Gas–solid reaction for in situ deposition of Cu₃SbS₄ on a mesoporous TiO₂ film

Abstract

Herein, a novel, facile, in situ gas–solid reaction method has been successfully employed for the deposition of famatinite (Cu₃SbS₄) semiconductor on a mesosporous TiO₂ film, where precursors CuCl₂ and SbCl₃ were first solution-coated on a TiO₂ substrate, followed by reaction with H₂S gas and further thermal annealing. Various precursor ratios, temperatures and heating atmospheres have been examined for the deposition. The phase-pure Cu₃SbS₄ has been obtained by coating a mixed solution of CuCl₂ and SbCl₃ at a molar ratio of 2.5 : 1, and annealing at 300 °C for 10 min in an H₂S atmosphere. The deposited Cu₃SbS₄ was uniformly distributed on the entire porous TiO₂ film with crystal grain sizes of about 3–4 nm. X-ray photoelectron spectroscopy (XPS) analysis revealed valence states of the synthetic samples for Cu⁺, Sb⁵⁺ and S²⁻, verifying phase-pure Cu₃SbS₄. The spectral absorption of the film ranges from 400 nm to 1000 nm with a band gap of ∼1.24 eV. The Cu₃SbS₄ film shows good and stable photoresponse performance, indicating its high potential as photovoltaic absorber.

---

Introduction

Thin film solar cells based on CdTe¹ and Cu(In_1−xGa_x)Se (CIGS)² have drawn much attention due to their excellent absorbance and high efficiencies of nearly 20%. Despite their success, these materials contain toxic (cadmium) and less-abundant elements (tellurium, indium, and gallium). Therefore, cheap, low-toxicity, and earth-abundant metal sulfide semiconductors, such as Cu₂ZnSn(S, Se)₄ (CZTSSe),^3–5 Fe₂S,^6,7 Cu₂FeSnS₄ (ref. 8–10) and SnS^11,12 have increasingly attracted attention as promising alternatives.

Cu–Sb–S ternary compounds, including four major phases: CuSbS₂ (chalcostibite),^13–16 Cu₁₂Sb₄S₁₃ (tetrahedrite),^17,18 Cu₃SbS₄ (famatinite),^19–21 and Cu₃SbS₃ (skinnerite),^22,23 have also emerged as potential absorber materials because of their appropriate optical, electrical properties and the presence of earth-abundant elements.²² Among them, Cu₃SbS₄ is suggested as a promising light absorber with an optimum band gap energy of ∼1.1 eV, comparable with those of high-efficiency CIGS and CZTSSe materials.²⁴ In the past years, various solution techniques have been explored for the preparation of nano/microscale Cu₃SbS₄ particles.^18,22,24,25 In 2013, J. Embden¹⁸ reported the synthesis of Cu₁₂Sb₄S₁₃ and Cu₃SbS₄ nanocrystals through optimizing the ligand chemistry in a hot-injection reaction. Moreover, colloidal nanocrystals of the four different Cu–Sb–S phases were obtained through adjusting the ratios of the precursors and reaction temperatures in the hot-injection method.²² Besides, G. Chen²⁶ studied the synthesis and formation mechanism of flower-like Cu₃SbS₄ particles via microwave irradiation. In contrast, there were few reports on the in situ preparation of Cu₃SbS₄ polycrystalline films. Franzer²⁷ reported fabrication of polycrystalline Cu₃SbS₄ films by magnetron sputtering. Very recently, U. Chalapathi²⁰ reported on the fabrication of Cu₃SbS₄ thin film by annealing the Sb₂S₃/CuS stacking layers, where both the Sb₂S₃ and CuS thin films with appropriate thicknesses are deposited by a chemical bath deposition technique. Both the methods, however, remain inherent disadvantages such as requisite high vacuum or difficulty in the control of the film with desired stoichiometric composition.

Recently, we developed a simple, low-cost, in situ gas–solid reaction method for the deposition of binary metal sulfide Sb₂S₃, CdS, and ternary CuSbS₂.^28–30 In this approach, a precursor salt was first introduced from solution into a nanoporous TiO₂ film, and subsequently transformed into sulfide through reaction with H₂S.^28–30 Herein, we further extended this method for the fabrication of Cu₃SbS₄ film. Through optimizing different precursor concentrations, temperatures and heating atmosphere, pure-phase Cu₃SbS₄ was achieved. The optical and electrical properties proved that the Cu₃SbS₄ film could be used as an attractive light absorber in photovoltaics.

Experimental

Fabrication of TiO₂/Cu₃SbS₄ device

First, the soda-lime glasses were cleaned by ultrasonic treatment with detergent, deionized water, ethanol and acetone successively, and then blow-dried by nitrogen gas. 20 nm-sized TiO₂ paste (DSL 18NM-T, Dyesol), diluted with ethanol as a mass ratio of 1 : 1.5, was blade coated on the cleaned glass substrates, followed by calcination at 125 °C for 5 min and 500 °C for 30 min. For the deposition of Cu₃SbS₄, various ethanolic solutions of CuCl₂ and SbCl₃ were prepared with the molar ratios of Cu : Sb set as 2 : 1, 2.3 : 1, 2.5 : 1, and 3 : 1. The precursor solution was first dropped on the mesoporous TiO₂ film and kept for 10 s, then spin-coated at 2000 rpm for 20 s. After the coating, the film was rapidly put in an enclosed quartz container full of H₂S gas (Fig. 1), which was generated by reaction of 20 wt% sodium sulfide (Na₂S·9H₂O) and 10 wt% dilute sulfuric acid (H₂SO₄). With the addition of Na₂S solution in the dilute H₂SO₄, H₂S gas was generated and entered into the reaction vessel. The TiO₂ film turned dark brown in color upon the addition. After 2 minutes, the film was taken out, washed by ethanol to remove the remnant and resulting by-products, then blown by nitrogen. Afterwards, the film was moved into the quartz container again, and annealed on a hot plate at different temperatures from 200 to 350 °C for 10 min.

Fig. 1 Schematic diagram of fabrication procedure for Cu₃SbS₄ film.

Fabrication of Cu₃SbS₄ device

Fluorine-doped SnO₂ (FTO) substrates with a resistivity of ∼10 Ω cm⁻¹ were cleaned with detergent, deionized water, ethanol and acetone successively. On the substrate, a dense titanium dioxide film was prepared as barrier layer to prevent short circuiting through sequentially deposition and annealing of 0.15 M and 0.3 M titanium (diisopropoxide) bis(2,4-pentanedionate) solutions in n-butanol. The TiO₂/Cu₃SbS₄ films were prepared with above-mentioned procedure. Finally, a 60 nm-thick silver (Ag) was deposited on the film as counter electrode by thermal evaporation.

Characterization

Cu₃SbS₄ thin films were characterized by X-ray diffraction (XRD) using a Bruker D8 Discover diffractometer (with Cu Kα radiation at 1.540598 Å). The surface morphologies and X-ray energy-dispersive spectrometer (EDS) of the Cu₃SbS₄ thin films were investigated by a field-emission scanning electron microscope (JSM-7500, Japan) and the transmission electron microscopy (TEM) images were acquired with a JEOL HT7700 microscope. X-ray photoelectron spectroscopy (XPS) were performed on the Thermo Scientific ESCA Lab 250Xi using 200 W monochromated Al Kα radiation, and 500 μm X-ray spot was used for XPS analysis. Typically the hydrocarbon C 1s line at 284.8 eV from adventitious carbon is used for energy referencing. The absorption spectra were measured using a UV/Vis/NIR spectrometer (PerkinElmer Lambda 950) and Raman spectrum were recorded on a Bruker Fourier Raman spectrometer (RFS100/S) with a laser excitation wavelength of 532 nm. Ultraviolet photoelectron spectroscopy (UPS) measurement was made on a Specs UVLS using He I excitation (21.15 eV, referenced to the Fermi edge of argon etched gold). The photoresponse performance was conducted on a computer-controlled Keithley 2400 under the illumination of AM 1.5 G. The applied voltage was 20 mV and active area of the device was 0.04 cm⁻².

Results and discussion

Preparation for pure-phase Cu₃SbS₄ film

Fig. 2 shows XRD patterns (a) and Raman spectra (b) of Cu₃SbS₄ film prepared with molar ratios of Cu : Sb set at 2 : 1, 2.3 : 1, 2.5 : 1, and 3 : 1, respectively. As shown in Fig. 2(a), the XRD pattern for the 2.5 : 1 film presents the diffraction peaks at 2θ = 28.70°, 47.78°, 56.66° and 64.75°, corresponding to (112), (204), (312) and (314) faces of famatinite Cu₃SbS₄ pattern (JCPDS 35-0581).²² All the peaks neither match with those of Sb₂S₃ or CuS,^31,32 nor the peaks of other crystalline structures, such as CuSbS₂ (chalcostibite), Cu₁₂Sb₄S₁₃ (tetrahedrite) and Cu₃SbS₃ (skinnerite),²⁴ indicating the complete conversion of CuCl₂ and SbCl₃ with the ratio at 2.5 : 1. At lower ratios of 2 : 1 and 2.3 : 1, several weak peaks from Sb₂S₃ can be distinguished, while at the higher ratio of 3 : 1, the peaks for Sb₂S₃ disappeared, but new peaks for CuS appeared together with those from Cu₃SbS₄. Thus, phase-pure Cu₃SbS₄ could be obtained only at the ratio of around 2.5 : 1. The deviation from the stoichiometric composition of Cu₃SbS₄ (3 : 1) could be explained as follows. In the initial stage of the reaction between the Cu²⁺/Sb³⁺ and H₂S, CuS and Sb₂S₃ formed at room temperature, as shown in eqn (1) and (2). During the annealing, the CuS and Sb₂S₃ reacted to form Cu₃SbS₄, as shown in eqn (3). In this process, Sb₂S₃ could gradually volatilize due to its effumability. Therefore, a little excess of Sb³⁺ precursor is essential to obtain pure Cu₃SbS₄ phase.

CuCl₂ + H₂S → CuS + 2HCl↑ (1)

2SbCl₃ + 3H₂S → Sb₂S₃ + 6HCl↑ (2)

6CuS + Sb₂S₃ → 2Cu₃SbS₄ + S↑ (3)

Fig. 2 XRD patterns (a) and Raman spectra (b) of Cu₃SbS₄ film prepared by different precursor ratio of 2 : 1 (1), 2.3 : 1 (2), 2.5 : 1 (3) and 3 : 1 (4) at 300 °C calcination.

The Raman spectrum of the 2.5 : 1 film shows shifts at 147, 198, 246, 273, 319, 345, 398, 515 cm⁻¹ and 637 cm⁻¹. Here, the peaks of 246, 273, 319, 345 cm⁻¹ and 637 cm⁻¹ are close to those of Cu₃SbS₄ phase,²⁰ while the peaks of 147, 198, 398 cm⁻¹ and 515 cm⁻¹ are from the TiO₂ substrate.³³ Notably, a distinct peak at 470 cm⁻¹ can be noticed in the spectrum of the 3 : 1 film, due to the presence of the excess CuS.³¹ In the case for the 2 : 1 and 2.3 : 1 films, a few of weak peaks are observed at 285 and 440 cm⁻¹ from Sb₂S₃.³⁴ The results are consistent with the phase analysis of XRD. Besides, the elemental composition and EDS spectra of the Cu–Sb–S films are presented in Table 1 and Fig. S1.† It is clear that at the ratio of 2.5 : 1, the average Cu/Sb/S elemental ratio (2.73 : 0.92 : 3.70) is very close to the stoichiometric value of 3 : 1 : 4 of Cu₃SbS₄. All the results indicate that phase-pure Cu₃SbS₄ film can be synthesized with CuCl₂ and SbCl₃ at the ratio of 2.5 : 1 by in situ gas–solid reaction method.

Table 1 Elemental composition of Cu₃SbS₄ film prepared by different precursor ratio of 2 : 1, 2.3 : 1, 2.5 : 1 and 3 : 1 (from SEM-EDX)

Ratio	2 : 1	2.3 : 1	2.5 : 1	3 : 1
Cu (%)	2.97	4.12	2.73	3.27
Sb (%)	1.15	1.43	0.92	0.95
S (%)	4.10	5.27	3.70	3.90

---

In previous reports, reaction temperature was regarded as an pivotal factor for controlling the ternary phase composition and crystallinity.^5,20 Here, various annealing temperatures ranged from 200 °C to 350 °C were examined, and the corresponding XRD patterns were shown in Fig. 3. The peak intensity for Cu₃SbS₄ at 2θ = 28.70°, 33.33°, 47.78°, 56.66° and 64.75° are very weak when annealing at 200 °C for 10 min, implying poor crystallinity under this condition.²⁶ Increasing the temperature to 300 °C, the crystallinity is evidently improved with the enhanced intensity of all the peaks. With further increasing the temperature to 350 °C, new peaks for CuS appeared. The observation may imply that at this high temperature annealing, the Cu₃SbS₄ was partially decomposed to CuS and Sb₂S₃, and the resulting CuS remained after volatilization of the Sb₂S₃. Besides, different atmosphere could affect the crystal phases during the annealing (Fig. 4). Here, the annealing was carried out under H₂S or N₂ atmosphere, respectively. Different from in H₂S atmosphere, where only pure Cu₃SbS₄ phase appeared, both Cu₃SbS₄ and CuSbS₂ phases were observed under N₂ annealing ambience. The result implies that H₂S environment is requisite to supply adequate amount of S element and avoid the formation of other Cu–Sb–S phases due to the loss of S element during the annealing. All the results demonstrated that pure famatinite Cu₃SbS₄ film was achieved by the in situ gas–solid reaction with Cu²⁺ : Sb³⁺ at the ratio of 2.5 : 1 and annealing in H₂S at 300 °C for 10 min. Here, the film prepared with the ratio at 2.5 : 1 is named as 2.5 : 1 film for further investigation.

Fig. 3 XRD patterns of Cu₃SbS₄ film as a precursor ratio of 2.5 : 1 heated at different temperatures for 10 min: 200 °C (a), 250 °C (b), 300 °C (c), 350 °C (d).

Fig. 4 XRD patterns of Cu₃SbS₄ film as a precursor ratio of 2.5 : 1 heated at H₂S and N₂ atmosphere at 300 °C for 10 min.

XPS analysis was carried out to investigate the elemental state and composition of the 2.5 : 1 film. From the full analysis of XPS spectrum (Fig. 5(a)), no peaks were observed for any other elements except Cu, Sb, S, Ti, O and C. Here, carbon element is adventitious used for energy referencing, and the peaks for Ti and O are attributed to porous TiO₂ substrate. The spectrum of Cu (Fig. 5(b)) shows two Gaussian peaks at 952.2 and 932.3 eV with a separation of 19.9 eV, which are corresponding to Cu 2p_1/2 and Cu 2p_3/2 binding state, respectively. The above binding energies are well consistent with reported values for Cu⁺.³⁵ The antimony presents doublet at the binding energies at 539.7 eV (3d_3/2) and 530.4 eV (3d_5/2), which are in good agreement with those of Sb⁵⁺, and no peaks are observed at 538.7 and 529.2 eV for Sb³⁺.^20,36 The peaks at 162.9 and 161.8 eV represent the S 2p_1/2 and 2p_3/2, respectively, which are consistent with the chemical state of sulfur (S²⁻) in the Cu₃SbS₄. Thus, XPS analysis indicates that the elemental state of the 2.5 : 1 film are (Cu⁺)₃(Sb⁵⁺)(S²⁻)₄.

Fig. 5 X-ray photoelectron spectra (XPS) survey scan (a), Cu (2p) (b), Sb (3d) (c), and S (2p) (d) core levels of 2.5 : 1 film and heat treatment in H₂S at 300 °C.

The microstructure of the 2.5 : 1 film was further evaluated by recording the SEM and TEM images (in Fig. 6). According to the top-view SEM, uniform morphology is evident with many tiny Cu₃SbS₄ particles homogeneously distributed within the TiO₂ film, where the specific particle size could hardly be distinguished. In the TEM, lots of small granules of Cu₃SbS₄ can be observed on the surface of 20 nm TiO₂ balls with crystal grain sizes of about 3–4 nm. Besides, the cross sectional SEM (Fig. 6(c)) displays the thickness of TiO₂/Cu₃SbS₄ is about 1.1 μm. From the cross section, the mapping images were recorded, as shown in Fig. S2,† manifesting the homogeneous deposition of Cu, Sb and S elements in the TiO₂ film.

Fig. 6 Top-view SEM (a), TEM (b) and cross sectional SEM (c) images of Cu₃SbS₄ film prepared by in situ gas–solid reaction method as a precursor ratio of 2.5 : 1 film and heat treatment in H₂S at 300 °C.

The determination of band gap and positions are significant for a potential photovoltaic absorber material. The UV-Vis-NIR absorption spectrum of Cu₃SbS₄ film shows a wide absorption range from 400 nm to 1000 nm. The (αhν)^1/2 versus hν curves (Tauc plots, inset of Fig. 7(a)) is plotted, and band gap is estimated to be 1.24 eV, which is closed to the reported band gap of 1.2 eV for Cu₃SbS₄.^22,26 Moreover, UPS was applied to determine valance band and Fermi energies of this film. From intercept of 16.54 eV for secondary electron onset and Au energy of 21.15 eV, Fermi energy position relative to vacuum is calculated to be −4.61 eV. The distance between Fermi energy and valance band is 0.25 eV by extrapolating and linear fitting the long tail of UPS spectrum. Combining with the band gap of 1.24 eV estimated from UV-Vis-NIR spectra, the valence band edge and conduction band edge are determined to be located at 4.86 eV and 3.62 eV relative to vacuum, respectively, which suggests its potential application as one kind of solar absorber materials.^17,37 The band alignment diagrams of TiO₂/Cu₃SbS₄ film were given as shown in Fig. S3.† It is clear that there exists some excessive driving energy (−0.6 eV) for electron transfer, potentially leading to relatively low theoretical voltage. Thus, it will be desirable to select other N-type electrodes with relatively lower conduction band edge for high open-circuit voltages in the future studies.

Fig. 7 UV-Vis-NIR absorption spectra and Tauc plots (a) and UPS full spectrum (b) of 2.5 : 1 film. Linear fitting of UPS spectrum in the range of 16.2–17.0 eV (c) and −1.0 to 2.0 eV (d).

Photoresponse performance was investigated to evaluate the quality of the 2.5 : 1 film, and the results are presented in Fig. 8. The current–potential (I–V) curves are nearly linear, indicating ohmic contact between silver counter electrode and Cu₃SbS₄ semiconductor. Compared with dark current, the corresponding photocurrent increased remarkably over the entire bias ranged from −700 mV to 700 mV, which is attributed to excited charge carriers by simulated sunlight, leading to high photocurrent. Besides, photocurrent response curve conducted under a bias of 20 mV presents an excellent photoresponse and high photostability. The dark current of 1.96 × 10⁻³ mA cm⁻² dramatically rises to 3.50 × 10⁻² mA cm⁻² when the light was on, and the light current drops rapidly as soon as light source was turned off. Despite repeating switch cycle for several times, photocurrent still remains at 3.48 × 10⁻² mA cm⁻² without observable attenuation. Above all, Cu₃SbS₄ made by in situ gas–solid reaction method exhibits high photoresponse performance and provides a possibility to be used as photovoltaic absorption layer.

Fig. 8 Current–potential curves (a) and photocurrent response curve (b) under darkness and light for Cu₃SbS₄ device prepared by in situ gas–solid reaction method.

Conclusions

In this report, an in situ gas–solid reaction method has been successfully employed for the deposition of pure-phase Cu₃SbS₄ film. Various precursor molar ratios, temperatures, and heating atmospheres have been explored for the deposition. The results show that phase-pure Cu₃SbS₄ can be obtained by the in situ gas–solid reaction with the molar ratio of CuCl₂ and SbCl₃ at 2.5 : 1, and annealing in H₂S at 300 °C. The Cu₃SbS₄ particles with crystal grain sizes of about 3–4 nm are uniformly distributed on the porous TiO₂ film. The Cu₃SbS₄ film shows a broad absorption ranged from 400 nm to 1000 nm with the band gap of 1.24 eV. Moreover, the Cu₃SbS₄ film presents high and stable photoresponse performance, indicating its potential application as solar absorber materials.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the 973 Program (No. 2013CB933004), the National Nature Science Foundation of China (Grant No. 61405207, 21174149, 21572235, 51473173, 91433202 and 21221002), and the “Strategic Priority Research Program” of Chinese Academy of Sciences (Grant No. XDA09020000 and XDB12010200). The measurements of (NMR, Mass Spectrometry, XRD, Photoelectron Spectroscopy) were performed at the Center for Physicochemical Analysis and Measurements in ICCAS.

References

1. M. Gloeckler, I. Sankin and Z. Zhao, IEEE J. Photovolt., 2013, 3, 1389–1393.
2. P. Jackson, D. Hariskos, R. Wuerz, O. Kiowski, A. Bauer, T. M. Friedlmeier and M. Powalla, Phys. Status Solidi RRL, 2015, 9, 28–31.
3. S. C. Riha, B. A. Parkinson and A. L. Prieto, J. Am. Chem. Soc., 2009, 131, 12054–12055.
4. H. Katagiri, K. Jimbo, S. Yamada, T. Kamimura, W. S. Maw, T. Fukano, T. Ito and T. Motohiro, Appl. Phys. Express, 2008, 1, 041201.
5. A. S. R. Chesman, N. W. Duffy, S. Peacock, L. Waddington, N. A. S. Webster and J. J. Jasieniak, RSC Adv., 2013, 3, 1017–1020.
6. J. Puthussery, S. Seefeld, N. Berry, M. Gibbs and M. Law, J. Am. Chem. Soc., 2011, 133, 716–719.
7. M. Wang, D. Xue, H. Qin, L. Zhang, G. Ling, J. Liu, Y. Fang and L. Meng, Mater. Sci. Eng., B, 2016, 204, 38–44.
8. C. Yan, C. Huang, J. Yang, F. Liu, J. Liu, Y. Lai, J. Li and Y. Liu, Chem. Commun., 2012, 48, 2603–2605.
9. X. Jiang, W. Xu, R. Tan, W. Song and J. Chen, Mater. Lett., 2013, 102, 39–42.
10. L. Li, X. Liu, J. Huang, M. Cao, S. Chen, Y. Shen and L. Wang, Mater. Chem. Phys., 2012, 133, 688–691.
11. P. Sinsermsuksakul, K. Hartman, S. B. Kim, J. Heo, L. Sun, H. H. Park, R. Chakraborty, T. Buonassisi and R. G. Gordon, Appl. Phys. Lett., 2013, 102, 053901.
12. K. T. R. Reddy, N. K. Reddy and R. W. Miles, Sol. Energy Mater. Sol. Cells, 2006, 90, 3041–3046.
13. S. Suehiro, K. Horita, M. Yuasa, T. Tanaka, K. Fujita, Y. Ishiwata, K. Shimanoe and T. Kida, Inorg. Chem., 2015, 54, 7840–7845.
14. L. Wan, C. Ma, K. Hu, R. Zhou, X. Mao, S. Pan, L. H. Wong and J. Xu, J. Alloys Compd., 2016, 680, 182–190.
15. T. Rath, A. J. MacLachlan, M. D. Brown and S. A. Haque, J. Mater. Chem. A, 2015, 3, 24155–24162.
16. Y. C. Choi, E. J. Yeom, T. K. Ahn and S. Seok II, Angew. Chem., Int. Ed., 2015, 54, 4005–4009.
17. L. Wang, B. Yang, Z. Xia, M. Leng, Y. Zhou, D.-J. Xue, J. Zhong, L. Gao, H. Song and J. Tang, Sol. Energy Mater. Sol. Cells, 2016, 144, 33–39.
18. J. Embden, K. Latham, N. W. Duffy and Y. Tachibana, J. Am. Chem. Soc., 2013, 135, 11562–11571.
19. C. T. Crespo, J. Phys. Chem. C, 2016, 120, 7959–7965.
20. U. Chalapathi, B. Poornaprakash and S.-H. Park, Ceram. Int., 2017, 43, 5229–5235.
21. K. Aup-Ngoen, T. Thongtem and S. Thongtem, Mater. Lett., 2012, 66, 182–186.
22. K. Ramasamy, H. Sims, W. H. Butler and A. Gupta, Chem. Mater., 2014, 26, 2891–2899.
23. L. Yu, R. S. Kokenyesi, D. A. Keszler and A. Zunger, Adv. Energy Mater., 2013, 3, 43–48.
24. K. Chen, J. Zhou, W. Chen, Q. Chen, P. Zhou and Y. Liu, Nanoscale, 2016, 8, 5146–5152.
25. S. Ikeda, S. Sogawa, Y. Tokai, W. Septina, T. Harada and M. Matsumura, RSC Adv., 2014, 4, 40969–40972.
26. G. Chen, W. Wang, J. Zhao, W. Yang, S. Chen, Z. Huang, R. Jian and H. Ruan, J. Alloys Compd., 2016, 679, 218–224.
27. N. D. Franzer, N. R. Paudel, C. Xiao and Y. Yan, in Photovoltaic Specialist Conference (PVSC), 2014 IEEE 40th, 2014, pp. 2326–2328.
28. L. Zheng, K. Jiang, J. Huang, Y. Zhang, B. Bao, X. Zhou, H. Wang, B. Guan, L. Yang and Y. Song, J. Mater. Chem. A, 2017, 5, 4791–4796.
29. Y. Zhang, J. Tian, K. Jiang, J. Huang, L. Zhang, H. Wang, B. Bao and Y. Song, J. Mater. Sci.: Mater. Electron., 2017 DOI:10.1007/s10854-017-7263-1.
30. Y. Zhang, J. Tian, K. Jiang, J. Huang, H. Wang and Y. Song, Mater. Lett., 2017, 209, 23–26.
31. T. Hurma and S. Kose, Optik, 2016, 127, 6000–6006.
32. C. D. Lokhande, B. R. Sankapal, R. S. Mane, H. M. Pathan, M. Muller, M. Giersig and V. Ganesan, Appl. Surf. Sci., 2002, 193, 1–10.
33. W. F. Zhang, Y. L. He, M. S. Zhang, Z. Yin and Q. Chen, J. Phys. D: Appl. Phys., 2000, 33, 912–916.
34. R. Parize, T. Cossuet, O. Chaix-Pluchery, H. Roussel, E. Appert and V. Consonni, Mater. Des., 2017, 121, 1–10.
35. M. Kundu, T. Hasegawa, K. Terabe, K. Yamamoto and M. Aono, Sci. Technol. Adv. Mater., 2008, 9, 035011.
36. C. An, Y. Jin, K. Tang and Y. Qian, J. Mater. Chem., 2003, 13, 301–303.
37. J. v. Embden and Y. Tachibana, J. Mater. Chem., 2012, 22, 11466–11469.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7ra08137e

---