Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 96 (2012) 305–309

Contents lists available at SciVerse ScienceDirect

Spectrochimica Acta Part A: Molecular and

Biomolecular Spectroscopy
journal homepage: www.elsevier.com/locate/saa

Performance of Kerria japonica and Rosa chinensis flower dyes as sensitizers

for dye-sensitized solar cells
K.V. Hemalatha, S.N. Karthick, C. Justin Raj, N.-Y. Hong, S.-K. Kim, H.-J. Kim ⇑
School of Electrical Engineering, Pusan National University, Jangjeon, Geumjeong, Busan 609-735, Republic of Korea

h i g h l i g h t s g r a p h i c a l a b s t r a c t

" The efficiency of Kerria japonica and

Rosa chinensis dye is 0.22%, 0.29%
respectively.
" In the presence of sugar molecule, it
was increased to 0.29% for K. japonica
and decreased to 0.27% for
R. chinensis.
" Implementation of natural dye in
DSSC is cost effective compared to the
metal based dyes.

a r t i c l e i n f o a b s t r a c t

Article history: The natural dyes carotenoid and anthocyanin were extracted from Kerria japonica and Rosa chinensis,
Received 26 December 2011 respectively, using a simple extraction technique without any further purification. They were then used
Received in revised form 23 April 2012 as sensitizers in dye-sensitized solar cells (DSSCs), and their characteristics were studied. The ranges of
Accepted 10 May 2012
short-circuit current (JSC) from 0.559 to 0.801 (mA/cm2), open-circuit voltage (VOC) from 0.537 to 0.584 V,
Available online 24 May 2012
and fill factor from 0.676 to 0.705 were obtained for the DSSCs made using the extracted dyes. Sugar mol-
ecules were added externally to the dye for stabilization and to increase the conversion efficiency. The
Keywords:
efficiencies of the K. japonica and R. chinensis dyes were 0.22% and 0.29%, respectively; after the addition
Anthocyanin
Carotenoid
of sugar, the efficiency increased to 0.29% for K. japonica and decreased to 0.27% for R. chinensis. Thus, the
Dye-sensitized solar cell addition of sugar molecules increased the conversion efficiency slightly with the carotenoid dye of K.
Photocurrent japonica, while there was no considerable change with the anthocyanin of R. chinensis. This paper briefly
discusses the simple extraction technique of these natural dyes and their performance in DSSCs.
Ó 2012 Elsevier B.V. All rights reserved.

Introduction employed to sensitize nanocrystalline TiO2 semiconductors is a

Dye-sensitized solar cells (DSSCs) are considered as alternative
energy sources to conventional inorganic semiconductor photovol-
taic devices because of the low fabrication costs involved. Dyes
play key roles in light harvesting and in the conversion of solar en-
ergy into electrical energy, and the dye molecules in DSSCs ‘‘sensi-
tize’’ wide-bandgap semiconductors to visible radiation. Therefore,
dyes are referred to as ‘‘sensitizers’’. The most efficient sensitizer
⇑ Corresponding author. Tel.: +82 51 510 2364; fax: +82 51 513 0212.
E-mail address: heeje@pusan.ac.kr (H.-J. Kim).
transition-metal coordination compound (ruthenium polypyridyl
complex). The ruthenium polypyridyl complexes anchored to
nanocrystalline TiO2 surfaces through carboxylic acid groups [1]
are the most successful sensitizers for applications in regenerative
solar cells. However, the high cost and long-term unavailability of
Ru complexes have led to a widening of research on low-cost, effi-
cient sensitizers. Consequently, metal-free sensitizers such as nat-
ural dyes have been investigated as alternative molecules for DSSC
applications [2]. The advantages of natural pigments as photosen-
sitizers are their intramolecular p–p⁄ transitions, the absence of
noble metals such as Ru, and their large absorption coefficients,

1386-1425/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.saa.2012.05.027
306 K.V. Hemalatha et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 96 (2012) 305–309
high light-harvesting efficiencies, easy preparation, environmen-
tally friendly nature, high availability, and low cost [3].
In our present investigation, DSSCs were assembled using
natural dyes extracted from the flowers of Kerria japonica and Rosa
chinensis as sensitizers. The performances of the raw natural dyes
are better than those of the commercial or purified analogs because
of the presence of natural extracts like alcohols, organic acids, etc.,
which act as a co-adsorbates, suppressing recombination with the
electrolyte, reducing dye aggregation, and favoring charge
injection [4]. Therefore, the dyes were extracted using a minimal
chemical procedure without any purification. The pigment
carotenoid is responsible for the color of the flower of K. japonica.
The total percentage of carotenoid in the form of carotene and
xanthophyll content present in K. japonica is 88% [5]. All carote-
noids are polyisoprenoids, possessing an extensive system of
conjugated double bonds, as shown in Fig. 1(a). This serves as
the light-absorbing chromophore responsible for the yellow,
orange, and red colors of the plant sources, and provides the visible
absorption spectrum that is the basis for their identification. The
long system of alternating double and single bonds giving this
distinct light-absorbing property is the most important feature of
the carotenoids [6]. In the flower of R. chinensis, the anthocyanin
molecule is the core component, which shows color in the range
of visible light from red to blue, so it is considered as a valuable
natural material to be used as an efficient sensitizer for wide-band-
gap semiconductors. The presence of carbonyl and hydroxyl groups
in the anthocyanin molecule (Fig. 1b) bound to the TiO2 helps the
photoelectric conversion efficiency. The extracted dyes were trea-
ted with sugar molecules, because, according to Filipa Queiroz
et al., 2009 [7], sugar has a positive effect on the stability of the
anthocyanin pigment. Therefore, we attempted to add sugar exter-
nally to the anthocyanin pigment to stabilize the dye, and then
evaluated the current conversion efficiency. Natural dyes are
promising alternative sensitizers for DSSCs because they are easy
to prepare, widely available, cheap, and eco-friendly.
Experimental part
Preparation of dye sensitizer solutions
The flowers of K. japonica and R. chinensis were collected fresh
and kept in a vacuum furnace for about 10 h at 65 °C to remove
the moisture. After being dried, the flowers were crushed in a mor-
tar to make them into powder. About 1 g of the powdered sample
was dissolved in 60 mL of ethanol and kept for two days at ambient
Fig. 1. Structure of (a) xanthophyll and (b) anthocyanin.
temperature for adequate extraction without exposure to sunlight.
After the extraction, the solid residues were filtered out, and the
clear solutions were used as prepared, without purification. Fur-
ther purification of the extracts was avoided so as to achieve effi-
cient sensitization using simple extraction procedures. The
extractants were properly stored, protected from direct sunlight,
and used further as sensitizers in DSSCs.
Solar-cell assembly
A fluorine-doped tin oxide (FTO) glass plate (13 O/sq), pur-
chased from Hartford Glass Co., Inc., was used as the current collec-
tor. The FTO plate was first cleaned using an ultrasonic bath with
acetone, ethanol, and water for about 10 min each. Scotch tape
was used as a spacer to control the film thickness and to provide
non-coated areas for electrical contact. The TiO2 paste (Ti-nanoxide
T/SP, Solaronix) was coated on the FTO plate using the doctor-
blade method. After being coated, the TiO2 films were air-dried
for about 5 min to reduce the surface irregularities. The films were
annealed at 450 °C in air for 30 min to remove the organic loads
and facilitate the interconnection of the TiO2 nanoparticles. The
thickness of the films was 20 lm, and the active area of the
TiO2 electrodes used was 0.27 cm2. The TiO2 film electrodes were
cooled to 80 °C, and were then immersed in an extracted caroten-
oid and anthocyanin dye solution separately for 24 h. After the dye
adsorption was completed, the film was cleaned with pure ethanol
and dried with hot air. The Pt electrode was prepared using plati-
num paste (platisol T/SP, Solaronix). The paste was coated on the
FTO glass plate using the doctor-blade technique and heated at
400 °C for 30 min. The dye-covered TiO2 electrode and Pt counter
electrode were assembled as a sandwich-type cell. A drop of Iodo-
lyte AN-50 (Solaronix) electrolyte solution was injected into the
cell via a hole at the back of the counter electrode. The hole was
then sealed with a hot-melt ionomer film (SX 1170, Solaronix)
and covered with glass. Finally, the edge of each side of the FTO
glass was cleaned and soldered (ultrasonic soldering system, Mod-
el-9200) with alloy #143 (Cerasolza) in order to achieve a good
electrical contact for measurement.
Results and discussion
FTIR spectra
Fig. 2 shows the FTIR spectra of the extracted dyes from
K. japonica and R. chinensis. The main pigment of the K. japonica
 K.V. Hemalatha et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 96 (2012) 305–309 307

Fig. 2. FT-IR spectra of natural dye extracted from K. japonica and R. chinensis.

flower is carotenoid. The strong absorption of carotenoid in the

blue–green spectral region is caused by the strongly allowed
S0–S2 transition. The peaks at 1080 and 1646 cm 1 are attributed
to the symmetric vibrational modes of C–C and C@C stretching,
respectively. These are important spectroscopic measurements
since they reflect the structural properties of the carotenoid
molecule [8]. The intense bands at 1066 cm 1 are attributed to
the C–O–C vibrational modes of acid and carbohydrate groups.
The strong and broad bands at 3000–3700 cm 1 are due to
the –OH groups of water. The two peaks at 2923 and 2850 cm 1
correspond to the –CH stretching mode [9,10]. The strong
characteristic absorption spectrum observed with R. chinensis
may be due to the superimposition of the absorptions of two main
anthocyanin pigments. The two peaks at 2922 and 2850 cm 1
correspond to the –CH stretching modes. The peak at a wavenum-
ber of 3393 cm 1 corresponds to the OH stretching vibration for
the anthocyanin dye of R. chinensis. The spectrum also contains a
peak at 1726 cm 1, corresponding to the C@O stretching vibration. Fig. 3. UV absorption spectra of natural dye solution extracted from (a) K. japonica,
R. chinensis and (b) TiO2 adsorbed dye of K. japonica and R. chinensis.
This indicates that the anthocyanin dye has a partial quinonoidal
form [11]. absorption spectra of the TiO2-adsorbed carotenoid and anthocya-
nin dyes. Broad absorption peaks in all the samples indicate that
UV–Vis absorption spectra the dyes adsorbed on the TiO2 particles well, increasing the
absorption of light by the TiO2 nanoparticles in the visible range.
Fig. 3(a) shows the absorption spectra of the K. japonica and The intensity of the absorption spectrum is high for R. chinensis
R. chinensis dye solutions with and without sugar molecules. compared to K. japonica. Among the absorption spectrum of K.
Absorption peaks were observed for K. japonica between 400 and japonica the intensity is lower for the carotenoid pigment treated
500 nm without the addition of sugar, coinciding with the absorp- with sugar molecules than for the untreated dye. In the case of
tion spectrum obtained by Liu et al., 2008, for yellow chrysanthe- TiO2-adsorbed anthocyanin also, the intensity is low for the
mum [12]. The spectrum obtained between 400 and 500 nm sugar-treated dye compared to the bare dye (Fig. 3b). Although
confirms the extraction of the carotenoid pigment. According to the sugar has a positive effect on the stabilization of the anthocy-
Gao et al., 2000 [13], light absorption in the range 380–520 nm anin pigment, it does not improve the efficiency. The efficiency is
and molar extinction coefficients larger than 105 make certain slightly higher for the bare dye than for the sugar-treated dye for
carotenoids potential sensitizer materials for solar photovoltaic anthocyanin sensitized device, whereas for the carotenoid
cells and other artificial photochemical devices. Upon the addition sensitized the efficiency is slightly high for the sugar treated dye
of sugar molecules to the extracted dye, the intensity of the peak compared to the bare dye.
increases. This is because, as per the Beer–Lambert law, the
amount of light absorbed by a body is proportional to the number
of absorbing particles in it. The increase in the number of sugar IV characteristics of DSSCs sensitized with natural dyes
molecules in the pigment slightly increases the amount of visible
light absorption of the carotenoid pigment. Absorption peaks were Table 1 shows that the conversion efficiencies of the DSSCs fab-
observed for R. chinensis between 500–600 nm confirms the ricated using carotenoid pigment extracted from the flower petals
extraction of anthocyanin pigment. The intensity of the absorption of K. japonica treated with and without sugar are 0.29% and 0.22%,
spectrum of the dye solution of R. chinensis, increased with the the open-circuit voltages (VOC) are 0.5526 and 0.5839 V, the short-
addition of sugar, and the absorption wavelength was around circuit current densities (JSC) are 0.7509 and 0.5597 mA/cm2
550 nm higher than that of bare dye (Fig. 3a). Fig. 3b shows the (Fig. 4a), and the fill factors (FF) are 0.70 and 0.67, respectively.
308 K.V. Hemalatha et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 96 (2012) 305–309

Table 1 performance, whereas they enhance the performance of the carot-

Characteristics of the dye sensitized solar cell fabricated by the carotenoid and enoid sensitized cell.
anthocyanin pigment of K. japonica and R. chinensis respectively with and without
sugar molecule.
The cells sensitized with extracts of Blue pea (0.05%), Annatto
(0.19%), Norbixin (0.13%), Spinach (0.13%), Frutus lyciia (0.17%),
Sample VOC (V) JSC (mA/cm2) Fill factor Efficiency (%) Lily (0.17%) [14–17], and so on showed poor performances (less
K. japonica 0.5839 0.5597 0.6775 0.22 than 0.2%) compared to our dyes. Some of the pigments extracted
K. Japonica + sugar 0.5526 0.7509 0.7045 0.29 from Ipomoea (0.27%), Begonia (0.24%), Tangerine peel (0.28%),
R. chinensis 0.5433 0.8017 0.664 0.29
R. chinensis + sugar 0.5373 0.7025 0.6938 0.27
Marigold (0.23%), yellow rose(0.26%), China loropetal (0.27%),
Chinese rose (0.27%), Dragon fruit(0.22%) [16–18] and so on
showed similar efficiencies to our pigments. The fill factors of
natural dyes range from 20 to 70%, the short-circuit currents from
0.2 to 0.8 mA/cm2, and the open-circuit voltages from 0.2 to 0.6 V.
With our dye, we obtained a better fill factor (up to 70%) compared
with other natural dyes.
From the above results, it is seen that most natural-dye-based
DSSCs produce efficiencies of less than 1%. The main reasons for
the low efficiencies of natural dyes are the structure of the pigment,
the anchoring group in the natural dye (interaction of dye with
TiO2), and the stability. If the natural dye structure has long chains
with R groups such as xanthophylls, they hinder the bonding of
the pigment with the oxide surface of the TiO2, thus preventing
the dye molecules from becoming arrayed effectively on the TiO2
film. This results in less electron transfer from the dye molecules
to the conduction band of TiO2. In the present study, the carotenoid
pigments have long alkyl groups that hinder the arraying of dye
molecules on the TiO2 compared to the anthocyanin, so the caroten-
oid pigments showed low efficiency. Most natural dyes have OH and
oxygen ligands and lack carboxylic groups. The carboxylic group is
mainly responsible for anchoring, combining with the hydroxyl of
the TiO2 particles to produce ester groups and boosting the coupling
effect of the electrons in the TiO2 conduction band to give a rapid
electron-transport rate [19]. Carotenoid pigments do not have the
carboxylic group for effective anchoring, and this is the main reason
for their lower efficiency compared to the anthocyanin dye. Another
reason is their stability. Natural dyes are less stable than inorganic
dyes (N719). The low stabilities of both carotenoid and anthocyanin
lead to their very low efficiencies as sensitizers in DSSCs.

Conclusion

We have demonstrated the generation of photovoltaic currents

Fig. 4. Photocurrent–voltage characteristics of the DSSC using the natural dye of (a)
K. japonica and (b) R. chinensis.
Here, the efficiency, short-circuit current density, and fill factor of
the sugar-treated dye were increased slightly compared to those of
the bare dye, but the open-circuit voltages were similar. The con-
version efficiency of the DSSC fabricated from the anthocyanin
dyes of R. chinensis extract is 0.29%, VOC is 0.5433 V, JSC is
0.8017 mA/cm2, and FF is 0.66. Furthermore, on treating the ex-
tract with sugar, the conversion efficiency, VOC, and JSC are reduced
to 0.27%, 0.5373 V, and 0.7025 mA/cm2, respectively. The fill factor
is similar to that of the bare dye (0.69) (Fig. 4b).
It can be seen from Table 1 that the anthocyanin dye of R. chin-
ensis gives a better photoelectric conversion efficiency in the DSSC
without the sugar treatment. Since the extracted dye was a crude
mixture with co-adsorbents that were involved in enhancing the
performance of the dye, little variation is evident in the conversion
efficiency on treating the pigment with sugar. Usually, co-sensitiz-
ers are used to stimulate the cell performance. However, as far as
anthocyanin is concerned, the sugar molecules suppress the cell
upon the illumination of FTO electrodes coated with TiO2 nano-
crystals and sensitized with the pigments carotenoid and anthocy-
anin, which were extracted from the flowers of K. japonica and
R. chinensis, respectively, using a simple extraction technique.
The peaks at 1080 and 1646 cm 1 for the carotenoid molecule
are attributed to the symmetric vibrational modes of the C–C and
C@C stretching. The partial appearance of the quinonoidal form
of the anthocyanin molecule is confirmed by the peak at
1726 cm 1, which corresponds to the C@O stretching vibration.
The broad absorption peak of both the pigments in the visible
range indicates the adsorption of dye molecules to the TiO2 nano-
particles. The addition of sugar molecules to the carotenoid pig-
ment enhanced the short-circuit current from 0.559 to
0.750 (mA/cm2) and the fill factor from 0.677 to 0.705, whereas
the addition of sugar molecules to the anthocyanin pigment sup-
pressed the current density, and thereby caused a decrease in effi-
ciency. Although the fill factors for the extracted pigments
remained high, the current conversion efficiencies were very low
for both K. japonica and R. chinensis compared to N719 dye. The
efficiency could be improved further by the addition of different
stabilizing agents or binders to the pigments in which the research
is under progress. The simple extraction procedure, low cost, wide
availability, and environmentally friendly nature make such natu-
ral dyes promising alternative sources of sensitizers for DSSCs.
 K.V. Hemalatha et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 96 (2012) 305–309
Acknowledgment
The first author would like to thank the ‘‘2011 Post-Doc. Devel-
opment Program’’ of Pusan National University, Busan, South Korea
for its financial support.
References
[1] G.K.R. Senadeera, T. Kitamura, Y. Wada, S. Yanagida, Sol. Energy Mater. Sol.
Cells 88 (2005) 315–322.
[2] J. Preat, D. Jacquemin, E.A. Perpète, Environ. Sci. Technol. 44 (2010) 5666–5671.
[3] S. Furukawa, H. Iino, T. Iwamoto, K. Kukita, S. Yamauchi, Thin Solid Films 518
(2009) 526–529.
[4] G. Calogero, G.D. Marco, S. Cazzanti, S. Caramori, R. Argazzi, A.D. Carlo, C.A.
Bignozzi, Int. J. Mol. Sci. 11 (2010) 254–267.
[5] O. Emter, H. Falk, P. Sitte, Protoplasma 157 (1990) 128–135.
[6] U.G. Chandrika, in: T. Bechtold, R. Mussak (Eds.), Handbook of Natural
Colourants, John Wiley & Sons, Ltd., United Kingdom, 2009, pp. 221–236.
309
[7] F. Queiroz, C. Oliveira, O. Pinho, I.M. Ferreira, J. Agric. Food Chem. 57 (2009)
0712–10717.
[8] T. Polvka, V. Sundstrom, Chem. Rev. 104 (2004) 2021.
[9] I.R. Bunghez, M. Raduly, S. Doncea, I. Aksahin, R.M. Ion, Dig. J. Nanomater. Bios.
6 (2011) 1349–1356.
[10] T.L. Rand, J.Z. Deli, P. Molnar, G. Toth, Helvetica Chim. Acta 85 (2002) 1691–
1697.
[11] P. Luo, H. Niu, G. Zheng, X. Bai, M. Zhang, W. Wang, Spectrochim. Acta, Part A
74 (2009) 936–942.
[12] B.Q. Liu, X.P. Zhao, W. Luo, Dyes pigments 76 (2008) 327–331.
[13] F.G. Gao, A.J. Bard, L.D. Kispert, J. Photochem. Photobiol. A. 130 (2000) 49–56.
[14] K. Wongcharee, V. Meeyoo, S. Chavadej, Sol. Energy Mater. Sol. Cells 91 (2007)
566–571.
[15] N.M. Gomez-Ortiz, I.A. Vazquez-Maldonado, A.R. Perez-Espadas, G.J. Mena-
Rejon, J.A. Azamar-Barrios, G. Oskam, Sol. Energy Mater. Sol. Cells 94 (2009)
40–44.
[16] H. Chang, H.M. Wu, T.L. Chen, K.D. Huang, C.S. Jwo, Y.J. Lo, J. Alloys Compd. 495
(2010) 606–610.
[17] H. Zhou, L. Wu, Y. Gao, T. Ma, J. Photochem. Photobiol. A. 219 (2011) 188–194.
[18] A. Ram, N. Nayan, Int. J. Integr. Eng. 2 (2010) 55–62.
[19] M.R. Narayan, Renew. Sust. Energ. Rev. 16 (2012) 208–215.