Abstract

CoS/nanocarbon (NC) composites were prepared via a one-pot hydrothermal method and were used as counter electrodes (CEs) in quantum dot-sensitized solar cells (QDSCs). The CoS/nanocarbon (NC) composite thin film CE has been prepared via a one-pot hydrothermal method. Addition of NC to the solution before hydrothermal treatment led to a CoS/NC composite with a good dispersion of conducting NC. The nanoscaled CoS in the composite CE provides abundant catalytic sites, and the carbon particle framework also acts as highly conductive paths for fast charge transport from the counter electrode (highly catalytic CoS active sites) to the photoanode. The optimized CoS/NC composite CE showed a two-order decrease in the charge-transfer resistance, compared to the pure CoS CE. The TiO2/CdS/CdSe/ZnS-based QDSC using the optimized CoS/NC composite CE shows enhanced photovoltaic performance with a power conversion efficiency of 4.46% and good stability (94.8% retention after 100 h continuous illumination).

1. Introduction

In past years, quantum dot-sensitized solar cells (QDSCs) have attracted significant attention as the next generation of dye-sensitized solar cells (DSSCs). Compared with traditional dyes employed in DSSCs, the strategic advantages of the use of low band gap QDs for light-trapping include widely tunable bandgap, broad spectral ranges, high extinction coefficients, and the expectation to obtain a considerable improvement in power conversion efficiency () by utilizing multiple exciton generation of QDs [13]. For most previously reported high-performance QDSCs, CoS was used as the CE, exhibiting remarkable catalytic activity [1, 46]. However, to the best of our knowledge, the pure CoS used as CE material alone is still very limited because of the relatively low electrical conductivity and poor stability for QDSCs. Therefore, many researchers have turned to combine CoS with other conducting materials to overcome the limitation for improving the electrical conductivity and cell performance [4, 5, 7]. Moreover, the specific surface area of the catalyst is another key factor of the counter electrode. The nanocarbon (NC) particles have advantages for photovoltaic applications because of their large surface-to-volume ratio and high electrical conductivity [8, 9], which would result in the effective interaction between the CEs and electrolyte solution and thus lead to an enhancement in charge-transfer efficiency [1, 10]. Accordingly, motivated by the above facts, it is envisioned that CoS composited with NC can provide superior electrocatalytic performance as a CE of a QDSC device by means of their positive synergistic effects.

In this paper, we report a facile one-pot method to prepare a CoS/NC composite CE for high-performance CdS/CdSe QDSCs. We compared the adding amount of NC on the performance of QDSCs with the CoS/NC composite CE under the same conditions. The nanoscaled CoS nanosheets in the CoS/NC composite CE provide an abundance of catalytic sites, and the NC framework acts as an excellent electrical tunnel for improved charge transport from an external circuit to highly catalytic CoS active sites. The optimized CoS/NC-30 composite displayed an enhanced catalytic activity toward polysulfide electrolyte reduction as a CE, which increased the value to 4.46%, in comparison to either CoS or NC CEs.

2. Experimental Section

2.1. Preparation of CEs

All of the chemical reagents were purchased from Sigma-Aldrich Shanghai Trading Co Ltd. and used without further purification. The CoS/NC composites were prepared by a facile one-pot hydrothermal method. CoNO3·6H2O (0.2328 g) and CH4N2S (0.1218 g) were dissolved in 20 mL of absolute ethanol. A calculated amount of NC powder was suspended in 20 mL ethanediol ultrasonically and then blended with the above solution and 3 mL ethylenediamine under magnetic stirring. The amounts of NC were set at five levels: 10 mg, 20 mg, 30 mg, 40 mg, and 50 mg. The above suspensions were poured into a Teflon-lined autoclave with a capacity of 100 mL and heated at 180°C for 12 h. Then, the product was collected and washed with deionized water and ethanol. The final product was dried at 80°C for 12 h to get the CoS/NC composite. Five different weight ratios of the NC to CoS samples were prepared by the same method and labeled as CoS/NC-10, CoS/NC-20, CoS/NC-30, CoS/NC-40, and CoS/NC-50. The comparison sample (without NC) was also prepared and denoted as CoS. The method used for the formation of the CEs is similar to that described by Yang et al. [11]. To prepare the CEs, 2 g of CoS/NC powder was dispersed into polyvinylidene fluoride (PVDF) with several drops of N-methyl pyrrolidone (NMP) as the solvent and 0.2 g of PVDF as the binder and then milled in a ball mill to form the slurries. And then the slurries were spread on FTO by the doctor blade method to form CE films. The films were air dried at 80°C for 1 h.

2.2. Preparation of Photoanodes and QDSCs

A TiO2 photoanode was constructed via layer-by-layer assembly of the TiO2 compact layer, porous TiO2 layer, and light-scattering layer. In a first step, the TiO2 compact layer was traditionally prepared by spray pyrolysis at 500°C [12, 13]. Subsequently, an 8 μm thick transparent layer with 20 nm sized anatase TiO2 particles, followed by another 3 μm thick light-scattering layer combined with 300 nm sized rutile TiO2 particles for the QD deposition, was screen printed onto the substrate. As-prepared films were dried at 80°C for 0.5 h. After sintering at 450°C for 0.5 h, the film was treated with 30 mM TiCl4 aqueous solution at 70°C for 0.5 h and then further sintered at 500°C for 0.5 h.

2.3. Photoanode Fabrication

The CdS/CdSe/ZnS QDSSCs were deposited onto the TiO2 photoanodes following the procedure from reference [7]. A CdS seed layer was in situ grown on the TiO2 SILAR process prior to CdSe deposition. For the preparation of the CdS seed layer, 5 SILAR cycles were conducted, employing Cd(CH3COO)2·2H2O aqueous solution (0.1 M) and Na2S aqueous solution (0.1 M) as the cation and anion source, respectively. CdSe QDs were grown on the surface of the TiO2/CdS film by a chemical bath deposition (CBD) technique. TiO2/CdS films were vertically immersed into an aqueous solution (a mixture of 80 mM CdSO4·8/3H2O, 80 mM Na2SeSO3, and 90 mM N(CH2CO2H)3) at 40°C for 2.5 h. The electrodes were sintered at 300°C in air for 1 h and cooled to room temperature. Finally, after CdSe QD deposition, the films were coated with 2 SILAR cycles of a ZnS passivation layer, by dipping alternatively into 0.1 M Zn(CH3COO)2·2H2O ethanol solution and 0.1 M Na2S methyl alcohol solution for 1 min. At last, the photoanode is labeled as TiO2/CdS/CdSe/ZnS.

2.4. QDSSC Device Fabrication

The QDSSC device was assembled with a TiO2/CdS/CdSe/ZnS photoanode, CoS/NC composites or commercial Pt CE, and a polysulfide electrolyte using a Surlyn spacer to maintain a distance of 30 μm between the two facing electrodes. The polysulfide electrolyte was a methanol/water (7:3) solution containing of 1 M Na2S, 2 M S, and 0.1 M KCl. The active area of the devices was 0.25 cm2. For stability tests, the assembled QDSSC devices were sealed with an EVA film under hot press.

2.5. Characterizations

The crystal structure was examined by X-ray diffraction (XRD) analysis (X’Pert MPD Pro diffractometer) with Cu Kα radiation operated at 40 kV and 40 mA. The morphologies and elemental compositions of the products were observed by field emission scanning electron microscopy (FE-SEM, Zeiss SUPRA 35VP) equipped with an energy dispersion X-ray spectroscopy (EDS). Transmission electron microscope (TEM) and high-resolution TEM images of the CoS/NC were performed on a JEM-1400 electron microscope. Electrochemical impedance spectroscopy (EIS) was executed on symmetrical cells using an electrochemical workstation (CHI660E, Shanghai Chenhua Device Company, China). The photovoltaic characterization of devices were measured using a solar simulator (Xe Lamp Oriel Sol3A™ Class AAA Solar Simulator 94023A, USA). The JV curves were also recorded by a Keithley 2400 source meter.

3. Results and Discussion

3.1. Structures and Morphologies of the Samples

In this section, we use CoS/NC-30 as a representative CoS/NC composite for the study of structures and morphologies. The difference between CoS/NC-10, CoS/NC-20, CoS/NC-30, CoS/NC-40, and CoS/NC-50 was only the different contents of NC. The phase purity and structural properties of as-prepared samples were determined by XRD. Figure 1 shows the XRD patterns of NC, CoS, and representative CoS/NC-30 powders. The diffraction peaks at 26.0° and 44.1° can be assigned to the (002) and (100) planes of the carbon phase (JCPDS no. 26-1080), while the other four major diffraction peaks appeared at are well indexed to the (100), (101), (102), and (110) planes of the hexagonal phase CoS (JCPDS no. 65-3418), respectively. After being composited with NCs, a small “bulge” near 26.0° for CoS/NC-30 is easily discerned, which reflects the mixed phase structure of its two components confirming the formation of the CoS/NC composite.

The microstructure and morphology of the CoS, NC, and CoS/NC-30 composites are demonstrated by SEM images in Figures 2(a)2(c). The SEM image shows that the CoS is composed of some irregular nanosheets ranged from several tens to hundreds of nanometers in edge size. NC presents as nanoparticles with a size of about 30 nm, as shown by the SEM image (Figure 2(b)) and TEM images (Figures 3 and 3(b)). The CoS nanosheets and NCs can be seen in the CoS/NC-30 composite. Clearly, CoS nanosheets are highly dispersed in the NC matrix. It was clear that the as-prepared CoS/NC-30 composite film had a fine porous structure and there was good electrical contact among the nanoparticles. Moreover, there is sufficient contact between CoS nanosheets and NCs as reflected in the TEM images (Figures 3(c) and 3(d)) of the CoS/NC-30. The microstructure of CoS nanosheets embedded in the NC framework can be clearly seen. This is favorable for electron transfer between CoS nanosheets and NC. The NC framework, together with CoS nanosheets, ensures the large contact surface area of the CoS/NC composite CE, thereby presenting abundant active sites for the reduction of the Sx2− ion [11]. This mesoporous structure will be favorable for improving catalytic activity as compared with the CoS counter electrode due to more active sites being exposed. Similar observations were also reported previously in the cases of NiCo2O4 or TiN nanoparticles adhering on carbon nanoparticles [14, 15]. Figures 2(d) and 2(e) show the EDS spectrum and EDS mapping images of CoS/NC-30 indicating the homogeneous distribution of Co, S, and C elements. There is a C element exiting in the EDS spectrum analysis, and C is apparently distinguished in the mapping image, indicating that the C materials are highly dispersed around CoS particles.

3.2. Electrochemical Properties

Previous research has already proved that NC can enhance the conductivity of the CE, while NC itself is not a suitable CE material for QDSCs [11, 16]. In order to estimate the effect of the NC content on the performance of CoS/NC composite CEs, five types of the CoS/NC composite with different NC content were successfully fabricated. Pure CoS and NC were also prepared as the reference. These thin film CEs have a similar thickness between 3.2 and 3.7 μm.

Electrochemical impedance spectroscopy (EIS) measurements have been done to study the charge-transfer processes between CEs and the electrolyte. Symmetric dummy cells with two symmetric CEs facing each other were prepared for EIS measurements. Figure 4 shows the Nyquist plots of Pt, pure NC, CoS, CoS/NC-10, CoS/NC-20, CoS/NC-30, CoS/NC-40, and CoS/NC-50 electrodes incorporating with a polysulfide redox electrolyte. One can see that the sheet resistance () of all CEs can be determined at high frequencies around 100 kHz. For our CEs, the first semicircle in the high-frequency region (between 10 and 100 kHz) corresponds to the solid−solid interface resistance () [17]. In the frequency region between 100 Hz and 10 kHz, the impedance was related to the carrier transport at the CE/electrolyte interface including the charge-transfer resistance () at the CE/electrolyte interface for I3 reduction and the double layer capacitance (constant phase element (CPE)) [18]. The parameters obtained by fitting the impedance spectra with the equivalent circuit shown in Figure 4(a) are summarized in Table 1. is a pivotal parameter reflecting the catalytic activity of the prepared CEs. The values of Pt and pure NC CEs are not satisfied (up to several hundred Ω cm2), indicating that their catalytic activity was unacceptably low in the polysulfide electrolyte. The value decreases in the order of . CoS/NC-30 CE achieves the lowest value of 14.47 Ω cm2, indicating that NC strongly enhances the charge-transfer ability of the CoS/NC composite-based CEs, though optimal amount of NC in CoS/NC composite CEs can enhance the conductivity properties of CE, and extremely excessive amounts will decrease the amount of effective catalytic sites of CoS nanosheets in the CoS/NC composite CE and therefore lead to a higher value of (up to 69.34 Ω cm2) for CoS/NC-50 CE. The similar finding was reported in the literature [11]. The largest differences between the CoS/NC composite-based CEs and pure CoS CE are probably related to the strength of bonding between CoS nanosheets or NCs within the film, which can be indicated from their microstructural properties presented above. As for the CoS/NC-based CEs, the CoS nanosheets have compact linkage with the NC particles, creating a valid framework to support the CoS nanosheets. The NC framework serves as the good electron transport tunnels and thus notably reduces the total internal resistance of the CoS/NC composite CEs, which ensures the efficient exploitation of highly catalytic CoS nanosheets. In the case of pure CoS CE, the poor connections between the loose stacked CoS nanosheets would create a bottleneck for the reflux electrons (the electrons flowing from the external circuit to the CE). This suggests that catalytic active sites on CoS nanosheets may not be efficiently utilized, thus leading to a higher overall value.

3.3. Photovoltaic Characteristics

Figure 5 shows the photocurrent–voltage (JV) curves for QDSCs based on various CEs measured under AM1.5 sunlight illumination (100 mW cm−2), and the photovoltaic parameters are summarized in Table 2. The QDSC using CoS/NC-30 CE shows optimal (), with improvement in the short-current density () and open-circuit voltage () compared to that of the pure CoS. In contrast, the QDSCs based on pure CoS CE exhibit a of 13.46 mA cm−2, a of 54%, and a of 3.12%, while NC by itself was not an appropriate CE material, showing relatively poor performance (). The Pt CE-based QDSC shows a much lower (1.36%), owing to the poor catalytic activity of these CE materials when immersed in polysulfide electrolyte. The other samples CoS/NC-10, CoS/NC-20, CoS/NC-30, CoS/NC-40, and CoS/NC-50 have also been investigated and are expected to get improved performances over pure CoS. Consistent with the EIS analysis, the devices with these five CEs (CoS/NC-10, CoS/NC-20, CoS/NC-30, CoS/NC-40, and CoS/NC-50) exhibit a certain increase in over pure CoS-based QDSCs, indicating the enhancement of charge transfer and a lowering of the overall internal resistance of the device by the NC. The negative effect of excessive amount of NC can be clearly seen by the detectable decrease of in QDSCs based on CoS/NC-50.

The long-term stability has been considered as one of the most crucial issues that limit the practical application of QDSCs. The CoS catalyst compositions together with the PVDF binder and conductive NC provide good physical contact between CE active materials and the FTO film, thereby resulting in the observed increase in the long-range stability of the CoS/NC CE. The stability of sealed QDSSCs with CoS and CoS/NC-30 composite CEs was investigated under continuous illumination for 100 h, as shown in Figure 6. The of CoS/NC-30 and CoS-based QDSSCs showed a stable increase in the first few hours, which can be partially attributed to a capillary effect of the nanochannel structure of TiO2, and improved ionic transport due to heating of the electrolyte [19]. The of QDSSCs based on the CoS/NC-30 and CoS CEs retained 94.8% and 83.2% of its initial value for continuous illumination for 100 h. This indicated that the CoS/NC-30 CE is highly stable working in the QDSC. Moreover, the existence of NC in the CoS/NC composite can serve as the framework to hold the CoS and act as a “nanobridge” to facilitate the electron transport and collection and ultimately improve the efficiency of QDSC. As shown in Table 3, the photovoltaic parameters obtained from our sample are comparable with those in the previous reports. The purpose of Table 3 is to serve as a relative comparison between the present QDSC and those QDSCs containing sulfide CEs. Compared to the previous reports, the of the CoS/NC-30-based QDSC reported in this manuscript achieves the outstanding value.

4. Conclusions

In summary, CoS/NC composite CEs with high electrocatalysis and stability have been synthesized using a low-cost and simple one-pot hydrothermal method. The CoS/NC composite catalyst with appropriate amount of NC can result in better enhancement of both electrical conductivity and catalytic activity. The solar cell obtained with the optimized CoS/NC-30-based CE reaches a high of up to 4.46%. The CoS/NC-30-based CE achieves the lowest value of 0.31 Ω cm2, which is two orders of magnitude lower than that of pure CoS. The present study shows that addition of NC directly into the solution prior to hydrothermal treatment prevents the increase in electronic conductivity of CoS. In the meantime, the stability test of sealed QDSC with CoS/NC-30 CE for 100 h proves that the CoS/NC-30 CE is highly stable working in the QDSC. In conclusion, the high-quality CoS/NC composite presented a promising application prospect in QDSCs.

Data Availability

The data used to support the findings of this study are included within the article.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was supported by the Key Applied Basic Research Program of Yunnan Province (Grant No. 2017FA024), the Program for Innovative Research Team (in Science and Technology) in University of Yunnan Province, the Natural Science Foundation of the Jiangsu Higher Education Institutions of China (Grant No. 17KJB140029), and the QingLan Project of Jiangsu Province.

Supplementary Materials

Figure S1 was added to provide the evidence that the cell realization procedure allows the realization of CdS/CdSe/ZnS quantum dots. The measured lattice spacing in Figure S1 is in accordance with the d-spacing of TiO2, CdS, ZnS, and CdSe. (Supplementary Materials)