Advanced Materials - 2019 - Li - Low Cost Counter Electrode Materials For Dye Sensitized and Perovskite Solar Cells

REVIEW
Next-Generation Solar Cells www.advmat.de
Low-Cost Counter-Electrode Materials for Dye-Sensitized

and Perovskite Solar Cells
Guo-Ran Li* and Xue-Ping Gao*
Dye-sensitized solar cells (DSSCs) and all-

It is undoubtable that the use of solar energy will continue to increase. Solar solid-state perovskite solar cells (PSCs) are
cells that convert solar energy directly to electricity are one of the most considered as the most promising alterna-
convenient and important photoelectric conversion devices. Though silicon- tives or supplements for the conventional
based solar cells and thin-film solar cells have been commercialized, devel- solar cells, due to relatively high solar-
to-electricity conversion efficiency and
oping low-cost and highly efficient solar cells to meet future needs is still a
affordable cost.[3] In recent years, besides
long-term challenge. Some emerging solar-cell types, such as dye-sensitized issues of power conversion efficiency and
and perovskite, are approaching acceptable performance levels, but their stability, the important topic is reducing
costs remain too high. To obtain a higher performance–price ratio, it is neces- fabrication cost of the emerging solar cells
sary to find new low-cost counter materials to replace conventional precious by introducing new materials and techno
logies based on development of nanomate-
metal electrodes (Pt, Au, and Ag) in these emerging solar cells. In recent
rials and nanotechnology.[4,5]
years, the number of counter-electrode materials available, and their scope In a typical dye-sensitized solar cell
for further improvement, has expanded for dye-sensitized and perovskite (DSSC), dye-sensitized titania is used
solar cells. Generally regular patterns in the intrinsic features and structural as photoanode, iodide/triiodide couple
design of counter materials for emerging solar cells, in particular from an as a redox shuttle, and platinum-coated
electrochemical perspective and their effects on cost and efficiency, are transparent conductive oxide (TCO)
glass as counter electrode.[6] In principle,
explored. It is hoped that this recapitulative analysis will help to make clear
photoanode generates electrons under
what has been achieved and what still remains for the development of cost- irradiation, counter electrode collects elec-
effective counter-electrode materials in emerging solar cells. trons from external circuit, and the redox
shuttle realizes internal charge exchange
in the cells by reduction of triiodide ions
1. Introduction on the interface between electrolyte and counter electrode, and
simultaneous oxidation of iodide ions by holes on the photo-
Energy is the fundamental necessity for life and develop- anode-electrolyte interface.[6] However, the reduction of triiodide
ment of human beings. There is growing awareness that solar ions almost does not occur spontaneously on a bare TCO glass,
energy will assuredly become more and more important in the and a catalyst has to be involved in the counter electrode. As
future energy-generation structure. Developing low-cost, high- a result, electrical conductivity and catalytic activity of counter
efficiency, and clean solar energy technologies will be signifi- electrode materials strongly influence DSSCs on photoelectric
cant long-term interests. A solar cell can directly convert solar conversion performance.[6–9] Platinum is a competent counter
energy to electricity that is one of the most convenient energy electrode material due to its good electrical conductivity and
forms in modern society.[1] The ongoing concerns for the appli- excellent electrocatalytic activity for reduction of triiodide ions.
cation of solar cells are cost performance, that is, to achieve However, as it is well known, platinum is one of the scarcest
both higher power conversion efficiency and lower cost. After resources with expensive price, which is obviously disadvan-
conventional Si-based and thin film solar cells, some emerging tageous for improving the cost performance of a DSSC.[8,9]
solar cells have attracted more and more attention due to Moreover, it is doubtable that platinum can keep a long stability
more room for the improvement of cost performance ratio.[2,3] because a complexation reaction between platinum and iodides
can occur under light irradiation.[8,9] In addition, platinum
is not advantageous in the case of flexible electrodes due to its
Prof. G.-R. Li, Prof. X.-P. Gao
Institute of New Energy Material Chemistry metallic brittleness. Therefore, it is very necessary to seek for
School of Materials Science and Engineering more low-cost and high-efficient alternatives to counter mate-
Renewable Energy Conversion and Storage Center rials for development of DSSCs. Actually, in the last decade,
Nankai University the horizon of low-cost counter electrode materials have widely
Tianjin 300350, China
E-mail: [email protected]; [email protected]
expanded from carbonaceous materials to many kinds of com-
pounds, as shown in some published reviews.[9–13]
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adma.201806478. A typical perovskite solar cell (PSC) usually has a solid-state
five-layer structure.[14,15] The core layer employs perovskite
DOI: 10.1002/adma.201806478 CH3NH2PbI3 as active material for generation and separation
Adv. Mater. 2020, 32, 1806478 1806478 (1 of 20) © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
15214095, 2020, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.201806478 by UFSJ - Universidade Federal de Sao Joao del Rei, Wiley Online Library on [13/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de
of electron–hole pairs under irradiation. A compact titania

electron-transport layer and a 2,2′,7,7′-Tetrakis[N,N-di(4-meth- Guo-Ran Li is a full pro-
oxyphenyl)amino]-9,9′-spiro-bifluorene (sprio-OMeTAD) hole- fessor at the Institute of New
transport layer, respectively, are overlayed on both sides of the Energy Material Chemistry,
core perovskite layer. A transparent fluorine-doped tin oxide Nankai University, China.
(FTO) glass supports the electron-transport layer to transfer His main current research
electrons to external circuit. On the other side, an Au or Ag interest is the study of func-
layer as counter electrode is next to the hole-transport layer tional materials for energy
to collect the electrons from external circuit that recombines generation, and in particular
with holes on the interface between counter electrode and for hybrid solar cells and
hole-transport layer. In the devices, the conventional counter rechargeable batteries. He
electrode materials Au or Ag have excellent electrical conduc- received a Bachelor’s degree
tivity and matched work function with the light absorber and in chemistry in 2000 and a
hole transport materials. However, Au and Ag as expensive Ph.D. degree in physical chemistry in 2005, both from
materials must be a limited for large-scale application of PSCs Nankai University.
in the future. Moreover, the metal layer is usually non-trans-
parent, which makes it impossible to use in a transparent or Xue-Ping Gao is a professor
bifacial solar cells. In recent years, some low-cost materials at the Institute of New
such as carbon, non-precious metal, and polymers have been Energy Material Chemistry,
investigated as an alternative to the precious metals, providing Nankai University, China. He
one potential approach to economic PSCs.[16] received his doctorate from
In this review, we summarize the research progress of low- the Department of Chemistry
cost counter electrode materials for two of the most promising of Nankai University in 1995.
emerging solar cells, DSSCs and PSCs. The contrast observa- Currently, his main research
tion and insight analysis are helpful for a better understanding is focused on energy storage
of commonness and otherness in low-cost counter electrode materials for power sources,
between DSSCs and PSCs. On the other side, researches on including Li-ion batteries, Li-S
low-cost counter electrode for PSCs are still on the initial state batteries, and solar recharge-
as compared with knowledge on those for DSSCs. Thus, it is able batteries.
possible to draw on the experience from the strategy of counter
electrode in DSSCs for developing low-cost counter electrode
materials for PSCs. Therefore, linking with previous works on simulated Nyquist plot can be obtained as shown in Figure 1c.
low-cost counter electrode materials for DSSCs and PSCs, we The Rct, Rr, and ZN are in inverse relation to current density at
follow with interest some of the following aspects in this review: counter electrode, photoanode side, and electrolyte, respectively,
the intrinsic requirements on electrochemistry and physics of and the final performance of DSSC depends on the part that
counter electrodes for DSSCs and PSCs, the expanded mate- has the lowest current density. Thus a qualified counter elec-
rial horizon of low-cost counter electrodes, and the efforts in trode should provide a comparable exchange current density
nanomaterials and nanotechnologies for improvement of the with that from photoanode. On counter electrode, the exchange
final performance. We hope to refine some generally regular current density (j0) actually refers to the rate of triiodide ion
patterns about intrinsic feature and structure design of low- reduction, which can be described using Equation (1):
cost counter materials for the emerging solar cells, and reveal
the internal relationships among electrochemical performance, RT
j0 = (1)
hierarchical structure and geometrical morphologies, and mate- nFRct
rial characteristics.
in which R is the gas constant, T is the absolute tempera-
ture, n is the number of electrons being transferred, and F
2. Intrinsic Features of Counter Electrodes is the Faraday constant. Obviously, Rct is the rate-determining
parameter for counter electrode. As mentioned above, the
in DSSCs and PSCs
reduction of triiodide ions on counter electrode is not a spon-
2.1. Counter Electrodes in DSSCs taneous reaction. This means that the reduction rate of trii-
odide ions is controlled by electrocatalytic ability of counter
A typical DSSC can be expressed in electrics and electrochem- electrode materials. In the other words, the intrinsic electro-
istry using the equivalent circuit shown in Figure 1a.[17] Rct and catalytic activity of counter electrode materials can be evalu-
Rr are charge-transfer resistance at the catalyst/electrolyte inter- ated using Rct.
face and at the oxide/electrolyte interface, respectively. Cdl and In researches on counter electrode materials, EIS are usu-
Cμ refer to double-layer capacitance and nanocrystalline oxide ally measured with a symmetrical configuration shown in
capacitance, respectively. ZN is electrolyte diffusion impedance, Figure 1b, in order to avoid any possible perturbations from
and Rs is sheet resistance. The parameters are usually meas- other parts in a full DSSC.[18] A typical Nyquist plot can be
ured using electrochemical impedance spectra (EIS), and a obtained as shown in Figure 1d, in which the semicircle in
Figure 1. Schematic representation of a) the impedances in DSSCs and the equivalent circuit and Nyquist plot (c), b) a symmetric cell and equivalent
circuit and Nyquist plot (d), and e) an energy-level diagram of a DSSC (Rs: sheet resistance; Rct and Rr: charge-transfer resistance at the catalyst/
electrolyte interface and at the oxide/electrolyte interface, respectively; ZN: electrolyte diffusion impedance; Cdl and Cµ: double-layer capacitance and
nanocrystalline oxide capacitance, respectively). (a) and (b) are reproduced with permission.[17] Copyright 2015, WILEY-VCH.
the high-frequency region corresponds to charge-transfer pro- a counter electrode with relatively low energy level is propi-
cess at the interface between electrolyte and counter electrode, tious to improve output voltage of DSSC. However, the Fermi
and the semicircle in the low-frequency region is assigned to level of counter electrode cannot be lower than redox potential
the Nernst diffusion process of triiodide ions. A low Rct means of the mediator. Otherwise, the electron transfer from counter
high intrinsic electrocatalytic activity, and a low ZN implies a electrode to the redox mediator cannot proceed to complete a
fast diffusion process. Besides fabrication route of electrode, full charge transport process from external circuit to internal
geometrical morphology and microstructure of active materials electrolyte. In practice, Fermi level of counter electrode is
have a great impact on the electrocatalytic activity and diffusion always higher than the redox potential of the mediator due
characteristics on counter electrode. to the inevitable electrode polarization. The polarization of
Electrocatalytic activity of counter electrode is a key factor counter electrode involves electrochemical polarization, ohmic
for influencing the current density of DSSC, at the meantime, polarization, and concentration polarization. The electrochem-
energy level of counter electrode is one of the two determi- ical polarization depends highly on electrocatalytic activity
nants of output voltage of DSSC. As shown in Figure 1e, the toward reduction of triiodide ions on counter electrode; ohmic
output voltage of a DSSC is determined by the energy differ- polarization is inversely proportional to electrical conductivity
ence between potential of iodide/triiodide redox couple and of counter electrode; and concentration polarization is mainly
Fermi level of counter electrode.[9,13] Fermi level of counter affected by migration rate of triiodide ions in electrolyte.
electrode is lower as compared with conductive band edge However, if a counter electrode material has strong adsorp-
of TiO2, so that photo-generated electrons can be passed to tion to triiodide ions, it can undoubtedly reduce concentration
counter electrode through the external circuit. This means that polarization.
2.2. Counter Electrodes in PSCs and HTM. Obviously, a low RBC means reduced charge-
transfer resistance at the interface between counter electrode
The equivalent circuit model of a typical perovskite solar and HTM, which is helpful for improving power conversion
cell, as shown in Figure 2a, can be applied to EIS analysis efficiency of PSCs.[19] Different from the counter electrode
for this type of devices.[19] Rx and Cx refer to resistance and process in DSSCs occurring at solid/liquid interface with
capacitance associated with perovskite, respectively. RHTM catalytic behavior, the counter electrode process in PSCs is
and CHTM are charge transport resistance and capacitance actually the recombination process of holes from HTM and
of hole-transport material (HTM), respectively. RBC and CBC electrons collected by counter electrode from external circuit
are charge transfer resistance and capacitance at the inter- at a solid/solid interface. In essence, the solid–solid interface
face between back contact (BC) or counter electrode and appears as metal–semiconductor contact, in which HTM is
HTM, respectively. The Nyquist plot of a typical solid-state a p-type semiconductor, and counter electrode materials are
perovskite-based device at 550 mV forward bias is presented usually consisted of metals or analogue.
in Figure 2b. The high-frequency response corresponds to To keep a low RBC value, a counter electrode for PSCs firstly
the back contact at the interface between counter electrode should have high electrical conductivity to reduce its own
ohmic resistance and support rapidly collecting electrons from
external circuit. More importantly, a suitable energy level or
work function is necessary for the workable counter electrode
materials. The work function corresponds to the energy differ-
ence between Fermi energy and vacuum level, and represents
the energy barrier to free space that prevents electrons at the
Fermi level from escaping the solid.[20] Therefore, it seems that
a low work function is helpful for electron output from counter
electrodes. However, for the metal–semiconductor contact
between counter electrode and HTM, Schottky barrier can
be generated when the counter electrode material has a higher
Fermi level (lower work function) than Fermi level of the p-type
semiconductor as HTM.[21] Electron transport can be blocked
or has to get over the barrier height with output voltage loss.
On the other hand, if Fermi level of counter electrode is lower
than that of valence band of HTM (corresponding to higher
work function), the holes as positive charges from HTM
cannot be transferred to counter electrode. In short, work-
able counter electrode materials needs a suitable energy level
region. As shown in Figure 2c, in principle, the valence band
and conduction band edges of perovskite CH3NH3PbI3 are at
−5.4 and −3.9 eV, respectively. A Fermi level, slightly higher
than the valence band edge of the perovskite, is workable for
the transport of holes. The typical HTM sprio-OMeTAD has
a valence band of −5.2 eV. Therefore, Au with a Fermi level
of −5.1 eV is an ideal choice as counter electrode for the typ-
ical PSCs. The conventional counter electrode Pt in DSSCs,
with a Fermi level of −5.65 eV, is obviously not applicable to
PSCs.
In practice, the influence of counter electrode materials is
much more complicated than the above theoretical analysis.
The electrical properties of metal-semiconductor contact, like
RBC, not only depend on work function of metal, but also on
surface states of semiconductor and doping concentration of
semiconductor.[20] As seen in many works, PSCs using Au and
Ag as counter electrode, respectively, show comparable power
conversion efficiency, even if the difference in work function
(Fermi level) between the two metals is up to 0.84 eV. Currently,
what one knows is that the workable Fermi level range for
Figure 2. Schematic representation of a) the impedances in PSCs and the counter electrode materials is −4.1 to −5.1 eV.
equivalent circuit and Nyquist plot (b), full model of a solid-state device In addition, to keep a low interface resistance RBC between
with perovskite (Rx/Cx, resistance and capacitance associated with perov-
counter electrode and HTM, besides electrical conductivity and
skite; RHTM/CHTM, charge transport resistance and capacitance of the
HTM; RBC/CBC, charge transfer resistance and capacitance at BC/HTM), work function of counter electrode materials, compatibility of
and c) energy levels of DSSC. (a) is reproduced with permission.[19] Copy- counter electrode materials and hole-transport materials is also
right 2014, American Chemical Society. one of the important factors.
3. Current Developments in Counter Electrode

Materials for DSSCs
A large number of materials have be investigated as counter
electrode for DSSCs, especially in the last decade.[22–39] Due to
the difference in device fabrication and measurement, it is not
reasonable to evaluate a counter material only by power con-
version efficiency of the corresponding device. As analyzed in
Section 2.1, the electrocatalytic activity and electrical conduc-
tivity of counter electrode determine its final performance.
Further, the electrocatalytic activity depends on exchange cur-
rent density of the counter electrodes. For a recapitulative
comparison, we summarize current density (j0) of some repre-
Figure 3. Current density (j0) of some representative counter-electrode
sentative counter electrode materials in Figure 3 based on the materials based on theoretical electrical conductivity (σ) of the materials.
theoretical electrical conductivity (σ) of the materials. It should j0 is calculated by Equation (1) and Rct is taken from the following refer-
be pointed out that a practical sheet resistance of counter elec- ences: metal and alloys: Refs. [22,23], carbon materials: Refs. [24–26],
trodes is more workable to reflect electrical conductivity of conductive polymers: Refs. [27–29], nitrides: Refs. [30,31], carbides:
counter electrodes than the theoretical values. However, sheet Ref. [31], oxides: Refs. [32–34], sulfides: Ref. [35], selenides: Refs. [36,37],
resistance depends heavily on electrode preparation technology, multiple compounds: Refs. [38,39].
leading to a variety for the same materials. Actually, a material
with high electrical conductivity can usually become a highly the position of the d-band center and the nature of the 3d ele-
conductive electrode by suitable preparation methods. In ment is linear, and Pt3Ni and Pt3Co provide d-band center very
Figure 3, the used electrical conductivity values are still efficient close to Pt. In the paper,[23] two crucial parameters including
to judge the different kinds of counter electrode materials. electronegativity and lattice constants of Pt and transition
In general, the electrical conductivity of the counter electrode metals are compared, yielding 2.28/3.92, 1.83/2.87, 1.88/2.51,
materials follows the order of metals and alloys > metal nitrides, 1.92/3.52, 1.90/3.61, 2.16/3.15, and 1.66/2.88 Å, for Pt, Fe, Co,
metal carbides > carbon materials > multiple compounds Ni, Cu, Mo, and Cr, respectively. The higher electronegativity of
(CuInS2 etc.) > conductive polymers, selenides > sulfides > transition metals can make an electron deviate from transition
oxides. Undoubtedly, the higher electrical conductivity of counter metal to Pt, leading to an electron redistribution on the surface
electrode materials is, the lower the energy loss on counter elec- of Pt species for providing more active catalytic sites. Gener-
trodes is. However, it can be observed from Figure 3 that electro- ally, the alloys of Pt and VIII metals or only VIII metals (such
catalytic activity of the counter electrodes has not a consequent as CoNi, and FeNi) show improved electrocatalytic activity.[41,42]
correlation with electrical conductivity. This provides the possi- Second, component proportion of alloys has a decisive effect
bility that one can individually consider electrocatalytic activity or on electrocatalytic activity. In the case of Co–Pt alloys, the elec-
electrical conductivity for fabrication of counter materials. In the trocatalytic activity is obviously improved by introducing very
next section, we place particular emphasis on how the different small amount of Pt into Co (Figure 4b).[43] CoPt0.02 exhibits
kinds of materials are improved in electrocatalytic activity. amazing electrocatalytic activity for the reduction of triiodide
ions, improving the corresponding DSSC to a power conver-
sion efficiency of 10.23% in comparison with that of 6.52%
3.1. Metals and Alloys using pure Pt. In the case of Pt–Ni alloy, Pt3Ni also shows a
higher performance than Pt.
In general, catalytic activity of a material depends on catalytically Third, it is concluded that absorption–desorption kinetics
active sites and adsorption–desorption kinetics. Compared with for triiodide and iodide ions on alloys is important in the elec-
single metal, an alloy can alter availability of active surface sites trocatalytic process. It is shown in Figure 4c that the absolute
and adsorption-desorption process of reactants, intermediates, energy (Eabs), which is defined as the difference between Ede
and products. Moreover, a manipulation of electronic structure and Ead, has a reverse trend with electrochemical performance
in alloys can also influence intrinsic catalytic ability. Some low Pt of the counter electrodes in DSSCs.[23] According to the EIS
and Pt-free alloys have been investigated and reviewed by Q. W. analysis in Section 2.1, the faster adsorption of triiodide ions
Tang etc.[22] As electrode materials, metals and alloys are obvious and desorption of iodide ions on alloys, the smaller the diffu-
different from porous materials with high specific surface area. sion impedance on the counter electrodes. Therefore, it can
This makes surface properties and electronic structure of metal improve electrocatalytic activity of alloys by adjusting the abso-
and alloys as counter electrodes more important. In general, sev- lute energy to minimization.
eral interesting and important suggestions from the researches
on alloy counter electrodes should be paid attention to.
First, VIII metals, Co and Ni, have comparative advantage 3.2. Carbon Materials
due to similarity with Pt. The surface electronic properties of
Pt3M (MFe, Co, Ni, Ti, and V) are characterized using syn- Carbon materials are one of the earliest low-cost counter elec-
chrotron-based high-resolution ultraviolet photo emission trode materials investigated as alternative to Pt. Based on the
spectra (Figure 4a).[40] It is shown that the relation between different hybridized structures, sp, sp2, and sp3, carbon gives
rise to many stable allotropes. This makes that the family of low-cost counter electrode materials. Generally, carbon counter
carbon materials possesses a large proportion in the horizon of electrodes have sufficient conductivity and high corrosion resist-
ance. However, intrinsic electrocatalytic activity of carbon mate-
rials is much lower than that of Pt.[24–26,43,44] This depends on the
essence of carbon, and has nothing to do with the various allo-
tropes. In fact, it can be found from Figure 5a that there are not
any clear correlations between the final electrochemical perfor-
mance and carbon allotropes. In the other words, the different
carbon allotropes show various electrocatalytic activities, which
is not because of different catalytic ability of active sites, but dif-
ferent amount of active sites and diversity of ion diffusion in
the counter electrodes. In general, the amount of active sites is
proportional to active area of electrodes or specific surface area
of the electrode materials. Figure 5b shows function of power
Figure 4. a) The d-band center position obtained from the valence- Figure 5. a) Summary of the efficiencies of different carbon-based DSSCs.
band spectra of Pt3M alloy surfaces (sputtered surfaces) and Pt-skin Reproduced with permission.[17] Copyright 2015, WILEY-VCH. b) Function
surfaces relative to the Fermi level. Reproduced with permission.[40] of power-conversion efficiency of DSSCs and exchange current density (j0)
Copyright 2007, Nature. b) The current density (j0) of the alloying with surface area of eight kinds of carbon material:[44] fullerene (C60), carbon
component.[43] c) A calculation of adsorption energy of triiodide ions, black (Cb), activated carbon (Ca), conductive carbon (Cc),carbon fiber (Cf ),
desorption energy of iodide ions , and corresponding absolute energy carbon dye (Cd), synthesized well-ordered mesoporous carbon (Com), and
with conventional Pt, Pt3Ni, PtNi, and PtNi3 counter electrodes. carbon nanotubes (Cn), in order of surface area. The discarded toner of a
Reproduced with permission.[23] Copyright 2014, WILEY-VCH. printer (Cp) in the original Ref. [44] is excluded since it is not pure carbon.
conversion efficiency of DSSCs using carbon counter electrodes 3.3. Conductive Polymers
or exchange current density of the counter electrodes with spe-
cific surface area of the used eight kinds of carbon materials. Conductive polymers provide charge conduction through a con-
Clearly, there is a positive correlation between power conver- jugated bond system along polymer backbone, and usually have
sion efficiency of DSSCs or exchange current density of couther good electrical conductivity in the range from a few S cm−1 to
electrodes and surface area of the used different kinds of carbon 500 S cm−1 in the doped state.[53,54] Since conductive polymers
materials. This suggests that the key to improve electrocatalytic have electrocatalytic activity for reduction of triiodide ions like
activity of carbon counter electrodes lies in increasing surface Pt, some attractive conducting polymers such as polyaniline
area of carbon materials. On the other hand, the different mor- (PANI), polypyrrole (PPy), poly(3,4-ethylene dioxythiophene)
phologies such as 1D carbon nanotubes, 2D graphene, and 0D (PEDOT), and other polythiophene (PTh) derivatives, have
porous carbon have an important effect on diffusion kinetics. been broadly investigated as counter electrode materials for
Actually, features of carbon materials is helpful for forming DSSCs.[51] As known well, the same kind of conductive poly-
mass-transfer networks easily in the electrodes to reduce diffu- mers can have great difference in properties because of diverse
sion impedance in the porous electrode. More discussion about polymerization degrees and doped states, causing the dispersive
this point will be continued in Section 3.6. and inconsistent results in evaluation of conductive polymer
counter electrodes in DSSCs. Figure 6a shows power conver-
sion efficiency of the DSSCs using PPY, PEDOT, PPrDOT,
3.2.1. Carbon Nanotubes and PANI as counter electrodes, respectively, based on col-
lecting data from the review.[51] Despite the dispersive results
1D tubular structure is the strongest characteristics of carbon for the same kind of conductive polymers, it can be found
nanotubes (CNTs). In view of counter electrodes, the large sur- that PProDOT and PEDOT should be the preferred materials
face area of CNTs can provide much electrocatalytic active sites, by comparison of the average values and the highest perfor-
good electrical conductivity favors fast electron transport, and mance. According to the average performance of the conduc-
the tubular structure is inclined to form network-like electrode tive polymers, the statistical result presents the following order:
for fast ion transport.[45,46] According to the amount of graphene PProDOT > PEDOT > PANI > PPY, in accordance with the
sheets in the walls of carbon nanotubes, there are single-walled, result investigated by Bay et al.[55] Further, it can be supposed
double-walled, or multiwalled carbon nanotubes (SWCNT, that intrinsic electrocatalytic activity of conductive polymers
DWCNT, or MWCNT), respectively.[45] In the three types of have something to do with the monomer structure. PProDOT
CNTs, DWCNTs show the highest electrocatalytic activity for and PEDOT have similar monomer framework structure, but
reduction of triiodide ions, because DWCNTs have a higher obvious difference from PPY and PANI (Figure 6a).
surface area of 450 m2 g−1 than SWCNTs of approximately Conductive polymers tend to form 3D network morphology
400 m2 g−1. The large surface area and the high chemical due to crosslink of bonds in the polymerization process. This
stability of DWCNTs improve the charge-transfer reaction. makes conductive polymers form a high quality and very thin
Interestingly, vertically aligned single-walled and multi- film electrode more easily than other electrode materials. As
walled carbon nanotubes with various lengths have been shown in Figure 6b–e, the porous and cross-linked structure is
investigated as counter electrodes, respectively, by different beneficial for not only fast charge transport, but also enough
groups.[47,48] Obviously, the highly ordered arrays are helpful contact with electrolyte. Moreover, the morphology and struc-
for diffusion and transport of ions to achieve an optimized ture of conductive polymers can easily tuned by controlling
electrode kinetics. EIS measurements result in lower interface preparation conditions.[52] SEM images in Figure 6b–e show the
resistance between the electrolyte and the arrays.[47,48] influence of the deposition charge capacity on the conforma-
tion of the PProDOT-Et2 films. The PProDOTEt2 film shows a
dense conformation at 20 mC cm−2, and then a porous struc-
3.2.2. Graphene ture emerged with a further increase to 40 mC cm−2. When the
deposition charge capacity increases to 80 mC cm−2 and larger
In theory, perfect graphene nearly provides electrocata- pores present in the films.[52] In the case, the dense appear-
lytic activity for reduction of triiodide ions. However, some ance and presence of small pores in the films are better for
reduced graphene oxides and graphene with some defects improving electocatalytic activity for reduction of triiodide ions.
can be active for catalyzing the counter electrode reaction. It
is supposed that defects and oxygen-containing groups such
as hydroxyl groups (OH), carbonyl groups (>CO), and car- 3.4. Metal Compounds
boxyl groups (COOH) are catalytically active sites.[49] In the
work, increasing the number of functional groups leads to an 3.4.1. Nitrides and Carbides
improvement in electrocatalytic activity.[49] Generally, it can be
considered that graphene counter electrodes for DSSCs actually Transition-metal nitrides and carbides belong to interstitial
have to depend on functionalized treatment of graphene. How- intermetallic compounds, and are generated by nitrogen and
ever, the special sheet structure can make graphene become an carbon atoms filling into interstitial voids in a metal lattice,
excellent matrix material for supporting highly active catalytic respectively.[56] The special structure makes interstitial com-
materials. In addition, graphene electrodes possibly have poten- pounds provide excellent electrical conductivity near to that of
tial advantage as transparent electrodes.[50] parent metals. Moreover, the infilling of nitrogen or carbon
Figure 6. a) Power-conversion efficiency of DSSCs using different types of conductive polymers as counter electrodes.[51] b–e) SEM images of PProDOT-
Et2 films with various deposition charge capacities; respectively, 20, 40, 80, and 120 mC cm−2. Reproduced with permission.[52] Copyright 2009, Elsevier.
atoms can change electronic structure of transition metals, with the device using a conventional Pt/FTO counter elec-
resulting in that metal nitrides and carbides provide catalytic trode. Subsequently, a similar TiN counter electrode with hier-
activity like the noble metals.[56] In 2009, our group successfully archical micro/nano TiN spheres supported on Ti foil is also
prepared highly ordered TiN nanotube arrays by the anodiza- proved to be effective.[58] Ma group investigated systematically
tion of Ti foil and subsequent nitridation in a NH3 atmosphere, many kinds of transition metal nitrides and carbides as counter
and used the metal nitride as counter electrode in DSSCs electrode materials.[31] The results are illustrated in Figure 7a.
for the first time.[57] The DSSC using TiN counter electrode In general, the metal nitrides show better performance than
achieves a power conversion efficiency of 7.73%, comparable the corresponding carbides, and the titanium/molybdenum
nitrides have outstanding competitive advantage among all the for reduction of triiodide ions, but the final power conversion
tested samples. efficiency of the corresponding DSSSs is lower than that of
In the investigation of MoN, WN, and Fe2N, it is found that device using Pt (Figure 7b,c).[30] The results are attributed to
MoN and WN have higher intrinsic electrocatalytic activity large adsorption impedance of iodine and triiodide ions and
Nernst diffusion impedance of triiodide ions for the metal
nitride counter electrodes, as shown in Figure 7c. This means
that there is much room for improvement of the metal nitride
electrodes by reducing mass-transfer impedance and electro-
chemical polarization, since that provides superior electrocata-
lytic activity and excellent electrical conductivity. As mentioned
above, TiN nanotube arrays and hierarchical micro/nanoTiN
spheres supported on Ti metallic foil actually improves diffu-
sion kinetics of the electrodes due to the structure feature.[57,58]
Also, the NiN electrode prepared by surface nitridation of Ni
foil shows a good performance, because the good distribution
of catalytically active sites on the surface decreases to a certain
extent importance of diffusion.[59]
In general, mass transport kinetics depend on the structure
and texture of electrodes. Enhancement of diffusion kinetics in
porous MoN nanorods-based counter electrode in DSSCs has
been displayed in the precious work.[60] In comparison with
MoN nanoparticle electrode (Figure 8a–d), the porous MoN
nanorod electrode shows higher performance, attributed to the
different mass transport kinetics. As illustrated in Figure 8e,
the unique morphology of the porous MoN nanorods produces
a larger porosity and a better interconnected electrolyte distri-
bution in the electrode film of the MoN nanorod electrode. As
for the MoN nanoparticle electrode, the sphere-like MoN nano-
particles are apt to closely pack, leading to a smaller porosity
and a poor electrolyte distribution with a higher tortuosity.
By analysis of surface area, it is concluded that the observed
enhancement of diffusion kinetics for the MoN nanorod elec-
trode is basically derived from the interconnected porous
texture in the electrode film. Actually, this provides an effective
strategy to solve the ion diffusion problem for nitride and car-
bide counter electrode materials by designing and optimizing
electrode structure.
Transition metal nitrides and carbides have very high simi-
larity. Interestingly, nitrogen and carbon atoms can be filled in
interstitial voids together to form carbon–nitrogen compounds.
An ordered mesoporous titanium nitride–carbon nanocom-
posite is used as counter electrode material for DSSC, and
obtain an efficiency of 8.41%, even higher than that using Pt
counter electrode. Perhaps, the co-existence of nitrogen and
carbon atoms is in favor of enhancing electrical conductivity
and increasing active sites of the metal compounds.
3.4.2. Sulfides and Oxides
Wang and Gratzel et al. introduce metal sulfides into counter

electrode for DSSC for the first time.[61] It is revealed that a
semi-transparent CoS film on ITO/PEN has a large exchange
current density that is comparable to the conventional Pt elec-
trode. Metal sulfides present various morphologies on nanom-
Figure 7. a) The PCEs of the DSSCs with different transition metal nitrides
eter and micrometer scale with different preparation routes.
and carbides as counter electrodes.[31] b) I–V characteristic curves of
DSSCs with different nitride counter electrodes. c) Nyquist plots of the As shown in Figure 9, many transition metal sulfides with
symmetric cells with two identical nitride counter electrodes. Reproduced different morphologies have been investigated as counter elec-
with permission.[30] Copyright 2011, Royal Society of Chemistry. trode materials for DSSCs.[62–67] The various morphologies with
Figure 8. a,b) SEM surface and cross-sectional images of a MoN NR-Ti electrode and c,d) MoN NP-Ti. Schemes of the electrode structure and the
electrolyte distribution of the porous MoN nanorod electrode and the MoN nanoparticle electrode (e). Reproduced with permission.[60] Copyright 2014,
Royal Society of Chemistry.
distinguished surface area are positive for improving electro- MoS2lieon the edges of the typical layered material, but not on
catalytic activity for reduction of triiodide ions. Therefore, it the basal planes.[68]
is not easy to find a general relation between electrochemical Compared to metal alloys and metal interstitial compounds,
performance and sulfide sorts or morphologies. However, most of metal sulfides belong to covalent compounds with low
in essence, the electrocatalytic activity depends on amount electrical conductivity and relatively low light reflection coeffi-
of active sites in the metal sulfides. As an example, Lei et al. cient. The features make metal sulfides more suitable for trans-
investigated graphene-like few-layered MoS2sheets as counter parent electrodes. On the one side, a very thin metal sulfide
electrode material.[68] Based on comparison with multi-layered film favors transport of charges. On the other side, the enough
MoS2 and nanoparticles with distinguished ratios of edge plane thin thickness is helpful for improving light transmittance of
to basal plane, it is concluded that the catalytically active sites of electrodes. We have developed a general strategy to prepare
Figure 9. SEM images of metal sulfides used as counter electrode materials in DSSCs. Reproduced with permission.[62–66] Copyright 2013, 2016,
Elsevier, 2012 Royal Society of Chemistry.
transparent and high-efficient metal sulfide counter electrodes these selenides, Co0.85Se and Ru0.33Se have outstanding per-
by an organic solution route to reach the balance between formance, even superior to Pt electrodes. In addition, NiSe2,[75]
optical transparency and electrochemical activity.[35] Among CoSe,[76] and NbSe2[77] also show acceptable performance as
the Group VIIIB metal sulfides, the nickel sulfide film with the reported in the literatures.[75–77] Similar to selenides, metal tel-
thickness of 100 nm shows the high transparency and energy lurides CoTe and NiTe2 as counter electrodes yields power con-
conversion efficiency of 7.37%, higher than that of the DSSC version efficiencies of 6.92% and 7.21%, respectively, which are
using Pt electrode.[35] comparable to the conventional Pt electrode with 7.04% under
In the family of metal oxides, many oxides belong to semicon- the same condition.[78,79]
ductors, some are insulators (i.e., ZrO2 and Cr2O3), and also a In our precious work, Ni12P5 was first investigated as counter
few of oxides provide good electrical conductivity, such as RuO2, electrode in DSSCs.[80] It is demonstrated that the phosphides
MoO2, and WO2. In particular, non-stoichiometric oxides contain can provide high intrinsic electrocatalytic activity but have
plenty of oxygen vacancies, which can provide catalytically active relatively large diffusion resistance, similar with nitrides and
sites and charge transport. It is reported that WO2 nanorods as carbides. Ni5P4 and MoP also are active but not high enough
counter electrode material in DSSCs hold excellent electrocata- for reduction of triiodide ions.[81]
lytic activity, and the corresponding DSSC has a high power con-
version efficiency of 7.25%, close to that of the device using Pt.[69]
Also, WO3 and WO2.72 show good electrocatalytic activity for 3.5. Multiple Compounds
reduction of triiodide ions.[70] In the investigation, RuO2, Fe3O4,
CuO, V2O3, Nb2O5, and TaOx as counter materials show accept- In the field of counter electrodes for DSSCS, multiple com-
able performance, while TiO2, ZrO2, Cr2O3, and MoO2 have poor pounds are mainly denoted as binary or trinary metal sulfides
electrocatalytic activity for reduction of triiodide.[31,71,72] and/or selenides. It is investigated first by Lin group that after
copper zinc tin sulfide (CZTS) is doped with Se, the obtained
CZTS with doping selenium CZTSSe results in a power con-
3.4.3. Selenides, Tellurides, and Phosphides version efficiency of 7.37% for the corresponding device, much
higher than that using CZTS as counter electrodes.[82] It is
In the cases of metal selenides, tellurides, and phosphides, pointed out that thickness of CZTSe is an important impact
many of the used materials belong to solid solution in essence. factor, and 1.2 µm is the optimal thickness value. Hou et al.
Similar with the interstitial compounds, solid solution pro- reported that the iron group thiospinels MIn2S4 (M = Fe, Co,
vides not only good electrical conductivity but also activated and Ni) counter electrodes with a vertically aligned and inter-
functions. It is reported that Co0.85Se and Ni0.85Se as counter connected nanosheet array structure shows excellent perfor-
electrode in DSSCs, the corresponding devices have power con- mance, as shown in Figure 10.[83] Among the case, the highest
version efficiencies of 9.40% and 8.32%,respectively.[73] Co0.85Se power conversion efficiency of the CoIn2S4 based DSSC is up
obviously surpasses the conventional Pt counter electrode with to 8.83%, higher than that of the device using Pt. The intercon-
power conversion efficiency of 8.64% under the same condi- nected nanosheet array structure shown in SEM in Figure 10
tions. Tang group systematically investigated many metal sele- provides large surface area and an expeditious charge transfer
nides such as Co0.85Se, Ni0.85Se, Cu0.5Se, FeSe, and Ru0.33Se as channel. Also, CuInS2 as counter electrodes shows excellent
transparent counter electrodes for bifacial DSSCs.[74] Among performance in the works.[84,85]
Figure 10. a–c) FESEM images of a) FeIn2S4, b) CoIn2S4, and c) NiIn2S4 counter electrodes. d) Nyquist plots and e) photocurrent density–voltage
characteristics of DSSCs based on MIn2S4 (M=Fe, Co, Ni). Reproduced with permission.[83] Copyright 2017, WILEY-VCH.
3.6. Hybrids fast diffusion pathway of electrolyte species. When active nan-
oparticles are embedded on the flat surface of layers and into
As summarized above, each kind of counter electrode mate- layers, functional composites with an unique structure could
rials has strengths and weaknesses, thus it is actually popular be formed. Except for 1D and 2D matrix materials, most other
to combine two or more components to prepare hybrid counter matrix materials are actually 3D matrix. Among those materials,
electrode materials. It is expected that the hybrid materials ordered mesoporous carbon with 3D ordered channels is very
can possess advantage of each individual by supplementing typical. In 3D matrix, electrons can fast transfer along the 3D
each other. For example, carbon materials have relatively low carbon backbone to the active nanoparticles embedded, while
intrinsic electrocatalytic activity for reduction of triiodide ions, the interconnected ordered channels is helpful for improving
but the network-like morphology and texture provide fast mass- ion diffusion. Figure 12 gives the examples for showing the
transfer in kinetics. On the other hand, transition metal com- effect of diffusion kinetics. To compare diffusion resistance of
pounds usually have excellent intrinsic electrocatalytic activity, bulk metal compounds and hybrids involving matrix materials,
but are not good enough for diffusion kinetics. Therefore, it it is obvious that diffusion kinetics of counter electrodes can be
is possible to obtain improved hybrid materials by dispersing improved to a large extent.
transition metal compounds nanoparticles on carbon materials.
In the past years, many kinds of carbon materials with excel-
lent structural features such as carbon nanotubes, graphene, 3.6.1. Hybrids Based on 1D Matrix Materials
and ordered mesoporous carbon, are used as matrix materials
to fabricate hybrid counter electrode materials. Besides carbon Using carbon nanotubes or silicon nanowires as a 1D matrix,
materials, other kinds of materials like conductive polymers are a much lower amount of Pt loading can obtain excellent per-
also used as matrix materials in hybrids, according to the struc- formance.[91,92] Joshi et al. prepared counter electrodes that
tural features of matrix materials.[30,80,86–89] Figure 11 shows contains nickel nanoparticles on carbon nanotubes and carbon
charge transfer mechanism in hybrid counter electrode mate- nanofibers, and obtained a comparable photovoltaic perfor-
rials based on 1D, 2D, and 3D matrix materials, respectively.[90] mance (η = 7.96%) to those based on Pt CE (η = 8.32%).[93]
1D materials like carbon nanotubes and nanofibers can be More examples are summarized in the review,[94] and efficiency
considered as a fast electron-transport network because of the enhancement by carbon nanotubes in many cases is illustrated
coexistence of ballistic and diffusive transport and the tubular in Figure 13.
morphology. In the counter reaction process, electrons from
outside can be dispersed into the catalytically active sites by
the fast electron-transport network. At the same time, electro- 3.6.2. Hybrids Based on 2D Matrix Materials
lyte can be more effectively distributed due to existence of 1D
materials to reduce ion diffusion resistance. For 2D materials Graphene and reduced graphene oxide (RGO) are typical 2D
like graphene, reduced oxide graphene, etc., the flat surface matrix materials for hybrid counter electrodes.[87,88,95–97] Li
can ensure better electrode–electrolyte contact and serve as a et al. prepared NiS2 nanoparticles with reduced graphene oxide
Figure 11. Schematic of the charge transfer mechanism in hybrid counter-electrode materials based on 1D, 2D, and 3D matrix materials. Reproduced
with permission.[90] Copyright 2010. American Chemical Society.
(NiS2@RGO) as counter electrode materials for DSSCs. The cell 4. Current Developments in Counter Electrode
with NiS2@RGO produces an efficiency of 8.55%, higher than Materials for PSCs
that of the cell with pure NiS2 (7.05%) or RGO (3.14%).[88] Zhang
et al. fabricated three kinds of hybrids, with MoN, TiN, and VN As analyzed in Section 2.1, the function of counter elec-
nanoparticles anchored on the surface of N-doped graphene, trode in PSCs is somehow different from those of counter
obtained obviously improved efficiencies compared to the pure electrode in DSSCs. For counter electrode in PSCs, elec-
metal nitrides.[87] Zhou et al. show nano–micro composite cata- trocatalytic activity is not needed, but energy level and
lysts (NMCCs) composed of highly dispersed Fe3O4 nanoparticles work function must be preferentially considered, and then
on RGO sheets (namely, Fe3O4@RGO-NMCC). The rigid electrical conductivity and stability. In recent years, some
and flexible Fe3O4@RGO-NMCC counter electrodes low-cost substitutive materials for Au and Ag as counter
achieved high power conversion efficiency, reaching 9% electrode materials such as carbon, Al, Cu, TCO, etc.,
and 8% on FTO/glass substrates and flexible polymer have been investigated.[103–114] Figure 14 shows energy
substrates, respectively.[97] Conductive polymers mixed with 2D level and price of the counter electrode materials that
matrix materials can also be improved in performance.[98,99] The have been tested. Among these materials, carbon, Ni, and
efficiency enhancement of 2D matrix materials is illustrated in PEDOT have energy level at around −5.0 eV, very near
Figure 13, and it is more outstanding compared to those of 1D to that of Au, Cu, and TCO at about −4.6 eV, and Mo,
and 3D cases. W, and Al provide relatively high energy level like Ag.
It cannot be concluded from the results what is the best
point of energy level for counter electrode in PSCs. However,
3.6.3. Hybrid-Based 3D Matrix Materials it has been known that the suitable energy level ranges from
−5.1 to −4.1 eV are workable for counter electrodes in PSCs.
MoC and WC embedded in ordered mesoporous carbon mate- As mentioned in Section 2.1, either too low or too high work
rials achieve excellent performance.[89] Ramasamy et al. prepared function for counter electrode materials, the charge transport
mesoporous of OM TiN-C as a counter electrode material and cannot occur in the interface between counter electrode and
obtain a 6.71% of energy conversion efficiency, much higher HTM. With regard to material price, it is obvious that the non-
than that of OMC (5.16%) or TiN nanoparticles (1.73%).[100] precious metals are much cheaper, and carbon totally keeps
In2.77S4 supported conductive carbon also has an improved per- the lowest cost (graphene as a non-commercial material is
formance.[101] Interestingly, some conductive polymers can also currently higher than metals in price).
be used as matrix materials. Xu et al. mixed TiN and PEDOT:PSS The details on the low-cost substitutive materials as
under ultrasonication to obtain the composite counter electrodes, counter electrode in PSCs have been recently summa-
delivering an effectively combined network of both high elec- rized by Gao group.[115] In this review, we will not discuss
trical conductivity and superior electrocatalytic activity.[102] each material one by one, but show comprehensive analysis
Figure 12. TEM images and Nyquist plots of hybrid counter electrodes based on a) 1D, b) 2D, and c) 3D matrix materials. Reproduced with
permission.[22,87,89] Copyright 2015, Elsevier, 2013, Royal Society of Chemistry, 2011, WILEY-VCH.
based on the reported data. As known, power conversion 4.1. Carbon Materials
efficiency (η) of a solar cell depends on open circuit voltage
(Voc), short circuit current (Isc), and fill factor (FF), that is, Some typical carbon materials such as graphite, carbon black,
η = Voc × Isc × FF. Among the three parameters, open cir- carbon nanotubes, and graphene have been investigated as
cuit voltage (Voc) and fill factor (FF) have direct relation with counter electrode materials for PSCs.[116] Different from metal
performance of counter electrodes, while short circuit cur- electrodes, carbon materials usually have rich micropores and
rent (Isc) is related to only generation efficiency and recom- mesopores, and relatively high specific surface area. The fea-
bination efficiency of photo-generated electrons in perovskite tures make carbon materials work well as counter electrodes in
layer and photoanode.[20] Electrical conductivity of counter some non-typical structured PSCs. Han’s group has developed
electrodes and the interface resistance between counter elec- monolithic structured PSCs with carbon-based electrodes,[116]
trode and hole transport layer are one of the decisive param- in which electronic transport layer (TiO2), scaffold layer (ZrO2
eters for open circuit voltage (Voc) and fill factor (FF). There- or Al2O3), carbon layer contact sequentially, and perovskite
fore, open circuit voltage (Voc) and fill factor (FF) can be used forms in the porous triple layers by infiltration. The mono-
to reasonably evaluate performance of the low-cost counter lithic-structured PSCs actually break through the conventional
electrode materials. layer-by-layer structure. In addition, some works based on the
level and good electrical conductivity.[106,124]

However, hole-transport materials usually
are required in the cases of CNTs electrodes
to reduce the recombination of holes at the
interface of perovskite layer and the CNTs
layers. In addition, graphene is also intro-
duced into counter electrodes, and is con-
sidered as an excellent transparent electrode
material for PSCs.[107]
Besides interface compatibility, the resist-
ance of counter electrode itself has influence
on final performance of PSCs. Figure 15
shows open voltage and fill factor of PSCs
using different counter electrodes versus
1/Rsq (Rsq stands for square resistance of
electrodes) of counter electrodes, based on
Figure 13. Enhancement of efficiency of DSSCs using hybrid counter electrode with carbon shown data in some works.[119,124–129] For
nanotubes as matrix materials.[87–89,95,96,99,100] the carbon electrodes, it is generally clear
that open voltage of PSCs increases with the
conventional structure have directly used carbon materials rising 1/Rsq value, that is, the lower square
to replace Au electrode and hole-transport layer.[117–120] In the
devices without hole-transport materials, carbon collects not
only the electrons from the external circuit but also the holes
separated in the perovskite layer. Therefore, interface compat-
ibility of carbon materials with the organic-inorganic hybrid
perovskite layer is especially important. For example, the
lamella nature of graphite tends to produce interspace between
the graphite flakes and perovskites.[121] The problem needs
to be solved by the solvent engineering strategy realizing an
enhanced interface contact.[121] Additives can also enhance the
interface contact. Single-walled carbon nanotubes introduced
into graphite and carbon black electrode can improve the inter-
face compatibility between carbon electrode and perovskite
materials and transport of charges.[122]
Similarly, multi-walled carbon nanotubes inserted into the
grain boundaries of perovskite can create a CNTs-perovskite
heterojunction, accelerating hole transfer from perovskite to
carbon electrode.[123] Also, CNTs can be employed as counter
electrode materials alone in PSCs due to the matched energy
Figure 14. Price and Fermi energy levels of counter-electrode materials Figure 15. a) Open voltage and b) fill factor of PSCs using different counter
for PSCs. electrodes versus 1/Rsq of counter electrodes.[110–112,114,119,124–129,136–143]
on the inverted structured PSCs (p-i-n PSCs).[110,111,136,137] In

the cases, C60/BCP as electron transport layer is effective due
to less damage during the thermal evaporation of Cu for the
device ITO glass/PTAA/perovskite/ICBA/C60/BCP/Cu. In
comparison, the Cu-electrode solar cell shows a mildly better
performance than Al-electrode and Au-electrode devices. In
addition, the Cu-based devices show dramatic stability for both
MAPbI3 and FA0.4MA0.6PbI3 devices. For the use of Cu counter
electrode, it is very important to avoid direct contact of perov-
skite layer and Cu electrode, because Cu is easily oxidized by
oxygen and moisture in air to accelerate the decomposition of
perovskite.[138]
Cu, Ni, W, and Mo, as well as Ag as reference, have been
investigated as counter electrode of PSCs under the same
conditions.[112] Clearly, the final photovoltaic performance of the
PSCs is influenced by the resistance of the as-prepared metal
films to a great extent.[112] Under the optimized conditions, Cu,
Figure 16. Fill factor and power conversion efficiency of PSCs using dif- Ni, W, and Mo films as metal counter electrodes present per-
ferent counter electrodes. formance with the best PCE of 13.04%,12.18%, 12.38%, and
11.38%, respectively, with 15.97% for the Ag counterpart.[112]
resistance of carbon electrodes is, the higher open voltage of In addition, a mesoporous Ni film is employed as a printable
the corresponding PSCs. From the results, it can be under- counter electrode for PSCs. Due to the high conductivity of the
stood that high resistance of counter electrode increases ohmic mesoporous Ni films, an initial power conversion efficiency of
polarization to cause voltage loss for the devices. However, 13.6% is achieved, and a unique reuse process can be realized
it is not clear in Figure 15 for the relation between fill factor by reloading the perovskite materials.[139]
and Rsq. In Figure 15, the open circuit voltage and fill factor of PSCs
using different metal counter electrodes is shown versus
1/Rsq of the counter electrodes. Generally, open circuit voltage
4.2. Metal Materials of PSCs slightly increases with the rising 1/Rsq value. The
influence is not obvious as that in the cases of carbon counter
To replace noble metals gold and silver, some low-cost metals, electrodes. Similarly, it seems that Rsq has not a consistent
such as Al, Cu, Ni, Mo, and W, have been investigated as influence on fill factor.
counter electrode materials for PSCs.
Al has been tested in many cases, especially in the inverted
structured PSCs (p-i-n PSCs), due to high electrical conductivity 4.3. TCO and Polymers
and low cost.[130–132] It can be observed in Figure 16 that the
performance of the devices using Al are widely varied. These In PSCs and DSSCs, TCO glass usually are used as bottom
phenomena are mainly because of matching of Al with var- electrodes for light irradiation. The TCO materials such as ITO
ious hole-transport materials (HTM) and buffer layers. If Al is (Indium tin oxide), IO:H (hydrogen-doped indium oxide), IZO
deposited directly on the top of the HTM layer, the power con- (indium zinc oxide), and AZO (aluminum-doped zinc oxide)
version efficiency of the PSC is unsatisfied.[132] It is considered can not only transfer electrons and holes, but also allow light
that an energy barrier can form at the HTM/Al interface and to transit. The features make TCO irreplaceable to fabricate
limit hole extraction.[133] However, in our view, the oxidation of tandem solar cells combining lower bandgap absorbers like
Al could be the main reason, because Al is extremely easy to silicon with PSCs for high efficiency devices.[114,140–143] Usu-
be oxidized, and alumina has a Fermi energy level in −3.2 to ally, for the PSCs with TCO as counter electrodes, a buffer
3.8 eV that is not workable for PSCs. Thus, some interlayers layer (MoO3, BCP etc.) should be used to protect HTM layer or
such as MoOx, LiF, ZnO, bathocuproin (BCP) are inserted perovskite layer from magnetron sputtering of TCO. Undoubt-
between HTM and Al counter electrode to improve power con- edly, this will increase the interface resistances to reduce the
version efficiency of PSCs.[130] In addition, alloying based on Al is final performance of the devices. This is why the power conver-
a potential approach to improving the performance of Al-based sion efficiencies of the PSCs with TCO as counter electrodes
counter electrode in PSCs.[134] Similarly, in organic solar cells, are relatively low level (lower than 12%). However, the corre-
Al is also usually used as metal electrode with matching buffer sponding tandem solar cells normally achieve a high power
layers for stable and effective interfacial contact.[135] conversion efficiency over 20%.[141–143]
Cu provides excellent electrical conductivity and com- Different from the results of carbon materials and metal
parable ductility with Au. In the typical architecture with materials, TCO does not show any relation between open
Spiro-OMeTAD as HTM, our group has used Cu as counter voltage and fill factor of PSCs with Rsq (Figure 15). This should
electrode to obtain a PSC with power conversion efficiency be attributed to that fact different buffer layers used in the dif-
of 13.04%.[112] Huang’s group has developed outstanding ferent cases have more influence than counter electrode itself
technologies for use of Cu counter electrodes mainly based on the final performance of the devices.
stable. In cases of carbon nanotubes (CNTs),

the high fill factor and high open voltage can
be achieved individually, implying promising
potential for further development in future.
The transition metals, such as Ni, Mo, and
W, have an acceptable performance; however,
the price of these metals is obviously higher
than carbon materials. TCO and PEDOT:PSS
are expected to have good development in
semitransparent devices and tandem solar
cells.
5. Conclusion
In this review, the research progress of low-
cost counter electrode materials for DSSCs
and PSCs are summarized, focusing on
comparison of the different sorts of low-cost
materials and commonness and otherness in
low-cost counter electrode between DSSCs
and PSCs. As a summary, the main conclu-
sions are generalized in Figure 17.
For DSSC counter electrodes, the electro
catalytic activity, electrical conductivity and
diffusion properties of materials have directly
influence on the final performance of solar
cells, and material cost are also important
for use in large scale. Metals and alloys
have outstanding conductivity and excellent
electrocatalytic activity. By exploring new
low-cost alloy materials and adjusting con-
trollably component ratios, more metal alloy
counter electrode materials with superior
Figure 17. Sumary of low-cost counter materials for a) DSSCs and b) PSCs. The different types performance are expected. Carbon materials
of materials are sequenced in 1–5 levels, representing their performance in the corresponding have obvious advantages in price compe-
factors from poor to excellent. tiveness, and abundance. Though the low
intrinsic electrocatalytic activity is a limita-
As a conductive polymer, PEDOT:PSS with high conductivity tion for achieving high-efficient counter electrodes, carbon
can be used as hole-transport material. Besides, work function materials are an excellent option as matrix to fabricate highly
of PEDOT:PSS is about 5.0 eV, very close to Au. Therefore, con- active hybrid counter electrode materials. Among transition
sider individual material itself, PEDOT:PSS should be an ideal metal compounds, nitrides and carbides have high intrinsic
replacing material for counter electrodes in PSCs. However, the electrocatalytic activity and excellent electrical conductivity, but
preparation process of PEDOT:PSS usually involves aqueous poor diffusion limits full utilization of active materials. To pre-
solution, which is a fatal influence on stability of perovskite pare smaller-sized and stable nanoparticles or hybridization
layer. Thus, developing suitable method for important to use with matrix materials like carbon is one of the most important
PEDOT:PSS as counter electrodes. Some works have shown approaches to improving nitrides and carbides counter elec-
effective pathways to obtain polymer-based PSCs such as the trodes. Though not attractive enough on electrocatalytic activity
transfer-lamination technique,[144] antireflection coating,[145] compared to other materials, metal sulfides and oxides may
and detachable stacking perovskite.[146] be a good choice for transparent and semitransparent counter
As mentioned above, the open circuit voltage (Voc) and fill electrodes, especially for tandem solar cells.
factor (FF) are key parameters for evaluating performance of For counter electrode of PSCs, energy level and compatibility
the low-cost counter electrode materials. Figure 16 shows fill must preferentially be considered, and then electrical conduc-
factor (FF) and open circuit voltage (Voc) of the PSCs with the tivity, stability, and price. Carbon materials have a close Fermi
conventional cell structure using the different low-cost counter energy level to Au and very low price, and are very stable. It is
electrode materials based on the reported data in the review.[115] believed that carbon materials will attract more attention and
In Figure 16, focusing on the area of open voltage higher get further development in future as counter electrode mate-
than 1.0 V and fill factor higher than 0.75, it can be found that rials for PSCs. Among transition metals, Cu is outstanding
performance of Cu counter electrodes is excellent and relatively based on excellent electrical conductivity and controlled contact
with different hole-transport layers. However, Ni, Mo and [22] Q. W. Tang, J. L. Duan, Y. Y. Duan, B. L. He, L. M. Yu, Electrochim.
W needs more evaluations and more researches to find their Acta 2015, 178, 886.
highlights. Similar to oxide materials in DSSCs, TCO and con- [23] J. W. Wan, G. J. Fang, H. J. Yin, X. F. Liu, D. Liu, M. T. Zhao,
ductive polymers like PEDOT are expected to have good per- W. J. Ke, H. Tao, Z. Y. Tang, Adv. Mater. 2014, 26, 8101.
[24] G. R. Li, F. Wang, Q. W. Jiang, X. P. Gao, P. W. Shen, Angew.
formance in transparent, semitransparent, and tandem devices.
Chem., Int. Ed. 2010, 49, 3653.
[25] G. R. Li, F. Wang, J. Song, F. Y. Xiong, X. P. Gao, Electrochim. Acta
2012, 65, 216.
Acknowledgements [26] M. J. Ju, I. Y. Jeon, J. C. Kim, K. Lim, H. J. Choi, S. M. Jung,
I. T. Choi, Y. K. Eom, Y. J. Kwon, J. Ko, J. J. Lee, H. K. Kim,
Financial Supports from the 973 Program (2015CB251100), NSFC J. B. Baek, Adv. Mater. 2014, 26, 3055.
(21875123) of China, Nankai University (2122018196), and Natural Science [27] R. Trevisan, M. Döbbelin, P. P. Boix, E. M. Barea, R. Tena-Zaera,
Foundation of Tianjin (18JCZDJC31000) are gratefully acknowledged. I. Mora-Sero, J. Bisquert, Adv. Energy Mater. 2011, 1, 781.
[28] F. Guo, H. Hua, Q. D. Tai, B. L. Chen, B. Sebo, C. H. Bu, J. H. Xu,
X. Z. Zhao, J. Renewable Sustainable Energy 2012, 4, 023109.
Conflict of Interest [29] W. J. Hou, Y. M. Xiao, G. Y. Han, D. Y. Fu, R. F. Wu, J. Power Sources
2016, 322, 155.
The authors declare no conflict of interest. [30] G. R. Li, J. Song, G. L. Pan, X. P. Gao, Energy Environ. Sci. 2011, 4,
1680.
[31] M. X. Wu, X. Lin, Y. D. Wang, L. Wang, W. Guo, D. D. Qi,
X. J. Peng, A. Hagfeldt, M. Grätzel, T. L. Ma, J. Am. Chem. Soc.
Keywords 2012, 134, 3419.
counter-electrode materials, dye-sensitized solar cells, electrochemistry, [32] F. Kaiser, P. Simon, U. Burkhardt, B. Kieback, Y. Grin,
perovskite solar cells I. Veremchuk, Crystals 2017, 7, 271.
[33] C. Nico, T. Monteiro, M. P. F. Graça, Prog. Mater. Sci. 2016, 80, 1.
Received: October 7, 2018 [34] L. Wang, Y. T. Shi, H. Zhang, X. G. Bai, Y. X. Wang, T. L. Ma,
Revised: January 31, 2019 J. Mater. Chem. A 2014, 2, 15279.
Published online: May 22, 2019 [35] J. Song, G. R. Li, C. Y. Wu, X. P. Gao, J. Power Sources 2014, 266,
464.
[36] Y. Y. Duan, Q. W. Tang, J. Liu, B. L. He, L. M. Yu, Angew. Chem.,
Int. Ed. 2014, 53, 1.
[1] Q. Schiermeier, J. Tollefson, T. Scully, A. Witze, O. Morton, Nature [37] J. H. Guo, Y. T. Shi, C. Zhu, L. Wang, N. Wang, T. L. Ma, J. Mater.
2008, 454, 816. Chem. A 2013, 1, 11874.
[2] A. Polman, M. Knight, E. C. Garnett, B. Ehrler, W. C. Sinke, Science [38] J. Yang, C. X. Bao, J. Y. Zhang, T. Yu, H. Huang, Y. L. Wei, H. Gao,
2016, 352, add4424. G. Fu, J. G. Liu, Z. G. Zou, Chem. Commun. 2013, 49, 2028.
[3] S. Chu, A. Majumdar, Nature 2012, 488, 294. [39] R. Y. Yao, Z. J. Zhou, Z. L. Hou, X. Wang, W. H. Zhou, S. X. Wu,
[4] J. W. Gong, K. Sumathy, Q. Q. Qiao, Z. P. Zhou, Renewable ACS Appl. Mater. Interfaces 2013, 5, 3143.
Sustainable Energy Rev. 2017, 68, 234. [40] V. R. Stamenkovic, B. S. Mun, M. Arenz, K. J. J. Mayrhofer,
[5] P. P. Boix, K. Nonomura, N. Mathews, S. G. Mhaisalkar, Mater. C. A. Lucas, G. Wang, P. N. Ross, N. M. Markovic, Nat. Mater.
Today 2014, 17, 16. 2007, 6, 241.
[6] B. O’Regan, M. Gratzel, Nature 1991, 353, 737. [41] X. J. Zheng, J. Deng, N. Wang, D. H. Deng, W. H. Zhang,
[7] A. Hagfeldtt, M. Gratzel, Chem. Rev. 1995, 95, 49. X. H. Bao, C. Li, Angew. Chem., Int. Ed. 2014, 53, 7023.
[8] N. Papageorgiou, Coord. Chem. Rev. 2004, 248, 1421. [42] X. X. Chen, Q. W. Tang, B. L. He, L. Lin, L. M. Yu, Angew. Chem.,
[9] A. Hagfeldt, G. Boschloo, L. C. Sun, L. Kloo, H. Pettersson, Chem. Int. Ed. 2014, 53, 10799.
Rev. 2010, 110, 6595. [43] B. L. He, X. Meng, Q. W. Tang, P. J. Li, S. S. Yuan, P. Z. Yang,
[10] J. H. Wu, Z. Lan, J. M. Lin, M. L. Huang, Y. F. Huang, L. Q. Fan, J. Power Sources 2014, 260, 180.
G. G. Luo, Y. Lin, Y. M. Xie, Y. L. Wei, Chem. Soc. Rev. 2017, 46, [44] M. X. Wu, X. Lin, T. H. Wang, J. S. Qiu, T. L. Ma, Energy Environ.
5975. Sci. 2011, 4, 2308.
[11] S. N. Yun, A. Hagfeldt, T. L. Ma, Adv. Mater. 2014, 26, 6210. [45] D. Zhang, X. Li, S. Chen, Z. Sun, X. J. Yin, S. Huang, Microchim.
[12] J. Theerthagiri, A. R. Senthil, J. Madhavan, T. Maiyalagan, Acta 2011, 174, 73.
ChemElectroChem 2015, 2, 928. [46] E. Ramasamy, W. J. Lee, D. Y. Lee, J. S. Song, Electrochem.
[13] S. Thomas, T. G. Deepak, G. S. Anjusree, T. A. Arun, S. V. Nair, Commun. 2008, 10, 1087.
A. S. Nair, J. Mater. Chem. A 2014, 2, 4474. [47] H. Anwar, A. E. George, I. G. Hill, Sol. Energy 2013, 88, 129.
[14] S. Q. Luo, W. A. Daoud, J. Mater. Chem. A 2015, 3, 8992. [48] P. Dong, C. L. Pint, M. Hainey, F. Mirri, Y. Zhan, J. Zhang,
[15] G. D. Niu, X. D. Guo, L. D. Wang, J. Mater. Chem. A 2015, 3, 8970. M. Pasquali, R. H. Hauge, R. Verduzco, M. Jiang, H. Lin, J. Lou,
[16] L. S. Liang, Y. Cai, X. Li, M. K. Nazeeruddind, P. Gao, Nano Energy ACS Appl. Mater. Interfaces 2011, 3, 3157.
2018, 52, 211. [49] J. D. Roy-Mayhew, D. J. Bozym, C. Punckt, I. A. Aksay, ACS Nano
[17] M. Kouhnavard, N. A. Ludin, B. V. Ghaffari, K. Sopian, S. Ikeda, 2010, 4, 6203.
ChemSusChem 2015, 8, 1510. [50] L. Kavan, J. H. Yum, M. Grätzel, ACS Nano 2011, 5, 165.
[18] L. Kavan, J. H. Yum, M. Grotzel, ACS Nano 2011, 5, 165. [51] K. Saranya, Md. Rameez, A. Subramania, Eur. Polym. J. 2015, 66,
[19] A. Dualeh, T. Moehl, N. Tétreault, J. Teuscher, P. Gao, 207.
M. K. Nazeeruddin, M. Gratzel, ACS Nano 2014, 8, 362. [52] K. M. Lee, P. Y. Chen, C. Y. Hsu, J. H. Huang, W. H. Ho,
[20] A. Kahn, Mater. Horiz. 2016, 3, 7. H. C. Chen, K. Chuan Ho, J. Power Sources 2009, 188, 313.
[21] C. H. Lin, T. Y. Li, B. Cheng, C. X. Liu, C. W. Yang, J. J. Ke, T. C. Wei, [53] M. Mastragostino, C. Arbizzani, F. Soavi, J. Power Sources 2001,
L. J. Li, A. Fratalocchi, J. H. He, Nano Energy 2018, 53, 817. 97–98, 812.
[54] J. J. M. Halls, C. A. Walsh, N. C. Greenham, E. A. Marseglia, [88] Z. Li, F. Gong, G. Zhou, Z. Wang, J. Phys. Chem. C 2013, 117,
R. H. Friend, S. C. Moratti, A. B. Holmes, Nature 1995, 376, 498. 6561.
[55] L. Bay, K. West, B. Winther-Jensen, T. Jacobsen, Sol. Energy Mater. [89] M. Wu, X. Lin, A. Hagfeldt, T. Ma, Angew. Chem., Int. Ed. 2011, 50,
Sol. Cells 2006, 90, 341. 3520.
[56] E. Furimsky, Applied Catalysis A: General. 2003, 240, 1. [90] N. Yang, J. Zhai, D. Wang, Y. Chen, L. Jiang, ACS Nano 2010, 4,
[57] Q. W. Jiang, G. R. Li, X. P. Gao, Chem. Commun. 2009, 6720. 887.
[58] X. Zhang, X. Chen, S. Dong, Z. Liu, X. Zhou, J. Yao, S. Pang, [91] J. W. Guo, B. Zhang, Y. Hou, S. Yang, X. H. Yang, H. G. Yang,
H. Xu, Z. Zhang, L. Lia, G. Cui, J. Mater. Chem. 2012, 22, 6067. J. Mater. Chem. A 2013, 1, 1982.
[59] Q. W. Jiang, G. R. Li, S. Liu, X. P. Gao, J. Phys. Chem. C 2010, 114, [92] F. Miao, B. Tao, P. K. Chu, Electrochim. Acta 2013, 96, 61.
13397. [93] P. Joshi, Z. P. Zhou, P. Poudel, A. Thapa, X. F. Wu, Q. Q. Qiao,
[60] J. Song, G. R. Li, K. Xi, B. Lei, X. P. Gao, R. Vasant Kumar, J. Mater. Nanoscale 2012, 4, 5659.
Chem. A 2014, 2, 10041. [94] S. Hwang, M. Batmunkh, M. J. Nine, H. Chung, H. Jeong,
[61] M. K. Wang, A. M. Anghel, B. Marsan, N. C. Ha, N. Pootrakulchote, ChemPhysChem 2015, 16, 53.
S. M. Zakeeruddin, M. Gratzel, J. Am. Chem. Soc. 2009, 131, [95] H. Zheng, C. Y. Neo, J. Ouyang, ACS Appl. Mater. Interfaces 2013,
15976. 5, 6657.
[62] J. H. Huo, J. H. Wu, M. Zheng, Y. G. Tu, Z. Lan, J. Power Sources. [96] M. H. Yeh, L. Y. Lin, C. L. Sun, Y. A. Leu, J. T. Tsai, C. Y. Yeh,
2016, 304, 266. R. Vittal, K. C. Ho, J. Phys. Chem. C 2014, 118, 16626.
[63] J. Y. Lin, G. T. Yue, S. Y. Tai, Y. M. Xiao, H. M. Cheng, F. M. Wang, [97] H. W. Zhou, J. Yin, Z. H. Nie, Z. J. Yang, D. J. Li, J. H. Wang,
J. H. Wu, Mater. Chem. Phys. 2013, 143, 53. X. Liu, C. Z. Jin, X. X. Zhang, T. L. Ma, J. Mater. Chem. A 2016, 4,
[64] H. J. Zhang, L. T. Yang, Z. Liu, M. Ge, Z. Zhou, W. Chen, Q. Z. Li, 67.
L. Liu, J. Mater. Chem. 2012, 22, 18572. [98] G. T. Yue, J. H. Wu, Y. M. Xiao, J. M. Lin, M. L. Huang, Z. Lan,
[65] P. P. Sun, M. Zhang, C. Z. Ai, Z. X. Wu, S. Lu, X. T. Zhang, L. Q. Fan, Energy 2013, 54, 315.
N. Huang, Y. H. Sun, X. H. Sun, J. Power Sources 2016, 319, [99] F. Gong, X. Xu, G. Zhou, Z. S. Wang, Phys. Chem. Chem. Phys.
219. 2012, 15, 546.
[66] J. H. Huo, J. H. Wu, M. Zheng, Y. G. Tu, Z. Lan, Electrochim. Acta. [100] E. Ramasamy, C. Jo, A. Anthonysamy, I. Jeong, J. K. Kim, J. Lee,
2016, 187, 210. Chem. Mater. 2012, 24, 1575.
[67] C. Kung, H. Chen, C. Lin, K. Huang, R. Vittal, K. Ho, ACS Nano [101] M. X. Guo, Y. G. Yao, F. L. Zhao, S. M. Wang, D. G. Wang,
2012, 6, 7016. S. H. Yin, H. M. Zhang, H. Gao, J. K. Xiao, J. Photo. & Photobio. A
[68] B. Lei, G. R. Li, X. P. Gao, J. Mater. Chem. A 2014, 2, 3919. Chem. 2017, 332, 87.
[69] M. Wu, X. Lin, A. Hagfeldt, T. Ma, Chem. Commun. 2011, 47, 4535. [102] H. Xu, X. Zhang, C. Zhang, Z. Liu, X. Zhou, S. Pang, X. Chen,
[70] H. Zhou, Y. Shi, L. Wang, H. Zhang, C. Zhao, A. Hagfeldt, T. Ma, S. Dong, Z. Zhang, L. Zhang, P. Han, X. Wang, G. Cui, ACS Appl.
Chem. Commun. 2013, 49, 7626. Mater. Interfaces 2012, 4, 1087.
[71] Y. Hou, Z. Chen, D. Wang, B. Zhang, S. Yang, H. Wang, P. Hu, [103] C. Zhang, Q. Luo, X. Deng, J. Zheng, W. O. Yang, X. Chen,
H. Zhao, H. Yang, Small 2014, 10, 483. S. Huang, Electrochim. Acta 2017, 258, 1262.
[72] L. Wang, Y. Shi, H. Zhang, X. Bai, Y. Wang, T. Ma, J. Mater. Chem. [104] S. Gholipour, J. P. Correa-Baena, K. Domanski, T. Matsui, L. Steier,
A 2014, 2, 15279. F. Giordano, F. Tajabadi, W. Tress, M. Saliba, A. Abate, A. Morteza
[73] F. Gong, H. Wang, X. Xu, G. Zhou, Z. Wang, J. Am. Chem. Soc. Ali, N. Taghavinia, M. Grätzel, A. Hagfeldt, Adv. Energy Mater.
2012, 134, 10953. 2016, 6, 1601116.
[74] Y. Y. Duan, Q. W. Tang, J. Liu, B. L. He, L. M. Yu, Angew. Chem., [105] Q. Q. Chu, B. Ding, Q. Qiu, Y. Liu, C. X. Li, C. J. Li, G. J. Yang,
Int. Ed. 2014, 53, 14569. B. Fang, J. Mater. Chem. A 2018, 6, 8271.
[75] F. Gong, X. Xu, Z. Li, G. Zhou, Z. Wang, Chem. Commun. 2013, [106] N. Ahn, I. Jeon, J. Yoon, E. I. Kauppinen, Y. Matsuo, S. Maruyama,
49, 1437. M. Choi, J. Mater. Chem. A 2018, 6, 1382.
[76] Z. Zhang, S. Pang, H. Xu, Z. Yang, X. Zhang, Z. Liu, X. Wang, [107] P. You, Z. Liu, Q. Tai, S. Liu, F. Yan, Adv. Mater. 2015, 27,
X. Zhou, S. Dong, X. Chen, RSC Adv. 2013, 3, 16528. 3632.
[77] J. Guo, Y. Shi, C. Zhu, N. Wang, T. Ma, J. Mater. Chem. A 2013, 1, [108] K. Y. Zhao, A. M. Nardes, K. Zhu, Appl. Phys. Lett. 2014, 104,
11874. 213906.
[78] J. Guo, Y. Shi, Y. Chu, T. Ma, Chem. Commun. 2013, 49, 10157. [109] H. Sung, N. Ahn, M. S. Jang, J.-K. Lee, H. Yoon, N.-G. Park,
[79] S. Patil, E. Kim, N. Shrestha, J. Chang, J. Lee, S. Han, ACS Appl. M. Choi, Adv. Energy Mater. 2016, 6, 1501873.
Mater. Interfaces 2015, 7, 25914. [110] X. Zheng, B. Chen, J. Dai, Y. Fang, Y. Bai, Y. Lin, H. Wei, X. C. Zeng,
[80] Y. Y. Dou, G. R. Li, J. Song, X. P. Gao, Phys. Chem. Chem. Phys. J. Huang, Nat. Energy 2017, 2, 17102.
2012, 14, 1339. [111] Y. Lin, B. Chen, F. Zhao, X. Zheng, Y. Deng, Y. Shao, Y. Fang, Y. Bai,
[81] M. Wu, C. Chung, Z. Ceng, RSC Adv. 2015, 5, 4561. C. Wang, J. Huang, Adv. Mater. 2017, 29, 1700607.
[82] X. Xin, M. He, W. Han, J. Jung, Z. Lin, Angew. Chem., Int. Ed. 2011, [112] L. Wang, G. R. Li, Q. Zhao, X. P. Gao, Energy Storage Mater. 2017,
50, 11739. 7, 40.
[83] W. J. Hou, Y. M. Xiao, G. Y. Han, Angew. Chem., Int. Ed. 2017, 56, [113] A. Rajagopal, Z. Yang, S. B. Jo, I. L. Braly, P. Liang, H. W. Hillhouse,
9146. A. K.-Y. Jen, Adv. Mater. 2017, 29, 1702140.
[84] J. Yang, C. Bao, J. Zhang, T. Yu, H. Huang, Y. Wei, H. Gao, G. Fu, [114] K. A. Bush, A. F. Palmstrom, Z. J. Yu, M. Boccard, R. Cheacharoen,
J. Liu, Z. Zou, Chem. Commun. 2013, 49, 2028. J. P. Mailoa, D. P. McMeekin, R. L. Z. Hoye, C. D. Bailie, T. Leijtens,
[85] R. Yao, Z. Zhou, Z. Hou, X. Wang, W. Zhou, S. Wu, ACS Appl. I. M. Peters, M. C. Minichetti, N. Rolston, R. Prasanna, S. Sofia,
Mater. Interfaces 2013, 5, 3143. D. Harwood, W. Ma, F. Moghadam, H. J. Snaith, T. Buonassisi,
[86] J. Guo, B. Zhang, Y. Hou, S. Yang, X. Yang, H. Yang, J. Mater. Z. C. Holman, S. F. Bent, M. D. McGehee, Nat. Energy 2017, 2,
Chem. A 2013, 1, 1982. 17009.
[87] X. Zhang, X. Chen, K. Zhang, S. Pang, X. Zhou, H. Xu, S. Dong, [115] L. S. Liang, Y. Cai, X. Li, M. K. Nazeeruddin, P. Gao, Nano Energy
P. Han, Z. Zhang, C. Zhang, G. Cui, J. Mater. Chem. A 2013, 1, 2018, 52, 211.
3340. [116] Z. Ku, Y. Rong, M. Xu, T. Liu, H. Han, Sci. Rep. 2013,3, 3132.
[117] H. N. Chen, Z. H. Wei, H. X. He, X. L. Zheng, K. S. Wong, [132] S. J. Park, S. Jeon, I. K. Lee, J. Zhang, H. Jeong, J.-Y. Park, J. Bang,
S. H. Yang, Adv. Energy Mater. 2016, 6, 1502087. T. K. Ahn, H. Shin, B. Kim, H. J. Park, J. Mater. Chem. A 2017, 5,
[118] X. Chang, W. Li, H. Chen, L. Zhu, H. Liu, H. Geng, S. Xiang, J. Liu, 13220.
X. Zheng, Y. Yang, S. Yang, ACS Appl. Mater. Interfaces 2016, 8, [133] J. Gao, C. L. Perkins, J. M. Luther, M. C. Hanna, H. Y. Chen,
30184. O. E. Semonin, A. J. Nozik, R. J. Ellingson, Nano Lett. 2011, 11,
[119] Q. Q. Chu, B. Ding, Y. Li, L. L. Gao, Q. Qiu, C. X. Li, 3263.
C. J. Li, G. J. Yang, B. Fang, ACS Sustainable Chem. Eng. 2017, 5, [134] Z. Jiang, X. Chen, X. Lin, X. Jia, J. Wang, L. Pan, S. Huang, F. Zhu,
9758. Z. Sun, Sol. Energy Mater. Sol. Cells 2016, 146, 35.
[120] Q.-Q. Chu, B. Ding, Q. Qiu, Y. Liu, C.-X. Li, C.-J. Li, G.-J. Yang, [135] L. S. Pali, S. K. Gupta, A. Garg, Sol. Energy 2018, 160, 396.
B. Fang, J. Mater. Chem. A 2018, 6, 8271. [136] J. Zhao, X. Zheng, Y. Deng, T. Li, Y. Shao, A. Gruverman, J. Shield,
[121] Y. Yang, J. Xiao, H. Wei, L. Zhu, D. Li, Y. Luo, H. Wu, Q. Meng, J. Huang, Energy Environ. Sci. 2016, 9, 3650.
RSC Adv. 2014, 4, 52825. [137] S. Tang, Y. Deng, X. Zheng, Y. Bai, Y. Fang, Q. Dong, H. Wei,
[122] H. Li, K. Cao, J. Cui, S. Liu, X. Qiao, Y. Shen, M. Wang, Nanoscale J. Huang, Adv. Energy Mater. 2017, 7, 1700302.
2016, 8, 6379. [138] Y. Deng, Q. Dong, C. Bi, Y. Yuan, J. Huang, Adv. Energy Mater.
[123] J. Ryu, K. Lee, J. Yun, H. Yu, J. Lee, J. Jang, Small 2017, 13, 1701225. 2016, 6, 1600372.
[124] K. Aitola, K. Sveinbjörnsson, J.-P. Correa-Baena, A. Kaskela, [139] Z. Ku, X. Xia, H. Shen, N. H. Tiep, H. J. Fan, Nanoscale 2015, 7,
A. Abate, Y. Tian, E. M. J. Johansson, M. Grätzel, E. I. Kauppinen, 13363.
A. Hagfeldt, G. Boschloo, Energy Environ. Sci. 2016, 9, 461. [140] S. Zhu, X. Yao, Q. Ren, C. Zheng, S. Li, Y. Tong, B. Shi, S. Guo,
[125] H. Wei, J. Xiao, Y. Yang, S. Lv, J. Shi, X. Xu, J. Dong, Y. Luo, D. Li, L. Fan, H. Ren, C. Wei, B. Li, Y. Ding, Q. Huang, Y. Li, Y. Zhao,
Q. Meng, Carbon 2015, 93, 861. X. Zhang, Nano Energy 2018, 45, 280.
[126] F. Zhang, X. Yang, M. Cheng, J. Li, W. Wang, H. Wang, L. Sun, [141] F. Fu, T. Feurer, T. Jäger, E. Avancini, B. Bissig, S. Yoon,
J. Mater. Chem. A 2015, 3, 24272. S. Buecheler, A. N. Tiwari, Nat. Commun. 2015, 6, 8932.
[127] S. Liu, K. Cao, H. Li, J. Song, J. Han, Y. Shen, M. Wang, Sol. Energy [142] F. Fu, T. Feurer, T. P. Weiss, S. Pisoni, E. Avancini, C. Andres,
2017, 144, 158. S. Buecheler, A. N. Tiwari, Nat. Energy 2017, 2, 16190.
[128] K. Aitola, K. Domanski, J.-P. Correa-Baena, K. Sveinbjörnsson, [143] F. Sahli, J. Werner, B. A. Kamino, M. Bräuninger, R. Monnard,
M. Saliba, A. Abate, M. Grätzel, E. Kauppinen, E. M. J. Johansson, B. Paviet-Salomon, L. Barraud, L. Ding, J. J. Diaz Leon,
W. Tress, A. Hagfeldt, G. Boschloo, Adv. Mater. 2017, 29, 1606398. D. Sacchetto, G. Cattaneo, M. Despeisse, M. Boccard, S. Nicolay,
[129] Y. Yang, H. Chen, X. Zheng, X. Meng, T. Zhang, C. Hu, Y. Bai, Q. Jeangros, B. Niesen, C. Ballif, Nat. Mater. 2018, 17, 820.
S. Xiao, S. Yang, Nano Energy 2017, 42, 322. [144] F. Jiang, T. Liu, S. Zeng, Q. Zhao, X. Min, Z. Li, J. Tong, W. Meng,
[130] V. Trifiletti, V. Roiati, S. Colella, R. Giannuzzi, L. De Marco, S. Xiong, Y. Zhou, Opt. Express 2015, 23, 3748.
A. Rizzo, M. Manca, A. Listorti, G. Gigli, ACS Appl. Mater. [145] Y. Jiang, B. Luo, F. Jiang, F. Jiang, C. Fuentes-Hernandez, T. Liu,
Interfaces 2015, 7, 4283. L. Mao, S. Xiong, Z. Li, T. Wang, B. Kippelen, Y. Zhou, Nano Lett.
[131] T. Y. Chiang, G. L. Fan, J. Y. Jeng, K. C. Chen, P. Chen, 2016, 16, 7829.
T. C. Wen, T. F. Guo, K. T. Wong, ACS Appl. Mater. Interfaces 2015, [146] S. Wang, K. Hou, Y. Xing, Q. Dong, K. Wang, Y. Lv, Y. Shi, Energy
7, 24973. Environ. Sci. 2017, 10, 2117.

Advanced Materials - 2019 - Li - Low Cost Counter Electrode Materials For Dye Sensitized and Perovskite Solar Cells

Uploaded by

Copyright:

Available Formats

Advanced Materials - 2019 - Li - Low Cost Counter Electrode Materials For Dye Sensitized and Perovskite Solar Cells

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Advanced Materials - 2019 - Li - Low Cost Counter Electrode Materials For Dye Sensitized and Perovskite Solar Cells

Uploaded by

Copyright:

Available Formats

REVIEW

Next-Generation Solar Cells www.advmat.de

Low-Cost Counter-Electrode Materials for Dye-Sensitized

Dye-sensitized solar cells (DSSCs) and all-

of electron–hole pairs under irradiation. A compact titania

3. Current Developments in Counter Electrode

3.4.2. Sulfides and Oxides

Wang and Gratzel et al. introduce metal sulfides into counter

level and good electrical conductivity.[106,124]

on the inverted structured PSCs (p-i-n PSCs).[110,111,136,137] In

stable. In cases of carbon nanotubes (CNTs),

You might also like