Oxidative Transformation of Controlled Substances by Manganese Dioxide

This study investigated the oxidative transformation of four controlled substances (ketamine, methamphetamine, morphine, and codeine) by synthesized MnO2 (δ-MnO2) in aqueous environments. The results indicated that ketamine and methamphetamine were negligibly oxidized by MnO2 and, thus, may be persistent in the aqueous environment. However, morphine and codeine were able to be oxidized by MnO2, which indicated that they are likely naturally attenuated in aqueous environments. Overall, lower solution pH values, lower initial compound concentrations, and higher MnO2 loading resulted in a faster reaction rate. The oxidation of morphine was inhibited in the presence of metal ions (Mn2+, Fe3+, Ca2+, and Mg2+) and fulvic acid. However, the addition of Fe3+ and fulvic acid enhanced codeine oxidation. A second-order kinetics model described the oxidation of morphine and codeine by MnO2; it suggested that the formation of a surface precursor complex between the target compound and the MnO2 surface was the rate-limiting step. Although the target compounds were degraded, the slow TOC removal indicated that several byproducts were formed and persist against further MnO2 oxidation.


Introduction
The presence of pharmaceuticals in aqueous environments is an important environmental issue because conventional wastewater treatment plants (WWTP) do not remove them, resulting in their release into the environment [1,2]. Controlled substances are one type of pharmaceuticals commonly used in hospitals. Besides their medical applications, controlled substances are also used illicitly and abusively. These substances can be classified as antidepressants, stimulants, and hallucinogens based on their effects on the central nervous system. In addition, these compounds can cause significant toxicity after prolonged exposure because they are chemically and biologically active [3].
Ketamine, methamphetamine, morphine, and codeine are four controlled substances and have been detected in various waterbodies in different countries. For example, they were detected in hospital effluents (maximum concentrations of 10,000, 260, 1,240, and 378 ng/L, resp.), and ketamine, methamphetamine, and codeine were detected in river water in Taiwan (maximum concentrations of 341, 405, and 57 ng/L, resp.) [4,5]. Boleda et al. [6] reported that morphine and codeine were detected at concentrations of up to 81 and 397 ng/L, respectively, in WWTP effluents in Spain. Hummel et al. [7] investigated the occurrence of morphine and codeine in Germany and found that both of these compounds were present in river water (Rhine water) (78 ng/L; 94 ng/L), WWTP influents (820 ng/L; 540 ng/L), and effluents (110 ng/L; 260 ng/L). In addition, Castiglioni et al. [8] investigated the occurrence of morphine and methamphetamine in Italy and Switzerland, maximum concentrations of 204 and 16 ng/L, respectively, in the WWTP influent and 55 and 4 ng/L, respectively, in the WWTP effluent. The release of these controlled substances into the aqueous environment may endanger aquatic life and result in ecosystem contamination.
Natural attenuation (hydrolysis, redox reaction, sorption, photolysis, and biodegradation) is considered to be one of the most important pathways for removing contaminants from the aqueous environment and is the most economically 2 The Scientific World Journal feasible method for further attenuation of treated wastewater [9]. However, very limited information is currently available regarding the environmental fates of these four controlled substances (ketamine, methamphetamine, codeine, and morphine), and the oxidative transformation of these four controlled substances by manganese oxide (MnO 2 ) has not been previously investigated. MnO 2 is abundant in natural sediments and soils and has been reported to possess a powerful oxidative capacity for removing phenol, aniline, aliphatic amine, and triazine in aquatic environments through oxidation and sorption [10,11]. The objective of this study was to investigate the oxidative transformation of four controlled substances (ketamine, methamphetamine, morphine, and codeine) by synthesized MnO 2 ( -MnO 2 ) in an aqueous environment. Several critical environmental factors that may affect the MnO 2 oxidation rate were studied, including the MnO 2 loading, initial compound concentration, and solution pH. In addition, the effects of cosolutes (metal ions (Mn 2+ , Fe 3+ , Ca 2+ , and Mg 2+ ) and natural organic matter) on MnO 2 oxidation were investigated, and total organic carbon removal and oxidative transformation byproducts were studied. Methamphetamine was purchased from the US Pharmacopeial Convention (USP). Without any further purification, the purities of all drugs were greater than 95%. All other stock solutions were prepared in deionized water at concentrations of 500-1,000 mg/L and were stored at 4 ∘ C in a refrigerator. Other chemicals were purchased from Merck Millipore (Guyancourt, France), Avantor Performance Materials (Phillipsburg, NJ, USA), Sigma-Aldrich (St. Louis, MO, USA), and Nacalai Tesque (Kyoto, Japan) with purities greater than 85% (Table S1 in Supplementary Material available online at http://dx.doi.org/10.1155/2015/364170). The physicochemical properties of ketamine, methamphetamine, codeine, and morphine are listed in Table 1.

Synthesis of Manganese
Dioxide. -MnO 2 was synthesized according to the methods of Murray [12]. First, 80 mL of 0.1 M KMnO 4 and 160 mL of 0.1 M NaOH were added to 1.64 L of N 2 -sparged reagent water, and 120 mL of 0.1 M MnSO 4 was added to the solution while continuously stirring. After the MnO 2 particles settled, the supernatant solution was replaced with deionized water until the conductivity of the MnO 2 suspension was less than 2 S-cm −1 . The final volume of the MnO 2 solution was adjusted to 1 L, and then the solution was stored at 4 ∘ C. used to analyze the concentrations of the four target controlled substances. The HPLC module consisted of a pump (Agilent 1200 Series Binary Pump), degasser (Agilent 1200 Series Micro Vacuum Degasser), and autosampler (Agilent 1200 Series Autosampler). The analytes were determined using chromatography with a ZORBAX Eclipse XDB-C 18 column (150 × 4.6 mm, 5 m particle size) and a flow rate of 1 mL min −1 . Mobile phases A and B consisted of 0.05% formic acid with 10 mM ammonium acetate in DI water and methanol, respectively.

Oxidation of Controlled
Substances. MnO 2 oxidation experiments were conducted using 250 mL screw-cap amber glass bottles with aluminum foil septa at room temperature (22 ∘ C). The solution pH was controlled using a 10 mM acetate buffer (pH 4 and 5), 4-morpholinepropanesulfonic acid (MOPS) (pH 6 and 7), 2-(cyclohexylamino)ethanesulfonic acid (CHES) (pH 9), and a phosphate buffer (pH 9). In addition, NaCl (0.01 M) was added to maintain an appropriate ionic strength. The volume of the reactor was fixed at 100 mL. The reaction aliquots were collected (1 mL) and quenched using two different methods: the filtration quenching method and the reductant quenching method. In the reductant quenching method, oxalic acid (0.5 g/L) was added to desorb the target compound from the MnO 2 surface. The filtration quenching method can only detect target compounds dissolved in solution, not those adsorbed on the surface of manganese dioxide.

Analysis of Oxidative Byproducts.
To analyze the oxidative byproducts of codeine, an Agilent LC-ESI-MS/MS combined with a ZORBAX Eclipse XDB-C 18 column (150 × 4.6 mm, 5 m particle size) was used to perform quantitative analyses. Three steps were required to analyze the byproducts. First, the electron-spray ionization (ESI) detector was operated to obtain a full mass spectrum scan (Q1 scan) ranging from 50 to 600. Then, the selected sample was compared with the control sample to determine the possible oxidative byproduct. Second, the multiple-reaction monitoring transition mode (MRM) was used to optimize the parameters for MS/MS detection, including the declustering potential (DP), collision energy (CE), and collision cell exit potential (CXP). Third, the LC parameters were optimized to separate the analyte, including the mobile phase, flow rate, and injection volume.

Results and Discussion
Background tests (hydrolysis and adsorption) were performed before the MnO 2 oxidation experiment. The four controlled substances remained stable in the water at all tested pH conditions (pH 4 and 5 for ketamine and methamphetamine; pH 5, 7, and 9 for morphine; and pH 5, 6, 7, 8, and 9 for codeine) ( Figure S1). No obvious differences were observed between the filtration quenching method and the reductant quenching method ( Figure S2) and no measurable adsorption by MnO 2 was observed by the four controlled substances. The MnO 2 experiments revealed that ketamine and methamphetamine were persistent and were not oxidized by 10 mg/L MnO 2 . However, morphine and codeine were oxidized by MnO 2 , indicating that they may undergo attenuation in natural soil and sediment environments ( Figure 1). Based on the kinetic model reported by Zhang et al. [11], the oxidation of morphine and codeine follows second-order kinetics with adsorption as the rate-limiting step. The second-order kinetics of this process are shown in (1), where respectively. Because the adsorption reaction is considerably slower than the electron transfer reaction, the amount of target compound (morphine and codeine) adsorbed on MnO 2 but unreacted would be negligible for detection as observed by the experiments. Consider (1)

Effects of pH.
An initial environmentally relevant concentration was selected for the four controlled drugs (100 g/L). Ketamine and methamphetamine remained stable against MnO 2 oxidation (4 mg/L) at pH 4 and 5 for 48 h ( Figure  S3). However, the degradation behaviors of morphine and codeine were different from that of ketamine and methamphetamine. In this study, different MnO 2 loadings were used for morphine and codeine (100 g/L MnO 2 for morphine and 8 mg/L MnO 2 for codeine) because morphine is oxidized by MnO 2 much more readily than codeine. The results showed that morphine and codeine exhibited lower degradation rates as the solution pH decreased (Figure 2). Morphine was >99%, 91.5%, and 82% degraded at pH 5, 7, and 9, respectively (reaction time = 4 h, MnO 2 = 100 g/L). Codeine was >99%, 96%, 67%, 36%, and 28% degraded at pH 5, 6, 7, 8, and 9, respectively (reaction time = 1 h, MnO 2 = 8 mg/L). Previous studies indicated that the oxidation of compounds by synthesized MnO 2 is a surface reaction and that the surface charge of MnO 2 is negative (point of zero charge (PZC) of MnO 2 was 2.25) [12,13]. Therefore, the intensity of the negative MnO 2 charge increased as the pH increased from 5 to 9. This increase caused the MnO 2 surface to become more hydrophilic and subsequently reduced the number of accessible active sites for the target compound because of obstruction from the nearby water molecules, The Scientific World Journal which further reduced the reaction rate [14]. Lin et al. [15] also demonstrated that the ability of MnO 2 to oxidize organic compounds is pH-dependent. Furthermore, Stumm and Morgan [16] reported that the MnO 2 reduction potential decreased from 0.99 V to 0.76 V when the pH increased from 4 to 8. Zhang and Huang [17] also reported that more surface species are produced at lower pH that subsequently react with the reductants. In addition, Klausen et al. [18] indicated that protons are required to dissociate the reaction product Mn(II) from the MnO 2 (Mn(IV)) mineral surface ((1/2)Mn IV O 2(s) +2H + + e − (1/2)Mn 2+ + H 2 O). Therefore, a higher solution pH may result in a lower degradation rate. MnO 2 Loading. The results showed that ketamine and methamphetamine were still present after MnO 2 oxidation for 48 hrs unless the MnO 2 loading reached 750 mg/L ( Figure S4). High MnO 2 loading (750 mg/L) indicates that ketamine and methamphetamine are not degraded by natural MnO 2 in soils and sediments. For morphine and codeine, higher MnO 2 loadings resulted in a faster oxidation rate (Figure 3). The degradation rates of morphine by 250 and 500 g/L MnO 2 were faster than that by 100 g/L MnO 2 . Similarly, when MnO 2 increased from 1 mg/L to 8 mg/L, the degradation efficiency of codeine increased from 12% to 65% in 1 h. Zhang et al. [11] reported that a precursor complex between the organic compound and MnO 2 is formed before electron transfer reaction occurs during the redox reaction. These two steps (precursor complex formation and electron transfer reaction) were hypothesized to be rate-limiting steps and to influence the reactivities of organic compounds and MnO 2 . Thus, the oxidation of organic compounds by MnO 2 resulted from surface reactions between the MnO 2 and compounds [10]. Consequently, when the MnO 2 concentration increased, the potential for contact between the target compounds and MnO 2 increased and resulted in the reaction. Furthermore, Zhang and Huang [17] reported that higher MnO 2 concentrations offer more active surface sites, which resulted in an increase in the rate of precursor complex formation.

Effects of the Initial Compound Concentration.
It has been hypothesized that the formation of a surface complex is the rate-limiting step in the oxidation of morphine and codeine by MnO 2 . The formation rate of such a surface complex is affected by the MnO 2 loading, solution conditions (pH and ionic strength), and the initial concentration of the target compound. In this study, the initial compound concentration (morphine and codeine) range was 20 to 1,000 g/L. The results showed that increasing compound concentrations resulted in decreased degradation efficiency (Figure 4) when the MnO 2 loading and pH were fixed. A higher compound concentration could result in self-competition for the fixed number of active sites on the surface of MnO 2 , thereby decreasing the removal rate of morphine and codeine [11].

Effects of Metal
Ions. The effects of metal cations (Mn 2+ , Fe 3+ , Ca 2+ , and Mg 2+ ) on the oxidation of morphine and codeine by MnO 2 were investigated (Figures 5 and 6). Previous studies showed that dissolved cations in real water matrices may inhibit the degradation of compounds by MnO 2 [11,17]. Our results showed a similar phenomenon, in which the coexistence of metal ions resulted in an inhibitory effect on the oxidation of morphine and codeine by MnO 2 (except for the effect of Fe 3+ on the oxidation of codeine). and Mg 2+ ) increased. However, the opposite phenomenon was observed regarding the effects of Fe 3+ on the degradation of codeine (Fe 3+ enhanced the degradation efficiency of codeine by MnO 2 ). As the Fe 3+ concentration increased from 1 to 200 M, the degradation efficiency of codeine increased from 54.6% to 100%. One possible explanation for this result is that the addition of FeCl 3 at pH 7 could result in coagulation of MnO 2 particles, which might reduce the codeine concentration in the solution. The induced coagulation is more likely to occur at the higher MnO 2 loading (8 mg/L) applied for codeine. Further investigations should be performed to understand the detailed mechanisms of Fe 3+ toward codeine degradation by MnO 2 .

Effects of Natural Organic
Matter. Natural organic matter (NOM) is present in aqueous environments and is also a factor that may influence the oxidation efficiency of MnO 2 [18]. In this work, fulvic acid (FA) was used to represent the NOM in the natural environment. The effects of FA on the oxidation of morphine and codeine by MnO 2 are shown in Figure 7. Results showed that FA decreased the degradation efficiency of morphine, potentially because FA competes with target compounds to the active surface hydroxyl groups of MnO 2 , resulting in slowing the removal of morphine [21]. In contrast, greater concentrations of FA resulted in slightly greater codeine removal. This result implied that FA did not compete with codeine for the same reactivity sites on MnO 2 . Another possibility is that the presence of FA might facilitate some aggregation of the MnO 2 particles under the experimental conditions used for codeine (with 8 mg/L MnO 2 ). In the absence of MnO 2 , FA alone did not react with morphine or codeine (Figure 7), which indicated that the interaction between FA and the target compound (morphine or codeine) did not result in the removal of the target compound.  or transformed into other oxidative transformation byproducts by MnO 2 . According to Figure 8(a), although 93% and 54% of the carbon in morphine and codeine were degraded within 5 h, the TOC removals were only 12% and 33%, respectively. This result indicated that various degradation byproducts of morphine and codeine were formed. The degradation byproducts of codeine were investigated in a preliminary experiment, as shown in Figure 8(b). The signal corresponding to the byproducts did not decrease during the oxidation of codeine by MnO 2 , indicating that these byproducts may be persistent and resistant to oxidation by MnO 2 . In addition, the signal corresponding to the byproducts reached a plateau once the codeine concentration stopped decreasing. This result implied that the formation of byproducts would be affected by the rate of surface complex formation. Among all of the byproducts investigated, the molecular weights of the five byproducts (m/z = 575.4, 470.1, 597.2, 540.0, and 366.0) were higher than the parent compound codeine, indicating that these intermediates may be formed by the interaction and combination of the byproducts in the solution. The MS/MS spectra of the oxidative byproducts of codeine are shown in Figure S5. Further work is required to comprehensively investigate the formation of the degradation byproducts and The Scientific World Journal 9 their environmental fate and behavior in aqueous environments.

Conclusions
This study investigated the oxidative transformation of ketamine, methamphetamine, morphine, and codeine by MnO 2 . Ketamine and methamphetamine were stable against oxidation by MnO 2 until MnO 2 loading of 750 mg/L was used, indicating that ketamine and methamphetamine are negligibly degraded by natural MnO 2 in soils and sediments and may persist in the environment if no other oxidative pathways are involved. Morphine and codeine were oxidized by MnO 2 , which indicated that they can be naturally attenuated in aquatic environments with the presence of MnO 2 . A higher MnO 2 loading, a lower pH, and a lower compound concentration resulted in a higher efficiency of morphine and codeine oxidation by MnO 2 . The kinetic modeling results showed that the oxidation of morphine and codeine followed second-order kinetics and was limited by the rate of the surface precursor complex formation. The presence of metal ions (Mn 2+ , Fe 3+ , Ca 2+ , and Mg 2+ ) and FA inhibited the oxidation of morphine by MnO 2 . Although Mn 2+ , Ca 2+ , and Mg 2+ suppressed the oxidative efficiency of codeine, Fe 3+ and FA enhanced its degradation, but more research is needed to elucidate the mechanisms. In addition, although the target compounds were degraded, the slow removal of TOC indicated that several byproducts were formed and were even more persistent against further oxidation.