Synthesis and properties of carboxyl-substituted coumarin–phthalocyanine dyad dyes: effect of linker unit

Article abstract of DOI:10.1007/s11164-016-2792-4

Two novel coumarin-substituted phthalocyanine dyes (CAPc and CSPc) with different linking units, using carboxylic groups as electron acceptor and injector, were designed, synthesized, and applied as panchromatic organic sensitizers in I^?/I

Full text of DOI:10.1007/s11164-016-2792-4

Res Chem Intermed
DOI 10.1007/s11164-016-2792-4
Synthesis and properties of carboxyl-substituted
coumarin–phthalocyanine dyad dyes: effect of linker
unit
Xuejun Zhang1
Chaoqiang Liao1 Li Zou1
Xilin Pei1
•
•
•
Received: 13 July 2016 / Accepted: 19 October 2016
Ó Springer Science+Business Media Dordrecht 2016
Abstract Two novel coumarin-substituted phthalocyanine dyes (CAPc and CSPc)
with different linking units, using carboxylic groups as electron acceptor and
injector, were designed, synthesized, and applied as panchromatic organic sensi-
tizers in I-/I3--based dye-sensitized solar cells (DSSCs). The asymmetric coumarin–
phthalocyanine dyad analogue (OAPc) and the symmetric coumarin-phthalocyanine
dyad analogue (COPc) were also synthesized for comparison. The structure–prop-
erty relationship for these dyes was investigated by absorption spectroscopy, cyclic
voltammetry, and thermogravimetric analysis. The ultraviolet–visible (UV–Vis)
absorption was dramatically affected by the nature of the linker. Notably, modu-
lation of the p-conjugated coumarin–phthalocyanine dyad CAPc and CSPc resulted
in a large red-shift (Q-band above 710 nm); in particular, the dyad CAPc showed
remarkably intense absorption in the spectral window of 450–650 nm. Electro-
chemical data for the dyes indicated that the azo double-bond-linked coumarin–
phthalocyanine dyad dyes possessed relatively low-lying EHOMO values, which may
be beneficial to suppress electron back-transfer from the conduction band of TiO2 to
oxidized dyes due to facile regeneration of the oxidized dye by the electrolyte.
Thermogravimetric analysis showed that the three coumarin–phthalocyanine dyad
sensitizers were stable above 200 °C. When the dyads were applied as light-har-
vesting sensitizers in dye-sensitized solar cells, the cell sensitized by OAPc showed
the best power conversion efficiency of 2.4%.
Keywords Coumarin Á Phthalocyanines Á Linker Á Absorption spectrum Á
Electrochemistry Á Dye-sensitized solar cells
&
Xuejun Zhang
zhangxuejun@nuc.edu.cn
1
College of Science, North University of China, Taiyuan 030051, Shanxi, China
123

X. Zhang et al.
Introduction
Dye-sensitized solar cells (DSSCs) based on dye-sensitized nanocrystalline TiO2
photoelectrodes with reduced manufacturing cost and facile processing have
potential to compete with more mature photovoltaic solar cells based on Si in the
future [1, 2]. Therefore, great efforts have been made to develop such low-cost solar
cells over the last few years. Satisfactory conversion efficiencies have been
achieved by using many organic or metalorganic molecules such as ruthenium
[3, 4], coumarins [5, 6], porphyrins [7, 8], phenothiazine [9], phthalocyanines
[10–12], and other small-molecule sensitizers [13–15]. The performance of DSSCs
is mainly affected by the properties of the sensitizer [16], especially its absorption
spectrum and energy levels.
Though organic dyes exhibit high molar extinction coefficient, they have narrow
visible-range absorption spectra [17, 18]. To improve the light-harvesting potential
of sensitizers, various notable attempts to achieve panchromatic absorption have
been made, including: (1) cosensitization with multiple, complementarily absorbing
dyes [8, 19], (2) integration of chromophores and auxiliary auxochrome (donors or
acceptors) in one dye configuration [6, 11, 20], and (3) extending the conjugation
structure of the dye molecule [6, 15, 17, 18]. On the other hand, intramolecular
charge transfer from donor to acceptor has also been demonstrated to broaden the
longer-wavelength absorption band to achieve panchromatic absorption and high
photoelectron conversion efficiency [12, 21].
Phthalocyanines can serve as a near-infrared (IR)/IR motif in DSSC sensitizers
owing to their intense absorption in the Q-band (around 700 nm) [10–12].
Furthermore, the intensity and region of light absorption by phthalocyanine
derivatives can be enhanced and broadened by decorating the phthalocyanine core
with complementary chromophores. Ren et al. [22, 23] reported phthalocyaninato
zinc(II) substituted with tetrakis- or octabenzothiadiazole moieties, linked with
alkynic bond or oxygen, achieving broad absorption covering the range of
300–900 nm. The superior red-shift and broadened absorption bands of these
phthalocyanines arise due to their efficient conjugation structure and electronic
interaction between the chromophores and phthalocyanine core. Yamamoto et al.
[20] and Kc et al. [24] reported covalently linked Zn phthalocyanine–Zn porphyrin
dyads with different linkages. These dyes combined the characteristic absorptions of
the phthalocyanine and porphyrin.
Coumarin dyes are attractive due to their photoresponse at around 450 nm,
excellent solubility, and fast electron injection [25]. When coumarin derivatives are
introduced into a phthalocyanine core via a thioether bond, oxygen, alkoxy or
phenoxy bridge, the absorption of the resulting coumarin-substituted phthalocyanine
compounds between 450 and 600 nm is weak, despite the increased absorption
range and solubility [26–29]. In previous study [30], we reported that integration of
azobenzene on phthalocyanine was useful to increase the light-harvesting ability in
the spectral window of 400–600 nm. As a result, dye-sensitized solar cells
fabricated using such dyes including naphthalenyl showed improved efficiency
compared with the dye including benzene. Also, the azobenzene branch provides the
123

Synthesis and properties of carboxyl-substituted…
opportunity to link other chromophores with various functional properties. Other
research groups have demonstrated use of azobenzene or imino as an auxiliary
chromophore, which was found to extend the absorption maximum and increase the
molar extinction coefficient [30–33]. However, to the best of the authors’
knowledge, coumarin–phthalocyanine dyad dyes covalently linked with oxygen,
azo, or imino have not yet been reported for use in DSSCs. We believed that such
covalently linked structures could be exploited for construction of coumarin–
phthalocyanine dyads featuring this molecular configuration (Fig. 1). We varied the
linker (oxygen, azo, and imino) to study its effects on the optical, aggregation,
electrochemical, thermal, and photovoltaic properties.
Experimental
General methods
The scaffolds required to synthesize the target dyes, viz. 7-hydroxycoumarin-3-
carboxylic acid, coumarin-3-carboxylic acid, 7-hydroxy-4-methylcoumarin, and
7-hydroxy-4-methylcoumarin-8-carbaldehyde, were obtained by Knoevenagel [28],
Pechmann [34], and Duff [35] reactions according to literature procedures.
4-Methyl-7-(3,4-dicyanophenoxy)coumarin (4) was synthesized according to liter-
ature [26], then COPc was prepared according to literature method with minor
modifications [27]. All other reagents were procured from commercial sources and
used without purification. Solvents used for absorption spectrum and cyclic
voltammetry measurements were distilled over suitable dehydrating agents accord-
ing to standard methods. Column chromatography was performed with silica gel
(COPc, CAPc, and OAPc) or activated alumina (CSPc).
Infrared (IR) spectra were recorded on a Shimadzu 4800S Fourier-transform
infrared spectrophotometer in the range from 4000 to 450 cm-1 from KBr pellets.
1
The structures of the target compounds were confirmed by H nuclear magnetic
resonance (NMR) spectra recorded on a Bruker AVANCE III 400-MHz spectrom-
eter in dimethylsulfoxide (DMSO)-d6, with chemical shifts reported relative to
tetramethylsilane (TMS) as internal standard. Matrix-assisted laser desorption/
ionization (MALDI) time-of-flight (TOF) mass-spectrometric measurements were
performed on a Bruker Biflex III. Optical spectra in the UV–Vis region were
Fig. 1 Structures of carboxyl coumarin–phthalocyanine dyad dyes
123

X. Zhang et al.
recorded on a Techcomp 2300 spectrophotometer using 1-cm-path-length cuvettes
at room temperature. Different concentrations of dyes in solvent were obtained by
stepwise dilution. Cyclic voltammetry measurements were carried out on an
electrochemical workstation (VSP-300, Bio-Logic, France) using a three-electrode
setup in which the working electrode and film counterelectrode were two platinum
wires, with a standard calomel electrode (SCE) as reference electrode. The highest
occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital
(LUMO) levels of the four coumarin-substituted phthalocyanine compounds were
estimated using the following equations reported by de Leeuw et al. [36]:
EHOMO ¼ ÀðEoOnxsyet!SCE þ 4:4 eVÞ;
ELUMO ¼ ÀðEoRnesdet!SCE þ 4:4 eVÞ:
Each dye was analyzed in dry dimethylformamide (DMF) containing tetrabuty-
lammonium perchlorate (0.1 M) as supporting electrolyte. Before testing, nitrogen
was inputted for 15 min to remove oxygen. Thermogravimetric analysis was carried
out from 30 to 750 °C using a ZCT-A thermogravimetric analyzer in flowing
nitrogen or flowing air (45 mL min-1) at heating rate of 10 °C min-1. The sample
was prepared at 60 °C under vacuum for 12 h. Device fabrication followed a
method similar to that reported recently [33]. Dye solutions were prepared at
concentration of 1 9 10-6 M in either THF or DMF containing 0.4 mM 3,7-di-
hydroxy-5b-cholic acid (CDCA). The electrolyte contained 0.05 M iodine and
0.5 M LiI in acetonitrile.
Synthesis
7-(3,4-Dicyanophenoxy)coumarin-3-carboxylic acid (1)
A mixture of 1.73 g (10 mmol) 4-nitrophthalonitrile and 2.04 g (10 mmol)
7-hydroxycoumarin-3-carboxylic acid was dissolved in 35 mL DMF, and the
solution was stirred and heated to 60 °C under nitrogen atmosphere. Subsequently,
4.1 g (30 mmol) fine anhydrous potassium carbonate was added in batches within
3 h and reacted for 32 h. Then, the reaction mass was poured into 120 mL ice water
and the pH value adjusted to 4. The precipitate was filtered, and washed with water
and ethanol several times until the filtrate became neutral, followed by recrystal-
lization from a mixture of acetone/methanol (1:4 v/v) to obtain yellow–brown
powder, which was finally dried under vacuum at 60 °C overnight. Yield: 67%
(2.3 g). M.p.: 243–245 °C. Anal. Calc. for C18H8N2O5: C, 65.07; H, 2.43; N, 8.43.
Found: C, 64.98; H, 2.44; N, 8.47. IR (KBr) mmax: 3146–2785 cm-1 (COOH),
3075 cm-1, 3040 cm-1 (C=C–H), 2234 cm-1 (C:N), 1708 cm-1 (C=O), 1589,
1569, 1485 cm-1 (C=C), 1310 cm-1 (C–N), 1252 cm-1 (Ar–O–Ar), 1202,
1093 cm-1 1H NMR (DMSO-d6, 400 MHz) (d, ppm): 10.73 (s, 1H, COOH),
.
8.53 (s, 1H, CH=C), 7.85 (d, 1H, Ar–H), 7.52 (s, 1H, Ar–H), 7.31 (d, 2H, Ar–H),
7.19 (d, 1H, Ar–H), 6.86 (s, 1H, Ar–H). MALDI-TOF MS calcd. for [M ? H]?
333.27; found 333.18.
123

Synthesis and properties of carboxyl-substituted…
6-(3,4-Dicyanophenylazo)coumarin-3-carboxylic acid (2)
A mixture of 4-aminophthalonitrile (1.56 g, 11 mmol), concentrated hydrochloric
acid (3 mL), and deionized water (30 mL) was stirred in an ice–water bath until
thoroughly mixed. Then, a solution of sodium nitrite (0.76 g, 11 mmol) was added
dropwise to the mixture during 20 min at approximately 0–5 °C, followed by
constant stirring for another 40 min. Then, the reaction mixture was filtered,
resulting in bright-yellow solution, which was kept at temperature around 0 °C for
the next reaction. At the same time, coumarin-3-carboxylic acid (1.92 g, 10 mmol)
was dissolved in 20 mL deionized water by adjusting the pH value to around 10
using sodium hydroxide solution (20%) in a three-necked flask. Then, previously
prepared diazonium salt solution was added slowly to a stirred solution of coumarin-
3-carboxylic acid as the coupling component in alkali medium with pH value
controlled to around 9 while maintaining temperature of 0–5 °C. The resulting
solution was stirred for 45 min in an ice–water bath, then removed to ambient
temperature. Two hours later, dilute hydrochloric acid solution was added to adjust
the pH value of the reaction solution to 5. The yellow–brown solid was filtered off,
washed with ice water repeatedly, purified by recrystallization from methanol and
acetone (1:1), and finally dried under vacuum at 40 °C overnight. Yield: 52%
(1.79 g). M.p.: 178 °C. Anal. Calc. for C18H8N4O4: C, 62.75; H, 2.34; N, 16.27.
Found: C, 62.70; H, 2.35; N, 16.32. IR (KBr) mmax: 3137–2785 cm-1 (COOH),
3073, 3038 cm-1 (Ar, C–H), 2228 cm-1 (C:N), 1732 cm-1 (C=O), 1590 cm-1
(Ar, C=C), 1490 cm-1 (N=N), 1391 cm-1 (C–N), 1374, 1316, 1249, 1132 cm-1
.
1H NMR (DMSO-d6, 400 MHz) (d, ppm): 11.07 (s, 1H, COOH), 8.14 (s, 1H,
CH=C), 7.93 (d, 1H, Ar–H), 7.81 (m, 2H, Ar–H), 7.68 (s, 1H, Ar–H), 7.60 (d, 1H,
Ar–H), 7.36 (d, 2H, Ar–H). MALDI-TOF MS calcd. for [M ? H]? 345.28; found
345.36.
7-Hydroxy-8-(3,4-cyanophenylimino)coumarin-3-carboxylic acid (3)
In a 250-mL three-necked flask, 1.17 g (5 mmol) 7-hydroxy-4-methylcoumarin-8-
carbaldehyde and 1.14 g (8.5 mmol) anhydrous aluminum chloride were mixed,
then 20 mL dichloromethane was added slowly until the aluminum chloride
completely dissolved. Then, dichloromethane solution containing 1.73 g (10 mmol)
4-aminophthalonitrile and 3.5 mL (25 mmol) triethylamine was added dropwise to
the reactor, with a lot of white mist; the colour of the mixture became red–brown.
The solution was stirred at ambient temperature for 24 h. After completion of the
reaction, 2.72 g (68 mmol, eight times that of Al2O3) sodium hydroxide solution
was added to the mixture. The white solid was filtered off, and the aqueous phase
was extracted twice with dichloromethane. The organic layers were combined and
dried over MgSO4, and the solvent evaporated. The residue was purified by column
chromatography on aluminum oxide using ethyl acetate/methanol (3:1) to obtain
yellow powder. Yield: 37% (0.62 g). M.p. 209–212 °C. Anal. Calc. for C19H9N3O5:
C, 63.51; H, 2.52; N, 11.70. Found: C, 63.47; H, 2.73; N, 11.68. IR (KBr) mmax
:
3172–2763 cm-1 (OH), 3065, 3026 cm-1 (Ar, C–H; N=C–H), 2227 cm-1 (C:N),
1714 cm-1 (C=O), 1682, 1632 cm-1 (alkene, C=C, N=C), 1587, 1526, 1462,
123

X. Zhang et al.
1
1317 cm-1 (C–N), 1254, 1108 cm-1. H NMR (DMSO-d6, 400 MHz) (d, ppm):
10.84 (s, 1H, COOH), 8.41 (s, 1H, CH=C), 8.26 (s, 1H, CH=N), 7.83 (m, 3H, Ar–
H), 7.21 (d, 1H, Ar–H), 6.79 (d, 1H, Ar–H), 4.7 (s, 1H, OH). MALDI-TOF MS
calcd. for [M ? H]? 360.29; found 360.37.
2(3),9(10),16(17),23(24)-Tetrakis(3-carboxycoumarin-7-oxy)phthalocyaninato
zinc(II) (COPc)
A mixture of 1 (0.67 g, 2 mmol), zinc acetate dihydrate (0.13 g, 0.6 mmol), and
catalytic amount of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) in n-pentanol
(15 mL) was heated to 140 °C and stirred under nitrogen atmosphere for 24 h.
After cooling to room temperature, the reaction mixture was precipitated by adding
methanol (60 mL). The resulting black suspension was filtered off, and washed with
deionized water and methanol. The black–green solid obtained after filtration was
subjected to Soxhlet extraction using acetone for 24 h. Finally, the solid material
was purified by silica gel column chromatography and eluted with trichloromethane,
collecting a blue–green band. The product was obtained under reduced pressure.
Yield: 37% (0.26 g). M.p. [300 °C. UV–Vis (DMF) kmax: 345, 677 nm. Anal.
Calc. for C72H32N8O20Zn: C, 61.98; H, 2.29; N, 8.03; found C, 61.89; H, 2.31; N,
8.06. IR (KBr) mmax: 3421 cm-1 (COOH), 3031 cm-1 (Ar–H), 1719 cm-1 (C=O),
1607–1469 cm-1 (Ar, C=C), 1233, 1101 cm-1. 1H NMR (DMSO-d6, 400 MHz) (d,
ppm): 11.33 (s, 4H, OH), 8.65 (s, 4H, CH=C), 7.93–7.86 (m, 12 H, Pc–H),
7.51–7.23 (m, 12H, Ar–H, J = 8.0 Hz). MALDI-TOF MS calcd. for [M ? H]?
1395.47; found 1395.29.
2(3),9(10),16(17),23(34)-Tetrakis(3-carboxycoumarin-7-azo)phthalocyaninato
zinc(II) (CAPc)
CAPc was synthesized following the procedure described above for COPc, except
using 0.69 g (2 mmol) 2 instead of 1. The solid material was purified using silica gel
column chromatography and eluted with DMF:H2O:conc. HCl = 30:10:1 (v/v),
collecting a black–green band. The product was obtained under reduced pressure.
1
Yield: 0.67 g, 44%, M.p. [300 °C. UV–Vis (DMF) kmax: 355, 553, 718 nm; H
NMR (DMSO-d6, 400 MHz) d: 11.56 (s, 4H, COOH), 8.37 (s, 4H, coumarin
unsaturation hydrogen CH=C), 8.18–8.29 (m, 12H, Pc–H), 8.12 (s, 4H, coumarin–
H), 8.02–8.04 (d, 4H coumarin–H, J = 7.8 Hz), 7.61–7.63 (d, 4H, coumarin–H,
J = 8.8 Hz); IR (KBr) mmax: 3407–3200 cm-1 (carboxyl OH), 1715 cm-1 (lactone,
C=O), 1599–1491 cm-1 (C=C), 1459 cm-1 (N=N), 1095 cm-1 (Ar–O–C). Anal.
Calc. for C72H32N16O16Zn: C 59.89, H 2.22, N 15.53; found C 59.82, H 2.27, N
15.55. MALDI-TOF MS calcd. for [M ? H]? 1443.53; found 1443.51.
2(3),9(10),16(17),23(24)-Tetrakis(3-carboxycoumarin-7-imino)phthalocyaninato
zinc(II) (CSPc)
CSPc was synthesized following the procedure described above for COPc, except
using 0.71 g (2 mmol) 3 instead of 1. The solid material was purified by aluminum
123

Synthesis and properties of carboxyl-substituted…
oxide column chromatography and eluted with trichloromethane:methanol = 3:1
(v/v), collecting a gray–green band. The product was obtained under reduced
pressure. Yield: 32% (0.24 g). M.p. [300 °C. UV–Vis (DMF) kmax: 363, 716 nm.
Anal. Calc. for C76H36N12O20Zn: C, 60.72; H, 3.42; N, 11.19; found: C, 60.67; H,
3.47; N, 11.07. IR (KBr) mmax: 3412, 3233 cm-1 (O–H), 3024 cm-1 (Ar–H),
1722 cm-1 (C=O), 1604–1281 cm-1 (Pc, C=N), 1265 cm-1 (C–O–C), 1188, 1093,
1
1048 cm-1. H NMR (DMSO-d6, 400 MHz) (d, ppm): 11.47 (s, 4H, COOH), 8.43
(s, 4H, CH=N), 8.29 (s, 4H, CH=C), 8.07 (m, 12H, PcH), 7.84 (d, 4H, Ar–H,
J = 8.0 Hz), 7.25 (d, 4H, Ar–H, J = 8.0 Hz), 5.21 (broad, 4H, OH) MALDI-TOF
MS calcd. for [M ? H]? 1503.57; found 1503.83.
2(3),9(10),16(17)-Tris(4-methylcoumarin-7-oxy)-23(24)-(3-carboxycoumarin-7-
azo)phthalocyaninato zinc(II) (OAPc)
OAPc was synthesized following the procedure described above for COPc, except
using a mixture of 0.46 g (1.5 mmol) 4 and 0.17 g (0.5 mmol) 3 instead of 1. The
solid material was purified by silica gel column chromatography and eluted with
DMF:THF = 1:3 (v/v), collecting a green band. The product was obtained under
reduced pressure. Yield: 12% (0.16 g). M.p. [300 °C. UV–Vis (DMF) kmax: 325,
500, 680 nm. Anal. Calc. for C72H38N10O13Zn: C, 65.65; H, 2.89; N, 10.64; found:
C, 65.60; H, 3.01; N, 10.70. IR (KBr) mmax: 3394 cm-1 (COOH), 3016 cm-1 (Ar–
H), 2948 cm-1 (CH3), 1718 cm-1 (C=O), 1643–1381 cm-1 (Pc skeleton, C=C and
1
N=N), 1236, 1091 cm-1. H NMR (DMSO-d6, 400 MHz) (d, ppm): 11.38 (s, 1H,
COOH), 8.24 (s, 1H, CH=C), 7.17–7.32 (m, 30H, 12PcH, 6ArH), 7.13 (m, 6H, Ar–
H, J = 8.4 Hz), 6.39 (s, 3H, CH=C), 2.17 (s, 9H, CH3). MALDI-TOF MS calcd. for
[M ? H]? 1317.54; found 1317.49.
Results and discussion
Synthesis and characterization
Coumarin–phthalocyanine dyad dyes were synthesized following the protocol
illustrated in Scheme 1. Alkylation of 4-nitrophthalonitrile with 7-hydroxy-
coumarin-3-carboxylic acid or 7-hydroxy-4-methylcoumarin in presence of
K2CO3 in DMF was performed to obtain coumarin-substituted phthalonitrile
precursors 7-(3,4-dicyanophenoxy)-coumarin-3-carboxylic acid (1) and 4-methyl-7-
(3,4-dicyanophenoxy)coumarin (4) [29, 32], respectively. Meanwhile, 4-aminoph-
thalonitrile was subjected to diazo coupling [30] or Schiff-base condensation [37]
reaction with coumarin-3-carboxylic acid or 7-hydroxy-4-methylcoumarin-8-car-
baldehyde to obtain the required coumarin-substituted phthalonitrile precursors
6-(3,4-dicyanophenylazo)coumarin-3-carboxylic acid (2) and 7-hydroxy-8-(3,4-
cyanophenylimino)coumarin-3-carboxylic acid (3). In the final step, the precursors
1, 2, 3 were condensed with zinc acetate by cyclotetramerization to produce the
target coumarin–phthalocyanine dyad dyes 2(3),9(10),16(17),23(24)-tetrakis(3-
carboxycoumarin-7-oxyl)phthalocyaninato zinc(II) (COPc), 2(3),9(10),16(17),
123

X. Zhang et al.
Scheme 1 Synthesis of carboxyl coumarin–phthalocyanine dyad dyes
23(34)-tetrakis(3-carboxycoumarin-7-azo)phthalocyaninato zinc(II) (CAPc), and
2(3),9(10),16(17),23(24)-tetrakis(3-carboxycoumarin-7-imino)phthalocyaninato
zinc(II) (CSPc), respectively. The target dyad dye 2(3),9(10),16(17)-tris(4-methyl-
coumarin-7-oxy)-23(24)-(3-carboxycoumarin-7-azo)phthalocyaninato zinc(II)
(OAPc) was obtained by reaction of 4-methyl-7-(3,4-dicyanophenoxy)coumarin
(4) with 6-(3,4-dicyanophenylazo)coumarin-3-carboxylic acid (2). The dyad dyes
were soluble in dimethyl formamide, dimethylsulfoxide, and pyridine, and sparingly
in trichloromethane, tetrahydrofuran, triethylamine, etc. Finally, all intermediates
and final targets were thoroughly analyzed by IR spectroscopy, 1H NMR, elemental
analysis, and mass spectroscopy.
In the IR spectra, formation of compounds 1, 2, 3, 4 was clearly confirmed by
C:N vibration at around 2230, 2234, 2228, and 2227 cm-1, a,b-unsaturated –
C=C– of lactone ring at 1569 cm-1, and ether group C–O at 1208 cm-1, without
vibrations of NO2 or NH2. After conversion of the dinitrile derivatives 1, 2, 3, 4 into
the phthalocyanines COPc, CAPc, CSPc, OAPc, the sharp peak of C:N vibration
at around 2223 cm-1 disappeared. The data are consistent with the proposed
1
molecular structures. In addition, the H NMR and elemental analysis results were
consistent with the proposed molecular structures; For example, the expected
1
signals of CAPc in the H NMR spectra were observed in accordance with the
proposed structure. Isomeric multiplet peaks located at 8.30–8.18 ppm with 12
protons were found for the Pc ring, making the calculation of coupling constants
inaccurate. Aromatic protons of coumarin were observed as two sets of doublets at
8.04–8.02 and 7.63–7.61 ppm, integrating to eight protons. Three singlets located at
11.56, 8.37, and 8.12 ppm appeared, being ascribed to carboxyl acid protons,
unsaturated protons, and coumarin C-5 protons, respectively, all of them indicating
123

Synthesis and properties of carboxyl-substituted…
four protons. The signal of carboxyl proton was observed to be broader and weaker
due to its exchange with solvent.
Absorption properties and aggregation behavior
The UV–Vis absorption spectra of the four dyes in DMF solution, exhibiting two
types of characteristic absorption bands (Q-band and Soret band), are displayed in
Fig. 2; the corresponding absorption data are listed in Table 1. The longer-
wavelength stronger Q-band absorption band at 650–750 nm is lacking due to
phthalocyanine ring p ? p* transitions (a1u ? e*g) [38], whereas the shorter-
wavelength broader Soret absorption band at 330–400 nm corresponds to transitions
from a2u and b1u to eg* orbitals [37]. This result suggests that the optical properties of
the phthalocyanine dyes can be finely tuned by altering the nature of the asymmetric
structure. Using the equation E0-0 = 1240/kmax, the E0-0 energy for COPc, CAPc,
CSPc, and OAPc was estimated to be 1.83, 1.73, 1.73, and 1.80 eV, respectively.
The decrease of the energy gap of the dyad dyes results in a red-shift of the
absorption band. The Soret bands of all the dyad dyes were broadened due to
superimposition of phthalocyanine and coumarin in the *450 nm region, suggest-
ing the absence of strong electronic interaction between the Pc and coumarin
components in the ground state, resulting in an intramolecular color match.
Meanwhile, the Q-band absorption (kmax) of OAPc showed a slight red-shift
(10 nm) or significant blue-shift (31 nm) compared with COPc and CAPc,
respectively. On changing the linkage between coumarin and phthalocyanine, the
diazo- and imino-conjugated coumarin–phthalocyanine dyad dyes CAPc and CSPc
displayed broadened and red-shifted absorption at 718 and 715 nm compared with
COPc, respectively. This can be explained by the fact that efficient incorporation of
the coumarin ring and phthalocyanine core in the conjugation pathway increased the
electron delocalization scope over the molecule, thus reducing the bandgap of the
Fig. 2 Absorption spectra for dyes COPc, CAPc, CSPc, and OAPc recorded in DMF
123

X. Zhang et al.
Table 1 Electrochemical properties of COPc, CAPc, CSPc, and OAPc
abs/nm (emax 9 105 M-1 cm-1
)
Egopg/eV
EHOMO/eV ELUMO/eV
Oxy
Onset
Dyad
k
E
/V
COPc 677 (2.47), 615 (0.56), 344 (1.60)
CAPc 718 (2.32), 551 (1.77), 358 (8.24)
CSPc 715 (2.42), 646 (0.71), 365 (2.29)
1.83
1.73
1.73
0.51
0.77
0.67
0.71
-4.91
-3.08
-3.44
-3.34
-3.31
-5.17
-5.07
-5.11
OAPc 777 (0.57), 687 (2.39), 644 (1.08), 350 (3.07) 1.80
dye. Interestingly, CAPc exhibited a new absorption in the range of 519–592 nm
compared with COPc featuring oxygen linkage, and CSPc displayed intense
absorption at around 400–550 nm. This is probably due to electron transfer from the
peripheral, tetraazo substituent p* orbital to the Pc 1s level [39]. This clearly
indicates that the azo linker is a better choice than oxygen or imino regarding
realization of panchromatic light harvesting.
It is also interesting to compare the aggregation behavior of the dyad dyes with
oxygen, azo, and imino linkage, which helps to reveal the impact of the linker
segment on the absorption profile. p–p* interactions between the p-electron clouds
surrounding Pc macrocycles result in aggregation of phthalocyanine molecules [40].
In the aggregated state, the electronic structure of the complex phthalocyanine rings
is perturbed, resulting in alteration of the ground- and excited-state electronic
structures, leading to change of the molecular extinction coefficient and position of
the Q-band [38]. In this study, the molar absorption coefficient of the stronger
Q-band (kmax = 677, 718, 715, and 687 nm for COPc, CAPc, CSPc, and OAPc,
respectively) was monitored in the concentration range from 1 9 10-6 to
8 9 10-6 mol L-1 in DMF to provide support for the change in aggregated
species (Fig. 3). With increasing concentration, the absorption intensity of the
Q-band maxima increased, the Beer–Lambert law was obeyed, and there was no
blue-shift typical of aggregated species for COPc, CSPc, and OAPc, suggesting
that the effects of the coumarin substituent on the phthalocyanine core prevented
aggregation [41]. Interestingly, the intensity of the Q-band absorption for CAPc
also increased, while the Beer–Lambert law was not obeyed as the concentration
was increased. Furthermore, there was a more obvious change of the Soret band
than the Q-band. The different bridge linkages in COPc, CAPc, and CSPc resulted
in different aggregation behavior, which is ascribed to the double-azo bond being
more rigid than the oxygen or imino bond [42], resulting in a more rigid structure
that hinders the coumarin substitution compared with the other linker units.
Electrochemical characteristics
To investigate the effect of the coumarin substituent on the electrochemical
characteristics of the synthesized dyes as well as the feasibility of electron injection
from the excited dye into the conduction band of TiO2 and dye regeneration from
I-/I3- redox electrolyte, the relative energy levels EHOMO and ELUMO are required.
123

Synthesis and properties of carboxyl-substituted…
Fig. 3 Absorption spectra for COPc, CAPc, CSPc, and OAPc dyes at different concentrations
To obtain these values, cyclic voltammograms were obtained in DMF at scan rate of
100 mV s-1 (Fig. 4); the corresponding data are presented in Table 1. In our
experiment, it was found that all of the dyad dyes showed one quasireversible
reduction couple, ascribed to reduction of four coumarin moieties, and a one-
electron irreversible oxidation wave, attributed to oxidation of the Pc core. The
ox
oxidation onset potentials E
determined from the intersection of two
onset!SCE
tangents drawn at the rising current and baseline charging current of the cyclic
voltammograms [13] for the dyes revealed the order: COPc (0.51 V) \ CSPc
(0.67 V) \ OAPc (0.71 V) \ CAPc (0.77 V). There are various interesting points,
irrespective of the phthalocyanine core and coumarin substituent: the p-conjugated
linked dyad dyes CAPc and CSPc showed higher oxidation potential than the
oxygen p-conjugated linked dyad dye COPc. This clearly indicates that electron of
p-conjugated dyes is more efficient than p-conjugated delocalization at the
coumarin and phthalocyanine moieties. Also, for the asymmetric coumarin-
substituted phthalocyanine OAPc, the coumarinazo moieties of the molecule
facilitate oxidation, while the triscoumarinoxy moieties showed the opposite trend.
The ELUMO level of the sensitizer corresponding to the excited-state oxidation
potential is closely related to the electron injection efficiency of the TiO2 surface,
and a higher ELUMO level is helpful for providing higher electron injection potential
[9, 21]. In general, the low EHOMO level of the dye sensitizers corresponding to the
oxidation onset potentials matches with the energy level of the I-/I3- redox couple to
123

X. Zhang et al.
Fig. 4 Cyclic voltammograms of phthalocyanines COPc, CAPc, CSPc, and OAPc
Fig. 5 Energy-level diagrams for COPc, CAPc, CSPc, and OAPc
achieve efficient regeneration from the oxidized dyes, limiting further improvement
of the energy conversion efficiency of DSSCs. The ELUMO level and EHOMO level of
all the dyad dyes were found to fall in the range of -3.44 to -3.08 and -5.17 V to
123

Synthesis and properties of carboxyl-substituted…
-4.91 V (versus SCE), respectively; the energy-level diagrams are shown in Fig. 5.
The ELUMO level was higher than the energy level for the conduction band (CB) of
TiO2 (-3.6 eV versus vacuum), suggesting that electron injection from the excited
dye to the CB of TiO2 is a thermodynamically favorable process. Besides, the
EHOMO level corresponding to the oxidation onset potentials of all the dyad dye
sensitizers was more positive than the iodine/iodide redox potential (0.2 V versus
SCE) [9, 13, 32], ensuring regeneration of the oxidized dyes by electron recapture
from the redox electrolyte. In general, the dyad dyes containing p-conjugated
moieties (CAPc, CSPc, and OAPc) possessed low EHOMO (Fig. 5; Table 1)
compared with their oxygen-linked counterpart (COPc). This low EHOMO may
suppress electron back-transfer from the CB of TiO2 to the oxidized dyes through
facile regeneration of the oxidized dye, which will eventually increase VOC due to
accumulation of more electrons in the CB of TiO2.
Thermal degradation
The thermal stability of the synthesized dyad dyes was tested by thermogravimetric
analysis under nitrogen atmosphere at heating rate of 10 °C min-1; the resulting
TGA curves are shown in Fig. 6. Seoudi et al. [43] compared a series of
unsubstituted metal phthalocyanines by thermogravimetric analysis, finding a first
minor decomposition reaction at 440 °C and a major decomposition reaction at
495 °C for the phthalocyaninato zinc compounds, but when additional functional
groups were introduced into the molecule, most of the substituted phthalocyanines
were not as stable as their unsubstituted counterparts [38]. Recently, we reported the
thermal behavior of a series of azo-coupled zinc phthalocyanine derivatives, whose
TGA curves showed peaks at around 400 °C [30].
The TGA curve for the oxygen-linked dyad dye COPc shows three mass loss
stages in the temperature range of 150–700 °C. The first stage onsets at 214 °C with
mass loss of 41%, corresponding to loss of some carboxylic coumarin moieties,
Fig. 6 TGA curves of compounds COPc, CAPc, CSPc, and OAPc
123

X. Zhang et al.
while the other two stages that onset at 425 and 503 °C correspond to
decomposition of phthalocyanine. The TGA curve of the azo-linked dye CAPc
exhibits three different degradation stages. The weight loss of 12% observed
between 200 and 309 °C corresponds to the theoretical content of the four carboxyl
groups [44], and two major decomposition stages occurred at around 339 and
474 °C. The former decomposition stage begins with scission of N=N bonds
[30, 33], while the latter maximum decomposition in the range of 437–474 °C is due
to degradation of the phthalocyanine core itself. The TGA curve of the imino-linked
dye CSPc exhibits degradation analogous to CAPc, consistent with the report of
Mohamed et al. [45]. The asymmetric dyad dye OAPc showed slight decomposition
with onset at around 205 °C, corresponding to loss of carboxyl group, and about
14% weight runaway ascribed to loss of azo moiety, with major decomposition in
the range of 435–560 °C, possibly due to decomposition of the phthalocyanine core
and 4-methylcoumarin substituent. Overall, it is clear that all the dyad sensitizers
were stable up to 200 °C, indicating that all the dyad dyes are sufficiently thermally
stable for rooftop applications of dye-sensitized solar cells.
Photovoltaic performance
To evaluate the photovoltaic performance of compounds COPc, CAPc, CSPc, and
OAPc as DSSC sensitizers, current–voltage curves for cells fabricated using TiO2
electrodes sensitized with the dyad dyes were measured in ambient air under
standard simulated AM1.5G illumination (100 mW cm-2) conditions (Fig. 7); the
corresponding short-circuit currents (JSC), open-circuit potentials (VOC), fill factors
(ff), and overall conversion efficiencies (g) are presented in Table 2. The g values
for the DSSCs based on these dyad dyes lay in the range of 0.9–2.4%. Among the
dyes, the cell fabricated with the asymmetric dye OAPc exhibited high g (2.4%)
with high JSC of 8.1 mA cm-2, VOC of 0.58 V, and fill factor (ff) of 0.53.
Obviously, the g and JSC values of cells sensitized using the p-conjugated dyad dyes
Fig. 7 Photocurrent–voltage curves of DSSCs under simulated sun illumination
123

Synthesis and properties of carboxyl-substituted…
Table 2 Photovoltaic
Compound
JSC (mA cm-2
)
VOC (V)
ff
g (%)
characteristics of DSSCs
sensitized by COPc, CAPc,
COPc
2.9
5.8
6.5
8.1
0.56
0.58
0.54
0.58
0.57
0.51
0.54
0.53
0.9
1.7
1.9
2.4
CSPc, and OAPc
CAPc
CSPc
OAPc
CAPc and CSPc were slightly better than that using the dye COPc with oxygen
bridge linkage. These results are ascribed to superior light-harvesting and effective
electron transfer. Arguably, the efficiency of the cells using CAPc and CSPc is
opposite to that expected based on their absorption spectra, which can be attributed
to the aggregation behavior of CAPc. All the dyad dyes showed similar ff values,
which is related to the fact that they contain both coumarin and phthalocyanine core.
The comparatively high efficiency obtained when using the dye OAPc may
originate from the following factors: (1) the presence of methylcoumarin as a steric
hindrance moiety effectively suppresses dye aggregation, (2) the boarder and more
efficient absorption spectrum, and (3) the asymmetric structure of the dye molecule
that promotes electron transfer.
Conclusions
A series of carboxyl coumarin–phthalocyanine dyad dyes conjugated by –O–, –
N=N– or –N=C– bond in a single structure were designed, synthesized, and
demonstrated as sensitizers in DSSCs. The influence of the linker on their
absorption spectrum and electrochemical properties was revealed, and their
photovoltaic parameters such as g, JSC, and VOC studied. The absorption spectra
demonstrate that the p-conjugated dyad dyes exhibited excellent absorption and
photocurrent response across the whole visible-light region in the range of
400–800 nm due to reduction of the energy gap. The p-conjugated coumarin–
phthalocyanine dyes CAPc and CSPc exhibited only moderate power conversion
efficiency, mainly due to insufficient electron transfer of the symmetric coumarin-
substituted phthalocyanine core. In contrast, the dyad dye OAPc exhibited
improved asymmetric structure and efficiency of 2.4%. CAPc showed panchromatic
response, but the CAPc-based DSSC showed low conversion efficiency. Further
work to develop new asymmetric panchromatic dyes and improve their electron
transfer is underway.
Acknowledgements The authors would like to acknowledge financial support from the Natural Science
Foundation of Shanxi Province, China (Grant No. 2015011011) and the Research Project Supported by
Shanxi Scholarship Council of China (Grant No. 2015-078).
References
¨
1. B. O’Regan, M. Gratzel, Nature 737, 353 (1991)
2. M. Gratzel, Nature 338, 414 (2001)
123

X. Zhang et al.
3. M.K. Nazeeruddin, F. De Angelis, S. Fantacci, A. Selloni, G. Viscardi, P. Liska, S. Ito, B. Takeru, M.
Gratzel, J. Am. Chem. Soc. 16835, 127 (2005)
´ ´
4. M.J. Bosiak, M. Rakowiecki, A. Wolan, J. Szlachta, E. Stanek, D. Cycon, K. Skupien, Dyes Pigments
79, 121 (2015)
´
5. E. Torres, S. Sequeira, P. Parreira, P. Mendes, T. Silva, K. Lobato, M.J. Brites, J. Photochem.
Photobiol. 1, 310 (2015)
6. C. Zhong, J. Gao, Y. Cui, T. Li, L. Han, J. Power Sources 831, 273 (2015)
7. S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, B.F. Curchod, N. Ashari-Astani, I. Tavernelli, U.
Rothlisberger, M.K. Nazeeruddin, M. Gratzel, Nat. Chem. 242, 6 (2014)
8. Y. Xie, Y. Tang, W. Wu, Y. Wang, J. Liu, X. Li, H. Tian, W.H. Zhu, J. Am. Chem. Soc. 14055, 137
(2015)
9. Z.-S. Huang, H. Meier, D. Cao, J. Mater. Chem. C 2404, 4 (2016)
10. T. Ikeuchi, H. Nomoto, N. Masaki, M.J. Griffith, S. Mori, M. Kimura, Chem. Commun. 1941, 50
(2014)
11. L. Giribabu, V.K. Singh, C.V. Kumar, Y. Soujanya, P.Y. Reddy, M.L. Kantam, Sol. Energy 1204, 85
(2011)
12. L. Tejerina, M.V. Martinez-Diaz, M.K. Nazeeruddin, T. Torres, Chem. Eur. J. 4369, 22 (2016)
13. G.S. Reddy, S. Ramkumar, A.M. Asiri, S. Anandan, Spectrochim. Acta A 531, 145 (2015)
14. G.B. Bodedla, K.R. Justin Thomas, M.S. Fan, K.C. Ho, J. Org. Chem. 640, 81 (2016)
15. K. Kakiage, Y. Aoyama, T. Yano, T. Otsuka, T. Kyomen, M. Unno, M. Hanaya, Chem. Commun.
6379, 50 (2014)
16. B. Basheer, D. Mathew, B.K. George, C.P. Reghunadhan Nair, Sol. Energy 108, 479 (2014)
17. Y. Ooyama, Y. Harima, Eur. J. Org. Chem. 2903, 2009 (2009)
18. P. Kluson, M. Drobek, A. Kalaji, M. Karaskova, J. Rakusan, Res. Chem. Intermed. 103, 35 (2009)
19. C.-P. Lee, R.Y.-Y. Lin, L.-Y. Lin, C.-T. Li, T.-C. Chu, S.-S. Sun, J.T. Lin, K.-C. Ho, RSC Adv.
23810, 5 (2015)
20. S. Yamamoto, S. Mori, P. Wagner, A.J. Mozer, M. Kimura, Isr. J. Chem. 175, 56 (2016)
21. M. Liang, J. Chen, Chem. Soc. Rev. 3453, 42 (2013)
22. F. Liang, F. Shi, Y. Fu, L. Wang, X. Zhang, Z. Xie, Z. Su, Sol. Energy Mater. Sol. Cells 1803, 94
(2010)
23. B. Ren, L. Zhu, G. Cui, Y. Sun, X. Zhang, F. Liang, Y. Liu, Tetrahedron Lett. 5953, 54 (2013)
24. C.B. Kc, K. Stranius, P. D’Souza, N.K. Subbaiyan, H. Lemmetyinen, N.V. Tkachenko, F. D’Souza, J.
Phys. Chem. C 763, 117 (2013)
25. J. Wang, M. Li, D. Qi, W. Shen, R. He, S.H. Lin, RSC Adv. 53927, 4 (2014)
26. A. Aslı Esenpınar, M. Bulut, Dyes Pigments 249, 76 (2008)
27. E. Dur, M. Bulut, Polyhedron 2689, 29 (2010)
28. O. Soyer Can, E.N. Kaya, M. Durmus¸, M. Bulut, J. Photochem. Photobiol. A 56, 317 (2016)
29. A. Tas¸temel, B.Y. Karaca, M. Durmus¸, M. Bulut, J. Lumin. 163, 168 (2015)
30. M. Han, X. Zhang, X. Zhang, C. Liao, B. Zhu, Q. Li, Polyhedron 864, 85 (2015)
31. L. Zhang, J.M. Cole, C. Dai, ACS Appl. Mater. Interfaces 7535, 6 (2014)
32. B. Zhu, X. Zhang, M. Han, P. Deng, Q. Li, J. Mol. Struct. 61, 1079 (2015)
˙
33. H. Dinc¸alp, F. Toker, I. Durucasu, N. Avcıbas¸ı, S. Icli, Dyes Pigments 11, 75 (2007)
34. J. Yang, C. Ji, Y. Zhao, Chin. J. Org. Chem. 1740, 28 (2008)
35. F. Chen, G. Liu, Y. Shi, P. Xi, J. Cheng, J. Hong, R. Shen, X. Yao, D. Bai, Z. Zeng, Talanta 139, 124
(2014)
36. D.M. de Leeuw, M.M.J. Simenon, A.R. Brown, R.E.F. Einerhand, Synth. Met. 53, 87 (1997)
37. Y. Liu, Q. Yang, D. Hao, W. Zhang, Aust. J. Chem. 1390, 65 (2012)
38. F. Ghani, J. Kristen, H. Riegler, J. Chem. Eng. Data 439, 57 (2012)
39. Y. Liu, H. Lin, X. Li, J. Li, H. Nan, Inorg. Chem. Commun. 187, 13 (2010)
40. R.Z.U. Kobak, M.U. Arı, A. Tekin, A. Gu¨l, Chem. Phys. 91, 448 (2015)
41. Z. Ning, Y. Fu, H. Tian, Energy Environ. Sci. 1170, 3 (2010)
42. M. Ikiz, E. Ispir, E. Aytar, E. Aytar, M. Ulusoy, S. Karabuga, M. Aslantas, O. Celik, New J. Chem.
8204, 39 (2015)
43. R. Seoudi, G.S. El-Bahy, Z.A. El Sayed, J. Mol. Struct. 119, 753 (2005)
44. M. Sreenivasu, A. Suzuki, M. Adachi, C.V. Kumar, B. Srikanth, S. Rajendar, D. Rambabu, R.S.
Kumar, P. Mallesham, N.V. Rao, M.S. Kumar, P.Y. Reddy, Chem. Eur. J. 14074, 20 (2014)
45. G.G. Mohamed, M.M. Omar, A.M. Hindy, Spectrochim. Acta A 1140, 62 (2005)
123