Synthesis of 2D-CZTS nanoplate as photocathode material for efficient PEC water splitting

Fabrication of economically feasible photocathode for hydrogen energy production through solar water splitting is a major research among the scientific community for a decade. P-type compound Cu2ZnSnS4 (CZTS) is very interesting material due to its absorption property, earth-abundant constituents and environmental friendliness that serves as a suitable candidate to act as a photocathode. In the present work, Cu2ZnSnS4 (CZTS) nanoparticles are synthesized by simple one-step chemical method and annealed at 350 °C for three different times (60 min, 90 min, and 120 min). The effect of annealing time on the structural, optical and photoelectrochemical properties are investigated. XRD pattern indicates the formation of tetragonal crystal structure and the crystallinity increases according to the annealing time. 2D nanoplate morphology is obtained for the sample that was annealed for 120 min. From the absorption spectra, it was found that the bandgap decreases with increase of annealing time. Further, the prepared nanoparticle thin films are used as a cathode for photoelectrochemical water splitting application. Among these, the nanoparticles that are annealed for 120 min showed higher photocurrent density when compared to nanoparticles annealed for 60 min and 90 min.


Introduction
Renewable energy resources procure substantial involvement in the area of alternative source of energy [1] and is seen as a best solution for addressing the current and future energy demands of our human society. Therefore working and exploring in the field of new materials is necessary to achieve the goal of cost-effective one, for example converting solar energy into electrical energy via photoelectrochemical reactions, photovoltaic and photocatalytic technology [2]. The most promising and emerging technology for converting solar energy into electrical energy is photocatalytic and electrocatalytic water splitting. Natural photosynthesis is inefficient because it captures less number of photons, so charge separation and charge transfer are limited. But artificial photosynthesis is a novel technology, which converts light energy (solar energy) into chemical energy using semiconductors as photocatalysts. Although, it is an alternative for the natural process as it captures more photons and increases the charge transportation. The main goal of artificial photosynthesis is to capture more sunlight and split the water molecules to produce renewable energy (hydrogen fuel). Hydrogen (H 2 ) is a clean fuel and it emits water upon utilization and it does not emit greenhouse gases, free from air pollution [3].
With the existing production mechanisms, cost and the upgrowing technology, utilization of hydrogen fuel has facing many practical difficulties. To overcome such issues, wide technical knowledge is required in theoretical and experimental research.
Recently, many researchers have been devoid of green and sustainable technology toward the clean environment for the production of hydrogen fuel through photocatalytic (PC) and photoelectrochemical (PEC) water splitting. The major hurdle in these methods is the evolution of a suitable semiconductorbased photocatalyst with good light absorption properties in the visible region [4]. Photoelectrochemical (PEC) water splitting is a complex process because efficient device design and several phenomena must be optimized. The most fundamental ones are the interaction of light with matter, generation of electron-hole pairs, separation of charge and its transport, transfer of charge from the catalyst to electrolyte and the water splitting reaction. A current flow is obtained between both the electrodes, when the water splitting occurs [5]. For PEC water splitting, numerous photoanodes (n-type semiconductors) have been investigated for water oxidation. However, relatively few work has been carried out on the photocathodes (p-type semiconductors) for water reduction [6,7]. The recent research in (PEC) water splitting is focused on p-type semiconductor including oxides and sulfides, which has received more attention than that of p-type silicon and III-V semiconductor photocathodes. To have a large photocurrent density, the conduction band edge potential of photocathode material needs to be more negative than the hydrogen redox potential in PEC.
Materials with direct bandgap energies in the range of 1-2 eV acquire maximum light absorption from the solar spectrum [8] because direct bandgap materials absorb light quickly before it goes deeper into the absorbing material compared to indirect bandgap materials. Copper Zinc Tin Sulfide (CZTS) is a good substitute for present photovoltaic and photocatalytic materials [9][10][11]. CZTS is better than Copper Indium Selenium (CIS) and copper indium gallium selenide (CIGS) since it is composed of scarce, expensive and relatively toxic materials. CZTS is a p-type quaternary semiconductor compound with the kesterite crystal structure. The components in CZTS are rich in the earth's crust and it is harmless, economic and non-hazardous [12][13][14]. It is a potential material for the photovoltaic absorber layer with direct bandgap energy of 1.4-1.6 eV and absorption coefficient * 10 4 cm -1 in the visible region [15]. The first report of CZTS-based photocathode for hydrogen production was reported by Yokoyama et al., [16]. Zhang et al. synthesized CZTS photocathodes via electrodeposition of metal precursors followed by sulfurization [17]. The recent approaches for the development and optimization of the photocathodes for PEC water spitting motivates to investigate this favorable photocathode material for PEC energy conversion.
The focus of present work was the preparation of CZTS nanoparticles by simple one-step chemical method using non-toxic solvent (water), because most organic and inorganic mixtures can favorably dissolved in water. CZTS nanoparticles are synthesized at an annealing temperature of 350°C with different annealing times. To optimize the annealing time for a particular annealing temperature, structural, morphological and optical properties are investigated. The PEC performance of CZTS nanoparticle thin films are studied by fabricating CZTS photocathode and its appropriateness for water splitting applications are measured.

Preparation of CZTS nanoparticles
CZTS nanoparticles have been synthesized through a simple one-step chemical method. 1 M of CuCl 2-2H 2 O, 0.5 M of ZnCl 2 , 0.5 M of SnCl 2 Á2H 2 O and 2 M of H 2 NCSNH 2 are dissolved in 40 ml of double distilled water for the preparation of initial solution. To dissolve the components completely, the solution was stirred on a magnetic stirrer for 1 h at 60°C. The precipitates are washed several times using ethanol and distilled water to bring out the traces of pollutants. The resultant solution was centrifuged and the final product in the form of paste was extracted. Further, to remove moisture, the precipitate was dried at 80°C in a hot air oven for 1 h and then annealed at 350°C for 60 min. Finally, black participates are collected and used for further investigations. The same investigation was repeated with same temperature but with different annealing time (90 min and 120 min) to study the effect of annealing time in the properties of CZTS nanoparticles. The samples annealed at 60 min, 90 min, and 120 min are named as CZTS/60, CZTS/90 and CZTS/120 respectively.
Formation of CZTS nanoparticle has taken place according to the given reaction mechanism.
The final ionic reaction is The stoichiometric reaction for the formation of Cu 2 ZnSnS 4 .

Characterization techniques
Crystallinity and the orientation of Cu 2 ZnSnS 4 nanoparticles are assessed by X-Ray Diffraction (XRD) technique using an XPERT-PRO X-ray diffractometer with CuKa radiation of k = 1.5406 Å . Horiba Jobin Yvon LABRAM -HR 800 spectrometer was used to record Raman spectrum. Surface morphology and compositional analysis of the CZTS nanoparticles are examined by scanning electron microscope(JEOL mode JSM 6390 SEM) equipped with the energy dispersive X-ray spectrometer (EDX). Shimadzu FTIR with ATR Spectrometer was employed to measure the IR spectrum of the sample over the range of 400-4000 cm -1 . The absorption spectra of the sample are measured using the JASCO Corp., V-570 spectrophotometer.

Preparation of photocathode
The CZTS photocathode was fabricated by the following procedure: the substrate (FTO glass plate) was washed with a mixed solution in the volume ratio 1:1:1 of deionized water, acetone, and ethanol for 10 min using ultrasonication. CZTS paste was prepared by dissolving the CZTS powder (2 g) in ethylene glycol at room temperature. The mixture was then sonicated for 1 h for homogeneous dispersal. CZTS nanoparticle paste was deposited on FTO glass plate using facile, cost-effective spin coating technique at 1500 rpm for 30 s. After that, the thin film was dried at 100°C in a hot air oven for a half an hour with a four cycle of deposition and drying. Finally, the CZTS nanoparticle thin films are annealed at 350°C for 1 h using a muffle furnace.

Photoelectrochemical water splitting experiment
E°( Ag/AgCl) = 0.1976 V at 25°C is the correction factor for reference electrode and E (Ag/AgCl) is the measured potential against the Ag/AgCl reference.

Results and discussion
3.1 X-ray diffraction (XRD) studies X-ray Diffraction is used as the major tool for the identification of the phase purity and crystallinity of CZTS nanoparticles. The Crystalline or Grain size of the prepared CZTS samples was confirmed through the powder XRD technique using CuKa radiation. Figure 1a shows the XRD pattern of CZTS nanoparticles annealed at 350°C with different annealing times (60 min, 90 min and 120 min  [20] and ternary phases such as Cu 2 SnS 3 and Cu 3 SnS 4 [21] are not present in the XRD pattern indicating the formation of Cu 2-ZnSnS 4 only. The intense and sharp peaks of the samples show the good crystallinity of nanoparticles. The crystalline size of CZTS nanoparticles was estimated using the Scherrer's formula [22], where k is the shape factor (k = 0.94), k is the wavelength of CuKa1 radiation source (k = 1.5406 Å ), b is the peak width (Full Width Half Maximum) and h is the Bragg angle.
The number of defects in the crystal was determined using dislocation density (d) and can be calculated using the formula [23], Microstrain of the CZTS nanoparticles can be estimated using the relation [23], where b is the peak width (Full Width Half Maximum) and h is the Bragg angle. The lattice parameters of the tetragonal crystal system are calculated using the relation [23], Fig. 1 a XRD patterns and b Raman spectra of CZTS nanoparticles annealed at 350°C with different annealing times where d is the inter planar spacing, h, k, and l are the Miller indices and a, c are lattice constants. Cell volume can be calculated by the formula, The calculated crystallite size (D), dislocation density (d), strain (e), lattice parameters (a),(c) and cell volume (v) are displayed in Table 1.
The average crystallite size of CZTS nanoparticles was found to be 23-30 nm. Crystallite size is increased with the increase of annealing time as 23.36 nm, 27.78 nm and 30.44 nm for CZTS/60, CZTS/90 and CZTS/120 respectively. The increase in crystallite size has an advantage for the photovoltaic and photocatalytic applications, as the photogenerated electron-hole recombination rate would be reduced with large agglomerated grains. Hassanien et al. [24] reported the increase in crystallite size value from 4.5 to 38.8 nm for CZTS with an annealing temperature of 400-550°C. Liping Chen et al. [25] observed the increase in crystallite size from 5.60 to 19.97 nm in CZTS when annealed from 250 to 550°C. Giedr_ e Grincien_ e et al. [26] reported the size of crystallites increases from 2.4 to 24.1 nm for CZTS with the increase of annealing temperature from 300 to 550°C. The decrease in microstrain and dislocation density with an increase in crystallite size implies a less number of lattice defects, i.e., good crystallinity. The lattice parameters calculated are in good concurrence with the report of Kishore et al. [27] and Mali et al. [28] and also matched with standard JCPDS data (26-0575).

Raman analysis
Raman spectrum is a very sensitive tool for phase identification often combined with XRD results. The phase purity of the CZTS nanoparticles was determined using the Raman spectrum, i.e., the existence of other phases, such as binary and ternary phases in CZTS nanoparticles will be investigated in detail using Raman spectra. The Raman spectrum of the CZTS nanoparticles for different annealing times is shown in Fig. 1b. The main peak at 339 cm -1 along with small peak at 289 cm -1 was observed for the samples annealed at 60 min and 90 min. The sample annealed at 90 min was observed the main peak at 339 cm -1 along with shoulder peaks at 289 cm -1 , 362 cm -1 . The peaks are matched with the main Raman peaks of CZTS nanoparticles, which are close to the earlier reported values of CZTS of Kannan et al. [29], Mkawi et al. [30], and Jing Wang et al. [31]. The major peaks are recognized as the main vibrational A1 symmetry modes from the kesterite CZTS nanocrystals [32]. The A1 phonon mode corresponds to the vibration of S atoms present in the sample and it is a pure anion mode [33]. There are no significant extra peaks related to the presence of secondary and ternary phases. Moreover, the intense major peak indicates the good crystalline quality of the sample. The results observed from Raman spectra are agreed with the XRD results.

Morphological analysis
Scanning electron microscope was used to examine the surface morphology of CZTS nanoparticles. Figure 2 shows the SEM image of CZTS nanoparticles prepared with different annealing times. The sample annealed at 60 min is agglomerated which is due to adhesion of particles to each other by weak forces leading to entities or structures of micrometer range. The sample annealed at 90 min composed of different types of shapes like plates, polygons, etc. 2D Nanoplate structure [34] is obtained for the sample annealed at 120 min. An increase in annealing time induces a change in morphology from agglomerated form to nanocrystallites and then to nanoplates which arises as a result of covalent bond breaking between the molecules that are not observed in the earlier stage of annealing (60 min).

Compositional analysis
The elemental composition of CZTS was analyzed using the EDAX technique. This is used to confirm the existence of Copper, Zinc, Tin, and Sulfur.

FTIR studies
The impact of annealing time on the vibrational modes of Copper, Zinc, Tin, and Sulfur was investigated using an FTIR spectrophotometer. FTIR spectra of CZTS nanoparticles annealed at different times are shown in Fig. 3. The stretching and bending of oxygen was observed around 900-1600 cm -1 . The peak at 626 and 664 cm -1 is assigned to the ZnS band. The presence of sulfide ions in vibrational modes are observed as broad bands around 858-893 cm -1 [35]. The peak observed around 1050-1100 cm -1 and 1400-1460 cm -1 is due to NH 2 vibrational mode of thiourea. The peaks at 2980, 2928, and 2980 cm -1 are ascribed to the S-H thiol functional group and N-C-N stretching respectively [36]. The stretching of S-H bond is revealed by the peaks at 2350 and 2990 cm -1 respectively [37]. Presences of water, as well as thiourea are observed in the range of 3550-3750 cm -1 [35]. As a common observation, the shifts indicate a rearrangement of the CZTS network toward that of the stoichiometric structure with a lower structural disorder which gives rise to 2D nanoplate morphologies witnessed in SEM analysis.

UV analysis
To study the optical absorption spectra and bandgap energy of the CZTS nanoparticles, absorption spectra are studied using UV-visible spectra in the wavelength range of 300-800 nm. The optical absorbance of CZTS nanoparticles annealed at 350°C with annealing time of 60 min, 90 min, and 120 min are shown in Fig. 4a. It is clearly noticeable from the Fig. 4a, all samples showed absorption in the visible region. The spectra revealed that CZTS nanoparticles annealed with 120 min indicating applicability as an absorbing material compared to 60 min and 90 min. Therefore, CZTS nanoparticles annealed at 350°C with an annealing time of 120 min is considered to be a suitable material for solar energy conversion.
Tauc and Davis Mott model was used to determine the optical bandgap energy E g as a function of photon energy hm, using the formula [38], where a-Absorption coefficient, hm-Incident photon energy, A-Constant, E g -Bandgap energy, and n is an index characterizing the nature of optical transition, i.e., for direct allowed transition n takes a value of 1/2 and for an indirect allowed transition n takes a value of 2. Figure 4b shows the Tauc plot of CZTS nanoparticles annealed at 350°C with an annealing time of 60 min, 90 min and 120 min. The bandgaps (E g ) are estimated to be 1.54 eV, 1.50 eV, and 1.48 eV for 60 min, 90 min and 120 min respectively. The decrease in bandgap with increase in annealing time demonstrates less crystal defects with good crystalline quality. With the increasing of annealing time from 60 to 120 min, the decrease in bandgap indicates that the grain size increase may increase the absorbance nature of materials. The obtained bandgap value is in good agreement with the earlier reports [39][40][41]. The bandgap observed was near the optimum value for photocatalytic and photovoltaic applications.

Photoelectrochemical properties
The photoelectrochemical response of CZTS thin films are measured to find their suitability for artificial photosynthesis. Photocurrent density generated by photocathode was measured to study its photoactivity, while the photocathode is dipped within the electrolyte solution. Figure 5a represents the LSV of CZTS photocathodes annealed at 350°C with an annealing time of 60 min, 90 min and 120 min. The current density under illumination represents the solar energy converted into electrical energy which is stored as chemical energy in the electrochemical cell. The direction of photocurrent is expected behavior for p-type semiconductor, where electron injection travel toward the electrolyte, while holes travel toward the counter electrode via an external circuit. The charge accumulation on the semiconductor is due to the rapid increase in cathodic current, which gives rise to an exponential capacity [42]. It can be noticed from LSV curve, all the photocathodes exhibit improved photocurrent density upon illumination. The enhancement of photocurrent exhibits the light-sensitive nature of CZTS material. The photocurrent density value increases very small with the increase in annealing time from 60 to 120 min.
Chronoamperometry measurements are carried out at 1sun (100 mW cm -2 ) illuminations under chopped light illumination conditions in 0.5 M Na 2-SO 4 (pH 8) aqueous solution for CZTS/60, CZTS/90, and CZTS/120 photocathodes. Figure 5b represents the i-t curves of CZTS photocathodes annealed at 350°C with an annealing time of 60 min, 90 min and 120 min. The results from the i-t curves showed a stable photoresponse in the aqueous Na 2 SO 4 solution for all the electrodes. As shown in Fig. 5b, CZTS/120 photocathode represents higher photocurrent density (41.8 lA/cm 2 ) than CZTS/90 (32.8 lA/cm 2 ) and CZTS/60 (15.5 lA/cm 2 ) was obtained at 0 V (vs RHE). The photocurrent density of prepared photocathodes was increased with increase of annealing time. The obtained morphology (2D nanoplate) for CZTS/120 gives good interfacial contact between the material and electrolyte thus offering improved photocurrent density. The photoactivity of the CZTS thin film was confirmed from photoelectrochemical measurement, which was already confirmed in UVvisible spectra.

Electrochemical impedance spectroscopy analysis
Electrochemical impedance spectroscopy (EIS) is used for a better understanding of electrochemical activity. EIS spectra of CZTS photocathodes are shown in Fig. 5c. From Fig. 5c, the semicircle of all the CZTS electrodes annealed at 350°C with different annealing times becomes narrower strategy. The lower semicircle indicates a higher conductivity and increase of charge transfer resistance due to the increase of annealing time, which enabled electrodes activation. For CZTS/120 the results shows better PEC performance.

Conclusion
Cu 2 ZnSnS 4 (CZTS) nanoparticles are successfully synthesized by simple one-step chemical method with post-annealing in air at different annealing time and with an annealing temperature of 350°C. The results of XRD and Raman spectrum confirm the formation of tetragonal kesterite structure of CZTS nanoparticles with good crystallinity and absence of secondary phases. Moreover, the grain size of the CZTS nanoparticles had a substantial growth when the annealing time was increased from 60 to 120 min. The 2D nanoplate morphology is obtained for the sample annealed at 350°C with an annealing time of 120 min. The chemical composition of the CZTS nanoparticles is near the stoichiometric ratio. FTIR spectra confirm the presence of functional groups present in the CZTS samples. Optical measurements show that the direct bandgap of the CZTS nanoparticles at different annealing time decreased from 1.54 to 1.48 eV. Further, the nanoparticle thin films were prepared to study the water splitting performance. The CZTS/120 photocathode exhibited higher photocurrent density when compared with other two photocathodes, because of larger grain size, the recombination of electron-hole pairs at the grain boundaries was reduced. The enhancement in the cathodic photocurrent density of CZTS/120 during the water splitting process was due to the lower bandgap and higher electrical conductivity of the sample.