4-Methoxy-1,3-thiazole based donor-acceptor dyes: Characterization, X-ray structure, DFT calculations and test as sensitizers for DSSC

Article abstract of DOI:10.1016/j.dyepig.2012.02.014

Four donor-(π-conjugated-bridge)-acceptor type dyes A-D were designed and synthesized. These new compounds use an unconventional 4-hydroxy-1,3- thiazole building block as an additional chromophore for light harvesting and to extend the π-conjugated system of the molecules. The synthetic route involved a double N-arylation Hartwig-Buchwald reaction using Pd(dba) ₂ as precatalyst and P(^tBu)₃ as ligand. Two different triarylamines and a 4-methoxyphenyl group were used as electron donor moieties. The electron acceptor (anchoring) group was 2-cyanoacrylic acid for all dyes, whereas the π-spacer was varied and the influence was investigated. The dyes were thoroughly characterized using photophysical and electrochemical methods and by density functional theory calculations. Additionally, they were evaluated in nanocrystalline TiO₂-based dye-sensitized solar cells (DSSCs). The DSSCs were prepared with and without deoxycholic acid (DCA) as a co-adsorbent to inhibit dye aggregation. The efficiencies obtained were low for DSSCs fabricated without DCA, but were significantly improved for DCCS with co-adsorbed DCA. Additionally, the X-ray structure of dye D was obtained, demonstrating the stereochemistry and planar geometry of the molecule. The DSSC based on dye A showed an efficiency of η = 1.70% (J_sc = 4.49 mA cm^-2, V_OC = 0.61 V, FF = 0.62) under 100 mW cm ^-2 simulated AM 1.5 G solar irradiation compared to η = 4.1% of the standard N₃ obtained under same conditions.

Full text of DOI:10.1016/j.dyepig.2012.02.014

Dyes and Pigments 94 (2012) 512e524
Contents lists available at SciVerse ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
4-Methoxy-1,3-thiazole based donor-acceptor dyes: Characterization, X-ray
structure, DFT calculations and test as sensitizers for DSSC
Roberto Menzel a, Daniel Ogermann b, Stephan Kupfer c, Dieter Weiß a, Helmar Görls d,
,
c
,
e,
a,
*
Karl Kleinermanns b , Leticia González
, Rainer Beckert
**
***
a Institute of Organic and Macromolecular Chemistry, Friedrich Schiller University Jena, Humboldtstraße 10, 07743 Jena, Germany
b Institute of Physical Chemistry I, Heinrich Heine University, Universitätsstraße 1, 40225 Düsseldorf, Germany
c Institute of Physical Chemistry, Friedrich Schiller University Jena, Helmholtzweg 4, 07743 Jena, Germany
d Institute of Inorganic and Analytical Chemistry, Friedrich Schiller University Jena, Lessingstr. 8, 07743 Jena, Germany
e Institute of Theoretical Chemistry, University of Vienna, Waehringer Str. 17, 1090 Vienna, Austria
a r t i c l e i n f o
a b s t r a c t
Article history:
Four donor-(
p-conjugated-bridge)-acceptor type dyes A-D were designed and synthesized. These new
Received 27 January 2012
Received in revised form
20 February 2012
Accepted 21 February 2012
Available online 25 February 2012
compounds use an unconventional 4-hydroxy-1,3-thiazole building block as an additional chromophore
for light harvesting and to extend the
p-conjugated system of the molecules. The synthetic route
involved a double N-arylation HartwigeBuchwald reaction using Pd(dba)2 as precatalyst and P(tBu)3 as
ligand. Two different triarylamines and a 4-methoxyphenyl group were used as electron donor moieties.
The electron acceptor (anchoring) group was 2-cyanoacrylic acid for all dyes, whereas the
p-spacer
was varied and the influence was investigated. The dyes were thoroughly characterized
using photophysical and electrochemical methods and by density functional theory calculations. Addi-
tionally, they were evaluated in nanocrystalline TiO2-based dye-sensitized solar cells (DSSCs). The DSSCs
were prepared with and without deoxycholic acid (DCA) as a co-adsorbent to inhibit dye aggregation.
The efficiencies obtained were low for DSSCs fabricated without DCA, but were significantly improved for
DCCS with co-adsorbed DCA. Additionally, the X-ray structure of dye D was obtained, demonstrating the
stereochemistry and planar geometry of the molecule. The DSSC based on dye A showed an efficiency of
Keywords:
4-Hydroxy-1,3-thiazoles
Double N-arylation
DSSC
DFT calculations
Donor-p-acceptordyes
X-ray structure
h
¼ 1.70% (Jsc ¼ 4.49 mA cmꢀ2, VOC ¼ 0.61 V, FF ¼ 0.62) under 100 mW cmꢀ2 simulated AM 1.5 G solar
irradiation compared to
h
¼ 4.1% of the standard N3 obtained under same conditions.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
In this paper, new thiazole dyes were designed, that consist of
a donor-( -conjugated-bridge)-acceptor structure (De eA), e.g.
p
p
The 4-hydroxy-1,3-thiazoles are a well known type of hetero-
cyclic compounds [1]. This building block was revived from our
group due to the similarity to the naturally occurring luciferine, the
light-emitting dye of fireflies, and their interesting optical proper-
ties. Since then, they have received interest due to their use as light
harvesting ligands in Ru(II) polypyridyl complexes [2], as blue-
emitting dyes in polymers [3], and because of their tuneable
absorption and emission spectra [4], high quantum yields and
extinction coefficients, and easy functionalization [5].
a main prerequisite of sensitizers in dye-sensitized solar cells
(DSSCs). Molecules with this special structural architecture have
attracted attention since they can be used as emissive materials in
molecular electronics [6,7], as fluorescent probes in biochemical
applications [8], as nonlinear optical (NLO) materials [9,10], in
organic light-emitting diodes (OLEDs) [11] and of course, in DSSCs.
DSSCs have received considerable attention as one possibility to
convert solar energy into electricity. Numerous compounds have
been tested [12e15] since their description by Grätzel et al. in 1991
[16], and the efficiency was driven up to 11.1% for small DSSCs
sensitized with N3 and up to 10.3% for DSSCs sensitized by an
organic-dye [17,18]. Recently, a new record was obtained from Yella
et al. using a porhpyrine dye as sensitizer and a cobalt (II/III) redox
electrolyte instead of the commonly used iodide/triiodide (Iꢀ/I3ꢀ)
redox couple, exceeding an efficiency of 12% [19]. The advantages of
organic sensitizers toward Ru(II) polypyridyl complexes, such as N3,
N719, and the black dye, are higher molar absorption coefficients,
* Corresponding author. Fax: þ49 3641 948212.
** Corresponding author. Fax: þ49 211 81 12179.
*** Corresponding author. Fax: þ49 3641 9 48302.
E-mail addresses: kleinermanns@uni-duesseldorf.de (K. Kleinermanns),
leticia.gonzalez@univie.ac.at (L. González), c6bera@uni-jena.de (R. Beckert).
0143-7208/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.dyepig.2012.02.014

R. Menzel et al. / Dyes and Pigments 94 (2012) 512e524
513
often simpler synthesis procedures and structure varieties and,
naturally, lower costs for the production of a DSSC [20]. The record
mentioned for a DSSC with an organic sensitizer was set by a dye
containing a triarylamine-based donor. This moiety is extraordi-
narily promising in various fields of organic dyes [21e26] and was,
therefore, predominantly used in this study.
measurement ferrocene (Fc/Fcþ) was added as an internal standard.
Starting materials were commercially obtained from SigmaeAldrich
and used as received. TLC was from Merck (Polygram SIL G/UV254,
aluminum oxide 60 F254). The material for Column chromatography
was also obtained from Merck (Silica gel 60).
One important possibility to tune the optical properties of
2.2. Synthesis of the compounds
organic sensitizers is by alternating the p-conjugated bridge which
typically consists of thiophenes and corresponding derivatives (e.g.
bithiophene, ethyldioxythiophene). They proved to be more
advantageous than an oligoene bridge, which undergoes cis-trans
isomerisation or oxidation reactions [27]. In this contribution, we
substituted the commonly used thiophene unit with a novel
2.2.1. 5-(4-Nitrophenyl)-2-(thiophen-2-yl)thiazol-4-ol (1)
A mixture of thiophene-2-carbothioamide (2.50 g, 17.5 mmol)
and ethyl 2-bromo-2-(4-nitrophenyl)acetate (5.46 g, 19.0 mmol, 1.1
equiv) in 100 mL toluene was heated under reflux for 24 h, and an
orange precipitate occurred. After the reaction was cooled down to
50 ꢁC, pyridine (5 mL) was added and stirring was continued for
additional 30 min. The mixture was cooled to 0 ꢁC, filtered through
a glass frit, washed with cold EtOH and pentane, and dried in vacuo.
The orange product was pure in terms of elemental analysis and
used without further purification (3.57 g, 11.7 mmol, 67%). mp
etherified 4-hydroxy-1,3-thiazole chromophore to build up D-p-A
dyes with high molar extinction coefficient and to expand the
absorption spectra further into the solar spectral region. Dyes with
a similar structure already proofed to be effective as sensitizers in
DSSCs [28,29]. Consequently, two different triarylamines and one
4-methoxyphenyl unit were used as electron donors and the
influence of the length of the conjugated bridge was verified. 2-
cyanoacrylic acid was used as the electron acceptor/anchoring
group for the nanocrystalline TiO2 surface. The dyes obtained were
characterized by UV/Vis and emission spectroscopy, electro-
chemical measurements, and quantum chemical methods.
Furthermore, they were tested regarding their application as
sensitizers in DSSCs.
279 ꢁC (decomp.). 1H NMR (250 MHz, DMSO-d6):
d 7.19 (dd,
J ¼ 4.0 Hz, J ¼ 4.8 Hz, 1H), 7.69 (d, J ¼ 3.2 Hz, 1H), 7.78 (d, J ¼ 5.0 Hz,
1H), 7.87 (m, 2H), 8.20 (m, 2H), 12.53 (s, 1H). 13C NMR (63 MHz,
DMSO-d6): d 160.42, 156.86, 144.09, 138.77, 136.05, 129.949, 128.76,
127.60, 125.63, 124.19, 104.65. MS-EI: m/z 304 (M$þ, 100%), 110 (M$þ
e C8H4NO3S, 60%). Anal. Calc. for C13H8N2O3S2: C, 51.30; H, 2.65; N,
9.20; S, 21.07. Found: C, 51.43; H, 2.84; N, 9.16; S, 20.92. UV/Vis
(DMSO): lmax (log ): 325 (3.78), 429 (4.10), 611 (4.39).
3
2. Experimental section
2.2.2. 4-Methoxy-5-(4-nitrophenyl)-2-(thiophen-2-yl)thizaole (2)
A mixture of 1 (1.86 g, 6.11 mmol) and KOH (0.41 g, 7.33 mmol,
1.2 equiv) in DMSO (50 mL) was stirred for 30 min followed by the
addition of CH3I (1.04 g, 7.33 mmol, 1.2 equiv). The deep blue
mixture was stirred for 24 h at room temperature. The solution was
poured into H2O (150 mL) and was extracted with CHCl3
(3 ꢂ 50 mL). The combined organic phases were additionally
washed with H2O (3 ꢂ 50 mL) to remove the DMSO, dried over
MgSO4 and concentrated in vacuo to give a brown oil which was
further purified by a short gel filtration (silica, CHCl3) to yield the
ether as an orange solid (1.3 g, 4.09 mmol, 67%). mp 144e145 ꢁC. 1H
2.1. General procedures and spectroscopic measurements
1H, 13C NMR and the corresponding correlation spectra were
recorded on a Bruker AC-250 (250 MHz) and AC-400 (400 MHz)
spectrometer. Chemical shifts (d) are given relative to solvents and all
coupling constants are given in Hz. UV/Vis data of the compounds
were collected on a Lambda 19 from PERKIN-ELMER and fluores-
cence spectra were measured on a Jasco FP 6500. Optical absorption
spectra of the adsorbed dyes were recorded by using a Cary 300 UV/
Vis spectrophotometer operated at a resolution of 1 nm. The
absorption spectra of the electrodes were measured in a reflection
arrangement and the absorption of pure TiO2 nanoparticles was
subtracted. Measurements of the fluorescence intensity were carried
out on a Perkin Elmer lambda16 UV/VIS spectrometer in the
perpendicular excitation-emission geometry, at excitation condi-
tions where the absorbance in the most red-shifted absorption was
<0.05. The emission quantum yields were determined by comparing
the corrected emission spectra with the reference spectrum of
NMR (250 MHz, CDCl3):
d
4.22 (s, 3H), 7.10 (dd, J ¼ 3.8 Hz, J ¼ 5.0 Hz,
1H), 7.44 (dd, J ¼ 1.1 Hz, J ¼ 5.1 Hz, 1H), 7.54 (dd, J ¼ 1.1 Hz,
J ¼ 3.7 Hz,1H), 7.81 (m, 2H), 8.20 (m, 2H). 13C NMR (63 MHz, CDCl3):
d
161.00, 157.02, 145.14, 138.48, 137.18, 128.63, 128.17, 126.71, 126.24,
124.12, 107.81, 58.12. MS-EI: m/z 318 (M$þ, 100%), 288 (M$þ - CH2O,
10%). Anal. Calc. for C14H10N2O3S2: C, 52.82; H, 3.17; N, 8.80; S, 20.14.
Found: C, 52.67; H, 2.90; N, 9.11; S 20.02. UV/Vis (CH2Cl2): lmax (log
3 ): 254 (3.80), 413 (4.30).
quinine sulfate (
absorption maximum. The fluorescence lifetimes of the thiazoles
were obtained by time-correlated single photon counting (TCSPC)
after excitation with
F
¼ 0.55). All compounds were excited in their
2.2.3. 4-(4-Methoxy-2-(thiophen-2-yl)thiazol-5-yl)aniline (3)
N2H5OH (1.7 mL, 18.7 mmol, 5 equiv, solution 80% in H2O) was
added to a suspension of 2 (1.18 g, 3.73 mmol) and freshly prepared
Raney nickel (catalytic amounts) in MeOH (100 mL) at 50 ꢁC. The
suspension turned green and became clear as the reaction pro-
ceeded. After the reaction had finished, as indicated by TLC (if the
conversion was not complete, more hydrazine was added), the
mixture was diluted with CH2Cl2 (100 mL) and filtered through
a frit on which a 2 cm-thick silica bed was applied to efficiently
remove the Raney nickel. The mixture was concentrated till dryness
and the crude product was either purified using flash chromatog-
raphy (silica, CHCl3/EtOAc 5:1) or by recrystallization from heptane/
CH2Cl2 by slow evaporation of the CH2Cl2, yielding the pure amine
as a yellow solid (1.02 g, 3.54 mmol, 95%). mp 114e115 ꢁC. 1H NMR
a frequency-doubled Ti-sapphire laser
(Tsunami, Newport Spectra-Physics GmbH), i.e. lex ¼ 650 nm. The
repetition rate of the laser was adjusted to 0.8 MHz by a pulse
selector (Model 3980, Newport Spectra-Physics GmbH). The emis-
sion of the thiazoles was detected using a streak-camera in TCSPC
mode. Elemental analysis was carried out on a Leco CHNS-932. Mass
spectra were measured either on a Finnigan MAT SSQ 710 (EI) or
MAZ 95 XL (FAB) system. MALDI-TOF MS was performed on a Bruker
Ultraflex TOF/TOF mass spectrometer equipped with a 337 nm
nitrogen laser operated in the reflectron mode using an acceleration
voltage of 25 kV, and dithranol (Bruker, #209783) was used as
matrix. Electrochemical measurements were performed on a Met-
(250 MHz, CDCl3):
d
7.56e7.47 (m, 2H), 7.44 (dd,J ¼ 3.7 Hz,
rohm Autolab PGSTAT30 potentiostat with
electrode configuration. The experiments were carried out in
degassed solvents containing 0.1 M Bu4NPF6 salt. At the end of each
a
standard three-
J ¼ 1.1 Hz, 1H), 7.34 (dd, J ¼ 5.1 Hz, J ¼ 1.1 Hz, 1H), 7.06 (dd, J ¼ 5.1,
J ¼ 3.7 Hz, 1H), 6.77e6.64 (m, 2H), 4.12 (s, 3H), 3.73 (s, 2H).13C NMR
(63 MHz, CDCl3): d 157.86, 152.60, 145.36, 138.27, 128.22, 127.99,

514
R. Menzel et al. / Dyes and Pigments 94 (2012) 512e524
126.98, 125.20, 121.92, 115.35, 111.55, 57.93. MS-EI: m/z 288 (M$þ
,
followed by warming up to room temperature and stirring for 24 h.
The mixture was concentrated and dried in vacuo and used without
further purification. The tin organyl thereby obtained was dissolved
in dry DMF (15 mL) and the mixture was degassed for 15 min. To
this solution, 2-bromothiophene (313 mL, 1.92 mmol, 1.5 equiv) and
freshly prepared Pd(PPh3)4 (60 mg, 0.052 mmol, 4 mol%) were
added, and the mixture was heated to 95 ꢁC for 16 h till no educt
was left, as indicated by TLC (silica, CHCl3/heptane 5:2, Rf
educt ¼ 0.6, Rf product ¼ 0.8). The mixture was concentrated and
directly applied to column chromatography yielding the product as
a red solid (578 mg, 1.05 mmol, 82%). mp 161e162 ꢁC. 1H NMR
70%), 136 (M$þ - C7H6NOS, 100%). Anal. Calc. for C14H12N2S2O: C,
58.31; H, 4.19; N, 9.71; S 22.24. Found: C, 57.98; H, 4.12; N, 9.51; S,
22.01. UV/Vis (CH3CN): lmax (log ): 233 (4.05), 248 (4.03), 283
3
(3.92), 397 (4.35).
2.2.4. 4-(4-Methoxy-2-(thiophen-2-yl)thiazol-5-yl)-N,N-dip-tolyl-
aniline (3a)
To a solution of 3 (740 mg, 2.57 mmol) in dry and degassed
toluene (30 mL) were added Pd(dba)2 (74 mg, 0.128 mmol, 5 mol%),
P(tBu)3 (130
mL of a 1 M solution in toluene, 0.130 mmol, 5 mol%), p-
bromotoluene (970 mg, 5.70 mmol, 2.2 equiv) and KOtBu (633 mg,
5.70 mmol, 2.2 equiv). The mixture was degassed for 30 min and
heated to 90 ꢁC for 2 days till no educt and monosubstituted
product was left as indicated by TLC (silica, CHCl3/heptane 3:1 Rf
educt ¼ 0.2, Rf monosubstituted product ¼ 0.6, Rf disubstituted
product ¼ 0.7). After the reaction was finished, the mixture was
allowed to cool down to RT, was washed with H2O (3 ꢂ 50 mL),
dried over MgSO4 and concentrated in vacuo till dryness. Purifica-
tion using column chromatography yielded the product as a yellow
solid (1.10 g, 2.34 mmol, 91%). mp 174e175 ꢁC. 1H NMR (250 MHz,
(300 MHz, CDCl3):
7.27 (d, J ¼ 4.4 Hz, 2H), 7.20e7.01 (m, 12H), 4.17 (s, 3H), 2.37 (s,
6H).13C NMR (75 MHz, CDCl3):
158.37, 152.55, 146.76, 145.13,
d
7.58 (d, J ¼ 8.7 Hz, 2H), 7.34 (d, J ¼ 3.9 Hz, 1H),
d
138.96, 137.02, 136.46, 132.72, 130.00, 128.10, 127.49, 125.91, 125.12,
124.78, 124.60, 124.35, 124.29, 122.52, 111.02, 57.84, 20.94. HRMS
Micro-ESI: m/z 551.12687 (calc. for C32H26N2OS3 þ Hþ: 551.12855).
UV/Vis (CH3CN): lmax (log ): 240 (3.93), 304 (4.15), 442 (4.36).
3
2.2.7. 5-(5-(4-(Dip-tolylamino)phenyl)-4-methoxythiazol-2-yl)thi-
ophene-2-carbaldehyde (4a, general procedure)
CDCl3):
J ¼ 5.1 Hz, J ¼ 1.0 Hz, 1H), 7.45 (dd, J ¼ 3.7 Hz, J ¼ 1.1 Hz, 1H),
7.57e7.49 (m, 2H). 13C NMR (63 MHz, CDCl3):
158.21, 153.04,
146.70, 145.05, 138.01, 132.64, 129.87, 127.87, 127.41, 127.03, 125.24,
d
2.32 (s, 6H), 4.12 (s, 3H), 7.13e6.96 (m, 11H), 7.35 (dd,
To a solution of 3a (626 mg, 1.34 mmol) in dry THF (10 mL)
at ꢀ78 ꢁC under a nitrogen atmosphere, was added n-BuLi (0.59 mL,
1.47 mmol, 1.1 equiv, 2.5 M solution in hexanes) and the mixture
was stirred for 30 min. After warming up to room temperature and
additional stirring for 30 min, the mixture was cooled again
to ꢀ78 ꢁC and approx. Dry DMF (1 mL) was added and stirring was
continued for 30 min followed by warming up to room temperature
and further stirring for 1 h. The mixture was neutralized with
saturated NH4Cl solution, diluted with CHCl3 (50 mL) and washed
with H2O (3 ꢂ 50 mL). The organic phase was dried over MgSO4 and
concentrated in vacuo. Column chromatography (silica, CHCl3 to
CHCl3/EtOAc 3:1) yielded the product as a red solid (600 mg,
d
124.65, 124.51, 122.47, 110.83, 57.77, 20.79. MS-EI: m/z 468 (M$þ
,
100%), 316 (M$þ - C7H6NOS, 90%). Anal. Calc. for C28H24N2OS2: C,
71.76; H, 5.16; N, 5.98; S, 13.68. Found: C, 71.98; H, 5.18; N, 5.82; S,
13.44. UV/Vis (CHCl3): lmax (log ): 311 (4.23), 417 (4.35).
3
If the reaction was aborted earlier (after 2 h), the mono-
substituted product, 4-(4-methoxy-2-(thiophen-2-yl)thiazol-5-yl)-
N-p-tolylaniline, was obtained in 70% yield after purification. 1H
NMR (250 MHz, CDCl3):
d
7.58 (d, J ¼ 8.7 Hz, 1H), 7.45 (d, J ¼ 3.4 Hz,
1H), 7.35 (d, J ¼ 4.9 Hz, 1H), 7.21e6.89 (m, 2H), 5.69 (s, 1H), 4.14 (s,
1H), 2.32 (s, 1H). MS-EI: m/z 378 (M$þ, 70%), 226 (M$þ - C7H6NOS,
100%).
1.21 mmol, 90%). mp 151e152 ꢁC. 1H NMR (400 MHz, CDCl3):
d 9.90
(s, 1H), 7.69 (d, J ¼ 4.0 Hz, 1H), 7.54 (d, J ¼ 8.8 Hz, 2H), 7.47 (d,
J ¼ 4.0 Hz, 1H), 7.08 (d, J ¼ 8.3 Hz, 4H), 7.05e6.97 (m, 6H), 4.13 (s,
3H), 2.33 (s, 6H).13C NMR (100 MHz, CDCl3):
d 182.59, 159.00,
2.2.5. 4-Methoxy-N-(4-(4-methoxy-2-(thiophen-2-yl)thiazol-5-yl)
phenyl)-N-(4-methoxyphenyl)aniline (3b)
150.45, 147.41, 146.60, 144.84, 143.33, 136.55, 133.01, 129.95, 127.66,
125.00, 124.93, 123.55, 121.96, 114.30, 57.88, 20.81. MS-EI: m/z 496
(M$þ, 100%), 468 (M$þ - CO, 20%), 316 (M$þ - C8H6NO2S, 70%). Anal.
Calc. for C29H24N2O2S2: C, 70.13; H, 4.87; N, 5.64; S, 12.91; Found: C,
The procedure was similar to that for 3a. 3 (170 mg, 0.59 mmol),
Pd(dba)2 (10.2 mg, 0.018 mmol, 3 mol%), P(tBu)3 (30
mL of a 1 M
solution in toluene, 0.030 mmol, 5 mol%), p-bromoanisole (243 mg,
1.29 mmol, 2.2 equiv), KOtBu (146 mg, 0.29 mmol, 2.2 equiv),
toluene (10 ml). After purification (silica, CHCl3/heptane 3:1 Rf
educt ¼ 0.2, Rf monosubstituted product ¼ 0.7, Rf disubstituted
product ¼ 0.8), the product was obtained as an orange oil which
solidified after a few hours (266 mg, 0.53 mmol, 90%). mp
70.08; H, 4.68; N, 5.92; S, 13.04. UV/Vis (CHCl3): lmax (log ): 311
(4.29), 332 (4.18), 485 (4.39).
3
2.2.8. 5-(5-(4-(Bis(4-methoxyphenyl)amino)phenyl)-4-methoxyth-
iazol-2-yl)thiophene-2-carbaldehyde (4b)
The procedure was similar to that for 4a. 3b (249 mg,
0.50 mmol), THF (10 mL), n-BuLi (0.22 mL, 0.55 mmol, 1.1 equiv),
DMF (1 mL). After purification (silica, CHCl3), the product was
obtained as a deep red solid (230 mg, 0.44 mmol, 87%). mp
131e132 ꢁC. 1H NMR (250 MHz, CDCl3):
d 7.56e7.47 (m, 2H), 7.45
(dd, J ¼ 3.7 Hz, J ¼ 1.0 Hz, 1H), 7.34 (dd, J ¼ 5.1 Hz, J ¼ 1.0 Hz, 1H),
7.13e7.02 (m, 5H), 6.93 (d,J ¼ 8.8 Hz, 2H), 6.89e6.79 (m, 4H), 4.12 (s,
3H), 3.81 (s, 6H).13C NMR (63 MHz, CDCl3):
d
158.19, 156.04, 152.88,
90e91 ꢁC. 1H NMR (250 MHz, CDCl3):
d 9.89 (s, 1H), 7.68 (d,
147.40, 140.86, 138.18, 128.00, 127.55, 127.10, 126.67, 125.31, 123.59,
120.73, 114.84, 111.15, 57.90, 55.62; HRMS Micro-ESI: m/z
500.123743 (calc. for C28H24N2O3S2: 500.12283). UV/Vis (CH3CN):
J ¼ 4.0 Hz, 1H), 7.55e7.47 (m, 2H), 7.45 (d,J ¼ 4.0 Hz, 1H), 7.13e7.00
(m, 4H), 6.96e6.88 (m, 2H), 6.88e6.78 (m, 4H), 4.12 (s, 3H), 3.80 (s,
6H).13C NMR (63 MHz, CDCl3):
d 182.79, 158.98, 156.27, 150.29,
lmax (log
3
): 238 (4.20), 302 (4.30), 414 (4.42).
148.05, 146.79,143.35,140.56,136.81,127.79,126.93,125.06,122.64,
120.21, 114.90, 114.62, 58.01, 55.63. MALDI-TOF: 528.1 (M$þ). Anal.
Calc. for C29H24N2O4S2: C, 65.89; H, 4.58; N, 5.30; O, 12.11; S, 12.13.
Found: C, 65.87; H, 4.54; N, 5.30; S, 12.13. UV/Vis (CH3CN): lmax (log
2.2.6. 4-(2-(2,20-Bithiophen-5-yl)-4-methoxythiazol-5-yl)-N,N-
dip-tolylaniline (3c)
To a solution of 3a (600 mg, 1.28 mmol) in dry THF (15 mL)
at ꢀ78 ꢁC under a nitrogen atmosphere was added n-BuLi
(0.615 mL, 1.54 mmol, 1.2 equiv, 2.5 M solution in hexanes) and the
mixture was stirred for 30 min. After warming up to room
temperature and additional stirring for 30 min, the mixture was
cooled again to ꢀ78 ꢁC and tributyltin chloride (625 mg, 1.92 mmol,
1.5 equiv) was added, and stirring was continued for 30 min more,
3
): 298 (4.29), 331 (4.18), 476 (4.41).
2.2.9. 5’-(5-(4-(Dip-tolylamino)phenyl)-4-methoxythiazol-2-yl)-2,2’-
bithiophene-5-carbaldehyde (4c)
The procedure was similar to that for 4a. 3c (390 mg,
0.715 mmol), THF (5 mL), n-BuLi (0.37 mL, 0.93 mmol, 1.1 equiv),
DMF (1 mL). After purification (silica, CHCl3), the product was

R. Menzel et al. / Dyes and Pigments 94 (2012) 512e524
515
obtained as a black solid (380 mg, 0.66 mmol, 93%). mp 193e194 ꢁC.
1H NMR (250 MHz, CDCl3):
2.33 (s, 6H), 4.13 (s, 3H), 7.03 (m, 10H),
(63 MHz, DMSO-d6):
d
163.50, 157.98, 151.22, 146.55, 146.11, 144.61,
d
144.33, 141.28, 137.79, 136.44, 134.67, 132.71, 130.14, 128.01, 127.43,
125.84, 125.82, 124.62, 123.26, 121.48, 116.53, 111.41, 98.71, 57.84,
20.45. MS-Micro-ESI (neg.): m/z 644 (M$þ). Anal. Calc. for
C36H27N3O3S3: C, 66.95; H, 4.21; N, 6.51; O, 7.43; S, 14.90. Found: C,
66.87; H, 4.31; N, 6.40; S, 15.01. IR (ATR, cmꢀ1): 3030 (w), 2930 (br),
2350 (m), 2315 (m), 1600 (s), 1500 (s), 1380 (s), 1320 (s), 1270 (s),
7.27 (d, J ¼ 4.3 Hz, 2H), 7.33 (d, J ¼ 4.0 Hz, 1H), 7.52 (d, J ¼ 8.7 Hz,
2H), 7.66 (d, J ¼ 4.0 Hz, 1H). 13C NMR (63 MHz, CDCl3):
d 182.29,
158.44, 151.38, 146.93, 146.35, 144.88, 141.99, 139.16, 137.21, 136.83,
132.78, 129.89, 127.42, 126.72, 125.76, 124.76, 124.48, 123.99, 122.14,
112.10, 57.77, 20.80. MS-EI: m/z 578 (M$þ, 100%), 316 (M$þ
e
C12H8NO2S2, 80%). Anal. Calc. for C33H26N2O2S3: C, 68.48; H, 4.53; N,
4.84; S, 16.62. Found: C, 68.46; H, 4.65; N, 4.77; S, 16.40. UV/Vis
1080 (m). UV/Vis (THF): lmax (log ): 278 (4.13), 307 (4.24), 400
(4.23), 490 (4.50).
3
(CHCl3): lmax (log ): 263 (4.13), 313 (4.32), 376 (4.24), 486 (4.50).
3
2.2.13. 5-(4-Methoxyphenyl)-2-(thiophen-2-yl)thiazol-4-ol (5)
To solution of thiophene-2-carbothioamide (3.25 g,
2.2.10. 2-Cyano-3-(5-(5-(4-(dip-tolylamino)phenyl)-4-
a
methoxythiazol-2-yl)thiophen-2-yl)acrylic acid (Dye A, general
procedure)
22.7 mmol) in DMF (50 mL) was added ethyl 2-bromo-2-(4-
methoxyphenyl)acetate (9.30 g, 34.1 mmol, 1.5 equiv). The solu-
tion was heated to 50 ꢁC and maintained there for 24 h under
continuous stirring. After cooling down to RT, H2O (100 mL) was
added and the product precipitated as a yellow solid. The mixture
was stirred for additional 30 min. The product was filtered off,
washed thoroughly with EtOH and Et2O, and dried in vacuo. The
light yellow solid was used without further purification (4.6 g,
15.9 mmol, 70%). mp 227e228 ꢁC. 1H NMR (250 MHz, DMSO-d6):
To a solution of 4a (143 mg, 0.29 mmol) and cyanoacetic acid
(27 mg, 0.32 mmol, 1.1 equiv) in CH3CN (5 mL) was added piperi-
dine (cat. amounts, one drop). The deep red mixture was heated to
reflux under a nitrogen atmosphere for 4 h till no starting material
was left as indicated by TLC. After cooling down to room temper-
ature, the suspension was diluted with CHCl3 (20 mL) and washed
with H2O (3 ꢂ 20 mL). The organic phase was dried over MgSO4,
concentrated in vacuo and the residue was purified by flash chro-
matography (silica, CHCl3 to CHCl3/MeOH 5:1) to yield the product
as a glasslike black solid (123 mg, 0.22 mmol, 76%). mp 190 ꢁC
d
3.75 (s, 3H), 6.96 (d, J ¼ 8.77 Hz, 2H), 7.15 (dd, J ¼ 4.73 Hz,
J ¼ 3.98 Hz, 1H), 7.65e7.54 (m, 3H), 7.68 (d,J ¼ 4.93 Hz, 1H), 11.41 (s,
1H). 13C NMR (63 MHz, DMSO-d6):
157.56, 156.64, 152.51, 136.71,
d
(decomp.). 1H NMR (400 MHz, DMSO-d6):
d
8.25 (s, 1H), 7.77 (d,
128.46, 128.33, 127.12, 125.98, 123.96, 114.23, 106.58, 55.05. MS-EI:
m/z 289 (M$þ, 100%), 274 (M$þ - CH3, 15%). Anal. Calc. for
C14H11NO2S2: C, 58.11; H, 3.83; N, 4.84; S, 22.16. Found: C, 58.27; H,
J ¼ 4.0 Hz,1H), 7.65 (d, J ¼ 4.0 Hz,1H), 7.50 (d, J ¼ 8.7 Hz, 2H), 7.10 (d,
J ¼ 8.2 Hz, 4H), 6.91 (d, J ¼ 8.3 Hz, 4H), 6.86 (d, J ¼ 8.7 Hz, 2H), 4.04
(s, 3H), 2.26 (s, 6H). 13C NMR (100 MHz, DMSO-d6):
d
163.43, 158.35,
4.08; N, 4.47; S, 21.80. UV/Vis (DMSO): lmax (log ): 394 (4.12).
3
150.87, 146.75, 144.24, 142.38, 141.84, 137.67, 137.47, 132.74, 130.07,
127.51, 126.39, 124.63, 122.98, 121.32, 117.93, 112.72, 106.36, 57.88,
20.37. MS-EI: m/z 563 (M$þ, 10%), 519 (M$þ - CO2, 30%). Anal. Calc.
for C32H25N3O3S2: C, 68.18; H, 4.47; N, 7.45; S, 11.38. Found: C,
68.27; H, 4.31; N, 7.49; S, 11.31. IR (ATR, cmꢀ1): 2970 (br), 2360 (m),
2330 (m), 1730 (s), 1560 (s), 1550 (s), 1495 (s), 1365 (s), 1320 (m),
2.2.14. 4-Methoxy-5-(4-methoxyphenyl)-2-(thiophen-2-yl)thiazole
(6)
The procedure was similar to that for 2. 5 (1.67 g, 5.78 mmol),
DMSO (40 mL), K2CO3 (96 g, 6.93 mmol, 1.2 equiv, red solution),
CH3I (0.98 g 6.93 mmol, 1.2 equiv), After purification, the product
was obtained as a yellow solid (1.01 g, 3.33 mmol, 58%). mp
1270 (m), 1230, (m), 1090 (w). UV/Vis (CHCl3): lmax (log
3 ): 308
(4.32), 360 (4.16), 517 (4.38). UV/Vis (THF): lmax (log
3 ): 302 (4.36),
75e76 ꢁC. 1H NMR (250 MHz, CDCl3):
d
7.63 (d, J ¼ 8.7 Hz, 2H), 7.45
354 (4.15), 520 (4.43).
(d, J ¼ 3.2 Hz, 1H), 7.35 (d, J ¼ 4.9 Hz, 1H), 7.11e7.01 (m, 1H), 6.92
(d,J ¼ 8.7 Hz, 2H), 4.13 (s, 3H), 3.83 (s, 3H). 13C NMR (63 MHz,
2.2.11. 3-(5-(5-(4-(Bis(4-methoxyphenyl)amino)phenyl)-4-
CDCl3):
d
¼ 158.48, 158.19, 153.35, 138.10, 128.21, 128.00, 127.19,
methoxyth-iazol-2-yl)thiophen-2-yl)-2-cyanoacrylic acid (Dye B)
The procedure was similar to that for Dye A. 4b (163 mg,
0.31 mmol), cyanoacetic acid (29 mg, 0.34 mmol, 1.1 equiv), CH3CN
(5 mL), piperidine. After purification, the product was obtained as
a black solid (164 mg, 0.28 mmol, 89%). mp 220 ꢁC (decomp.). 1H
125.41, 124.19, 114.30, 110.62, 57.92, 55.45. HRMS Micro-ESI: m/z
304.04572 (calc. for C15H13NO2S2 þ Hþ: 304.0466). UV/Vis (CHCl3):
lmax (log ): 252 (3.98), 385 (4.26).
3
2.2.15. 5-(4-Methoxy-5-(4-methoxyphenyl)thiazol-2-yl)thiophene-
2-carbaldehyde (7)
The procedure was similar to that for 4a. 6 (477 mg, 1.57 mmol),
THF (20 mL), n-BuLi (0.69 mL, 1.73 mmol, 1.1 equiv), DMF (2 mL).
After purification (silica, CHCl3), the product was obtained as an
orange solid (389 mg, 1.17 mmol, 75%). mp 128e129 ꢁC. 1H NMR
NMR (400 MHz, DMSO-d6):
d
8.16 (s,1H), 7.67 (t, J ¼ 7.9 Hz,1H), 7.61
(d, J ¼ 4.0 Hz, 1H), 7.45 (d, J ¼ 8.8 Hz, 2H), 7.03e6.98 (m, 4H),
6.93e6.87 (m, 4H), 6.75 (d,J ¼ 8.8 Hz, 2H), 4.03 (s, 3H), 3.73 (s, 6H).
13C NMR (100 MHz, DMSO-d6):
d
¼ 158.05, 155.96, 150.77, 147.45,
140.90, 140.84, 139.64, 137.94, 136.46, 127.48, 126.83, 126.23, 121.69,
119.07,118.72,115.00,112.59,109.27, 57.87, 55.25. MALDI-TOF: 595.1
(M$þ). Anal. Calc. for C32H25N3O5S2: C, 64.52; H, 4.23; N, 7.05; O,
13.43; S, 10.77. Found: C, 64.36; H, 4.20; N, 7.12; O, 13.32; S, 10.82. IR
(ATR, cmꢀ1): 3030 (w), 2940 (br), 2840 (w), 2350 (m), 2325 (m),
1741 (s), 1600 (m), 1500 (s), 1370 (s), 1240 (s), 1090 (m), 1030 (m).
(250 MHz, CDCl3):
d
¼ 9.91 (s, 1H), 7.69 (d, J ¼ 4.0 Hz, 1H), 7.68e7.61
(m, 2H), 7.48 (d,J ¼ 4.0 Hz, 1H), 7.01e6.87 (m, 2H), 4.15 (s, 3H), 3.82
(s, 3H).13C NMR (100 MHz, CDCl3): 182.69, 158.90, 150.80, 146.53,
143.41, 136.60, 130.64, 128.35, 125.09, 123.44, 114.30, 113.80, 57.91,
55.35. MS-EI: m/z: 331 (M$þ, 50%), 316 (M$þ - CH4, 10%), 151 (M$þ
-
UV/Vis (THF): lmax (log
3
): 300 (4.37), 364 (4.16), 505 (4.46).
C8H6NO2S, 70%). Anal. Calc. for C16H13NO3S: C, 57.99; H, 3.95; N,
4.23; S, 19.35. Found: C, 58.16; H, 3.99; N, 4.23; S, 19.36.
2.2.12. 2-Cyano-3-(50-(5-(4-(dip-tolylamino)phenyl)-4-
methoxythiazol -2-yl)-2,20-bithiophen-5-yl)acrylic acid (Dye C)
The procedure was similar to that for Dye A. 4c (230 mg,
0.40 mmol), cyanoacetic acid (37 mg, 0.44 mmol, 1.1 equiv), CH3CN
(10 mL), piperidine. After purification, the product was obtained as
a black solid (230 mg, 0.36 mmol, 90%). mp 243 ꢁC (decomp.). 1H
2.2.16. 2-Cyano-3-(5-(4-methoxy-5-(4-methoxyphenyl)thiazol-2-
yl)th-iophen-2-yl)acrylic acid (Dye D)
The procedure was similar to that for Dye A. 7 (324 mg,
0.98 mmol), cyanoacetic acid (92 mg, 1.08 mmol, 1.1 equiv), CH3CN
(15 mL), piperidine. The product precipitated as a red solid during
the reaction. The mixture was diluted with CHCl3 (50 mL) and
washed with H2O (3 ꢂ 50 mL). The organic phase was dried over
MgSO4 and concentrated in vacuo. The crude product was
NMR (250 MHz, DMSO-d6):
7.60 (d, J ¼ 4.0 Hz, 1H), 7.58e7.43 (m, 4H), 7.12 (d,J ¼ 8.2 Hz, 4H),
d
8.05 (s, 1H), 7.66 (d, J ¼ 4.0 Hz, 1H),
6.99e6.81 (m,J ¼ 8.6 Hz, 6H), 4.04 (s, 3H), 2.26 (s, 6H). 13C NMR

516
R. Menzel et al. / Dyes and Pigments 94 (2012) 512e524
recrystallized from a mixture THF/EtOH by slow evaporation of the
THF, yielding a deep red-black fine crystalline compound (351 mg,
0.88 mmol, 90%). mp 260 ꢁC (decomp.). 1H NMR (400 MHz, DMSO-
Here MPP is the “maximum power point” and Voc the open-
circuit voltage [30].
d6):
(d, J ¼ 8.7 Hz, 2H), 6.99 (d, J ¼ 8.7 Hz, 2H), 4.08 (s, 3H), 3.78 (s, 3H).
13C NMR (100 MHz, DMSO-d6):
163.25, 158.63, 158.38, 150.54,
d
8.47 (s, 1H), 7.96 (d, J ¼ 4.1 Hz,1H), 7.75 (d, J ¼ 3.9 Hz, 1H), 7.62
2.5. Computational methods
d
All calculations were performed with the GAUSSIAN 09 program
145.94, 143.98, 140.50, 136.33,128.03, 126.58, 122.53, 116.35, 114.49,
113.32, 99.99, 57.96, 55.19. MS-EI: m/z 398.5 (M$þ, 50%), 354.4 (M$þ
- CO2, 60%), 151.5 (M$þ - C11H7N2O3S 100%). Anal. Calc. for
C19H14N2O4S2: C, 57.27; H, 3.54; N, 7.03; S, 16.09. Found: C, 57.31; H,
3.49; N, 6.82; S, 16.06. IR (ATR, cmꢀ1): 3000 (w), 2970 (m), 2930
(w), 2360 (m), 2320 (m), 1730 (s), 1580 (w), 1500 (m), 1420 (m),
[31] in the presence of solvent (tetrahydrofuran,
3
¼ 7.4257,
n ¼ 1.4070) via the integral equation formalism of the polarizable
continuum model [32]. The ground state equilibrium structures
were optimized with the long-range corrected CAM-B3LYP XC
functional [33] and the 6e31G(d,p) double-x basis set. [34].
Harmonic vibrational frequencies at the same level of theory
proved that the stationary points obtained corresponded to the
minima of the potential energy surfaces. Excited state properties,
such as excitation energies, oscillator strengths and excited states
geometries, were computed via TDDFT and the same XC functional
and basis set as was employed for the ground state. The absorption
spectra in the FC region were simulated by the first ten singlet
excited states. According to Kasha’s rule fluorescence occurs from
the lowest excited singlet state [35], the S1 state was optimized to
obtain the emission energies from the S1.
1370 (s), 1230 (s), 1215 (s), 1200 (s). UV/Vis (THF): lmax (log ): 278
3
(4.11), 338 (4.00), 478 (4.60).
2.3. Preparation of the DSSC
TiO2 nanopowder (5 g, AEROXIDEÒ TiO2 P25, Evonik) was sus-
pended in HNO3 (1 N, Fluka) to prepare a suitable TiO2 paste. This
suspension was heated at 80 ꢁC for 24 h. The HNO3 was evaporated
and the TiO2 solid was dried for three days at 100 ꢁC. Finally, the
TiO2 solid was treated with 25 mL H2O, acetylacetone (2.5 g,
25.0 mmol, Merck), Triton X-100 (1.25 g, 1.92 mmol, Avocado), and
polyethylene oxide (M.W. 100 000, Alfa Aesar). Aluminoborosilicate
glass, coated with fluorine doped tin oxide (FTO, SnO2:F), was used
as a conducting, transparent substrate (resistance w 10 U/cm2,
Solaronix SA, Switzerland). The active area was masked with scotch
tape and was coated with TiO2 suspension by using a glass scraper
and dried at 80 ꢁC for 10 min. After removal of the scotch tape,
3. Results and discussion
3.1. Synthesis and structure of the dyes
The synthetic route and molecular structures of the triarylamine-
based dyes are depicted in Scheme 1. The starting material, 5-(4-
nitrophenyl)-2-(thiophen-2-yl)thiazol-4-ol (1), was synthesized
using a Hantzsch cyclization of thiophene-2-carbothioamide with
ethyl 2-bromo-2-(4-nitrophenyl)acetate. The 4-hydroxy-1,3-thiazole
obtained was alkylated following a Williamson-type etherification
with K2CO3 as the base and methyl iodide as the nucleophile to yield
the 4-methoxy-5-(4-nitrophenyl)-2-(thiophen-2-yl)thiazole (2).
The reduction of the nitro group was realized following a stand-
ard protocol using freshly prepared Raney nickel and hydrazine
hydrate solution as the hydrogen source in ethanol to almost quan-
titatively yield the amine 4-(4-methoxy-2-(thiophen-2-yl)thiazol-5-
yl)aniline (3). The crucial step in the synthetic route was the
HartwigeBuchwald coupling of the aryl amine 3 with an appropriate
aryl halide using a Pd-catalyst. Although thoroughly described in
literature, this type of coupling is difficult and strongly depends on
the nature of the catalysts, ligands and reaction conditions [36,37]. In
addition, the coupling has to take place twice to obtain the desired
product. This double N-arylation of primary amines is not often
described in literature. Examples exist for simple triarylamines
basically consisting of tolyl or anisole groups [38], or for a catalytic
process with a second, intramolecular coupling to generate different
types of carbazoles [39, 40]. Tri-tert-butylphosphine (P(tBu)3) was
chosen as a promising ligand, which has already been successfully
applied in the synthesis of different indole derivatives [41]. For the
couplingreaction, bis(dibenzylideneacetone)palladium(0) (Pd(dba)2)
was used as the precatalyst, KOtBu as the base to deprotonate the
amine in the catalytic cycle and toluene as solvent. The aryl halides
chosenwere p-bromotolueneand p-bromoanisole because ofa better
stability of the corresponding triarylamines toward oxidative
coupling reactions in para-position after excitation and radical
formation (charge separation), which is well known for these radicals
[42,43]. Surprisingly, the coupling reaction with P(tBu)3 as ligand
(several attempts using, for example, triphenylphosphine or 1,10-
bis(diphenylphosphino)ferrocene as ligands failed completely)
produced the coupling products in very good yields (>86%).
Furthermore, the two-step process of the reactionwas confirmed and
the monosubstituted product formed was isolated and identified for
3a (see experimental section). A noteworthy observation was that
approx. 11 mm were measured as deposit thickness. This primed
photo electrode was sintered in a muffle furnace for w45 min at
450 ꢁC. The TiO2 electrode was then immersed in the dye solution
(0.2 mM in THF) for 24 h at room temperature. A thin platinum
layer was spread on the FTO coating of the counter electrode
(Platisol, Solaronix SA, Switzerland). A few drops of the active redox
couple iodine/iodide (0.1 M LiI (Fluka), 0.05 M I2 (Aldrich), 0.6 M
Bu4NI (Aldrich) and 0.5 M 4-tert-butylpyridine (Aldrich) in CH3CN)
were added to the photoelectrode. Finally, the counter electrode
was clamped to the photoelectrode/electrolyte system.
2.4. Measurements
Measurements of the wavelength dependence of the short
circuit current density (JSC) were carried out in a light-proof box
with a 70 W Xenon lamp (Oriel, Germany) and a grating mono-
chromator (Zeiss, Germany) in the spectral range of 300e800 nm
with a resolution of 1 nm. The light intensity incident (Iinc) on the
electrode was measured with a power meter (Coherent, USA). The
IPCE was calculated using the following expression:
Â
Ã
JSC A cmꢀ2 1240
Â
Ã
IPCEð%Þ ¼
100
l
½nmꢃIinc W cmꢀ2
The photocurrent-voltage (IeV) curves were measured using
a focused 120 W Xenon lamp (Oriel, Germany) and a special filter to
simulate solar irradiation (100 mW cmꢀ2, 1 sun, 1.5 air mass (AM)
global). The fill factors (FF) and overall efficiencies (h) were calcu-
lated according to following equations.
Â
Ã
MPP W cmꢀ2
JSC A cmꢀ2 VOC½Vꢃ
Â
Ã
FF ¼
Â
Ã
JSC A cmꢀ2 VOC½VꢃFF
Â
Ã
h
ð%Þ ¼
100
Iinc W cmꢀ2

R. Menzel et al. / Dyes and Pigments 94 (2012) 512e524
517
Scheme 1. Synthesis and structure of the triarylamine-based dyes A-C.
when a high excess of coupling agent together with longer reaction
times were applied, an additional spot (TLC) eluting in front of the
coupling products, occurred. This by-product was identified (MS) as
the coupling product of the thiophene moiety at the electron rich 5-
position with the aryl halide.
In order to vary the conjugation lengths of the dyes, a further
thiophene ring was introduced in 3a via a Stille coupling with 2-
bromothiophene to yield 4-(2-(2,20-bithiophen-5-yl)-4-methoxyt-
hiazol-5-yl)-N,N-di-4-tolylaniline (3c) in very good yields (86%). The
introduction of the aldehyde group via a VilsmeiereHaack reaction
with DMF and POCl3 was not successful. Therefore, the thiophene was
deprotonated with n-BuLi followed by the addition of DMF and
quenching with saturated NH4Cl solution to give the corresponding
aldehydes in excellent yields(>87%). No side product due to a possible
directed ortho-metalation at the 4-methoxyphenyl ring for 3b was
observed if n-BuLi wasadded veryslowlyat ꢀ78ꢁC and the excesswas

518
R. Menzel et al. / Dyes and Pigments 94 (2012) 512e524
kept small. The last step, the introduction of the 2-cyanoacrylic acid
acceptor group, was achieved by a Knoevenagel condensation of the
aldehyde group with cyanoacetic acid and catalytical amounts of
piperidine in acetonitrile to give the dyes A-C in moderate to good
yields (>75%) Similar to the synthesis of 1, the starting material for D,
5-(4-methoxyphenyl)-2-(thiophen-2-yl)thiazol-4-ol
(5),
was
obtained by a cyclization reaction of the thiophene-2-carbothioamide
and ethyl 2-bromo-2-(4-methoxyphenyl)acetate. The reaction in
toluene only resulted in a sluggish conversion to 5, therefore a second
method, with DMF as the solvent and slightly elevated temperature
was used, which gave the product in a satisfactory yield (70%). All the
following reactions (alkylation, introduction of the aldehyde and
Knoevenagel condensation) were similar to that for dyes A-C and the
synthetic route is depicted in Scheme 2.
3.2. X-ray structure of dye D
Single Crystals suitable for X-ray structure analysis were
obtained from a mixture ethylene glycol/THF in an NMR tube by
slow evaporation of the THF. The presented structure is a rare
example in this field, because of the weak crystallization tendency
Fig. 1. Ortep plot of dye D. Ellipsoids are at 50% probability level. Shown are the
numbering scheme (above) and the two hydrogen bonds to form the dimer (below).
of this D-
p
-A type compounds. Usually, crystallization leads to
3.3. Photophysical properties
amorphous precipitates or very thin plates because of the planarity
of the molecules (aggregation).
The UV/Vis spectra of the protonated and deprotonated dyes are
depicted in Fig. 2 and the spectroscopic data are summarized in
Table 1. The small unmodified etherified 4-hydroxy-1,3-thiazoles
usually show an intense band in the region of 380 nm (see exper-
The crystal structure of D is depicted in Fig. 1 and a more
detailed look, selected bond lengths and angles in comparison with
the calculated values and additional crystallographic data are given
in the ESI (S1 and S2). The structure reveals the almost planar
arrangement of the dye. Two hydrogen bonds between the
carboxylic acid groups of two adjacent molecules lead to the
formation of a dimer. The distance between the two oxygens
forming the hydrogen bonds is 2.593(3) Å. The cyano group of the
2-cyanoacrylic acid acceptor unit is in cis-position to the sulfur of
the thiophene moiety. The NMR spectrum of D shows only one
signal for the ethylene hydrogen bound to C16. (ESI, S3). Therefore,
it can be considered that the presented isomer is exclusively
formed in the Knoevenagel condensation reaction. This is similar to
one infrequent literature example where the cyano group is also in
cis-position to the sulfur of the thiophene moiety [44]. As expected,
the conjugation pathway of the molecule shows an almost planar
geometry. The distance between two layers formed form the
imental section for 2, 3 and 6) due to p-p* transition of the thiazole
moiety. These lmax of absorption were successfully shifted bath-
ochromically after the formation of the electron-donating triaryl-
amine moieties for 3a and 3b, furthermore, after the introduction of
the electron accepting aldehyde and up to 505 nm for B after the
generation of the electron-accepting cyanoacrylic acid group due to
an intense intramolecular charge transfer (ICT) transition. The
absorption maximum of A-C is at moderate longer wavelengths
compared to very similar triarylamines using thiophenes and oli-
gothiophenes [46], thiophenes and 3,4-ethylenedioxythiophene
[47], and thiophenes and 3,4- thienothiophene [23] instead of 4-
methoxy-1,3-thiazoles units as a
p-spacer. This is promising in
terms of the light harvesting properties of the dyes presented. The
influence of the 4-methoxy group at the triarylamine moiety for B
compared to the tolyl group (A) resulted in a bathochromic shift of
dimers is only 3.5 Å, which is typical for strong
p-stacking inter-
actions between the molecules [45]. The torsion angles between
the three aromatic rings and the anchoring group are 4.47(2) (C20-
ꢁ
C8-C7-S2), 7.21(2) (N1-C5-C4-S1) and 8.99(2) (S1-C1-C16-C17).
This planarity allows an electron flow from the donor to the
acceptor group.
Fig. 2. UV/Vis spectra of the protonated and deprotonated dyes A-D in THF solution at
Scheme 2. Synthesis and structure of the 4-metoxyphenyl-based dye D.
room temperature.

R. Menzel et al. / Dyes and Pigments 94 (2012) 512e524
519
Table 1
Spectroscopic properties, and redox potentials of the protonated and deprotonated (dep) dyes A-D.
b
Dye
lAbs (nm) [
3
(102 Mꢀ1 cmꢀ1)]
lFl (nm)a
Stokes S. (cmꢀ1
)
s
(ns)c
F
(%)
4
E1/2 Ox (V)d
0.35, 0.59g
E1/2 red (V)d
E0e0 (eV)e
2.30
E (Sþ/S*) vs NHE (V)f
A
305 [220], 355 [133], 503 [247]
277 [400], 457 [295]
300 [230], 363 [145], 505 [289]
297 [280], 450 [460]
307 [175], 400 [170], 490 [320]
309 [210], 474 [360]
277 [130], 339 [100], 477 [400]
276 [150], 433 [390]
611
592
640
558
622
598
599
507
3500
5000
4200
4300
4300
4400
4300
3400
2.5
1.1
1.4
2.9
ꢀ1.95
e
ꢀ1.32
ꢀ1.52
ꢀ1.32
ꢀ1.02
A (dep)
B
B (dep)
C
C (dep)
D
D (dep)
e
< 1
1
0.20, 0.56g
ꢀ1.93
e
2.35
2.23
2.34
e
0.28, 0.68g
ꢀ1.99
e
54
0.69
e
ꢀ1.53
a
Excitation at the longest wavelength absorption.
Relative to maximum of the absorption.
b
c
d
e
f
Excitation at
l
¼ 650 nm.
E1/2 ¼ (EpaeEpc)/2 vs Fc/Fcþ.
Calc from the onset (10%) of the photoluminescence spectra.
The excited-state oxidation potential E (Sþ/S*) ¼ E1/2 (first oxidation) - E0-0 with Fc/Fcþ vs NHE ¼ 0.63 V.
Peaks are irreversible, E derived from differential pulse polarographic (DPP) measurements.
g
only 2 nm (0.01 eV), which is in accordance with the quantum
chemical calculations. In the energetic representations of the
HOMO, it can be seen that participation of the 4-methoxy groups is
very small and, therefore, the influence to the ICT is negligible (see
Fig. 5 dye B). Surprisingly, the extension of the conjugation length
for C did not lead to a bathochromic shift of the absorption
maximum. This is contrary to the literature examples mentioned
and can be attributed to an insufficient ICT transition, which is also
in accordance with the quantum chemical calculations (see Fig. 5
dye C). The bands in the UV-A region at 300e310 nm with
a diminished ICT character of the transition and, consequently, to
a hypsochromic shift of the absorption [49]. This is also in accor-
dance with the small change of the absorption maximum for C and
the lower ICT character of the transition in the visible part of the
spectra. Furthermore, after deprotonation the extinction coeffi-
cients for all triarylamine dyes increased (up to 46 000 Mꢀ1 cmꢀ1
for B), whereas for D, it slightly decreased. The triarylamine-based
dyes show moderate room temperature fluorescence in THF and
DMSO, but not in CHCl3 and MeOH. Their emission spectra in THF
solution are shown in Fig. 3.
extinction coefficients (
3
) from 17 500 to 23 000 Mꢀ1 cmꢀ1
* transitions in triarylamines and
The large Stokes Shift is typical for these D-p-A dyes. In accor-
correspond to a mixture of
pe
p
dance with the UV/Vis spectra, the emission is also hypsochromi-
cally shifted after deprotonation. The fluorescence lifetimes vary
from 1.1 to 2.9 ns for B and D, respectively. The dye emissions
showed the expected monoexponential decay curves, which are
presented in the ESI (S4). The values obtained are consistent with
ICT states (see ESI, S11). Accordingly, this band is missing in the
spectra of C, which only shows a weak UV absorption at 277 nm due
to the 4-methoxyphenyl ring. The extinction coefficients of the
absorption bands in the visible region are up to 40 000 Mꢀ1 cmꢀ1
,
which is characteristic for an ICT and 2½ times higher than those of
Ru(polypyridyl) complexes [48]. Deprotonation of the dyes was
carried out by adding the appropriate amount (1.1 equiv) of a tet-
rabutylammonium hydroxide (TBAOH) solution directly to the
cuvette, followed instantaneously by the measurement.
The longest wavelength absorptions for all dyes are hyp-
sochromically shifted (46 nm or 0.25 eV for A, to 16 nm or 0.09 eV
for C). The energy acceptor strength (electronegativity) of the
cyanoacrylic acid after deprotonation is lowered, which leads to
the measured fluorescence quantum yields (F) of the dyes. They are
very low for the triarylamines A-C (below 4%) and higher for the 4-
methoxyphenyl-based dye D (54%).
3.4. Electrochemical properties
In order to gain deeper insights into the electrochemical prop-
erties of the dyes, cyclic voltammetry (CV) and differential puls
polarography (DPP) measurements were applied. They were
Fig. 4. Cyclic voltammetry spectra of dyes A-D in THF; experimental conditions: 0.1 M
TBAPF6, c z 10ꢀ4 M, 25 ꢁC, scan rate 200 mV sꢀ1, RE: Ag/AgCl, WE: carbon, AE:
platinum.
Fig. 3. Emission spectra of the protonated and deprotonated dyes A-D in THF solution
at room temperature after excitation in the longest wavelength absorption band.

520
R. Menzel et al. / Dyes and Pigments 94 (2012) 512e524
Fig. 5. Frontier orbitals of the four dyes optimized at the CAM-B3LYP/6e31G(d,p) level of theory.
carried out in THF with 0.1 M TBAPF6 at 25 ꢁC with a scan rate of
200 mV sꢀ1, with Ag/AgCl as the reference, carbon as the working
and platinum as the auxiliary electrode. Ferrocene was added as the
internal standard after every measurement. All potentials discussed
here are given relative to the normal hydrogen electrode (NHE)
with Fc/Fcþ vs NHE ¼ 0.63 V [50]. The cyclic voltammetry spectra of
the compounds are depicted in Fig. 4 and data are summarized in
Table 1.
The dyes A-C show a reversible first oxidation wave, caused by
the oxidation of the triarylamine donor in the region from 0.98 to
0.83 V for A and B, respectively. This suggests that the first oxidized
state of these dyes is stable, which is one prerequisite for a good
long term stability of DSSCs [13]. The oxidation potential for B is
lowered by 0.15 V compared to A due to the introduction of the
electron-donating 4-methoxy group at the triarylamine core, which
facilitates the oxidation of the amine. Additional, compound C is
oxidized more easily than A after the incorporation of the thio-
phene unit in the conjugation pathway, leading to an increase of the
electron density of the aromatic linker which, consequently, lowers
the oxidation potential of the end-capped amine unit [51]. The
oxidation wave for compound D is superimposed by the oxidation
of the solvent and, therefore, it is not possible to assign if the
oxidation process is reversible or not.
E (Sþ/S*) was calculated from the first oxidation potential and the
zeroezero transition energy E0ꢀ0, which was determined from the
onset (10%) of the emission spectra (E (Sþ/S*) ¼ E1/2 Ox e E0ꢀ0). A
schematic representation of the energy levels is given in the ESI
(S5). The calculated potentials vary from ꢀ1.52 to ꢀ1.02 V for B and
D, respectively. They are sufficiently more negative (DE > 0.2 V)
[55] than the conducting band (CB) edge of the TiO2 (standard
potential ECB ¼ ꢀ0.5 V vs NHE) [56] and electron injection should be
possible. The regeneration of the oxidized dyes is also energetically
favored. The oxidation potential, the energy of the HOMO, lies in
the range of 1.32 V (D) to 0.98 (A), which is high enough for an
efficient reduction by the Iꢀ/Iꢀ3 redox couple (E ¼ 0.35 V vs NHE)
[57]. Thus, electron injection of the excited molecules and, subse-
quently, regeneration of the oxidized species is energetically
permitted. This allows the application of the dyes in DSSC.
3.5. Computational results
In order to gain more detailed insight into the structural and the
spectroscopic properties of the chromophores, density functional
theory (DFT) and its time-dependent version (TDDFT) was applied.
The four dyes were investigated in the protonated form and in the
trans-arrangement of the nitrile group to the thiophene. TDDFT
calculations on both the trans and the cis-isomer were performed
and details are summarized in the ESI. It was shown that both
isomers exhibit very similar photophysical properties. Therefore an
influence of the different isomers on the spectroscopic properties
can be neglected. The ground state equilibrium geometries of the
protonated species show similar structural features. The conju-
Nonetheless, due to the missing triarylamine moiety, the
oxidation occurs at a significant higher potential (1.32 V) compared
to the dyes A-C, but it could not be clarified where the oxidation
and radical formation takes place.
The first reduction peaks appear to be irreversible for all dyes.
They are very similar for the triarylamines (ꢀ1.30 for B to ꢀ1.33 V
for C) and shifted to a higher potential for compound D (ꢀ0.90 V). In
a water-free environment, it could be attributed to the reduction of
the double bond forming a radical which then undergoes dimer-
ization or hydrodimerization reactions (if traces of water are
present) [52,53]. This would explain the overall irreversible nature
of the process. The higher potential for D is in agreement with the
lower ICT character and a reduced dipole moment, which leads to
a less pronounced electron density at the cyanoacrylic acid group,
which makes it easier to reduce [54].
gated p-system from the bridge to the anchoring group is planar in
all dyes. The phenyl groups of the triphenylamine fragment (A-C)
are twisted out of planarity due to sterical hindrance. All dyes
exhibit a torsionated phenyl ring connected to the 4-methoxy-1,3-
thiazole fragment, with a dihedral angle independent of the length
of the bridge, as well as of the size of the donor system. It varies
only marginally from 18.3 (C) up to 20.4ꢁ (D). The analysis of the
harmonic vibrational modes corresponding to the torsion indicates
a shallow minimum in this normal coordinate. Wavenumbers of ca.
30 cmꢀ1 have been determined for this vibrational mode for all
In order to judge the possibility for efficient charge injection in
the conducting band of TiO2, the excited-state oxidation potential
compounds. For this reason,
a thermal distribution in this

R. Menzel et al. / Dyes and Pigments 94 (2012) 512e524
521
coordinate is to be expected. According to the low frequencies, the
HOMOs show a non-constructive overlap of the -systems from
the 4-methoxy-1,3-thiazole and the adjacent phenyl moiety
leading to a single bond in these positions.
The first absorption band of D exhibits the largest measured
molecular extinction coefficient, while the calculated oscillator
strength is considerably too small. This is caused by the more local
HOMO/LUMO transition for D, and the fact that CAM-B3LYP tends
to underestimate the oscillator strength of such systems.
p
The optimized geometries of the S1 states hold similarities to the
ground state equilibrium structures. The reason is the
p
e
p
* nature
The adiabatic emission spectra were obtained using the opti-
mized geometry of the first excited singlet state. The measured and
simulated fluorescence wavelengths are also collected in Table 2. A
comparison of the theoretical and experimental fluorescence
spectra, as well as the molecular orbitals contributing to the rele-
vant transitions, can be found in the ESI (S12 and S13). In agreement
with the configurations of the S1 in the ground state equilibrium
structure, the weight of the LUMO/HOMO transitions decreases
of the S1 and the partial occupation of the LUMOs. Therefore, the
excited state structures show only minor shifts in the bond lengths
and angles. However, the 4-methoxy-1,3-thiazole-phenyl frag-
ments are planarized as a result of the bonding character of the
LUMOs in these positions. The most important molecular orbitals
involved in the electronic excitations are shown in Fig. 5. The
absorption spectra of the four dyes were calculated using the
Franck-Condon (FC) geometry and the first ten singlet excited
states. Table 2 collects vertical and adiabatic excitations from the S1
state. The absorption energies to higher excited states, together
with an overlay of the experimental and calculated absorption
spectra and the molecular orbitals contributing to those states, are
summarized in the ESI (S6-S11). In all dyes the S1 is assigned to
with the enhancement of the p-system. Nevertheless, the weight is
significantly increased (A: 91, B: 89, C: 85, and D: 95%) compared to
the vertical excitations in the FC region. The emission wavelengths
obtained are in excellent agreement with the experimental data.
Only minor deviations of ꢀ0.02, ꢀ0.06 and 0.01 eV for A, B and D,
respectively, were generated, while C shows a slight underesti-
mation of ꢀ0.15 eV.
a
p
ep
* excitation, with the main transition occurring from the
HOMO to the LUMO. The main contribution to the HOMO of A, B
and C is due to the -systems of the donor and thiazole fragments,
p
3.6. Photovoltaic devices and solar cell performance
while in the case of D, the donor and the entire bridge contribute to
the HOMO. The LUMO for all dyes is spread over the anchoring
group and the bridge. Hence, an ICT from the donor to the acceptor
occurs.
In order to test the dyes for a possible application as sensitizers,
DSSCs with nanocrystalline TiO2 (70% anatase, 30% rutile) were
prepared. The exact protocol is given in the experimental part and
the results, together with the standard N3 (cis-bis(isothiocyanato)
bis(2,20-bipyridyl-4,40-dicarboxylato)-ruthenium(II)), are summa-
rized in Table 3. The incident photon-to-current conversion effi-
ciencies (IPCE) spectra obtained are shown in Fig. 6. Only moderate
IPCEs were achieved, with values varying from 25 (D) to 13% (C) for
the absorption maximum of the dyes, though several different
preparation conditions (different solvents, dye concentrations)
were utilized. IPCEs of up to 80% are usually expected for these
types of compounds [12,23]. Therefore, it can be assumed that the
electron injection efficiency in the electron acceptor states of the
TiO2 is low. The amount of dyes adsorbed on the TiO2 was deter-
mined by measuring the absorption spectra of a solution of the
corresponding dye before and after soaking of the TiO2 plate. After
correlation with the measured extinction coefficient, the adsorbed
amount was calculated to be 5.0 ꢂ 10ꢀ8 (A and B), 6.6 ꢂ 10ꢀ8 (D)
and 4.0 ꢂ 10ꢀ8 mol cmꢀ2 for the largest sensitizer C. The values
correspond very well with the size of the molecules and are in good
agreement with literature examples and cannot account for the low
IPCEs [58,59]. The absorption spectra of the adsorbed dyes are
depicted in the ESI (S18). They were similar to the absorption
spectra of the dyes in solution. The main difference is the broad-
ening of the longest wavelength ICT band (absorption from approx.
370e700 nm compared to 425e600 nm for the dyes in solution)
The expansion of the aromatic
p-system tends to increase the
weight of further configurations. The HOMO/LUMO transition for D
has a weight of 88%, while the implementation of the N,N-di-4-tolyl
(A) and N,N-bis(4-methoxyphenyl) (B) donor groups decreases the
weight to 72 and 71%, respectively. The weight is decreased further
to 54% by the introduction of an additional thiophene moiety (C). As
a consequence, the S1 of C features a less pronounced ICT leading to
an increase of the excitation energy and, consequently, to a hyp-
sochromic shift of the lmax of the absorption compared to A and B.
Additionally, D shows a blueshift of the S1. The reason is the more
local p-p* transition between the frontier orbitals. The wavelength
dependency of the first absorption band is in good agreement to the
experiment. In general, the excitation energies of the S1 are
marginally overestimated (A: 0.12, B: 0.09, C: 0.11, and D: 0.16 eV),
showing that the CAM-B3LYP functional is very well suited to
describe ICT in these dyes. By comparing the oscillator strengths (f)
with the experimentally obtained molecular extinction coefficients,
a general trend of increasing absorption is obvious for A, B and C.
Table 2
Calculated absorption/emission (fluorescence) properties to/from the first excited
state S1. Main contributions to the wavefunction (weight), vertical (absorption) or
adiabatic (emission) excitation energies (
(f) and deviations from experimental values (
D
Ee in eV and nm), oscillator strengths
D
EExp). H ¼ HOMO, L ¼ LUMO.
Transition
Weight (%)
D
Ee
f
DEExp (eV)
Table 3
(eV)
(nm)
Photovoltaic performance of DSSCs based on dyes A-D without and with the addi-
Absorption
tion of DCA (10 mM) as a co-adsorbent.
A
B
C
H / L
H-1 / L
H / L
H-1 / L
H / L
H-1 / L
H / L þ 1
H / L
72
20
71
20
54
26
12
88
2.58
2.54
2.64
479
487
470
1.511
1.502
1.958
0.12
0.09
0.11
Dye
Jsc (mA cmꢀ2
)
Voc (mV)
FF
h (%)
Without DCA
A
B
C
D
N3
1.81
1.42
1.37
1.03
11.73
519
558
524
468
690
0.46
0.51
0.47
0.52
0.51
0.44
0.41
0.36
0.25
4.1
D
2.75
451
1.364
0.16
Emission
With DCA
A
B
C
D
H ) L
H)L
H ) L
H ) L
91
89
85
95
1.91
1.88
1.85
2.12
650
658
669
586
e
e
e
e
ꢀ0.02
ꢀ0.06
ꢀ0.15
0.01
A
B
C
D
4.49
3.55
3.27
2.23
612
611
616
511
0.62
0.50
0.51
0.52
1.70
1.08
1.02
0.61

522
R. Menzel et al. / Dyes and Pigments 94 (2012) 512e524
Fig. 6. Spectra of the incident photon-to-current conversion efficiencies obtained for
nanocrystalline TiO2 solar cells sensitized by the dyes A-D without (d) and with
(d-d) co-adsorption of DCA (10 mM).
Fig. 7. Photocurrent-voltage characteristics for A-D sensitized DSSCs under illumina-
tion of simulated solar light (100 mW cmꢀ2, 1.5 AM global) without (d) and with
(d-d) co-adsorption of DCA (10 mM).
due to the absorption on the TiO2 surface. This ensures a sufficient
light harvesting over a broad wavelength range. Two reasons can be
This equals a remarkable increase of a factor 4 of the efficiency in
the case of dye A. Keeping in mind that the efficiency of a DSSC
prepared from the standard N3 with the preparation technique
described was 4.1%, the best performance DSSC using A as sensi-
tizer achieved reasonable 42% of this value. The trend of the effi-
ciencies of the cells equals the trend without co-adsorbed DCA. For
A and B the efficiency is very similar, whereas for C it is decreased.
Most likely electron recombination is enhanced for compound C
due to the extended bridge and, therefore, electron injection in the
CB of TiO2 is less favored leading to lower IPCE values [69]. Most
likely due to the missing strong electron donor and the low ICT
character of the first excited state, DSSCs sensitized with D showed
the lowest performance.
assumed for this broadening: Firstly, the energy of the
p* level is
lowered due to interaction of the carboxylic acid with the Ti4þ ions
on the TiO2 surface, which consequently leads to a bathochromic
shift of the absorption spectra [60], and (more important) secondly,
unfavorable dye aggregation and excimer formation (J-aggregates)
would lead to this behavior [61,62]. The DSSCs prepared only
showed a low short circuit photocurrent density JSC (Fig. 7) from
1.81 to 1.03 mA cmꢀ2 for A and D, respectively. Because of the low
JSC, the maximum power outputs (MPP ¼ jmp $ Vmp) and, conse-
quently, the fill factors (FF) were small. The resulting solar energy-
to-electricity conversion yield (h) is low and follows the trend
A > B > C > D with values from 0.44 to 0.25%.
In summary, the dyes tend to aggregate on the surface of TiO2.
This effect was more pronounced for the triarylamine-based dyes
than for D. Further improvements to overcome this aggregation
could be the introduction of long alkyl chains [70,71], for example,
via the alkylation reaction or by bridging two dye molecules [72].
Also the introduction of the strong electron-withdrawing benzo-
thiadiazole unit in the conjugation pathway to further shift the
absorption maximum will be considered in the construction of
novel dyes [73]. Also photophysical experiments, like electro-
chemical impedance spectroscopy (EIS) and electron lifetime
measurement of the adsorbed dyes, have to be carried out to
account the photovoltaic properties further [74,75].
Two possible reasons for the weak solar cell performance can
be: (i) an electron accepting property of the 4-methoxy-1,3-
thiazole, which can effect an efficient electron injection in the CB
of TiO2 as observed for anthraquinone-based dyes [63] or (ii) the
formation of aggregates leading to a loss of excitation energy
through intermolecular processes [64,65]. The first fact is sup-
ported from the TDDFT calculations. The LUMO is not only spread
over the anchoring group, but also includes contribution from the
thiophene and also the thiazole moieties, leading to an insufficient
electron injection. In order to verify the second reason as crucial for
the low performance, DSSCs were prepared using deoxycholic acid
(DCA) as a co-adsorbent. It has been proven several times that co-
adsorption of either DCA or chenodeoxycholic acid (CDCA) can
efficiently suppress dye aggregation [61,66,67]. Therefore, DCA
(10 mM in THF) was chosen for the preparation of more elaborated
DSSCs (Table 3). The IPCE values (Fig. 6) were improved by about
the factor two (from 14 to 30% for A and up to 35% for D) at the
longest wavelength absorption band. Consequently, the short
circuit photocurrent densities and the open-circuit voltages were
also significantly improved (Fig. 7). The FF was increased by 35% for
A (FF ¼ 0.62), whereas they remained almost unchanged for the
other dyes. A reason for the poor FFs of all dyes can be a poor
performance of the counter electrode, which was made of TiO2 P25.
This will affect the IeV characteristics of the DSSC and, conse-
quently, lowering the FF. For further experiments mesoporous
anatase TiO2 beads will be used to improve the cell performance as
described in literature [68]. The overall efficiencies of the optimized
DSSCs obtained varied from 0.61 to 1.7% for D and A, respectively.
4. Conclusion
A series of donor-p-acceptor dyes were successfully synthe-
sized. Instead of the commonly used thiophene derivatives, an
unconventional chromophore based on a 4-hydroxy-1,3-thiazole
was used in the conjugation pathway of the dyes, to efficiently
shift the absorption maximum toward longer wavelengths
compared to similar triarylamines.
A
double N-arylation
HartwigeBuchwald reaction, rarely described in literature, was
successful applied in the synthetic route for the triarylamine-based
dyes. All dyes have been fully characterized including a series of
spectroscopic and theoretical methods. The quantum chemical
TDDFT calculations used proofed to be well suited to reproduce the
experimental values as supported by the X-ray structure of the D
and the absorption and emission characteristics.

R. Menzel et al. / Dyes and Pigments 94 (2012) 512e524
523
[18] Zeng W, Cao Y, Bai Y, Wang Y, Shi Y, Zhang M, et al. Efficient dye-sensitized
solar cells with an organic photosensitizer featuring orderly conjugated eth-
ylenedioxythiophene and dithienosilole blocks. Chem Mater 2010;22(5):
1915e25.
[19] Yella A, Lee H-W, Tsao HN, Yi C, Chandiran AK, Nazeeruddin MK, et al.
Porphyrin-sensitized solar cells with cobalt (II/III)ebased redox electrolyte
exceed 12 Percent efficiency. Science 2011;334(6056):629e34.
[20] Tian Z, Huang M, Zhao B, Huang H, Feng X, Nie Y, et al. Low-cost dyes based on
methylthiophene for high-performance dye-sensitized solar cells. Dyes Pigm
2010;87(3):181e7.
Electrochemical measurements and DFT calculations approved
a possible application of the synthesized dyes as sensitizers in
DSSCs. The efficiencies obtained of DCCSs without co-adsorption of
DCA proved to be very low. Nevertheless, the performance was
significantly increased after the addition of DCA. This demonstrated
the distinct tendency of the dyes to form aggregates on the surface
of nanocrystalline TiO2. The best performance solar cell sensitized
with dye A achieved 42% of the performance of a DSSC prepared
with the standard N3 dye. In further investigations, the NLO
properties of this class of molecules will be reviewed. For molecules
of type D this property has already been proven in experiment.
Furthermore, the use as fluorescent probes, bearing a reactive eOH
group instead of the methoxy ether to interact with biological
systems, will be the subject of further research.
[21] Ning Z, Tian H. Triarylamine: a promising core unit for efficient photovoltaic
materials. Chem Commun 2009;37:5483e95.
[22] Tian H, Yang X, Chen R, Zhang R, Hagfeldt A, Sun L. Effect of different dye baths
and dye-structures on the performance of dye-sensitized solar cells based on
triphenylamine dyes. J Phys Chem C 2008;112(29):11023e33.
[23] Li G, Jiang K-J, Li Y-F, Li S-L, Yang L-M. Efficient structural modification of
triphenylamine-based organic dyes for dye-sensitized solar cells. J Phys Chem
C 2008;112(30):11591e9.
[24] Liang M, Xu W, Cai F, Chen P, Peng B, Chen J, et al. New triphenylamine-based
organic dyes for efficient dye-sensitized solar cells. J Phys Chem C 2007;
111(11):4465e72.
[25] Hagberg DP, Marinado T, Karlsson KM, Nonomura K, Qin P, Boschloo G, et al.
Tuning the HOMO and LUMO energy levels of organic chromophores for dye
sensitized solar cells. J Org Chem 2007;72(25):9550e6.
[26] Hwang S, Lee JH, Park C, Lee H, Kim C, Park C, et al. A highly efficient organic
sensitizer for dye-sensitized solar cells. Chem Commun 2007;46:4887e9.
[27] Hara K, Kurashige M, Dan-Oh Y, Kasada C, Shinpo A, Suga S, et al. Design of
new coumarin dyes having thiophene moieties for highly efficient organic-
dye-sensitized solar cells. New J Chem 2003;27(5):783e5.
Acknowledgment
The authors thank the Thuringian Ministry for Education,
Science and Culture (grant #B514-09049, Photonische Mizellen
(PhotoMIC)) for financial support. Johann Schäfer is also gratefully
acknowledged for measuring the fluorescence lifetimes and the
quantum yields of the dyes.
[28] Chen C-H, Hsu Y-C, Chou H-H, Thomas KRJ, Lin JT, Hsu C-P. Dipolar
compounds containing fluorene and a heteroaromatic ring as the conjugating
bridge for high-performance dye-sensitized solar cells. Chem Eur J 2010;
16(10):3184e93.
Appendix. Supporting information
[29] He J, Wu W, Hua J, Jiang Y, Qu S, Li J, et al. Bithiazole-bridged dyes for dye-
sensitized solar cells with high open circuit voltage performance. J Mater
Chem 2011;21(16):6054e62.
Supplementary data related to this article can be found online at
10.1016/j.dyepig.2012.02.014.
[30] Ooyama Y, Harima Y. Molecular designs and Syntheses of organic dyes for
dye-sensitized solar cells. Eur J Org Chem 200 2009;18:2903e34.
[31] Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, et al.
Gaussian 09, Revision A.02. Wallingford CT2009.
References
[32] Tomasi J, Mennucci B, Cammi R. Quantum mechanical continuum solvation
models. Chem Rev 2005;105(8):2999e3093.
[1] Liebscher J. Methoden Org Chem (Houben-Weyl). 4 ed. Stuttgart-New York:
Georg Thieme; 1993.
[33] Yanai T, Tew DP, Handy NC. A new hybrid exchange-correlation functional
using the Coulomb-attenuating method (CAM-B3LYP). Chem Phys Lett 2004;
393(1e3):51e7.
[34] Hariharan PC, Pople JA. The influence of polarization functions on molecular
orbital hydrogenation energies. Theor Chem Acc 1973;28(3):213e22.
[35] Kasha M. Characterization of electronic transitions in complex molecules.
Discuss Faraday Soc 1950;9:14e9.
[36] Fors BP, Krattiger P, Strieter E, Buchwald SL. Water-mediated catalyst pre-
activation: an efficient protocol for CꢀN cross-coupling reactions. Org Lett
2008;10(16):3505e8.
[37] Surry DS, Buchwald SL. Dialkylbiaryl phosphines in Pd-catalyzed amination:
a user’s guide. Chem Sci 2011;2(1):27e50.
[2] Menzel R, Täuscher E, Weiß D, Beckert R, Görls H. The combination of 4-
hydroxythiazoles with azaheterocycles: efficient bidentate ligands for novel
ruthenium complexes. Z Anorg Allg Chem 2010;636(7):1380e5.
[3] Menzel R, Breul A, Pietsch C, Schäfer J, Friebe C, Täuscher E, et al. Blue-emitting
polymers based on 4-hydroxythiazoles incorporated in a methacrylate back-
bone. Macromol Chem Phys 2011;212(8):840e8.
[4] Täuscher E, Weiß D, Beckert R, Fabian J, Assumpção A, Görls H. Classical
heterocycles with surprising properties: the 4-hydroxy-1,3-thiazoles. Tetra-
hedron Lett 2011;52(18):2292e4.
[5] Täuscher E, Weiß D, Beckert R, Görls H. Synthesis and characterization of new
4-hydroxy-1,3-thiazoles. Synthesis-Stuttgart 2010;10:1603e8.
[6] Kim E, Park SB. Chemistry as a prism: a review of light-emitting materials
having tunable emission wavelengths. Chem Asian J 2009;4(11):1646e58.
[7] Otsubo T, Aso Y, Takimiya K. Functional oligothiophenes as advanced
molecular electronic materials. J Mater Chem 2002;12(9):2565e75.
[8] Jose J, Burgess K. Benzophenoxazine-based fluorescent dyes for labeling
biomolecules. Tetrahedron 2006;62(48):11021e37.
[38] Harris MC, Buchwald SL. One-pot synthesis of unsymmetrical triarylamines
from aniline precursors. J Org Chem 2000;65(17):5327e33.
[39] Nozaki K, Takahashi K, Nakano K, Hiyama T, Tang H-Z, Fujiki M, et al. The
double N-arylation of primary amines: toward multisubstituted carbazoles
with unique optical properties. Angew Chem 2003;115(18):2097e9.
[40] Zhou Y, Verkade JG. Highly efficient ligands for the palladium-assisted double
N-arylation of primary amines for one-sep construction of carbazoles. Adv
Synth Catal 2010;352(4):616e20.
[9] Reinhardt BA, Brott LL, Clarson SJ, Dillard AG, Bhatt JC, Kannan R, et al. Highly
active two-photon dyes: design, synthesis, and characterization toward
application. Chem Mater 1998;10(7):1863e74.
[41] Newman SG, Lautens M. The role of reversible oxidative addition in selective
palladium(0)-catalyzed intramolecular cross-couplings of polyhalogenated
substrates: synthesis of brominated Indoles. J Am Chem Soc 2010;132(33):
11416e7.
[10] Oliva MM, Casado J, Raposo MMM, Fonseca AMC, Hartmann H, Hernández V,
et al. Structure-property relationships in push-pull amino/cyanovinyl end-
capped oligothiophenes: quantum chemical and experimental studies. J Org
Chem 2006;71(20):7509e20.
[42] Seo ET, Nelson RF, Fritsch JM, Marcoux LS, Leedy DW, Adams RN. Anodic
oxidation pathways of aromatic amines. electrochemical and electron para-
magnetic resonance studies. J Am Chem Soc 1966;88(15):3498e503.
[43] Zhao H, Tanjutco C, Thayumanavan S. Design and synthesis of stable triaryl-
amines for hole-transport applications. Tetrahedron Lett 2001;42(27):4421e4.
[44] Olivier C, Sauvage F, Ducasse L, Castet F, Grätzel M, Toupance T. Fine-tuning of
triarylamine-based photosensitizers for dye-sensitized solar cells. Chem-
SusChem. 2011;4(6):731e6.
[11] Duan L, Qiao J, Sun YD, Qiu Y. Strategies to design bipolar small molecules for
OLEDs: donor-acceptor structure and non-donor-acceptor structure. Adv
Mater 2011;23(9):1137e44.
[12] 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(14):2474e99.
[13] Hagfeldt A, Boschloo G, Sun L, Kloo L, Pettersson H. Dye-sensitized solar cells.
Chem Rev 2010;110(11):6595e663.
[45] Coates GW, Dunn AR, Henling LM, Dougherty DA, Grubbs RH. Spezifische
phenyl-perfluorphenyl-stapelwechselwirkungen: eine neue Strategie zum
Aufbau supramolekularer Strukturen. Angew Chem 1997;109(3):290e3.
[46] Thomas KRJ, Hsu Y-C, Lin JT, Lee K-M, Ho K-C, Lai C-H, et al. 2,3-Disubstituted
thiophene-based organic dyes for solar cells. Chem Mater 2008;20(5):
1830e40.
[14] Li L, Yang X, Gao J, Tian H, Zhao J, Hagfeldt A, et al. Highly efficient CdS
quantum dot-sensitized solar cells based on a modified polysulfide electro-
lyte. J Am Chem Soc 2011;133(22):8458e60.
[15] Wu TY, Tsao MH, Chen FL, Su SG, Chang CW, Wang HP, et al. Synthesis and
characterization of organic dyes containing various donors and acceptors. Int J
Mol Sci 2010;11(1):329e53.
[47] Li G, Jiang K-J, Bao P, Li Y-F, Li S-L, Yang L-M. Molecular design of triarylamine-
based organic dyes for efficient dye-sensitized solar cells. New J Chem 2009;
33(4):868e76.
[48] Nazeeruddin MK, Kay A, Rodicio I, Humphry-Baker R, Mueller E, Liska P, et al.
Conversion of light to electricity by cis-X2bis(2,2’-bipyridyl-4,4’-
[16] O’Regan B, Grätzel M. A low-cost, high-efficiency solar cell based on dye-
sensitized colloidal TiO2 films. Nature 1991;353(6346):737e40.
[17] Chiba Y, Islam A, Watanabe Y, Komiya R, Koide N, Han L. Dye-sensitized solar
cells with conversion efficiency of 11.1%. Jpn
45(24e28):L638e40.
J Appl Phys Part 2 2006;

524
R. Menzel et al. / Dyes and Pigments 94 (2012) 512e524
dicarboxylate)ruthenium(II) charge-transfer sensitizers (X ¼ Cl-, Br-, I-, CN-,
[62] Pastore M, Angelis F. Aggregation of organic dyes on TiO2 in dye-sensitized
solar cells models: an ab initio investigation. ACS Nano 2009;4(1):556e62.
[63] Li C, Yang X, Chen R, Pan J, Tian H, Zhu H, et al. Anthraquinone dyes as
photosensitizers for dye-sensitized solar cells. Sol Energy Mater Sol Cells
2007;91(19):1863e71.
[64] Lee J-K, Yang M. Progress in light harvesting and charge injection of dye-
sensitized solar cells. Mater Sci Eng B 2011;176(15):1142e60.
[65] Hara K, Dan-Oh Y, Kasada C, Ohga Y, Shinpo A, Suga S, et al. Effect of additives
on the photovoltaic performance of coumarin-dye-sensitized nanocrystalline
TiO2 solar cells. Langmuir 2004;20(10):4205e10.
[66] Chen R, Yang X, Tian H, Wang X, Hagfeldt A, Sun L. Effect of tetrahy-
droquinoline dyes structure on the performance of organic dye-sensitized
solar cells. Chem Mater 2007;19(16):4007e15.
[67] Xu W, Pei J, Shi J, Peng S, Chen J. Influence of acceptor moiety in
triphenylamine-based dyes on the properties of dye-sensitized solar cells.
J Power Sources 2008;183(2):792e8.
and SCN-) on nanocrystalline titanium dioxide electrodes. J Am Chem Soc
1993;115(14):6382e90.
[49] Lee JY, Kim KS, Mhin BJ. Intramolecular charge transfer of pi-conjugated push-
pull systems in terms of polarizability and electronegativity. J Chem Phys
2001;115(20):9484e9.
[50] Pavlishchuk VV, Addison AW. Conversion constants for redox potentials
measured versus different reference electrodes in acetonitrile solutions at
25 ꢁC. Inorg Chim Acta 2000;298(1):97e102.
[51] Thomas KRJ, Singh P, Baheti A, Hsu YC, Ho K-C, Lin JT. Electro-optical prop-
erties of new anthracene based organic dyes for dye-sensitized solar cells.
Dyes Pigm 2011;91(1):33e43.
[52] Zhdanov SI, Feoktistov LG. The reductive dimerization of acrylonitrile on the
mercury cathode. Russ Chem Bull 1963;12(1):45e8.
[53] Baizer MM. Electrolytic reductive coupling. IV.1 mixed couplings among
derivatives of a,b-unsaturated acids. J Org Chem 1964;29(7):1670e3.
[54] Dietz F, Olbrich G, Karabunarliev S, Tyutyulkov N. Photoswitching of dipole
moments, charge-transfer and spectroscopic properties. Chem Phys Lett 2003;
379(1e2):11e9.
[55] Hagfeldt A, Grätzel M. Light-induced redox reactions in nanocrystalline
systems. Chem Rev 1995;95(1):49e68.
[68] Sauvage Fdr, Chen D, Comte P, Huang F, Heiniger L-P, Cheng Y-B, et al. Dye-
sensitized solar cells employing a single film of mesoporous TiO2 beads ach-
ieve power conversion efficiencies over 10%. ACS Nano 2010;4(8):4420e5.
[69] Marinado T, Nonomura K, Nissfolk J, Karlsson MK, Hagberg DP, Sun L, et al.
How the nature of triphenylamine-polyene dyes in dye-sensitized solar cells
affects the open-circuit voltage and electron lifetimes. Langmuir 2009;26(4):
2592e8.
[56] Angelis F, Fantacci S, Selloni A. Alignment of the dye’s molecular levels with
the TiO2 band edges in dye-sensitized solar cells:
Nanotechnology 2008;19(42):424002.
a DFTeTDDFT study.
[70] Shen P, Liu Y, Huang X, Zhao B, Xiang N, Fei J, et al. Efficient triphenylamine
[57] Boschloo G, Hagfeldt A. Characteristics of the iodide/triiodide redox mediator
in dye-sensitized solar cells. Acc Chem Res 2009;42(11):1819e26.
[58] Liang M, Zong X-P, Han H-Y, Chen C, Sun Z, Xue S. An efficient dye-sensitized
solar cell based on a functionalized-triarylamine dye. Mater Lett 2011;65(9):
1331e3.
[59] Stergiopoulos T, Ghicov A, Likodimos V, Tsoukleris DS, Kunze J, Schmuki P,
et al. Dye-sensitized solar cells based on thick highly ordered TiO2 nanotubes
produced by controlled anodic oxidation in non-aqueous electrolytic media.
Nanotechnology 2008;19(23):235602.
[60] Wang Z-S, Hara K, Dan-Oh Y, Kasada C, Shinpo A, Suga S, et al. Photophysical
and (photo)electrochemical properties of a coumarin dye. J Phys Chem B
2005;109(9):3907e14.
[61] Wang Z-S, Cui Y, Dan-Oh Y, Kasada C, Shinpo A, Hara K. Thiophene-func-
tionalized coumarin dye for efficient dye-sensitized solar cells: electron
lifetime improved by coadsorption of deoxycholic acid. J Phys Chem C 2007;
111(19):7224e30.
dyes for solar cells: effects of alkyl-substituents and p-conjugated thiophene
unit. Dyes Pigm 2009;83(2):187e97.
[71] Ning Z, Fu Y, Tian H. Improvement of dye-sensitized solar cells: what we
know and what we need to know. Energy Environ Sci 2010;3(9).
[72] Sirohi R, Kim DH, Yu S-C, Lee SH. Novel di-anchoring dye for DSSC by bridging
of two mono anchoring dye molecules: a conformational approach to reduce
aggregation. Dyes Pigm 2012;92(3):1132e7.
[73] Zhu W, Wu Y, Wang S, Li W, Li X, Chen J, et al. Organic D-A-p-A solar cell
sensitizers with improved stability and spectral response. Adv Funct Mater
2011;21(4):756e63.
[74] Wang C, Li J, Cai S, Ning Z, Zhao D, Zhang Q, et al. Performance improvement
of dye-sensitizing solar cell by semi-rigid triarylamine-based donors. Dyes
Pigm 2012;94(1):40e8.
[75] Ko S-B, Cho A-N, Kim M-J, Lee C-R, Park N-G. Alkyloxy substituted organic
dyes for high voltage dye-sensitized solar cell: effect of alkyloxy chain length
on open-circuit voltage. Dyes Pigm 2012;94(1):88e98.