Role of Borate Buffer in Organic Degradation by Peroxymonosulfate in the Presence of Metal Oxides

: The effects of borate ions on the reactivity of peroxymonosulfate (PMS) during organic degradation in the presence of metal oxides were examined. The metal oxides exhibited catalytic abilities for the degradation of carbamazepine (CBZ) but not for phenol (PN). Scavenging experiments revealed the absence of radical generation during PN degradation in the presence of the various metal oxides and borate buffer. This indicated that the borate buffer hindered the catalytic abilities of the metal oxides for producing radicals via the PMS oxidant, especially during the faster degradation of compounds such as PN. Various concentrations of borate ions were assessed for enabling pH control and permitting catalytic activity. Fe 2 O 3 was found to accelerate and inhibit PN degradation at borate-ion concentrations of 2 mM and 5–20 mM, respectively. Only the 20 mM borate-ion solutions were successful at maintaining the initial pH for 2 d. Phosphate buffer, which was examined as an alternative, also disrupted radical formation but not as considerably as that of the borate ions with metal oxides. This study demonstrates the signiﬁcance of enabling pH control and permitting the catalytic activity for ensuring the effective use of oxyanions as buffers.

Several studies have used borate ions as a typical buffer solution to mimic groundwaterrelevant pH conditions (6.0-8.5) owing to their pH buffering capacity being in a range of 8.0-10.0 [26,27]. Borate ions were recently found to stimulate PMS activity or induce PMS depletion resulting in accelerating or decelerating the degradation of target compounds depending on its concentrations [24,28].
The in situ presence of naturally occurring minerals such as Fe(II/III)-(1-20%, avg. 3%) and Mn(IV)-containing oxides (20-3000 mg/kg, avg. 600 mg/kg) [29][30][31] in subsurface systems can presumably enable the activation of injected PMS [32,33]. Although the effects of borate buffer on PMS activation in the absence of catalysts have been systematically investigated, the interactions between borate buffer and PMS in the presence of catalysts such as metal oxides have not yet been explored. Therefore, the present study was aimed at investigating the effects of borate buffer on organic degradation by the PMS oxidant in the presence of various metal oxides, identifying the major reactive species of PMS oxidation in the presence of metal oxides and borate ions, and optimizing the concentration of borate buffer solution to maintain a target pH and ensure indifference toward PMS reactivity.

Experimental Procedure
The sample preparation and treatments were all conducted using 125-mL crimped amber bottles with airtight butyl-rubber stoppers in an anaerobic chamber (>95% N 2 , <5% H 2 ) to mimic the anaerobic conditions of aquifer systems. All solutions were purged with N 2 gas for 1 h prior to use. The pH of borate buffer solution (~25 mM) was adjusted to 7.5-8.0 using 0.1 M HCl and 0.1 M NaOH, accordingly, and 25 mL of the borate buffer solution (25 mM) was added to separate vessels containing 0.006 g of the respective metal oxides. In addition, 15 mL of CBZ (90 µM) or phenol (90 µM) solution was added in accordance with the specific experiments. Finally, the addition of PMS stock solution (20 mL, 600 µM) initiated the experiments. Additionally, the effects of borate buffer concentration on PMS reactivity were examined by varying the borate-ion concentration. Moreover, the identification of the predominant reactive species during PMS oxidation in the presence of the borate buffer solution and the metal oxides was conducted using 0.5 mL of scavengers such as TBA (10.46 M) and IPA (13.07 M), which are known to eliminate • OH and SO 4 •− , respectively, with a 500:1 molar ratio of the scavenger to the PMS oxidant [34]. In addition, NB (9.71 M) was used to detect • OH (k •OH = 3.9 × 10 9 M −1 s −1 ) [35]. The effect of the borate buffer solution on PMS reactivity in the presence of catalysts was elucidated using Fe(II)SO 4 as a positive control instead of the metal oxides. Triplicate samples were mixed using an end-over-end rotary shaker (45 rpm) during the reaction. Seven milliliters Water 2021, 13, 2698 3 of 11 of the samples were periodically drawn using a syringe and filtered using 0.22 µm nylon filter and stored prior to analysis.
The degradation rates of each target compound were described using the pseudo-firstorder reaction or second-order reaction, with the reaction rate constant (k) being determined using Equation (1) or (2), respectively.

Analytical Methods
The concentrations of CBZ, PN, and NB were quantified using an ultra-performance liquid chromatography apparatus (Agilent Technologies-1260 Infinity) equipped with a UV detector (UPLC/UV). Samples were diluted with MeOH (1:1, v/v) to quench the reactions and reduce the matrix effects. Ultrapure water (18 MΩ water, solvent A) and MeOH (solvent B) were used as mobile solutions, with chromatographic separation being achieved using an Eclipse XDB-C18 column (5 µm, 4.6 × 150 mm) at an injection volume of 20 µL. For CBZ analysis, the gradient was initiated at 20% of solvent A, ramped to 70% in 12 min, maintained for 8 min, returned to 20% at 25 min, and maintained for 30 min at a flow rate of 1.0 mL min −1 under 286-nm UV radiation. PN and NB were analyzed using an isocratic mobile phase with 45% and 40% of solvent A at wavelengths of 254 nm and 275 nm, respectively, and a flow rate of 0.8 mL min −1 .

Organic Degradation by the Peroxymonosulfate (PMS) Oxidant in the Presence of Metal Oxides and Borate Buffer
The effects of borate buffer on organic degradation by PMS were assessed in the presence of various metal oxides. Two different organic compounds, CBZ and PN, were individually applied, and their degradation rates were quantified. The contents of the metal oxides were optimized by conducting CBZ degradation experiments in the range of 0.1-100 g L −1 of metal oxides ( Figure 1). The metal oxides appear to exhibit the highest removal of CBZ at 0.1 g L −1 among the investigated concentrations (1-100 g L −1 ), which contradicts the results obtained in the previous studies that revealed a positive effect of metal oxides on the degradation of target compounds in a concentration range of 0.1-2.0 g L −1 which showed the accelerated degradation of contaminants as the dosage of metal oxides increased [36,37]. This is presumably because of agglomeration of the excess metal-oxide catalysts, which decreases the number of active sites and consequently reduces the catalytic activity. In addition, a large quantity of catalysts can possibly reduce the amount of PMS available for the reactions [38][39][40][41]. Therefore, a metal-oxide concentration of 0.1 g L −1 was selected for the subsequent experiments. The degradation rates of CBZ and PN can be adequately described using the first-and second-order kinetics models, respectively; the rapid degradation reaction of PN enabled the latter ( Figure 2). Figure 2a shows that the presence of metal oxides accelerates CBZ degradation rates, with FeOOH exhibiting the fastest degradation, followed by Fe 3  be readily accessible for reacting with PMS by generating radicals in accordance with Equations (3)- (7), because of their crystalline structures [42,43].
Water 2021, 13, x FOR PEER REVIEW 4 of 12 the fastest degradation, followed by Fe3O4, Fe2O3, and MnO2. The Fe(III) component of FeOOH possibly exhibits enhanced catalytic activity for PMS activation compared to that of the other metal oxides, and/or the surface-hydroxylated FeOOH may be readily accessible for reacting with PMS by generating radicals in accordance with Equations (3)- (7), because of their crystalline structures [42,43].  However, optimal PN degradation was found to occur in the absence of metal oxides, with values of 0.29, 0.21, 0.19, and 0.14 × 10 −2 M −1 d −1 corresponding to Fe2O3, MnO2, Fe3O4, and FeOOH, respectively (Figure 2b). This indicates the potential of metal oxides as PMS activators for CBZ treatment, and their insignificant catalytic capability for PN degradation. Figure 3 suggests that the FeOOH-PMS treatment for CBZ degradation yields the highest concentration of SO4 2− ions, indicating the ability of FeOOH to effectively activate PMS by producing SO4 2− ions as products of PMS oxidation. Given that PN can be degraded relatively faster than CBZ by PMS-based technologies owing to its selective reactivity toward phenolic compounds [23], the amount of PMS added in the present study is presumed to be sufficient for independently decomposing PN, thereby not requiring additional catalytic assistance. Alternatively, this result indicates that the presence of borate buffer possibly obscures the capability of metal oxide catalysts during the relatively easier/faster PN degradation by the PMS-metal oxide systems. However, optimal PN degradation was found to occur in the absence of metal oxides, with values of 0.29, 0.21, 0.19, and 0.14 × 10 −2 M −1 d −1 corresponding to Fe 2 O 3 , MnO 2 , Fe 3 O 4 , and FeOOH, respectively (Figure 2b). This indicates the potential of metal oxides as PMS activators for CBZ treatment, and their insignificant catalytic capability for PN degradation. Figure 3 suggests that the FeOOH-PMS treatment for CBZ degradation yields the highest concentration of SO 4 2− ions, indicating the ability of FeOOH to effectively activate PMS by producing SO 4 2− ions as products of PMS oxidation. Given that PN can be degraded relatively faster than CBZ by PMS-based technologies owing to its selective reactivity toward phenolic compounds [23], the amount of PMS added in the present study is presumed to be sufficient for independently decomposing PN, thereby not requiring additional catalytic assistance. Alternatively, this result indicates that the presence of borate buffer possibly obscures the capability of metal oxide catalysts during the relatively easier/faster PN degradation by the PMS-metal oxide systems.

Identification of the Major Reactive Species in PMS Oxidation Systems in the Presence of Metal Oxides and Borate Buffer
The predominant reactive species during PN degradation by the PMS-metal oxide systems in borate buffer were determined by conducting scavenging experiments to clarify the role of borate buffer (Figure 4). The addition of TBA, which is known to efficiently eliminate • OH radicals (k •OH = 3.8-7.6 × 10 8 M −1 s −1 ), does not affect PN degradation, except with FeOOH (Figure 4a-d); moreover, no NB degradation occurs in any of the metal oxide systems (Figure 4e), which implies the lack of • OH radical generation in the assessed systems. Figure 4a-d suggest that the spiking of IPA, a scavenger for both • OH and SO 4 •− radicals (k •OH = 1.9 × 10 9 M −1 s −1 and k SO4•− = 8.5 × 10 7 M −1 s −1 ) [20,21], also does not cause any changes with respect to the PN degradation rates achieved without IPA. Therefore, the scavenging experiments reveal that the • OH and SO 4 •− radicals are not the primary reactive species produced in the PN-PMS systems with four different metal oxides and borate buffer. This observation contradicts previous studies that have confirmed • OH and SO 4 •− to be the significant reactive species participating in PN degradation by the PMS oxidant with additional catalysts such as CeVO 4 , coal-fly-ash-supported Co 3 O 4 , and α-Mn 2 O 3 [44][45][46]. This discrepancy can be attributed to the inhibition of the catalytic capability of metal oxides owing to the presence of borate buffer solution. This was confirmed by conducting a positive-control scavenging experiment using an Fe(II)SO 4 catalyst instead of the metal oxides. Figure 5 shows a clear difference in the degrees of PN degradation achieved in systems without the scavenger and the samples with TBA/IPA without the borate buffer system; however, the impact of the scavengers on the degree of PN degradation in the borate-buffered systems is insignificant. According to previous studies [24,28], it has been known that borate ions can stimulate PMS reactivity accelerating the degradation of target compounds at certain range of borate concentration (<100 mM) [28]. However, above 200 mM of borate ions hindered the degradation of contaminants since borate ions function as a consumer of PMS rather than a catalyst of PMS [24,28]. Based on the results in this study, it is hypothesized that borate ions (XO n m− ) may react with Lewis acid of metal oxides (M n+ ) instead of reacting with PMS [47]. Thus, when borate ions and metal oxides co-exist, metal oxides disable to exhibit its catalyst capability due to the interaction of metal oxides with borate ions.

Identification of the Major Reactive Species in PMS Oxidation Systems in the Presence of Metal Oxides and Borate Buffer
The predominant reactive species during PN degradation by the PMS-metal oxide systems in borate buffer were determined by conducting scavenging experiments to clarify the role of borate buffer (Figure 4). The addition of TBA, which is known to efficiently eliminate • OH radicals (k•OH = 3.8-7.6 × 10 8 M −1 s −1 ), does not affect PN degradation, except with FeOOH (Figure 4a-d); moreover, no NB degradation occurs in any of the metal oxide systems (Figure 4e), which implies the lack of • OH radical generation in the assessed systems. Figure 4a-  compounds at certain range of borate concentration (<100 mM) [28]. However, above 200 mM of borate ions hindered the degradation of contaminants since borate ions function as a consumer of PMS rather than a catalyst of PMS [24,28]. Based on the results in this study, it is hypothesized that borate ions (XOn m-) may react with Lewis acid of metal oxides (M n+ ) instead of reacting with PMS [47]. Thus, when borate ions and metal oxides coexist, metal oxides disable to exhibit its catalyst capability due to the interaction of metal oxides with borate ions.

Optimization of Borate Buffer Concentration
The optimal concentration of borate ions for enabling their use as a buffer solution and simultaneously not affecting PMS reactivity was determined by comparing the rates of PN degradation with respect to the presence and absence of metal oxides (Figure 6a-d) and monitoring changes in pH in terms of the various concentrations of borate ions investigated (Figure 6e). Borate-ion concentrations of 2 mM and 5 mM result in systems with Fe2O3 exhibiting accelerated PN degradation or no differences compared to that in systems without it, respectively. However, an increase in the borate-ion concentration from 10 mM to 20 mM leads to the metal-oxide-deprived systems exhibiting accelerated degra-

Optimization of Borate Buffer Concentration
The optimal concentration of borate ions for enabling their use as a buffer solution and simultaneously not affecting PMS reactivity was determined by comparing the rates of PN degradation with respect to the presence and absence of metal oxides (Figure 6a-d) and Water 2021, 13, 2698 8 of 11 monitoring changes in pH in terms of the various concentrations of borate ions investigated (Figure 6e). Borate-ion concentrations of 2 mM and 5 mM result in systems with Fe 2 O 3 exhibiting accelerated PN degradation or no differences compared to that in systems without it, respectively. However, an increase in the borate-ion concentration from 10 mM to 20 mM leads to the metal-oxide-deprived systems exhibiting accelerated degradation compared to that of systems with metal oxides. With respect to changes in pH during the reaction, the investigated borate-ion concentrations fail at maintaining the initial pH for 2 d, except for the borate concentration of 20 mM (Figure 6e). Therefore, while the catalytic activity of the metal oxides is not hindered at a borate-ion concentration of 2 mM, it cannot function as a buffer solution, which necessitates the use of an alternative buffer solution in PN-PMS degradation to avoid experimental artifacts. In addition, the ability of phosphate buffer solution (PBS), which is ubiquitously used to control pH in the range of 5. 8-7.4, to maintain the target pH during the BZA decomposition reaction and permit PMS reactivity was investigated (Figure 7). Although the presence of phosphate ions appears to hinder BZA degradation at a PBS concentration of 0.02 mM, the degree of reduction is not as significant as that with the borate buffer solution (Figure 7a). PMS and phosphate anions are known to react with each other and produce •OH radicals as the predominant radicals, thereby allowing the generation of radicals, unlike the behavior of borate buffer [48]. With respect to the time-evolution of pH, a PBS concentration of 0.2-2.0 mM can maintain the target pH for 2 d (Figure 7b). Therefore, the impact of oxyanions, such as borate and phosphate ions, on PMS reactivity must be considered in their application as a buffer solution with the PMS oxidant and metal oxide catalysts.

Conclusions
This study investigated the effects of borate buffer on the degradation of organics (CBZ and PN) by the PMS oxidant in the presence of various metal-oxide catalysts. The metal oxides acted as catalysts of PMS in CBZ degradation; however, they failed to exhibit catalytic activity for PN degradation. Scavenging experiments revealed that neither • OH

Conclusions
This study investigated the effects of borate buffer on the degradation of organics (CBZ and PN) by the PMS oxidant in the presence of various metal-oxide catalysts. The metal oxides acted as catalysts of PMS in CBZ degradation; however, they failed to exhibit catalytic activity for PN degradation. Scavenging experiments revealed that neither • OH nor SO 4 •− radicals were generated in PN degradation by the PMS oxidant in the presence of the borate buffer and metal oxides. This observation meant that the borate buffer solution was hypothesized to hinder radical generation by the PMS oxidant in the presence of metal oxides, especially during the relatively faster/easier degradation of target compounds such as PN. It is likely that borate ions may react with Lewis acid of metal oxides, thus, both borate and metal oxides disable playing a role as an activator of PMS. The optimal concentration of borate ions for controlling the pH and simultaneously facilitating catalytic activity was determined. Starting at 2 mM, borate ion already showed hindering PN degradation in the presence of Fe 2 O 3 and only 20 mM borate buffer solution was able to maintain the initial pH; thus, we failed to find the optimal borate concentration controlling pH simultaneously not affecting PMS reactivity. Phosphate ions were used as an alternative buffer, and exhibited a similar trend by affecting radical formation to an extent not as significant as that of the borate ions. Therefore, the use of an optimal concentration of oxyanion buffer (borate and phosphate ions) was found to be vital for maintaining a target pH and simultaneously permitting the catalytic activity of metal oxides in PMS oxidation, especially during the faster/easier degradation of target compounds.

Data Availability Statement:
The data presented this study are available upon request from the corresponding author.