The materials and reagents were used as received without additional purification; they encompass titanium dioxide (TiO₂, Hombikat UV100 (Sachtleben Chemie GmbH) and Aeroxide P25 (Evonik)), tungsten(VI) oxide (WO₃, Sigma-Aldrich), ranitidine hydrochloride (C₁₃H₂₂N₄O₃S·HCl, Sigma, ≥98.0%), cimetidine (C₁₀H₁₆N₆S, 100%, Sigma), iron(II) perchlorate hydrate (Fe(ClO₄)₂· xH₂O, ≥98.0%, Aldrich), hydrogen peroxide (H₂O₂, 30.0%, Daejung), sodium acetate (C₂H₃NaO₂, 99.0%, Sigma-Aldrich), malonic acid (C₃H₄O₄, ≥99.0%, Sigma-Aldrich), sodium citrate tribasic dihydrate (C₆H₅Na₃O₇·2H₂O, ≥99.0%, Sigma-Aldrich), potassium oxalate monohydrate (C₂K₂O₄·H₂O, ≥99.0%, Fluka), Horseradish peroxidase (POD, type VI-A, Sigma-Aldrich), N,N-diethyl-p-phenylenediamine (DPD, C₁₀H₁₆N₂, 97.0%, Aldrich), ascorbic acid (C₆H₈O₆, 99.6%, Junsei), sodium azide (NaN₃, ≥99.5%, Sigma-Aldrich), tetranitromethane (CN₄O₈, 100%, Aldrich), coumarin (C₉H₆O₂, ≥99.0%, Sigma), phosphoric acid (H₃PO₄, 85.0%, Junsei), acetonitrile (CH₃CN, ≥99.0%, J. T. Baker), perchloric acid (HClO₄, 60%, Sigma-Aldrich), and sulfuric acid (H₂SO₄, ≥95.0%, Sigma-Aldrich). All suspensions and solutions were prepared using deionized water (resistivity > 18.0 MΩ·cm) produced by a water purification system (Millipore Milli-Q).

Hombikat UV100 TiO₂ powder was dispersed in deionized water through sonication for 2 min using an ultrasonic cleaner (SciLab). Aliquots of oxalate and ranitidine (or cimetidine) stock solutions were then introduced into the TiO₂ suspension. Subsequently, the pH of the suspension was adjusted to 5.0 using an HClO₄ solution. The suspension was stirred for 30 min before the addition of H₂O₂ to exclude the removal of ranitidine by adsorption from the total degradation. Finally, H₂O₂ was introduced into the TiO₂ suspension containing oxalate and ranitidine. This moment was considered the start of the reaction. Generally, the initial concentrations of TiO₂, H₂O₂, oxalate, and ranitidine were 15 mg/30 mL, 1 mM, 500 μM, and 100 μM, respectively.

The degradation reaction proceeded in a 50 mL beaker covered with aluminum foil, serving as the reactor, with continuous magnetic stirring under dark conditions. At specified time intervals, sample aliquots (1.5 mL) were extracted from the reactor and filtered through a 0.45 μm polytetrafluoroethylene filter (Millipore) to eliminate TiO₂ particles. The filtered samples were promptly measured. All experiments were replicated two or three times to validate data reproducibility, and the figures represent average values with standard deviations.

The concentration of ranitidine and cimetidine was assessed using high-performance liquid chromatography (HPLC, Agilent 1260 Infinity II) equipped with a Poroshell 120 EC-C18 column (4.6 mm × 150 mm) and a variable wavelength detector. The eluent comprised a binary mixture of acetonitrile and a 0.1% phosphoric acid solution, with a volume ratio of 5:95. The detection wavelength was set at 218 nm for cimetidine and 228 nm for ranitidine. The column temperature was maintained at 30 °C, and the eluent flowed at a rate of 1.0 mL/min. The concentration of total organic carbon (TOC) was determined using a TOC analyzer equipped with a nondispersive infrared sensor, serving as a CO₂ detector (Shimadzu TOC-L_CPH). The TOC concentration was calculated by subtracting the concentration of inorganic carbon from the total carbon concentration.

The concentration of H₂O₂ was determined using the colorimetric DPD method employing the DPD/POD reagent, which consists of 100 mg of DPD, 10 mg of POD, 10 mL of water, and 10 mL of H₂SO₄ (0.1 M) [26, 27]. Specifically, 3 mL of the 20-fold diluted sample was mixed with 0.4 mL of phosphate buffer solution (0.5 M) and 0.1 mL of DPD/POD reagent. The mixture in a conical tube was vigorously shaken for 1 min, and its absorbance was measured at 551 nm (ɛ of DPD^•+ = 21,000 M⁻¹ cm⁻¹) using a UV–visible spectrophotometer (Agilent Cary 60). To assess the production of ^•OH, the fluorescence emission intensity of 7-hydroxycoumarin was measured at 460 nm using a spectrofluorometer (Horiba FluoroMax 4), with 332 nm UV light employed as the excitation source [28, 29]. This compound is formed through the reaction of ^•OH with coumarin (1 mM) (^•OH + coumarin + O₂ → 7-hydroxycoumarin + HO₂^•). For measuring the production of HO₂^•, the absorbance at 350 nm of nitroform (ɛ = 14,600 M⁻¹ cm⁻¹ at 350 nm), formed from the reaction between tetranitromethane (50 μM) and HO₂^• (HO₂^• + tetranitromethane → nitroform + NO₂ + O₂ + H⁺), was measured using a UV–visible spectrophotometer (Agilent Cary 60) [30, 31].

The effect of oxalate on the degradation of ranitidine in the presence of TiO₂ and H₂O₂ was investigated and compared with control systems (Fig. 1a). The concentration of ranitidine diminished slowly in the TiO₂/H₂O₂ system without oxalate. Remarkably, the introduction of oxalate to the TiO₂/H₂O₂ system notably boosted the decomposition of ranitidine. After 4 h of reaction, 44.2% of ranitidine was degraded in the absence of oxalate, while the degradation reached 94.0% in the presence of oxalate. In terms of the pseudo-first-order degradation rate constant (k), the presence of oxalate resulted in a 5.7-fold increase in the degradation rate constant (i.e., 0.144 h⁻¹ without oxalate → 0.823 h⁻¹ with oxalate). In addition, the concentration of total organic carbon (TOC) in the TiO₂/H₂O₂/oxalate system decreased from 27.4 ppm to 24.0 ppm after 4 h of reaction, resulting in a TOC removal efficiency of 12.4%. Other control systems (i.e., TiO₂ alone, H₂O₂ alone, oxalate alone, TiO₂ + oxalate, and oxalate + H₂O₂) did not induce any degradation of ranitidine. These results indicate that the presence of oxalate in the TiO₂/H₂O₂ system could synergistically enhance the production of oxidant species, consequently resulting in the enhanced degradation of ranitidine.

We also tested the degradation of cimetidine, another model pharmaceutical pollutant, to verify the pervasiveness of the positive effect of oxalate in the TiO₂/H₂O₂ system (Fig. 1b). After 3 h of reaction, cimetidine was completely degraded in the presence of oxalate, while 40.3 μM of cimetidine remained in the absence of oxalate. Similar to the case of ranitidine, the existence of oxalate markedly elevated the degradation rate constant of cimetidine from 0.304 to 1.470 h⁻¹. This result implies that the positive effect of oxalate on the degradation process in the TiO₂/H₂O₂ system is pervasive and is not restricted to a specific pollutant.

It has been reported that the catalytic decomposition of H₂O₂ on metal oxides can generate HO₂^• and ^•OH [17, 18]. The generation of HO₂^• is induced by the formation of peroxo surface complexes of H₂O₂ on metal oxides (–OOH groups) (Eq. (1)), followed by subsequent inner-sphere electron transfer from the –OOH groups to the metal centers (Eq. (2)) [32]. In addition, upon the adsorption of H₂O₂ on the metal oxide surface through hydrogen bonding, the metal center can induce the homolytic cleavage of the O–O bond in H₂O₂, resulting in the production of ^•OH (Eq. (3)) [33].

To establish the role of oxalate in the TiO₂/H₂O₂ system, the production of radical species, such as HO₂^• and ^•OH, was measured and compared in the absence and presence of oxalate. The production of HO₂^•, directly associated with the formation of nitroform in the presence of tetranitromethane, was significantly accelerated in the presence of oxalate (Fig. 2a).

We also investigated the effect of oxalate in the TiO₂/H₂O₂ system on the production of ^•OH using fluorescence spectroscopy, with coumarin employed as the ^•OH-trapping reagent. Although the production of ^•OH was observed in both the absence and presence of oxalate, the quantity was minimal, and the difference was not significant. In comparison to the Fenton reaction (Fe²⁺ + H₂O₂ → Fe³⁺ + OH⁻ + ^•OH) [34], the production of ^•OH in the TiO₂/H₂O₂ system, regardless of the presence of oxalate, was much smaller (Fig. 2b). In contrast to this behavior, the degradation of ranitidine was more significant in the TiO₂/H₂O₂/oxalate system than in the Fe²⁺/H₂O₂ system (Fig. 2c). This result suggests that the involvement of ^•OH in the degradation of ranitidine within the TiO₂/H₂O₂/oxalate system was marginal.

To confirm that HO₂^• acts as the primary oxidant for the degradation of ranitidine in the TiO₂/H₂O₂/oxalate system, ROS-quenching experiments were conducted using various ROS scavengers, such as methanol for ^•OH (k = 9.7 × 10⁸ M⁻¹ s⁻¹) [35], ascorbic acid for HO₂^• (k = 1.6 × 10⁴ M⁻¹ s⁻¹) [36], and sodium azide for singlet oxygen (¹O₂) (k = 1 × 10⁹ M⁻¹ s⁻¹) [37]. As illustrated in Fig. 3, methanol (^•OH scavenger) exhibited an insignificant effect on the degradation of ranitidine, suggesting that ^•OH was not the primary oxidant in the TiO₂/H₂O₂/oxalate system. This finding is in agreement with the marginal production of ^•OH in the TiO₂/H₂O₂/oxalate system (Fig. 2b). In contrast, the addition of ascorbic acid (HO₂^• scavenger) reduced the degradation of ranitidine. This result implies that HO₂^• plays a significant role in the degradation of ranitidine in the TiO₂/H₂O₂/oxalate system. In addition, when sodium azide, acting as a ¹O₂ scavenger, was present, the degradation of ranitidine was retarded. The reaction between HO₂^• and O₂^•− (also two O₂^•−) produces ¹O₂ as another oxidant for ranitidine degradation, as described by Eqs. (4)–(6) [38, 39]. Based on the results of ROS production and ROS-quenching experiments, both HO₂^• and ¹O₂ can be considered as the primary oxidants in the TiO₂/H₂O₂/oxalate system. Although the oxidation power of ^•OH is higher than those of HO₂^• and ¹O₂ [40], weak oxidants, HO₂^• and ¹O₂, contribute more significantly to the degradation process in the TiO₂/H₂O₂/oxalate system because the production of ^•OH through Eq. (3) is minimal (Fig. 2b).

Fig. 4a depicts the time profiles of H₂O₂ decomposition induced by TiO₂, both in the absence and presence of oxalate, over reaction time. Interestingly, the catalytic decomposition of H₂O₂ was lower in the presence of oxalate compared to its absence. The contrasting behaviors observed in the presence of oxalate, higher production of HO₂^• (Fig. 2a) but lower decomposition of H₂O₂ (Fig. 4a), suggest the existence of another pathway for the production of HO₂^• in the presence of oxalate.

Oxalate can form a complex with the TiO₂ surface through monodentate and bidentate carboxylate linkages [41]. Due to the competition between H₂O₂ and oxalate for complex formation on the TiO₂ surface, a lower decomposition of H₂O₂ was observed in the presence of oxalate (Fig. 4a). Although >Ti^III (surface Ti^III species on TiO₂), which can act as a reductant, is generated through electron transfer from the –OOH groups to >Ti^IV (>Ti^IV–OOH → >Ti^III–^•O₂H), the reduction of oxygen (i.e., the formation of HO₂^•) by >Ti^III is not favored due to the more positive reduction potential of >Ti^III (E⁰(Ti⁴⁺/Ti³⁺) = 0.1 V_NHE) compared to the reduction potential of oxygen (E⁰(O₂/HO₂^•) = −0.05 V_NHE) [42, 43]. However, it has been reported that the complexation of oxalate with metal species decreases the reduction potential of metal species (i.e., increases the thermodynamic driving force for electron transfer) [22, 23]. For example, the reduction of oxamyl by Fe²⁺ was very slow, but it proceeded rapidly after complexation with oxalate (i.e., after the formation of Fe^II-oxalate complexes) [25]. Similarly, the complex between >Ti^III and oxalate (i.e., >Ti^III(oxalate)_n^(2n–3)−) would have a more negative reduction potential, facilitating the formation of HO₂^• through oxygen reduction (Eq. (7)).

To provide evidence for this hypothesis, the effects of various chelating agents in the TiO₂/H₂O₂ system on the degradation of ranitidine were investigated (Fig. 4b). Although all chelating agents, such as acetate, malonate, citrate, and oxalate, enhanced the degradation of ranitidine, the extent of the positive effect varied significantly depending on the type of chelating agent. The positive effects of chelating agents followed the order: oxalate > citrate > malonate > acetate. This order exactly corresponds with the sequence of lower reduction potential values of Fe^II-organic complexes (oxalate < citrate < malonate < acetate) [25]. The observed behaviors support the hypothesis that the enhanced production of HO₂^• in the presence of oxalate is attributed to the negative shift in reduction potential resulting from the complexation between >Ti^III and oxalate.

The degradation kinetics of ranitidine in the TiO₂/H₂O₂/oxalate system were investigated as functions of TiO₂ dosage, H₂O₂ concentration, and oxalate concentration (Fig. 5). The degradation rate (k) of ranitidine displayed a positive correlation with increasing TiO₂ dosage until it reached saturation at [TiO₂] = 1.0 g/L (Fig. 5a and b). A higher dosage of TiO₂ provides more interaction sites among TiO₂, H₂O₂, and oxalate, thereby enhancing the formation of TiO₂-H₂O₂ complexes and TiO₂-oxalate complexes. However, because the interaction sites on TiO₂ were completely occupied by H₂O₂ and oxalate at [TiO₂] = 1.0 g/L, a further increase in TiO₂ dosage had a minimal impact on the degradation kinetics of ranitidine.

The k value increased linearly with increasing H₂O₂ concentration (Fig. 5c and d). Higher concentrations of H₂O₂, serving as an HO₂^• precursor and producing >Ti^III as an oxygen reductant for HO₂^• generation, can increase the production of HO₂^• and ¹O₂, which can efficiently react with ranitidine. Therefore, the degradation was accelerated at higher concentrations of H₂O₂ in the TiO₂/H₂O₂/oxalate system.

Contrary to the degradation patterns observed with changes in TiO₂ dosage and H₂O₂ concentration, the k value increased drastically with increasing oxalate concentration up to 1 mM, but gradually decreased from 1 mM (Fig. 5e and f). Oxalate can lower the redox potential of >Ti^III by forming a complex with TiO₂, which facilitates the generation of HO₂^• through oxygen reduction. However, oxalate can compete with H₂O₂ for the formation of complexes on the TiO₂ surface and act as a scavenger for HO₂^• and ¹O₂ [31]. The dual function of oxalate helps to explain the observed initial rise and subsequent decline in the k value as the oxalate concentration increases in the TiO₂/H₂O₂/oxalate system.

To determine whether the positive effect of oxalate is observed in other metal oxide/H₂O₂ systems, the effects of oxalate on the degradation of ranitidine were investigated in the Aeroxide P25 TiO₂/H₂O₂ and WO₃/H₂O₂ systems. These results were then compared with those obtained in the Hombikat UV100 TiO₂/H₂O₂ system. Regardless of the type of metal oxide, the presence of oxalate enhanced the degradation of ranitidine (Fig. 6a). However, the extent of the positive effect of oxalate varied considerably depending on the type of metal oxide. The relative first-order degradation rate constants (k_rel), which represent the ratio of k with oxalate to k without oxalate, were 5.715, 4.114, and 1.443 in the cases of Hombikat UV100 TiO₂, Aeroxide P25 TiO₂, and WO₃, respectively (Table 1). The positive effect of oxalate in the TiO₂/H₂O₂ system was more pronounced than that in the WO₃/H₂O₂ system.

The points of zero zeta potential (PZZP) of Hombikat UV100 TiO₂ (pH 6.0) and Aeroxide P25 TiO₂ (pH 6.3) are higher, whereas that of WO₃ (pH 2.0) is lower than the pH of the catalyst suspension containing H₂O₂ and oxalate (5.0) [44, 45]. The positive surface charges of Hombikat UV100 TiO₂ and Aeroxide P25 TiO₂ can create favorable conditions for anionic oxalate complexation. However, the complexation of anionic oxalate on the negatively charged surface of WO₃ is not favored, resulting in the low positive effect of oxalate on the degradation of ranitidine. As expected, the adsorption of oxalate on the Hombikat UV100 TiO₂ surface was significantly higher than on the WO₃ surface (Fig. 6b). This behavior confirms that when the metal oxide possesses a more positive surface charge, the positive effect of oxalate on the degradation process in the metal oxide/H₂O₂ system is more pronounced by enhancing the complexation of anionic oxalate.

The positive effect of oxalate in the Hombikat UV100 TiO₂/H₂O₂ system was observed under acidic conditions (i.e., k_rel > 1.0 at pH 3.0 and 5.0); however, the effect of oxalate was negative under basic conditions (i.e., k_rel < 1.0 at pH 9.0 and 11.0) (Fig. 6c). This phenomenon is also associated with the pH-dependent surface charge of TiO₂. When the suspension pH is lower than the PZZP of Hombikat UV100 TiO₂ (pH 6.0), the surface charge becomes more positive, thereby enhancing the complexation of oxalate as the pH decreases. Therefore, the extent of the positive effect of oxalate on ranitidine degradation at pH 3.0 (k_rel = 7.679) was greater than that at pH 5.0 (k_rel = 5.715). However, the surface charge becomes negative when the suspension pH is higher than the PZZP of Hombikat UV100 TiO₂. At pH 9.0 and 11.0, the electrostatic repulsion between anionic oxalate and the negatively charged surface banishes the positive effect of oxalate on ranitidine degradation by preventing its complexation. Under these circumstances, the function of oxalate as a scavenger for HO₂^• and ¹O₂ inhibits the degradation of ranitidine, leading to k_rel values lower than 1.0 (k_rel = 0.698 at pH 9.0 and k_rel = 0.452 at pH 11.0).