Enhancing Aqueous Chlorate Reduction Using Vanadium Redox Cycles and pH Control

Abstract

Chlorate (ClO₃^–) is a toxic oxyanion pollutant from industrial wastes, agricultural applications, drinking water disinfection, and wastewater treatment. Catalytic reduction of ClO₃^– using palladium (Pd) nanoparticle catalysts exhibited sluggish kinetics. This work demonstrates an 18-fold activity enhancement by integrating earth-abundant vanadium (V) into the common Pd/C catalyst. X-ray photoelectron spectroscopy and electrochemical studies indicated that V^V and V^IV precursors are reduced to V^III in the aqueous phase (rather than immobilized on the carbon support) by Pd-activated H₂. The V^III/IV redox cycle is the predominant mechanism for the ClO₃^– reduction. Further reduction of chlorine intermediates to Cl^– could proceed via V^III/IV and V^IV/V redox cycles or direct reduction by Pd/C. To capture the potentially toxic V metal from the treated solution, we adjusted the pH from 3 to 8 after the reaction, which completely immobilized V^III onto Pd/C for catalyst recycling. The enhanced performance of reductive catalysis using a Group 5 metal adds to the diversity of transition metals (e.g., Cr, Mo, Re, Fe, and Ru in Groups 6–8) for water pollutant treatment via various unique mechanisms.

Short abstract

Chlorate reduction is substantially accelerated by vanadium redox cycles upon adding V^V or V^IV salts to Pd catalysts.

Introduction

The global annual production of sodium chlorate (NaClO₃) has exceeded 4 million tons for pulp and paper bleaching, pyrotechnics, weed control, and various other applications.¹ ClO₃^– is also a common byproduct from drinking water disinfection using NaClO or ClO₂² and wastewater treatment with electrochemical^3,4 or photochemical processes.^5,6 Upon entering the environment, ClO₃^– can adversely affect ecosystems. For instance, it exhibits high toxicity toward specific algal species in aquatic environments⁷ and can potentially suppress the nitrification process in soil.⁸ Human exposure to ClO₃^– through drinking water, dairy supply chain, and agricultural products can lead to various health effects, including thyroid dysfunction and methemoglobinemia.⁹ The World Health Organization,¹⁰ European Union (EU),¹¹ and China¹² have established the drinking water limit of 0.7 mg L^–1 for ClO₃^–, while the United States has set a health reference level of 0.21 mg L^–1, with a minimum reporting level of 0.02 mg L^–1.^13,14 Additionally, undesired formation of ClO₃^– also takes place in both essential and emerging electrochemical processes such as chloralkali,^15,16 water splitting,¹⁷ and seawater valorization.¹⁸ Thus, technological advances in chlorate reduction hold immense importance across various fields.

The exploration of platinum group metals (PGM) as catalysts for aqueous ClO₃^– reduction has been ongoing since the early 1990s.¹⁹ While PGM nanoparticles catalyze the clean reduction of ClO₃^– by H₂ into Cl^– and H₂O (eq 1), they typically demonstrate limited activity, necessitating high catalyst loadings.²⁰⁻²⁵

1

Previously, we have enhanced the reduction of oxyanions by incorporating Group 6–8 transition metals into PGM catalysts.²⁶⁻²⁹ Each secondary metal imparts novel and unique functionalities. For instance, Cr(OH)₃ formed in various PGM catalysts (e.g., Pd or Rh nanoparticles supported by carbon, alumina, or silica) remarkably enhanced the adsorption of oxyanions.²⁸ The reduced Mo and Re oxide species on the PGM catalysts substantially accelerated oxyanion reduction via Mo^IV/VI and Re^V/VII redox cycles, respectively.^27,30 The immobilization of Ru⁰ nanoparticles onto Pd/C or Rh/C initiated a synergistic working mode, in which Ru rapidly reduces ClO₃^– while Pd or Rh scavenges the Ru-passivating chlorine intermediates.²⁹ We were thus motivated by extending the exploration to Group 5 metals.

Vanadium (V) is renowned for its easy accessibility to four adjacent oxidation states ranging from +2 to +5. This feature makes V a versatile redox catalyst and an active participant in various chemical and electrochemical processes.³¹⁻³³ V is also attractive for application because it is the sixth most abundant element among the transition metals and the 20th most abundant element overall in the Earth’s crust.³⁴ Particularly for environmental application scenarios, V is found in various natural and polluted waters.³⁵⁻³⁸ In this study, by leveraging the redox property of V and integrating it with a common Pd/C catalyst, we substantially accelerated the reduction of aqueous ClO₃^–. In contrast to Group 6–8 metals, V exhibits distinct behavior by serving as an electron shuttle in the aqueous phase for ClO₃^– reduction. After the reaction, reduced V^III can be readily immobilized on Pd/C through pH adjustment. This work shows that simple chemical innovations can substantially enhance the catalyst activity and effectively capture the potentially harmful metal species.

Materials and Methods

General Information

Ultrahigh-purity H₂ gas (99.999%) was supplied by Airgas. Activated carbon (no. L11860) for Pd/C preparation, NaVO₃, VOSO₄, and V₂O₃ were used as received from Alfa Aesar. NaClO₃ and Na₂PdCl₄ were purchased from Sigma-Aldrich. All aqueous solutions were prepared with 18.2 MΩ cm^–1 Milli-Q water. The Pd/C catalyst was prepared using our recently developed all-in situ method, which has been validated in our previous studies.^28,29,39,40 Specifically, a magnetic stir bar, 40 mg of activated carbon, 400 mL of DI water, and the desired amount of Na₂PdCl₄ (dissolved in a stock solution) were sequentially placed in a 500 mL round-bottom flask. The flask was capped with a rubber stopper and sonicated for 1 min. Next, the suspension was stirred at 350 rpm for 5 min to facilitate the adsorption of Pd^II onto the carbon support. The suspension was then sparged with 1 atm H₂ for 5 min at room temperature to reduce adsorbed Pd^II to Pd⁰ on the carbon support. The H₂ gas was introduced through a 16 gauge stainless-steel needle penetrating the stopper. Another needle served as the sampling port and the gas outlet to a fume hood. The solid catalyst was collected on filter paper in air, dried by airflow in the fume hood, and used for the subsequent steps.

Catalytic ClO₃^– Reduction

In a 50 mL round-bottom flask, 25 mg of Pd/C and the desired amount of the V precursor (NaVO₃ or VOSO₄, dissolved in a stock solution) were mixed together with 50 mL of Milli-Q water. The solution pH was adjusted to 3.0 with 1 N (i.e., 0.5 M) H₂SO₄. The mixture was stirred under H₂ sparging for 15 min to ensure the complete reduction of the V precursor to lower oxidation states. Once the V reduction was complete, NaClO₃ was added to the flask to initiate the reaction. Aliquots were collected periodically and filtered using a 0.22 μm cellulose membrane. The ClO₃^– and Cl^– in the aliquots were quantified using a Dionex ICS-5000 ion chromatography (IC) with a conductivity detector. The anions were separated using an IonPac AS19 column at 30 °C, with a 20 mM KOH eluent at 1 mL min^–1. The dissolved V in the aqueous phase was monitored by inductively coupled plasma-mass spectroscopy (ICP-MS, Agilent 7700 series).

Characterization of Pd/C-Reduced Vanadium Species

After the NaVO₃ addition to the Pd/C suspension and 15 min of H₂ sparging, the mixture was promptly transferred to an anaerobic glovebag (98% N₂, 2% H₂; Coy Laboratories). The ICP-MS analysis indicated limited adsorption of V by Pd/C after H₂ exposure (see Results and Discussion). To facilitate X-ray photoelectron spectroscopy (XPS) characterization, the dissolved V was forced to be immobilized on Pd/C by evaporating the water at 110 °C (sand bath) within the glovebag. The dried powder was mixed with TiO₂ (as an internal standard with the Ti 2p_3/2 binding energy (BE) of 458.7 eV). The mixture was loaded onto copper conductive tape and placed inside an anaerobic tube. The tube was transferred into a glovebox connected to an XPS instrument (AXIS Supra Kratos Analytical) equipped with a monochromatized Al Kα source. Inside the glovebox, the sample was retrieved from the tube and directly loaded into the instrument for characterization. The XPS spectra were fitted using CasaXPS software (version 2.3.19), with constrained separations of spin–orbit coupling doublet peaks (7.33 eV for V 2p, 5.76 eV for Ti 2p, and 5.27 eV for Pd 3d) and constrained ratios of the doublet peak areas (2:1 for V and Ti 2p and 3:2 for Pd 3d).

To gain further insights into aqueous V speciation under the reducing environment by H₂–Pd/C, we employed a combination of rotating ring-disk electrode (RRDE) techniques and static cyclic voltammetry (CV) to characterize the oxidation states of V species at various reduction potentials. We used a three-electrode system, which included a gold RRDE working electrode, a platinum counter electrode, and a Ag/AgCl reference electrode, following a previously established procedure.⁴¹ The electrochemical experiments were performed using NaVO₃ and VOSO₄ solutions at pH 3.0 (the same pH for ClO₃^– reduction), with 0.5 M Na₂SO₄ as the background electrolyte. Prior to the experiments, the V precursor solutions underwent a 30 min purge with N₂ gas to remove dissolved O₂. Throughout the experiments, a continuous flow of N₂ gas was maintained above the solution to prevent the ingress of O₂. In the CV experiments, the disk electrode potential was scanned over the range from +0.40 to −1.35 V, with the scan rate ranging from 25 to 200 mV s^–1. The disk current was recorded. In the RRDE study, the potential of the disk electrode was scanned within a targeted range, while the ring electrode potential remained fixed at a specific value. The rotating speed of the ring-disk electrode ranged from 400 to 2700 rpm. Both the disk and ring currents were recorded. Details regarding the assignment of V reduction peaks in the voltammograms are provided in Text S1 of the Supporting Information.

Results and Discussion

Vanadium Substantially Enhances ClO₃^– Reduction

The addition of either a V^V (NaVO₃) or V^IV (VOSO₄) precursor to Pd/C substantially enhanced ClO₃^–reduction by 18-fold compared to bare Pd/C, with both precursors exhibiting equal activity (Figure 1a and Figure S1). At pH 3.0 and 20 °C, the 1 wt % V–1 wt % Pd/C catalyst at 0.5 g L^–1 accomplished >99.9% reduction of 1 mM ClO₃^– within 0.5 h. We postulated that Pd/C could potentially reduce V^V and V^IV to the same active species (i.e., V^III or V^II, characterized in the next section) for ClO₃^– reduction. NaV^VO₃ was used as the catalyst precursor for further studies. To optimize the catalyst formula, we investigated how the varying Pd and V contents affect ClO₃^– reduction activity using our recently developed method, which allows for the “instant” preparation of Pd/C with any metal content.³⁹ By maintaining a constant V content of 1 wt %, we observed that the increasing Pd content raised the apparent rate constant (Figure 1b). However, surpassing 1 wt % Pd did not result in a proportional acceleration of ClO₃^– reduction. In fact, the rate constant normalized by the total Pd mass decreased, most likely due to decreased Pd dispersion.^29,39 Notably, no ClO₃^– reduction was observed when Pd was absent (i.e., carbon only, Figure 1b), indicating the role of Pd nanoparticles in harnessing electrons from H₂.

Figure 1.

(a) Time profiles of ClO₃^– reduction by Pd/C added with V^V and V^IV precursors, relationships between ClO₃^– reduction rate constants and (b) Pd and (c) V contents of the catalysts, (d) time profiles of ClO₃^– reduction with V–Pd/C catalysts prepared by allowing various time intervals between adding NaVO₃ under H₂ and adding ClO₃^–, (e) chlorine balance during ClO₃^– reduction, and (f) time profiles of ClO₃^– reduction by V–Pd/C at various pH. Unless specified, all reactions used 0.5 g L^–1 1 wt % V–1 wt % Pd/C, 1 mM ClO₃^–, pH 3.0, 1 atm H₂, and 20 °C.

When the Pd content was fixed at 1 wt % and the V content was progressively increased from 0 to 1 wt %, we observed a proportional increase in the reaction rate. However, beyond 1 wt % V, a disproportional acceleration in ClO₃^– reduction was observed (Figure 1c). For the subsequent studies, we used 1 wt % for both Pd and V. The highest ClO₃^– reduction activity was observed when V^V–Pd/C was exposed to H₂ for 15 min or longer before adding ClO₃^– (Figure 1d). Shorter H₂ exposure (e.g., 5 min) led to a slower reaction. Hence, the reduction of V^V by Pd-activated H₂ is crucial for ClO₃^– reduction, and it can be achieved within 15 min at ambient pressure and temperature.

The V–Pd/C ranks among the top-tier catalysts in terms of the ClO₃^– reduction rate (Table S1). The only catalyst known to exhibit greater activity is our previously reported Ru–Pd/C.²⁹ However, V is a much more earth-abundant and readily available metal than Ru. V^V oxyanions are commonly found in natural waters³⁵⁻³⁸ and may enhance catalysis. A good mass balance was observed between ClO₃^– and Cl^–, indicating that ClO₃^– was reduced to Cl^– without accumulation of Cl-containing intermediates (Figure 1e). At the challenging ClO₃^– concentrations of 10 and 100 mM, the catalyst exhibited remarkable performance, achieving >99% ClO₃^– reduction within 4 and 8 h, respectively (Figure S2). We also evaluated the performance of V–Pd/C in synthetic brines containing 2 M Cl^– and 1 M SO₄^– ions, respectively. Despite the observed activity loss in both scenarios, ClO₃^– reduction still completed within 4 h (Figure S3). Thus, V–Pd/C can be potentially applied to treat concentrated ClO₃^– under specific scenarios. The activity of V–Pd/C was pH-dependent with faster reactions at lower pH (Figure 1f). Similar trends have been observed from Re–Pd/C and Mo–Pd/C catalysts.^27,30,42 The proton can facilitate electron transfer in the deoxygenation of V-coordinated ClO₃^–27,30,42 and the reduction of the V site to lower oxidation states.^43,44

Characterization of Reduced Vanadium Species

ICP-MS analysis revealed that Pd/C had a limited adsorption capacity for V^V (Figure 2a). When 1 wt % V^V (corresponding to 0.1 mM in solution versus 0.5 g L^–1 Pd/C) was added, it was fully adsorbed by Pd/C. However, after a brief H₂ exposure, most adsorbed V was released into solution again (Figure 2b). Thus, the reduced V species could not be readily immobilized on Pd/C. To characterize the prevailing oxidation states of V in the H₂+Pd/C system, we evaporated the water at 110 °C in an anaerobic glovebag and forced V to deposit on the Pd/C powder. XPS analysis observed the V 2p_3/2 BE at 515.6 eV (Figure 2c), which is consistent with the previously reported V^III.⁴⁵ The BE value also matched the major peak BE at 515.6 eV of the V^III₂O₃ standard (a minor BE peak at 516.5 eV corresponded to V^IV due to oxidation).⁴⁶ Therefore, the V^V precursor was reduced by H₂+Pd/C to V^III. XPS also confirmed metallic Pd⁰ (3d_5/2 BE of 335.44 eV, Figure S4) under the H₂ atmosphere.

Figure 2.

(a) Fraction of V^V adsorbed on Pd/C (0.5 g L^–1, 1 wt % V^V if fully adsorbed), (b) aqueous V concentration after adding NaVO₃ (5 mg L^–1 as V) into the Pd/C suspension under varying conditions, (c) V 2p XPS spectra of the V^III₂O₃ reference and V–Pd/C prepared from NaVO₃, and (d) linear sweep voltammetry of V^IVO²⁺ on an RDDE with a scan rate of 50 mV s^–1. The ring electrode potential was fixed at 0.8 V. Dashed lines represent the linear sweep voltammetry of the background electrolyte at 50 mV s^–1 and 2700 rpm; (e) cyclic voltammetry of V^IVO²⁺ on a gold disk electrode. Dashed lines represent the voltammograms of the background electrolyte at 50 mV s^–1. Cyclic scan on the electrode started in the cathodic direction followed by the anodic direction. Other electrochemical study conditions for (d) and (e): 2 mM VOSO₄, 0.5 M Na₂SO₄, 1.5 M ionic strength, and pH 3.0.

To corroborate the XPS results, we used an electrochemical method to investigate the oxidation state of aqueous V after reduction. To simulate the H₂+Pd/C system, V^V was incrementally reduced on an electrode with a potential range that covered the actual redox potential range of the H₂+Pd/C system. The speciation of aqueous V^V varies depending on pH and concentration.^41,47 Visual MINTEQ simulation (Figure S5) suggested that the dominant species was VO₂⁺ (93.2%) in our optimized reaction system (pH 3.0 and 1 wt % V, equivalent to 0.1 mM V). The fraction of VO₂⁺ decreased as the V concentration increased (Figure S5). To enhance the signal and maintain VO₂⁺ as the main species (>90%), we performed the initial electrochemical study at pH 3.0 with a concentration of 0.2 mM NaV^VO₃, which was twice the optimum value. CV and RRDE results demonstrated that the reduction of V^V started at 0.35 V (Figures S6–S9). We analyzed the correlation between peak currents and scan rates in CV and Koutecky–Levich plots (Text S1, Calculation 1).⁴⁸ Results showed that V^V underwent one-electron transfer to V^IV, predominantly as V^IVO²⁺.

However, V^IV was not the reactive species responsible for the rapid reduction of ClO₃^– (see evidence in the next section). When we attempted to further reduce V^IV at a more negative potential, the low concentration of V (0.2 mM) failed to generate a notable signal due to the pronounced interference from H₂ evolution. Consequently, we used 2 mM V^IVOSO₄ to further resolve other V species. CV and RRDE analysis (Text S1, Calculation 2) indicated that V^IV was initially reduced to V^III within the range of −0.78 V to −0.55 V, prior to H₂ evolution, and subsequently reduced to V^II at −1.21 V after H₂ evolution (Figure 2d,e). Thus, V^II cannot be generated by using H₂ as the reductant. Therefore, V^III was the final reduction product from V^V and was responsible for the rapid reduction of ClO₃^–. This is consistent with the XPS characterization.

Vanadium Redox Cycles and ClO₃^– Reduction Mechanisms

Once the V precursor is reduced to V^III, it could rapidly reduce ClO₃^–. The reaction between V^IV and ClO₃^– could occur, but it was very slow. The homogeneous reaction between V^IVO²⁺ (1 mM) and ClO₃^– (0.1 mM) was negligible within 0.5 h (Figure 3a,b), which is the time frame for the complete reduction of 1 mM ClO₃^– using 1 wt % V (0.1 mM)–1 wt % Pd/C (0.5 g L^–1, Figure 1a). In stark contrast, the homogeneous reaction between 1 mM V^III (reduced from V^V by H₂+Pd/C and then separated by filtering off Pd/C) and 0.1 mM ClO₃^– was observed to reach completion within the same time frame (Figure 3a). Given that the reduction of ClO₃^– by bare Pd/C is sluggish (Figure 1a), the activity of V–Pd/C is primarily attributed to V^III. According to a previous report, the reaction between V^III and ClO₃^– predominantly occurs through one-electron transfer, where V^III is oxidized to V^IVO²⁺ and ClO₃^– is reduced to ClO₂.⁴⁹ Notably, the reported second-order rate constant for ClO₃^– reduction in the previous homogeneous system (pH ∼0 and 20 °C) was 3.2 M^–1 s^–1, lower than that from our V–Pd/C system (15.0 M^–1 s^–1 at pH 3 and 20 °C, Figure 1f). This comparison also shows that integration with H₂+Pd/C enables efficient catalysis without highly acidic conditions. The reduction of V^IVO²⁺ to V^III by H₂+Pd/C completes redox cycle 1 (Figure 3c). V^III can also react with ClO₂ and further intermediates (e.g., ClO₂^–, ClO^–, and Cl₂), through one-electron transfer,⁴⁹ ultimately to Cl^– (Figures 1e and 3d).

Figure 3.

(a,b) Time profiles for 0.1 mM ClO₃^– reduction by 1 mM V^III and V^IV, (c) proposed catalytic cycles for the reduction of ClO₃^– and the reaction intermediates, and (d) overall reaction pathway for the reduction of ClO₃^– to Cl^–.

However, the degradation of those intermediates does not solely use V^III in cycle 1. First, V^IVO²⁺ generated from cycle 1 has substantially higher reactivity with the intermediates compared to ClO₃^–. For example, the rate constants for the reaction of V^IVO²⁺ with ClO₂ and ClO^– are 500 and 550 times higher, respectively, than that with ClO₃^–.⁴⁹ The reaction of V^IVO²⁺ with these intermediates also follows a one-electron transfer mechanism and yields V^VO₂⁺.⁵⁰ The reduction of V^VO₂⁺ to V^IVO²⁺ by H₂+Pd/C closes redox cycle 2 (Figure 3c). Second, it is worth noting that while Pd/C is sluggish in reducing ClO₃^–, it rapidly reduces the intermediates.²⁹ In summary, the reduction of parent ClO₃^– primarily occurs through cycle 1, yielding ClO₂. The degradation of the intermediates (e.g., ClO₂, ClO₂^–, ClO^–, and Cl₂) can proceed via redox cycle 1 or 2, or they can be directly reduced by Pd/C, eventually generating Cl^– (Figure 3d).

Aqueous Vanadium Capture and Reuse

As mentioned earlier, while 0.1 mM V^V could be completely adsorbed on the Pd/C, V^III was released into the aqueous phase during the reaction (Figure 2b). To contain the potentially toxic V species in the catalytic system, we aimed to immobilize V onto carbon after the ClO₃^– reduction. V^III readily hydrolyzes⁵¹ and the addition of base (e.g., NH₄OH)⁵² can effectively precipitate V^III(OH)₃. Our Visual MINTEQ simulation found that at pH 8.0, over 99% of V^III existed as a V^III(OH)₃ solid (Figure 4a). We adjusted the solution pH from 3.0 to 8.0 and achieved the rapid and complete immobilization of V^III onto Pd/C. To prevent V^III oxidation by O₂, pH adjustment and filtration were performed under anaerobic conditions. ICP-MS analysis found that the aqueous V went below the detection limit of 1 μg L^–1. Although a small portion of V^III may still persist in its hydrolyzed form (e.g., [V(OH)]²⁺, [V(OH)₂]⁺, and [V₂(OH)₂]⁴⁺), the increased pH can modify the catalyst surface charge to more negative values,²⁴ potentially facilitating the adsorption of those cations. After the pH adjustment, we collected the V^III–Pd/C solid by filtration. The subsequent redispersion into water at pH 3 resumed the high activity (Figure 4b).

Figure 4.

(a) Calculated aqueous V^III speciation (5 mg L^–1 as V, Visual MINTEQ ver. 3.1) and (b) reduction of 1 mM ClO₃^– by fresh and recycled V–Pd/C catalysts. Reaction conditions: 0.5 g L^–1 1 wt % V–1 wt % Pd/C, pH 3.0, 1 atm H₂, and 20 °C.

Implications to Catalyst Development for Water Treatment

In the aqueous phase, V precursors (NaV^VO₃ and V^IVOSO₄) can be efficiently reduced to V^III by Pd-activated H₂. The rapid and complete reduction of ClO₃^– is primarily achieved by the redox cycling between V^III and V^IVO²⁺ as a homogeneous process, but V^III can be effectively contained in the catalytic system via pH control. V is naturally present in a diverse range of earth materials^53,54 and can enter the water environment through mechanical and chemical weathering processes.^35,36 The global demand for high-grade steel also keeps releasing V-enriched industrial byproducts into the environment.^{37,38,55−57} Therefore, V from either natural or anthropogenic sources in the affected water could accelerate ClO₃^– reduction. In the absence of V cocontaminants, the simple addition of 0.1 mM V can substantially facilitate the reduction of up to 100 mM ClO₃^–. The addition, capture, and recycling of V afford a low-cost and sustainable technical option for ClO₃^– reduction. The reduction and immobilization of V also suggest a novel approach for V removal from water using PGM catalysts and H₂.

Compared to the original Pd/C, the added earth-abundant and low-cost V reduced the amount of expensive Pd/C by 18-fold (Figure 1a) while maintaining the same reaction rate, thus enhancing the cost-effectiveness of the catalyst system. This study demonstrates the great application potential of incorporating redox-active metals to enable new reaction pathways for water pollutant degradation. Our research has investigated various metals in Groups 5 (V), 6 (Cr and Mo), 7 (Re), and 8 (Ru) for enhanced oxyanion reduction with supported PGM catalysts. Each metal follows unique mechanisms through enhancing oxyanion adsorption (e.g., Cr²⁸), creation of novel reaction pathways (e.g., V, Mo,²⁷ and Re³⁰), or synergistic work (e.g., Ru²⁹). Notably, the reaction pathways created by V, Mo, and Re differ remarkably. V reacts with ClO₃^– through a one-electron transfer process (Figure 3c), whereas Mo and Re oxometallates reduce ClO₃^– via oxygen atom transfer (eqs 2 and 3).

2

3

These findings affirm the opportunities for developing effective and sustainable water technologies with chemical innovation,¹⁹ particularly the exploration of the periodic table. Notably, the oxide species of the other two Group 5 metals, niobium (Nb) and tantalum (Ta), have very different properties from V. For example, NaNbO₃ and NaTaO₃ are not water-soluble.⁵⁸ Generating aqueous Nb and Ta species requires excessive acid,⁵⁹ base,^60,61 or complexing agents.⁶² Thus, integrating Nb and Ta with PGM catalysts via the add-and-use strategy is challenging and warrants future research efforts.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.est.3c06519.

• Comparison with other catalysts; kinetic figures for the effects of V precursors, chlorate concentrations, and common inert anions; XPS spectra; calculated V speciation using Visual MINTEQ; details of electrochemical studies (PDF)

Author Contributions

J.G. conducted kinetic experiments, analyzed the data, and drafted the manuscript; G.C. and H.L. conducted electrochemical studies; Q.F. conducted kinetic experiments; C.R. assisted in XPS characterization; C.T. measured aqueous V concentrations; J.L. conceived the idea, conducted initial kinetic studies with Y.W., supervised the research, and revised the manuscript.

The authors declare no competing financial interest.