Recent advances in counter electrodes of quantum dot-sensitized solar cells

Abstract

Over the past few decades, various types of solar cells provided alternative ways for solar energy conversion. Among them, quantum dot-sensitized solar cells (QDSCs) have gained significant interest due to the advantages of quantum dots (QDs) including easy fabrication, multiple exciton generation, band-gap energy controllability and high absorption coefficient. A QDSC consists of a metal oxide photoanode, QDs, electrolyte and a counter electrode (CE). In comparison with the photoanode and QDs, the CE has not been paid much attention. As an essential part of QDSCs, the CE plays an important role in the charge transport and collection of the device. Here, the recent progress in the development of CEs is reviewed, and the key issues for the materials, structures and performance evaluation of CEs are also addressed.

---

Shixun Wang

Shixun Wang obtained his Bachelor degree from University of Science and Technology Beijing (USTB) in 2014, and he is currently a postgraduate student under the supervision of Professor Jianjun Tian in the Institute for Advanced Materials Technology at USTB. His research interests are focused on the functional materials and solar cells.

Jianjun Tian

Jianjun Tian is a professor of the Institute for Advanced Materials Technology at University of Science and Technology Beijing (USTB). He earned a PhD degree in material science from USTB in 2007. He worked as visiting professor at University of Washington from 2011 to 2012. Now he is director of Institute of Functional Materials at USTB. His current research is focused mainly on quantum dots and their application for the power conversion devices.

---

1 Introduction

The development of cost-effectiveness, easy fabrication and environment-friendly energy sources is an urgent subject to replace fossil fuels and solve the emergence of global warming and emission of greenhouse gases.^1,2 As a type of third generation solar cell, quantum dot-sensitized solar cells (QDSCs) are regarded as a derivation of dye-sensitized solar cells (DSCs) and have attracted considerable attention from scientists worldwide. DSCs typically consist of a TiO₂ mesoporous photoanode, a dye sensitizer, electrolyte and a Pt counter electrode (CE). Since the outstanding work done by O'Regan and Grätzel in 1991,³ this cell structure has received quite a lot of attention over the past 24 years. Although a power conversion efficiency (PCE) of 14.3% has been recorded for optimized DSCs,⁴ the high cost and complicated synthesis procedures of organic dyes along with limited photoelectric conversion efficiency restrict the commercialization of DSCs. Unlike DSCs, QDSCs have their own distinct advantages, such as tunable band gap upon the size of quantum dots (QDs),⁵ high extinction coefficients,⁶ large intrinsic dipole moment,⁷ multiple exciton generation,^8,9 and facile fabrication processes.¹⁰ QDs are a special class of semiconductors with diameter less than 10 nm. Thus, QDs have various advantages including tunable bandgap and the ability to form intermediate bands, which could make the theoretical maximum conversion efficiency as high as 44%.¹¹ So QDSCs using QDs as sensitizers to replace organic dyes have been considered as promising alternatives to DSCs.¹² Normally, QDs are composed of periodic groups of II–VI, III–V, or IV–VI materials such as CdS,¹³ CdSe,¹⁴ CdTe,¹⁵ PbS,¹⁶ InP,^17,18 InAs,¹⁹ Ag₂S,²⁰ Ag₂Se,²¹ Bi₂S₃ (ref. 22) and ternary CuInS₂ QD.^23,24 In the past five years, the highest PCE records from about 5% to ∼11.6% have been continuously created by researchers.^25,26 Besides, QD materials for sensitized solar cells should have a high absorption coefficient and matched band gap energy with metal oxide semiconductor (MOS) to effectively absorb sun-light.

The representative structure of QDSC consists of a metal oxide semiconductor (MOS) photoanode (usually TiO₂ or ZnO), QDs (sensitizer), an polysulfide electrolyte and a CE, as shown in Fig. 1(a).²⁷ Owing to an enlarged surface area to load massive sensitizer for the enhancement of light absorption and provide electrically smooth pathways for electrons transferring to the end of photo-electrode, nanostructured MOS usually used as the supporting frame of QDs. Commonly, metal oxide semiconductors such as TiO₂ and ZnO are extensively used in QDSCs.^28–30 The MOS photoanode is fabricated by coating nanostructured wide bandgap semiconductor with an optimum thickness about 11 μm on the conducting glass.³¹ Then, the quantum-dot sensitizers are adsorbed onto the surface of MOS to harvest photons. The widely used electrolytes in DSCs are organic electrolytes with I⁻/I₃⁻ redox couple. However, I⁻/I₃⁻ redox couple is corrosive to most QDs and leading to the deterioration of QDSCs performance. Thus, it is critical to explore an appropriate non-corrosive electrolyte for stable QDSCs. Electrolyte with polysulfide redox couple (S²⁻/S_x²⁻) is introduced to provide a stable environment for QDs. And the composition of polysulfide electrolyte was tailored by methanol to improve penetration of the electrolyte into photoanode.^32,33 Quasi-solid-state polysulfide electrolyte with the assistant of polymer was also reported to improve the cell stability.^34,35 In order to improve the performance of QDSCs, several electrolytes have been reported based on different redox couple such as the Fe²⁺/Fe³⁺, Fe(CN)₆³⁻/Fe(CN)₆⁴⁻, Co²⁺/Co³⁺ and (CH₃)₄N₂S/(CH₃)₄N₂S_n.^34–42 QDSCs assembled with electrolytes based on the redox system above got relative high open circuit voltage than cells packaged polysulfide electrolyte. However, QDSCs employing polysulfide electrolyte have shown the highest short circuit current and power efficiency among them.^43,44 A Pt-based CE has been widely studied in a typical DSCs device due to its excellent catalytic ability to reduce I₃⁻ to I⁻.⁴⁵ Nevertheless, Pt-based CEs are unsuitable in a polysulfide based QDSC device for their strong chemisorption with polysulfide redox couple (S²⁻/S_x²⁻), leading to a reduced catalytic activity of CEs.^46–48 Recently, the copper chalcogenides are used as CEs for the highly efficient QDSCs due to excellent catalytic performance and good compatibility with polysulfide electrolyte.^49,50

Fig. 1 (a) Schematic illustration of a typical QDSC device; (b) interfacial charge transfer processes following a laser pulse excitation. Reproduced with permission from ref. 27. Copyright 2015 American Chemical Society.

Fig. 1(b) shows photoinduced charge transfer process. When the cell is exposed to continuous illumination, QDs will absorb light within a certain range of wavelengths based on their band gap energy and (①) excite electrons from the valence band (VB) of QDs to their conduction band (CB). (②) Electrons in the CB of QDs are injected into the CB of MOS, and then transferred along the TiO₂ nanocrystallites and the external circuit to the CE. At the same time, (③) the holes are released by redox couples in the electrolyte. (④) The oxidized electrolyte formed as the oxidation proceeds is reduced at the CE. Eventually, the reduced electrolyte will be transported to the photoanode to combine with the holes again.^27,51 With the driving force provided by the energy-level difference between the aligned Fermi levels at the QDs/MOS and the redox potential of the electrolyte, the electrons are theoretically able to circulate the cell over and over again. However, (⑤) recombination of electrons from QDs and the oxidized form of the redox couple, and (⑥) interfacial recombination of electrons from MOS (TiO₂ in Fig. 1(b)) and the oxidized form of the redox couple will be a major factor in limiting the overall efficiency. Thus, many researchers devote themselves to the modification of the photoanode to reduce undesired recombination. A tremendous amount of work has been done to optimize the performance of QDSCs including exploiting various QD sensitizer,^{25,26,52–54} electrolytes,^32,55,56 and counter electrodes (CEs).^50,57–59 Among them, CEs have been the least studied.

Actually, CE plays an extremely important role in the structure of QDSC to collect electrons from the external circuit and actuate catalysis of the reduction of oxidized species in electrolyte. That is, CE should catalyze the regeneration of the redox in electrolyte after its action to take back the oxidized sensitizers to the ground state and thus keep the device active and stable. Generally, CEs with high catalytic activity, large surface area, chemical durability and low production cost are indispensable to enhance the performance of QDSC. Besides, the conductivity of CEs corresponding to the kinetics of charge transfer at the CE/electrolyte interface will significantly affect the fill factor (FF) and short circuit current. Thus, the research of alternative catalytic materials with high conductivity to substitute conventional Pt-based CE is an urgent way to improve the conversion efficiency of QDSCs. In brief, the optimization of CEs with developed photoanode may contribute to a prominent result in increasing PCE and stability of QDSCs in the foreseeable future. This short review will concentrate on an overall retrospection of QDSCs counter electrode.

2 Counter electrode materials and structures

In this section, we will make a summary of the categories of CEs for QDSCs, and then recount typical research details further. Generally, the reported CEs for QDSCs can be roughly classified into five categories: metal-based counter electrodes, metallic compounds, carbon derivatives, polymers, and composite counter electrodes (combination of the four categories above).

2.1 Metal-based counter electrodes

Pt was first used as counter electrode (CE) in the work about CdS based QDSC done by Vogel et al. in 1990.⁶¹ As more attention paid to the study of CE, Bisquert et al. found that polysulfide redox couple (S²⁻/S_x²⁻) can absorb onto the surface of Pt CE, leading to the depression of electrocatalytic activity due to the decrease of the catalyst active surface and increase of charge-transfer resistance at the electrolyte/CE interface.^62,63 And then researchers dedicate to solve the problem by the structural improvements and surface-reinforcement of Pt CE. Wu et al. improved the efficiency of QDSCs by using Pt counter electrode with rough surface.⁶⁴ Raj et al. fabricated Pt CE coated with a thin passivation layer of copper sulfide, which can efficiently restrain the chemisorption between Pt and polysulfide redox couple (S²⁻/S_x²⁻).⁶⁵ Besides, Au CE was proved to be more tolerant to the polysulfide redox couple with high PEC (CdS/CdSe based QDSCs) up to 4.22% in the work done by Lee and Lo.⁶⁶ After surveying a range of such metal-based CE, Au has shown better performance in QDSCs when used as CE compared with Pt. Compared with simple metal, alloyed metal may achieve higher efficiency. Du et al. reported Co–Ru alloyed CE which delivered enhanced PCE of 3.04% based on CdS QDSC.⁶⁷ Dao et al. fabricated AuPt bimetallic nanoparticles (AuPt-BNPs) on the surface of (fluorine doped tin oxide) FTO glass using dry plasma reduction (DPR) as shown in Fig. 2(a). This kind of CE can be applied as a transparent CE for QDSC. Fig. 2(b) clearly depicts the high-light transmission of metal nanoparticles (NPs) based CEs except sputtered Au CE. With the use of AuPt-BNPs CE, they finally achieved enhanced PCE of 3.4% of CdS/CdSe based QDSCs in comparison with Au NPs (1.1%), Pt NPs (3.2%), and Au sputtered which has bad transmittance.⁶⁰ The highlight of the work introduced above is the fabrication of high-light transmission CE with considerable catalytic activity and conductivity which can further develop the transparent and efficient QDSCs. However, the research of alternative CE is still imperative, taking the high costs and poor catalytic ability of noble metal into considerations.

Fig. 2 (a) The schematic diagram of synthesis of (Pt NPs, Au NPs, and AuPt-BNPs) CEs using DPR method and a QDSC with a CdSe/CdS/ZnO-nano wire photoanode and NPs/FTO CE; (b) the transmission spectra in a range of 300–800 nm of the different CEs; (c) current density–voltage (I–V) characterization of QDSC assembled with different CEs. Reproduced with permission from ref. 60. Copyright 2014 Elsevier B.V. All rights reserved.

2.2 Polymer-based counter electrodes

Conducting polymers has become alternative materials of Pt as CE materials on account of the outstanding advantages including high conductivity and large electrochemical surface area.⁶⁹ Shu et al. used TiO₂ as a porous support material to fabricate TiO₂–PEDOT CE and achieved PCE of 1.56% which is higher than cells assembled with the pure PEFOT CE (1.30%) as shown in Fig. 3.⁷⁰ Besides, Shu et al. utilized an organic thiolate/disulfide redox couple [C₇H₅N₄S⁻/C₁₄H₁₀N₈S₂, (AT⁻/BAT)] and a polymer CE [poly(3,4-ethylenedioxythiophene), PEDOT] to assemble with CdS based QDSCs and obtained PCE of 1.53%. The PEDOT_HDC CE and PEDOT_LDC CE in this work were prepared by applying a constant potential until a charge capacities of 100 mC cm⁻² and 2 mC cm⁻² was reached, respectively. The obtained polymer-based CE are depicted in Fig. 3(a). The porous structure of PEDOT_HDC enhanced the electrolyte penetration and increased the electrocatalytic area, which in turn favors higher short circuit current density (J_sc, 3.88 mA cm⁻²) than of PEDOT_LDC. However, polymer films obtained by electro-polymerization tends to form large pores and big aggregation and easy to crack and fall with increasing thickness, leading to the reduction of electrocatalytic ability and robustness.⁷¹

Fig. 3 (a) The SEM images of PEDOT_HDC and PEDOT_LDC CEs; (b) I–V characterization of CdS based QDSCs assembled with different CEs. Reproduced with permission from ref. 68. Copyright 2014 Elsevier Ltd. All rights reserved. Attention: the curves in (b) should exchange the colors with each other.

2.3 Carbon-based counter electrodes

Carbon is an alternative candidate for Pt-based CEs due to its competitive prices, heat resistance, sufficient conductivity and corrosion–inertness towards polysulfide redox couple. To achieve high efficiency cells with carbon-based CEs, the CE should have large surface area to compensate the relatively slow intrinsic reaction kinetics of carbon with polysulfide electrolyte, meanwhile, provide enough seepage sites for electrolyte into the carbon layer. Thus, various form of carbon-based CE used in QDSCS emerged, such as activated carbon,⁷² carbon nanofibers,⁷³ fullerene,⁷⁴ mesocellular carbon foam,^75,76 hollow core/mesoporous shell carbon (HCMSC)⁷⁷ and carbon nanotubes (CNT), which achieved relatively high PEC assembled with CdS/CdSe based QDSCs (4.67%) among them according the study of Gopi et al.⁷⁸ In their work, the CNT film was prepared by the doctor blade method on the well-cleaned FTO glass. The structure of assembled QDSCs and obtained I–V curve is shown in Fig. 4. QDSCs using CNT-based CE achieved a higher PEC of 4.6% than that with CuS CE (3.6%).

Fig. 4 I–V characterization of QDSC assembled with CNT-based CE and CuS CE. The schematic diagram of QDSC assembled with CNT-based CE is inserted. Reproduced with permission from ref. 78. Copyright 2016 Elsevier B.V. All rights reserved.

Gopi et al. take a series of researches to investigate the reason of the superior performance of CNT-based CE. As shown in Fig. 5, the cyclic voltammetry (CV) was firstly carried out. The negative and positive peaks correspond to the reduction of S_n²⁻ and the oxidation of nS²⁻, respectively. The CE in the QDSCs serves as a catalyst for reducing the redox couple in the electrolyte. Thus, the cathodic reduction peak current in CV curves can be applied to evaluate the electrocatalytic ability of the CEs. From Fig. 5(a), CNT-based CE obtained the largest cathodic reduction peak current, which indicates the superior performance of it. Besides, the peak-to-peak separation (E_pp) value should also be taken into consideration. CNT-based CE has lower value of E_pp (1.16 V) than CuS CE (1.27 V). It can mainly be attributed to the large surface area which supplied enough seepage sites for electrolyte into the carbon nanotube layer of CNT-based CE besides the good conductivity, electrocatalytic ability and corrosion–inertness towards polysulfide electrolyte. Accordingly, CNT-based CE has better electrochemical-catalytic activities in comparison of conventional CuS CE.

Fig. 5 (a) Cyclic voltammetry (CV) of polysulfide redox couple on CNT, CuS and Pt CEs; (b) enlarged part of the CNT and CuS electrodes CV plots. Reproduced with permission from ref. 78. Copyright 2016 Elsevier B.V. All rights reserved.

2.4 Metallic compound counter electrodes

Metallic compounds have their own bandgap energy, which have drawn great attention as a highly useful component of CEs in QDSCs in comparison with Pt based CEs. Thus, researchers devote to investigate kinds of metallic compounds and their applications in photovoltaic cells. Diverse types of metallic compounds including Cu_xS,⁵⁸ CuS,⁷⁹ PbS,⁸⁰ FeS,⁴⁷ CoS₂,⁸¹ NiCo₂S₄,⁸² Cu_xSe,⁸³ Cu₂SnSe₃ (ref. 84) and Cu₂ZnSnS(Se)₄ (ref. 85) have been reported as candidates for the CE material. In this section, some major results of the researches on metallic compound CEs especially about metal chalcogenide are covered.

2.4.1 Copper chalcogenide counter electrodes. Copper sulfides have received enough attention as an important component in photovoltaic cells, since the discovery of CdS/Cu₂S heterojunction solar cell in the work done by Reynolds et al. in 1954.⁸⁶ Owing to the relatively high conductivity and catalytic activity to polysulfide electrolyte, copper sulfides are commonly used in QDSCs in comparison with traditional CE materials.

Cu₂S, a p-type semiconductor, with bandgap of 1.1–1.4 eV has been widely used in QDSCs as CE materials. Generally, Cu₂S CEs are fabricated by dipping the etched brass foils into polysulfide electrolyte to form compact Cu₂S film.¹³ But, the poor stability and relatively low surface of the Cu₂S compact film still affect its performance as a CE. On account of those problems, novel Cu₂S CE preparation methods were developed in order to achieve CE with better stability, conductivity and catalytic activity. Meng et al. applied electroplating method to deposit a thin layer of copper on the surface of FTO and then immersed the copper coated glass in polysulfide solution to prepare Cu₂S CE.^87,88 As shown in Fig. 6, the final flower-like Cu₂S film has a thickness between 100 and 200 nm. The favorable porous structure makes the film better catalytic ability. Fig. 7 shows that the interface charge transfer resistance of electrolyte and Cu₂S film is calculated to be a relatively low value of 5.7 Ω cm² and the CdS based QDSC assembled with this CE obtained a promising PCE of 2.6%.

Fig. 6 SEM images of (a) bare FTO glass, (b) copper coated FTO glass, (c) Cu₂S coated FTO glass and (d) cross-sectional SEM image of Cu₂S coated glass. Reproduced from ref. 88 with permission from the Royal Society of Chemistry.

Fig. 7 (a) Nyquist plots of symmetric cells based on different CEs; (b) I–V curves of cells assembled with different kinds of CEs. Reproduced from ref. 88 with permission from the Royal Society of Chemistry.

According to the idea of increasing the surface area of copper sulfides to provide more active catalytic sites through novel preparation method, copper sulfide with various morphologies were reported including porous hierarchical structure,⁸⁹ nanoplates,⁹⁰ nanosheets,⁹¹ nanoparticals⁹² and so on. Hu et al. reported a novel ITO@Cu₂S based CE for QDSC with the ability of divergent thinking.⁹³ It is another way to increasing the contact area between electrolyte and CE. The preparation process is shown in Fig. 8: (a) gold NPs coated FTO glasses were firstly prepared as catalysis via a sputtering coater and then (b) high-purity metallic indium and tin powders were inducted to synthesis ITO nanowire arrays; (c) chemical bath deposition (CBD) technique was used to grow CdS shell on the surface of ITO nanowires; (d) a solution based cation exchange was then carried out to convert CdS into Cu₂S; (e) the obtained ITO@Cu₂S nanowire arrays was used as CE to assemble with CdS/CdSe/ZnS photoanode. The formation of high-quality tunnel junctions with short carrier transport path (<100 nm) between core and sheath significantly reduced the sheet resistant of CE and enhanced electrons transfer from ITO to Cu₂S. The CdS/CdSe based QDSCs assembled with ITO@Cu₂S CE which have ITO nanowires of 8–10 μm achieved enhanced PCE of 4.06% in comparison with Au-based CE (2.20%).

Fig. 8 (a) Scheme for preparing ITO@Cu₂S nanowire arrays on FTO glass; (b) SEM image of ITO nanowires; (c) low-magnification and (d) high-magnification SEM images of ITO@Cu₂S. Reproduced with permission from ref. 93. Copyright 2014 American Chemical Society.

Hu et al. further developed the preparation method of ITO@Cu₂S CE.⁹⁴ Hierarchically assembled ITO@Cu₂S nanowire arrays were fabricated as efficient CdSe_xTe_1−x based QDSCs CEs. The main differences of the preparation process lie in the following aspects: the hierarchical ITO nanowire arrays were prepared by repeating the procedure (a) two times (ITO-II) or three times (ITO-III). And the performance of QDSCs assembled with ITO-I@Cu₂S, ITO-II@Cu₂S, and ITO-III@Cu₂S CE were throughout investigated in their work. As shown in Fig. 9, the smaller third-generation branches of ITO-III were sequentially grown on the surfaces of the second-generation branches of ITO-II via an additional (chemical vapor deposition) CVD process. After the CBD and cation exchange process, the smooth surfaces of ITO nanowire arrays' stems and branches became rough. The length of ITO nanowire stem and the branches of ITO-II and ITO-III were 8–10, 4–5, and 0.5–1 μm. As depicted in Fig. 10(a), the QDSC with ITO@Cu₂S nanowire CEs exhibited higher short circuit current density (J_sc) and open circuit voltage (V_oc) than that with the brass/Cu₂S CE. The monochromatic incident photo-to-electron conversion (IPCE) measurements in Fig. 10(b) indicated the enhanced IPCE between wavelengths of 450–700 nm for QDSCs with ITO@Cu₂S CEs compared with the QDSCs assembled with brass/Cu₂S CE. The increased J_sc can be attributed to the improved utilization of the incident light because of the light scattering caused by the three-dimensional nanowire arrays of the ITO@Cu₂S CEs.

Fig. 9 SEM imaged of hierarchical ITO nanowire arrays before and after the coaxial growth of the Cu₂S shell: SEM image of (a) ITO-II, (b) ITO@Cu₂S-II, (c) ITO-III, and (d) ITO@Cu₂S-III. Reproduced with permission from ref. 94. Copyright 2015 American Chemical Society.

Fig. 10 (a) I–V characterization and (b) IPCE spectra of the QDSCs with various ITO@Cu₂S and brass/Cu₂S CEs. Reproduced with permission from ref. 94. Copyright 2015 American Chemical Society.

To further evaluate the properties of ITO@Cu₂S CEs, resistance–voltage (R–V) measurement were performed. The results of R–V measurement indicate the decrease in the series resistance (R_s) which reduced the loss of photovoltage and the increase in the shunt resistance (R_sh) which reduced the loss of photocurrent of QDSC assembled with brass/Cu₂S CE. According to the eqn (1) below, the increase in the light generated current caused the increase in the V_oc. Thus, the QDSCs with ITO@Cu₂S-III CE achieved increased V_oc compared with that with the brass/Cu₂S.

To evaluate the catalytic activity of the various ITO@Cu₂S CEs, CV measurement were carried out. QDSCs with ITO@Cu₂S-III CE obtained highest peak current at reductive potentials which indicated better electrocatalytic ability of it. Besides, 10 times cycling was performed at a scan rate of 50 mV while little change was observed in the curve shape or current density demonstrating the excellent chemical ability of ITO@Cu₂S CEs. A Tafel polarization measurement was further conducted on various CEs. The ITO@Cu₂S-III CE symmetric cells got the largest exchange current density (J₀) which is inversely proportional to R_ct. Thus, the smallest R_ct was achieved for ITO@Cu₂S-III CE indicating the superior catalytic ability. Moreover, ITO@Cu₂S-III CE had higher value of the limiting current density (J_lim) among them which shown larger diffusion coefficient in the electrolyte and better catalytic ability. According to the electrochemical impedance spectroscopy (EIS) measurements on symmetrical cells, ITO@Cu₂S-III CE exhibited relatively small intrinsic sheet resistance (R_h, 5.05 Ω) and R_ct (1.30 Ω). They can be attributed to the formation of conductive networks which enhanced the efficient electronic transmissions in the three-dimensional nanowire array structure as well as to the formation of an effective tunneling junction between the n-type degenerate ITO nanowires and the p-type degenerate Cu₂S nanocrystals.⁹³ Therefore, QDSC assembled with ITO@Cu₂S-III CE achieved higher PCE of 6.12% compared with brass/Cu₂S based QDSCs (5.05%).

However, the electrical conductivity ability and erosion problem caused by polysulfide electrolyte still limit the further development of QDSCs assembled with copper sulfide CE. Recently, researchers report that copper selenide (Cu_xSe) shows the potential to be the CEs of QDSCs due to the excellent catalytic activity and lower transfer resistance than that of copper sulfide.^95–98 In the work reported by Zhong et al.,⁵⁰ Cu_xSe nanoparticles were prepared through aqueous solution method and further made into Cu_xSe pastes to cover on the surface of FTO (to fabricate Cu_xSe/FTO CE) via screen-printing method. The experimental conditions for the preparation of Cu_xSe/FTO CE were further optimized through the resultant optimal values of 1 : 4 for Cu : Se ratio, annealing at 450 °C for 30 min and 2.6 μm for film thickness. The achieved maximum PCE Cu_xSe/FTO based QDSCs are 6.49%, higher than Cu_xS/brass CE based cells (6.35%). As depicted in Fig. 11(a), the statistical results show the excellent performance Cu_xSe/FTO CEs based QDSCs with PECs generally locating at 5.8–6.1%. Besides, Nyquist plots and Tafel polarization obtained using symmetric dummy cells ascertained that the catalytic activity of Cu_xSe/FTO CE were better than Cu_xS/brass CE. The outstanding stability of Cu_xSe/FTO CE based QDSCs were also presented in Fig. 11(d). Devices assembled with Cu_xSe/FTO still maintained their initial activity due to the stability of CEs. The stability of CE mainly reflects in its corrosion-inertness towards polysulfide electrolyte. In recent years, researchers have devoted themselves to producing CE material with large specific surface area which have the ability to maintain its chemical stability towards the corresponding electrolyte.

Fig. 11 (a) Statistical results of PCEs for CdSe QDSCs assembled with Cu_xSe/FTO and Cu_xS/brass CEs; (b) Nyquist plots of symmetric cells using Cu_xSe/FTO and Cu_xS/brass CEs; (c) Tafel polarization curves of symmetric cells used in EIS measurements; (d) high stability of CdSe QDSCs assembled with Cu_xSe/FTO CE characterized by PCE and photocurrent. Reproduced with permission from ref. 50. Copyright 2016 Elsevier Ltd. All rights reserved.

Unlike the aqueous solution method introduced above, our group reported a Cu₃Se₂ nanostructured CE which directly grow on FTO to form double-layer morphologies through successive ionic layer adsorption reaction (SILAR) and CBD process (Fig. 12(a)).⁹⁹ The CBD duration which would directly affect the formation of the double-layer structure was firstly studied. The electrochemical characteristics of Cu₃Se₂ CE were investigated by EIS analysis and compared with CEs prepared by different CBD hour. In Fig. 12(d), the semicircle corresponds to the electron transfer at the TiO₂/QDs/electrolyte interface and transport in TiO₂ film (R_ct). QDSCs assembled with Cu₃Se₂-3 h CE achieved the highest R_ct value (172.9 Ω) among them. The Bode plots in Fig. 12(e) shown the considerable electron lifetime of QDSCs with Cu₃Se₂-3 h CE (10.07 ms). The Tafel spectra of symmetrical dummy cells in Fig. 12(f) depicts the J_o value of Cu₃Se₂-3 h CEs is the highest among them, which means the excellent electrocatalytic ability of them. The measurement results can be ascribed to the high crystallinity, proper crystal defects and large surface area and outstanding electron transfer pathway in counter electrodes.

Fig. 12 (a) The preparation processes of Cu₃Se₂; (b) SEM image and (c) cross-sectional SEM image of Cu₃Se₂-based CE treated by 3 h of CBD process; (d) electrochemical impedance spectra and (e) Bode phase plot of the QDSCs based on the three different CEs under forward bias (−0.6 V) and dark conditions; (f) Tafel polarization curves for the different CEs. Reproduced from ref. 99 with permission from the Royal Society of Chemistry.

In the system of CdS/CdSe based QDSCs, the performance of Cu₃Se₂-based CE was further studied in comparison with brass/Cu_xS based CE. Fig. 13(a) depicted that the device assembled with Cu₃Se₂ CEs achieved higher fill factor, larger short circuit current and open circuit voltage than that with Cu_xS CE. The possible reasons are ascribed to the large surface area, proper crystal defects, high conductivity and good electrocatalytic activity of Cu₃Se₂ nanostructure, which further accelerate the electron transfer and the reaction of redox in polysulfide electrolyte, and finally result in low charge recombination and high conversion efficiency (best PCE of 5.05%). Generally, both approaches above mentioned are effective in case of Cu₃Se₂ CEs preparation. The performance of QDSC using Cu₃Se₂ CE has been studied in our work, in which the PCE shown excellent stability in comparison with Cu_xS/brass based devices. As shown in Fig. 13(b), the PCE of QDSCs with Cu₃Se₂ CE changed from 4.95% to 4.81% with a negligible decline of about 2.8%. However, the QDSCs assembled with CuS CE shown relatively low stability and their PCE changed from the initial 4.03% to the final 2.24% with an obvious decline of about 44%. It is evident that Cu₃Se₂/FTO CE which has larger specific surface area shows remarkable stability.

Fig. 13 (a) J–V curves of QDSCs assembled with different CEs, the inset is the photographs of Cu_xS and Cu₃Se₂/FTO CEs; (b) stability of QDSCs assembled with different CEs. Reproduced from ref. 99 with permission from the Royal Society of Chemistry.

2.4.2 Other metallic compound counter electrodes. As regards metallic compound CE materials for QDSCs, usually metal sulfides and selenides, the most important abilities required are high conductivity and good catalytic activity and particularly the ability to resist the corrosion of polysulfide electrolyte. Other metal sulfides here, including lead sulfide, nickel sulfide, cobalt sulfide and iron sulfide are all investigated in QDSCs when applied as CEs.^{47,80–82,100,101} PbS based CE shown promising catalytic ability among them. Lin et al. reported a photoactive PbS film synthesized by successive ionic solution and reaction (SISCR) and finally achieved excellent performance in CuInS₂/CdS based QDSCs (PCE of 4.7%).⁸⁰

As depicted in Fig. 14, the PbS film outperforms Pt and CuS film even though CuS has a much higher catalytic ability in polysulfide electrolyte than PbS. It can be attributed to the photoactive characteristics of PbS CE which increased the photocurrent of the resulting QDSCs. The PbS film indicated its absorption of the entire range of visible light and even into the IR region to compensate the shortage of photoanode on light absorption. Thus, the PCE of cells assembled with PbS based CE increased due to the marked increment of J_sc and V_oc.

Fig. 14 (a) Schematic energy diagram representing the entire charge-transfer process in the QDSCs assembled with the PbS CE; (b) SEM image of PbS/FTO CE deposited by 5 SISCR cycles; (c) Nyquist impedance plots of different CEs; (d) I–V curves of PbS, CuS and Pt based QDSCs. Reproduced from ref. 80 with permission from the Royal Society of Chemistry.

Semiconducting metal selenides, as an essential class of chalcogenides, have attracted enormous attention on account of their distinctive electronic properties and conspicuous chemical behaviors. Recently, metal selenides have emerged as a new class of CE electrocatalysts to replace conventional CE materials of QDSCs. Choi et al. performed a throughout investigation on binary metal selenides to assess their feasibility as CE materials for CdS/CdSe based QDSCs.¹⁰² Eight different types of binary selenides (MnSe, CoSe₂, NiSe₂, Cu_1.8Se, MoSe₂, WSe₂, PbSe, and Bi₂Se₃) were randomly chosen as electrocatalyst candidates for CE materials in their study. And the photoelectric transition property of solar cells assembled with these metal selenides were compared with conventional Pt CE based QDSCs. As shown in Fig. 15, QDSCs assembled with Cu_1.8Se or PbSe CEs shows better overall photoelectric property than that with Pt, while inferior performances were obtained when using other CEs according to the I–V curves. The study on Cu_1.8Se and PbSe CEs were further carried out to investigate their physical and electrocatalyst properties. Cu_1.8S has a considerable stability while PbSe was partly oxidized to PbO and SeO₂ in air. Besides, EIS measurement was also performed to approve Cu_1.8S CE had better catalytic ability and electrochemical stability when functioning in polysulfide electrolyte in comparison with PbSe CE. PCE of 5.01% was achieved, applying Cu_1.8S CE.

Fig. 15 I–V curves of QDSCs assembled with various metal selenide CEs under 1 sun illumination. Reproduced with permission from ref. 102. Copyright 2014 American Chemical Society.

2.5 Composite counter electrodes

The categories of QDSCs' CE materials introduced above comprise noble metals, metallic compounds, carbon derivatives and conducting polymers. However, all of these types have their distinctive characteristics on conductivity, electrocatalytic ability and stability to polysulfide electrolyte. In this aspect, researchers combine two (or more) of them together to achieve CE materials with higher electrochemistry performance than any of them alone. Generally, they were combination of sulfide/metal, sulfide/sulfide, sulfide/carbon derivation, and metal/carbon derivation. Thus, various research about composite CEs have been reported including CuS/Pt,⁶⁵ CuS/CoS,¹⁰³ NiS/PbS,¹⁰⁴ Cu₂S/RGO,⁴⁸ PbS/carbon black,¹⁰⁵ CuInS₂/Carbon composite,¹⁰⁶ carbon dot/Au,¹⁰⁷ PVP–CuS.¹⁰⁸ Among various reported composite CEs, the combination of Mo-compound particles (Mo₂N, Mo₂C, and MoS₂) and carbon nanotube (CNT)–reduced graphene oxide (RGO) has drawn greatly attention.¹⁰⁹

It is known to us that carbon composites have the advantages of large surface area, excellent electrical conductivity and relatively low costs. And Mo compounds are widely used as catalysts in scientific research about hydro-processing which proves their excellent electrocatalytic ability.¹¹⁰ Seol et al. successfully combined the advantages of both CNT and RGO and applied the composite in QDSCs as CE.¹⁰⁹ The CNT–RGO complex offered an excellent electron pathway due to the high electrical conductivity and a large surface area by preventing stacking of RGO and bundling of CNT. The reported performance of these composite CEs achieved a PCE of greater than 5% assembled with CdS/CdSe/ZnO nanowire photoanode. The research about Mo-compound/CNT–RGO provided the convincing evidence of the clearly enhanced electrocatalytic activities of the CE. The Mo-compound/CNT–RGO complex was fabricated according to a facile method involving a modified urea-glass route.¹¹¹ In brief, the Mo-compound (Mo₂N, Mo₂C, and MoS₂) anchored onto the CNT–RGO structure could be prepared by the reaction of controlling the application of urea or thiourea precursor and the ratio of the urea precursor to Mo precursor. As shown in Fig. 16(a) and (b), the achieved multiwalled CNTs are randomly distributed on the RGO layer and the Mo compounds are well-spread over the CNT–RGO support. The 1.5 μm-thick mesoporous film covering on the FTO was used as CE for CdS/CdSe based QDSCs containing polysulfide electrolyte. Besides, the charge transfer process that occurred in the Mo compound/CNT–RGO CE is clearly illustrated by Fig. 16(b). Electrons from external circuit are delivered to the Mo compound along the CNT–RGO composite structure and are used to reduce the oxidized electrolyte. The Mo compound here work as the critical factor in facilitating the catalytic reaction, whereas CNT–RGO composite acts as an excellent electron pathway to reveal its outstanding electron conductivity. Moreover, the larger surface area of CNT–RGO complex provided plenty of load sites and contribute to an increased amount of loading onto Mo compound catalysts.

Fig. 16 (a) Cross-sectional SEM image of Mo₂N/CNT–RGO CE; (b) electron-transfer mechanism of the MoX/CNT–RGO CE; (c) I–V curves and (d) photocurrent stability of QDSCs assemble with Au and the Mo-compound/CNT–RGO CE. Reproduced with permission from ref. 109. Copyright 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

To evaluate the performance of these novel composite CE, PCE was firstly carry out by using a solar simulator. The resultant I–V curves are presented in Fig. 16(c). Except for the MoS₂/CNT–RGO CE based QDSCs, the other QDSCs all shown improved performance compared to the conventional Au CE based QDSCs. QDSCs assembled with Mo₂N/CNT–RGO CE achieved the best performance due to the excellent catalytic ability of Mo₂N which greatly contribute to the fill factor and thus considered to be the most promising catalyst for use in QDSC as CE material. Moreover, Fig. 16(d) clearly demonstrate QDSCs with Mo₂N and Mo₂N maintained their photocurrent even after 5000 s of irradiation due to the noble metal-like catalytic behavior. EIS measurements and Tafel polarization tests were performed after the stability measurement to assure the encouraging effect of the introduction of the Mo₂N/CNT–RGO CE.

Recently, researchers further developed the application of CNT as CE in QDSCs. Sayan et al. reported a composite of a 15 wt% graphene oxide nanoribbon (GOR) and Cu_1.18S composite CE and achieved a record PCE of 3.55% for CdS based QDSCs compared with Pt CE or brass/Cu₂S CE.¹¹² In their work, GOR was produced by oxidative cleavage of the commercial MWCNT whereas graphene oxide (GO) was synthesized by the modified Hummers method. I–V curve depicted in Fig. 17 shows the excellent performances of QDSCs assembled with Cu_1.18S–RGO CE and Cu_1.18S–GO CE which achieved PCE of 3.38% and 3.22%, respectively. Besides, QDSCs with Cu_1.18S–RGO CEs obtained higher FF (0.57) compared with cells assembled with brass/Cu₂S CEs (0.50). It indicated the superior catalytic ability and conductivity of Cu_1.18S–RGO CEs to some degree. To further evaluate the properties of CEs, EIS measurement, Tafel polarization plots and CV analysis were subsequently carried out which shown excellent agreement with the conclusions above. As shown in Fig. 17(d), the Fermi level (E_F) of p-type Cu_1.18S is close to the valence band and thus will be in equilibrium with the reduction potential of the polysulfide under dark condition. Since the Highest Occupied Molecular Orbital (HOMO) level of GOR matches well with the reduction potential of polysulfide, it closely coincides with the E_F of Cu_1.18S. In QDSCs, when the electrons move from the photoanode to the CE to reduce the oxidized electrolyte, if the redox potential for electrolyte is close to the HOMO level of GOR, it will greatly facilitate the electrons migration than Cu_1.18S itself. Thus, the particular band alignment will reduce the carrier loss and finally enhance PCE. In addition, the HOMO level of GO is little higher above the redox potential of the electrolyte, and therefore the transfer of electrons will be hindered.

Fig. 17 (a) I–V characterization of CdS based QDSCs assembled with different CEs; (b) GOR synthesized by unzipping CNT; (c) GO; (d) energy band diagram of Cu_1.8S–GOR CE showing electron transfer. The energy levels of GO are also indicated. Reproduced from ref. 112 with permission from the Royal Society of Chemistry.

2.6 Summary of CE materials for QDSCs

For a handy overview of recent advances of CEs, the referenced CE materials in this review have been summarized in Table 1. As to a certain type of cells assembled with the elementally similar CEs, the details of cell which achieved the highest PCE are recorded in the table. Blanks present that the exact values of R_ct for CEs were not given. CEs based on metal chalcogenides shown relatively better performance without consideration of photoanode. In addition, the metal chalcogenide would be combined with and the carbon derivatives to increase the conductivity of CEs.

Table 1 Summary of CEs used in QDSCs and their corresponding parameters

CE materials	Preparation	Sensitizer	Isc/mA cm^−2	Voc/V	FF	Efficiency (%)	Rct
Pt^113	Thermal evaporation	CdS/CdSe	16.2	0.49	0.49	3.90
Au^66	Sputtering	CdS/CdSe	16.8	0.51	0.49	4.22
AuPt NPs^60	Dry plasma reduction	CdS/CdSe	16.5	0.65	0.31	3.40	54.2 Ω cm^2
Co–Ru^67	Hydrothermal method	CdS	9.6	0.59	0.53	3.04	21.7 Ω
PEDOT^70	Electro-polymerization	CdS	3.2	0.59	0.72	1.37	123.1 Ω cm^2
Polypyrrole (PPy)^69	Electro-polymerization	CdS	3.9	0.39	0.27	0.41	337.5 Ω cm^2
Polythiophene (PT)^69	Electro-polymerization	CdS	0.9	0.33	0.26	0.09	1267.5 Ω cm^2
TiO2–PEDOT^70	Electro-polymerization	CdS	5.5	0.42	0.67	1.56	4.3 Ω cm^2
PVP–CuS^108	Solvothermal	CdS/CdSe	17.5	0.58	0.51	5.22	1.4 Ω
Activated carbon^74	Doctor blade technique	CdS/CdSe	10.9	0.54	0.57	3.48	0.2 Ω cm^2
CNT^78	Doctor blade technique	CdS/CdSe	16.0	0.58	0.49	4.67	0.8 Ω
Carbon nanofibers^73	Nanocasting technology	CdSe	11.9	0.62	0.60	4.81	2.6 Ω cm^2
Fullerene^74	Doctor blade technique	CdS/CdSe	12.6	0.54	0.60	4.18	84 Ω
Mesocellular carbon foam^75	Carbonization from silica template	CdS/CdSe	12.6	0.68	0.42	3.60
Hollow core/mesoporous shell^77	Carbonization from silica template	CdSe	12.4	0.60	0.52	3.90	12 Ω cm^2
N-doped hollow carbon NPs^114	Detonation-assisted chemical vapor deposition (CVD)	CdS/CdSe	13.5	0.51	0.40	2.67	2.0 Ω cm^2
AuPt NP/reduced graphene nanoplatelets^115	dry plasma reduction	CdS/CdSe	15.2	0.72	0.40	4.50	34.3 Ω
Carbon dot/Au nanoraspberries^107	Doctor blade technique	CdS/CdSe	16.6	0.70	0.46	5.40	2.0 Ω cm^2
Mesoporous carbon/Ti^25	Silica template nanocasting/screen printing	Zn–Cu–In–Se	25.1	0.74	0.62	11.61
PbS/carbon black^105	Solvothermal/ball mill	CdS/CdSe	13.3	0.50	0.58	3.91	10.3 Ω
FeS/carbon^116	Electrochemical deposition	CdS/CdSe	20.3	0.44	0.51	4.58	8.6 Ω cm^2
CuInS2/carbon^106	Hot-injected/ball milling	CdS/CdSe	14.1	0.51	0.60	4.32	18.8 Ω cm^2
graphene/PbS^117	Hummer's method/SILAR	CdS/CdSe	11.0	0.55	0.42	2.63	4.1 Ω cm^2
RGO–Cu2S^48	Sonication	CdS/CdSe	18.4	0.52	0.46	4.40	1.6 Ω cm^2
Cu1.18S–GOR^112	Doctor blade technique	CdS/CdSe	18.4	0.52	0.58	5.42	1.3 Ω
Mo2N/CNT–RGO^109	Modified urea-glass route	CdS/CdSe	16.9	0.68	0.47	5.41	23.8 Ω
Mo2C/CNT–RGO^109	Modified urea-glass route	CdS/CdSe	16.4	0.67	0.44	4.84	34.7 Ω
MWCNT–CZTSe^118	Hot-injection/surface modification	CdSe	17.0	0.53	0.51	4.60	2.5 Ω
CuxS^50	Etching brass foil by HCL solution	CdS/CdSe	16.1	0.62	0.63	6.35	2.1 Ω
CuxSe^50	Aqueous solution method	CdS/CdSe	16.0	0.64	0.62	6.49	1.3 Ω
CoS2 (ref. 81)	Thermal sulfidation	CdS/CdSe	14.4	0.51	0.56	4.16	40.6 Ω cm^2
Co9S8 (ref. 100)	Template assisted method	CdS/CdSe	16.8	0.48	0.46	3.72	2.3 Ω cm^2
FeS^47	Sulfidation of carbon steel	CdS/CdSe	20.4	0.42	0.40	3.34	8.9 Ω
PbS^80	SILAR	CuInS/CdS	18.3	0.58	0.45	4.70	30 Ω cm^2
PbSe^102	SILAR	CdS/CdSe	16.7	0.59	0.48	4.71	36.3 Ω
NiSe2 (ref. 102)	Successive ionic layer adsorption and reaction	CdS/CdSe	14.3	0.48	0.52	3.63
NiCo2S4 (ref. 82)	Chemical synthesis	CdS/CdSe	13.1	0.52	0.49	3.30	2.9 Ω
Cu2SnSe3 (ref. 84)	Solvothermal	CdSe	16.3	0.56	0.54	4.93	0.3 Ω cm^2
Cu2ZnSnSe4 (ref. 85)	Hot-injected method	CdSe	15.5	0.54	0.52	4.35	8.9 Ω
FeS/nickel foam^101	Electrochemistry deposition	CdS/CdSe	18.7	0.41	0.58	4.39	2.0 Ω cm^2
NiS/PbS^104	CBD	CdS/CdSe	14.5	0.59	0.54	4.52	10.1 Ω
ITO@Cu2S^92	CVD/CBD/cation exchange	CdSeTe	15.2	0.68	0.58	6.12	1.3 Ω cm^2
SKE–Cu7S4 (ref. 119)	Template assisted method	CdS/CdSe	18.5	0.51	0.47	4.43	2.0 Ω cm^2

---

3 Performance evaluation of counter electrodes

As depicted in Fig. 18(a), current density–voltage (I–V) measurement is frequently used in the investigation of the performance of QDSCs. We can get four vital parameters obtained from I–V measurements as follows: the open circuit voltage (V_oc), the short circuit current density (J_sc), fill factor (FF) and the PCE. PCE is defined as the output power divide by the input power, as eqn (2) shows:where the η is PCE and P_in represent the input power of irradiated light (usually 100 mW cm⁻²). FF varies from zero to one in the actual courses of solar cell due to recombination between electrons and holes and corrosion of sensitizers. There are many factors that directly determine the value of FF including sheet resistance (R_s) and the shunt resistance (R_sh). CE owns the ability to collect electrons from the external circuit and cause catalysis of the reduction reaction of oxidized polysulfide that will also obviously affect the FF of solar cells and thus determine the PCE of the device. Besides, the resistance–voltage (R–V) curves, shown in Fig. 18(b), are derived from I–V curves which can reflect the properties of CEs to a certain extent. The resistant close to the short circuit and open circuit in the R–V curves are represented by R_sh and R_s, respectively.¹²⁰ The R_sh and R_s in the typical equivalent diode circuit model of QDSC can be estimated according to the following eqn (3). R_s, as well as intrinsic sheet resistance (R_h), Warburg impedance (Z_w) and the charge transfer resistance (R_ct) mainly represent the charge transportation within QDSCs. And R_sh indicates the efficiency of electron–hole separation compared with recombination.

R = −(dJ/dV)⁻¹ (3)

Fig. 18 (a) The typical I–V curve and (b) the R–V curve of a solar cell with related parameters. The inset is the equivalent circuit.

The decrease in the value of R_s will reduce the loss of the photovoltage, and the increase in the value of R_sh will reduce the loss of photocurrent, both of which will enhance the value of fill factor (FF). Moreover, the larger value of R_sh will cause an increase in the V_oc according to the eqn (4) below:

in which I_L, I₀, n, q, k, and T represent the light generated current, diode saturation current, ideality factor, elementary charge, Boltzmann constant, and temperature, respectively. The results of R–V measurement can just be used as an assistant means in evaluation of the performance of CEs.

Furthermore, I–V measurement is not a rigorous research method of the performance of CEs but accurate characterization of the overall performance of solar cells. The most significant analytic methods for investigating the electric and chemical characterization of CEs are briefly introduced below.

3.1 Electrochemical impedance spectroscopy

The kinetics and energetics of transport and recombination in QDSCs have been investigated over the past years by using EIS. An important advantage of EIS is the possibility of using tiny AC voltage amplitudes exerting a very small perturbation.¹²¹ That is to say, EIS is operated by using the AC voltage system to measure impedance. The responding output current is measured while the AC voltage, which usually has the intensity of an open circuit voltage, is applied to the cell under test. In this process, the frequency of the AC voltage is varied, and thus the impedance values at each frequency are obtained by the ratio of the applied voltage to the output current. As shown in Fig. 19, the Nyquist plot consists of an imaginary part (y axis) and a real part (x axis) which represent the impedance originated from both resistance and capacitance in the cell.

Fig. 19 Nyquist plot obtained from the EIS measurement of symmetric cells. CPE: constant phase element.

The Nyquist diagram, which is widely used to represent the measurement results of EIS, features typically two semicircles that in the order of increasing frequency are attributed to the Nernst diffusion within the electrolyte, the electron transfer at the QDs/electrolyte interface and the redox reaction at the CE. With the EIS measurement of QDSCs, we can indirectly get the property of CE. In order to directly study the performance of CE, the concept of symmetric cell ought to be brought up. In the system of symmetric cell, various interfacial and internal impedances of photoanode are removed. The existent impedances in the symmetric cell are the charge-transfer impedance at electrolyte/CE interface and the Nernst diffusion impedance (Z_N) of the redox couple in the electrolyte. Thus, in the Nyquist diagram, the high-frequency feature is attributed to the charge-transfer resistant (R_ct) which provides the most concrete criterion for CEs' performance. The value of R_ct which is calculated by the radius of the first semicircle can present the properties of CE.¹²² And CE with low R_ct value means that charge transfer occurs more easily. However, Z_N which originated from the diffusion the redox couple in electrolyte is rarely concerned with the performance of CE.

3.2 Tafel polarization measurement

Tafel polarization measurement is another powerful tool to evaluate the electrochemical catalytic ability of CE. The test come from the Tafel equation which is an important equation in electrochemical kinetics relating the rate of an electrochemical reaction to the overpotential. The equation is named after Swiss chemist Julius Tafel.¹²³ The electrocatalytic ability of CE which works as a catalyst for the oxidation–reduction reaction of electrolyte can be thorough studied according to the results of a Tafel measurement. A symmetric cell is also used in this system for the measurement of catalytic ability. The responding output current density of the symmetric cell is measured under a varying bias voltage.

As depicted in Fig. 20, the applied voltage (x axis) and the logarithmic form of current density make the Tafel curve. Theoretically, the Tafel curve can be divided into three different zones. The curve at low potential curve (|U| < 120 mV) is attributed to the polarization zone, the curve at middle potential (with a sharp slope) is attributed to the Tafel zone and the curve at high potential can be attributed to the diffusion zone. Among them, Tafel zone and diffusion zone which shows a horizontal shape are closely relate to the catalytic ability of CE. In the investigation of electrochemical catalytic ability, two critical parameters are required: J₀ and J_lim.

Fig. 20 Schematic diagram of Tafel polarization curve.

J₀ here refers to current density at equilibrium which means the oxidized and reduced electrolyte can simultaneously transfer electrons to the electrodes. Thus, the state of equilibrium occurs when the overpotential is zero. Besides, the value of J₀ can obtained from the intersection point of the tangent of cathodic branch in the Tafel zone and the equilibrium potential line. The relationship between J₀ and R_ct can be formulated clearly by eqn (5):

where R is the gas constant, T is the temperature, F is the Faraday constant, and n is the number of electrons involved in the reaction. According to eqn (5), J₀ varies inversely with R_ct. Therefore, the catalytic ability of CE increase with increasing value of J₀. The J_lim which represent the limiting value of the faradaic current is determined by the diffusion of redox couple in the electrolyte.¹²⁴ At the diffusion zone, the charge transfer rate limited by the deposition of electrolyte onto the electrode by diffusion. Although the overpotential is increasing, the current density is saturated to J_lim. The value of J_lim is determined by eqn (6):where D is diffusion coefficient of the electrolyte, l is the spacer thickness, C is the concentration of electrolyte and n and F retain their established meanings. Therefore, CE with higher value of J_lim shows larger diffusion coefficient in the electrolyte and thus have better catalytic ability.

3.3 Cyclic voltammetry

Generally, cyclic voltammetry (CV) is an electrochemical technique which measures the current that develops in an electrochemical cell under conditions where voltage is in excess of that predicted by the Nernst equation. CV is performed by cycling the potential of a working electrode and measuring the resulting current. The effectiveness of CV results from its capability for rapidly observing the redox behavior over a wide potential range.^55,125

CV is an efficient tool to evaluate electrode kinetics.¹²⁶ CV is performed by using an electrode connected to a dynamic potentiostat and immersed in electrolytes that contain the redox couple. Then, the redox reaction occurs at the interface between electrolyte and CE in the measurement process. A resultant CV curve is obtained as shown Fig. 21. The reduction process occurs from (a) the initial potential to (d) the switching potential. In this region the potential is scanned negatively to cause a reduction. The resulting current is called cathodic current (I_pc). The E_pc, named cathodic peak potential, is reached when all of the oxidized electrolyte at the surface of the electrode has been reduced. After the switching potential has been reached (d), the potential scans positively from (d) to (g). The resulting current is called anodic current (I_pa). The E_pa, named anodic peak potential, is reached when all of the reduced electrolyte at the surface of the electrode has been oxidized.

Fig. 21 Resultant graph of a typical cyclic voltammetry measurement. Points d and i represent the cathodic and anodic peaks, respectively. I_pc = cathodic peak current, I_pa = anodic peak current, E_c = cathodic potential, E_a = anodic potential.

Thus, the cathodic and anodic peak position indicate the rate of charge transfer at the interface of electrolyte/CE. In brief, the lower value of peak-to-peak separation (E_pp), the better the CE is at facilitating charge-transfer characteristics under a certain fixed scan rate. Besides, the intensities of the cathodic/anodic peak currents should be evaluated. CE with higher peak current means there is more electrons exist and participate in the redox reaction at the surface of electrode. That is, higher peak current indicates better electrocatalytic ability of CE. And the stability of CE can also be estimated according to the degree of degradation of the curves by repeating the cycles of CV measurement.

4 Summary and outlook

In recent years, QDSCs have drawn considerable attention of researchers. As an important constituent of QDSCs, various CEs and their properties have been introduced in the review. The basic role of CE in QDSCs is to transfer electrons to realize the reduction of the oxidized electrolyte. To optimize the performance of CE, researchers dedicate to enhance its conductivity, catalytic ability and stability. All the CEs reported are classified into five categories: metal-based CE, carbon-based CE, polymer-based CE, metallic compound CE and composite CE. Among them, metallic compound CE and composite CE have shown relatively excellent performance due to the high conductivity and better catalytic activity. In the long run, composite CEs show promising ability to make a breakthrough of QDSCs' performance by combining the advantages of metal chalcogenides and carbon derivatives. The high catalytic abilities of metal chalcogenide and the excellent conductivity of carbon derivatives with large surface area will greatly promote the development of QDSCs. However, there is still many challenges in the study of composite CE based on the combination of different catalytic materials. What's more, the structural stability and intricate preparation of composite CEs will limit their application and industrialization of QDSCs. All of this highlights the importance of metallic compound CEs with facile preparation, good stability and excellent electrochemical catalytic ability. With the further advances in the study of metallic compound CEs, we expect major breakthroughs in CE research and even in developing QDSCs in the future.

Acknowledgements

This work was supported by the National Science Foundation of China (51374029, 51611130063), Program for New Century Excellent Talents in University (NCET-13-0668), Fundamental Research Funds for the Central Universities (FRF-TP-14-008C1).

References

1. J. P. Holdren, Science, 2008, 319, 424.
2. J. J. Tian and G. Z. Cao, Nano Rev., 2013, 4, 22578.
3. B. O'Regan and M. Grätzel, Nature, 1991, 353, 737–740.
4. K. Kakiage, Y. Aoyama, T. Yano, K. Oya, J. Fujisawa and M. Hanaya, Chem. Commun., 2015, 51, 15894–15897.
5. A. Kongkanand, K. Tvrdy, K. Takechi, M. Kuno and P. V. Kamat, J. Am. Chem. Soc., 2008, 130, 4007–4015.
6. W. W. Yu, L. H. Qu, W. Z. Guo and X. G. Peng, Chem. Mater., 2003, 15, 2854–2860.
7. R. Vogel, P. Hoyer and H. Weller, J. Phys. Chem., 1994, 98, 3183–3188.
8. Y. Yang, W. Rodriguez-Cordoba and T. Lian, Nano Lett., 2012, 12, 4235–4241.
9. J. M. Luther, M. C. Beard, Q. Song, M. Law, R. J. Ellingson and A. J. Nozik, Nano Lett., 2007, 7, 1779–1784.
10. P. V. Kamat, J. Phys. Chem. Lett., 2013, 4, 908–918.
11. M. C. Hanna and A. J. Nozik, J. Appl. Phys., 2006, 100, 074510.
12. J. L. Duan, H. H. Zhang, Q. W. Tang, B. L. He and L. M. Yu, J. Mater. Chem. A, 2015, 3, 17497–17510.
13. T. Shen, J. J. Tian, L. L. Lv, C. B. Fei, Y. J. Wang, T. Pullerits and G. Z. Cao, Electrochim. Acta, 2016, 191, 62–69.
14. D. Katz, T. Wizansky, O. Millo, E. Rothenberg, T. Mokari and U. Banin, Phys. Rev. Lett., 2002, 89, 086801.
15. Z. Li, C. Dong, L. Tang, X. Zhu, H. Chen and J. Ren, Luminescence, 2011, 26, 439–448.
16. B. S. Kim, D. C. Neo, B. Hou, J. B. Park, Y. Cho, N. Zhang, J. Hong, S. Pak, S. Lee, J. I. Sohn, H. E. Assender, A. A. Watt, S. Cha and J. M. Kim, ACS Appl. Mater. Interfacesa, 2016, 8, 13902–13908.
17. A. Zaban, O. I. Mićić, B. A. Gregg and A. J. Nozik, Langmuir, 1998, 14, 3153–3156.
18. A. J. Nozik, Phys. E, 2002, 14, 115–120.
19. P. Yu, K. Zhu, A. G. Norman, S. Ferrere, A. J. Frank and A. J. Nozik, J. Phys. Chem. B, 2006, 110, 25451–25454.
20. A. Tubtimtae, K.-L. Wu, H.-Y. Tung, M.-W. Lee and G. J. Wang, Electrochem. Commun., 2010, 12, 1158–1160.
21. A. Tubtimtae, M.-W. Lee and G.-J. Wang, J. Power Sources, 2011, 196, 6603–6608.
22. I. Zumeta-Dubé, V.-F. Ruiz-Ruiz, D. Díaz, S. Rodil-Posadas and A. Zeinert, J. Phys. Chem. C, 2014, 118, 11495–11504.
23. T.-L. Li, Y.-L. Lee and H. Teng, Energy Environ. Sci., 2012, 5, 5315–5324.
24. P. K. Santra, P. V. Nair, K. George Thomas and P. V. Kamat, J. Phys. Chem. Lett., 2013, 4, 722–729.
25. J. Du, Z. L. Du, J.-S. Hu, Z. X. Pan, Q. Shen, J. K. Sun, D. H. Long, H. Dong, L. T. Sun, X. H. Zhong and L.-J. Wan, J. Am. Chem. Soc., 2016, 138, 4201–4209.
26. P. K. Santra and P. V. Kamat, J. Am. Chem. Soc., 2012, 134, 2508–2511.
27. J. J. Tian and G. Z. Cao, J. Phys. Chem. Lett., 2015, 6, 1859–1869.
28. J. J. Tian, Q. F. Zhang, E. Uchaker, R. Gao, X. H. Qu, S. G. Zhang and G. Z. Cao, Energy Environ. Sci., 2013, 6, 3542.
29. X. Y. Wang, J. J. Tian, C. B. Fei, L. L. Lv, Y. J. Wang and G. Z. Cao, RSC Adv., 2015, 5, 8622–8629.
30. J. J. Tian and G. Z. Cao, Coord. Chem. Rev., 2016, 320–321, 193–215.
31. J. J. Tian, R. Gao, Q. F. Zhang, S. G. Zhang, Y. W. Li, J. Lan, X. H. Qu and G. Z. Cao, J. Phys. Chem. C, 2012, 116, 18655–18662.
32. Y. L. Liao, J. Zhang, W. G. Liu, W. X. Que, X. T. Yin, D. N. Zhang, L. H. Tang, W. D. He, Z. Y. Zhong and H. W. Zhang, Nano Energy, 2015, 11, 88–95.
33. Y.-L. Lee and C.-H. Chang, J. Power Sources, 2008, 185, 584–588.
34. Z. X. Yu, Q. X. Zhang, D. Qin, Y. H. Luo, D. M. Li, Q. Shen, T. Toyoda and Q. B. Meng, Electrochem. Commun., 2010, 12, 1776–1779.
35. H.-Y. Chen, L. Lin, X.-Y. Yu, K.-Q. Qiu, X.-Y. Lü, D.-B. Kuang and C.-Y. Su, Electrochim. Acta, 2013, 92, 117–123.
36. R. M. Evangelista, S. Makuta, S. Yonezu, J. Andrews and Y. Tachibana, ACS Appl. Mater. Interfacesa, 2016, 8, 13957–13965.
37. Y. Tian and T. Tatsuma, J. Am. Chem. Soc., 2005, 127, 7632–7637.
38. H. J. Lee, J.-H. Yum, H. C. Leventis, S. M. Zakeeruddin, S. A. Haque, P. Chen, S. I. Seok, M. Grätzel and M. K. Nazeeruddin, J. Phys. Chem. C, 2008, 112, 11600–11608.
39. H. J. Lee, P. Chen, S. J. Moon, F. Sauvage, K. Sivula, T. Bessho, D. R. Gamelin, P. Comte, S. M. Zakeeruddin, S. I. Seok, M. Gratzel and M. K. Nazeeruddin, Langmuir, 2009, 25, 7602–7608.
40. S. Y. Chae, Y. J. Hwang and O.-S. Joo, RSC Adv., 2014, 4, 26907.
41. D. Karageorgopoulos, E. Stathatos and E. Vitoratos, J. Power Sources, 2012, 219, 9–15.
42. L. Li, X. C. Yang, J. J. Gao, H. N. Tian and J. Z. Zhao, J. Am. Chem. Soc., 2011, 133, 8458–8460.
43. K. Zhao, Z. X. Pan, I. Mora-Sero, E. Canovas, H. Wang, Y. Song, X. Gong, J. Wang, M. Bonn, J. Bisquert and H. X. Zhong, J. Am. Chem. Soc., 2015, 137, 5602–5609.
44. J. Wang, Y. Li, Q. Shen, T. Izuishi, Z. X. Pan, K. Zhao and X. H. Zhong, J. Mater. Chem. A, 2016, 4, 877–886.
45. A. Hauch and A. Georg, Electrochim. Acta, 2001, 46, 3457–3466.
46. Z. Tachan, M. Shalom, I. Hod, S. Rühle, S. Tirosh and A. Zaban, J. Phys. Chem. C, 2011, 115, 6162–6166.
47. L. L. Quan, W. P. Li, L. Q. Zhu, H. F. Geng, X. W. Chang and H. C. Liu, J. Power Sources, 2014, 272, 546–553.
48. J. G. Radich, R. Dwyer and P. V. Kamat, J. Phys. Chem. Lett., 2011, 2, 2453–2460.
49. H. Zhang, H. L. Bao and X. H. Zhong, J. Mater. Chem. A, 2015, 3, 6557–6564.
50. H. Zhang, C. Wang, W. X. Peng, C. Yang and X. H. Zhong, Nano Energy, 2016, 23, 60–69.
51. P. V. Kamat, Acc. Chem. Res., 2012, 45, 1906–1915.
52. J. J. Tian, L. L. Lv, C. B. Fei, Y. J. Wang, X. G. Liu and G. Z. Cao, J. Mater. Chem. A, 2014, 2, 19653–19659.
53. J.-Y. Kim, J. Yang, J. H. Yu, W. Baek, C.-H. Lee, H. J. Son, T. Hyeon and M. J. Ko, ACS Nano, 2015, 9, 11286–11295.
54. C. Justin Raj, S. N. Karthick, S. Park, K. V. Hemalatha, S.-K. Kim, K. Prabakar and H.-J. Kim, J. Power Sources, 2014, 248, 439–446.
55. F. M. Ren, S. J. Li and C. L. He, Sci. China Mater., 2015, 58, 490–495.
56. J. J. Tian, T. Shen, X. G. Liu, C. B. Fei, L. L. Lv and G. Z. Cao, Sci. Rep., 2016, 6, 23094.
57. A. Zyoud, R. S. Al-Kerm, R. S. Al-Kerm, M. Waseem, H. S. H. Mohammed, D. Park, G. Campet, N. Sabli and H. S. Hilal, Electrochim. Acta, 2015, 174, 472–479.
58. M. D. Ye, X. R. Wen, N. Zhang, W. X. Guo, X. Y. Liu and C. J. Lin, J. Mater. Chem. A, 2015, 3, 9595–9600.
59. Y. B. Chen, Z. X. Qin, X. X. Wang, X. Guo and L. J. Guo, RSC Adv., 2015, 5, 18159–18166.
60. V.-D. Dao, Y. Choi, K. Yong, L. L. Larina, O. Shevaleevskiy and H.-S. Choi, J. Power Sources, 2015, 274, 831–838.
61. R. Vogel, K. Pohl and H. Weller, Chem. Phys. Lett., 1990, 174, 241–246.
62. I. Mora-Sero, S. Gimenez, T. Moehl, F. Fabregat-Santiago, T. Lana-Villareal, R. Gomez and J. Bisquert, Nanotechnology, 2008, 19, 424007.
63. C. Yu, Z. Q. Liu, Y. W. Chen, X. T. Meng, M. Y. Li and J. S. Qiu, Sci. China Mater., 2016, 59, 104–111.
64. C. F. Wu, Z. X. Wu, J. Wei, H. Dong and Y. C. Gao, ECS Electrochem. Lett., 2013, 2, H31–H33.
65. C. Justin Raj, K. Prabakar, A. Dennyson Savariraj and H.-J. Kim, Electrochim. Acta, 2013, 103, 231–236.
66. Y.-L. Lee and Y.-S. Lo, Adv. Funct. Mater., 2009, 19, 604–609.
67. N. Du, L. Ren, W. F. Sun, X. Jin, Q. Zhao, Y. Y. Cheng, T. H. Wei and Q. H. Li, J. Power Sources, 2015, 284, 162–169.
68. T. Shu, X. Li, Z.-L. Ku, S. Wang, S. Wu, X.-H. Jin and C.-D. Hu, Electrochim. Acta, 2014, 137, 700–704.
69. M.-H. Yeh, C.-P. Lee, C.-Y. Chou, L.-Y. Lin, H.-Y. Wei, C.-W. Chu, R. Vittal and K.-C. Ho, Electrochim. Acta, 2011, 57, 277–284.
70. T. Shu and Z.-L. Ku, J. Alloys Compd., 2014, 586, 257–260.
71. K.-M. Lee, P.-Y. Chen, C.-Y. Hsu, J.-H. Huang, W.-H. Ho, H.-C. Chen and K.-C. Ho, J. Power Sources, 2009, 188, 313–318.
72. Q. X. Zhang, Y. D. Zhang, S. Q. Huang, X. M. Huang, Y. H. Luo, Q. B. Meng and D. M. Li, Electrochem. Commun., 2010, 12, 327–330.
73. B. Z. Fang, M. Kim, S.-Q. Fan, J. H. Kim, D. P. Wilkinson, J. Ko and J.-S. Yu, J. Mater. Chem., 2011, 21, 8742.
74. Q. X. Zhang, S. J. Zhou, Q. Li and H. G. Li, RSC Adv., 2015, 5, 30617–30623.
75. M. Seol, E. Ramasamy, J. Lee and K. Yong, J. Phys. Chem. C, 2011, 115, 22018–22024.
76. P. Sudhagar, E. Ramasamy, W.-H. Cho, J. Lee and Y. S. Kang, Electrochem. Commun., 2011, 13, 34–37.
77. S.-Q. Fan, B. Fang, J. H. Kim, J.-J. Kim, J.-S. Yu and J. Ko, Appl. Phys. Lett., 2010, 96, 063501.
78. C. V. V. M. Gopi, M. Venkata-Haritha, S.-K. Kim and H.-J. Kim, J. Power Sources, 2016, 311, 111–120.
79. A. D. Savariraj, K. K. Viswanathan and K. Prabakar, ACS Appl. Mater. Interfacesa, 2014, 6, 19702–19709.
80. C.-Y. Lin, C.-Y. Teng, T.-L. Li, Y.-L. Lee and H. Teng, J. Mater. Chem. A, 2013, 1, 1155–1162.
81. M. S. Faber, K. Park, M. Cabán-Acevedo, P. K. Santra and S. Jin, J. Phys. Chem. Lett., 2013, 4, 1843–1849.
82. Q. Luo, Y. C. Gu, J. B. Li, N. Wang and H. Lin, J. Power Sources, 2016, 312, 93–100.
83. X. Q. Chen, Z. Li, Y. Bai, Q. Sun, L. Z. Wang and S. X. Dou, Chemistry, 2015, 21, 1055–1063.
84. F. Liu, J. Zhu, Y. Li, J. F. Wei, M. Lv, Y. F. Xu, L. Zhou, L. H. Hu and S. Y. Dai, J. Power Sources, 2015, 292, 7–14.
85. X. W. Zeng, W. J. Zhang, Y. Xie, D. H. Xiong, W. Chen, X. B. Xu, M. K. Wang and Y.-B. Cheng, J. Power Sources, 2013, 226, 359–362.
86. D. C. Reynolds, G. Leies, L. L. Antes and R. E. Marburger, Phys. Rev., 1954, 96, 533–534.
87. K. Meng, P. K. Surolia, O. Byrne and K. R. Thampi, J. Power Sources, 2014, 248, 218–223.
88. K. Meng, G. Chen and K. R. Thampi, J. Mater. Chem. A, 2015, 3, 23074–23089.
89. W. T. Liang, L. Q. Zhu, H. C. Liu, F. X. Xi and W. P. Li, Electrochim. Acta, 2015, 184, 285–294.
90. C. V. Thulasi-Varma, S. S. Rao, C. S. Kumar, C. V. Gopi, I. K. Durga, S. K. Kim, D. Punnoose and H. J. Kim, Dalton Trans., 2015, 44, 19330–19343.
91. Y. Lei, H. M. Jia, Z. Zheng, Y. H. Gao, X. W. Chen and H. W. Hou, CrystEngComm, 2011, 13, 6212.
92. S. M. Wang, W. W. Dong, X. D. Fang, S. Zhou, J. Z. Shao, Z. H. Deng, R. H. Tao, Q. L. Zhang, L. H. Hu and J. Zhu, Electrochim. Acta, 2015, 154, 47–53.
93. Y. Jiang, X. Zhang, Q. Q. Ge, B. B. Yu, Y. G. Zou, W. J. Jiang, W. G. Song, L. J. Wan and J. S. Hu, Nano Lett., 2014, 14, 365–372.
94. Y. Jiang, B. B. Yu, J. Liu, Z. H. Li, J. K. Sun, X. H. Zhong, J. S. Hu, W. G. Song and L. J. Wan, Nano Lett., 2015, 15, 3088–3095.
95. F. Bo, C. F. Zhang, C. L. Wang, S. H. Xu, Z. Y. Wang and Y. P. Cui, J. Mater. Chem., 2014, 2, 14585.
96. Y. Bai, C. Han, X. Q. Chen, H. Yu, X. Zong, Z. Li and L. Z. Wang, Nano Energy, 2015, 13, 609–619.
97. F. Liu, J. Zhu, L. H. Hu, B. Zhang, J. X. Yao, M. K. Nazeeruddin, M. Grätzel and S. Y. Dai, J. Mater. Chem. A, 2015, 3, 6315–6323.
98. J. Xu, Q. D. Yang, W. P. Kang, X. Huang, C. Y. Wu, L. Wang, L. B. Luo, W. J. Zhang and C.-S. Lee, Part. Part. Syst. Charact., 2015, 32, 840–847.
99. S. X. Wang, T. Shen, H. W. Bai, B. Li and J. J. Tian, J. Mater. Chem. C, 2016, 4, 8020–8026.
100. C. Chen, M. D. Ye, N. Zhang, X. R. Wen, D. J. Zheng and C. J. Lin, J. Mater. Chem. A, 2015, 3, 6311–6314.
101. H. F. Geng, L. Q. Zhu, W. P. Li, H. C. Liu, L. L. Quan, F. X. Xi and X. W. Su, J. Power Sources, 2015, 281, 204–210.
102. H. M. Choi, I. A. Ji and J. H. Bang, ACS Appl. Mater. Interfacesa, 2014, 6, 2335–2343.
103. Z. Yang, C.-Y. Chen, C.-W. Liu, C.-L. Li and H.-T. Chang, Adv. Energy Mater., 2011, 1, 259–264.
104. C. V. V. M. Gopi, M. Venkata-Haritha, S. Ravi, C. V. Thulasi-Varma, S.-K. Kim and H.-J. Kim, J. Mater. Chem. C, 2015, 3, 12514–12528.
105. Y. Y. Yang, L. F. Zhu, H. C. Sun, X. M. Huang, Y. H. Luo, D. M. Li and Q. B. Meng, ACS Appl. Mater. Interfacesa, 2012, 4, 6162–6168.
106. X. L. Zhang, X. M. Huang, Y. Y. Yang, S. Wang, Y. Gong, Y. H. Luo, D. M. Li and Q. B. Meng, ACS Appl. Mater. Interfacesa, 2013, 5, 5954–5960.
107. V.-D. Dao, P. Kim, S. Baek, L. L. Larina, K. Yong, R. Ryoo, S. H. Ko and H.-S. Choi, Carbon, 2016, 96, 139–144.
108. H. J. Kim, L. Myung-Sik, C. V. Gopi, M. Venkata-Haritha, S. S. Rao and S. K. Kim, Dalton Trans., 2015, 44, 11340–11351.
109. M. Seol, D. H. Youn, J. Y. Kim, J.-W. Jang, M. Choi, J. S. Lee and K. Yong, Adv. Energy Mater., 2014, 4, 1300775.
110. B. Hinnemann, J. K. Nørskov and H. Topsøe, J. Phys. Chem. B, 2005, 109, 2245–2253.
111. C. Giordano, C. Erpen, W. Yao and M. Antonietti, Nano Lett., 2008, 8, 4659–4663.
112. D. Ghosh, G. Halder, A. Sahasrabudhe and S. Bhattacharyya, Nanoscale, 2016, 8, 10632–10641.
113. H. J. Lee, J. Bang, J. Park, S. Kim and S.-M. Park, Chem. Mater., 2010, 22, 5636–5643.
114. J. H. Dong, S. Jia, J. Z. Chen, B. Li, J. F. Zheng, J. H. Zhao, Z. J. Wang and Z. P. Zhu, J. Mater. Chem., 2012, 22, 9745.
115. V.-D. Dao, Y. Choi, K. Yong, L. L. Larina and H.-S. Choi, Carbon, 2015, 84, 383–389.
116. H. F. Geng, L. Q. Zhu, W. P. Li, H. C. Liu, X. W. Su, F. X. Xi and X. W. Chang, Electrochim. Acta, 2015, 182, 1093–1100.
117. P. Parand, M. Samadpour, A. Esfandiar and A. Iraji Zad, ACS Photonics, 2014, 1, 323–330.
118. X. W. Zeng, D. H. Xiong, W. J. Zhang, L. Q. Ming, Z. Xu, Z. F. Huang, M. K. Wang, W. Chen and Y. B. Cheng, Nanoscale, 2013, 5, 6992–6998.
119. W. L. Chen, M. Wang, T. Y. Qian, H. L. Cao, S. S. Huang, Q. Q. He, N. Liang, C. Wang and J. T. Zai, Nano Energy, 2015, 12, 186–196.
120. K. Yan, W. Chen and S. H. Yang, J. Phys. Chem. C, 2013, 117, 92–99.
121. M. Adachi, M. Sakamoto, J. Jiu, Y. Ogata and S. Isoda, J. Phys. Chem. B, 2006, 110, 13872–13880.
122. Q. Wang, J. E. Moser and M. Gratzel, J. Phys. Chem. B, 2005, 109, 14945–14953.
123. G. T. Burstein, Corros. Sci., 2005, 47, 2858–2870.
124. M. X. Wu, X. Lin, A. Hagfeldt and T. L. Ma, Chem. Commun., 2011, 47, 4535–4537.
125. P. T. Kissinger and W. R. Heineman, J. Chem. Educ., 1983, 60, 702.
126. R. S. Nicholson, Anal. Chem., 1965, 37, 1351–1355.

---