Photochemical Oxidation Process of Copper from Electroplating Wastewater: Process Performance and Kinetic Study

: An investigation of the process of ozone combined with ultraviolet radiation has been carried out in order to establish the kinetics for photochemical oxidation of copper (Cu) from electroplating wastewater. The e ﬀ ects of operating parameters, including initial Cu concentration, ozone dosage, UV irradiation intensity, and pH value on the photochemical oxidation of Cu have been studied comprehensively. The Cu concentration during the reaction was identiﬁed using atomic absorption spectroscopy (AAS) method. The solid product was analyzed using X-ray di ﬀ raction (XRD) and scanning electron microscope–energy-dispersive X-ray (SEM–EDX) methods. It was found that the UV-Ozone process has high performance on Cu removal compared to UV and Ozone processes due to the high production rate of HO • radicals. It was also found that the solid product from the UV-Ozone process was CuO monoclinic crystal phase. The initial Cu concentration, ozone dosage, and pH value were signiﬁcantly a ﬀ ected the Cu removal e ﬃ ciency. On the other hand, the UV irradiation intensity was not signiﬁcant; however, it has responsibility in promoting the ozone photolysis. The kinetics model for the photochemical oxidation of Cu was established following the ﬁrst-order kinetic model. Furthermore, the reaction mechanism was also developed.


Introduction
Heavy metals have become a global issue of environmental and public health concern because of their toxicity and bioaccumulation in the human body and food chain [1]. The effects of urbanization and industrialization cause an increase in heavy metal pollution to the environment [2]. High toxicity and nonbiodegradability of heavy metals caused a number of environmental problems [3]. The accumulation of heavy metals in the atmosphere is responsible for both natural and anthropogenic activities [4]. Copper (Cu), as an essential trace element, is required by biological systems for the activation of some enzymes during photosynthesis. However, at higher concentrations, it shows harmful effects on the human body. Continuous exposure may lead to kidney damage and even death. Cu is also toxic to a variety of aquatic organisms even at very low concentrations. Mining, metallurgy, and industrial applications are the major sources of Cu exposure in the environment [5].
There were several techniques for the treatment of industrial wastewater containing heavy metals, including chemical precipitation, ion exchange, coagulation-flocculation, flotation, membrane filtration, electrochemical treatment, magnetic separation and purification, biosorption, and nanotechnology [6]. Advanced oxidation processes (AOPs) are promising, efficient, and environmentally friendly methods

Experimental System Setup
The experimental study on photochemical oxidation of copper by ozone combined with UV irradiation was conducted in a bubble column reactor made of borosilicate glass. The reactor was equipped with a low-pressure mercury UV light (Philips-TUV 8 Watt, Koninklijke Philips N.V., Amsterdam, The Netherlands) with main emission line at 253.7 nm. Ozone gas generated from ozone generator type dielectric barrier discharge (Dipo Technology, Diponegoro University, Jawa Tengah, Indonesia) connected to reactors UV-Ozone ( Figure 1). The ozone dosage was varied at 5, 10, and 15 mg/h. The UV-Ozone photoreactor was performed under a semibatch condition reaction. The reactor was an open-top cylindrical tank with a fixed top suspension with a low-pressure mercury UV light. UV irradiation intensity was varied at 20, 40, and 60 mW/cm 2 . The UV irradiation intensity was adjusted by varying the number of UV light. The UV irradiation intensity was measured using a J-225 Black Ray intensity meter (Analytik Jena US, Upland, CA, USA). A fritted glass diffuser was placed at the bottom of the reactor to allow the continuous injection of ozonized gas. Ozone was produced from the air, which flows into the ozone generator. The reaction was adjusted for different time periods (i.e., 10,20,30,40,50, and 60 min) and initial pH (3, 6, 8, and 10) during UV-Ozone treatment. All oxidation reaction processes were conducted at room temperature and atmospheric pressure. The initial pH was adjusted using NaOH and HCl solutions.

Analytical Methods
A pH meter analyzer (EZDO PH-5011A) was used during the procedure for assessing the pH of the response solution. An atomic absorption spectroscopy (AAS) (Shimadzu AA 6300, Kyoto, Japan) was used to analyze Cu concentration and wastewater samples. Even though the UV-VIS method can be used to analyzed Cu concentration in the solutions, the shift on maximum wavelength absorbance at different pH could affect the inaccuracy in the measurement [19][20][21] since we variated the pH as one of the process parameters investigated. The solid product was also analyzed using X-ray diffraction (XRD) (Shimadzu XRD-7000, Kyoto, Japan) method. The Cu Kα radiation (λ = 1.54 Å) was operated at 30 mA and 30 kV. The diffraction patterns were generated at 2θ angle ranges of 20−70° with a scanning speed of 4° min −1 . The surface morphology and metal oxide composition of the solid product were analyzed using scanning electron microscope-energy dispersive X-ray (SEM-EDX) (SEM-EDX JEOL JSM-6510LA, Tokyo, Japan) method. The removal efficiency was calculated using

Analytical Methods
A pH meter analyzer (EZDO PH-5011A) was used during the procedure for assessing the pH of the response solution. An atomic absorption spectroscopy (AAS) (Shimadzu AA 6300, Kyoto, Japan) was used to analyze Cu concentration and wastewater samples. Even though the UV-VIS method can be used to analyzed Cu concentration in the solutions, the shift on maximum wavelength absorbance at different pH could affect the inaccuracy in the measurement [19][20][21] since we variated the pH as one of the process parameters investigated. The solid product was also analyzed using X-ray diffraction (XRD) (Shimadzu XRD-7000, Kyoto, Japan) method. The Cu Kα radiation (λ = 1.54 Å) was operated at 30 mA and 30 kV. The diffraction patterns were generated at 2θ angle ranges of 20−70 • with a scanning speed of 4 • min −1 . The surface morphology and metal oxide composition of the solid product were analyzed using scanning electron microscope-energy dispersive X-ray (SEM-EDX) (SEM-EDX JEOL JSM-6510LA, Tokyo, Japan) method. The removal efficiency was calculated using Equation (1), when η is the Cu removal efficiency (%), C 0 is the initial Cu concentration (mg/L), and C t (mg/L) is the Cu concentration at time t. η(%) = C 0 − C t C 0 × 100. (1)

Comparation of UV, Ozone. and Combination UV-Ozone
In order to study the different effects of UV, Ozone, and UV-Ozone combination processes on Cu removal from electroplating wastewater, these three processes are compared. The result is shown in Figure 2. Equation (1), when η is the Cu removal efficiency (%), C0 is the initial Cu concentration (mg/L), and Ct (mg/L) is the Cu concentration at time t. 0 0 (%) 100

Comparation of UV, Ozone. and Combination UV-Ozone
In order to study the different effects of UV, Ozone, and UV-Ozone combination processes on Cu removal from electroplating wastewater, these three processes are compared. The result is shown in Figure 2.  Figure 2 depicted the comparison of UV, Ozone, and combined UV-Ozone treatment on Cu removal efficiency from electroplating wastewater. As can be seen, at 60 min of reaction, the Cu removal efficiency obtained from the UV irradiation process is 69.08%, whereas the Cu removal efficiency obtained from Ozone treatment is 72.64%. Interestingly, the Cu removal efficiency can be increased up to 78.8% when the UV and Ozone processes are combined. This high Cu removal efficiency can be obtained by the enhancement of hydroxyl radicals (HO•) during the combination of UV and Ozone processes. As reported by Hanela et al. [10], the combination of UV irradiation and ozone could enhance the production of hydroxyl radicals; therefore, the oxidation process could be accelerated. Furthermore, it is suggested that the combination of UV irradiation and Ozone treatment can increase the Cu removal efficiency in the Cu removal process from electroplating wastewater. It is true since the mechanism of Cu oxidation through AOP is initiated by the presence of HO• radicals. Equations (2) and (3) represent the oxidation process of Cu through AOP. As can be seen in Equation (2), the Cu 2+ ion is oxidized by HO• radicals producing solid CuO and water molecules.
Based on Equation (2), the important thing in Cu removal through AOP is the presence of HO• radicals. Therefore, it is suggested that the process of producing more HO• radicals is the preferred  Figure 2 depicted the comparison of UV, Ozone, and combined UV-Ozone treatment on Cu removal efficiency from electroplating wastewater. As can be seen, at 60 min of reaction, the Cu removal efficiency obtained from the UV irradiation process is 69.08%, whereas the Cu removal efficiency obtained from Ozone treatment is 72.64%. Interestingly, the Cu removal efficiency can be increased up to 78.8% when the UV and Ozone processes are combined. This high Cu removal efficiency can be obtained by the enhancement of hydroxyl radicals (HO•) during the combination of UV and Ozone processes. As reported by Hanela et al. [10], the combination of UV irradiation and ozone could enhance the production of hydroxyl radicals; therefore, the oxidation process could be accelerated. Furthermore, it is suggested that the combination of UV irradiation and Ozone treatment can increase the Cu removal efficiency in the Cu removal process from electroplating wastewater. It is true since the mechanism of Cu oxidation through AOP is initiated by the presence of HO• radicals. Equations (2) and (3) represent the oxidation process of Cu through AOP. As can be seen in Equation (2), the Cu 2+ ion is oxidized by HO• radicals producing solid CuO and water molecules.
Based on Equation (2), the important thing in Cu removal through AOP is the presence of HO• radicals. Therefore, it is suggested that the process of producing more HO• radicals is the preferred process. As can be seen in Figure 2, the Cu removal efficiency obtained by the UV irradiation process is the lowest compared to the other processes. It is true since the UV irradiation process without the presence of ozone in the system only produces HO• radicals through the photolysis of the H 2 O molecule [12,22]. The photolysis of the H 2 O molecule by UV irradiation is shown in Equation (4) [12]. The other possible mechanisms in HO• production through the UV irradiation process are shown in Equations (5) and (6) [23].
For the Ozone treatment, as depicted in Figure 2, the Cu removal efficiency is higher than the Cu removal efficiency obtained by the UV irradiation process. This is due to the fact that the Cu oxidation through Ozone treatment can be initiated by HO• radicals and directly oxidized by the ozone molecule (Equation (3)). These two reactions occurred in Ozone treatment, therefore; the Cu removal efficiency obtained by Ozone treatment is higher than obtained by UV treatment. The chemical reaction of HO• radicals formation through Ozone treatment is shown in Equation (7) [12].
As mentioned before, the highest Cu removal efficiency is obtained by the UV-Ozone process due to the high HO• radical formation during the combination of the UV irradiation process and the Ozone process. Besides, the presence of ozone can also increase the oxidation process due to the direct oxidation process. The formation of HO• radicals in the UV-Ozone process has been widely reported. Some proposed HO• radical formation during the UV-Ozone process follow the reaction mechanism as described in Equations (8)-(10) [12,24]: As can be seen, in the UV-Ozone process, an additional oxidant, H 2 O 2 , is generated through O 3 photolysis (Equation (8)) [12]. O 3 reacts with H 2 O to form H 2 O 2 under UV irradiation. This oxidant can produce more HO• radicals. H 2 O 2 absorbs UV light to generate HO• radicals. It should be noted that this process can occur only at the presence in light sources with wavelength below 300 nm [17]. Therefore, UV light is preferred rather than visible light in this study. By this fact, the Cu removal efficiency can be increased following the reaction shown in Equation (2) in the combination of UV and Ozone process. This finding is in accordance with several previous studies. In dyehouse wastewater subjected to the combination of UV-Ozone treatment, a significant color reduction, up to 98.3%, was obtained [13]. For another reason, Bes-Piá et al. [11] studied the UV-Ozone process for textile wastewater treatment. They reported that the combination of UV irradiation with ozone could significantly reduce the operating time to reach the same COD removal efficiency.

Characterization of the Solid Product
As reported in Section 3.1, the UV-Ozone process has the highest Cu removal efficiency compared to the other processes. The Cu removal through this process produces a solid product that is proposed as CuO, as shown in Equations (2) and (3). In order to prove that the solid product is CuO, the solid product was analyzed. Figure 3 shows the XRD pattern of the solid product obtained from the UV-Ozone process. Based on Figure 3, the characteristic peaks appear at 2θ of 32 These also indicate that the formation of monoclinic crystal structure [25]. On the other hand, several studies reported that two peaks at 2θ = 35.6 • (002) and 2θ = 38.8 • (111) observed in the diffraction patterns are ascribed to the formation of the CuO (space group C2/c) monoclinic crystal phase [26,27]. In this study, these peaks appear at 2θ = 35.56 • and 2θ = 38.74 • . Therefore, it is confirmed that the solid product produced from the UV-Ozone process is CuO in the monoclinic crystal phase.

Characterization of the Solid Product
As reported in Section 3.1, the UV-Ozone process has the highest Cu removal efficiency compared to the other processes. The Cu removal through this process produces a solid product that is proposed as CuO, as shown in Equations (2) and (3). In order to prove that the solid product is CuO, the solid product was analyzed. Figure 3 shows the XRD pattern of the solid product obtained from the UV-Ozone process. Based on Figure 3, the characteristic peaks appear at 2θ of 32.54°, 35.56°, 38.74°, 48.68°, 53.62°, 58.16°, 61.62°, 66.12°, and 68.12°. These peaks correspond to the CuO (tenorite phase) according to the JCPDS card number 96-900-8962. These also indicate that the formation of monoclinic crystal structure [25]. On the other hand, several studies reported that two peaks at 2θ = 35.6° (002) and 2θ = 38.8° (111) observed in the diffraction patterns are ascribed to the formation of the CuO (space group C2/c) monoclinic crystal phase [26,27]. In this study, these peaks appear at 2θ = 35.56° and 2θ = 38.74°. Therefore, it is confirmed that the solid product produced from the UV-Ozone process is CuO in the monoclinic crystal phase. The morphology and the metal oxide composition of the solid product from the UV-Ozone process are identified using SEM analysis. Based on the result of SEM-EDX analysis (Figure 4a), it is confirmed that the most metal oxide present in the solid product is CuO. The CuO content detected by SEM-EDX analysis in the solid product is 83.52%. The other components detected may be produced from the impurities in the wastewater. However, it is confirmed that the solid product is CuO. Figure 4b shows the appearance of the solid product surface (right) and the model of the CuO monoclinic crystal structure (left). As can be seen, the surface morphology of the CuO is agglomerated nanorods. Manyasree et al. [28] reported that the surface morphology of the CuO nanoparticle, which was synthesized from copper sulfate and sodium hydroxide through the coprecipitation process, is a flower-shaped structure. 20   The morphology and the metal oxide composition of the solid product from the UV-Ozone process are identified using SEM analysis. Based on the result of SEM-EDX analysis (Figure 4a), it is confirmed that the most metal oxide present in the solid product is CuO. The CuO content detected by SEM-EDX analysis in the solid product is 83.52%. The other components detected may be produced from the impurities in the wastewater. However, it is confirmed that the solid product is CuO. Figure 4b shows the appearance of the solid product surface (right) and the model of the CuO monoclinic crystal structure (left). As can be seen, the surface morphology of the CuO is agglomerated nanorods. Manyasree et al. [28] reported that the surface morphology of the CuO nanoparticle, which was synthesized from copper sulfate and sodium hydroxide through the coprecipitation process, is a flower-shaped structure.

Effect of Initial Concentration
In order to study the effect of the initial Cu concentration, the photochemical oxidation process was conducted with the variation of the initial Cu concentration. The initial Cu concentration was varied at 145.73, 72.86, and 36.9 mg/L. The monitored parameter is the Cu removal efficiency. The effect of initial Cu concentration on Cu removal efficiency is depicted in Figure 5.

Effect of Initial Concentration
In order to study the effect of the initial Cu concentration, the photochemical oxidation process was conducted with the variation of the initial Cu concentration. The initial Cu concentration was varied at 145.73, 72.86, and 36.9 mg/L. The monitored parameter is the Cu removal efficiency. The effect of initial Cu concentration on Cu removal efficiency is depicted in Figure 5.

Effect of Initial Concentration
In order to study the effect of the initial Cu concentration, the photochemical oxidation process was conducted with the variation of the initial Cu concentration. The initial Cu concentration was varied at 145.73, 72.86, and 36.9 mg/L. The monitored parameter is the Cu removal efficiency. The effect of initial Cu concentration on Cu removal efficiency is depicted in Figure 5. As can be seen in Figure 5, the Cu removal efficiency is significantly affected by the initial Cu concentration. During the oxidation process for 60 min, the Cu removal efficiency obtained at the initial Cu concentration of 145.73 mg/L is 34.39%, the Cu removal efficiency obtained at the initial Cu concentration of 72.86 mg/L is 54.36%, while the Cu removal efficiency obtained at the initial Cu concentration of 36.9 mg/L is 68.46%. Based on these results, it is suggested that the initial Cu concentration has a significant effect on the Cu removal efficiency in the Cu removal process from electroplating wastewater. Furthermore, the Cu removal efficiency decreases significantly with the increase in the initial Cu concentration from 36.9 to 145.73 mg/L. The decrease in Cu removal efficiency at high initial Cu concentration is due to the presence of a high amount of hydroxyl radical scavengers. When the initial Cu concentration is high, the photochemical oxidation process is overloaded. As a result, the Cu in the solution competes with each other during the oxidation reaction process [29]. On the other hand, in the UV irradiation process, the permeation of photons is reduced at high solution concentration; therefore, the concentration of hydroxyl radical produced is low [29]. Therefore, the oxidation of Cu decreases at high initial Cu concentration. This finding is in accordance with several previous studies in AOP. Hassan et al. [30] reported that the decolorization of Direct Yellow 50 dye in seawater through the UV-Ozone process was highly affected by the initial dye concentration, i.e., the decolorization rate decreased by increasing the dye concentration. Dai et al. [31] also reported a similar finding in the degradation of carbamazepine in water through AOP. They reported that the degradation percentage of carbamazepine decreased from 34% to 13% with an increase in the initial carbamazepine concentration from 4.2 to 42.3 µM. Jing et al. [32] also reported that the initial aniline concentration affected the aniline degradation process through ozonation. They reported that aniline degradation decreased with the increase in the initial aniline concentration due to the overload of the ozonation process.

Effect of Ozone Dosage
In this study, the effect of ozone dosage on the Cu removal process through the photochemical oxidation process is studied by varying the ozone dosage. The ozone dosage was varied at 5, 10, and 15 mg/h. Cu removal efficiency was monitored to study this matter. The effect of ozone dosage on Cu removal efficiency is depicted in Figure 6. As can be seen in Figure 5, the Cu removal efficiency is significantly affected by the initial Cu concentration. During the oxidation process for 60 min, the Cu removal efficiency obtained at the initial Cu concentration of 145.73 mg/L is 34.39%, the Cu removal efficiency obtained at the initial Cu concentration of 72.86 mg/L is 54.36%, while the Cu removal efficiency obtained at the initial Cu concentration of 36.9 mg/L is 68.46%. Based on these results, it is suggested that the initial Cu concentration has a significant effect on the Cu removal efficiency in the Cu removal process from electroplating wastewater. Furthermore, the Cu removal efficiency decreases significantly with the increase in the initial Cu concentration from 36.9 to 145.73 mg/L. The decrease in Cu removal efficiency at high initial Cu concentration is due to the presence of a high amount of hydroxyl radical scavengers. When the initial Cu concentration is high, the photochemical oxidation process is overloaded. As a result, the Cu in the solution competes with each other during the oxidation reaction process [29]. On the other hand, in the UV irradiation process, the permeation of photons is reduced at high solution concentration; therefore, the concentration of hydroxyl radical produced is low [29]. Therefore, the oxidation of Cu decreases at high initial Cu concentration. This finding is in accordance with several previous studies in AOP. Hassan et al. [30] reported that the decolorization of Direct Yellow 50 dye in seawater through the UV-Ozone process was highly affected by the initial dye concentration, i.e., the decolorization rate decreased by increasing the dye concentration. Dai et al. [31] also reported a similar finding in the degradation of carbamazepine in water through AOP. They reported that the degradation percentage of carbamazepine decreased from 34% to 13% with an increase in the initial carbamazepine concentration from 4.2 to 42.3 μM. Jing et al. [32] also reported that the initial aniline concentration affected the aniline degradation process through ozonation. They reported that aniline degradation decreased with the increase in the initial aniline concentration due to the overload of the ozonation process.

Effect of Ozone Dosage
In this study, the effect of ozone dosage on the Cu removal process through the photochemical oxidation process is studied by varying the ozone dosage. The ozone dosage was varied at 5, 10, and 15 mg/h. Cu removal efficiency was monitored to study this matter. The effect of ozone dosage on Cu removal efficiency is depicted in Figure 6. As can be seen in Figure 6, the Cu removal efficiency increases significantly with increasing the ozone dosage. At 60 min reaction, the Cu removal efficiency obtained at ozone dosage of 5 mg/h is As can be seen in Figure 6, the Cu removal efficiency increases significantly with increasing the ozone dosage. At 60 min reaction, the Cu removal efficiency obtained at ozone dosage of 5 mg/h is 48.31%, the Cu removal efficiency obtained at ozone dosage of 10 mg/h is 68.46%, while the Cu removal efficiency obtained at ozone dosage of 15 mg/h is 78.80%. Based on these results, it is suggested that the Cu removal efficiency in Cu removal from electroplating wastewater through the photochemical oxidation process is highly affected by the ozone dosage. As the ozone dosage increases, the Cu removal efficiency increases. It is true since in the ozonation process, the oxidation of Cu can occur through two different mechanisms, i.e., oxidation by HO• radicals (Equation (2)) and direct oxidation by ozone (Equation (3)). As reported by Wang et al. [33], in the ozone-based oxidation process, there are two methods, namely, indirect reaction of free radicals and direct reaction.
Pertaining to the effect of ozone dosage on HO• radical formation, when ozone dosage increases, more HO• radicals are formed. Under UV irradiation, H 2 O 2 can be formed through the reaction of ozone with H 2 O (Equation (8)). Furthermore, this H 2 O 2 molecule absorbs the UV light to generate the HO• radicals (Equation (9)) [24]. Ozone molecules also can react with H 2 O 2 molecules to produce HO• radicals. Besides, the ozone also can also directly react with H 2 O to generate HO• radicals (Equation (7)). Then, Cu is oxidized by HO• radicals. Therefore, the photochemical oxidation rate of Cu increases with the increase in ozone dosage. As a comparison, Wang et al. [33] reported that the decomplexation of electroplating wastewater by the ozone-based oxidation process increased with the increase in ozone dosage. Ren et al. [24] reported that the removal efficiency of polyacrylamide through the photochemical oxidation process increased with the increase in ozone dosage. Guo et al. [34] also reported the same result in the degradation process of sulfadiazine in water by the UV-Ozone process.

Effect of UV Irradiation Intensity
In order to study the effect of UV irradiation intensity on Cu removal from electroplating wastewater, the photochemical process was conducted by varying the UV irradiation intensity. The UV irradiation intensity was varied at 20, 40, and 60 mW/cm 2 . Figure 7 shows the result of the effect of UV irradiation intensity on Cu removal efficiency. 48.31%, the Cu removal efficiency obtained at ozone dosage of 10 mg/h is 68.46%, while the Cu removal efficiency obtained at ozone dosage of 15 mg/h is 78.80%. Based on these results, it is suggested that the Cu removal efficiency in Cu removal from electroplating wastewater through the photochemical oxidation process is highly affected by the ozone dosage. As the ozone dosage increases, the Cu removal efficiency increases. It is true since in the ozonation process, the oxidation of Cu can occur through two different mechanisms, i.e., oxidation by HO• radicals (Equation (2)) and direct oxidation by ozone (Equation (3)). As reported by Wang et al. [33], in the ozone-based oxidation process, there are two methods, namely, indirect reaction of free radicals and direct reaction.
Pertaining to the effect of ozone dosage on HO• radical formation, when ozone dosage increases, more HO• radicals are formed. Under UV irradiation, H2O2 can be formed through the reaction of ozone with H2O (Equation (8)). Furthermore, this H2O2 molecule absorbs the UV light to generate the HO• radicals (Equation (9)) [24]. Ozone molecules also can react with H2O2 molecules to produce HO• radicals. Besides, the ozone also can also directly react with H2O to generate HO• radicals (Equation (7)). Then, Cu is oxidized by HO• radicals. Therefore, the photochemical oxidation rate of Cu increases with the increase in ozone dosage. As a comparison, Wang et al. [33] reported that the decomplexation of electroplating wastewater by the ozone-based oxidation process increased with the increase in ozone dosage. Ren et al. [24] reported that the removal efficiency of polyacrylamide through the photochemical oxidation process increased with the increase in ozone dosage. Guo et al. [34] also reported the same result in the degradation process of sulfadiazine in water by the UV-Ozone process.

Effect of UV Irradiation Intensity
In order to study the effect of UV irradiation intensity on Cu removal from electroplating wastewater, the photochemical process was conducted by varying the UV irradiation intensity. The UV irradiation intensity was varied at 20, 40, and 60 mW/cm 2 . Figure 7 shows the result of the effect of UV irradiation intensity on Cu removal efficiency. As shown in Figure 7, at 60 min reaction, the Cu removal efficiency obtained at a UV irradiation intensity of 20 mW/cm 2 is 65.09%, the Cu removal efficiency obtained at UV irradiation intensity of 40 mW/cm 2 is 68.46%, while the Cu removal efficiency obtained at UV irradiation intensity of 60 mW/cm 2 is 69.08%. It can be concluded that the Cu photochemical oxidation rate slightly increases As shown in Figure 7, at 60 min reaction, the Cu removal efficiency obtained at a UV irradiation intensity of 20 mW/cm 2 is 65.09%, the Cu removal efficiency obtained at UV irradiation intensity of 40 mW/cm 2 is 68.46%, while the Cu removal efficiency obtained at UV irradiation intensity of 60 mW/cm 2 is 69.08%. It can be concluded that the Cu photochemical oxidation rate slightly increases with the increase in UV radiation intensity. The Cu removal efficiency slightly increases for UV irradiation of 20-40 mW/cm 2 . Furthermore, the Cu removal efficiency remains unchanged for UV irradiation intensity of 40-60 mW/cm 2 .
The increase in Cu removal efficiency with the increase in UV irradiation intensity is due to the high possibility to produce HO• radicals at high UV irradiation intensity. The possible process of HO• radical formation during UV irradiation is the photolytic dissociation of water by UV irradiation at a wavelength of 254 nm [35]. However, this process is still unclear since some studies reported that the photolytic dissociation of water to HO• radicals only can be conducted at a wavelength of less than 242 nm. Deng and Zhao [12] claimed that this process occurs at a wavelength of less than 242 nm. Furthermore, Jin et al. [22] reported that the HO• radicals could be formed through irradiation on liquid water by UV light in the range of 150-200 nm. As the UV irradiation wavelength used in this study is around 253.7 nm, the increase in Cu removal efficiency with the increase in UV irradiation intensity from 20 to 40 mW/cm 2 is attributed to the formation of HO• radicals through ozone photolysis producing H 2 O 2 (Equation (8)). However, high UV irradiation intensity in the ozonation process is not good. It is true since, at high UV irradiation intensity, the ozone molecule could be degraded to oxygen [36]. Therefore, the formation of HO• radicals is low at high UV irradiation intensity. Furthermore, it is suggested that the unchanged Cu removal efficiency at a UV irradiation intensity of 60 mW/cm 2 is due to the degradation of the ozone molecule resulting in the low formation of HO• radicals.

Effect of pH
The efficiency of AOP can be influenced by various factors, such as the pH of the solution [37][38][39]. In order to study the effect of pH on the Cu removal process from electroplating wastewater through the photochemical oxidation process, the pH of solution was varied at 3, 6, 8, and 10. The result is depicted in Figure 8. with the increase in UV radiation intensity. The Cu removal efficiency slightly increases for UV irradiation of 20-40 mW/cm 2 . Furthermore, the Cu removal efficiency remains unchanged for UV irradiation intensity of 40-60 mW/cm 2 . The increase in Cu removal efficiency with the increase in UV irradiation intensity is due to the high possibility to produce HO• radicals at high UV irradiation intensity. The possible process of HO• radical formation during UV irradiation is the photolytic dissociation of water by UV irradiation at a wavelength of 254 nm [35]. However, this process is still unclear since some studies reported that the photolytic dissociation of water to HO• radicals only can be conducted at a wavelength of less than 242 nm. Deng and Zhao [12] claimed that this process occurs at a wavelength of less than 242 nm. Furthermore, Jin et al. [22] reported that the HO• radicals could be formed through irradiation on liquid water by UV light in the range of 150-200 nm. As the UV irradiation wavelength used in this study is around 253.7 nm, the increase in Cu removal efficiency with the increase in UV irradiation intensity from 20 to 40 mW/cm 2 is attributed to the formation of HO• radicals through ozone photolysis producing H2O2 (Equation (8)). However, high UV irradiation intensity in the ozonation process is not good. It is true since, at high UV irradiation intensity, the ozone molecule could be degraded to oxygen [36]. Therefore, the formation of HO• radicals is low at high UV irradiation intensity. Furthermore, it is suggested that the unchanged Cu removal efficiency at a UV irradiation intensity of 60 mW/cm 2 is due to the degradation of the ozone molecule resulting in the low formation of HO• radicals.

Effect of pH
The efficiency of AOP can be influenced by various factors, such as the pH of the solution [37][38][39]. In order to study the effect of pH on the Cu removal process from electroplating wastewater through the photochemical oxidation process, the pH of solution was varied at 3, 6, 8, and 10. The result is depicted in Figure 8. As can be seen in Figure 8, at 60 min reaction, the Cu removal efficiency obtained at pH of 3 is 46.02%, the Cu removal efficiency obtained at pH of 6 is 68.46%, the Cu removal efficiency obtained at pH of 8 is 80.09%, while the Cu removal efficiency obtained at pH of 10 is 70.46%. It can be seen that the highest Cu removal efficiency is obtained at a pH of 8. Hence, it can be concluded that the photochemical oxidation of Cu from electroplating wastewater increases with the increase in pH from 3 to 8, then it decreases with a further increase in pH to 10. As can be seen in Figure 8, at 60 min reaction, the Cu removal efficiency obtained at pH of 3 is 46.02%, the Cu removal efficiency obtained at pH of 6 is 68.46%, the Cu removal efficiency obtained at pH of 8 is 80.09%, while the Cu removal efficiency obtained at pH of 10 is 70.46%. It can be seen that the highest Cu removal efficiency is obtained at a pH of 8. Hence, it can be concluded that the photochemical oxidation of Cu from electroplating wastewater increases with the increase in pH from 3 to 8, then it decreases with a further increase in pH to 10.
The increase in Cu removal efficiency with the increase in pH value is attributed to the high rate of HO• radical formation at high pH conditions. As reported by Muniyasamy et al. [39], in the oxidation process, pH influences the process by altering the chemical nature of ozone. At low pH conditions (acidic conditions), the decomposition of the ozone molecule to produce HO• radicals is relatively slow [40]. It was reported that the decomposition rate of ozone was sluggish below the pH of 4 [41]. Furthermore, ozone tends to remain in the molecular state at acidic conditions; however, ozone can react directly as ozone radical with the contaminants at alkaline conditions [38,42]. Therefore, the Cu removal efficiency is low at a low pH value. Furthermore, the higher the pH value, the higher the Cu removal efficiency. This tendency is attributed to the high formation of HO• radicals at the alkaline conditions. It was reported that the decomposition rate of ozone in water is better at higher pH values [43]. At alkaline conditions, ozone is unstable and rapidly decomposes into HO• radical [39,40]. The HO• radical formation through ozone decomposition occurs as follows: As can be seen in Equations (16) and (17), the HO• radical, which is the primary oxidant in indirect oxidation [39], is produced at high pH level represented as OH -. Therefore, the Cu removal efficiency increases with the increase in pH value since the HO• radical formation is favored at a high pH level. However, the Cu removal efficiency decrease at a pH of 10. At higher pH value, CuO can dissolve to the aqueous phase due to the formation of soluble hydroxy and hydroxide complexes [44]. Therefore, Cu removal efficiency decreases. The possible dissolution mechanism of CuO at high pH value was proposed by Khan et al. [44] as follows:

Determination of the Kinetic Rate Order
Three classical kinetic models are proposed to describe the Cu removal process and to determine the order or kinetic rate. These kinetic models include first-order model, second-order model, and pseudo-first-order model. The compatibility of these models is evaluated by the R 2 value as reported elsewhere [45]. In this work, the kinetic study of Cu removal from electroplating wastewater is considered for various initial Cu concentration (C 0 ), ozone dosage (C O3 ), UV irradiation intensity (I UV ), and initial pH condition.
In this advanced oxidation process for Cu removal from electroplating wastewater, the mass conservation of Cu in the process can be generally expressed as: where C t (mg/L) is the concentration of Cu at time t (min) and (−r) (mg/L.min) is the rate of Cu removal. For first order model (−r = k 1 C t ), integration of Equation (20) at the initial concentration of C 0 , gives: where k 1 (1/min) is the kinetic rate constant for first order. Further simplification of Equation (21) will give the time-dependent concentration of Cu (Equation (22)).
For second-order model (−r = k 2 C t 2 ), the integration of Equation (20) will give: where k 2 (L/mg·min) is the kinetic rate constant of the second-order model. For the pseudo-first-order model (−r = k p (C t − C e )), the time-dependent concentration of Cu through the integration of Equation (20) is obtained as: where C e (mg/L) is the concentration of Cu at equilibrium condition and k p (1/min) is the kinetic rate constant of pseudo-first-order model. The kinetic parameters were determined using the least-square method. As can be seen, the obtained equation for first-order and second-order kinetic models (Equations (21) and (23)) are linear equations. Therefore, the kinetic parameters (k 1 and k 2 ) can be obtained from the linear plot relating to these equations using the linear regression method. However, the equation derived from the pseudo-first-order kinetic model is not linear. Therefore, the kinetic parameters of the pseudo-first-order kinetic model (C e and k p ) are calculated using a nonlinear regression method. On the other hand, to measure the goodness of the kinetic models proposed, the squared-correlation coefficient, R 2 , was used as the parameter [40,41]. The obtained kinetic parameters and R 2 values for first-order, second-order, and pseudo-first-order kinetic models are shown in Table 1. Table 1. Kinetic parameters of the first order, second-order, and pseudo-first-order models.  As can be seen in Table 1, all proposed models fit the data as the R 2 values are close to unity. However, compared to the other proposed models, the R 2 value of the second-order model is far enough from unity. Besides, the values of R 2 obtained are not uniform for all data. Therefore, it is suggested that the second-order model is excluded as the proposed model to describe the kinetic rate of Cu removal. Hence, the first order and pseudo-first-order models are then considered as the most suitable proposed models to describe the kinetics of Cu removal. Considering the R 2 values of the first order and pseudo-first-order models, both these two models have high goodness in describing the kinetic rate of Cu removal. However, at the initial Cu concentration, C 0 , of 145.73 and 72.86 mg/L, the value of the equilibrium concentrations, C e , obtained are zero. It indicates that at a high initial concentration of Cu in the wastewater, the kinetic rate will be increased. Furthermore, some of the C e values are relatively low. In some cases, the theoretical value of C e might be negative when the pseudo-first-order kinetic model is forcibly used [46]. In those cases, the pseudo-first-order model cannot be used to describe the kinetic rate. Obviously, if the value of the equilibrium concentration is zero or low enough, the pseudo-first-order model gets back to the first-order model [45,46]. Therefore, it is suggested that the most suitable model to describe the kinetic rate of Cu removal from electroplating wastewater through the photochemical oxidation process, combined UV-Ozone process, is the first-order model.
The kinetic study of Cu removal using an advanced oxidation process has been widely reported. However, the kinetic study of Cu removal using the photochemical oxidation process, UV-Ozone process, is still limited. The most-reported process is the electrochemical/electrocoagulation process. Al-Shannag et al. [45] reported that the electrocoagulation of heavy metals from wastewater, including Cu, followed a pseudo-first-order model. Using the same method, Vasudevan and Lakshmi [47] reported that the electrocoagulation of Cu from water follows the second-order model. Furthermore, Khattab et al. [48] reported that Cu removal through the electrochemical process follows the first-order kinetic model.

The Effect of Operating Parameters on Kinetic Rate Constant
The kinetic rate behavior of Cu removal using the photochemical oxidation process, combined UV-Ozone process, follows the first-order kinetic model as reported in Section 3.3.1. The value of the kinetic rate constant of the first-order model, k 1 , is presented in Table 1. As can be observed, the value of k 1 is varied as the variation of operating parameters, including C 0 , C O3 , I UV , and pH. The value of k 1 increases with a decrease in C 0 . Furthermore, the value of k 1 increases with the increase in C O3 , I UV , and pH. However, at a pH of 10, the value of k 1 decreases. These phenomena indicate that the value of the observed k 1 is affected by the operating parameters. This is in accordance with some previous studies that the kinetic rate constant of the advanced oxidation process, especially the UV-Ozone process, is affected by the operating parameters [24,49]. The operating parameters-dependent of the kinetic rate constant can be mathematically written as Equations (25) and (26) where ε is the pre-exponent constant, whereas a, b, c, and d are the exponent constant characteristic of C 0 , C O3 , I UV , and pH, respectively. The linear form of Equation (26) is shown as Equation (27). ln Based on Equation (27), the value of ε, a, b, c, and d can be obtained using multiple regression analysis which also has been used elsewhere [49]. However, the value of k 1 at a pH of 10 is not included in this calculation because it does not follow the tendency. Table 2 shows the result of the multiple regression analysis. As can be seen, the p-value of the coefficients obtained for each parameter is lower than 0.05. This indicates that the coefficients obtained are significant. However, the p-value of ln I UV coefficient is higher than 0.05, which indicates that this coefficient is not significant. Table 3 shows the result of the analysis of variance (ANOVA) of the multiple regression analysis. As can be observed, the value of multiple R, R 2 , and adjusted R 2 is close to unity. These indicate that the obtained regression equation fits the data. Furthermore, the obtained F-value (113.4928) is higher than the F-table or theoretical F-value. The theoretical F-value (F 0.05(4,7) ) is 6.09.
Based on the result of the multiple regression analysis (Table 2), Equation (27) can be rewritten as Equation (28). Furthermore, Equation (26) can be rewritten as Equation (29) with the value of ε constant calculated from the natural exponential of the intercept of Equation (28). The obtained value of k 1 was then compared to the calculated value of k 1 using Equation (29). The comparison of the observed k 1 and calculated k 1 is depicted in Figure 9. It is clear that the observed and calculated values of k 1 are in good fit as the R 2 is close to unity (0.9848). By combining the Equations (1), (22) and (29), the calculated Cu removal efficiency can be rewritten as Equation (30). Furthermore, the comparison of the Cu removal efficiency between the observed value and the calculated value is also depicted in Figure 10.
ln k 1 = −5.1661 − 0.7401 ln C 0 + 0.8154 ln C O 3 + 0.1441 ln I UV + 0.8407 ln pH (28)   Based on the result of the multiple regression analysis (Table 2), Equation (27) can be rewritten as Equation (28). Furthermore, Equation (26) can be rewritten as Equation (29) with the value of ε constant calculated from the natural exponential of the intercept of Equation (28). The obtained value of k1 was then compared to the calculated value of k1 using Equation (29). The comparison of the observed k1 and calculated k1 is depicted in Figure 9. It is clear that the observed and calculated values of k1 are in good fit as the R 2 is close to unity (0.9848). By combining the Equations (1), (22) and (29), the calculated Cu removal efficiency can be rewritten as Equation (30). Furthermore, the comparison of the Cu removal efficiency between the observed value and the calculated value is also depicted in Figure 10. As can be seen in Table 2 and Equation (29), the exponent characteristic value of C0 is in a negative value. It indicates that the value of the initial Cu concentration is inversely proportional to the k1 value. This finding is in accordance with the finding of Ren et al. [24]. They reported that the initial concentration of polyacrylamide in the UV-Ozone process of polyacrylamide oxidation is inversely proportional to the kinetic rate constant value. Concerning the effect of UV and ozone treatment in this oxidation process, the value of the exponent characteristic of CO3 and IUV can be used to study this matter [49]. As can be seen, the exponent characteristic value of CO3 is higher than IUV. It is suggested that the ozone dosage is more significant in this oxidation process than UV irradiation intensity. This finding is in accordance with the previous study of the UV-Ozone process. The UV-Ozone process was conducted for κ-Carrageenan treatment, and the result showed that the ozone dosage was more significant than UV irradiation intensity [49]. Furthermore, the value of the exponent characteristic of IUV is low enough, and it is close to zero. It indicates that the UV irradiation intensity is not significant enough for the Cu removal process from wastewater. In the case of the effect of pH on k1 value, it is clear that the pH has a significant effect on k1 value as the value of the exponent characteristic of pH is high. It is true because the metal oxidation process in an aqueous solution is affected by the presence of OHions, which can promote the formation of HO• radicals. On the other hand, the solubility of metal in aqueous solution is also affected by the pH value [44].

Proposed Mechanism
Based on the explanation in previous sections, the mechanism of Cu removal from electroplating wastewater through the UV-Ozone process is developed. Figure 11 shows the proposed mechanism in this study. As reported in several studies, the oxidation through AOPs is initiated by the formation of HO• radical because the main oxidating agent is HO• radical [50,51]. Based on several findings As can be seen in Table 2 and Equation (29), the exponent characteristic value of C 0 is in a negative value. It indicates that the value of the initial Cu concentration is inversely proportional to the k 1 value. This finding is in accordance with the finding of Ren et al. [24]. They reported that the initial concentration of polyacrylamide in the UV-Ozone process of polyacrylamide oxidation is inversely proportional to the kinetic rate constant value. Concerning the effect of UV and ozone treatment in this oxidation process, the value of the exponent characteristic of C O3 and I UV can be used to study this matter [49]. As can be seen, the exponent characteristic value of C O3 is higher than I UV . It is suggested that the ozone dosage is more significant in this oxidation process than UV irradiation intensity. This finding is in accordance with the previous study of the UV-Ozone process. The UV-Ozone process was conducted for κ-Carrageenan treatment, and the result showed that the ozone dosage was more significant than UV irradiation intensity [49]. Furthermore, the value of the exponent characteristic of I UV is low enough, and it is close to zero. It indicates that the UV irradiation intensity is not significant enough for the Cu removal process from wastewater. In the case of the effect of pH on k 1 value, it is clear that the pH has a significant effect on k 1 value as the value of the exponent characteristic of pH is high. It is true because the metal oxidation process in an aqueous solution is affected by the presence of OHions, which can promote the formation of HO• radicals. On the other hand, the solubility of metal in aqueous solution is also affected by the pH value [44].

Proposed Mechanism
Based on the explanation in previous sections, the mechanism of Cu removal from electroplating wastewater through the UV-Ozone process is developed. Figure 11 shows the proposed mechanism in this study. As reported in several studies, the oxidation through AOPs is initiated by the formation of HO• radical because the main oxidating agent is HO• radical [50,51]. Based on several findings obtained in this study (reported in the previous sections), the HO• radical formation takes place through three different steps. In Figure 11, these steps are represented in orange, blue, and red arrows. The orange arrows represent the HO• radical formation through the UV-Ozone process. The blue arrow represents the direct ozonation of water producing HO• radicals releasing O 2 molecules. Furthermore, the red arrows represent the ozone degradation producing HO• radicals in alkaline condition releasing O 2 molecules. The red arrow is depicted in dash-line arrow type, this indicates that this process occurs in alkaline condition only due to the presence of OH − ions. At acidic condition, this process does not occur as OH − ions are not present. obtained in this study (reported in the previous sections), the HO• radical formation takes place through three different steps. In Figure 11, these steps are represented in orange, blue, and red arrows. The orange arrows represent the HO• radical formation through the UV-Ozone process. The blue arrow represents the direct ozonation of water producing HO• radicals releasing O2 molecules. Furthermore, the red arrows represent the ozone degradation producing HO• radicals in alkaline condition releasing O2 molecules. The red arrow is depicted in dash-line arrow type, this indicates that this process occurs in alkaline condition only due to the presence of OH − ions. At acidic condition, this process does not occur as OH − ions are not present. Figure 11. Proposed mechanism of Cu removal from electroplating wastewater through UV-Ozone process.
After the formation of HO• radicals occurred, the next step is the oxidation process. This process, depicted in black arrows, includes direct and indirect oxidation since the oxidating agents in this UV-Ozone process are HO• radicals and ozone molecules. Both HO• radicals and O3 react with Cu 2+ to form solid CuO. The oxidation process of Cu 2+ to CuO by HO• radicals is called indirect oxidation, whereas the oxidation process of Cu 2+ by O3 molecules is called direct oxidation. Based on Figure 11, as the HO• radicals and O3 molecules increase in the system, the CuO product produced increases. It means that the Cu removal from the electroplating wastewater increases. However, this CuO product can dissolve in an alkaline condition. As can be seen in Figure 11, if the OH − ions are excessively present in the system, the CuO product will dissolve. This process is depicted as a dasharrow in green indicating that this process occurs in alkaline condition only due to the presence of excessive OH − ions. The dissolution of CuO to the aqueous phase in the alkaline condition is due to the formation of soluble hydroxy and hydroxide complexes [44].
Based on this proposed mechanism (Figure 11), it can be observed that the presence of the ozone molecule is important in this oxidation process. The ozone molecules have roles both to generate the HO• radicals and to oxidize the Cu directly. This is in accordance with the finding reported in Section 3.4 that the ozone dosage significantly affected the kinetic rate. On the other hand, the presence of UV irradiation assists the ozone photolysis process to produce HO• radicals through the formation of an intermediate oxidating agent, H2O2. Furthermore, the pH is also important in this mechanism. As can be seen in Figure 11, suitable alkaline condition accelerates the formation of HO• radicals. However, the high alkaline condition has a negative effect on the Cu removal process due to the CuO dissolution process in high alkaline condition. The excessive OH − ions catch the solid CuO to form soluble hydroxy and hydroxide complexes decreasing the Cu removal efficiency.

Conclusions
The photochemical oxidation process of Cu from electroplating wastewater has been investigated. It is shown that the UV-Ozone process has high performance on Cu removal compared to UV and Ozone processes due to the high production rate of HO• radicals as the oxidant and due Figure 11.
Proposed mechanism of Cu removal from electroplating wastewater through UV-Ozone process.
After the formation of HO• radicals occurred, the next step is the oxidation process. This process, depicted in black arrows, includes direct and indirect oxidation since the oxidating agents in this UV-Ozone process are HO• radicals and ozone molecules. Both HO• radicals and O 3 react with Cu 2+ to form solid CuO. The oxidation process of Cu 2+ to CuO by HO• radicals is called indirect oxidation, whereas the oxidation process of Cu 2+ by O 3 molecules is called direct oxidation. Based on Figure 11, as the HO• radicals and O 3 molecules increase in the system, the CuO product produced increases. It means that the Cu removal from the electroplating wastewater increases. However, this CuO product can dissolve in an alkaline condition. As can be seen in Figure 11, if the OH − ions are excessively present in the system, the CuO product will dissolve. This process is depicted as a dash-arrow in green indicating that this process occurs in alkaline condition only due to the presence of excessive OH − ions. The dissolution of CuO to the aqueous phase in the alkaline condition is due to the formation of soluble hydroxy and hydroxide complexes [44].
Based on this proposed mechanism (Figure 11), it can be observed that the presence of the ozone molecule is important in this oxidation process. The ozone molecules have roles both to generate the HO• radicals and to oxidize the Cu directly. This is in accordance with the finding reported in Section 3.4 that the ozone dosage significantly affected the kinetic rate. On the other hand, the presence of UV irradiation assists the ozone photolysis process to produce HO• radicals through the formation of an intermediate oxidating agent, H 2 O 2 . Furthermore, the pH is also important in this mechanism. As can be seen in Figure 11, suitable alkaline condition accelerates the formation of HO• radicals. However, the high alkaline condition has a negative effect on the Cu removal process due to the CuO dissolution process in high alkaline condition. The excessive OH − ions catch the solid CuO to form soluble hydroxy and hydroxide complexes decreasing the Cu removal efficiency.

Conclusions
The photochemical oxidation process of Cu from electroplating wastewater has been investigated. It is shown that the UV-Ozone process has high performance on Cu removal compared to UV and Ozone processes due to the high production rate of HO• radicals as the oxidant and due to the direct oxidation by ozone. Furthermore, this process produced a CuO monoclinic crystal phase as a solid product. The initial Cu concentration, ozone dosage, and pH value have a significant effect on Cu removal efficiency. The initial Cu concentration significantly reduces Cu removal efficiency. As expected, the ozone dosage increases the Cu removal efficiency since it is responsible for HO• radical formation and direct oxidation. Furthermore, the Cu removal efficiency increases with the increase in pH value from 3 to 8. However, a further increase in pH value reduces the Cu removal efficiency due to the formation of soluble hydroxy and hydroxide complexes of Cu at the alkaline conditions. Interestingly, it is found that the UV irradiation intensity is not significant in the photochemical oxidation process on Cu from electroplating wastewater. However, it is responsible to promote the ozone photolysis producing HO• radicals through H 2 O 2 formation. It is found that the kinetic behavior of the photochemical oxidation of Cu follows the first-order kinetic model. Furthermore, the relation between the operating parameters and kinetic rate constant is also established. In addition, the mechanism of Cu removal through the UV-Ozone process was also proposed concerning the findings obtained in this study.