Cu₂ZnSnS₄ synthesized through a low-cost reflux method

1 Introduction

With an increasing demand for nonconventional energy sources, highly efficient but low-cost solar cells play an important role for future energy requirements. Although silicon-based technology has established itself strongly with the high-power conversion efficiencies, the need for the thick layers owing to the indirect band-gap of silicon leads to higher costs. In contrast, thin-film technologies use a few tens of micrometers of absorber materials with a direct band-gap. Among various types of thin-film absorber materials, CuIn_xGa_1-xSe (CIGS) has exhibited high conversion efficiency (20%) [1]. The scarcity and high cost of the elements involved is the main drawback for their development. The chalcopyrite quaternary semiconductor Cu₂ZnSnS₄ (CZTS), a structural analogue to CIGS but with earth-abundant, low-cost elements, is an ideal substitute for CIGS. CZTS is an upcoming absorber material with an optimum band-gap of the range 1.4 to 1.5 eV [2–5]. CZTS has a high optical absorption coefficient of the order 10⁴ cm^-1 [6], which is one order higher in magnitude compared to CIGS, leading to the use of reduced absorption layer thickness. To date, the absorber material in devices that have achieved best efficiencies in CZTS-based solar cells has been synthesized through liquid-phase methods [7]. Because the secondary phases play a negative role in the efficiency of the cells [8], it is therefore necessary to obtain a single-phase material, which is still a major challenge in this research. CZTS has been synthesized through various methods, such as spray deposition [9], coevaporation [10], sputtering [11], chemical bath deposition [12], and electrodeposition [13]. The above-mentioned techniques either are ultrahigh vacuum based or use toxic solvents. This will lead to both an increase in the cost and a negative impact on the environment. Hence, a method that does not involve high vacuum and adopts water as solvent is highly desired.

Reflux is an established and efficient method for obtaining powders. It employs water as solvent and is very clean. Another advantage of this method is the ability to control the size and morphology of the products by varying the time of reaction. In this paper, we report the use of the reflux method for a cost-effective synthesis of pure-phase quaternary CZTS chalcogenide nanoparticles.

2 Materials and methods

The starting materials used are CuSO₄·H₂O, (CH₃COO)₂Zn, SnCl₂·2H₂O, and thiourea. All materials are pure and used without any further purification. Separate aqueous solutions of 1 m cupric sulfate, 0.5 m zinc acetate, and 0.5 m stannous chloride were prepared. These solutions were then added to 4 m aqueous solution of thiourea. The contents were transferred to a round-bottomed flask fitted with a reflux condenser. The flask was heated with the help of an electric heating mantle. (The schematic is shown in Figure 1A.) The solution was refluxed for 8 h and allowed to gradually cool to room temperature. Initially, the solution was milky white and, upon heating, turned grayish black. The products were filtered and washed with distilled water and ethanol several times. Then, the obtained powder was dried under vacuum. The flow chart of the reaction sequence is shown in Figure 1B.

Figure 1:

(A) Schematic of the reflux setup and (B) flow chart of reaction sequence of CZTS synthesis.

3 Reaction mechanism

The probable mechanism of the reaction is as follows: 2CuSO₄·5H₂O+(CH₃COO)₂Zn+SnCl₂·2H₂O+4SC (NH₂)₂+8H₂O→Cu₂ZnSnS₄+CO₂+8NH₄Cl·2Cu²⁺+Zn²⁺+Sn²⁺+4S^2-→Cu₂ZnSnS₄.

4 Results

4.1 Phase identification

Phase identification was done by using a panalytical X-ray diffractometer with CuKα radiation λ=0.15406 nm and step size=2°/min. The X-ray diffraction (XRD) pattern of the as-synthesized powder is shown in Figure 2. The peaks at 2θ=28.78°, 47.97°, and 57.00° correspond to the (112), (220), and (312) planes of the kesterite CZTS (JCPDS card no. #00-026-0575), which is in tune with the literature [14]. There are some additional peaks that might correspond to the formation of a small amount of secondary phase. The shoulder peaks at 26.6° and 32° correspond to the (-1 -3 1) and (-2 0 6) planes of the mohite Cu₂SnS₃ (copper tin sulfide) given by the reference pattern JCPDS card no. #00-035-0684.

Figure 2:

XRD pattern of the as-synthesized powder.

From the XRD data, the average crystallite size is calculated using the Scherrer formula:

D=0.9λβcosθ

where D is the average crystallite size (diameter), λ is the wavelength of the incident radiation, θ is the Bragg angle, and β is the full-width (in radians) of the peak at half-maximum intensity.

The average crystallite size calculated from the above formula is 2.5 nm. The small size of the crystallites is the reason for the peak broadening in the XRD.

Because the XRD patterns of CZTS and ZnS are similar, the formation of the phase needed further confirmation, which was achieved by Raman spectroscopy.

For the Raman spectra, a Horiba iHR 550 Raman spectrophotometer illuminated with a 532 nm laser beam was used. The peaks at 282 and 338 cm^-1, respectively, correspond to the CZTS phase. The absence of peaks at 348 and 356 cm^-1 indicates that ZnS is not present in the sample. The other peaks at 267 and 310 cm^-1 and a shoulder peak at 352 cm^-1 correspond to the Cu₂SnS₃ phase [15]. Figure 3 shows the Raman spectrum of the as-synthesized sample.

Figure 3:

Raman spectrum of the as-synthesized powder.

4.2 Band-gap calculations

The optical band-gap studies were done with a Shimadzu UV 2450 UV-visible spectrophotometer fitted with an integrating sphere unit.

Diffuse reflectance spectroscopy is a technique that enables one to find the band gap of any powder sample. In this method, the sample is ground into fine powder and placed in a sample holder. The reflectance of the sample is measured with respect to the reflectance of a standard (BaSO₄). The reflectance versus wavelength plot is then obtained.

The diffuse reflectance spectra are converted to the equivalent absorption spectra using the Kubelka-Munk equation [16]:

(1)f(R)=(1−R)22R

where f(R) (remission or Kubelka-Munk function) is the equivalent of absorption coefficient and R is the reflectivity.

The band gap (E_g ) of the sample and the remission function are related through the following equation:

(2)f(R)=C(hv−Eg)nhv

where C is an arbitrary constant, hν is the photon energy, and n=1/2 for the direct allowed transition band-gap materials.

The band-gap energy of the sample is obtained by following these steps:

1. The remission function f(R) is determined using Equation (1) and the measured reflectivity.

2. A graph of [f(R)hν]² versus hν is plotted.

3. The linear part of the curve is extrapolated to the x-axis.

This point of intersection with the x-axis is the band-gap energy of the material. This procedure has been repeated three times, and the average value is reported in Table 1.

Table 1

Band-gap values of the different batches of the as-synthesized sample.

Batch no.	Band gap (eV)
1	1.37
2	1.41
3	1.42
Average	1.40

The band gap of the material was derived from the UV-visible reflective measurements. The Kubelka-Munk function [Equation (1)] is used to convert the reflectance into absorbance data. Plotting (αhν)² versus hν upon extrapolation, the average value of the band gap is found to be 1.4 eV with a standard deviation of 0.02. This indicates that CZTS absorbs at approximately 860 nm, which is desirable for photovoltaic applications. Figure 4 shows the band-gap measurement plot.

Figure 4:

Plot showing the determination of the band-gap energy from diffuse reflectance measurements of the as-synthesized powder.

4.3 Field-emission scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDX)

The electron micrographs were taken using a Zeiss SEM fitted with an EDX detector. The elemental ratios as obtained from the EDX detector are presented in Table 2. The corresponding electron micrograph is shown in Figure 5. The particle size analysis from the SEM micrograph is done by using ImageJ software. From the SEM image, the morphology is known to be spherical with an average particle size of 140 nm (standard deviation 17 nm) formed from smaller spherical particles of average size 24±5 nm. The EDX results indicate that the elements in the sample are stoichiometric, indicating a pure phase. The metal ratios are as follows:

Table 2

Relative elemental ratios of the synthesized sample as detected by EDX detector.

Element	Weight %	Atomic %
S K	17.61	32.47
Cu L	49.57	46.12
Zn L	12.46	11.27
Sn L	20.36	10.14
Total	100.00	100.00

Figure 5:

SEM of the as-synthesized sample showing the formation of CZTS nanoparticles.

ZnSn=1.1

CuSn+Zn>1

This indicates that the sample is slightly Cu rich but is in tune with the literature for the formation of stoichiometric CZTS [17].

5 Discussion

CZTS nanoparticles were synthesized using the reflux method. The reflux method was chosen because of its ability to produce powders with uniform morphology and narrow size distribution [18]. For the fact that the equipment involved is minimal and uses water as solvent, reflux is also a cost-effective method for achieving low-cost solar cells. The XRD results show that kesterite CZTS has formed. Given the complexity of the system, the likelihood of the formation of secondary phases is high. There is a small amount of secondary-phase Cu₂SnS₃ as indicated by the XRD data. The EDX results show that the sample is Cu rich. As per the literature, solvent treatment can be employed to remove the secondary phases [17]. Moreover, Cu₂SnS₃ itself is used as an absorber material for photovoltaic applications [19]. The morphology of the sample as seen from the SEM image is spherical in nature with an average particle size of 140 nm, and these agglomerates consisted of smaller particles of average size 25 nm. The crystallite size as calculated from the Scherrer formula using the XRD data is 2.5 nm, indicating that the sample is nanocrystalline and the peak broadening in XRD also is in conformity with this. As seen from the image, there is uniform size distribution and morphology. This gives the possibility of controlling the size of the particles by varying parameters, such as reaction time. The band gap of the absorber material can be tuned by varying the particle size, which opens up a possibility of obtaining a wide range of absorption of the solar radiation.

6 Conclusion

In this study, the CZTS powder has been synthesized via a low-cost, environment-friendly reflux method. Phase identification was done using powder XRD (given by the reference pattern JCPDS card no. #00-026-0575) and Raman spectroscopy analysis (corresponding Raman shifts for CZTS: 282 and 338 cm^-1). The phase was found to be kesterite CZTS. We have determined the absorbance onset by plotting (αhν)² versus hν (α=absorbance, h=Planck constant, and ν=frequency). From the long wavelength extrapolation of the band edge, the band gap is determined to be 1.4 eV, which corresponds well with that reported in the literature [17]. The above-mentioned results from XRD, transmission electron microscopy, Raman spectroscopy, and UV-visible characterization techniques for CZTS indicate that these have good absorption characteristics for photovoltaic application. This observation also eliminates the existence of the secondary phase of ZnS, CuS, and Cu₂SnS₃. The good absorption in visible light region may find its interesting application in solar cells.

To further investigate the photoelectric property of CZTS nanoparticles, we fabricated CZTS film using the as-synthesized CZTS. FTO glass was cleaned by sonication in acetone, 2-propanol, and methanol, rinsed with deionized water, and finally dried in nitrogen stream. Then, 200 μL CZTS ink was doctor blade coated on FTO glass for uniform CZTS film and subsequently dried in vacuum at room temperature for 12 h. Work on detailed cell performance with such films is being carried out and will be reported elsewhere.

The convenience and low cost of reflex processing make it an attractive preparation method for chalcogenide solar cell materials. Hence, this makes them suitable as potential candidates for the development of solution processable, low-cost photovoltaic devices.