Synthesis of lead-free Cu/CuFeO₂/CZTS thin film as a novel photocatalytic hydrogen generator from wastewater and solar cell applications

The Cu/CuFeO₂ synthesis is performed using a combustion process. Firstly, the Cu foil (0.5 mm in thickness) was ultrasonically washed using soap and distilled water for 15 min, then it was soaked in concentrated H₂SO₄ for 15 min until complete polishing. Finally, it was cleaned with distilled water, ethanol, and acetone for five minutes each in an ultrasonic bath. The cleaned Cu foil was immersed in 0.2 M Fe(NO₃)₂ for 30 min, then heated at 60 °C for 30 min. The Cu foil is calcined in the air environment at 500 °C for 10 min to produce a Cu/CuFeO₂ micro-/nano-structured substrate.

CZTS was prepared through the solvothermal technique (Patil et al. 2016). Copper acetate anhydride, 0.22 g, zinc acetate dehydrate, 0.073 g, and tin(IV) acetate, 0.095 g, are mixed in 50 mL of anisole. The mixture was kept under stirring until a light blue color was obtained. Then, 1.2 mL of oleylamine was added and maintained under stirring till a dark blue color was observed. Carbon disulfide (0.6 mL) was added for 5 min until the dark yellow color was formed. The prepared solution is placed in an autoclave for 24 h at 170 °C. The black precipitate is separated by centrifugation once it has reached room temperature and washed sequentially with 20 mL of toluene and 40 mL of methanol.

The prepared CZTS (0.04 g) was dissolved in 2 mL of chloroform under ultrasonication, then 150 µL of CZTS was spin-coated over the prepared Cu/CuFeO₂ substrate and dried for 15 min at 70 °C. The sample is then annealed using the Muffle furnace (Chem. Tech) model for 25 min at 200 °C.

The surficial morphologies of the prepared samples were examined by scanning electron microscopy (SEM), Auriga Zeiss FIB, with a 5 kV accelerating voltage. A Bruker/Siemens D5000 diffractometer is applied for pattern recording of the X-ray diffraction (XRD). The chemical structures of the prepared samples were investigated using X-ray photoelectron spectroscopy (XPS, K-ALPHA, USA). The optical characteristics have been investigated through a Perkin Lamba 950 double-beam UV–vis spectrophotometer (USA).

The CHI660E electrochemical station (USA) is applied to investigate the photoelectrochemical (PEC) behavior of the Cu/CuFeO₂/CZTS. The electrodes with three parts made up the PEC system, in which the Cu/CuFeO₂/CZTS (1 cm²), Pt plate, and Ag/AgCl acted respectively as the working, counter, and reference electrodes. Wastewater (Egypt, Beni-Suef City, sewage water, third stage) was used as a testing electrolyte. The chemical composition of the sewage water is shown in Table 1. In the solar simulator, an Xe lamp (Newport) is applied throughout the measurements with a 100 mW cm⁻² power density for light illumination.

Cu/CuFeO₂, CZTS, and Cu/CuFeO₂/CZTS samples are characterized through the SEM analysis and presented in Fig. 1a–c, respectively. The highly crystalline polyhedral CuFeO₂ delafossite, Fig. 1a, and the kesterite CZTS, Fig. 1b, are free of cracks with a smooth surface; the dense uniform grains are expected to have a great advantage in enhancing the optical properties (Wang et al. 2012). Moreover, the compact and free-cracked morphology indicates the generation of electrons and photocatalytic features because the cracked surface favors the dispersion of electrons. This highly crystalline morphology matches well with other previous literature reports (Singh et al. 2012; Yu et al. 2014). The SEM morphology of the investigated photocatalyst (Cu/CuFeO₂/CZTS) was seen in Fig. 1c. The inset of Fig. 1c shows the cross-sectional SEM of the Cu/CuFeO₂/CZTS multilayers. The thickness of the Cu/CuFeO₂/CZTS films is around 400 nm that is estimated in Fig. 1d through the cross section simulated image. After spin coating of the CZTS, the formed Cu/CuFeO₂/CZTS shows a compact morphology related to the great ordering of the particles. Also, the surface of the investigated photocatalyst becomes homogeneous and less rough. In conclusion, there are no apparent pinholes at the surface of Cu/CuFeO₂/CZTS film, indicating that the investigated photocatalyst has great stability (Zhou et al. 2011).

The Cu/CuFeO₂ XRD pattern is displayed in Fig. 2a. The size of the crystal is estimated to be 24 nm. All Cu/CuFeO₂ signals confirmed the presence of the hexagonal 2H-CuFeO2 (JCPDS card No. 01-079-1546) and the rhombohedral 3R-CuFeO₂ (JCPDS card No. 00-039-0246) structural phases of CuFeO₂, based on the distinctive peaks. Although the 2H–CuFeO₂ phase is present, the 3R–CuFeO2 phase dominates the main XRD pattern (Gao et al. 2021). In Fig. 2a, the main characteristic peaks of CuFeO2 appeared at 36.02°, 39.12°, 58.3°, 61.78°, 67.70°, 74.96°, and 83.02°, which correspond to the (006), (012), (110), (1010), (0012), (202), and (024) lattice planes, respectively. This suggested the successful synthesis of the Cu/CuFeO₂ layers. The XRD of pure CZTS in Fig. 2b showed signals of CZTS tetragonal phase, and it is quite similar to the normal pattern of kesterite CZTS. It comprises major peaks at 32.20°, 28.29°, 56.08°, and 47.84° due to the (200), (112), (312), and (220) planes, respectively. The results are consistent with the previously disclosed kesterite CZTS findings (Nishi et al. 2013). This suggested the successful synthesis of CZTS film. The XRD of the investigated photocatalyst (Cu/CuFeO2/CZTS) is presented in Fig. 2c. The crystal of the investigated photocatalyst has a calculated size of 1.73 nm. The plot comprises diffraction peaks at 32.2°, 28.29°, 56.08°, and 47.84°. Of note, these peaks are nearly in the same locations as those of the CZTS kesterite structure; this implies that these peaks might be due to the (200), (112), (312), and (220) kesterite planes of CZTS. The other diffraction peaks at 36.02°, 39.12°, 58.3°, 61.78°, 67.7°, 74.96°, and 83.02° are assigned respectively to the (006), (012), (110), (1010), (0012), (202), and (024) lattice planes of CuFeO₂. This suggested the successful synthesis of the Cu/CuFeO₂/CZTS multilayers.

XPS is typically accomplished by exciting a sample surface with mono-energetic Al kα x-rays, causing photoelectrons to be emitted from the sample surface. An electron energy analyzer is used to measure the energy of the emitted photoelectrons. The surface types of Fe, Cu, O, and C atoms in Cu/CuFeO₂ are shown in the survey X-ray photoelectron spectrum in Fig. 3a, which displays the different elements and their corresponding binding energies (BE). The survey XPS plot comprises Fe 2p, C 1s, Cu 2p, and O 1s at 711.25, 285.0, 932.61, and 530.98 eV, respectively. The survey XPS spectrum of Cu/CuFeO₂ is discussed in detail from the high-resolution XPS plots in Fig. 4a–d. The Cu 2p XPS high-resolution plot is depicted in Fig. 4a, where the peaks around 952.0 and 932.0 eV arise from the Cu⁺ 2p(1/2) and the Cu⁺ 2p(3/2) orbital binding energies, respectively. The Cu²⁺ satellite characteristic peak at around 943.0 eV confirmed the presence of Cu²⁺ near the CuFeO₂ surface (Jain et al. 2019; Yamashita and Hayes 2008). In Fig. 4b, the Fe 2p XPS high-resolution spectrum could be curve-fitted with a peak around 711.12 eV due to the 2pFe(3/2) orbital binding energy, which is ascribed to the Fe³⁺ in CuFeO₂, indicating that the Fe³⁺ is the characteristic intense peak (Yamashita and Hayes 2008). Figure 4c showed the O 1s spectrum around 530.0 eV, which is typical for O²⁻ ions in the CuFeO₂ matrix. Figure 4d shows the C 1s spectrum, which is well approximated by the peak around 285.0 eV, which is assigned to adventitious carbon raised from the CO₂ adsorption (Sarabia et al. 2016). The above findings suggest that the Cu/CuFeO₂ thin film is produced through the combustion approach.

The survey XPS spectrum of the CZTS nanostructure is featured in Fig. 3b. The XPS plot comprises Sn 3d, Cu 2p, S 2p, and Zn 2p at 486.75, 932.7, 160.34, and 1022.06 eV, respectively. The survey XPS spectrum of CZTS has been discussed in detail from the high-resolution XPS plots in Fig. 4e–h. Figure 4e displays the spectrum of the Cu 2p, which can be deconvoluted into two main components due to Cu2p(1/2) and Cu2p(3/2) signals around 952.12 and 932.18 eV, respectively, giving an indication of the existence of Cu (I) (Zhou et al. 2011). Figure 4f showed that the Zn 2p peak comprises two characteristics, respectively, Zn2p(1/2) and Zn2p(3/2) signals around 1045.2 and 1022.1 eV, matching the 23.0 eV typical splitting, indicating the existence of Zn (II). Figure 4g shows the Sn 3d pattern, which exhibits two signals around 486.2 and 495.0 eV, evidence for Sn(IV), indicating the existence of Sn3d (5/2) and Sn3d (3/2), respectively. The XPS spectrum of the sulfur peaks is displayed in Fig. 4h. There are two expected signals from SO42-; the signal from sulfur with a zero charge S0 (2pS 3/2 and 2pS 1/2) at energies around 160.0–164.0 eV, and S⁶⁺ (2pS 3/2 and 2pS 1/2) at energies around 168.0–172.0 eV. The above findings suggest the formation of a CZTS thin film through the solvothermal method.

From the above results, we can conclude the binding energies of the investigated photocatalyst (Cu/CuFeO₂/CZTS) from the survey XPS in Fig. 3c. The peaks around 530.0, 285.0, 711.0, 932.0, 160.0, 486.0, and 1022.0 eV correspond to O 1s, C 1s, Fe 2p, Cu 2p, S 2p, Sn 3d, and Zn 2p, respectively. The above findings suggest the successful synthesis of Cu/CuFeO₂/CZTS photocatalyst.

Examining the optical properties of the Cu/CuFeO₂/CZTS sample is crucial for its potential application in photocatalytic hydrogen generation. Figure 5a displays the absorption curves of Cu/CuFeO₂/CZTS, CZTS, and Cu/CuFeO₂. These curves reveal a sharp absorbance peak at approximately 300 nm, aligning with absorption spectra observed in previous studies on CuFeO₂ and CZTS films. Understanding these optical features is essential for harnessing the photocatalytic properties of the Cu/CuFeO₂/CZTS sample in hydrogen generation applications. (Das, et al. 2018; Yengantiwar et al. 2018). Furthermore, all three prepared films exhibit the capability to absorb light across the near-infrared (NIR) and ultraviolet–visible (UV–vis) regions spanning from 400 to 800 nm. Notably, it is evident that the combination of CZTS with the Cu/CuFeO₂ sample leads to a noticeable enhancement in the absorption intensity of Cu/CuFeO₂. This enhancement can be attributed to the fact that the investigated photocatalyst, specifically Cu/CuFeO₂/CZTS, possesses a narrower bandgap in comparison to the CuFeO₂ layer.

In simpler terms, all three films exhibit the ability to absorb light in both the near-infrared and UV–visible regions, which is vital for their potential as photocatalysts. The amalgamation of CZTS with the Cu/CuFeO₂ sample results in a significant boost in the absorption capacity of Cu/CuFeO₂. This increase can be explained by the fact that the Cu/CuFeO₂/CZTS photocatalyst has a narrower energy bandgap when compared to the pure CuFeO₂ layer. This narrower bandgap enables the material to capture a broader range of light wavelengths, making it a more efficient candidate for photocatalytic applications (Ju et al. 2018). Moreover, the function of Cu plasmonic materials is recognized as highly effective in capturing photons. This effectiveness stems from the generation of hot electrons, which induce a significant resonance within the surrounding CuFeO₂ and CZTS semiconductor materials.

From Tauc’s equation (Shaban et al. 2018), we can calculate the energy gap, Eg, based on the direct allowed transitions:

In this equation, the parameter v is the light frequency, α is the absorption coefficient, h is the Planck constant, and β is a constant. The relation between α and the absorbance, A, is given in Eq. (2):where d is the thickness. Extrapolation of the hν versus (αhv)² linear part allows for an estimate of the band gap value through the X-axis intercept, as seen in Fig. 5b. The band gap values of Cu/CuFeO₂/CZTS, CZTS, and Cu/CuFeO₂ were determined as 1.15, 1.43, and 1.97 eV, respectively, which gives an enhancement in the optical absorption of the investigated photocatalyst (Cu/CuFeO₂/CZTS). Despite the low band gap of CZTS (1.43 eV), it showed high photocatalytic activity. This can be recently interpreted by S. Dursunin and co-authors, who have investigated the mechanism that performs photocatalysis depending on scavenger experiments, and it was shown that holes had little impact on the process, whereas O⁻² radical ions have a significant influence in the reaction (Dursun et al. 2023). Figure 5a, which is consistent with previously reported absorption spectra of CuFeO₂ (Yengantiwar et al. 2018; Read et al. 2012; Prevot et al. 2015a) and CZTS films (Das, et al. 2018). In that approach, the Cu/CuFeO₂/CZTS photocatalyst shifted to a longer wavelength (1.15 eV) compared to that of CZTS and CuFeO₂ at 1.43 and 1.97 eV, respectively. Therefore, the Cu/CuFeO₂/CZTS photocatalyst enhanced visible light harvesting and hence showed significant photocatalytic activities when illuminated by visible light. So, the investigated Cu/CuFeO₂/CZTS photocatalyst can be applied in H₂ generation systems.

The PEC study of the Cu/CuFeO₂/CZTS photoelectrode is recorded at 25 °C and a 20 mV s⁻¹ sweep rate using wastewater solution. The PEC results under dark and light conditions using a Xenon lamp (100 mW cm⁻²) in the absence of an optical filter are seen in Fig. 6a, b for Cu/CuFeO₂ and Cu/CuFeO₂/CZTS, respectively. In Fig. 6a, the photocurrent density (J_ph) values of the Cu/CuFeO₂ photoelectrode are − 0.36 and − 0.89 mA cm⁻² under dark and light, respectively. After CZTS coating and the formation of the Cu/CuFeO₂/CZTS photoelectrode composite, the recorded J_ph values for Cu/CuFeO₂/CZTS (Fig. 6b) under light and dark conditions are respectively − 8.0 and − 0.7 mA cm⁻². The efficient PEC water splitting is characterized by the high J_ph value achieved through the light illumination. The significant current under dark conditions arises from the transferred charge through the ionic currents of the solution. Moreover, the increase in voltage applied under the illumination of a 100 mW ⁻² Xenon lamp gives an increased J_ph value due to the photocatalytic nature of the prepared composite nanomaterial, in which a splitting in the external energy levels under light is expected. The incident photons are captured inside the material and cause the excitation of the external electrons that are collected on the surface of the composite material, from which these electrons reach the wastewater for additional reactions for the generation of hydrogen gas fuel.

Figure 6c depicts the stability of the Cu/CuFeO₂/CZTS photoelectrode. To avoid any external losses to the measuring system, a very low bias voltage of − 0.8 V (RHE) was supplied between the counter and working electrodes under simulated white light of 100 mW cm⁻². The first 500 s showed a sharp decrease in the J_ph value from − 2.25 to − 0.31 mA cm⁻², and then the J_ph value was almost constant up to 3200 s due to an increase in the accumulated ionic charges. The H₂ generation rate for Cu/CuFeO₂/CZTS photoelectrode is 0.049 mA/h. The reason for the decrease in J_ph value in the first 500 s is related to the reaction between the electrode and the electrolyte, which causes a very small corrosion process. Also, the increase in the rate of ionic charge accumulation led to a longer lifetime of the Cu/CuFeO₂/CZTS with time. This result confirms that Cu/CuFeO₂/CZTS has a longer lifetime and higher chemical stability for H₂ production from wastewater. The J_ph-V measurements are repeated many times for Cu/CuFeO₂/CZTS multilayers under a white light of 100 mW cm⁻², as seen in Fig. 6d. The value of J_ph decreases from − 8.0 to − 7.69 mA cm⁻² after four successive repeated measurements. The small change in J_ph indicates high reproducibility.

Under light illumination, Fig. 6b demonstrates the generation of photocurrent at V = 0. This indicates that the Cu/CuFeO₂/CZTS multilayer has applications in solar cells, where this value of photocurrent can be indicated as short circuit current (J_SC).

To measure the IPCE percent value, the study of the illuminating power density effect, P_light, on the prepared Cu/CuFeO₂/CZTS photoelectrode is studied. Figure 7a exhibits the J_ph variation with applied potential in the presence of P_light from 25.0 to 100 mW cm⁻². As P_light increased from 25 to 100 mW cm⁻², the J_ph results increased, as seen in Fig. 7b, where the highest J_ph of − 8.0 mA cm⁻² was recorded at 10² mW cm⁻².

The relation between the J_ph and the applied potential was studied at 100 mW cm⁻² for the Cu/CuFeO₂/CZTS in the presence of different monochromatic filters from 390 to 636 nm, as seen in Fig. 8a. From Fig. 8b, with a 390–636 nm increase in optical wavelength, the J_ph values decreased. The IPCE percent value was measured for Cu/CuFeO₂/CZTS under monochromatic illumination (Fig. 8c) to confirm its enhanced PEC features for efficient H₂ production from wastewater. Such measurements provide useful information on the film's contribution to the conversion of incident photons into charge carriers. Equation 3 is used to calculate the IPCE at a 1 V applied potential.where the wavelength of the monochromatic photons is λ, J_ph represents the photocurrent density, and P_light denotes the illuminating light power density (100 mW cm⁻²) (Shi et al. 2016). As seen in Fig. 8c, the IPCE recorded a remarkable decrease from 25.44% at 390 nm to 11.09% at 636 nm. These high IPCE values in a broad light region confirm the high-efficiency properties of the prepared Cu/CuFeO₂/CZTS photocatalytic electrode to use the incidence light for the generation of hydrogen gas.

Figure 9a displays the effect of temperature on the H₂ production of the prepared photocatalyst (Cu/CuFeO₂/CZTS). It could be observed that with increasing temperature, the J_ph values increased due to enhanced mobility of the ions. Therefore, the J_ph can be used as an indication of the rate of H₂ production at various temperatures.

To calculate the reaction activation energy (E_a), the Arrhenius law is applied as follows:

In this equation, T denotes the temperature in absolute terms, R corresponds to the standard gas constant, and k denotes the rate constant. Figure 9b shows the relationship between the reciprocal temperature and the values of the ln (J_ph) (rate of reaction). The E_a of the Cu/CuFeO₂/CZTS film was determined to be 21.0 kJ mol⁻¹, indicating that the film has an efficient application for H₂ production. The entropy ΔS^* and the enthalpy ΔH^* were calculated using the Eyring Eq. (6):

Figure 9c presents the relationship between (1/T) and Ln (J_ph/T). From the intercept and the slope results of the straight line, the ΔS^* and ΔH^* were respectively estimated to be 28.76 kJ mol⁻¹ K⁻¹ and 18.28 kJ mol⁻¹.

Figure 10a depicts the Nyquist impedance plot of the Cu/CuFeO₂/CZTS under white light illumination. It was observed that the impedance spectra under light conditions had two semicircles, indicating that there are two-time constants that controlled the reaction. For more accurate impedance analysis, an equivalent circuit model is suggested in Fig. 10b. It includes a double-layer capacitor, C_dl, in parallel with R_s (solution resistance) and R_ct (charge transfer resistance). An extra capacitor, C_f, and a resistance, R_f, are added to account for the capacitance and resistance of the film. The R_s value under light conditions was calculated to be 0.36 Ω cm². The small R_s value under light conditions was deduced from the small loop diameter, indicating the high current density, and this is a pre-indicator for the application of Cu/CuFeO₂/CZTS in the H₂ production from wastewater.

The impact of incident light on the photoelectrons serves as the foundation for the photocatalytic mechanism for the sewage water-splitting process. The photoelectrons produced as a result of the light incident go through two stages. In the first, called electron–hole generation, the created electrons leave the holes and ascend to a higher level. The localized surface plasmonic resonance (LSPR) is the second step, which results in energy transmission (Cortes et al. 2020). Because the Cu and CuFeO₂ nanomaterials don't have any depletion zones, and due to the small Cu/CuFeO2 band gap value, these two steps occurred easily. As a result, there is no restriction on the transfer of electrons from Cu to CuFeO₂, and there is no constraint on the continuous flow of electrons. Moreover, the Cu material can act as a plasmonic material for light capture and the generation of an electric field that covers all the neighboring photocatalytic materials. The final electron cloud is collected over CZTS under the lower conducting band level, through which the electrons are transferred to the sewage water-electrolyte. Under the applied voltage, these hot electrons cause the generation of Jph (Hussain et al. 2016; Wang et al. 2014). The optical analyses (Fig. 5) and electrochemical measurements feature the experimental image of electron transfer processes under several effects, such as light wavelength and light intensity. The light capture and electron resonance processes represent the rate of hydrogen generation reactions (Cao et al. 2021; Qin et al. 2016). In Fig. 11, the energy diagram of the photocatalysis system is described for the generation of H2. Due to the low barrier height between the CuFeO₂ Fermi level and the CZTS conduction band, the heated electrons can move from CuFeO₂ to the CZTS after crossing the barrier (Malerba et al. 2014; Khammar et al. 2020). This prevents the carrier's recombination and permits a constant flow of electrons into the system, leading to the generation of photocurrent. This photocurrent represents the efficiency of these materials for solar cell applications and hydrogen generation. The J_ph values and the H₂ production reaction rate are used to represent these electrons (S. S et al. 2016). Finally, Table 2 compares the results of the current investigation to the previously published literature.