-
2,7-diyl)bis(2-cyanoacrylic acid) (DK-11)
[4] A. Mishra, M. K. R. Fischer, P. Bäuerle, Angew. Chem. Int. Ed.
2009, 48, 2474.
[5] D. Kumar, K.-T. Wong, Mater. Today Energy 2017, 5, 243.
[6] K. D. Seo, B. S. You, I. T. Choi, M. J. Ju, M. You, H. S. Kang,
H. K. Kim, ChemSusChem 2013, 6, 2069.
To a solution of 4- (0.23 g, 0.5 mmol) and 2-cyanoacetic
acid (0.13 g, 1.5 mmol) in 10 ml of ethanol, two drops of
piperidine was added and the mixture was refluxed over-
night. The reaction mixture was cooled and poured into
water to precipitate the solid. The solid was collected by
filtration and washed with water. The crude product was
dissolved in THF and passed through silica gel column to
afford pure product. Black solid; Yield 0.21 g (72%); mp
240ꢀC; IR (KBr, cm−1) 2218 (νC=N), 1,720 (νC=O); 1H
NMR (400 MHz, DMSO-d6) δ 7.98 (s, 2H), 7.37 (d,
J = 8.8 Hz, 2H), 7.31 (s, 2 H), 6.69 (d, J = 8.4 Hz, 2H),
3.54 (d, J = 6.8 Hz, 4H), 1.84–1.89 (m, 2H), 1.15–1.43 (m,
16H), 0.79–0.87 (m, 12H); 13C NMR (100 MHz, CDCl3) δ
164.2, 152.6, 143.0, 135.5, 129.9, 124.4, 117.6, 112.7, 112.6,
97.7, 47.3, 34.8, 29.3, 27.7, 22.8, 22.5, 13.8, 10.4; HRMS
(ESI) m/z calcd. For C36H44N4O4 [M + H]+ 597.3435,
found 597.3431.
[7] Y. Hong, J.-Y. Liao, D. Cao, X. Zang, D.-B. Kuang, L. Wang,
H. Meier, C.-Y. Su, J. Org. Chem. 2011, 76, 8015.
[8] H. Li, Y. Hou, Y. Yang, R. Tang, J. Chen, H. Wang, H. Han,
T. Peng, Q. Li, Z. Li, ACS Appl. Mater. Interfaces 2013, 5, 12469.
[9] Z.-S. Huang, C. Cai, X.-F. Zang, Z. Iqbal, H. Zeng, D.-
B. Kuang, L. Wang, H. Meier, D. Cao, J. Mater. Chem. A 2015,
3, 1333.
[10] Y. Hong, Z. Iqbal, X. Yin, D. Cao, Tetrahedron 2014, 70, 6296.
[11] C.-Y. Lo, D. Kumar, S.-H. Chou, C.-H. Chen, C.-H. Tsai, S.-
H. Liu, P.-T. Chou, K.-T. Wong, ACS Appl. Mater. Interfaces
2016, 8, 27832.
[12] H. Shang, Y. Luo, X. Guo, X. Huang, X. Zhan, K. Jiang,
Q. Meng, Dyes Pigm. 2010, 87, 249.
[13] Y.-F. Chen, J.-M. Liu, J.-F. Huang, L.-L. Tan, Y. Shen, L.-M. Xiao,
D.-B. Kuang, C.-Y. Su, J. Mater. Chem. A 2015, 3, 8083.
[14] J. E. Savage, J. F. O'Sullivan, B. M. Zeis, R. Anderson,
J. Antimicrob. Chemother. 1989, 23, 691.
[15] M. Holzapfel, C. Lambert, C. Selinka, D. Stalke, J. Chem, Soc.,
Perkin Trans. 2002, 2, 1553.
4 | CONCLUSIONS
[16] Z. Zheng, Q. Dong, L. Gou, J.-H. Su, J. Huang, J. Mater. Chem.
C 2014, 2, 9858.
In summary, in this work a new DHP-based double-
anchoring dye DK-11 for serving as a sensitizer for DSSC
was reported. The unique and simple structure can be
economically synthesized without involving expensive
metal-catalyzed coupling reactions. The dye DK-11
showed a blue shifted absorption after the addition of a
base due to the outcome of the deprotonation of -COOH,
leading to the reduced charge transfer transition. The dye
was proposed to adopt a butterfly-like structure that
allows the efficient binding of both carboxylic units on
TiO2, which was confirmed by IR experiments. The di-
anchoring feature of DK-11 on TiO2 suppresses the dye
[17] T. Okamoto, E. Terada, M. Kozaki, M. Uchida, S. Kikukawa,
K. Okada, Org. Lett. 2003, 5, 373.
[18] Z. Zhang, Y.-S. Wu, K.-C. Tang, C.-L. Chen, J.-W. Ho, J. Su,
H. Tian, P.-T. Chou, J. Am. Chem. Soc. 2015, 137, 8509.
[19] L. Shi, W. Song, C. Lian, W. Chen, J. Mei, J. Su, H. Liu,
H. Tian, Adv. Optical Mater. 2018, 6, 1800190.
[20] C.-H. Lim, M. D. Ryan, B. G. McCarthy, J. C. Theriot,
S. M. Sartor, N. H. Damrauer, C. B. Musgrave, G. M. Miyake,
J. Am. Chem. Soc. 2017, 139, 348.
[21] A. Vilsmeier, A. Haack, Ber. Dtsch. Chem. Ges. 1927, 60, 119.
[22] E. Knoevenagel, Ber. Dtsch. Chem. Ges. 1896, 29, 172.
[23] D. Kumar, K. R. J. Thomas, C.-P. Lee, K.-C. Ho, J. Org. Chem.
2014, 79, 3159.

KUMAR ET AL.
9
[24] J. Park, C. Barolo, F. Sauvage, N. Barbero, C. Benzi,
P. Quagliotto, S. Coluccia, D. D. Censo, M. Grätzel,
M. K. Nazeeruddin, G. Viscardi, Chem. Commun. 2012, 48,
2782.
[25] H.-C. Ting, C.-H. Tsai, J.-H. Chen, L.-Y. Lin, S.-H. Chou, K.-
T. Wong, T.-W. Huang, C.-C. Wu, Org. Lett. 2012, 14, 6338.
[26] S.-S. Park, Y.-S. Won, Y.-C. Choi, J. H. Kim, Energy Fuel 2009,
23, 3732.
How to cite this article: Kumar D, Chen Y-L,
Tsai C-H, Wong K-T. Dihydrophenazine-based
double-anchoring dye for dye-sensitized solar cells.
J Chin Chem Soc. 2019;1–9. https://doi.org/10.
1002/jccs.201900410