Synthesis of Cu-Doped TiO₂ Nanocatalyst for the Enhanced Photocatalytic Degradation and Mineralization of Gabapentin under UVA/LED Irradiation: Characterization and Photocatalytic Activity

Abstract

Light-harvesting of titanium oxide (TiO₂) was enhanced by copper (Cu) doping, and its performance was evaluated by gabapentin (GBP) degradation under UVA-LED irradiation. The morphology and structure of TiO₂ and Cu-TiO₂ were characterized using XRD, FTIR, FE-SEM, EDX, TEM, PL, DRS, and BET analysis. The complete degradation of 10 mg/L GBP was obtained in the developed photocatalytic process under the optimal conditions: catalyst loading, 0.4 g/L; pH solution, 8; and reaction time, 20 min. The reactive species trapping was studied to identify the degradation mechanism in this system. Among the water matrix experiments, phosphate (PO₄³⁻) anion indicated an inverse effect in increasing efficiency. Finally, the main intermediates generation during the GBP degradation was investigated based on LC-MS analysis, and a decomposition pathway was proposed. Accordingly, doping TiO₂ with Cu resulted in the development of a UVA-activated photocatalyst for efficiently degrading and mineralizing GBP as a model of a pharmaceutical compound.

1. Introduction

Nowadays, emerging pollutants are widely consumed, but despite their beneficial effects, their adverse efficacy on human health and the environment cannot be denied [1]. The pharmaceutical industries produce a large number of medicinal materials for the prevention and therapy of human illnesses, and thus water resources are polluted by industrial wastewater. Additionally, the disposal of these therapeutic materials from the human body into the sewage will be another way of entering these micro-pollutants into aqueous environments [2,3]. Gabapentin (GBP) is widely used as an analgesic for kidney disorders and for treating neuropathic pain and epilepsy [4,5]. GBP has been detected at concentrations of up to 10 ηg/L, along with other drugs in raw wastewater, but the noteworthy point is that its destruction does not take place entirely, and no-metabolite structure can be dispersed in various solutions [6,7]. Additionally, this drug is ranked as an emerging micro-pollutant with adverse effects on human health [6,8]. Hence, the main issue is developing an effective method for treating contaminated water with different pharmaceutical pollutants. The selection of the heterogeneous photocatalytic oxidation (PCO) method based on the advanced oxidation technique is an attractive technology that has been applied in the degradation of various recalcitrant organic pollutants [9,10]. In the PCO process, photo-generated intermediates, such as electron/hole pairs, can enhance photoreaction under light irradiation. Many PCO processes have been studied to find effective photocatalysts for the decomposition of a wide range of contaminants [11]. Among the various semiconductors as photocatalysts, TiO₂ has been most highlighted due to its remarkable properties, including great oxidation potential, chemical durability, nontoxicity, a high proportion of surface to volume, and economical trait [12,13]. Nevertheless, the high recombination rate of photo-induced charge carriers and poor ionic and electrical transmission are the significant intrinsic disadvantages of unmodified TiO₂ [14]. As a solution, doping semiconductors with transition metal and nonmetal or coupling with other materials have been proposed in previous studies [2,15,16,17,18]. Doping TiO₂ with metals has shown high photocatalytic activity that enhanced the separation capability of e⁻/h⁺ pairs and reduced the band gap energy compared to pure TiO₂ [14,19]. For instance, the application of copper (Cu₂O, Cu²⁺, CuO, etc.) as a dopant to improve the photocatalytic activity of TiO₂ has been efficiently practiced for the degradation of contaminants and is applicable due to its cost-effectiveness and abundance [20,21]. It has been reported that when a semiconductor is doped with copper, it can create an extra energy level, leading to increasing photocatalytic activity [22,23]. Moreover, Vaiano et al. reported that when Cu was introduced to ZnO, the band gap energy decreased from 3.2 to 2.92 eV, resulting in decreasing the recombination rate of e^-/h⁺ pairs, thus the photocatalytic performance improved [24].

In this context, the Cu-TiO₂ material was prepared and characterized as an improved form of TiO₂ catalyst for the degradation and mineralization of Gabapentin under UVA irradiation emitted from LEDs. LEDs are environmentally benign, have a long life span, and provide flexibility in designing the photoreactor; therefore, they are considered an appropriate source of UV photons to drive the PCO [25]. Accordingly, the effect of the main experimental parameters of pH, catalyst concentration, water anions, radical scavengers, and reaction time was investigated on the degradation of Gabapentin. A pathway was also proposed for the degradation of Gabapentin in the developed photocatalytic process operated under optimal conditions.

2. Results and Discussion

2.1. Characterization of As-Prepared TiO₂ and Cu-TiO₂

The crystal structure of plain TiO₂ and Cu-TiO₂ products are shown in Figure 1a using XRD analysis. The characteristic peaks of (101), (004), (200), (105), (211), (204), (116), (220), and (215) planes are identified in the XRD pattern of both TiO₂ and Cu-TiO₂ products that can be attributed to the anatase phase of TiO₂ based on (JCPDS Card No. 73–1764). Additionally, the diffraction peaks of (101), (110), (111), and (220) planes are observed in the XRD pattern of the Cu-TiO₂ materials, which corresponds to the rutile phase of the TiO₂ structure (JCPDS Card No. 78–1510). On the other hand, Cu-TiO₂ with mixed phases (anatase and rutile) was prepared through the designed synthesizing method. Any characteristic peak is observed in the XRD pattern of Cu-TiO₂ product for CuO, suggesting that Cu might be doped in the lattice of the TiO₂ [26]. After Cu doping, the peak broadening in the XRD pattern qualitatively described the change in particle size and lattice expansion [27]. The crystallite size of TiO₂ and Cu-TiO₂ was estimated from the XRD pattern achieved using the Scherrer equation. The crystallite size of pure TiO₂ and Cu-TiO₂ was 29.3 and 45.6 nm. This matter indicates that introducing Cu into TiO₂ increased the crystallite size of Cu-TiO₂. The literature review [27,28] reported that the crystallite size of Cu-TiO₂ was slightly increased due to the presence of Cu²⁺ and CuO species. The different ionic radii of Cu and Ti ions caused lattice distortion and strain field, changing the size of particles of Cu-TiO₂.

The FTIR spectra of TiO₂ and Cu-TiO₂ are displayed in Figure 1b. The FTIR spectrum of the catalyst shows absorption peaks between 400 and 600 cm⁻¹, which is ascribed to Ti-O-Ti stretching. In addition, other peaks of about 1623 and 2347 cm⁻¹ are related to C = C and C = O functional groups, respectively. The medium peaks at about 3485 and 3500 cm⁻¹ confirm the presence of OH bands in the sample structure due to surface adsorb water and hydroxyl agent. The vibration bands are observed between 1000 to 1250 cm⁻¹ after Cu-doped TiO₂ indicated lattice vibration of TiO₂ [29,30,31,32].

The surface morphology of the as-prepared TiO₂ and Cu-TiO₂ materials was investigated by FESEM, and the elemental distribution was evaluated based on EDX and mapping analyses; the results are presented in Figure 2a–f for Cu-TiO₂ and in Figure S1 for TiO₂. FESEM images confirmed the almost spherical shape of the Cu-TiO₂ product with uniform size distribution. Therefore, doping with Cu affected the morphology of TiO₂. Norris et al. [33] reported that the photocatalyst’s morphology, size, and electronic structure could be modified by doping with copper. The EDX analysis demonstrated the presence of considerable percentages of Ti, O, and Cu in the structure of the as-prepared Cu-TiO₂ catalyst. In addition, the distribution of elements with colored dots in the mapping images (Figure 2d–f) confirms the presence of Cu in the Cu-TiO₂ product. The EDX mapping shows that the Cu-TiO₂ seems to be homogenous. To further examine the morphology of the prepared materials, the TEM analysis was conducted, and images of TiO₂ and Cu-TiO₂ can be observed in Figure 2c. The average particle sizes of TiO₂ and Cu-TiO₂ determined using Digimizer software based on the TEM images were around 35 and 47 nm, respectively. The results confirmed that doping TiO₂ with Cu increases the size of the particles. Tasbihi et al. [34] reported that Cu-TiO₂ was fully crystallized at lower doping concentrations, but at higher concentrations of Cu (15 wt.%), the prepared catalyst had amorphous phases.

BET analysis estimated the surface areas of prepared TiO₂ and Cu-TiO₂ to be 52.1 and 9.9 m²/g, respectively (Figure 3). It can be related to changes in the morphology and crystalline structure of the TiO₂ after doping with Cu [30]. This result agreed with Tasbihi et al. [34] and Bensouici et al. [22], who reported a decrease in the surface area of TiO₂ after doping with Cu. From the pore size distribution curve, Figure 3a,b (inset), and IUPAK categorization, both catalysts’ isotherms were observed to be type IV and mesoporous structures [35].

The optical properties of pure TiO₂ and Cu-TiO₂ products were measured using the UV-Vis DRS analyses, and the results are given in Figure 4. Based on Figure 4a, the role of Cu doping on the spectra profile of Cu-TiO₂ is visible by shifting the absorption edge of TiO₂ to the higher wavelengths upon doping with Cu. It can be related to the surface plasmon resonance properties of copper [36], resulting in increased light absorption potential of the product and therefore in the accelerated generation of electron-hole pairs under light irradiation. Moreover, the band gap energy (E_g) of TiO₂ and Cu-TiO₂ was determined using the Tauc/David–Matt method [37]. According to Figure 4b, the E_g values of TiO₂ and Cu-TiO₂ were estimated to be about 3.2 and 2.9 eV, respectively. The results indicate that doping TiO₂ with Cu could decrease its band gap energy, reconfirming the higher light absorption potential and thus the photocatalytic activity in the Cu-TiO₂ material compared to that in the plain TiO₂. Tasbihi et al. [34] observed that pure TiO₂ exhibited E_g of 3.05 eV, while the E_g of Cu/TiO₂ was lower due to the presence of Cu²⁺ ions in the structure of TiO₂ lattice. Additionally, they [34] reported that low Cu concentration (0.2 wt.%) could not decrease the E_g of Cu-TiO₂. Sahu and Biswas [27] reported that the E_g of TiO₂ decreased from 3.31 to 2.51 eV with increasing the Cu concentration at the highest dopant concentration (15 wt.%). This change in E_g was related to incorporating Cu²⁺ ions into the TiO₂ structure and forming CuO as a layer on the TiO₂ particle surface [27]. The change in the optical absorption of Cu-TiO₂ was attributed to the defect centers created by the replacement of Ti⁴⁺ by Cu²⁺ atoms in the TiO₂ crystal lattice. A previous study [38] showed that doping TiO₂ with aliovalent ions could change the local lattice symmetry and defect characteristics, changing the absorption properties. In Cu-TiO₂, when Cu⁺² ions are located inside or on the surface sites of TiO₂, a rearrangement of the neighbor atoms to compensate for the charge deficiency is caused by lattice deformation. The lattice deformation influences the electronic structure of Cu-TiO₂ and shifts the E_g. Therefore, these Cu-doped materials can be applied for diverse photocatalytic applications, which have been reported in numerous other studies.

The PL spectra of the pure TiO₂ and Cu-TiO₂ samples were investigated to comprehend the separation of e⁻/h⁺ pairs and the charge–carrier transfer efficiency. Figure 5 shows a significant peak at around 388 nm in pure TiO₂, which is assigned to the band-to-band recombination because it is near-band-edge luminescence. The peak in PL spectra of Cu-TiO₂ is observed at 376 nm, but with a drop in intensity. This intensity reduction refers to the inhibition of the recombination of the photo-generated electron from the conduction band (CB) to the valence band (VB) of TiO₂. The higher PL peak intensity in pure TiO₂ refers to the reduction of the lifetime of the electron-hole pairs due to the fast recombination rate. Reda et al. [39] reported the highest reduction in PL intensity of doped-TiO₂ with Cu and N, revealing the examined samples’ efficacy in reducing photo-generated electron-hole recombination compared with pure TiO₂. The intensity of the light absorption of TiO₂ was also decreased upon doping with Cu, implying that the rate of electron-hole recombination decreased in the doped product. It might result in the formation of higher reactive radical species and, thus, higher degradation performance in the developed photocatalytic process [40]. Therefore, it can be concluded that doping with copper significantly modified the properties of TiO₂, which increased its photoactivity [41].

2.2. The Photocatalytic Activity of Cu-Doped TiO₂

2.2.1. Effect of Initial pH Solution

The solution pH is an essential factor effective in GBP removal in the photocatalytic process. Moreover, the solution pH affects photocatalytic activity through two mechanisms. Firstly, the solution pH influences the extent of pollutant ionization and, therefore, the ionizable compounds’ adsorption on the photocatalyst’s surface. On the other hand, the ionizable compound’s absorption rate could control the photocatalytic degradation amount [42]. The removal of pollutant compounds affected by pH in photocatalytic processes depends on the solution pH, compound pKa, and catalyst pHpzc. The interaction between these factors promotes or inhibits the removal rate of the desired compound. Additionally, the literature review showed that change in the solution pH had altered the adsorption capacity of pollutants on the photocatalyst surfaces [43]. The experiment was evaluated under various pH levels (4–9), while the GBP concentration, catalyst dosage, and reaction time were considered to be 10 mg/L, 0.5 g/L, and 20 min, respectively. As shown in Figure 6a, the complete photo-activity was obtained at pH 8 and 9 while it was reduced at acidic pH. Additionally, it is interesting that in all pH values, the GPB degradation was higher than GBP adsorption. GBP is a compound with an isoelectric point at pH 7.14 and has two pKa values of 3.68 for the carboxylic acid and 10.7 for the primary amine [44]. On the other hand, the pHpzc value of prepared Cu-TiO₂ was estimated at about 7.32, which depicts that the catalyst surface had a negative charge higher than pHpzc. As a result, at pH 8 and 9, electrostatic attraction between GBP molecules (positive) and Cu-TiO₂ photocatalyst (negative) enhanced the destruction efficiency. Therefore, pH 8 was selected for the subsequent reaction. Nevertheless, at a pH value lower than 7, electrostatic repulsion is boosted due to the positive charge of GBP and as-made catalyst, resulting in low photocatalytic activity [45]. Ahmad and Yasin [46] investigated the Cu/TiO₂/bentonite composite for the photocatalytic degradation of deltamethrin. The results show that a maximum degradation rate was observed at pH 12. They concluded that modifying pH is an effective strategy to improve photocatalytic performance. Sharma et al. [47] observed that the degradation rate of tetracycline with Cu₂O-TiO₂ nanotube was complete under acidic, neutral, and alkaline conditions, and it is related to the adsorption of pollutants on the surface catalyst.

2.2.2. Effect of Cu-TiO₂ Concentration

Catalyst concentration directly influences the degradation efficiency. Figure 7 depicts GBP degradation under reaction conditions of pH = 8, reaction time = 20 min and pollutant concentration 10 mg/L with different catalyst dosages (0.1–0.5 g/L) in Cu-TiO₂/UVA process. Without the Cu-TiO₂, the removal efficiency was about 21%, but after the addition of the catalyst, the degradation rate was enhanced. In general, the degradation efficiency of GBP is enhanced with an increased concentration of Cu-TiO₂. As can be seen, removal efficiencies were obtained to be 49.8, 56.2, 69.3, 100, and 100% for 0.1, 0.2, 0.3, 0.4, and 0.5 g/L of the as-prepared catalyst, respectively. These results prove that the degradation yield enhanced with increasing Cu-TiO₂ loading to 0.4 g/L. Therefore, the optimal concentration of Cu-TiO₂ for the following experiments was chosen as 0.4 g/L.

The number of active sites on the catalyst surface increased, which, in turn, boosted the formation of main reactive species, such as electron-hole pairs or radical hydroxyl during the photo process. Additionally, catalyst concentrations higher than 0.5 g/L tend to accumulate in a system where the penetration of light decreases, resulting in no change in the removal efficiency [48]. It is worth saying that adsorption was evaluated along with GBP photo-degradation, and increasing catalyst loading reduced the adsorption rate, indicating the role of light on the Cu-TiO₂ activation [49]. Reda et al. [38] reported that 0.14 g/L of Cu-TiO₂ was the optimal concentration for dye removal. By increasing the amount of photocatalyst up to the optimal level, the reaction rate due to more active sites was increased. Additionally, when the Cu-TiO₂ dosage was increased beyond the optimal concentration, the aggregation of the suspended particles took place, and this phenomenon reduced the amount of light reaching the active sites of the catalyst, and subsequently, the rate of degradation decreased.

2.2.3. Catalytic Activity and Mechanism of GBP Degradation in the Cu-TiO₂/UVA Process

The catalytic activity of TiO₂ and Cu-TiO₂ materials were studied after selecting the optimal conditions, and the results are presented in Figure 8a. As-made Cu-TiO₂ catalyst exhibited poor adsorption toward GBP molecules under dark conditions. When no catalyst was introduced into the reactor, the GBP removal was increased by 21% under light irradiation. In addition, the pure TiO₂ indicated poor degradation efficiency of about 41% under the UVA-LED system compared to Cu-TiO₂. By utilizing the as-prepared Cu-TiO₂ catalyst, the degradation of GBP molecules significantly improved and complete removal was achieved. In addition, the kinetics of the removal of GBP in the selected processes was evaluated based on the pseudo-first-order (PFO) reaction model [50], and the plots are

$$ln\frac{C_0}{C_t}=k_{obs}t$$

(1)

in which C₀ and C_t are the initial and residual concentrations of GBP at time t (min) of the reaction, and k_obs is the observed PFO rate constant values for each initial concentration that was obtained from the slope of plotting ln (C₀/C_t) vs. reaction time.

The PFO rate constants of GBP removal in the UVA, TiO₂/UVA, and Cu-TiO₂/UVA processes were 0.004, 0.029, and 0.122 min⁻¹, respectively. It is seen that the degradation by direct photolysis was inefficient for the removal of GBP under the selected conditions. In addition, the removal rate of GBP in the photocatalytic process with Cu-TiO₂ was over four times that with plain TiO₂, implying that doping TiO₂ considerably improved its photocatalytic activity for the removal of GBP. The improvement of photocatalytic activity of TiO₂ upon doping with Cu can be related to the decrease in band gap and the recombination rate of the electron/hole. Indeed, Cu dopant as a free electron trap could reduce the recombination rate of electron-hole pairs, improve the light-harvesting capacity_, and accelerate the photo-activity of modified TiO₂, which is consistent with DRS and PL analysis. Additionally, in the presence of an appropriate light source, electrons are excited from the VB to the CB of TiO₂ to create VB holes (positive charge carriers) and CB electrons (negative charge carriers). By scavenging the oxidizing species, these electron-hole pairs produce different reactive oxygen (ROS), further degrading the GBP on the surface of the Cu-TiO₂ [47].

To understand the mechanisms involved in the removal of GBP in the Cu-TiO₂/UVA process, the GPB removal was compared in the absence and presence of several scavengers, and the related results are reported in Figure 9. The radical scavengers were tert-butanol (TBA) as the scavenger of ${\mathrm{HO}}^{•}$, p-benzoquinone (pBQ) as the scavenger of ${\mathrm{HO}}^{•}$ and ${\mathrm{O}}_2^{•-}$, sodium azide as the scavenger of ${\mathrm{HO}}^{•}$ and ¹O₂, and oxalate as the scavenger of h⁺ [34]. As seen in Figure 9, the removal of GBP was less affected in the presence of TBA, pBQ, and sodium azide, but it was strongly suppressed in the presence of oxalate. It indicates that photo-generated holes played the main role in GBP oxidation, which could be quenched by water molecules or oxygen atoms to enhance the formation of hydroxyl and superoxide radicals, respectively [51].

Regarding the FTIR curve given above, the presence of the functional group (TiO₂-O-Cu) confirmed the CuO formation that we expected: the CuO-TiO₂ heterojunction led to the best performance of modified TiO₂ [13,51]. Therefore, the low band gap of CuO (1.7 eV) compared to TiO₂ (3.2 eV) is a reason for driving photo-generated electrons from the CB of metallic copper to TiO₂, and at the same time, photo-induced holes jumped from the VB TiO₂ to VB CuO [13]. On the other hand, the synergistic role between anatase–rutile heterophase in the as-prepared catalyst induced the separation of electron–hole pairs accelerated for the greater transfer of photo agents in the CB and VB of Cu-TiO₂ nanoparticles (Equations (2)–(5)) [30].

$$\mathrm{Photocatalyst} \left({\mathrm{Cu-TiO}}_2\right)+\mathrm{h}\mathsf{ʋ} \to \mathrm{Photocatalyst} (e_{CB}^{-}+h_{VB}^{+})$$

(2)

$$e_{CB}^{-}+O_2\to O_2^{•-}$$

(3)

$$h_{VB}^{+}+H_2\mathrm{O}\to H{O}^{•}$$

(4)

$$\mathrm{GBP}+h_{VB}^{+}/H{O}^{•}/O_2^{•-} \to \mathrm{degradation} \mathrm{products}$$

(5)

Sharma et al. [47] reported that the main reactive species involved in tetracycline degradation in the photocatalytic process using Cu₂O-TiO₂ was the superoxide radical (O₂^•−), followed by ${\mathrm{HO}}^{•}$ and valence band holes (VB_h+). The effect of radical scavengers on carbamazepine (CBZ) degradation was investigated in the CuWO₄-TiO₂ process [52]. The results showed that the mechanism for the degradation of CBZ was the surface charge process driven by ${\mathrm{HO}}^{•}$, O₂^•−, the generation of VB_h+, and CB_e−. Under an appropriate light source, the excitation process can generate holes in the VB and electrons in the CB accompanied by redox reactions such as Cu⁺²/Cu⁺. The generated electrons of the CB of TiO₂ drift into the CB of CuWO₄. This phenomenon suppresses electron-hole recombination and enhances CBZ degradation. Besides electron-hole generation, ROS production, including O₂^•− and ${\mathrm{HO}}^{•}$, becomes available to drive the photodegradation activity [52].

2.2.4. Effect of Water Matrix on GBP Photo-Degradation

To investigate the as-prepared catalyst in practical applications, the presence of common water anions, such as chloride (Cl⁻), nitrate (NO₃⁻), carbonate (CO₃²⁻), phosphate (PO₄³⁻), and sulfate (SO₄²⁻), were evaluated on the GBP decomposition in the Cu-TiO₂/UVA process. Most studies have reported that anions have an inhibitory role in the photocatalytic processes due to (1) the competition between the anions and the target pollutant toward the available active sites on the catalyst surface; (2) the radicals’ scavenging properties of the anions, which caused the formation of ionic radicals such as Cl^•−, NO₃^•−, and HCO₃^•− with less degradation potential; and (3) the hole scavenging proprieties of anions [52]. Therefore, the experiment was conducted with no anion sample. As can be seen in Figure 10, the addition of PO₄³⁻ and SO₄²⁻ decreased the efficiency to 49 and 91%, respectively, under conditions of pH = 8, catalyst concentration = 0.4 g/L, GBP concentration = 10 mg/L, and anions = 1 mM, during 20 min. Based on the results shown in Figure 10, a negligible negative impact was observed after adding Cl⁻, NO₃⁻, and CO₃²⁻ compared with the control sample on the GBP removal in the Cu-TiO₂/UVA photocatalytic process, indicating no considerable interaction between these anions, and the reactive species contributed in removing GBP under the selected conditions. PO₄³⁻ and SO₄²⁻ at the higher levels had a limiting effect on GBP removal in the Cu-TiO₂/UVA photocatalytic process. It can be related to the interaction of these anions with the surface of the catalyst and occupying the active sites of the catalyst, resulting in a decrease in the GBP degradation rate (Equations (6)–(8)) [53,54]. The introduction of SO₄²⁻ decreased CBZ degradation, and a possible reason was the adherence of SO₄²⁻ to the surface of TiO₂ via Van der Waals forces and hydrogen bonds and perhaps the displacement of the surface hydroxyl group of TiO₂ via the ligand exchange mechanism. Additionally, SO₄²⁻, due to the double charge, has a high adsorption ability onto the surface of the catalyst and a high scavenging affinity to ${\mathrm{HO}}^{•}$, and these reasons decreased the GBP degradation rate as compared to the other anions [52].

$$P{O}_4^{3-}+H_2\mathrm{O}\to HP{O}_4^{2-}+O{H}^{-}$$

(6)

$$HP{O}_4^{2-}+O{H}^{•}\to O{H}^{-}+H_2+HP{O}_4^{•-}$$

(7)

$$S{O}_4^{2-}+h_{VB}^{+}\to S{O}_4^{•-}$$

(8)

2.2.5. Mineralization and Pathway Study of GBP Degradation in the Cu-TiO₂/UVA Process

To evaluate the mineralization efficiency of GBP molecules in the Cu-TiO₂/UVA photocatalytic process, the TOC concentration was measured when the process was operated under optimum conditions. According to the obtained result (Figure 11a), the removal of TOC increased with the reaction time and reached 63% at the reaction time of 60 min. In addition, the PFO removal rate of TOC under the selected condition was determined to be 0.016 min⁻¹ (Figure 11b). These results clearly indicate that the developed process efficiently mineralized GBP. The high degree of TOC removal at 60 min can be demonstrated that the GBP molecules were changed to simple materials, which might be completed by continuing the reaction time. The mineralization of adsorbed GBP on Cu-TiO₂ is possible either by direct oxidation of GBP on the catalyst surface or by produced hydroxyl radicals (Equation (9)).

$$H{O}^{•}+RH\to R^{•}+H_{2}O\to C{O}_2+H_{2}O+mineral acids$$

(9)

RH and R^• are organic substances and organic radicals, respectively. Evgenidou et al. [55] examined the mineralization of a mixture of eight antibiotics using Cu-TiO₂ photocatalysts. Against complete degradation of antibiotic compounds, the reduction in TOC after 30 min was not more than 20%, indicating the formation of intermediate products. There is a need for a long time in photocatalyst processes to achieve complete mineralization, approving the formation of recalcitrant intermediates. In their study, defluorination, the release of nitrite and nitrate ions, and desulfurization are the main steps in the mineralization of antibiotic compounds.

The difference between GBP degradation and mineralization rate proved the formation of some intermediates identified in the reaction solution. For this purpose, the degradation intermediates of the GBP were conducted by LC-MS spectra under the optimal condition at 60 min. Accordingly, four transformation intermediates were measured based on the m/z values, and the degradation pathway was proposed in Figure 12. Gabapentin is unstable and can be spontaneously converted to gabapentin lactam (m/z 154) during dehydration [56]. Then, the deoxygenation of the C-O bond resulted in forming an intermediate with m/z 128. The recorded fragment of an intermediate, such as cyclohexane (m/z 84) and isopropyl radical (m/z 43), is related to the presence of reactive species (${\mathrm{h}}_{\mathrm{VB}}^{+}/{\mathrm{HO}}^{•}/{\mathrm{O}}_2^{•-})$ that destroyed the N-C, C-C, and N-H bonds to low molar mass. Finally, simple products, such as ${\mathrm{CO}}_2$ and ${\mathrm{NH}}_4^{+}$, were observed in this reaction, confirming the effective mineralization of GBP in the photocatalytic process with an as-made Cu-TiO₂ catalyst. Dal Bello et al. [57] investigated GBP degradation in the presence of a TiO₂ photocatalyst. The degradation of GBP (m/z 172) consisted of several stages. Firstly, the loss of a water molecule at m/z 154; secondly, simultaneous loss of NH₃ (m/z 137) and carbon monoxide (m/z 126); and finally, losses of water and carbon monoxide.

3. Materials and Methods

3.1. Chemicals

Gabapentin (molecular formula: C₉H₁₇NO₂) was obtained from Local Pharmaceutical. Titanium isopropoxide (C₁₂H₂₈O₄Ti), isopropyl alcohol (CH₃O₈), acetic acid (CH₃COOH), copper acetate Cu(CH₃COO)₂, potassium dihydrogen phosphate (KH₂PO₄), dipotassium dihydrogen phosphate (K₂HPO₄), HPLC grade water and acetonitrile were purchased from the local market. In this work, all solutions were supplied with deionized water.

3.2. Synthesis and Characterization of Cu-TiO₂ Nanoparticles

First, 10 mL of isopropyl alcohol and 3 mL of acetic acid were mixed in a glass container, and then 2.5 mL of titanium isopropoxide as a precursor was added dropwise into the above solution. The reaction was carried out at an ambient temperature and stirred at 80 rpm for 30 min. Then 2 mL of distilled water was added drop by drop to the mixture under the hood. Finally, after 3 to 5 min mixing, the obtained gel was dried at 100 °C and calcined at 500 °C for 2 h under an air atmosphere. To prepare the Cu-doped TiO₂, 0.319 g copper acetate was dissolved in 10 mL of distilled water; the obtained solution was added dropwise to the titanium dioxide gel under mixing, and the prepared gel was calcined under conditions similar to the plain TiO₂.

The crystalline phase of the prepared catalyst was detected based on the X-ray diffraction (XRD) analysis in the 2θ scanning range from 10 to 80°, and Cu as anode material with ʎ radiation of 1.54 Å. The surface functional groups were identified by a Fourier transform infrared spectroscopy (FTIR) in the frequency range of 400–4000 cm⁻¹. The surface structure and elemental distribution of as-the prepared catalyst were determined by field emission scanning electron microscopy (FE-SEM) coupled with energy-dispersive X-ray spectroscopy (EDX). The actual size of Cu-TiO₂ was obtained with transmission electron microscopy (TEM) using Philips CM30. The specific surface area was characterized by Brunauer–Emmett–Teller (BET) in the condition of nitrogen adsorption/desorption at 77 °k. Energy band gap and photoelectronic quality were evaluated by differential reflectance spectroscopy (DRS) and photoluminescence (PL) methods, respectively. Finally, the identification of the intermediate through the gabapentin photodegradation was conducted by liquid chromatography-mass spectrometry (LC-MS) analysis.

3.3. Experimental and Analytical Procedures

The photodegradation efficiency of the as-prepared catalyst for gabapentin removal was examined under UVA-LED irradiation in a batch-quartz photoreactor with an inner diameter and volume of 10 cm and 50 mL, respectively. A setup equipped with 18 LEDs (3 W, SUN LED, Seoul Semiconductor, Korea) and emitted UVA photons (365 nm) was applied as a light source. Generally, a given range of catalysts was dispersed into the reactor with GBP solution (10 mg/L in all tests) at the target pH, stirred at an interval determined times for degradation of Gabapentin by Cu-TiO₂ under UVA-LED irradiation. Then 5 mL of sample was filtered through a 0.2-micron syringe filter and the residual concentration of the contaminant was monitored by high-performance liquid chromatography (HPLC, Eclipse Plus C18 column; 3.5 µm, 4.6 × 100 mm, Agilent) with UV detection at 210 nm. The mobile phase consisting of phosphate buffer (20 mM, pH = 6.2) and 25% acetonitrile, was conducted at a 1 mL/min flow rate [5]. The removal efficiency of the GBP was calculated using the following equation:

$$\mathrm{Removal} \mathrm{efficiency} (\%)=\frac{C_0-C_t}{C_0}\times 100$$

(10)

which C₀ and C_t are the initial and final concentrations of GBP, respectively. All the experiments were conducted in duplicate, and the mean of the results was reported. The standard deviation was within ±5%.

4. Conclusions

Doping Cu-TiO₂ was successfully performed by the sol-gel approach for GBP decomposition under the UVA-LED system. Based on the DRS technique, the band gap of the as-made catalyst was calculated at about 1.48 eV, and the pL results indicated a low recombination rate of charge carriers compared to pure TiO₂ nanoparticles. The Cu-TiO₂ particles/UVA-LED system indicated high GBP removal efficiency with a reaction rate constant of about 0.122 min⁻¹ under optimal conditions. In addition to investigating the common parameters (catalyst concentration, solution pH, and reaction time), the effect of scavengers and water matrix were also checked on the trend of GBP removal under the photocatalytic process. The order of the negative effect of the water matrix on GBP degradation was PO₄³⁻ > SO₄²⁻, and other anions did not affect the reaction. Scavenger experiments revealed that the h⁺ and ${\mathrm{HO}}^{•}$ had a significant role in the reaction mechanism of the Cu-TiO₂ particles/UVA-LED system. Finally, the mineralization rate by TOC measurement was obtained at about 63% at 60 min reaction time, and the identified intermediates suggested that the pollutant converted to simple compounds. As a result, the as-made Cu-TiO₂ nanocomposite showed high photocatalytic properties under UVA irradiation; thus, the Cu-TiO₂/UVA process was confirmed under the bench-scale condition as a promising and efficient method for the removal of pharmaceutical compounds from the contaminated water and wastewater.