Enhanced open-circuit voltage of p-type DSC with highly crystalline NiO nanoparticles

Article abstract of DOI:10.1039/c0cc05445c

Highly crystalline NiO nanoparticles synthesized through a facile thermolysis route were used as photocathode materials in p-type dye-sensitized solar cells resulting in open-circuit voltages exceeding 300 mV with an iodide/triiodide mediator. The Royal Society of Chemistry 2011.

Full text of DOI:10.1039/c0cc05445c

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 4808–4810
www.rsc.org/chemcomm
COMMUNICATION
Enhanced open-circuit voltage of p-type DSC with highly crystalline
NiO nanoparticlesw
a
a
a
a
a
Xiao Li Zhang,z Fuzhi Huang,z Andrew Nattestad, Kun Wang, Dongchuan Fu,
b
b
a
a
Amaresh Mishra, Peter Bauerle, Udo Bach* and Yi-Bing Cheng
¨
Received 8th December 2010, Accepted 1st March 2011
DOI: 10.1039/c0cc05445c
Highly crystalline NiO nanoparticles synthesized through a
facile thermolysis route were used as photocathode materials
in p-type dye-sensitized solar cells resulting in open-circuit
voltages exceeding 300 mV with an iodide/triiodide mediator.
Open-circuit voltages of up to 218 mV under full sun illumina-
tion were recorded, in conjunction with conventional iodide/
triiodide based electrolytes.
The maximum photovoltage produced by the photocathode
strongly influences the theoretical efficiency gain that can be
afforded by a tandem configuration. A straightforward strategy
towards the development of more efficient photocathodes
involves the replacement of NiO with alternative p-type
A number of tandem configurations for dye-sensitized solar
cells (DSCs) have recently been described with the aim
to boost the efficiencies of DSCs beyond their current limits.
One of these approaches involves the combination of a dye-
sensitized photoanode (n-DSC) and photocathode (p-DSC) in
a simple two-electrode sandwich configuration (pn-DSC). This
concept has first been described by Lindquist and co-workers
and represents the only two-terminal tandem DSC configura-
1
4
semiconductor materials with lower conduction band edges.
However, the choice of potential wide-bandgap p-type materials
with suitable band edge positions is very limited.
While the open-circuit potential is ultimately dependent on
the valence band edge position, the quality of the NiO used to
produce the mesoporous p-type films also strongly affects the
performance of p-DSCs. Here we show that the maximum
open-circuit potential of NiO-based p-DSCs can be significantly
improved to beyond 300 mV when highly crystalline NiO
materials synthesized via a novel synthetic route are used for
the construction of p-DSCs. This provides a new upper limit
for the theoretical efficiency limitation of NiO-based tandem
pn-DSCs. Especially when combined with a low-bandgap
NIR p-DSC sensitizers, this Voc improvement will be vital
for the development of highly efficient NiO p-DSCs based on
low-bandgap NIR sensitizers.
1
tion known to date. Here, a conventional photoanode, typically
consisting of a dye-sensitized mesoporous n-type TiO2
electrode is combined with a dye-sensitized mesoporous p-type
semiconductor electrode, composed of e.g., p-type NiO.
Photoexcitation of the sensitizer on the p-DSC results in
hole-injection into the valence band of the p-type semi-
conductor. The photoreduced dye is then regenerated by a
conventional iodide/triiodide electrolyte. The series connection
of both dye-sensitized junctions implies that the photocurrent
and energy conversion efficiency of the tandem device will
be limited by the weaker of the two photoelectrodes. Until
recently p-DSCs produced very limited photocurrent densities
Highly crystalline NiO nanoparticles were synthesized
through a high temperature thermolysis. Typically, nickel oxalate
and sodium chloride were ground into fine powders before
thermolysis at 950 1C for 2 hours. The thermolysis in molten
salt media provides a facile and efficient synthesis technique to
prepare nickel oxide nanoparticles in a large quantity. X-Ray
diffraction pattern (XRD) in Fig. 1(a) reveals the overall
crystal structure and phase purity of the products. All sharp
peaks match well with cubic rock salt structured NiO and
indicate high crystallinity of the particles.
ꢀ
2
and open-circuit voltages of typically o1 mA cm
and
2–7
o150 mV. Consequently, the poor photovoltaic performances
generally observed for p-DSCs represented a major impediment
for the realization of highly efficient pn-DSCs.
Recently the development of novel sensitizers for meso-
porous p-type NiO electrodes has led to strong improvements
8
–13
in the performance of dye-sensitized photocathodes.
In a
recent paper we could show that absorbed photon to charge
carrier conversion efficiencies of up to 96% can be observed in
12
p-DSCs based on a novel generation of donor–acceptor dyes.
The nickel oxide nanoparticles were analyzed by a field-
emission scanning electron microscopy (FESEM). In Fig. 1(b),
the as-synthesized NiO nanoparticles exhibit regular octa-
hedral shape with particle size range of 20–200 nm. The inset
selected area electron diffraction (SAED) rings well accord
with XRD patterns on the cubic NiO. An individual octahedral
NiO particle with smooth surfaces and straight arris is clearly
identified in Fig. 1(c), similar to the model cubic structure with
eight enclosed {111} facets, as shown in Fig. 1(d).
a
Department of Materials Engineering, Monash University,
Clayton 3800, Australia. E-mail: udo.bach@monash.edu
Institute for Organic Chemistry II and Advanced Materials,
University of Ulm, Albert-Einstein-Allee 11, 89081 Ulm, Germany
b
w Electronic supplementary information (ESI) available: Experimental
details and figures. See DOI: 10.1039/c0cc05445c
zw oD r rk .X iao Li Zhang and Dr Fuzhi Huang contributed equally to the
4
808 Chem. Commun., 2011, 47, 4808–4810
This journal is c The Royal Society of Chemistry 2011

View Article Online
Fig. 2 (a) Chemical structure of dye 3 and (b) photocurrent density–
voltage curve of dye-sensitized NiO photocathodes with and without
underlayer.
Fig. 1 (a) XRD pattern and (b) SEM observation of NiO nano-
particles with SAED pattern (inset). (c) SEM image of a single NiO
particle and (d) schematic illustration of an octahedral NiO crystal.
(
(
e) HRTEM image of NiO particle and corresponding SAED spot.
f) SEM images of cross sectional view of NiO film.
The p-type DSC with highly crystalline NiO nanoparticles and
no adhesion layer exhibited a markedly enhanced open-circuit
ꢀ
ꢀ
voltage (Voc) of 350 mV, using I /I
the recently reported NiO DSC using Co /Co . The current
3
which is comparable with
II
III 11
In Fig. 1(e), a high resolution transmission electron micro-
ꢀ
2
.
density (Jsc) value is however quite low, only 0.04 mA cm
scopy (HRTEM) image reveals a lattice spacing of 0.21 nm
which can be indexed to the lattice fringe of d200 of cubic NiO.
Furthermore, the HRTEM image clearly identified lattice
fringes and smooth surface (without any amorphous shell).
The clarity of diffraction spots in the inset and the corresponding
SAED pattern confirm the high crystallinity and defect free
native of the as-synthesized NiO nanoparticle.
Since the NiO nanoparticles were obtained at 950 1C, these
particles have considerable stability at lower temperature such as
the sintering temperature of 500 1C. Contacts between these
highly crystalline NiO nanoparticles and between the NiO film
and FTO glass formed at 500 1C are therefore very weak.
To improve contact, the FTO glass was pretreated by
dip-coating in a nickel acetate ethanolic solution (0.05 M)
Nickel(II) oxide is generally known as a non-stoichiometric
compound (Ni1ꢀxO with 0.003 o x o 0.17) with color varying
however, in our experiments, the octa-
5
times before screen printing NiO paste. During the sintering
15,16
from gray to black,
process, the underlayer of nickel acetate thermally decomposes
by the following reaction:
hedral NiO nanoparticles synthesized via thermolysis are light
green in color, which suggests that the as-synthesized NiO is close
to 1 : 1 stoichiometry with less nickel vacancies and holes.
Energy-dispersive X-ray spectroscopy (EDS) confirmed that
the NiO sample contains nickel and oxygen with a molecular
ratio of 1 : 1, with no trace of sodium or chlorine (Fig. S1, ESIw).
It has been reported that nanoparticles with less surface
defects can reduce the probability of recombination between
injected electrons (or holes) and the electrolytes, improving
Ni(CH COO) +4O - NiO+4CO +3H O
3
2
2
2
2
This will form a compact NiO blocking layer between NiO
film and FTO glass. The improvement was confirmed by
enhanced p-type NiO DSC performance.
With a compact underlayer, the NiO DSC shows a dramatic
ꢀ2
sc
increase on the current density (J ) to 1.32 mA cm with a
oc
slight drop in the open-circuit voltage (V ) to 305 mV. As a
1
7
DSC performance. The high crystalline stoichiometrical NiO
nanoparticles from the above preparation might be able to
reduce the recombination due to its less surface defects.
A finely dispersed colloidal solution was generated by
ball-milling the NiO nanoparticles. The paste for screen printing
was made by evaporating a fine mixture of the NiO colloidal
solution, ethyl cellulose and terpinol. The films were sintered at
result, the photovoltaic-conversion efficiency (Z) is significantly
increased to 0.14% under full sunlight illumination. Incident
photon-to-current efficiency (IPCE) spectra of the NiO DSC
are shown in Fig. S3 (ESIw). Photovoltaic performance
characteristics of the p-type NiO DSCs with and without
underlayer are listed in Table S1 (ESIw).
It could be conjectured that the sensitized underlayer could
ꢀ
2
add substantially to the photocurrent (from 0.4 mA cm to
5
00 1C for 1 h before soaking them in the dye solution. These
ꢀ
2
films were also examined by FESEM. Fig. 1(f) is a cross-sectional
view of the NiO film. The compact underlayer can be clearly
identified being approximately 100 nm thickness.
1.26 mA cm ) due to additional surface area for dye loading.
UV-vis absorption spectroscopy was used to determine the
absorption difference of dyed NiO films between with and
without underlayer (Fig. S4, ESIw). The small variation in
absorption suggests that the underlayer has negligible effect on
dye loading. The dramatic increase of current density is more
likely due to the improved contact between the NiO film
The chemical structure of dye 3 is shown in Fig. 2(a).
Fig. 2(b) shows the photocurrent-density vs. voltage behavior
of the NiO DSCs without and with compact underlayer
ꢀ2
measured at 100 mW m under AM 1.5 irradiation.
This journal is c The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 4808–4810 4809

View Article Online
and FTO glass. This was also confirmed by the following
electrochemical impedance spectroscopy measurements.
Compared with commercial NiO nanoparticles, the highly
crystalline octahedral NiO nanoparticles exhibited a signifi-
Hence, the initial photovoltaic efficiency was retained even after a
continuous 127 days aging test under above conditions.
A maximum open-circuit voltage (V ) of 320 mV and a
oc
solar to electric conversion efficiency (Z) of 0.14% had been
found during the stability testing. The reason of the increase
on the open-circuit photovoltage is probably due to the slow
diffusion of the electrolyte into inner pores of the NiO films
and the accompanied rearrangement of the adsorbed dye.
In summary, we demonstrated a facile thermolysis process
for producing large quantities of octahedral NiO nanoparticles
with high crystallinity. The highly crystalline NiO yielded an
1
2
cantly increased open-circuit voltage (V ) by about 85 mV.
oc
The high voltages presented by the octahedral NiO nano-
particles are probably due to the decreased recombination
paths owing to the low surface defect concentration of the high
crystalline particles. The current density (Jsc) is majorly limited
2
ꢀ1
by lower surface area of the NiO nanoparticle (10.288 m g
)
due to broad size range from 20 to 200 nm.
ꢀ ꢀ
In order to investigate charge transport and recombination
in the p-type NiO DSCs, electrochemical impedance spectro-
scopy (EIS) measurements were carried out as a function of
frequency measured at 1 sun with an applied potential bias of
Voc (Fig. S5, ESIw). From the Bode phase plots in Fig. S5(a),
the electron lifetime with the NiO DSC using underlayer
is 20 ms, a fourfold increase compared to the non-underlayer
NiO DSC (5 ms). In Fig. S5(b) and (c), the central arcs are
attributed to the impedance based on charge-transfer processes
open-circuit photovoltage (Voc) of 350 mV in a NiO, I /I
3
p-DSC system. A compact NiO underlayer was formed and
employed to improve charge transfer in p-type NiO DSCs. A
solar to electric conversion efficiency (Z) of 0.14% was
achieved. Further size selection of NiO nanoparticles is likely
to allow device improvement. Additionally, the p-type NiO
based devices have also shown excellent stability during a
long-term testing.
This work was financially supported by the ARC Centre
of Excellence for Electromaterials Science (ACES). The
authors acknowledge use of facilities within the Monash
Centre for Electron Microscopy and the Centre’s provision
of scientific and technical assistance. The authors would like to
acknowledge the ARC for providing equipment support
through LIEF (LE0883019).
ꢀ
ꢀ
occurring at the dye-sensitized NiO/dye/redox (I /I
3
) inter-
face. The radius of the intermediate-frequency semicircle in the
Nyquist plot decreased with using underlayer, which indicates
the improved electron (hole) generation and transport. This
result is also in good agreement with the trend of the overall
cell efficiency.
The stability of DSCs performance, including photochemical,
chemical, and physical stability of the DSCs, is another impor-
tant issue. Aging tests were performed to scrutinize the photo
stability of the p-type photocathode system employing the highly
crystalline NiO nanoparticles along with dye 3. Fig. 3 presents
the photo-voltaic data of the p-type DSC during aging at room-
temperature in the dark. The performance was remarkably stable
Notes and references
1 J. He, H. Lindstrom, A. Hagfeldt and S.-E. Lindquist, J. Phys.
Chem. B, 1999, 103, 8940.
2
A. Nakasa, H. Usami, S. Sumikura, S. Hasegawa, T. Koyama and
E. Suzuki, Chem. Lett., 2005, 34, 500.
3 S. Mori, S. Fukuda, S. Sumikura, Y. Takeda, Y. Tamaki,
E. Suzuki and T. Abe, J. Phys. Chem. C, 2008, 112, 16134.
4 H. Zhu, A. Hagfeldt and G. Boschloo, J. Phys. Chem. C, 2007, 111,
ꢀ
2
over this period, with only a 0.09 mA cm decay in Jsc, which
was compensated by an increase of 15 mV in Voc (320 mV).
1
7455.
Y. Mizoguchi and S. Fujihara, Electrochem. Solid-State Lett.,
008, 11, K78.
A. Nattestad, M. Ferguson, R. Kerr, Y.-B. Cheng and U. Bach,
Nanotechnology, 2008, 19, 295304.
L. L. Pleux, B. Chavillon, Y. Pellegrin, E. Blart, L. Cario, S. Jobic
and F. Odobel, Inorg. Chem., 2009, 48, 8245.
L. Li, E. A. Gibson, P. Qin, G. Boschloo, M. Gorlov, A. Hagfeldt
and L. Sun, Adv. Mater., 2010, 22, 1759.
5
6
7
8
9
2
P. Qin, H. Zhu, T. Edvinsson, G. Boschloo, A. Hagfeldt and
L. Sun, J. Am. Chem. Soc., 2008, 130, 8570.
1
0 P. Qin, M. Linder, T. Brinck, G. Boschloo, A. Hagfeldt and
L. Sun, Adv. Mater., 2009, 21, 2993.
1
1 E. A. Gibson, A. L. Smeigh, L. L. Pleux, J. Fortage, G. Boschloo,
E. Blart, Y. Pellegrin, F. Odobel, A. Hagfeldt and
L. Hammarstro
2 A. Nattestad, Y.-B. Cheng, A. J. Mozer, M. K. R. Fischer,
U. Bach and P. Bauerle, Nat. Mater., 2010, 9, 31.
3 A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo and H. Pettersson,
Chem. Rev., 2010, 110, 6595.
4 A. Nattestad, X. Zhang, U. Bach and Y.-B. Cheng, Journal of
Photonics for Energy, 2011, 1, 011103.
5 N. N. Greenwood and A. Earnshaw, Chemistry of the Elements,
Oxford Science Publication, Oxford, 1998.
6 K. Nakaoka, J. Ueyama and K. Ogura, J. Electroanal. Chem.,
¨
m, Angew. Chem., Int. Ed., 2009, 48, 4402.
1
1
1
1
1
1
¨
2
004, 571, 93.
7 J. Weidmann, T. Dittrich, E. Konstantinova, I. Lauermann,
I. Uhlendorf and F. Koch, Sol. Energy Mater. Sol. Cells, 1999,
56, 153.
Fig. 3 Variations of Voc, Jsc and efficiency (Z) as a function of the
life-time of the NiO photocathode.
4
810 Chem. Commun., 2011, 47, 4808–4810
This journal is c The Royal Society of Chemistry 2011