Effect of the initial pH on the removal of ammonia nitrogen
The photocatalytic conversion of ammonia nitrogen (500 mg/L) in aqueous solution with different initial pH is shown in Fig. 2. The initial pH greatly affected the photocatalytic oxidation of ammonia nitrogen. As shown in Fig. 2a, the ammonia removal rate was improved with the increase of the initial pH, which reaches an optimum pH value of 10.0 − 11.0. However, when the pH is 7.0, the ammonia nitrogen concentration is basically unchanged. It is known that when the solution is acidic and neutral, the ammonia nitrogen exists in the form of NH4+ which hydroxyl radicals generated by photocatalysis cannot easily attack (Lee et al. 2002; Yao et al. 2020). Fig. S1 shows that under different pH conditions, the process of photocatalytic ammonia nitrogen is fitted with the first − order kinetic reaction process. The pH not only influences on the conversion of ammonia nitrogen, but also the oxidation products. It can be seen from Fig. 2b that when the pH increases from 7.0 to 10.0, the amount of gaseous − N species generated increases. However, the conversion rate of ammonia nitrogen into N2 and other gases decreases with the increase of pH. Table S1 shows that the gas conversion rate is 94.0% under the pH = 8.0 and the gas conversion rate drops to 61.2% under the pH = 11.0. The reason is probably that the increase in pH facilitates generation of hydroxyl radicals ●OH, which mainly produces nitrite and nitrate nitrogen. According to the previous work (Feng et al. 2021), the key to generation of gaseous − N species is superoxide radical ●O2−. As the pH increases, superoxide radical ●O2− was gradually depleted due to the limited dissolved oxygen in the closed reaction system. The generated ionic species were also monitored. As shown in Fig. 2c and Fig. 2d, nitrite ions were found as the major products and nitrate ions are negligible. With the increase of pH, more nitrite ions were produced in 6 h reaction.
Effect of temperature on the removal of ammonia
The photocatalytic conversion of ammonia nitrogen in aqueous solution with different temperatures is shown in Fig. 3. As shown in Fig. 3a, the removal efficiency of ammonia nitrogen does not change significantly when the temperature rises from 20 oC to 30 oC. However, when the temperature is increased to 40 oC and 50 oC, the removal efficiency of ammonia nitrogen is improved. Fig. S2 shows that under different temperature conditions, the process of photocatalytic ammonia nitrogen is consistent with the first − order kinetic reaction process. When the temperature is greater than 30 oC, the reaction rate increases with the increase of the reaction temperature. Therefore, it can be considered that the increase in temperature promotes the reaction rate mainly because of the increase in the number of molecular collisions, rather than the increase in the number of thermally activated molecules. As the reaction temperature increases, the amount of ammonia nitrogen converted into N2 and other gases gradually increases, probably due to the reduced gas solubility in water accelerates the release of gaseous products into gas phase.
According to the data in Table S1, the Arrhenius Eq. (1) can be used to calculate the activation energy of photocatalytic ammonia nitrogen (Jung and Kruse 2017).
In the formula, k is the reaction rate constant at temperature T (h− 1), Ea is the activation energy of the reaction (J/mol), R is the molar gas constant (J/mol·K), and T is the absolute temperature (K). According to the Arrhenius Eq. (1), with ln(k2/k1) as the ordinate and (1/T2 − 1/T1) as the abscissa, using the least squares method to fit, the slope − Ea/R can be obtained, and the activation energy of the reaction is finally calculated to be 8.421 kJ/mol.
Effect of the high salinity on the removal of ammonia nitrogen
Generally, industrial wastewater with high concentration of ammonia nitrogen has a high salinity. Therefore, the effect of excessive Cl− and SO42− on the photocatalytic oxidation of ammonia nitrogen was investigated. It can be seen from Fig. 4a that the removal of ammonia nitrogen is basically not affected by the high concentration of SO42−. Under alkaline conditions, part of SO42− could be converted into sulfate radicals (SO4●−) by photogenerated ●OH (Wang et al. 2021). Although high concentration of SO42− consumes part of ●OH, SO4●− has a higher oxidation − reduction potential (E0 = 2.5 − 3.1 eV), and its oxidation capacity is equivalent to ●OH (E0 = 2.8 eV) (Song et al. 2019; Wang et al. 2019; Wang et al. 2019). Therefore, the high concentration of SO42− basically does not affect the removal rate of ammonia nitrogen. As shown from Fig. 4b, under the condition of high concentration of Cl−, the removal effect of ammonia nitrogen is obviously suppressed and as the concentration of Cl− increases, the removal rate of ammonia nitrogen gradually decreases. The reason may be that Cl− has the ability to scavenge ●OH, which convert part of Cl− into chlorine radical Cl● (E0 = 2.4 eV) that can not efficiently degrade ammonia nitrogen directly, thereby inhibiting the removal rate of ammonia nitrogen (Han et al. 2021; Liou and Dodd 2021).
Treatment of wastewater by photocatalyst film
The Cu/TiO2 photocatalyst powder was immobilized onto some solid substrates including titanium sheet, copper sheet, glass and ITO conductive glass. Figure 5a shows that the removal rate of ammonia nitrogen on glass substrate is the highest. In contrast, the efficiencies of photocatalytic oxidation of ammonia using titanium sheet, copper sheet and ITO supported photocatalytic films are relatively lower. Unlike the slurry tests in the closed reactor, the circulated film test system is open to the air, which is benefit for O2 supply and improved mass transfer between aqueous and gaseous phases. It can be seen from Fig. 5b and Fig. S3 that only a small part of NH3 is converted into NO2− and NO3−, and most of NH3 is removed in the form of gaseous species. This is because superoxide radicals ●O2− are proposed as the key oxidant to gas production (Feng et al. 2021), and the increased supply of oxygen in the open reactor system will increase the superoxide radicals ●O2−, and thus N2 production will be increased.
The flow rate of circulated water on the surface of catalyst film has strong influence on its removal efficiency of ammonia nitrogen and catalyst stability. It can be seen from Fig. 6a and Fig. S4 − S6 that when the flow rate is 40 mL/min, the morphology of Cu/TiO2@Glass before and after the photocatalytic reaction is basically unchanged, and the removal efficiency of ammonia nitrogen is the highest. When the flow rate increases, the catalyst film on the glass surface is damaged due to the excessive surface velocity, which leads to the weakening of the photocatalytic efficiency. However, when the flow rate decreases, the catalyst film on the glass surface is also damaged. The reason may be that the residence time of the solution on the surface of the catalyst film increases, and anions tend to accumulate on the surface of TiO2 and destroy the bonding structure of TiO2/SiO2, thereby reducing the removal rate of ammonia (Levchuk et al. 2019). Therefore, the flow rate 40 mL/min was selected for subsequent experiments.
The circulated treatment of different types of high − concentration ammonia wastewater by Cu/TiO2@Glass was investigated. In the control experiment (Fig. 7a), the ammonia was rapidly removed in the first hour, and then the oxidation rate gets slower. Because solution pH gradually decreases along with the ammonia removal, the photocatalytic oxidation of NH4+ in acidic or neutral condition is inhibited. Compared with Fig. 7a, the removal of ammonia in high salinity wastewater decreased significantly (Fig. 7b), the ammonia no longer drops after 4 h and about 165 mg/L of NH3 − N can be removed finally. The reason is that excessive Cl− consumes the h+ and ●OH on the surface of the photocatalyst film, resulting in a decrease of photocatalytic oxidation efficiency (Wang et al. 2021). Figure 7c shows that the NH3 − N in the copper − ammonia complex wastewater no longer drops after 2 h and about 160 mg/L of NH3 − N can be removed. The reason is that copper ions play a role in capturing e− in the photocatalytic process, inhibiting the generation of ●O2−, resulting in a decrease of photocatalytic oxidation efficiency. However, Cu2+ ions in water can be reduced to Cu+ or Cu depositing on photocatalyst film by the photogenerated electrons, which can be considered as simultaneous removal of NH3 − N and Cu2+ according to the Fig. S7. Figure 7d shows that the NH3 − N in the liquid − ammonia mercerization wastewater was slowly removed during 6 h reaction and about 135 mg/L of /NH3 − N can be removed finally, since the organic matters in liquid − ammonia mercerization wastewater consume the h+ and ●OH produced on the surface of the photocatalyst film.
To investigate the stability of Cu/TiO2@Glass during treatment of different wastewater, the experiments were repeated five times and the removal rates were shown in Fig. 8a. Cu/TiO2@Glass film before and after the reaction was weighed, and the mass change was calculated as shown in Fig. 8b. For high salinity ammonia wastewater, as the number of reuses of photocatalyst film increases, the removal rate of NH4+/NH3 − N gradually decreases and it is dropped by about 6% after five repeated experiments. In addition, the mass of photocatalyst film is continuously decreased because the excessive anions damages the surface structure of the film. Regarding the liquid − ammonia mercerization wastewater, although the organic matter has a greater inhibitory effect on the removal efficiency of NH3 − N, the removal rate was only dropped by about 1.5% after five repeated experiments, indicating that organic matter has a small impact on the reusability of Cu/TiO2@Glass. For copper − ammonia wastewater, as the number of reuses of Cu/TiO2@Glass increases, the removal rate of NH3 − N increases unpredictably. The removal of NH3 − N increased from 56.6–67.7% during the first three cycles and then slightly decreased, but the removal rate is still higher than that of the first photocatalytic process. Besides, as the number of reuses increases, the mass of Cu/TiO2@Glass continues to increase, because copper ions in the solution are continuously reduced and deposited on the surface of photocatalyst film. The deposited Cu species could be work as cocatalyst to enhance the photocatalytic performance (Mingmongkol et al. 2021).
Mechanism of photocatalytic conversion of ammonia
There are many pathways involved for photocatalytic oxidation of ammonia on Cu/TiO2 as illustrated in Fig. 9. In the closed reactor condition, it has been proposed that ●OH, h+ and ●O2− are main oxidants for ammonia conversion, in which ●O2− radical strongly influences the gaseous products (pathway I). However, due to the depletion of dissolved O2 content, the reactions prefer to further proceed via pathways II and III, thus the conversion rate and selectivity to N2 gradually decreases. In contrast, in the open system, the NH3 abatement resulting from both the large portion of volatilization and photocatalytic oxidation would decrease the pH of the solution gradually, so that the reaction driven by ●OH (pathway II and III) are limited. Moreover, due to the continuous supply of O2 within 6 hours, the ammonia conversion driven by ●O2− (pathway I) would be proceeded continuously, and the selective oxidation to gaseous products was enhanced as shown in Fig. 9. It can be found that no matter what kind of Cu/TiO2 films used in the open system, the removal of ammonia and the selectivity of gaseous products are higher than those of Cu/TiO2 powder in the closed reactor, which further demonstrates the role of dissolved O2 in the ammonia conversion.