The Roles of the CuO Buffer Layer on the Photocatalytic Activity of the p-Si/p-CuO/n-ZnO Composite Films

Abstract

The photocatalytic activity of p-Si/p-CuO buffer layer/n-ZnO nanorod (NR) composite films was studied using experimental and simulation methods. The simulation results indicated that in the p-Si/p-CuO/n-ZnO composite film, the 250 nm CuO buffer layer contributes the highest value of 11% to the total current density compared to the p-Si/n-ZnO composite film. Besides that, the experimental results also indicated that by introducing the p-CuO buffer layer, the pseudo-order rate constant (k) could be enhanced up to 12% compared to the composite film without the p-CuO buffer layer - the p-Si/n-ZnO composite film. Furthermore, the recycle result indicated that the pseudo-order rate constant - k value decreased sharply after the first three reaction cycle times and gradually stabilized at a value of 0.88 s⁻¹ after the fourth reaction cycle. Therefore, it can be concluded that by introducing the 250 nm thick of p-CuO buffer layer in the p-Si/p-CuO/n-ZnO NRs composite film, the photocatalytic activity could be improved up to 12% compared to that without the p-CuO buffer layer. In addition, the composite film, p-Si/p-CuO buffer layer/n-Zno, is a reusable photocatalyst with high photostability.

Fig. 1 The constructed p-Si/p-CuO buffer layer/n-ZnO NRs composite film.

1. Introduction

In recent years, the advanced oxidation method has been an up-and-coming method for the treatment of wastewater.^1–3) Semiconductor oxides are used as photocatalysts, and under the illumination of suitable radiation, electron-hole pairs are created through the photoelectric effect.^4,5) These electron-hole pairs will participate in a series of redox reactions to generate reactive radicals such as (^*O²⁻, HO^2*) to degrade pollutants.^6,7) Among several semiconductors, ZnO material is considered as a potential photocatalyst because of its non-toxicity, low cost, and high redox potential.^{8–10,31–33)} However, ZnO material is a wide and direct bandgap semiconductor leading to ZnO being photo-catalytically active under ultraviolet (UV) radiation and short carrier lifetime material. Therefore, the activity of the ZnO photocatalyst could be improved by broadening the optical absorption region and increasing the carrier lifetime.^11,12) Recently, ZnO has been studied as a heterojunction material based on combining with a p-type narrow bandgap semiconductor such as Si, CuO, Cu₂O, NiO, and CdS.^12–16) The built-in electric field at the heterojunction is considered as a promising solution to solve the problem of electron-hole recombination rate. Furthermore, by combining with narrow bandgap semiconductors, the optical absorption band of the heterojunctions could be expanded to the near-infrared region. Although the bandgap of Si is much narrower than that of CuO (1.1 eV vs. 1.5 eV), some studies indicate that ZnO/CuO heterojunctions have much higher photo-generated current compared to ZnO/Si heterojunctions.⁶⁾ Based on this method, the solar energy conversion efficiency of the n-ZnO NWs/p-CuO/p-Si and n-ZnO NWs/p-Si heterojunctions were obtained to be 3.92% and 0.008%, respectively, under 100 mW/cm² input incident power. Thus, the n-ZnO NWs/p-CuO/p-Si heterojunctions exhibit superior photovoltaic performance in comparison to the n-ZnO NWs/p-Si system.^17,18,29,30)

In this work, the composite film of p-Si/p-CuO (buffer layer)/n-ZnO nanorods (NRs) was fabricated by sputtering, thermal annealing, spin coating, and simple hydrothermal methods. The p-CuO buffer layer in this composite film is introduced to enhance the photo-generated current between the CuO/ZnO junction. In contrast, the p-Si layer plays the role of extending the adsorption band towards the near-infrared region. For comparison, p-Si/p-CuO, p-Si/n-ZnO, and p-Si/p-CuO/n-ZnO heterojunctions were fabricated using the same methods. The photocatalytic activities of all fabricated samples were investigated. The results demonstrated that the p-Si/p-CuO/n-ZnO NRs heterojunction showed the highest photocatalytic activities compared to the other samples. The highest photocatalytic activities of the p-Si/p-CuO/n-ZnO NRs heterojunction would be ascribed to the extension of the optical absorption range and the efficient separation of photo-generated electron-hole pairs. Based on the simulation results, characteristics of the p-Si/p-CuO/n-ZnO heterojunction were also investigated to find the optimal composite film. With the optimum thickness of 250 nm, the CuO buffer layer contributed 11% to the total photocurrent density of the p-Si/p-CuO/n-ZnO composite film. This result was also experimentally retested, and the results showed that the pseudo-order rate constant value of the p-Si/p-CuO/n-ZnO composite film improved by 12% compared to that of the p-Si/n-ZnO composite film.

2. Experimental Details

2.1 The simulation program SCAPS

The simulation program SCAPS (Solar Cell Simulation Program in One Dimension) was utilized in this work to investigate the p-CuO buffer layer. This program has normally been designed to simulate thin film solar cells such as CIGS, CdTe, etc.^19–21) However, the photocatalytic activity strongly depends on the number of photo-generated electron-hole pairs which can reach the counterparts and do the degradation work. These active electron-hole pairs also relate to the photon-induced current density. Therefore, this simulation program can be utilized to find the optoelectrical characteristics of the heterojunctions. The simulation results were also compared with experimental results to find the optimal parameters of the p-CuO buffer layer.

In this work, SCAPS was used to simulate the Current density – Voltage (J–V) and Quantum Efficiency (QE) characteristics of the p-Si/p-CuO buffer layer/n-ZnO NRs. Based on these characteristics, the most critical parameters, such as short-circuit current density (Jsc) and/or the number of photo-generated carriers by each layer under standard illumination (AM 1.5 G, 100 mW cm⁻², 300 K) were calculated and analyzed. The characteristics of the materials used in the simulations were acquired from the literature.^22–28) The parameters for Si, CuO, and ZnO used in the simulations were provided in Table 1. Figure 1 shows the p-Si/p-CuO buffer layer/n-ZnO composite film used for the SCAPS simulations. The influences of the CuO buffer layer’s thickness were analyzed to determine the optimum composite film for photocatalytic application.

Table 1 Parameters of ZnO, CuO, and Si.

Fig. 1

The constructed p-Si/p-CuO buffer layer/n-ZnO NRs composite film.

2.2 Materials

p-Si wafer, Zn(NO₃)₂·6H₂O, C₆H₁₂N₄, NaOH, and methanol were all purchased from Sigma-Aldrich. All reagents were analytic reagent grade and utilized without further purification.

2.3 Preparation of the p-Si/p-CuO/n-ZnO-NRs composite films

The p-CuO buffer layer was synthesized on the p-Si wafer by the sputtering and thermal annealing methods. To specify, copper sputtered for 30 min and under the pressure of 2.6 × 10⁻³ Torr to make an approximately 180 nm copper film on the clean p-type Si wafer. And then, the copper film on the p-Si wafer was annealed at 500°C for 3 h to obtain the composite film of the p-Si/p-CuO buffer layer. The thickness of the fabricated p-CuO buffer layer is approximately 230 nm, as shown in Fig. 6(b).

n-ZnO NRs were grown on the p-CuO buffer layer by the hydrothermal method. Particularly, the ZnO seed layer was synthesized on the p-CuO buffer layer by uniformly spin coating at a speed of 3000 rpm for 2 min following thermal annealing at 500°C for 1 hour. Then, n-ZnO NRs were grown over the p-CuO buffer layer under the hydrothermal process. A growth solution that contains 50 mL of 20 mM zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O) and 20 mM hexamethylenetetramine (C₆H₁₂N₄) was transferred into an autoclave. Afterwards, the composite film of the p-Si/p-CuO buffer layer coated with a ZnO seed layer was immersed into the growth solution and baked at 100°C for 2 h. After 2 hours of baking, the autoclave was allowed to cool down naturally.

The composite film of p-Si/p-CuO buffer layer/n-ZnO NRs was cleaned ultrasonically in ethanol and deionized water (DI water) for 30 min and then dried at 60°C for 12 h in an oven under atmospheric conditions. The thickness of the fabricated n-ZnO NRs layer is approximately 500 nm, as shown in Fig. 6(d).

For comparison purposes, p-Si/p-CuO, p-Si/n-ZnO, and p-Si/p-CuO/n-ZnO composite films were also fabricated using the same process.

2.4 Characterization

The crystal phases, surface morphologies, and optical absorption spectrum of the fabricated samples were determined using an XRD (D5000), FESEM (Hitachi, S-4800), and UV-Vis spectrophotometer (Jasco, V-670), respectively.

2.5 Photocatalytic activity measurement

The photocatalytic activity measurement was described in our previous report.^6,18)

3. Results and Discussion

3.1 SCAPS analysis

Simulation of the p-Si/p-CuO buffer layer/n-ZnO was performed without introducing additional defects. The thickness of each layer was set as 350 µm, 10 ÷ 500 nm, and 500 nm for the p-Si, p-CuO, and ZnO layers, respectively. The J–V characteristics for the p-Si/p-CuO buffer layer/n-ZnO were shown in Fig. 2. Based on the variation in the thickness of the CuO buffer layer, the short circuit current density can reach to 28.5 mA.cm⁻² when the thickness of the CuO buffer layer is 250 nm. It could be concluded that this p-Si/p-CuO buffer layer/n-ZnO works well in the photo-conversion to electron-hole pairs and the photo-induced current density strongly depend on the thickness of the CuO buffer layer. For further investigation of the influence of the thickness of the CuO buffer layer on the photo-induced current, quantum efficiencies (QE) of the p-Si/p-CuO buffer layer/n-ZnO related to the thickness of the CuO buffer layer were calculated and shown in Fig. 3. This indicated that QE of the ZnO layer, in range of wavelength from 300 nm to 375 nm (3.3 eV), is nearly unchanged and contributed to the photo-induced current density of 1.1 mA.cm⁻² that is about 4% of total photo-induced current density. This small contribution is explained by the wide band gap of ZnO (3.3 eV), so it can only absorb photons in the ultraviolet region of the solar spectrum, which accounts for only 4% of the total solar energy spectrum. In the range of wavelength from 375 nm (3.3 eV) to 850 nm (1.5 eV – bandgap of CuO), QE increases as well as the thickness of the CuO buffer layer increases and reaches the approximate saturation value of 25.8 mA.cm⁻² when the thickness of the CuO buffer layer is 400 nm. The total photo-induced current density is about 28.5 mA.cm⁻². This result indicates that the CuO buffer layer mainly contributes to the photo-induced current density in this composite film. The increasing photo-induced current density in this range of wavelength could be attributed to improving photo absorption resulting from enhanced optical path length in the CuO buffer layer. The dependence of the photo-induced current density on the thickness of the CuO buffer layer can be expressed by the equation $y = y_{0} + A.e^{R_{0}.x}$ as a fit function according to the simulated data, where y and x are related to the photo-induced current density and the thickness of the CuO buffer layer, respectively. y₀, A, and R₀ are coefficiencies depending on the composite film. When the thickness of the CuO buffer layer is higher than 250 nm, in the p-Si region, the QE is reduced as the thickness of the CuO buffer layer increases. In this composite film, the inner electric field is mainly contributed by the junction between n-ZnO and p-CuO. Therefore, if the thickness of the CuO buffer increases, the photo-generated electron-hole pairs in the p-Si region are weakly controlled by the inner electric field, resulting in recombination and reduction in photo-induced current density. The calculated photo-induced current densities versus the thickness of the CuO buffer layer for each layer were shown in Fig. 4. When the thickness of the CuO buffer layer is 250 nm, based on the photo-induced current densities, the number of effective photogeneration electron-hole pairs can be calculated of 3.45 × 10¹⁵ cm⁻², 7.85 × 10¹⁶ cm⁻², and 7.20 × 10¹⁵ cm⁻² for n-ZnO layer, p-CuO buffer layer, and p-Si, respectively. These electron-hole pairs will participate in the decomposition of organic compounds in the photocatalysis process, which will be discussed in more detail in the next section.

Fig. 2

The J–V characteristics of the p-Si/p-CuO buffer layer/n-ZnO NRs. Inset: current density related to the thickness of the CuO buffer layer.

Fig. 3

The quantum efficiency characteristics of the p-Si/p-CuO buffer layer/n-ZnO NRs.

Fig. 4

The calculated photo-induced current density versus the thickness of the CuO buffer layer for each layer for the p-Si/p-CuO buffer layer/n-ZnO NRs composite film.

Furthermore, to further clarify the role of the CuO buffer layer, the p-CuO/n-ZnO and p-Si/n-ZnO composite films were also evaluated and compared with the p-Si/p-CuO/n-ZnO composite film based on the quantum efficiency characteristic as shown in Fig. 5. The evaluation results showed that the quantum efficiency of the pSi/p-CuO/n-ZnO composite film exhibits outstanding photoelectric conversion in particular distinct in the wavelength range from 470 nm to 850 nm - this is the absorption region of CuO and Si. However, by adding a CuO buffer layer, the absorption capacity of CuO is higher than that of Si, leading to a higher photo-electrical conversion efficiency than that of the p-Si/n-ZnO composite film.^6,17,18) To further elucidate the role of the CuO buffer layer, the generated current densities of 20.0 mA.cm⁻², 22.7 mA.cm⁻², and 25.3 mA.cm⁻² in the wavelength region from 380 nm to 850 nm were calculated for the p-CuO/n-ZnO, p-Si/n-ZnO, and p-Si/p-CuO/n-ZnO composite films, respectively. Therefore, in the p-Si/p-CuO/n-ZnO composite film, the CuO buffer layer contributes about 11% to the total current density compared to the p-Si/n-ZnO composite film, which means 11% of added carrier density in the decomposition of organic compounds process.

Fig. 5

The quantum efficiency characteristics of the p-CuO/n-ZnO, p-Si/n-ZnO, and p-Si/p-CuO buffer layer/n-ZnO.

Based on the simulation results, the composite films p-Si/p-CuO, p-Si/n-ZnO, p-Si/p-CuO/n-ZnO, and p-Si/p-CuO/n-ZnO NRs were fabricated and evaluated for photocatalytic ability. The CuO buffer layer was fabricated by sputtering and annealing methods. The obtained thickness of the CuO buffer layer is 230 nm. The ZnO and ZnO nanorods were fabricated by sol-gel and hydrothermal processes.

3.2 SEM analysis

Figure 6 shows the SEM morphology evolution of the composite films. The composite films of p-Si/n-ZnO, p-Si/p-CuO, p-Si/p-CuO/n-ZnO, and p-Si/p-CuO/n-ZnO NRs can be observed in Fig. 6(a)–(d), respectively. The thickness of ZnO film, CuO buffer layer, and ZnO NRs are 120 nm, 230 nm, and 500 nm, respectively. The ZnO and CuO layers are formed from nanoparticles of about 50 nm. The ZnO NR diameters are in the range of 50–70 nm and well-aligned. This result indicated that the p-Si/p-CuO/n-ZnO NRs are well constructed via the processes in the experimental part. The top-view SEM image of the p-Si/p-CuO/n-ZnO NRs composite film is also shown in Fig. 6(e).

Fig. 6

Cross-view SEM images of the composite films (a) p-Si/n-ZnO, (b) p-Si/p-CuO, (c) p-Si/p-CuO/n-ZnO, (d) p-Si/p-CuO/n-ZnO NRs, and (e) Top-view SEM image of the p-Si/p-CuO/n-ZnO NRs.

3.3 XRD analysis

Figure 7 presents the X-ray diffraction pattern of the p-Si/n-ZnO, p-Si/n-CuO, p-Si/p-CuO/n-ZnO, and p-Si/p-CuO/n-ZnO NRs composite films. The diffraction peaks of all samples are well defined, revealing the good crystallinity of the fabricated samples, and no peak of other phases and impurity is detected. The result indicated that the p-Si/p-CuO film has diffraction peaks in the lattice plane families of (0 0 2), (1 1 1), $(\bar{3}\ 1\ 1)$, $(\bar{2}\ 0\ 2)$, (2 0 2), $(\bar{1}\ 1\ 3)$, $(\bar{3}\ 1\ 1)$ corresponds to the diffraction peaks of CuO (JCPDS No. 80-1917). This demonstrated that CuO films have been successfully fabricated by sputtering and annealing methods. Based on the X-ray diffraction pattern, the lattice constants of CuO are determined as a, b, and c are 0.468 nm, 0.342 nm, and 0.513 nm, respectively, and the crystal grains formed with a size are about 22 nm. In the X-ray diffraction patterns of the p-Si/p-CuO/n-ZnO and p-Si/p-CuO/n-ZnO NRs composite films, besides the diffraction peaks corresponding to the diffraction peak of CuO, the lattice plane families of (1 0 0), (1 0 1), (1 0 2), (1 1 0), (1 0 3), (1 1 2), (2 0 1) corresponding to the diffraction peak of ZnO (JCPDS 36-1451) are observed. Furthermore, the crystalline composite film of the p-Si/p-CuO/n-ZnO NR synthesized by the hydrothermal method is in a hexagonal wurtzite composite film. Besides that, the much higher intensity of the (002) diffraction peaks indicates the excellent c-axis orientation of these ZnO NRs, which is consistent with the observation in the SEM images.

Fig. 7

XRD patterns of the p-Si/ZnO, p-Si/CuO, p-Si/CuO/ZnO, and p-Si/CuO/ZnO NRs composite films.

3.4 Optical absorption analysis

Figure 8 shows the optical absorption spectra of the p-CuO, n-ZnO NRs, and CuO/n-ZnO NRs. The n-ZnO NRs showed a band gap energy at around 3.3 eV. The p-CuO film shows a high and broad range of light absorption up to 1.5 eV. Meanwhile, the p-CuO/n-ZnO NRs composite film behaviours not only a broad range absorption but also the highest optical absorption in comparison to that of the n-ZnO NRs and the p-CuO. This phenomenon can be explained by attributions of the high surface roughness of the ZnO NRs in the ZnO NRs/CuO composite film, which can reduce the optical reflection at the surface of the composite film.⁴⁰⁾

Fig. 8

Absorption spectra of the CuO, ZnO NR, and ZnO NRs/CuO.

3.5 Photocatalytic activity

The photocatalytic activities of the composite films were examined by the degradation rate of RhB contamination under Xenon lamp irradiation, as shown in Fig. 9. After 30 min of maintenance in the dark, the concentration change of the RhB solution is negligible, which indicates that the adsorption of RhB in the composite films is negligible. During the optical degradation investigation, the variation of RhB concentration was analyzed by the RhB absorption peak intensity at a wavelength of 554 nm. The photocatalytic activity mechanism of the samples was well described in our previous publication.^6,18) The obtained results are included in equation (eq. (1)) to calculate the decomposition efficiency for different samples shown in Fig. 9(a). After 4 hours of illumination, the degradation efficiency of the p-Si/n-ZnO, p-Si/p-CuO, p-Si/p-CuO/n-ZnO, and p-Si/p-CuO/n-ZnO NRs composite films are 68, 79, 85 and 98%, respectively. This result indicated that the decomposition efficiency of the p-Si/n-ZnO composite film can be enhanced when introduced with the CuO buffer layer. Furthermore, the degradation efficiency of the p-Si/p-CuO/n-ZnO NRs shows the best performance when it reaches 80% after 1 hour of illumination. This can be attributed to the larger effective surface area of the ZnO NRs compared to the ZnO film layer. When the composite films are irradiated under light radiation with photon energy higher than or equal to the band gap of ZnO, CuO, and Si, leading to process of electron-hole pairs is created (eq. (2)). Under the effect of the internal electric field, electrons generated in the conduction band of p-Si and p-CuO move into the conduction band of the ZnO nanorod layer. In the valence band, the holes move in the opposite direction, and the electrons combine with O₂ to form the radical $O_{2}^{ - }$ (eq. (3)). The radical $O_{2}^{ - }$ reacts with H₂O and electrons to give the ⁻OH radical (eq. (4)). The holes in the valence band of p-CuO and p-Si react with H₂O to form ⁻OH radicals (eq. (5)). These ⁻OH radicals react with the RhB solution to form CO₂ and H₂O (eq. (6)). Based on this technique, electrons and holes are effectively separated. The recombination rate of electron-hole pairs is reduced. Furthermore, electrons and holes will play an equal role in photocatalytic activities.

Fig. 9

(a) Photodegradation of RhB under Xenon lamp, (b) the first-order kinetic plot for RhB photodegradation, (c) pseudo-order rate constant, (d) recycling photodegradation of the p-Si/p-CuO/n-ZnO NRs composite film, and (e) pseudo-order rate constant versus action cycle.

The degradation efficiency of RhB molecules was calculated from the following equation:¹⁸⁾   

\begin{equation} \% \textit{Degradation} = \frac{C_{o} - C_{t}}{C_{o}} \times 100\% \end{equation} (1)

where C_o and C_t are the initial absorbance and the absorbance at a certain time t, respectively.

The degradation of the RhB contamination process can be proposed as the following equations.⁶⁾   

\begin{align} &\text{p-Si/p-CuO/n-ZnO} + \text{h}\nu \\ &\quad \to \text{p-Si/p-CuO/n-ZnO}\ (\text{h$^{+}$} + \text{e$^{-}$}) \end{align} (2)

  

\begin{equation} \text{e$^{-}$} + \text{O$_{2}$} \to O_{2}^{-} \end{equation} (3)

  

\begin{equation} O_{2}^{-} + \text{2H$_{2}$O} + \text{e$^{-}$} \to \text{$^{-}$OH} + \text{3OH$^{-}$} \end{equation} (4)

  

\begin{equation} \text{h$^{+}$} + \text{OH$^{-}$} \to \text{$^{-}$OH} \end{equation} (5)

  

\begin{equation} \text{$^{-}$OH} + \text{RhB} \to \text{CO$_{2}$} + \text{H$_{2}$O} \end{equation} (6)

Table 2 shows the photocatalytic activity of the previously reported compounds and p-Si/p-CuO/n-ZnO composite films. The assessment indicates that our p-Si/p-CuO/n-ZnO composite films are suitable for application in wastewater treatment compared with the other photocatalysts.

Table 2 Comparison of the photocatalytic decomposition of the p-Si/p-CuO buffer layer/n-ZnO NRs and ZnO/CuO based composite.

The first-order kinetics of RhB photodegradation were calculated and depicted in Fig. 9(b). The pseudo-order rate constant (k) is determined from the slope of the first-order kinetics of RhB photodegradation and is shown in Fig. 9(c). The results indicated that the k value of the p-Si/n-ZnO and p-Si/p-CuO/n-ZnO composite films in the decomposed RhB concentration was about 0.41 s⁻¹ and 0.46 s⁻¹, respectively. By introducing the p-CuO buffer layer, the pseudo-order rate constant (k) can be enhanced up to 12% compared to the composite film without the p-CuO buffer layer - the p-Si/n-ZnO composite film. This result is completely satisfied with the simulation results obtained from Fig. 5 where the CuO buffer layer contributes up to 11% to the total carrier density compared with the p-Si/n-ZnO composite film. Figure 9(d) shows the photocycle degradation indicating that the p-Si/p-CuO/n-ZnO NRs composite film maintains a good degradation efficiency of above 80% for RhB contamination after four recycles. Figure 8(e) shows the decay of the pseudo-order rate constant of the p-Si/p-CuO/n-ZnO NRs composite film versus the reaction cycle time. The results indicated that the k value decreased sharply after the first three reaction cycle times and gradually stabilized at a value of 0.88 s⁻¹ after four reaction cycle times. Therefore, it can be concluded that the p-Si/p-CuO/n-ZnO NRs composite film is a reusable photocatalyst and high photostability.

4. Conclusions

In this study, the role of the CuO buffer layer in the p-Si/p-CuO/n-ZnO NRs composite films is studied based on experimental and simulation results. The results indicated that the CuO buffer layer plays a very active role in the photoelectric conversion of the p-Si/p-CuO/n-ZnO composite film. With the optimum thickness of 250 nm, the CuO buffer layer contributed 11% to the total photocurrent density of the p-Si/p-CuO/n-ZnO composite film. This result was also experimentally retested, and the results showed that the k value of the p-Si/p-CuO/n-ZnO composite film increased by 12% compared with that of the p-Si/n-ZnO composite film. In addition, with the p-Si/p-CuO/n-ZnO NRs composite film, the decomposition efficiency can reach over 80% after 1 h of illumination, and the k-value can reach up to 1.1 s⁻¹. The p-Si/p-CuO/n-ZnO NRs composite film was also tested for reusability and the results showed that, after the first three cycles, the k value decreased sharply, however, from the fourth onwards, the k value is stable around the 0.88 s⁻¹. Therefore, the composite film of p-Si/p-CuO/n-ZnO NRs has high decomposition efficiency, a reusable photocatalyst, and high photostability.

Acknowledgements

The authors are thankful to Prof. Marc Burgelman, University of Gent, Belgium for providing the SCAPS software for our study. This research was supported by Asia Research Center (ARC) and Chey Institute for Advanced Studies (CHEY) under Grant Number CA.21.02A.

REFERENCES

• 1)   Q.  Zhai,  L.  Song,  X.  Ji,  Y.  Yu,  J.  Ye,  W.  Xu and  M.  Hou: Environ. Sci. Pollut. Res. 29 (2022) 40449–40461. doi:10.1007/s11356-022-19792-w
• 2)   X.  Wang,  J.  Jing,  M.  Zhou and  R.  Dewil: Chin. Chem. Lett. (2022). doi:10.1016/j.cclet.2022.06.044
• 3)   S.  Panwar,  G.K.  Upadhyay and  L.P.  Purohit: Mater. Res. Bull. 145 (2022) 111534. doi:10.1016/j.materresbull.2021.111534
• 4)   S.  Mao,  J.-W.  Shi,  G.  Sun,  D.  Ma,  C.  He,  Z.  Pu,  K.  Song and  Y.  Cheng: Appl. Catal. B 282 (2021) 119550. doi:10.1016/j.apcatb.2020.119550
• 5)   L.  Wu,  X.  Yang,  T.  Chen,  Y.  Li,  Q.  Meng,  L.  Zhu,  W.  Zhu,  R.  He and  T.  Duan: Chem. Eng. J. 427 (2022) 131773. doi:10.1016/j.cej.2021.131773
• 6)   N.T.  Hoa,  V.  Van Cuong and  N.D.  Lam: Mater. Chem. Phys. 204 (2018) 397–402. doi:10.1016/j.matchemphys.2017.10.070
• 7)   J.T.  Calow,  P.J.  Deasley,  S.J.T.  Owen and  P.W.  Webb: J. Mater. Sci. 2 (1967) 88–96. doi:10.1007/BF00550056
• 8)   N.Q.  Thang,  A.  Sabbah,  L.-C.  Chen,  K.-H.  Chen,  C.M.  Thi and  P.  Van Viet: Chemosphere 282 (2021) 130971. doi:10.1016/j.chemosphere.2021.130971
• 9)   M.  Yilmaz,  N.  Mengelizadeh,  M.  Saloot,  S.  Shahbaksh and  D.  Balarak: Mater. Sci. Semicond. Process. 144 (2022) 106593. doi:10.1016/j.mssp.2022.106593
• 10)   S.  Goktas and  A.  Goktas: J. Alloy. Compd. 863 (2021) 158734. doi:10.1016/j.jallcom.2021.158734
• 11)   I.  Ahmad,  S.  Shukrullah,  M.  Ahmad,  E.  Ahmed,  M.Y.  Naz,  M.S.  Akhtar,  N.R.  Khalid,  A.  Hussain and  I.  Hussain: Mater. Sci. Semicond. Process. 123 (2021) 105584. doi:10.1016/j.mssp.2020.105584
• 12)   Y.  Lv,  J.  Liu,  Z.  Zhang,  W.  Zhang,  A.  Wang,  F.  Tian,  W.  Zhao and  J.  Yan: Mater. Chem. Phys. 267 (2021) 124703. doi:10.1016/j.matchemphys.2021.124703
• 13)   S.B.  Mitta,  P.  Murahari,  K.R.  Nandanapalli,  D.  Mudusu,  R.  Karuppannan and  D.  Whang: Int. J. Hydrogen Energ. 43 (2018) 16015–16023. doi:10.1016/j.ijhydene.2018.07.038
• 14)   W.Z.  Tawfik,  M.A.  Hassan,  M.A.  Johar,  S.-W.  Ryu and  J.K.  Lee: J. Catal. 374 (2019) 276–283. doi:10.1016/j.jcat.2019.04.045
• 15)   Z.  Chen,  T.  Dedova,  N.  Spalatu,  N.  Maticiuc,  M.  Rusu,  A.  Katerski,  I.O.  Acik,  T.  Unold and  M.  Krunks: Colloids Surf. A Physicochem. Eng. Asp. 648 (2022) 129366. doi:10.1016/j.colsurfa.2022.129366
• 16)   T.  Senasu,  T.  Chankhanittha,  K.  Hemavibool and  S.  Nanan: Mater. Sci. Semicond. Process. 123 (2021) 105558. doi:10.1016/j.mssp.2020.105558
• 17)   S.  Paul,  J.  Sultana,  A.  Bhattacharyya,  A.  Karmakar and  S.  Chattopadhyay: Optik 164 (2018) 745–752. doi:10.1016/j.ijleo.2018.03.076
• 18)   D.T.  Nguyen,  M.D.  Tran,  T.  Van Hoang,  D.T.  Trinh,  D.T.  Pham and  D.L.  Nguyen: Sci. Rep. 10 (2020) 7792. doi:10.1038/s41598-020-64784-w
• 19)   A.F.  Violas,  A.J.N.  Oliveira,  J.P.  Teixeira,  T.S.  Lopes,  J.R.S.  Barbosa,  P.A.  Fernandes and  P.M.P.  Salomé: Sol. Energy Mater. Sol. Cells 243 (2022) 111792. doi:10.1016/j.solmat.2022.111792
• 20)   C.-H.  Huang and  W.-J.  Chuang: Vacuum 118 (2015) 32–37. doi:10.1016/j.vacuum.2015.03.008
• 21)   G.  Pindolia,  S.M.  Shinde and  P.K.  Jha: Sol. Energy 236 (2022) 802–821. doi:10.1016/j.solener.2022.03.053
• 22)   K.  Kim,  J.  Gwak,  S.K.  Ahn,  Y.-J.  Eo,  J.H.  Park,  J.-S.  Cho,  M.G.  Kang,  H.-E.  Song and  J.H.  Yun: Sol. Energy 145 (2017) 52–58. doi:10.1016/j.solener.2017.01.031
• 23)   A.  Houimi,  S.Y.  Gezgin,  B.  Mercimek and  H.Ş.  Kılıç: Opt. Mater. 121 (2021) 111544. doi:10.1016/j.optmat.2021.111544
• 24)   B.  Zhou,  X.  Yin,  J.  Zhang,  G.  Zeng,  B.  Li,  J.  Zhang and  L.  Feng: Opt. Mater. 110 (2020) 110505. doi:10.1016/j.optmat.2020.110505
• 25)   P.  Sawicka-Chudy,  Z.  Starowicz,  G.  Wisz,  R.  Yavorskyi,  Z.  Zapukhlyak,  M.  Bester,  Ł.  Głowa,  M.  Sibiński and  M.  Cholewa: Mater. Res. Express 6 (2019) 085918. doi:10.1088/2053-1591/ab22aa
• 26)   V.  Srikant and  D.R.  Clarke: J. Appl. Phys. 83 (1998) 5447. doi:10.1063/1.367375
• 27)   J.I.  Sohn,  S.N.  Cha,  B.G.  Song,  S.  Lee,  S.M.  Kim,  J.  Ku,  H.J.  Kim,  Y.J.  Park,  B.L.  Choi,  Z.L.  Wang,  J.M.  Kim and  K.  Kim: Energy Environ. Sci. 6 (2013) 97–104. doi:10.1039/C2EE23404A
• 28)   F.  Anwar,  R.  Mahbub,  S.S.  Satter and  S.M.  Ullah: Int. J. Photoenergy 2017 (2017) 9846310. doi:10.1155/2017/9846310
• 29)   A.  Dhara,  A.  Baral,  S.  Chabri,  A.  Sinha,  N.R.  Bandyopadhyay and  N.  Mukherjee: J. Inst. Eng. India Ser. D 98 (2017) 1–8. doi:10.1007/s40033-016-0111-2
• 30)  S. Ahmmed, A. Aktar, A. Kuddus and A.B.Md. Ismail: 2018 International Conference on Computer, Communication, Chemical, Material and Electronic Engineering (IC4ME2), (2018) pp. 1–4. doi:10.1109/IC4ME2.2018.8465662.
• 31)   B.-S.  Kim,  E.  Kim,  C.-K.  Kim,  H.-I.  Lee and  J.-S.  Sohn: Mater. Trans. 49 (2008) 1192–1198. doi:10.2320/matertrans.MER2007311
• 32)   L.  Zhao,  M.  Xia,  Y.  Liu,  B.  Zheng,  Q.  Jiang and  J.  Lian: Mater. Trans. 53 (2012) 463–468. doi:10.2320/matertrans.M2011345
• 33)   S.  Shimizu and  H.  Shiomi: Mater. Trans. 63 (2022) 636–643. doi:10.2320/matertrans.MT-M2021235
• 34)   R.  Saravanan,  S.  Karthikeyan,  V.K.  Gupta,  G.  Sekaran,  V.  Narayanan and  A.J.M.S.  Stephen: Mater. Sci. Eng. C 33 (2013) 91–98. doi:10.1016/j.msec.2012.08.011
• 35)   M.  Hassanpour,  H.  Safardoust-Hojaghan,  M.  Salavati-Niasari and  A.  Yeganeh-Faal: J. Mater. Sci. Mater. Electron. 28 (2017) 14678–14684. doi:10.1007/s10854-017-7333-4
• 36)   L.  Xu,  Y.  Zhou,  Z.  Wu,  G.  Zheng,  J.  He and  Y.  Zhou: J. Phys. Chem. Solids 106 (2017) 29–36. doi:10.1016/j.jpcs.2017.03.001
• 37)   A.M.  Cahino,  R.G.  Loureiro,  J.  Dantas,  V.S.  Madeira and  P.C.R.  Fernandes: Ceram. Int. 45 (2019) 13628–13636. doi:10.1016/j.ceramint.2019.03.239
• 38)   Z.  Dong,  Y.  Wu,  N.  Thirugnanam and  G.  Li: Appl. Surf. Sci. 430 (2018) 293–300. doi:10.1016/j.apsusc.2017.07.186
• 39)   M.A.  Hanif,  J.  Akter,  Y.S.  Kim,  H.G.  Kim,  J.R.  Hahn and  L.K.  Kwac: Catalysts 12 (2022) 151. doi:10.3390/catal12020151
• 40)   L.  Chen,  Y.  Cui,  Z.  Xiong,  M.  Zhou and  Y.  Gao: RSC Advances 9 (2019) 21831–21843. doi:10.1039/C9RA03484F