Bandgap tuning of TiO₂ by Cu nanoparticles applied in photocatalytic antifouling-coated PES membranes through PAA-plasma grafted adhesive layer

Membrane filtration has been studied and applied in practice for a long time in the filtration and separation of water and wastewater (Sójka-Ledakowicz et al. 1998). Membranes are used in one or more stages of the water treatment process and have proven to be effective in terms of economy, performance, environment, etc. (Kamali et al. 2019). However, an unavoidable limitation of membrane filtration is the phenomenon of membrane fouling during the long-term operation of membranes. This reduces the performance of membranes and forces operators to clean fouled membranes frequently or even replace them with new ones (Shi et al. 2014). Measures such as controlling the membrane operating mode or incorporating membrane pretreatment methods make significant changes to the water treatment system, requiring more space and cost, but have proved inefficient (Le-Clech et al. 2006). Another approach that seems reasonable and is being widely researched is changing the nature of the membrane surface. According to studies on the mechanism of membrane fouling, this phenomenon occurs due to the strong interaction of the similar hydrophobic nature of the hydrophobic foulants and hydrophobic membrane surfaces (Horseman et al. 2020). The organic membranes used in membrane filtration are mostly hydrophobic, so converting these surface properties to hydrophilic will increase their water permeability and decrease membrane fouling. In general, inorganic modifiers proved to be more effective in increasing the hydrophilicity of membrane surfaces than organic modifiers (Rana & Matsuura 2010). This is why inorganic catalysts are most often used to modify membrane surfaces. Therefore, catalytic membranes in general and photocatalytic membranes in particular are considered superior in mitigating membrane fouling because, in addition to enhancing the hydrophilicity of the membrane surface, it also provides catalytic activity to degrade the foulants under certain conditions, such as proper light irradiation (Yang et al. 2016).

The antifouling ability of photocatalytic membranes depends mainly on the photocatalytic performance and hydrophilicity of the photocatalyst applied (Wang et al. 2022). There have been many different photocatalysts incorporated into membranes, but in general, they are catalysts based on semiconductors such as TiO₂, ZnO, g-C₃N₄, etc. (Acharya & Parida 2020). The TiO₂ and ZnO catalysts themselves are UV-activated catalysts and were used in the initial studies. Their photocatalytic activity has been enhanced by doping or compositing stable metals or other semiconductors (Pelaez et al. 2012). However, the major drawback of ZnO is its sensitivity to large changes in the pH of the medium because it is an amphoteric oxide. g-C₃N₄ is a visible light-activated photocatalyst and has a particularly high photocatalytic performance when combined with several suitable semiconductors (Low et al. 2021). However, in terms of hydrophilicity, g-C₃N₄ does not have the desired effect because the composition is non-polar nitrogen carbons (Song et al. 2019). TiO₂ proved to be superior overall when considering factors such as chemical stability, environmental friendliness, photocatalytic performance, hydrophilicity, commercial availability in large quantities, and so on (Dong et al. 2015). The antifouling activity of TiO₂-based photocatalytic membranes will be significantly improved if the disadvantage of the high bandgap of TiO₂ is overcome. There have been many published bandgap tuning of TiO₂ but generally doping it by metallic or non-metallic elements or compositing it with other low bandgap semiconductors (Eddy et al. 2023). The doping of TiO₂ with metal nanoparticles (NPs) shows the advantage of being easily performed by chemical reduction reactions, and the bandgap of the resulting catalyst is also significantly reduced (Moma & Baloyi 2019). The metals commonly used in published works are durable metals, such as platinum, gold, or silver. Although the efficiency of photocatalysts is high, from an economic point of view, they do not seem feasible (Ijaz & Zafar 2021). Another fairly durable metal that has received little attention is copper. In terms of the economy and commercial availability in large quantities, copper is a tough candidate to beat (Ma et al. 2014). Therefore, this is the first time in this study that copper metal is used to tune the bandgap of TiO₂ for application in Cu@TiO₂-based photocatalytic antifouling PES membranes.

Membrane fouling is caused by organic compounds in wastewater that accumulate on the membrane surface over a long period of operation (Wang & Li 2008). Depending on the source of the treated wastewater, these organic compounds have very different compositions. Therefore, using actual wastewater to assess the antifouling ability of membranes makes it difficult to accurately conclude the performance of membranes (Zhang 2022). Laboratory studies often use simulated foulants because of the ease of concentration control and analysis and the exclusion of undesirable byproducts from actual wastewater (Drews 2010). Because the primary antifouling mechanism of photocatalytic membranes is to catalyze the degradation of foulants, these model foulants must be carefully selected for ease of analysis and not too readily degraded. Photocatalyst studies often use methylene blue as the model pollutant; however, this dye is readily degraded and therefore unsuitable for model foulants (Khan et al. 2022). In this study, the azo dye was chosen because it is considered to be a persistent compound and is difficult to degrade. In addition, we found that the antifouling efficiency of the modified membrane was significantly enhanced by incorporating H₂O₂ activation into the photocatalysis. A very small amount of H₂O₂ helps to achieve equilibrium of reactive free radical-generating reactions and, thus, a faster and more efficient breakdown of foulants.

In addition to the performance of the photocatalyst itself, the performance of the photocatalytic membranes is highly dependent on how the photocatalyst is incorporated into the membranes (Parrino et al. 2018). The photocatalysts in photocatalytic membranes must ensure maximum exposure to pollutants and be persistently immobilized on the membranes (Subramaniam et al. 2022). In terms of the second aspect, the photocatalytic membrane fabricated by the blending method has outstanding advantages. However, the photocatalysts located deep inside the membrane cannot have photocatalytic activity and greatly affect the permeability of the membrane, while the number of photocatalysts on the membrane surface is not enough to work efficiently (Zhang et al. 2016). This is because coating methods have been studied more extensively in the fabrication of photocatalytic membranes. In this case, the photocatalysts do not affect the porous pore structure of the membrane but are abundantly concentrated on the membrane surface. The high concentration of the photocatalyst on the membrane surface ensures efficient light absorption and contact with the pollutant (Zango et al. 2023). The problem is how to make the photocatalyst stably bound to the membrane surface. This is a key issue in the field of photocatalytic membranes that has attracted considerable attention from researchers (Gnanasekaran et al.). Until now, the solution to this problem has been to introduce a cohesive intermediate layer between the photocatalyst and the membrane surface. This intermediate layer is usually polar polymers that can form strong bonds to the membrane surface and immobilize inorganic photocatalysts through polar functional groups such as hydroxyl, ammonium, sulphonate, or carboxylate (Barclay et al. 2017). Because these polymers are polar and membranes are generally hydrophobic, a strong chemical bond is required between these polymers and the membrane surface (Van der Bruggen 2009). To the best of our knowledge, a lack of investigation on photocatalytic antifouling Cu@TiO₂/polyacrylic acid (PAA)/PES membranes has been performed previously. Therefore, the present study incorporates the use of oxygen plasma to facilitate covalent chemical bonding between the PAA and the PES membrane. Characteristic and photocatalysts of the prepared membranes were determined using Fourier-transform infrared (FTIR), scanning electron microscopy (SEM), transmission electron microscopy–energy-dispersive X-ray spectroscopy (TEM–EDS), ultraviolet–visible diffuse reflectance spectroscopy (UV)–Vis DRS, X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), and electron spin resonance spectroscopy (ESR). The catalytic performance of photocatalyst membrane was evaluated by the removal of Acid Blue 260 (AB260) from the aqueous solution.

Tisch Scientific Company (Ohio, USA) provided polyethersulfone (PES) membranes with a pore size of 0.22 μm and a water contact angle (WCA) of 0°. AB260 (C₃₁H₂₇N₃O₆SNa) was obtained from Alfa-Chemistry (New York, USA). Titanium dioxide (TiO₂, 99%), copper sulfate pentahydrate (CuSO₄.5H₂O, 98%), and hydrogen peroxide (H₂O₂, > 30%) were supplied by Merck (Germany). L-ascorbic acid (98.5%), acrylic acid (CH₂ = CHCOOH, 99%), Pluronic^® F-127 ((C₃H₆O·C₂H₄O)_x, non-ionic surfactant), KI (potassium iodide), IPA (isopropanol), HCOOH (formic acid), and NaN₃ (sodium azide) and other chemicals were provided by Sigma Aldrich (Missouri, USA). The chemicals were used as received without further purification. The solutions and standards were diluted using deionized water.

The Cu@TiO₂ photocatalyst were prepared as follows: First, 2 g of TiO₂ NPs were added to a 500 mL three-necked flask containing 350 mL of deionised water (DI) water and fitted with a thermometer, condenser, and magnetic bar, and the flask was placed on a magnetic stirring heater. This mixture in a flask was vigorously stirred and subjected to ultrasonic treatment for 1 h to achieve good dispersion of TiO₂ NPs in water. Next, a certain amount of CuSO₄·5H₂O (molar ratios of Cu to total moles of Cu and Ti of 1, 3, 5, 10, 15, and 25%) was transferred into the vigorously stirred reaction system. One gram (1 g) of Pluronic^® F-127 was added to the reactor, and the temperature of the system was slowly raised to 80 °C. After the solids in the reaction system were dissolved, 100 mL of the ascorbic acid solution was slowly added to the mixture, while the system was vigorously stirred at 80 °C for 2 h. At the end of the reaction, the color of the mixture changed to the characteristic color of the Cu NPs dispersed in water. The solid product in the reaction was allowed to cool naturally and then vacuum-filtered. The solid product was washed several times with DI water and ethanol before being dried at 70 °C for 24 h. The resulting photocatalysts are denoted as x% Cu@TiO₂, where x% is the molar ratio of the Cu mentioned above. The composite photocatalyst was then characterized by FTIR, SEM, TEM, EDS, UV–Vis DRS, XRD, XPS, and ESR.

The photocatalyzed degradation of AB260 by Cu@TiO₂ powder (20 mg) was carried out in 150 mL of AB260 solution with concentrations of 10 mg/L (without H₂O₂) and 20 mg/L (with H₂O₂). This dye degradation catalyzed by Cu@TiO₂/PES membrane was carried out in 150 mL of AB260 solution at a concentration of 20 mg/L in the presence of hydrogen peroxide. The Cu@TiO₂ powder was transferred to AB260 solution in a 250 mL beaker placed on a magnetic stirrer at 200 rpm. The light source is two ultraviolet-C (UVC) lamps with a power of 8 W, each placed 10 cm from the solution surface. All photocatalysis was performed at room temperature and atmospheric pressure. At each predetermined time, 5 mL of the reaction mixture was withdrawn, centrifuged at 2,500 rpm for 10 min to remove solids, and analyzed by UV–Vis spectroscopy at a characteristic wavelength of AB260. When using H₂O₂ (30%) in the photocatalytic decomposition of AB260, a very small amount (100 μL) of it was used. While in membrane's trapping tests approximately 0.0015 mole of scavengers, i.e. KI, IPA, HCOOH, and NaN₃ were employed to evaluate the contributions of holes, hydroxyl radicals, free electrons, and superoxide anion radicals to the photocatalysis. The photocatalyst with the highest efficiency in catalyzing the degradation of AB260 was selected to fabricate the photocatalytic membrane. The degradation of AB260 catalyzed by the photocatalytic membrane was carried out in the same way as above but with the help of a membrane holder. The membrane holder helps keep the membrane in place and ensures light exposure of the membrane when the AB260 solution is stirred.

The scavengers will capture these active agents and make the degradation reaction caused by them virtually nonexistent, which greatly reduces the degradation reaction efficiency. Figure 11(a) shows the dye degradation efficiency of 30, 60, 68, and 85% using IPA, NaN₃, HCOOH, and KI scavengers, respectively. Therefore, it can be concluded that the main mechanism causing the dye degradation catalyzed by this photocatalytic membrane is related to the free radical ^•OH. When comparing the performance of the photocatalyst and the photocatalytic membrane (Figure 11(b)), it can be seen that the degradation reaction rate at the initial time when using the photocatalyst is higher than when using the photocatalytic membrane. However, at the later stage of the reaction, this difference in performance was not very significant (about 7%). In addition, when the photocatalytic membrane is used with and without H₂O₂, the high difference in dye degradation efficiency is about 35%. Therefore, it proves the role of H₂O₂ activation in the antifouling activity of this photocatalytic membrane. The stability of the antifouling performance of the photocatalytic membrane was also evaluated through cycle tests (Figure 11(c) and 11(d)). The dye degradation efficiency decreased slightly with each cycle but was not significant and can be considered stable. In detail, the dye degradation efficiency over five consecutive cycles was 91, 89, 86, 82, and 82%, respectively; i.e. this efficiency decreases by 9% after five photocatalysis cycles.

In this investigation, Cu NPs were employed to adjust the high bandgap energy of TiO₂, resulting in the production of a composite photocatalyst that was coated onto a PES membrane with a plasma-grafted PAA layer. The catalytic activity of both powdered photocatalysts and coated photocatalytic membranes was fully examined. Various characterizations were utilized, including FTIR, SEM, TEM–EDS, UV–Vis DRS, XRD, XPS, and ESR, to demonstrate the successful synthesis of the composite photocatalyst, which has a reduced bandgap (3.32 eV) and a small and uniform particle size (100–200 nm), capable of generating reactive free radicals under light irradiation. The highest catalytic efficiency (73%) was observed in the powdered photocatalyst at a molar ratio of 25%Cu@TiO₂. The activation of H₂O₂ contributed greatly to the AB260 dye degradation efficiency catalyzed by the composite photocatalyst (96%). The coated photocatalytic 25%Cu@TiO₂/PAA/PES membrane catalyzed the dye degradation with an efficiency of 91% and was quite stable over five continuous cycles. The water permeability of the coated photocatalytic membrane was still quite high (1,868 L·m⁻²·h⁻¹), stable, and only reduced by about 9% compared to the pristine PES membrane. The FRR of the coated photocatalytic membrane after photocatalytic treatment was very high, at about 98%. Although coated photocatalytic membranes have higher surface roughness than pristine PES membranes, the antifouling ability may not be significantly affected because of the high photocatalytic activity of modified membranes. This study shows that Cu@TiO₂/PAA/PES photocatalytic membrane has the potential to be researched and applied in real textile wastewater or other contaminant in industry wastewater.

This research work was partly supported by the International Atomic Energy Agency (IAEA) under Research Contract No. 24715.

All data generated or analyzed during this study are included in this article.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.