This study investigated the generation of reactive oxygen species (ROS) (O2−•, H2O2, and HO•) by promoting the Cu(II)/Cu(I) redox cycle with certain reducing agents (RAs) in aerobic aqueous solution, and benzoic acid (BA) was employed as indicator for the hydroxyl radical (HO•). Hydroxylamine (HA) can reduce Cu(II) to Cu(I) to induce chain reactions of copper species resulting in the generation of the superoxide radical (O2−•) and hydrogen peroxide (H2O2), and the intermediate Cu(I) can further activate H2O2 via a Fenton-like reaction to produce HO•, creating the remarkable BA degradation. O2 is indispensable, and unprotonated HA is the motive power in the O2/Cu/HA system. Moreover, pH is a crucial factor of the O2/Cu/HA system due to the protonated HA not being able to reduce Cu(II) into Cu(I). The oxidation of HA can be effectively induced by trace amounts of Cu(II), and both a higher HA dosage and a higher Cu(II) dosage can enhance H2O2 generation and BA degradation. In addition, some other RAs that can reduce Cu(II) into Cu(I) could replace HA in the O2/Cu/HA system to induce the generation of these ROS in aerobic aqueous solution.
Advanced oxidation processes (AOPs) based on Fenton and Fenton-like reactions have fascinated practitioners of water treatment as effective methods to degrade refractory organic contaminants in recent years (Ganiyu et al. 2018). Some highly reactive oxidants such as the hydroxyl radical (HO•) can be produced during the decomposition of hydrogen peroxide (H2O2) via the catalytic redox cycle of Fe(III)/Fe(II). Similar to the circulation of Fe(III)/Fe(II), copper can also decompose H2O2 into HO• via the Fenton-like copper redox (Pham et al. 2013). Previous studies have reported that cuprous copper [Cu(I)] is a strong oxygen scavenger and some reactive oxygen species (ROS) were produced during the redox reactions of Cu(I), such as the superoxide radical (O2−•) and H2O2 (Wen et al. 2014; Zhou et al. 2016a; Zhou et al. 2018). Then, a sequence of reactions between the ROS and other components in aerobic aqueous solution would be induced (Yuan et al. 2012). As a further product, HO• can be produced via a Fenton-like reaction between Cu(I) and H2O2 (Yuan et al. 2012; Lee et al. 2016; Zhang et al. 2017). It is remarkable that Cu(I) is a well-known activator of H2O2 to produce HO•, whereas Cu(I) was seldom investigated to activate H2O2 due to its instability and naturally occurring copper being mostly in the cupric copper [Cu(II)] oxidation state.
However, it has been proposed that the oxygen-dependent oxidation of 1,4-hydroquinone undergoing copper-redox reactions can produce ROS such as O2−• and H2O2 (Yuan et al. 2013), and the further generation of HO• has been proved based on electron spin resonance (ESR) spectroscopy (Lee et al. 2016). Nonetheless, the product HO• gained little attention and the previous literatures mainly focused on its biological effects, such as oxidative DNA damage (Mobley et al. 1999; Kim et al. 2015). In summary, the capacity of 1,4-hydroquinone to reduce Cu(II) into Cu(I) is the decisive prerequisite to induce the chain reactions of copper species and accelerate the Cu(II)/Cu(I) redox cycle.
Therefore, some reducing agents (RAs), which are capable of reducing Cu(II) into Cu(I), can be quickly oxidized in aerobic aqueous solutions and undergo copper-redox reactions. More importantly, several ROS (O2−•, H2O2, and HO•) which can oxidize organics and inorganics in aqueous solution could be produced during these oxidation processes. Hydroxylamine (HA) is a distinguished reducing agent and was commonly considered as an antioxidant for reactive oxidants (Krishna et al. 1998). Recently, HA was introduced into some metal-catalyzed AOPs to enhance the production of reactive radicals (e.g. HO• and SO4−•) by accelerating the Fe(III)/Fe(II) and Cu(II)/Cu((I) redox cycles (Chen et al. 2011; Zou et al. 2013; Zhou et al. 2016b; Hou et al. 2017). On the strength of the strong reducibility, in this study, HA and some other RAs were employed to induce the chain reactions of oxygen species for generating HO• and some other ROS without the external supply of H2O2.
Thus, the objective of this study was to evaluate the degradation of organic contaminants in the O2/Cu/HA system while benzoic acid (BA) was employed as the target organic pollutant. The generation of ROS (O2−•, H2O2, and HO•) by promoting the Cu(II)/Cu(I) redox cycle with HA in aerobic aqueous solution was investigated, and mainly focused on the mechanism of the ROS (O2−•, H2O2, and HO•) generation and the effect of pH. The effects of HA dosage and Cu(II) dosage on H2O2 generation and BA degradation in the O2/Cu/HA system were also explored. Moreover, in order to expand the common mechanism of the O2/Cu/RAs system, some RAs (e.g. L-ascorbic acid, N-methylhydroxylamine, p-hydroquinone, and sodium nitrite) were introduced into the O2/Cu/RAs system to replace the role of HA in the O2/Cu/HA system.
HA (≥99.0%), copper sulfate pentahydrate (≥99.0%), BA (≥99.5%), and nitro blue tetrazolium (NBT, ≥98.5%) were supplied by Sigma-Aldrich. Neocuproine hemihydrate (NCP, ≥98%), coumarin (≥99.5%), 7-hydroxycoumarin (≥99.0%), acetone oxime (≥98.0%), and methyl tert-butyl ether (MTBE, (≥99.0%)), and 5,5-Dimethyl-1-pyrroline N-oxide (DMPO, ≥98.0%) were purchased from the Aladdin Industrial Corporation. N-methylhydroxylamine (NMHA, ≥99.0%), L-ascorbic acid (L-AA, ≥99.7%), p-hydroquinone (p-HQ, ≥99.0%), sodium nitrite (SN, ≥99.0%), tert-butyl alcohol (TBA), H2O2 (30%, v/v), acetone, and potassium titanium oxalate were purchased from the Sinopharm Chemical Reagent factory. Ammonium acetate and methanol were of HPLC grade and supplied by Sigma-Aldrich. The other chemical reagents in these experiments were all of analytical grade.
For experiments requiring aerobic or anoxic condition, pure O2 (≥99.2%) or pure N2 (≥99.99%) was bubbled into the reactor for 10 min prior to the initiation of the reaction and throughout the entire experiment at 25 ± 1 °C. Some other experiments were performed in a 250 mL sealed Florence flask with a constant stirring rate at 25 ± 1 °C to explore the consumption of DO in the aqueous solution. Each reactive aqueous solution was prepared with deionized water, the desired dosages of Cu(II) and BA or coumarin were firstly added into the reaction solution, and then adjusted to the desired pH by phosphoric acid-phosphate buffering. The reaction was started by adding a small volume of the desired fresh RAs solution. To analyze the concentrations of H2O2, BA, and Cu(I), the samples were withdrawn from the reaction solution and mixed with excess potassium titanium oxalate, sodium sulfite, and NCP at predetermined time intervals, respectively. Moreover, NBT was introduced into the O2/Cu/HA system to verify the generation of O2−• through the qualitative detection of monoformazan, which is the characteristic product of the reaction between NBT and O2−•.
To distinguish the contribution of HO• on BA degradation, the excess TBA was added to the reaction solution to perform the quenching experiment. Resulting from that the 7-hydroxycoumarin was the characteristic product of the reaction between coumarin and HO•, coumarin was added into the O2/Cu/HA systems to act as a HO• catching agent to identify the generation of HO• through detecting 7-hydroxycoumarin. All experiments were independently repeated at least 3 times, and the average values along with one standard deviation (error bars) are provided in the figures.
The BA concentration was measured by high performance liquid chromatography (HPLC, Waters e2695) equipped with a 2,489 λ UV absorbance detector and a reverse-phase C18 column (4.6 × 150 mm). The mobile phase composition and flow rate were methanol/0.1% ammonium acetate (40:60, v/v) and 1 mL/min, respectively, and the detection wavelength was 227 nm. The concentration of 7-hydroxycoumarin was analyzed by a HPLC (Waters e2695) equipped with a 2475 fluorescence (FLR) detector and a reverse-phase C18 column (4.6 × 150 mm). The mobile phase composition and flow rate were methanol/0.1% ammonium acetate (50:50, v/v) and 1 mL/min, respectively. The fluorescence intensity of 7-hydroxycoumarin was measured using a 2475 FLR detector with excitation wavelength of 346 nm and detection wavelength of 456 nm. The pH and DO concentration were measured by a pH meter (Shanghai Leici Apparatus Fac., China) and a DO meter (JPB-607A), respectively. The O2−• was identified by the NBT method (Liu et al. 2009; Cheng et al. 2017) on a UV-vis spectrometer (UV-1800, MAPADA). The concentrations of Cu(I) and H2O2 were determined by spectrophotometric methods on a UV-vis spectrometer (UV-1800, MAPADA), which is described in detail in Text S1 (available with the online version of this paper). HA was derived to acetone oxime by reaction with acetone and extracted by MTBE, and then measured by gas chromatography (QP2010 Plus) using a FID detector (Zou et al. 2013). Electron spin resonance (ESR, Bruker EMX plus) experiments were performed using DMPO as a spin-trapping agent to identify the generation of HO• (Text S2, available online).
RESULTS AND DISCUSSION
H2O2 generation and BA degradation in the O2/Cu/HA system
The generation of H2O2 and the degradation of BA during the copper-catalyzed oxidation of HA in aerobic and anoxic aqueous solutions was investigated. As shown in Figure 1(a), the generation of H2O2 was significant in the O2/Cu/HA system, and the H2O2 concentration became relatively stable after an initial rise and later drop. With the addition of 2 mM HA in the aerobic aqueous solution catalyzed by copper, the concentration of H2O2 was 0.58 mM at 120 min. Moreover, as a reactive radical probe, 79.9% BA was degraded in the O2/Cu/HA system as shown in Figure 1(b). On the contrary, the generation of H2O2 was seriously weakened in anoxic aqueous solution (the N2/Cu/HA system), resulting in less than 4% BA being degraded at 120 min, which shows that O2 is indispensable in the O2/Cu/HA system.
HA reduce Cu(II) into Cu(I)
Consumption of O2 and generation of O2−•
Moreover, the previous studies (Liu et al. 2009; Cheng et al. 2017) showed that the reduction reaction between O2−• with NBT and the inherent spectral characteristics of its reduction product, monoformazan (MF), provide a universal detection and quantitation method for O2−• that has high sensitivity. Therefore, NBT was introduced into the reaction solution to verify the generation of O2−• in the O2/Cu/HA system. As shown in Figure 2(c), the spectra clearly illustrate a peak at 660 nm, which is the characteristic absorption peak of MF. The absorbance of the characteristic reaction product (MF) increased gradually with the growth of the HA dosage. Thus, it reveals the generation of O2−• in the O2/Cu/HA system.
Generation of H2O2 and HO•
Based on the strong inhibition of BA degradation by TBA, the specific spectrum of DMPO-OH, and the significant generation of 7-hydroxycoumarin in the O2/Cu/HA system, it can be concluded that the primary reactive oxidant of the O2/Cu/HA system is HO•.
Effect of pH
Based on the above results, it can be inferred that NH3OH+ can not reduce Cu(II) into Cu(I), resulting in very little or no H2O2 and HO• generated. Moreover, higher pH may decrease the utilization ratio of HA due to the higher reaction rate between HO• and HA with pH increase (pH 4.0, k < 5.0 × 108 M−1s−1; pH 8.0, k = 9.5 × 109 M−1s−1) (Buxton et al. 1988). More HA was utilized to reduce Cu(II) into Cu(I) at pH 5.5 than that at pH 7.0 and 8.2, resulting in sustained decomposition of H2O2 to produce HO• (Figure 4(c)). For this reason, the BA degradation rate after 120 min decreased with an increase of the pH range from 5.5 to 8.2. Therefore, unprotonated HA is the motive power of the O2/Cu/HA system, and pH is a critical factor.
Effect of HA dosage
Figure 5 shows the effect of HA dosage on the generation of H2O2 and the degradation of BA in the O2/Cu/HA system. The growth in the final H2O2 yield was observed with the increase of HA dosage in the range of 0.5 mM to 2 mM. However, exceeding this concentration resulted in reduction of the final yield of H2O2. When the dosage of HA is under 3 mM, concentration of H2O2 upgrades significantly rapidly in the first phase (0–10 min), descends gradually from 10 min to 60 min, and remains stable after 60 min. Whereas, when the dosage of HA is greater than 5 mM, the H2O2 concentration fluctuates firstly then remains stable resulting from the competition between the H2O2 generation and consumption. A high concentration of HA can accelerate the transformation from Cu(II) to Cu(I) to enhance the formation of H2O2 (Equations (2) and (3)) (Yuan et al. 2012; Kim et al. 2015; Zhou et al. 2016a); meanwhile, Cu(I) can consume H2O2 as a greater activator for H2O2 (Equation (5)). The final BA degradation in 120 min increased with the growth of HA dosage, whereas excessive HA dosage, such as 8 mM or 10 mM, can inhibit the BA degradation in the initial phase. Resulting from that, the generated HO• could be quenched by excess HA (Equation (6)), especially when the dosage of HA is high enough (Buxton et al. 1988).
The role of reducing agents in the O2/Cu/RAs systems
Based on the results and discussion above, the key step to form reactive oxygen species in the O2/Cu/HA system is the generation of Cu(I), which can induce further chain reactions between the copper species and the oxygen-containing substances (O2, H2O2, and O2−•). It could be surmised that some other RAs, which can reduce Cu(II) into Cu(I), may take on the role of HA in the O2/Cu/HA system. Therefore, the effects of some RAs (NMHA, L-AA, p-HQ, and SN) were investigated in the O2/Cu/RAs systems.
Figure 6 shows the generation of Cu(I) and H2O2, BA degradation, and 7-hydroxycoumarin generation in the O2/Cu/RAs systems after adding NMHA, L-AA, p-HQ and SN, respectively. As shown in Figure 6(a), Cu(I) was produced in the O2/Cu/RAs systems with the addition of 1 mM HA, NMHA, or L-AA, and nearly 100% of Cu(II) was rapidly reduced into Cu(I) in the initial phase in these three systems. Corresponding to the generation of Cu(I), adding these three RAs significantly enhanced the H2O2 formation as shown in Figure 6(b), and the H2O2 concentrations dramatically increased in the initial phase and then gradually decreased with concentration peaks of 0.36, 0.37, and 0.43 mM, respectively. Furthermore, HO• was obviously produced in these systems, and the removal rates of BA reached 57.9%, 39.5%, and 39.3%, respectively, and the yields of 7-hydroxycoumarin were respectively 5.1, 8.8, and 6.2 μM at 120 min (Figure 6(c) and 6(d)). Moreover, Cu(II) can not be reduced into Cu(I) by SN in the O2/Cu/SN system, resulting in the chain reactions between copper species and oxygen-containing substances not being able to be induced. As such, the generation of H2O2, removal of BA, and generation of 7-hydoxycoumarin were also not detected in the O2/Cu/SN system.
In addition, almost all the Cu(II) was reduced into Cu(I) by p-HQ in a short time in the initial phase (less than 2 min). According to the previous literature (Yuan et al. 2013), p-HQ can be oxidized into p-benzoquinone (p-BQ) catalyzed by Cu(II)/Cu(I) redox couple in aerobic solution (Figure S1, available with the online version of this paper). Due to p-BQ being an O2−• scavenger (Kishikawa et al. 2009; Kawashima et al. 2011), the intermediate O2−• in the O2/Cu/p-HQ system may be quenched by p-BQ to inhibit the formation of H2O2 via Equation (3). Thus, the generation of H2O2 in the O2/Cu/p-HQ system was relatively lower than that of the O2/Cu/HA, O2/Cu/NHMA, and O2/Cu/L-AA systems. However, the generation of H2O2 was also detected and 0.07 mM H2O2 was produced at 120 min in the O2/Cu/p-HQ system. Low yield of H2O2 inhibited the generation of HO• via Equation (5); only 3% BA was degraded and 0.25 μM 7-hydroxycoumarin was generated. Thus, the p-HQ can also play the same role that HA played in the O2/Cu/RAs system; notwithstanding, p-BQ can reduce the yield of H2O2.
Based on the results and discussion above, it can be concluded that some other RAs that can reduce Cu(II) into Cu(I) can act as HA in the O2/Cu/HA system so as to induce the chain reactions of the copper species to induce the generation of ROS (e.g. O2−•, H2O2, and HO•). The mechanism of the O2/Cu/RAs systems was proposed as Figure 7.
This study set up the O2/Cu/HA system to utilize HA to reduce Cu(II) into Cu(I) so as to initiate the chain reactions of the oxygen species, resulting in the generation of O2−•, H2O2, and HO• in the aerobic aqueous solutions. The mechanism proposes that unprotonated HA can reduce Cu(II) into intermediate Cu(I) to react with O2 producing O2−•, and O2−• can react with Cu(I) to form H2O2. Then, H2O2 can be further activated by intermediate Cu(I) via a Fenton-like reaction to produce HO•, resulting in BA degradation. O2 is indispensable due to less H2O2 being detected and less BA being degraded in the anoxic (N2) aqueous solution. Moreover, pH is a crucial factor in the O2/Cu/HA system; resulting from that, the protonated HA (NH3OH+) can not reduce Cu(II) into Cu(I), and Cu(I), which can initiate the chain reactions, is the pivotal intermediate of the O2/Cu/HA system. A high concentration of HA can accelerate the Cu(II)/Cu(I) redox cycle to initiate the HO• generation, but excessive HA dosage could inhibit the BA degradation in the initial phase. Although a higher Cu(II) dosage could enhance the HO• generation to degrade BA, a lower dosage of Cu(II) is more beneficial for the accumulation of H2O2 (Figure S2, available with the online version of this paper). In addition, some other RAs (e.g. NMHA, L-AA, and p-HQ) which can reduce Cu(II) into Cu(I) could act as HA in the O2/Cu/HA system so as to induce the chain reactions of the copper species to induce the generation of ROS (e.g. O2−•, H2O2, and HO•) with strong oxidation ability.
Therefore, these ROS generated in this new advanced oxidation system (O2/Cu/RAs system) can effectively degrade refractory organic contaminants in reclaimed water and wastewater with O2 as the original oxidant. These systems using RAs and Cu(II) cover wider pH ranges (e.g. 4.2–8.0 for the O2/Cu/HA system), comparing to the iron-catalyzed Fenton-like system. However, the concentration and type of copper chelating compounds existing in the reaction solution are potential variable factors (both negative and positive effects) to influence the performance of the O2/Cu/RAs system. Structural stable heterogeneous copper catalysts need to be investigated to reduce the environmental risk of homogeneous copper, although the effective dosage of Cu(II) in the O2/Cu/RAs system is less than 15.6 μM (the maximum level for copper set by the Chinese Ministry of Health). In addition, some other metals with multiple valences which are eco-friendly may be employed to replace copper in the O2/Cu/RAs system.
This work was supported by the National Natural Science Foundation of China (No. 51508353) and Graduate Student's Research and Innovation Fund of Sichuan University (No. 2018YJSY074).