NiS composite counter electrode for

Journal of Electroanalytical Chemistry 777 (2016) 123–132

Contents lists available at ScienceDirect

Journal of Electroanalytical Chemistry journal homepage: www.elsevier.com/locate/jelechem

Investigation on novel CuS/NiS composite counter electrode for hindering charge recombination in quantum dot sensitized solar cells Hee-Je Kim, Seong-Min Suh, S. Srinivasa Rao, Dinah Punnoose, Chebrolu Venkata Tulasivarma, Chandu.V.V.M. Gopi, Nagabhushanam Kundakarla, Seenu Ravi, Ikkurthi Kanaka Durga ⁎ School of Electrical Engineering, Pusan National University, Busandaehak-ro 63beon-gil, Geumjeong-gu, Busan 46241, Rep. of Korea

a r t i c l e

i n f o

Article history: Received 8 May 2016 Received in revised form 26 July 2016 Accepted 27 July 2016 Available online 29 July 2016 Keywords: CuS/NiS composite counter electrodes Quantum dot solar cells Jasmine flower Cyclic voltammogram

a b s t r a c t To make quantum dot-sensitized solar cells (QDSSCs) more attractive, it is necessary for the power conversion efficiency (PCE) to be comparable to those of other emerging solar cells. Currently, copper sulfide (CuS) and nickel sulfide (NiS) are commonly used counter electrodes (CEs) in high-efficiency QDSSCs because of their low toxicity, environmental compatibility, and superior electrocatalytic activity in the presence of polysulfide electrolyte. For the first time, novel CuS/NiS electrodes were prepared by facile chemical bath deposition method. This article describes the effect of NiS layer on CuS film for preventing the recombination process to enhance the performance of QDSSCs. Under one sun illumination, the CE with the optimized CuS/NiS composite film exhibits higher short-circuit current density (Jsc), open-circuit voltage (Voc), and PCE of 12.47 mA cm−2, 0.599 V, and 4.19%, respectively. These values are much higher than those of bare CuS (2.73%), NiS (1.82%), and Pt CEs (1.16%). This enhancement is mainly attributed to the improved surface morphology, higher sulfur atomic percentage with Cu vacancies, rapid electron transport, and lower electron recombination rate for the polysulfide electrolyte. Characterization with, cyclic voltammetry, and Tafel polarization was performed to study the reasons for efficient CE performance. © 2016 Elsevier B.V. All rights reserved.

1. Introduction In modern society it is possible to observe an everlasting permeation of electron devices and appliances and depletion of fossil fuels force exploration of alternative clean regeneration energy. According to the Koomey's law that the computing efficiency doubles every 1.57 years, encroaching the surge performance of rechargeable energy sources, which is essentially stable. Many researchers see the dye-sensitized solar cells (DSSCs) as one of the ways to deal with the modern requirements of energy and it is extensively considered as a renewable energy technology to harness electricity from solar energy. It is considered as an alternative to conventional solid-state solar cells due to its environmental friendliness, low cost, easy fabrication, and acceptable power conversion efficiency (PCE) [1–3]. In general, a typical DSSC is composed of a dye-sensitized nanocrystalline semiconductor as a photoanode, a redox electrolyte consisting of the I−/I− 3 redox couple, and a counter electrode (CE) [4]. The bifunctional CE collects electrons from the external circuit and reduces the tri-iodide (I− 3 ) ions generated after dye regeneration. To improve conversion efficiency, the CE material should have excellent electrocatalytic properties [5,6]. The conventional platinum (Pt) is deposited on transparent conductive oxide (TCO) substrate ⁎ Corresponding author. E-mail address: [email protected] (I.K. Durga).

http://dx.doi.org/10.1016/j.jelechem.2016.07.037 1572-6657/© 2016 Elsevier B.V. All rights reserved.

due to its superior electrocatalytic activity and electrochemical stability for the I−/I− 3 redox couple [7–9]. In addition to DSSCs, several approaches have been developed for solar cells such as quantum-dot-sensitized solar cells (QDSSCs), organic solar cells, amorphous, nanocrystalline Si and now perovskite-based meso-super structured solar cells; [10]. However, the perovskite solar cells fabrication procedures make use of high-temperature and/or vacuum based processing and high cost precursor materials limit the commercial success of certain systems [11]. In particular, the QDSSCs promises to deliver one of the lowest cost technologies and greater conversion efficiency that are capable of converting sunlight to electricity. Conversely, QDSSCs shows partial efficiency because of the poor electro-catalytic activity among the electrolyte and the CE compared to challenging systems, for example DSSCs, thin-film-based solar cells, and also perovskite solar cells. The outstanding characteristics of QDs include tunable light-harvesting range, cost-effectiveness, high absorption coefficient, large intrinsic dipole moment, and generation of multiple electron carriers under high energy excitation [12]. Due to the above advantages, the QDSSC is perceived as one of the promising candidates in low-cost third generation solar cells. QDSSCs borrow the concept of DSSCs with the exception that dye molecules replace the inorganic QD absorbers. A typical configuration of QDSSC devices includes a photoanode based on a mesoporous metal oxide (mainly TiO2) loaded with a sensitizer (mainly CdS, CdSe, SnS, Sb2S3 or PbS), an electrolyte

124

H.-J. Kim et al. / Journal of Electroanalytical Chemistry 777 (2016) 123–132

(the optimal electrolyte for QDSSCs is a polysulfide redox couple (S2− n / S2−)), and a CE. However, jin wang et al. fabricated with Mn doped QDs on TiO2 film electrode at room temperature and achieved PCE and FF of 9.40% and 0.655 but the reported PCE of QDSSCs is still far below their theoretical value of 44% and that of DSSCs (12–13%) [13–15]. One of the possible reasons being the narrow absorption range of the quantum dots, especially in the case of CdS QDs, as well as charge recombination at the QD/electrolyte interface [16]. Other possible reasons include inferior optoelectronic properties of QD sensitizers and electron loss occurring through charge recombination at the interface of the counter electrode and electrolyte, which results in a drastic decrement in the short circuit current and open-circuit photo voltage [17–20]. To exploit the potential of QDSSCs and increase the efficiencies, a novel electrocatalytic material for use in CEs is essential. Pt CEs is suitable electro-catalyst when iodide/triiodide redox electrolyte is used in DSSCs. It exhibits tremendous electrocatalytic activity and low charge transfer resistance (RCE) at the interface of the CE and the electrolyte [21]. However, Pt CEs are unsuitable with the polysulfide electrolyte in QDSSCs because sulfur-containing compounds (S2 − or thiol) are preferentially adsorbed and lead to photo corrosion of the QDs. This decreases the surface activity, conversion efficiency, and short lifetimes of the QDSSCs [22]. Recently, various non-Pt-based materials have been reported as CEs to improve the electrocatalytic activity and performance of QDSSCs, such as CoS, NiS, CuS, PbS, hollow carbon, conducting polymer, and carbon-based materials (nanotubes, graphene, and carbon black). Among them, metal sulfide electrodes exhibit higher conversion efficiency (CuS N CoS N NiS), low charge transfer resistance, and superior catalytic activity, but shorter lifetimes because of sulfur corrosion for the redox reaction of polysulfide electrolyte [22–24]. While there are many alternatives, CuS and NiS are the most desired since they are not only cost-effective but also produce high PCE, superior photo and electrocatalytic properties, and conductivity. Even though CuS and NiS are p-type semiconductors, they show mixed ionic and electronic conduction due to the presence of Cu or Ni vacancies. The vacancies contribute to increasing the free holes in the material, which act as electron acceptors that exhibit excessive conductivity. Moreover, Cu2S and Ni3S2 are unstable phases due to the formation of Cu and Ni vacancies, even in thermodynamic equilibrium with bulk copper or nickel metal [25]. Our previous studies have shown that CuS and NiS electrodes can act as the most effective catalytic materials, respectively showing conversion efficiencies of 1.72% and 2.61% in QDSSCs, which are higher than that of cells based on Pt (0.82%) [26,27]. But CuS electrodes show less stability than Pt-based cells because Cu substrate continually reacts with polysulfide electrolyte, which can contaminate the electrolyte and the photoanode [22]. The lower PCE of CuS and NiS CEs are due to the higher charge recombination at the interfaces of photoanode and CE. A promising strategy to reduce charge recombination at the CE/ electrolyte interface is to modify the intrinsic properties of CE material is to fabricate composite electrode or metal ion dopants such as Co, Ni and Mn. The composite layer creates new electronic states in the host material, thus alters the recombination dynamics and charge separation. Zusing and co-workers pioneered the concept of CuS/CoS onto fluorine-doped tin oxide glass substrate and pushed the PCE of QDSSCs beyond 4% for the first time. Since then, comprehensive efforts have been paid on the composite material used as a CE in QDSSCs for improving the performance of QDSSCs. To overcome this problem, we fabricated a hybrid CE with a composite of copper sulfide/nickel sulfide (CuS/ NiS CE) using a simple chemical bath deposition (CBD) method. The CBD method is the most simple, convenient, and common method for fabricating CEs for large-scale production. Both the photo-electrochemical performance and the catalytic properties of the CuS/NiS CE were much higher than that of bare CuS, NiS, and Pt-based cells. By combining the optimized TiO2-QD photo electrode and the CuS/NiS CE with

polysulfide electrolyte, a power conversion efficiency of 4.19% was achieved under 1 sun illumination, which is attractive for use in QDSSCs. 2. Experimental section 2.1. Materials For the preparation of the CE and photoanodes, all materials were bought from Sigma-Aldrich and used without further purification. Copper sulfate pentahydrate [CuSO4·5H2O], sodium thiosulfate [Na2S2O3], urea [CH4N2O], nickel sulfate hexahydrate [NiSO4·6H2O], thioacetamide [CH3CSNH2], acetic acid, cadmium acetate dehydrate [Cd(CH3COO)2· 2H2O], sodium sulfide [Na2S], sodium sulfite [Na2SO3], zinc acetate dehydrate [Zn(CH3COO)2·2H2O], sulfur [S], selenium powder [99.99%], potassium chloride [KCl], Pt, and TiO2 paste [Ti-nanoxide HT/SP] were used without further purification. 2.2. Fabrication of TiO2/CdS/CdSe/ZnS photoanodes Ultrasonically cleaned FTO substrate was used to prepare the photoanodes and CEs. The commercial TiO2 paste with 20-nm particle size was coated on a FTO substrate by the doctor blade method (with an active area of 0.27 cm2). The coated substrate was gradually annealed at 450 °C for 30 min. The total thickness of the TiO2 film was 7 μm after evaporation of the solvent. Thereafter, the TiO2 electrodes were sensitized with CdS and CdSe QDs by successive ion layer absorption and reaction (SILAR) technique. In this method, TiO2 photoanodes were immersed in an aqueous solution of 0.1 M Cd(NO3)2 for 5 min, rinsed with DI water and ethanol for 1 min, dipped into an aqueous solution of 0.1 M Na2S for 5 min, rinsed again with DI water and ethanol, and dried with a drier. The process was repeated for 5 cycles and conducted at room temperature. For the deposition of CdSe, the prepared TiO2/CdS electrodes were dipped in a aqueous cationic precursor solution of 0.1 M Cd(NO3)2 for 5 min to allow Cd2+ ions to adsorb onto the TiO2/CdS and then dipped into the aqueous anionic precursor solution of Na2SeSO3 for 5 min, where the pre-adsorbed Cd2 + ions react with Se2 − to form the desired CdSe. This process was conducted at 50 °C and repeated for 8 SILAR cycles. Finally, the TiO2/CdS/CdSe electrodes were dipped into an aqueous solution containing 0.1 M Zn(NO3)2 for 5 min and then dipped in a solution of 0.2 M Na2S for 5 min to form the desired ZnS. The film was rinsed with DI water and ethanol and dried with a drier. The two-step process comprises one SILAR cycle and was repeated two times. 2.3. Preparation of CuS, NiS, CuS/NiS and Pt counter electrodes For the preparation of CEs, we used the CBD method, which is a simple and convenient method for the deposition of metal sulfide thin films for large-scale production. Prior to the metal sulfide deposition (CuS, NiS, and CuS/NiS), FTO substrates were cleaned with acetone, ethanol, and DI water and dried with hair drier. For the preparation of CuS electrodes, FTO substrates were horizontally dipped into a solution of CuSO4·5H2O (0.1 M), CH4N2O (0.4 M), and thioacetamide (0.4 M), kept in a hot air oven at 65 °C for 2 h, and then cleaned with DI water and ethanol. To fabricate NiS electrodes, FTO substrates were horizontally dipped into a solution containing cationic and anionic precursors of nickel sulfate hexahydrate (0.1 M) and thioacetamide (0.4 M), which act as sources of Ni2 + and S2 −, respectively. CH4N2O (0.8 M) and acetic acid (0.6 M) were used as strong reagents. The prepared solution with FTO substrates was kept in a hot air oven at 90 °C for 90 min. The CuS/NiS counter electrodes were fabricated through deposition of NiS nanoparticles on the surface of CuS. The as-prepared CuS electrodes were placed horizontally in the aforementioned NiS solution at 90 °C for 90 min. For comparison, the cleaned FTO substrate was coated with commercial Pt paste (Solaronix, Pt catalyst T/SP) by the doctor-


blade method and heat treated at 400 °C for 10 min in air (active area of Pt ~0.7 cm2). 3. Fabrication and characterization of QDSSCs Finally, the synthesized photoanodes and counter electrodes were sealed using a 60-μm sealing film (SX 1170-60, Solaronix). The space between the electrodes was filled with polysulfide electrolyte solution containing 1 M sodium sulfide, 0.2 M sodium hydroxide, and 0.1 M sulfur in methanol and water (7:3 ratio). X-ray diffraction (XRD, D/Max2400, Rigaku) was performed to identify the crystalline phase of the CuS/NiS thin films at 40 kV and 30 mA. The surface features, morphology, thickness, and elemental composition of CEs were characterized using a field emission scanning electron microscope (FE-SEM) equipped with energy-dispersive X-ray spectroscopy (EDX). An atomic force microscope (JPK instruments, Berlin, Germany) was used to identify the surface roughness of the substrate. Cyclic voltammetric (CV) measurements were performed using a BioLogic potentiostat galvanostat/EIS analyzer (SP-150, France) with a conventional three-electrode system. The surface compositions of the CuS, NiS, and CuS/NiS films were also verified by X-ray photoelectron spectroscopy. Photocurrent densityvoltage (J-V) measurements were performed using an ABET technology (USA) solar simulator at 1 sun condition (100 mW cm− 2, AM 1.5G). Electrochemical impedance spectroscopy (EIS) under illumination in the frequency range of 100 mHz–500 kHz and amplitude was kept at 10 mV in all cases. Tafel polarization analysis was performed using symmetric dummy cells under dark conditions. Detailed fabrication of the symmetric cells is reported elsewhere. 4. Results and discussion Fig. 1 shows the transport pathways of electrons and photo electric conversion configuration of the TiO2/CdS/CdSe/ZnS-based QDSSC with the CuS/NiS composite CE. The configuration of QDSSCs consists of a wide-band-gap mesoporous oxide film photoanode (such as the commonly used TiO2), QD sensitizer, polysulfide redox couple, and FTO/ CuS/NiS CE. During the operation, photons are captured by QDs, yielding electron-hole pairs that are rapidly separated into electrons and holes at the interface between the nanocrystalline oxide and QDs. The electrons are injected from the excited state of CdS/CdSe/ZnS QDs into the TiO2 film, and the holes are simultaneously scavenged by the CE via the hole transporting redox couple of the polysulfide electrolyte. Kamat et al. reported that the electron transfer between CdS/CdSe/ZnS QDs and TiO2 film is an ultrafast process with a rate constant on the order of 1010 to 109 s−1 [28–29]. The electron-transfer is very fast compared with the hole transfer (107 to 109 s−1). The electron transport within the n-type TiO2 film is very slow compared to the electron and hole

125

transfer. The charge recombination rate thus becomes a major limiting factor in the PCE. However, the electrons in the FTO move to the cathode (CuS/NiS CE) via hole transport by the redox couple of the polysulfide electrolyte and participate in the redox reaction of the CuS/NiS CE/poly− 2− sulfide redox electrolyte interface (S2− ). As a result, an effin + e nS cient CuS/NiS CE should have high conductivity, electrocatalytic activity, and low charge recombination rate for the reduction of charge carriers in the polysulfide electrolyte.

4.1. Morphology characterization Fig. 2 shows the top-view scanning electron microscopy (SEM) images of the CuS, NiS, and CuS/NiS thin films deposited on the FTO substrates by CBD technique. It reveals that the surface morphology is prominently affected by the varying material compositions on the FTO for the preparation of ideal CEs. The CuS sample is composed of many tiny nanoparticles with a smooth surface aligned together to form spherical structures with fairly uniform diameters of 200 nm and arranged randomly on the FTO substrate (Fig. 2a). Fig. 2b shows the surface morphologies of the NiS film, indicating clustered nanoparticles formed on the substrates with particle sizes in the range of 20–30 nm. In the CuS/NiS film (Fig. 2c), the deposition of NiS on the surface of CuS greatly affects the surface morphology, and the spherical structures of the CuS film and clustered nanoparticles (NiS film) were converted to flower-like morphology. Each flower structure is composed of many very thin mesoporous sheets (~200 to 250 nm in length), as clearly shown in the high-magnification images in Fig. 2d, e, and f. The deposition of NiS on the CuS film makes it denser, and the voids in between the spherical structures and nanoparticles are reduced with uniform pores. This may increase the electrocatalytic activity to efficiently transfer electrons in QDSSCs. Based on the surface morphology; we concluded that there is good adhesion of the CuS/NiS film between the material and the substrate. The improved surface morphology will also enhance the charge transfer at the interface of the CE and polysulfide redox electrolyte [30]. The SEM results indicate only the structural features and density of the CE film particles but do not distinguish the surface roughness. To study the surface roughness and for quantification of the films of the CuS, NiS, and CuS/NiS electrodes, high-resolution tapping-mode atomic force microscope (AFM) was carried out, as shown in Fig. 3. The estimated average root-mean-square (RMS) roughness of CuS electrode is about 44.39 nm, which is higher than that of the NiS (33.82 nm) electrode. The grains have almost uniform sizes of about 2 μm for CuS film with white spots observed on the surface. The grain size of NiS electrodes is about 2–1 μm.

Fig. 1. Schematic view of CuS/NiS based quantum-dot sensitized solar cells with the TiO2/CdS/CdSe/ZnS photo anode and the polysulfide ionic liquid electrolyte.

126


Fig. 2. HR-SEM images of (a) CuS, (b) NiS and (c) CuS/NiS thin films deposited on FTO substrates and (d), (e), and (f) are the high-magnified images of CuS/NiS thin film.

The variation of grain boundaries and grain sizes is due to the difference in film quantity, crystallographic orientation, and contaminants [24]. AFM studies indicated that the RMS surface roughness of the CuS/NiS electrode is higher than that of the bare CuS and NiS electrodes. The higher surface roughness is mainly due to the good contact between the composite electrodes, and this higher surface roughness completely satisfies the roughness requirements. The AFM results clearly support that the greater surface roughness of the CE is responsible for the higher

electrocatalytic activity for the polysulfide electrolyte [31,32]. These results also imply that the charge transfer resistance at the counter electrode/polysulfide electrolyte interface would be lower in the case of the CuS/NiS composite electrode than for the CuS and NiS electrodes. From the AFM results, it can be concluded that the deposition of composite films on FTO not only affects the surface morphology and roughness of the electrode, but also increases the area between the electrode and the electrolyte. The PCE of the CuS/NiS solar cell mainly depends on

Fig. 3. Atomic microscopy images of the CuS (a, a1), NiS (b, b1) and CuS/NiS (c, c1) counter electrodes. The left one corresponds to 2D and the right one corresponds to 3D images.


the higher surface roughness, which enhances the power conversion efficiency (PCE) of the QDSSCs and affects the electrocatalytic activity for the reduction.

4.2. Characterization of the films X-ray diffraction (XRD) analysis was conducted to investigate the crystallinity of the composite CuS/NiS film, and the result is shown in Fig. 4a. The composite CuS/NiS films were cleaned with DI water and ethanol and dried in air after the CBD method and then analyzed without calcination. In The XRD patterns of CuS, the peaks at 2θ values of 27.3o, 39.0o, 53.5o, and 61.1o are assigned to the (110), (1 0 5), (1 1 4), and (2 0 4) crystal planes of the hexagonal structure (JCPDS card No. 03-065-3561). The remaining diffraction peaks are observed at 2θ values of 30.8o and 79.6o assigned to the (101) and (4 4 0) planes of NiS film (JCPDS card No. 00-012-0041). The XRD results confirmed that the composite CuS/NiS film was successfully formed on the FTO substrate. EDX analysis was conducted to investigate the elemental compositions of CuS, NiS, and CuS/NiS films on FTO substrate, as shown in the supplementary information (Fig. S1). The elements in the FTO were also detected due to the lower thickness of the film. Atomic percentages of 50.94% and 49.06% were observed for Cu and sulfur in the CuS thin film (Fig. S1 (a)). The film in Fig. S1 (b) contains atomic percentages of 50.77% Ni and 49.23% sulfur, corresponding to a ratio of nearly 1:1 for Ni: S. The Ni atomic percentage was higher and the Cu content was lower when depositing NiS on CuS film, indicating the occurrence of ion exchange from Cu to Ni. Surface information for the CuS, NiS, and flower-like CuS/NiS composite films was acquired by X-ray photoelectron spectroscopy (XPS). XPS was conducted to better understand the other impurities. Chemical bond configuration and composition of the as-prepared films, and the XPS spectrograms are shown in Fig. 4b. The Fig. 4b displays the survey spectra of the CuS, NiS, and CuS/NiS films, indicating five elements for CuS CE (S, C, Sn, O, and Cu) four elements for NiS CE (S, C, O, and Ni), and five elements for CuS/NiS CE (S, C, O, Ni, and Cu). O, C, and Sn came from the binder and substrate. The XPS results confirmed that there are no peaks for other impurities. The distinction between NiS and Ni2S or CuS and Cu2S is difficult [25]. The high-resolution XPS spectrum of Cu2p (shown in Fig. 5a) illustrates two strong peaks with binding energies of 932.68 and 952.61 eV corresponding to the Cu(2p3/2) and Cu(2p1/2) of the Cu2p phase, respectively [33]. The Cu2p photoelectron in Fig. 5a band displays shake-up characteristics on the high binding energy side of the Cu(2p3/2) and Cu(2p1/2) for Cu2+ species, and cupric elements can be simply distinguished from cuprous and pure Cu or Ni. However, in Fig. 5a, no shake-up satellites appear in the Cu2p spectra because of electron orbital interactions.

127

The XPS spectrum of S2p shown in Fig. 5b illustrates two strong peaks with binding energies of 161.96 and 169.44 eV corresponding to 2p3/2 and 2p1/2. This confirms the presence of CuS [33]. Fig. 5c exhibits the high-resolution XPS spectra of Ni2p. The four strong peaks at 855.51, 861.014, 873.014, and 879.214 eV are assigned to the binding energies of Ni2p3/2 (855.51 eV) and Ni2p1/2 (873.014 eV), respectively, indicating the existence of Ni+ [34]. The two S2p3/2 and S2p1/2 peaks located at around 161.96 and 169.44 eV (Fig. 5d) are assigned to sulfur anions in the lattice of NiS. The corresponding high-resolution XPS spectra of Ni2p and S2p for the CuS/NiS CE are shown in Fig. 5e and f.

4.3. Electrocatalytic ability of the counter electrodes toward polysulfide redox couple The charge transfer resistance is related to the electrocatalytic activity of a material. To further support the electrocatalytic ability and enhanced redox reaction at the CE/electrolyte interface, cyclic voltammetry (CV) was performed using a three-electrode system. The as-prepared CE served as the working electrode, a standard calomel electrode (SCE) was used as the reference electrode, and Pt wire was used as the counter electrode. A mixture of 0.1 M Na2S, 0.1 M sulfur, and 0.01 M KCl was used as the electrolyte. Fig. 6a shows the CV curves of CuS, NiS, and CuS/NiS CEs. It is notable that there are no significant redox peaks for the CuS and NiS CE, suggesting that they are poorer at catalyzing the reduction of electrolyte [35,36]. This is in good agreement with the EIS results. The CuS/NiS CE shows the highest redox peaks, current density, and reversibility, meaning better electrocatalytic activity and photovoltaic performance in QDSSCs. The high catalytic activity for the CuS/NiS CE is caused by the better surface morphology, surface roughness, and conversion rate of S2− ions from the oxidized polysulfide S2−n ions and tends to result in high current density [36]. Fig. 6b shows the cyclic voltammograms of the CuS/NiS CE at different scan rates varying from 20 to 200 mV s−1 within the potential range of −1.0 to 1.0 V. In Fig. 6b, each voltammograms has a similar form but the total current density or peak current increases with increasing square root of the scan rates. At low scan rates, the diffusion layer grows much farther from the CuS/NiS electrode, and the flux to the electrode surface is much smaller than at higher scan rates. Moreover, CV takes longer when the scan rate is lower. To elucidate the electrocatalytic behavior of the CuS, NiS, and CuS/ NiS electrodes, Tafel polarization curves were measured with symmetric dummy cells under dark conditions. The exchange current density is directly related to the electrocatalytic ability of an electrode, which can be estimated from the extra plated intercepts of the anodic and cathodic branches of the Tafel polarization curves, as shown in Fig. 7 [37, 38]. In the Tafel curves, the anodic branches (βa) and cathodic branches 2− (βc) indicate the oxidation of S2 to S2− n ions and the reduction of Sn to

Fig. 4. (a) X-ray diffraction spectrum of CuS/NiS thin film deposited on FTO substrate and (b) XPS spectra of the as-prepared CuS, NiS, and CuS/NiS films survey spectra on FTO substrate.

128


Fig. 5. High-resolution XPS spectra of Cu2p and S2p for CuS, (b) Ni2p and S2p for NiS, and (c) Ni2p and S2p for CuS/NiS film.

S2− ions. The slopes for the anodic or cathodic branches are in the order of CuS/NiS CE N CuS CE N NiS CE N Pt CE. The CuS/NiS CE performs very well at the interface of the CE and polysulfide electrolyte (with a

reduction rate almost equal to the oxidation state) and exhibits higher electrocatalytic activity. For the Pt, CuS, and NiS CEs, the reduction rate is much slower than the oxidation rate, which indicates that the

Fig. 6. (a) CV curves of the CEs with the films of CuS, NiS and CuS/NiS, obtained polysulfide electrolyte containing 0.1 M Na2S, 0.1 M sulfur, and 0.01 M KCl at a scan rate of 50 mV s−1. (b) CV curves of the CE with the film of CuS/NiS at different scan rates.


129

R is the gas constant, T is the temperature, F is Faraday's constant, and n is the number of electrons exchanged in the reaction at the CE/ electrolyte interface. The intersection of the cathodic branch with the Y-axis can be regarded as the limiting diffusion current density, Jlim. This value depends on the diffusion coefficient of the redox couple in the symmetrical dummy cells, which can be expressed by Eq. (2), where C is the concentration of the electrolyte, D is the diffusion coefficient of the polysulfide electrolyte, and l is the spacer thickness. The value of Jlim increases in the order of CuS/NiS CE N CuS CE N NiS CE N Pt CE, which is directly proportional to D and indicates a higher diffusion rate for the redox couple in the electrolyte. Theses Tafel polarization experiments are consistent with the CV and EIS analysis.

4.4. Photovoltaic performance of QDSSC devices Fig. 7. Tafel polarization curves of symmetrical cells fabricated with two identical counter electrodes.

reduction of the polysulfide electrolyte is poorer at the interface of the CE and electrolyte, and Jo was lower. The exchange current density mainly depends on Rct (the charge transfer resistance) and can be calculated using the following formulas [39].

Jo ¼ RT=nFRct

ð1Þ

D ¼ 1=2nFC J lim −1

ð2Þ

Fig. 8a shows the resultant current density vs. voltage curves of the QDSSCs assembled with the CuS, NiS, and CuS/NiS CEs and TiO2/CdS/ CdSe/ZnS photoanodes under simulated one-sun illumination with a power density of 100 mW cm−2. The corresponding photovoltaic parameters are listed in Table 1. The measurement of three QDSSCs each featuring a Pt CE exhibited Jsc, Voc, and FF values of 9.26 mA cm−2, 0.36 V, and 0.29, respectively, resulting in a very low PCE of 1.16%. The low PCE is due to the poor catalytic activity of the Pt in the polysulfide electrolyte or strong absorption of S2− on the surface, which leads to a low FF [40]. Compared with the Pt CE, FF and Voc were significantly improved for the CuS (0.58 and 0.42 V) and NiS CEs (0.51 and 0.35 V, respectively). The improved Voc and FF are primarily due to the relatively low RCE. When the optimized conditions of CuS and NiS are used to prepare efficient composite CEs of CuS/NiS, the QDSSCs exhibited better

Fig. 8. (a) Photocurrent-voltage characteristics of QDSSCs using TiO2/CdS/CdSe/ZnS as photoanode and CuS, NiS, CuS/NiS, and Pt CEs under the illumination of AM 1.5G, (b) Nyquist plots of TiO2/CdS/CdSe/ZnS QDSSCs consisting of CuS, NiS, CuS/NiS, and Pt CEs; (c) the Bode plot and (d) Open-circuit voltage decay curves for four different CEs: CuS CE, NiS CE, CuS/NiS CE, and Pt CE.

130


Table 1 Photovoltaic parameters and EIS results of CuS, NiS, CuS/NiS and Pt CEs Based on QDSSCs under One Sun. Parameters

CuS CE

NiS CE

CuS/NiS CE

Pt CE

Voc (V) Jsc (mA cm−2) FF PCE (%) Rs (Ω) RCE (Ω) Rct (Ω) Cμ (μF) Zw (Ω) τe (ms)

0.58 9.10 0.51 2.73 8.59 53.72 16.59 805.6 27.51 0.0732

0.42 12.15 0.35 1.82 8.66 71.92 12.24 469.5 23.91 0.0700

0.599 12.47 0.545 4.19 8.51 49.53 21.86 964.2 15.21 0.0876

0.36 9.26 0.29 1.16 8.67 85.55 22.41 267.6 21.89 0.0690

performance (PCE = 4.19%) than the cells with CuS (2.73%) and NiS (1.82%) due to increased Jsc, Voc, and FF values. Voc and FF of the QDSSCs are dependent on the series resistance (Rs) and shunt resistance (Rp) of the devices. A higher Rp contributes to a higher Voc, while a larger Rp or low Rs would lead to a higher FF in QDSSCs and DSSCs. The Rs value can be expressed as: Rs ¼ RTCO þ RCE þ Rdiff

ð3Þ

Where Rct is the charge transfer resistance at the CE/electrolyte interface, RTCO is the FTO substrate resistance, and Rdiff is the diffusion impedance in the electrolyte. A higher Rs can decrease the photo current (Isc) under short-circuit conditions according to the Eq. (4). Isc ¼ Iph –Io ½ððq: Isc : Rs Þ nKTÞ−1−Isc :Rs IRsh

ð4Þ

Where Iph is the photo current, Io is the reverse saturation current, n is the ideality factor, and Rsh is the shunt resistance. The higher Voc of the CuS/NiS CE is due to the considerable reduction in the electron recombination rates, which would also shift up the electron Fermi level of the TiO2 photo electrode. The better performance of the CuS/NiS electrode is related to that of the CuS and NiS and is due to the NiS particles in direct contact with the polysulfide electrolyte, the wide absorption spectrum, and different surface morphologies. The CuS/NiS electrodes show high electrocatalytic activity, which are in good agreement with the CV and Tafel results. We also performed EIS to study the electron charge behavior at the CE/electrolyte and photo electrode/electrolyte interfaces, as well as the Warburg diffusion resistance and the internal resistance. Fig. 8b shows Nyquist plots for QDSSCs based on CuS, NiS, CuS/NiS, and Pt as CEs with TiO2/CdS/CdSe/ZnS as the photoelectrode, which resembles a sandwich structure. Gaps between electrodes were filled with polysulfide electrolyte. EIS was carried out under the illumination of AM 1.5G simulated sunlight at an applied bias of Voc and various AC frequencies (100 mHz–500 kHz). The photoelectrode and CE had active areas of 0.25 and 0.7 cm2. The resulting electrochemical parameters are shown in Table 1. The equivalent series circuit comprises a series resistance (Rs) added to account for the non-zero intercepts on the real axis of the impedance plot. The left circle in the high frequency region and middle semicircle signify the electron transport at the CE/electrolyte interface and charge recombination at the photoanode/electrolyte interface. In the lower frequency region, the semi-circle on the Nyquist plot is associated with the Warburg diffusion coefficients in the redox electrolyte. The Rs values of the CuS, NiS, and Pt CEs are 8.59, 8.66, and 8.67 Ω, respectively, which are much higher than the Rs value of the CuS/NiS CEs (8.51 Ω). The very low value for CuS/NiS compared to the other three electrodes demonstrates good bonding strength between the CuS/NiS and FTO substrate, which effectively conducts electrons from the CE to the electrolyte. Smaller Rs and higher Cμ values at the photoelectrode and polysulfide electrolyte interface denote a lower

charge recombination rate and an upward shift of the Fermi level, which gives a higher Voc [41]. The RCE values of CuS, NiS, CuS/NiS, and Pt CEs were 53.72, 71.92, 49.53, and 85.55 Ω, respectively. The low RCE indicates an increase in the electrocatalytic activity of the CE, which results in acceleration of the electron transfer process at the CE/electrolyte interfaces and enhanced conversion efficiency. Note that higherefficiency QDSSCs or DSSCs usually exhibit a low value of RCE and Rct at the electrodes interfaces [42,43]. The results show that the QDSSCs made with the CuS/NiS had a lower Zw (Zw = 15.21 Ω), which indicates higher electrolyte diffusion, fast mass transfer of the electrons, and improved performance of the QDSSCs. The Cμ values for the CuS, NiS, CuS/NiS, and Pt CEs are 805.6, 469.5, 964.2, and 267.6 μF, respectively. The very high value for CuS/ NiS compared to the other electrodes demonstrates a higher surface area, which is advantageous for higher electrocatalytic activity. With the low values of RCE, Rs, and Zw and the higher value of Cμ for the CuS/NiS CEs, the electrocatalytic activities and the electronic and ionic transport processes were much higher than that of the bare CuS and NiS CEs [44]. This implies that the roughness of the electroactive material and compositional variations also influence the electrocatalytic activity. According to the EIS model, the electron lifetime (τe) can be estimated from the maximum angular frequency of the impedance semi-circle arc at mid-frequency. τe can be calculated using Eq. (5). τe ¼ 1=2π f max

ð5Þ

Where fmax is the peak frequency among the mid-frequency peaks. The resulting electron lifetime values are listed in Table 1 and shown in Fig. 8c. The τe values for the CuS, NiS, CuS/NiS, and Pt CEs are 0.0732, 0.0700, 0.0876, and 0.0690 ms, respectively. The CuS/NiS QDSSCs show a mid-frequency peak shifting toward the lower frequency side, which represents an enhanced electron lifetime. A shift to lower frequency represents a higher τe, and vice-versa. The higher τe supports the lower recombination rate of injected electrons for reduction of the polysulfide electrolyte, and the lower τe of the CuS, NiS, and Pt CEs is due to the higher back reaction of electrons with the electrolyte [45, 46]. τe of the QDSSCs is dependent on the longer diffusion length (Ln) and electron recombination rate. The Ln value can be calculated using Eq. (6). Ln 2 ¼ Deff xτ e

ð6Þ

Deff ¼ ðRk =Rw ÞL2 Keff

ð7Þ

Where Deff is the effective electron diffusion coefficient, Rk is the charge transfer resistance related to recombination, and Rw, Keff, and L are the electron transport resistance in the photoelectrode, the effective rate constant for recombination, and the photo electrode thickness, respectively. The improved QDSSC performance using the CuS/NiS CE is due to the positive influence of Ni ions on the CuS film and the improved surface morphology with improved electrocatalytic activity of the CEs. The open-circuit voltage decay (OCVD) was studied to investigate the electron recombination process in the QDSSCs based on CuS, NiS, CuS/NiS, and Pt CEs with the same type of photo-electrodes for all conditions. Fig. 8d shows the OCVD curves as a function of time (s) under open-circuit conditions. The QDSSCs based on the CuS/NiS CE showed a much slower Voc decay rate that was the best among our experimental results. This may be attributed to the adequate Ni particles on the CuS surface or the internal radial electric field that developed within the CuS/NiS nanoparticles. The primary reason for higher electron lifetime in the CuS/NiS compared to the other three CEs is the intrinsic material properties and prevention of electrons from flowing in the opposite direction, which indicates a lower recombination rate and higher electron lifetime. From the OCVD, the electron lifetime can be determined using


Eq. (8): −1

−ðKB T=eÞxðdVoc =dtÞ

ð8Þ

Where KB and T are the Boltzmann constant and temperature. Long-Term Stability Test in the Polysulfide Electrolyte. To verify that the CuS/NiS CEs are more reliable than the bare CuS CE in QDSSCs, the long-term stability of the CE with CuS/NiS film was studied in polysulfide electrolyte with constant scan rate (50 mV s−1). The CV was acquired in the same polysulfide electrolyte, and curves obtained for 50 cycles of scanning for the CEs with CuS and CuS/NiS are shown in the supplementary information (Fig. S2). The CV curve for the CuS/NiS CE shows a lower decrement in the cathodic peak current density than the CuS CE, indicating approximately constant electrocatalytic reduction. However, the CV curves of the CuS CE displays a substantial difference between the first CV cycle and the 50th cycle, as shown in Fig. S2 (b). The cathodic peak current density decreased dramatically after the 50th scan, indicating the deactivation of CuS in the polysulfide electrolyte. The CV analysis shows that the CE with CuS/NiS is more stable than the bare CuS CE. Moreover, stability of the CuS, NiS and CuS/NiS cells is carried out by exposing it to light illumination for 10 h continuously. From Fig. S3 in supporting information, the cell with the CuS, NiS and CuS/NiS shows the PCE of 2.73, 1.82 and 4.19%. When the illumination is continued to 10 h, the PCE of the QDSSC based on the CuS and NiS CE decreases from 2.73 to 1.71% and 1.82 to 1.41%. Fig. S3, it is vivid that the decrease in the PCE was found to be 7.05% for the CuS/NiS cell while the other cells such as CuS and NiS decreased drastically (62.64% and 22.53%). The CV, EIS, Tafel polarization, and J-V curves showed that the high electrocatalytic activity and charge transfer resistance of CuS/NiS CEs depend on the elemental composition, adequate NiS ions on CuS, the deposition temperature, and the adhesion between the material and FTO substrate. Hee-Je et al. prepared a cobalt sulfide/nickel sulfide composite by a simple CBD method and achieved a PCE of 3.40% under one-sun illumination [47]. In this study, a PCE of 4.19% was achieved using the CE with the CuS/NiS composite. The CE with the CuS/NiS composite renders higher efficiency than the CE with CoS/NiS reported in the literature. The CuS and NiS electrodes used in conjunction with a photoanode may not poison the photoanode surface, thereby increasing the PCE of the QDSSC. The RCE varied due to the surface morphology of the flower structure with uniform pores, the positive influence of the NiS composition on the CuS layer, and the increased conductivity enabling better reduction of the polysulfide electrolyte. We trust that through our continuous efforts on additional optimization of the composite based CE by changing the concentration of materials, thickness, morphology and fabrication technique, a new record PCE for QDSSCs will be simply achievable, which is progress in our laboratory. 5. Conclusion In summary, we have demonstrated for the first time that NiS coated on CuS films can be used as a counter electrode in QDSSC applications. A facile and inexpensive CBD method was used to prepare CuS/NiS CEs with densely packed flower-like structures. Interestingly, when NiS nanoparticles were deposited on CuS film, the morphology changes completely, and flower-like structure was observed on the CuS/NiS composite film. The QDSSCs based on CuS/NiS CE shows an impressive PCE of 4.19% with Voc of 0.599 V and Jsc of 12.47 mA cm−2. This PCE is higher than the values of bare CuS (PCE = 2.73%), NiS (PCE = 1.82%), and Pt CEs (PCE = 1.16%) in conjunction with polysulfide electrolyte and a TiO2/CdS/CdSe/ZnS photoanode. The conductivity is increased with deposition of NiS on CuS film, which produced very low Rs, RCE, Rct, and Zw for the CuS/NiS CE. This gives it better electrocatalytic activity, photocatalytic activity, and lower charge transfer resistance toward the reduction of polysulfide

131

electrolyte than the CuS, NiS, and Pt CEs, with adequate electron lifetime. These results confirm that the composite film is very effective at drastically decreasing the charge transfer resistance and enhancing the PCE. The presented preparation method and characterization of the CuS/NiS composite CE in QDSSCs will help us to design and manufacture ideal and potentially high-efficiency QDSSCs in the near feature.

Acknowledgements This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2011-0014437).

Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jelechem.2016.07.037.

References [1] B. O'Regan, M. Gratzel, A low-cost, high efficiency solar cell based on dye-sensitized colloidal TiO2 films, Lett. Nat. 353 (1991) 737–740. [2] L.M. Peter, Characterization and modeling of dye-sensitized solar cells, J. Phys. Chem. C 111 (2007) 6601–6612. [3] P.V. Kamat, Meeting the clean energy demand: nanostructure architectures for solar energy conversion, J. Phys. Chem. C 111 (2007) 2834–2860. [4] G.K. Mur, K. Shankar, M. Paulose, O.K. Varghese, C.A. Grimes, Use of highly-ordered TiO2 nanotube arrays in dye-sensitized solar cells, Nano Lett. 6 (2006) 215–218. [5] T.N. Murakami, M. Gratzel, Counter electrodes for DSC: application of functional materials as catalysts, Inorg. Chem. Acta 361 (2008) 572–580. [6] R. Easwaramoorthi, W.J. Lee, D.Y. Lee, J.S. Song, Spray coated multi-wall carbon nanotube counter electrode for tri-iodide (I− 3 ) reduction in dye-sensitized solar cells, Electrochem. Commun. 10 (2008) 1087–1089. [7] V. Jovanovski, V. Gonzalez-Pedro, S. Gimenez, E. Azaceta, G. Cabanero, H. Grande, R. Tena-Zaera, I. Mora-Sero, J. Bisquert, A sulfide/polysulfide-based ionic liquid electrolyte for quantum dot-sensitized solar cells, J. Am. Chem. Soc. 133 (2011) 20156–20159. [8] J.G. Radich, R. Dwyer, P.V. Kamat, Cu2S reduced graphene oxide composite for high2− efficiency quantum dot solar cells. Overcoming the redox limitations of S− at 2 /Sn the counter electrode, J. Phys. Chem. Lett. 2 (2011) 2453–2460. [9] C.Y. Lin, C.Y. Teng, T.L. Li, Y.L. Lee, H. Teng, Photoactive p-type PbS as a counter electrode for quantum dot-sensitized solar cells, J. Mater. Chem. A 1 (2013) 1155–1162. [10] P.V. Kamat, Quantum dot solar cells. Semiconductor nanocrystals as light harvesters, J. Phys. Chem. C 112 (2008) 18737–18753. [11] P.V. Kamat, K. Tvrdy, D.R. Baker, J. Radich, Beyond photovoltaics: semiconductor nanoarchitectures for liquid-junction solar cells, Chem. Rev. 110 (2010) 6664–6688. [12] A.J. Nozik, Nanoscience and nanostructures for photovoltaics and solar fuels, Nano Lett. 10 (2010) 2735–2741. [13] M. Shalom, S. Buhbut, S. Tirosh, A. Zaban, Design rules for high-efficiency quantumdot-sensitized solar cells: a multilayer approach, J. Phys. Chem. Lett. 3 (2012) 2436–2441. [14] I. Mora-sero, S. Gimenez, F. Fabregat-santiago, R. Gomez, Q. Shen, T. Toyoda, J. Bisquert, Recombination in quantum dot sensitized solar cells, Acc. Chem. Res. 42 (2009) 1848–1857. [15] S. Buhbut, S. Itzhakov, E. Tauber, M. Shalom, I. Hold, T. Geiger, Y. Garini, D. Oron, A. Zaban, Built-in quantum dot antennas in dye-sensitized solar cells, ACS Nano 4 (2010) 1293–1298. [16] O. Niitsoo, S.K. Sarkar, C. Pejoux, S. Ruhle, D. Cahen, G. Hodes, Highly efficient conversion of carbon dioxide catalyzed by half-sandwich complexes with pyridinol ligand: the electronic effect of oxyanion, J. Photochem. Photobiol., A 181 (2006) 306–309. [17] B.E. Hardin, H.J. Snaith, M.D. McGehee, The renaissance of dye-sensitized solar cells, Nat. Photonics 6 (2012) 162–169. [18] T.L. Li, Y.L. Lee, H. Teng, High-performance quantum dot-sensitized solar cells based on sensitization with CuInS2 quantum dots/CdS heterostructure, Energy Environ. Sci. 5 (2012) 5315–5324. [19] G. Hodes, J. Manassen, Electrocatalytic electrodes for the polysulfide redox system, J. Electrochem. Soc. 127 (1980) 544–549. [20] S.A. Sapp, C.M. Elliott, C. Contado, S. Caramori, C.A. Bignozzi, Highly efficient CdS quantum dot-sensitized solar cells based on a modified polysulfide electrolyte, J. Am. Chem. Soc. 133 (2011) 8458–8460. [21] S.S. Rao, C.V.V.M. Gopi, S.K. Kim, M.K. Son, M.S. Jeong, A.D. Savariraj, K. Prabakar, H.J. Kim, Cobalt sulfide thin film as an efficient counter electrode for dye-sensitized solar cells, Electrochim. Acta 133 (2014) 174–179. [22] H.J. Kim, J.H. Kim, C.H.S.S.P. Kumar, D. Punnoose, S.K. Kim, C.V.V.M. Gopi, S.S. Rao, Facile chemical bath deposition of CuS nano peas like structure as a high efficient counter electrode for quantum-dot sensitized solar cells, J. Electroanal. Chem. 739 (2015) 20–27.

132


[23] M.S. Faber, K. Park, M.C. Acevedo, P.K. Santra, S. Jin, Earth-abundant cobalt pyrite (CoS2) thin film on glass as a robust, high-performance counter electrode for quantum dot-sensitized solar cells, J. Phys. Chem. Lett. 4 (2013) 1843–1849. [24] C.V.T. Varma, C.V.V.M. Gopi, S.S. Rao, D. Punnoose, S.K. Kim, H.J. Kim, Time varied morphology controllable fabrication of NiS nanosheets structured thin film and its application as a counter electrode for QDSSC, J. Phys. Chem. C 119 (2015) 11419–11429. [25] A.D. Savariraj, K.K. Viswanathan, K. Prabakar, Influence of Cu vacancy on knit coir mat structured CuS as counter electrode for quantum dot sensitized solar cells, ACS Appl. Mater. Interfaces 6 (2014) 19702–19709. [26] C.J. Raj, K. Prabakar, A.D. Savariraj, H.J. Kim, Surface reinforced platinum counter electrode for quantum dots sensitized solar cells, Electrochim. Acta 103 (2013) 231–236. [27] H.J. Kim, T.B. Yeu, S.K. Kim, S.S. Rao, A.D. Savariraj, K. Prabakar, C.V.V.M. Gopi, Optimal-temperature-based highly efficient NiS counter electrode for quantum-dotsensitized solar cells, Eur. J. Inorg. Chem. (2014) 4281–4286. [28] S. Ruhle, M. Shalom, A. Zaban, Quantum-dot-sensitized solar cells, ChemPhysChem 11 (2010) 2290–2309. [29] S. Ruhle, M. Shalom, A. Zaban, Quantum-dot-sensitized solar cells, ChemPhysChem 11 (2010) 2290–2304. [30] C. Wu, Z. Wu, J. Wei, H. Dong, Y. Gao, Improving the efficiency of quantum dots sensitized solar cell by using pt counter electrode, ECS Electrochem. Lett. 2 (2013) H31–H33. [31] S.Q. Fan, B. Fang, J.H. Kim, J.J. Kim, J.S. Yu, J. Koa, Spontaneous induction of the uniform lying helix alignment in bimesogenic liquid crystals for the flexoelectro-optic effect, Appl. Phys. Lett. 96 (2010) 063501. [32] C. Jian-Ging, W. Hung-Yu, H. Kuo-Chuan, Using modified poly(3,4-ethylene dioxythiophene): poly(styrene sulfonate) film as a counter electrode in dye-sensitized, Sol. Energy Mater. Sol. Cells 91 (2007) 1472–1477. [33] C. Xing, Y. Zhang, Z. Wu, A. Jiang, M. Chen, Ion-exchange synthesis of Ag/Ag2S/ Ag3CuS2 ternary hollow microspheres with efficient visible-light photocatalytic activity, Dalton Trans. 43 (2014) 2772–2780. [34] H.W. Nesbitt, D. Legrand, G.M. Bancroft, Interpretation of Ni2p XPS spectra of Ni conductors and Ni insulators, Phys. Chem. Miner. 27 (2000) 357–366. [35] S.S. Rao, I.K. Durga, C.V.V.M. Gopi, C.V. Tulasivarma, S.K. Kim, H.J. Kim, The effect of TiO2 nanoflowers as a compact layer for CdS quantum-dot sensitized solar cells with improved performance, Dalton Trans. 44 (2015) 12852–12862.

[36] C.W. Kung, H.W. Chen, C.Y. Lin, K.C. Huang, R. Vittal, K.C. Ho, CoS acicular nanorod arrays for the counter electrode of an efficient dye-sensitized solar cell, ACS Nano 6 (2012) 7016–7025. [37] M. Wang, A.M. Anghel, B. Marsan, N.L.C. Ha, N. Pootrakulchote, S.M. Zakeeruddin, M. Gratzel, CoS supersedes Pt as efficient electrocatalyst for triiodide reduction in dyesensitized solar cells, J. Am. Chem. Soc. 131 (2009) 15976–15977. [38] M.X. Wu, X. Lin, T.H. Wang, J.S. Qiu, T.L. Ma, Low-cost dye-sensitized solar cell based on nine kinds of carbon counter electrodes, Energy Environ. Sci. 4 (2011) 2308–2315. [39] S.S. Rao, I.K. Durga, C.V.T. Varma, D. Punnoose, S.K. Kim, H.J. Kim, Enhance the performance of quantum dot-sensitized solar cell by manganese-doped ZnS films as a passivation layer, Org. Electron. 26 (2015) 200–207. [40] H. Chen, L. Zhu, H. Liu, W. Li, ITO porous film-supported metal sulfide counter electrodes for high-performance quantum-dot-sensitized solar cells, J. Phys. Chem. C 117 (2013) 3739–3746. [41] T. Shu, P. Xiang, Z.M. Zhou, H. Wang, G.H. Liu, H.W. Han, Y.D. Zhau, Mesoscopic nitrogen-doped TiO2 spheres for quantum dot-sensitized solar cells, Electrochim. Acta 68 (2012) 166–171. [42] Y.L. Lee, C.H. Chang, Efficient polysulfide electrolyte for CdS quantum dot-sensitized solar cells, J. Power Sources 185 (2008) 584–588. [43] H.K. Mulmudi, S.K. Batabyal, M. Rao, R.R. Prabhakar, N. Mathews, Y.M. Lam, S.G. Mhaisalkar, Solution processed transition metal sulfides: application as counter electrodes in dye sensitized solar cells (DSCs), Phys. Chem. Chem. Phys. 13 (2011) 19307–19309. [44] S.S. Rao, D. Punnoose, C.V. Tulasivarma, C.H.S.S. Pavan Kumar, C.V.V.M. Gopi, S.K. Kim, H.J. Kim, A strategy to enhance the efficiency of dye-sensitized solar cells by the highly efficient TiO2/ZnS photoanode, Dalton Trans. 44 (2015) 2447–2455. [45] R. Kern, R. Sastrawan, J. Ferber, R. Stangl, J. Luther, Modeling and interpretation of electrical impedance spectra of dye solar cells operated under open-circuit conditions, Electrochim. Acta 47 (2002) 4213–4225. [46] S.S. Rao, D. Punnoose, S.K. Kim, H.J. Kim, Low-cost solution processed nano millet like structure CoS2 film superior to Pt as counter electrode for quantum dot sensitized solar cells, Electron. Mater. Lett. 11 (2015) 485–493. [47] H.J. Kim, S.W. Kim, C.V.V.M. Gopi, S.K. Kim, S.S. Rao, M.S. Jeong, Improved performance of quantum dot-sensitized solar cells adopting a highly efficient cobalt sulfide/nickel sulfide composite thin film counter electrode, J. Power Sources 268 (2014) 163–170.