Boosting Toluene Oxidation over Ru-Doped CoMn₂O₄ Spinel Catalysts by Constructing Ru–O–Mn/Co Chains

Abstract

The development of efficient spinel oxide catalysts for low-temperature oxidation of volatile organic compounds (VOCs) remains an important research objective. In this work, Ru was doped into a CoMn₂O₄ spinel to enhance its catalytic activity toward toluene oxidation and the underlying promotion mechanism of Ru doping was systematically investigated. The resulting Ru-CoMn₂O₄ catalyst showed remarkable performance, with T₉₀ reaching approximately 224 °C at a WHSV of 60,000 cm³ g⁻¹ h⁻¹ and nearly 100% CO₂ selectivity above 200 °C. Mechanism studies revealed that the reaction followed both Mars–van Krevelen (MvK) and Eley–Rideal (E–R) pathways. The reaction rates were strongly influenced by the oxidizing capacity of the catalyst, the abundance of highly valent surface species (namely Co³⁺, Mn⁴⁺, and Ru⁴⁺), adsorbed toluene, lattice oxygen, gaseous toluene, and adsorbed oxygen. With Ru doping, new Ru–O–Mn and Ru–O–Co chains formed in the CoMn₂O₄ spinel structure, leading to a moderate enhancement in oxidizing ability and a moderate increase in the concentration of highly valent surface species, adsorbed toluene, and lattice oxygen. Although a slight reduction in adsorbed oxygen was observed, Ru doping significantly boosted the overall toluene oxidation activity of CoMn₂O₄. In summary, Ru-CoMn₂O₄ represented a promising catalyst for the efficient oxidation of VOCs.

1. Introduction

Volatile organic compounds (VOCs), which originate from a wide range of industrial processes and human activities, play a critical role in the formation of secondary pollutants such as fine particulate matter and ozone [1,2]. These compounds exert considerable adverse impacts on human health and the environment [3]. Catalytic oxidation has emerged as one of the most promising technologies for VOCs removal owing to its high efficiency and operational reliability [4]. Nevertheless, the central challenge of this technology lies in developing highly active catalysts capable of completely mineralizing VOCs into harmless CO₂ and H₂O at relatively low temperatures [5].

Recently, spinel oxides, composed mainly of earth-abundant and low-cost transition metals, have been widely investigated as efficient and stable catalysts for the oxidation of VOCs [6]. Their attractiveness stems from a tunable elemental composition, flexible structure, and high thermal and chemical stability [7]. Among various spinel oxides, CoMn₂O₄ has emerged as a particularly prominent catalyst for VOCs oxidation, owing to its outstanding activity and robust stability [8,9]. However, the practical application of CoMn₂O₄ spinel is still constrained by the requirement for relatively high temperatures to achieve complete VOCs conversion and high CO₂ selectivity [10]. Therefore, engineering novel reaction pathways within the spinel structure is critical to improving low-temperature catalytic performance.

In spinel-type oxides (AB₂O₄), the A and B metal sites, occupying tetrahedral and octahedral positions, respectively, can be modulated by incorporating different metal cations, which directly influences the catalytic performance [11]. For instance, Ren et al. found that La doping not only improved the mobility and reactivity of lattice oxygen species by weakening metal–oxygen bond, but also facilitated the activation and replenishment of gaseous oxygen [12]. Hence, La-doped CoMn₂O₄ exhibited enhanced toluene oxidation performance. Similarly, Dong et al. reported that Pt doping modified the local electronic environment of Co sites, promoting toluene adsorption and the low-temperature activation and transformation of lattice oxygen species in CoMn₂O₄, leading to a notable improvement in toluene oxidation [10]. Despite these advances in metal cation doping, these doped CoMn₂O₄ materials still exhibit inadequate performance in the catalytic oxidation of VOCs, alongside prohibitive material costs.

Owing to its strong redox capability, low light-off temperature, and superior tolerance to H₂O, Ru has been extensively studied as a catalyst for the oxidation of VOCs [13], offering a cost-effective advantage over other precious metals like Pt, Au, and Pd [14]. Gao et al. deposited Ru nanoparticles of varying sizes on a CeO₂ support and observed that smaller Ru nanoparticles exhibited a tendency to readily redisperse, exposing a higher proportion of active Ruⁿ⁺ species and oxygen vacancies [15]. These smaller nanoparticles also demonstrated an enhanced ability to promote dehydrogenation and aromatic ring cleavage, ultimately leading to a significant improvement in toluene oxidation activity. Ding et al. employed a Ru/SnO₂ catalyst for the oxidation of vinyl chloride [16]. Given that both RuO₂ and SnO₂ shared a rutile crystal structure, the Ru species formed highly dispersed RuO₂ particles on SnO₂. These dispersed particles efficiently adsorbed chlorine species during the reaction, thus preserving the oxygen vacancies from chlorination. This synergistic mechanism, where oxygen vacancies facilitated oxidation and RuO₂ sites captured chlorine, resulted in significantly enhanced catalytic activity, stability, and byproduct selectivity.

In this work, Ru was doped into the CoMn₂O₄ spinel structure to further enhance its catalytic performance toward toluene oxidation, which served as a model VOCs, and the promotional effect of Ru doping on the reaction mechanism was systematically explored. The oxidation rates of toluene over CoMn₂O₄ and Ru-CoMn₂O₄ were found to be strongly influenced by the catalysts’ oxidizing capacity, the abundance of highly valent surface species (i.e., Co³⁺, Mn⁴⁺, and Ru⁴⁺), as well as the concentrations of surface adsorbed toluene, gaseous toluene, surface lattice oxygen, and surface adsorbed oxygen. The incorporation of Ru promoted the formation of Ru–O–Mn and Ru–O–Co linkages within the spinel framework, which moderately enhanced the oxidative capability and increased the concentration of highly valent surface species, surface adsorbed toluene, and surface lattice oxygen. Although a slight decrease in surface adsorbed oxygen was detected, Ru doping significantly improved the overall catalytic activity of CoMn₂O₄ toward toluene oxidation. These results demonstrate that Ru-CoMn₂O₄ may be a highly promising catalyst for efficient VOCs removal.

2. Results and Discussion

2.1. Performance for Toluene Oxidation

2.1.1. Activity

CoMn₂O₄ exhibited moderate activity in toluene oxidation, with a T₉₀ value (the temperature required for 90% conversion) of approximately 255 °C (Figure 1a). However, its performance was slightly inferior to that of the CoMn₂O₄ (fresh) (see Figure S1a), suggesting that the chemical reagents used during the reduction process may have slightly suppressed its catalytic activity. After loading Ru via conventional impregnation, the catalytic activity of Ru/CoMn₂O₄ decreased significantly, showing a T₉₀ value substantially above 260 °C (Figure 1a). In contrast, Ru-CoMn₂O₄ demonstrated excellent activity, achieving a T₉₀ of only about 224 °C (Figure 1a), significantly lower than those of CoMn₂O₄, Ru/CoMn₂O₄, and CoMn₂O₄ (fresh). These results indicate that incorporating Ru via doping, rather than by conventional surface loading or exposure to chemical agents, markedly enhanced the activity of CoMn₂O₄ in oxidizing toluene. Additionally, Ru-CoMn₂O₄ demonstrated superior activity for toluene oxidation compared to other catalysts reported in the literature (see Table S1).

Since H₂O is commonly present alongside VOCs in flue gas [17], its influence on toluene oxidation over Ru-CoMn₂O₄ was also evaluated. As shown in Figure 1a, the introduction of H₂O led to a slight increase in the T₉₀ value for Ru-CoMn₂O₄ to approximately 231 °C, indicating that the presence of H₂O mildly inhibited the catalytic oxidation of toluene. Nevertheless, Ru-CoMn₂O₄ still maintained excellent performance in toluene oxidation even in the presence of H₂O.

2.1.2. CO₂ Selectivity

The ideal products of toluene oxidation are CO₂ and H₂O [18]; therefore, the CO₂ selectivity of Ru-CoMn₂O₄ was investigated. As depicted in Figure 1b, CoMn₂O₄ showed low CO₂ selectivity (below 25%) in the temperature range of 160–220 °C, likely due to the formation of intermediate products. When the temperature was raised to 240 °C, the CO₂ selectivity of CoMn₂O₄ reached nearly 100%. Impregnation with Ru moderately improved the CO₂ selectivity of CoMn₂O₄, and the CO₂ selectivity of Ru/CoMn₂O₄ reached approximately 18% and 60% at 200 and 220 °C, respectively (Figure 1b). In comparison, Ru-CoMn₂O₄ exhibited excellent CO₂ selectivity during toluene oxidation, achieving nearly 100% even at 200 °C (Figure 1b). These findings indicate that incorporating Ru via doping also significantly enhanced the CO₂ selectivity of CoMn₂O₄. However, its CO₂ selectivity was slightly inferior to that of the CoMn₂O₄ (fresh) (see Figure S1b), suggesting that the chemical reagents used during the reduction process may have slightly suppressed its CO₂ selectivity. Additionally, the introduction of H₂O had negligible impact on the CO₂ selectivity of Ru-CoMn₂O₄ (Figure 1b).

2.1.3. Stability

To evaluate the stability of Ru-CoMn₂O₄ in toluene oxidation, a long-term test was conducted at 240 °C. As shown in Figure 2a, Ru-CoMn₂O₄ exhibited excellent stability, maintaining a toluene conversion of approximately 99%. Meanwhile, the CO₂ selectivity also remained stable at around 99% throughout the reaction (Figure 2b). Furthermore, the introduction of H₂O did not compromise the catalytic performance. Both toluene conversion and CO₂ selectivity were consistently maintained at approximately 99% under humid conditions (Figure 2), demonstrating the robust water-resistant stability of Ru-CoMn₂O₄.

2.2. Characterization

2.2.1. Crystal Structure

The crystal structures of CoMn₂O₄, Ru/CoMn₂O₄, and Ru-CoMn₂O₄ were analyzed by XRD. The XRD pattern of CoMn₂O₄ (Figure 3a) exhibited a perfect match with the reflection peaks of the pure tetragonal CoMn₂O₄ phase (JCPDS: 77-0471), and its space group was I4₁/amd [19]. Furthermore, no discernible differences were observed between the XRD patterns of CoMn₂O₄ and CoMn₂O₄ (fresh) (see Figure S2), indicating that the chemical reagents used in the reduction process had minimal influence on the crystal structure. After conventional impregnation with Ru, the XRD pattern still corresponded well to that of CoMn₂O₄, with no detectable peaks attributable to Ru species (Figure 3a). Although the XRD pattern of CoMn₂O₄ remained largely unchanged after Ru incorporation via doping, a slight shift toward higher angles was observed (Figure 3a). Concurrently, the lattice parameter of CoMn₂O₄ decreased from a = 5.72 Å and c = 9.24 Å to a = 5.67 Å and c = 9.15 Å. These results suggest that Ru was successfully incorporated into the surface crystal structure of CoMn₂O₄ spinel.

2.2.2. Surface Analysis

The surface elemental compositions and valence states of CoMn₂O₄ and Ru-CoMn₂O₄ were analyzed by XPS. The Co 2p_3/2 binding energies for CoMn₂O₄ were observed at 779.7, 781.5, and 786.1 eV (Figure 4a), corresponding to Co³⁺, Co²⁺, and a satellite peak associated with Co²⁺, respectively [20,21]. The Mn 2p_3/2 binding energies for CoMn₂O₄ were observed at 641.2 and 643.0 eV (Figure 4b), assigned to Mn³⁺ and Mn⁴⁺, respectively [22,23]. The O 1s binding energies for CoMn₂O₄ were observed at 529.5 and 531.5 eV (Figure 4c), attributed to surface lattice oxygen (O_latt) and surface adsorbed oxygen (O_ad), respectively [24,25]. In the spinel structure, Mn³⁺ cations preferentially occupied octahedral sites. Partial oxidation of Mn³⁺ to Mn⁴⁺ at these sites, driven by the Jahn-Teller effect, induced significant local lattice distortion [26]. This distortion weakened the Mn–O bond strength and enhanced the mobility of lattice oxygen, facilitating its participation in the oxidation reaction [27]. In contrast, Co²⁺ cations predominantly resided in tetrahedral sites. Due to stronger orbital overlap between metal d-states and O 2p orbitals in octahedral coordination, the octahedral sites exhibited higher oxygen binding capacity compared to tetrahedral sites [26]. Therefore, Mn cations at octahedral sites were likely the primary active sites for toluene oxidation, while Co cations may serve as auxiliary active sites.

After the introduction of Ru via doping, no significant changes were observed in the XPS spectra of CoMn₂O₄ corresponding to the Co 2p_3/2, Mn 2p_3/2, and O 1s regions (Figure 4a–c). In contrast, the Ru 3p_3/2 region of Ru-CoMn₂O₄ displayed weak peaks at 463.3 and 464.9 eV (Figure 4d), indicative of the presence of Ru⁰ and Ru⁴⁺ species [28,29]. Moreover, the HAADF-STEM image (Figure 5a) and EDX elemental mapping (Figure 5b–e) confirmed the homogeneous distribution of Ru, Co, Mn, and O across the surface of Ru-CoMn₂O₄. These findings suggest that Ru species were highly dispersed on the CoMn₂O₄ surface.

The surface element compositions of CoMn₂O₄ and Ru-CoMn₂O₄, as determined by XPS, were summarized in

Table 1. Percentages of Co, Mn, Ru, and O species on CoMn₂O₄, Ru-CoMn₂O₄, and spent Ru-CoMn₂O₄/%.

	Co3+	Co2+	Mn4+	Mn3+	Ru4+	Ru0	Olatt	Oad
CoMn2O4	4.12	3.81	8.97	10.9	-	-	37.5	34.7
Ru-CoMn2O4	4.82	3.21	10.1	10.5	0.95	0.52	38.5	31.4
spent Ru-CoMn2O4	4.60	3.92	9.58	10.8	-	-	46.4	24.7

. Following Ru doping, the percentage of Co³⁺ in CoMn₂O₄ moderately increased from 4.12% to 4.82%, while that of Co²⁺ moderately decreased from 3.81% to 3.21% (Table 1). Concurrently, the binding energies of Co³⁺ and Co²⁺ in CoMn₂O₄ slightly increased from 779.7 eV and 781.5 eV to 780.0 eV and 781.8 eV, respectively (Figure 4a). These results indicate an interaction between Ru and Co, involving electron transfer from Co to Ru, which led to a decrease in electron density around Co sites. Similarly, the percentage of Mn⁴⁺ also showed a slight increase after Ru doping, accompanied by a corresponding decrease in Mn³⁺ (Table 1). This change was also associated with a slight shift of the Mn³⁺ and Mn⁴⁺ binding energies to 641.1 eV and 642.8 eV, respectively (Figure 4b), suggesting analogous electron transfer from Mn to Ru and a corresponding reduction in electron density around Mn. To further probe the influence of Ru on the local coordination environments of Co and Mn in CoMn₂O₄, structural models of both CoMn₂O₄ and Ru-CoMn₂O₄ were constructed and geometrically optimized. Previous studies have shown that super-exchange interactions between Mn and Co ions via bridging oxygen atoms facilitated electron delocalization in the CoMn₂O₄ lattice, thereby reducing overall electron localization—a feature inherent to the Co–O–Mn structural motifs in the spinel framework (Figure 6) [30]. Moreover, the Ru₄ cluster was found to form chemical bonds at the CoMn₂O₄ interface (Figure 6). The Ru–O–Mn bridge exhibited a significantly higher formation energy (6.44 eV) than the Ru–O–Co bridge (4.66 eV), indicating stronger bonding interaction via the Mn–O bond in stabilizing the Ru₄ cluster (Figure 6). The charge density difference analysis (see Figure S3) revealed that in pristine CoMn₂O₄, the Bader charges of Mn and Co were −1.76 e and −1.30 e, respectively (see Table S2). After Ru incorporation and the formation of Ru–O–Mn and Ru–O–Co bridges, these values increased to −1.72 e and −1.28 e, respectively (see Table S2), consistent with a reduction in their oxidation states. This confirmed that Ru promoted the oxidation of Mn³⁺ to Mn⁴⁺ and Co²⁺ to Co³⁺. Furthermore, more pronounced charge redistribution was observed along the Ru–O–Mn pathway. Bader charge analysis further indicated that upon Ru₄ adsorption, Mn experienced greater electron depletion (Δρ = 1.70 e) than Co (Δρ = 1.28 e), suggesting stronger electronic coupling between Ru₄ and Mn (Figure 6). This result aligned with the enhanced charge redistribution observed in the Δρ map along the Ru–O–Mn path, as well as the higher oxygen vacancy formation energy on the Mn bridge (Figure 6). Collectively, these findings demonstrated that the Ru–O–Mn bridge played a dominant role in the interfacial interaction, while the Ru–O–Co bring played a secondary role.

After toluene oxidation at 240 °C for 20 h, the percentages of Mn⁴⁺ and Co³⁺ in Ru-CoMn₂O₄ both decreased significantly (Table 1). However, no distinct Ru 3p signal was detected for Ru–CoMn₂O₄ after the reaction (Figure 4d), likely due to the low Ru content (approximately 0.75%). As a result, the valence state of Ru species could not be unambiguously determined, and their percentages were therefore not included in Table 1.

2.2.3. Redox Behavior

The reduction behaviors of CoMn₂O₄ and Ru-CoMn₂O₄ were characterized by H₂-TPR. The H₂-TPR profile of CoMn₂O₄ exhibited three distinct reduction peaks (Figure 3b). The peak at 171 °C was assigned to the reduction of Mn⁴⁺ to Mn³⁺, the one at 306 °C was attributed to the concurrent reductions of Co³⁺ to Co²⁺ and Mn³⁺ to Mn²⁺, and the peak at 420 °C was ascribed to the reduction of Co²⁺ to Co⁰ [31]. Upon incorporation of Ru, a new shoulder peak emerged at 120 °C (Figure 3b), corresponding to the reduction of Ru⁴⁺ to Ru⁰ [32]. Moreover, the reduction peaks associated with Mn⁴⁺ to Mn³⁺, Co³⁺ to Co²⁺, and Mn³⁺ to Mn²⁺ all shifted slightly to lower temperatures. These results indicate a synergistic interaction between Ru and CoMn₂O₄, which moderately enhanced the redox capability of CoMn₂O₄.

2.3. Mechanism of Toluene Oxidation

The mechanism of toluene oxidation over Ru-CoMn₂O₄ primarily followed two pathways: the Mars–van Krevelen (MvK) mechanism and Eley–Rideal (E–R) mechanism [33,34]. In the MvK mechanism, gaseous toluene was first adsorbed onto the catalyst surface and then oxidized by surface lattice oxygen, leading to the formation of final products. The surface lattice oxygen consumed was subsequently replenished by gaseous oxygen. In contrast, the E–R mechanism involved the direct reaction between gaseous toluene and adsorbed oxygen on the catalyst surface to yield the oxidation products.

Toluene oxidation over Ru-CoMn₂O₄ via the MvK mechanism can be approximated as:

$${\mathrm{C}}_7{\mathrm{H}}_{8\left(\mathrm{g}\right)}\to {\mathrm{C}}_7{\mathrm{H}}_{8\left(\mathrm{a}\mathrm{d}\right)}$$

(1)

$${\mathrm{C}}_7{\mathrm{H}}_{8\left(\mathrm{a}\mathrm{d}\right)}+36\equiv {\mathrm{M}}^{n+}+18{\mathrm{O}}_{\mathrm{latt}}^{2-}\to 7{\mathrm{CO}}_2+4{\mathrm{H}}_2\mathrm{O}+36\equiv {\mathrm{M}}^{(n-1)+}$$

(2)

$$4\equiv {\mathrm{M}}^{(n-1)+}+{\mathrm{O}}_{2\left(\mathrm{g}\right)}\to 4\equiv {\mathrm{M}}^{n+}+2{\mathrm{O}}_{\mathrm{latt}}^{2-}$$

(3)

where Mⁿ⁺ represents the highly valent surface species, namely Co³⁺, Mn⁴⁺, and Ru⁴⁺.

Toluene oxidation over Ru-CoMn₂O₄ via the E–R mechanism can be approximated as:

$${\mathrm{O}}_{2\left(\mathrm{g}\right)}\to 2{\mathrm{O}}_{\left(\mathrm{ad}\right)}$$

(4)

$${\mathrm{C}}_7{\mathrm{H}}_{8\left(\mathrm{g}\right)}+18{\mathrm{O}}_{\left(\mathrm{ad}\right)}\to 7{\mathrm{CO}}_2+4{\mathrm{H}}_2\mathrm{O}$$

(5)

To explore the contribution of the MvK mechanism to the oxidation of toluene over Ru-CoMn₂O₄, in situ DRIFTS analysis was performed by introducing O₂ over Ru-CoMn₂O₄ that was pre-adsorbed with toluene at temperatures ranging from 160 to 240 °C. When gaseous toluene was introduced at 160 °C, twelve distinct bands were observed on the Ru-CoMn₂O₄ surface at 1020, 1073, 1144, 1178, 1305, 1416, 1448, 1492, 1538, 1558, 1595, and 3073 cm⁻¹ (Figure 7a). The bands at 1448, 1492, and 1595 cm⁻¹ were assigned to the stretching vibrations of the skeletal C=C bonds in the aromatic ring of toluene [31], while the band at 3073 cm⁻¹ was associated with the C–H stretching vibration of toluene [35], indicating toluene adsorption onto the Ru-CoMn₂O₄ surface. The bands at 1020, 1073, and 1144 cm⁻¹ were attributed to C–O stretching vibrations in benzyl alcohol [36,37], suggesting the oxidation of toluene to benzyl alcohol by surface lattice oxygen. Furthermore, the band at 1178 cm⁻¹ was ascribed to benzaldehyde species [38], suggesting the oxidation of benzyl alcohol to benzaldehyde. Additionally, the bands at 1416 and 1538 cm⁻¹ corresponded to symmetric and asymmetric stretching vibrations of the C=O bond in benzoate species [39,40], indicating further oxidation of benzyl alcohol to benzoate. The band at 1305 cm⁻¹ was assigned to the symmetric C=O stretching vibration in maleic anhydride [41], pointing to subsequent oxidation of benzoate. Moreover, the band at 1558 cm⁻¹ was associated with the stretching vibration of COO^– groups from acetate species [25], suggesting further oxidation of maleic anhydride. As the temperature gradually increased to 260 °C, the intensities of these intermediate-related bands initially increased and then diminished (Figure 7a), indicating that the intermediates were further oxidized into final products (primarily CO₂ and H₂O). These results demonstrate that toluene adsorbed on Ru-CoMn₂O₄ can be oxidized by surface lattice oxygen, ultimately yielding end products. Therefore, the MvK mechanism played a crucial role in the oxidation of toluene over Ru-CoMn₂O₄.

According to Reaction 2, the toluene oxidation rate over Ru-CoMn₂O₄ via the MvK mechanism (denoted as δ_MvK) can be approximated as:

$${\delta }_{\mathrm{MvK}}=-{\displaystyle \frac{\mathrm{d}[{\mathrm{C}}_7{\mathrm{H}}_{8\left(\mathrm{g}\right)}]}{\mathrm{d}t}}=k_1{[\mathrm{C}}_7{\mathrm{H}}_{8\left(\mathrm{ad}\right)}]{[\equiv {\mathrm{M}}^{n+}]}^{\alpha }{[{\mathrm{O}}_{\mathrm{latt}}^{2-}]}^{\beta }$$

(6)

where k₁, [C₇H_8(ad)], [Mⁿ⁺], [O_latt²⁻], α, and β represent the rate constant of Reaction 2, the number of surface adsorbed toluene molecules, Mⁿ⁺ cations, surface lattice oxygen, and the reaction orders with respect to Mⁿ⁺ cations and surface lattice oxygen, respectively.

Based on Reaction 5, the toluene oxidation rate over Ru-CoMn₂O₄ via the E–R mechanism (denoted as δ_E–R) can be approximated as:

$${\delta }_{\mathrm{E}–\mathrm{R}}=-{\displaystyle \frac{\mathrm{d}[{\mathrm{C}}_7{\mathrm{H}}_{8\left(\mathrm{g}\right)}]}{\mathrm{d}t}}=k_2[{\mathrm{C}}_7{\mathrm{H}}_{8\left(\mathrm{g}\right)}]{[{\mathrm{O}}_{\left(\mathrm{ad}\right)}]}^{\gamma }$$

(7)

where k₂, [C₇H_{8 (g)}], [O_(ad)], and γ denote the rate constant of Reaction 5, the number of gaseous toluene molecules in the flue gas, surface adsorbed oxygen, and the reaction order with respect to surface adsorbed oxygen, respectively.

Therefore, the overall toluene oxidation rate over Ru-CoMn₂O₄ can be described as:

$$\begin{array}{l}\delta ={\delta }_{\mathrm{MvK}}+{\delta }_{\mathrm{E}–\mathrm{R}}\\ =k_1{[\mathrm{C}}_7{\mathrm{H}}_{8\left(\mathrm{ad}\right)}]{[{\mathrm{M}}^{n+}]}^{\alpha }{[{\mathrm{O}}_{\mathrm{latt}}^{2-}]}^{\beta }+k_2{[\mathrm{C}}_7{\mathrm{H}}_{8\left(\mathrm{g}\right)}]{[{\mathrm{O}}_{(\mathrm{ad})}]}^{\gamma }\end{array}$$

(8)

The toluene concentration in the flue gas was typically high (~500 ppm), indicating that the adsorption of toluene on Ru-CoMn₂O₄ was nearly saturated. Thus, the number of surface adsorbed toluene molecules can be considered constant. Moreover, Mⁿ⁺ and O_latt²⁻ on Ru-CoMn₂O₄ can be speedily recovered through Reaction 3, and thus the amounts of Mⁿ⁺ and O_latt²⁻ on Ru-CoMn₂O₄ were rapidly replenished via Reaction 3, so their numbers were also regarded as constants. Furthermore, the O₂ concentration in the flue gas was significantly higher (~5%) than that of toluene, meaning that the consumption of O_(ad) due to toluene oxidation (Reaction 5) was negligible. Therefore, the number of O_(ad) can be treated as constant. As indicated by Equation 8, the overall oxidation rate was postulated to vary linearly with the gaseous toluene concentration. The slope and intercept of this linear correlation were attributed to the kinetic constants of the E–R (k_E–R) and MvK (k_MvK) mechanisms, respectively.

Thus, Equation (8) can be rewritten as:

$$\delta ={\delta }_{\mathrm{MvK}}+{\delta }_{\mathrm{E}–\mathrm{R}}=k_{\mathrm{MvK}}+k_{\mathrm{E}–\mathrm{R}}[{\mathrm{C}}_7{\mathrm{H}}_{8\left(\mathrm{g}\right)}]$$

(9)

To determine the kinetic parameters for toluene oxidation via the E–R and MvK mechanisms, a kinetic study of toluene oxidation over Ru-CoMn₂O₄ was performed in the temperature range of 180–260 °C, with toluene conversion kept below 15% (Figure 8). A linear regression analysis based on Equation (9) was applied, and the resulting intercept, slope, and regression coefficient were summarized in

Table 2. Reaction kinetic parameters for toluene oxidation over Ru-CoMn₂O₄.

	Temperature/°C	/μmol g−1 min−1
		δE–R	δMvK	R2
Ru-CoMn2O4	180	0.0044	4.02	0.998
	200	0.0095	5.94	0.998
	220	0.064	7.85	0.998
	240	0.158	12.2	0.998
	260	0.364	19.5	0.998

.

As shown in Table 2, the intercept values, corresponding to the MvK rate constant (k_MvK) were approximately 4.02, 5.94, 7.85, 12.2, and 19.5 µmol g⁻¹ min⁻¹ at 180, 200, 220, 240, and 260 °C, respectively. Moreover, the slope values, representing the E–R rate constant (k_E–R) were about 0.0044, 0.0095, 0.064, 0.158, and 0.364 µmol g⁻¹ min⁻¹ at the corresponding temperatures (Table 2). These results indicate that toluene oxidation over Ru-CoMn₂O₄ proceeded through both the MvK and E–R mechanisms. According to Equation (7), the oxidation rate via the E–R mechanism showed a linear dependence on toluene concentration. At toluene concentrations of approximately 914, 625, 123, 77, and 54 ppm at 180, 200, 220, 240, and 260 °C, respectively, the oxidation rates via the E–R and MvK mechanisms were roughly equal, suggesting comparable contributions from both pathways under these conditions. When the toluene concentration fell below these threshold values at each temperature, the oxidation rate via the MvK mechanism exceeded that via the E–R mechanism, indicating a more dominant role of the MvK pathway at lower concentrations. Conversely, at concentrations above these levels, the E–R mechanism became the predominant pathway. Given that typical toluene concentrations in flue gas were around 500 ppm, the MvK mechanism contributed more significantly at 200 °C, whereas the E–R mechanism dominated from 220 to 260 °C. Therefore, the oxidation of toluene over Ru-CoMn₂O₄ was influenced by both temperature and reactant concentration: the MvK mechanism dominated under lower temperatures and concentrations, while the E–R mechanism gained prominence at higher elevated temperatures and concentrations.

To further elucidate the reaction pathway of toluene oxidation over Ru-CoMn₂O₄ via the E–R mechanism, in situ DRIFTS measurements were performed during exposure of the catalyst to a mixture of toluene and O₂ at 160–260 °C. In addition to characteristic bands of adsorbed toluene (1448, 1492, 1595, and 3073 cm⁻¹), benzyl alcohol (1020, 1073, and 1144 cm⁻¹), benzaldehyde (1178 cm⁻¹), benzoate species (1416 and 1538 cm⁻¹), maleic anhydride (1305 cm⁻¹), and acetate species (1558 cm⁻¹), one band emerged at 1662 cm⁻¹ on Ru-CoMn₂O₄ (Figure 7b), corresponding to the C=O stretching vibration of benzaldehyde [42]. This observation suggests that the intermediate products formed during toluene oxidation over Ru-CoMn₂O₄ via the E–R mechanism included mainly benzyl alcohol, benzaldehyde, benzoate, maleic anhydride, and acetate species. Therefore, the reaction pathway of toluene oxidation via the E–R mechanism was probably similar or less intricate than that involving the MvK mechanism.

On basic of combined in situ DRIFTS and kinetic analyses, it was determined that a dual-mechanism coupling, involving both MvK and E–R routes, governs toluene oxidation over Ru-CoMn₂O₄. The corresponding pathway, depicted in Figure 9, comprised the following elementary steps: (1) Adsorption of gaseous toluene onto the Ru-CoMn₂O₄ surface. (2) Oxidation of both surface adsorbed and gaseous toluene to benzyl alcohol, activated by surface lattice oxygen and surface adsorbed oxygen, respectively. (3) Replenishment of surface consumed oxygen by gaseous oxygen from the feed. (4) Subsequent oxidation of benzyl alcohol to intermediates including benzaldehyde, benzoate, maleic anhydride, and acetate species. (5) Final oxidation of these intermediates, releasing CO₂ and H₂O as the end products.

2.4. Promotion Mechanism of Ru Doping on Toluene Oxidation over CoMn₂O₄

As indicated by Equation 8, the oxidation rate of toluene over Ru-CoMn₂O₄ exhibited a strong dependence on several parameters: the rate constants k₁ (for Reaction 2) and k₂ (for Reaction 5), as well as the concentrations of Mⁿ⁺ (i.e., Co³⁺, Mn⁴⁺, and Ru⁴⁺) cations, surface adsorbed toluene, surface lattice oxygen, gaseous toluene, and surface adsorbed oxygen. Since the concentration of gaseous toluene was unaffected by Ru doping, the enhancement in toluene oxidation activity on CoMn₂O₄ due to Ru doping was primarily attributed to an increase in the values of k₁ and k₂, enhanced toluene adsorption, and higher concentrations of Mⁿ⁺ cations, surface lattice oxygen, or surface adsorbed oxygen.

The rate constants k₁ and k₂ were closely associated with the oxidizing capabilities of CoMn₂O₄ and Ru-CoMn₂O₄. H₂-TPR analysis revealed a moderate improvement in the oxidizing ability of CoMn₂O₄ after Ru doping (Figure 3b), implying that both k₁ and k₂ for CoMn₂O₄ increased moderately following Ru modification. Furthermore, XPS analysis showed that the percentages of Co³⁺, Mn⁴⁺, and Ru⁴⁺ were all marginally higher on Ru-CoMn₂O₄ than those on CoMn₂O₄ (Table 1), indicating a moderate increase in the number of Mⁿ⁺ cations after Ru doping. Additionally, the percentage of O_latt was slightly greater for Ru-CoMn₂O₄ compared to CoMn₂O₄ (Table 1), suggesting a minor increase in surface lattice oxygen and a corresponding slight decrease in surface adsorbed oxygen after Ru doping.

To evaluate the effect of Ru doping on toluene adsorption, toluene-TPD profiles of CoMn₂O₄ and Ru-CoMn₂O₄ were compared. As shown in Figure 10, the amount of toluene adsorbed (calculated from the total desorbed toluene in Figure 10a and formed CO₂ in Figure 10b) on CoMn₂O₄ was approximately 66.5 μmol g⁻¹, whereas it increased to about 78.2 μmol g⁻¹ for Ru-CoMn₂O₄, representing a 1.2-fold enhancement. Hence, toluene adsorption onto CoMn₂O₄ was moderately promoted by Ru doping. To gain further insight into the interaction between toluene and CoMn₂O₄ under the influence of Ru doping, the toluene adsorption properties of both CoMn₂O₄ and Ru-CoMn₂O₄ were investigated via DFT calculations. On basic of the tetragonal CoMn₂O₄ phase identified by XRD (Figure 3a), the (211) crystal surface was identified as the dominant facet of the spinel structure. The adsorption energy (E_ads) of toluene molecules on this surface was evaluated with different exposed atoms. The most stable configurations were those in which carbon atoms of the aromatic ring (C_a) interacted with Mn sites, forming Mn–C_a bonds with lengths of 2.83 Å and 2.94 Å, yielding an E_ads of −1.47 eV for CoMn₂O₄. In contrast, for Ru-CoMn₂O₄, the E_ads value was −1.64 eV, and the corresponding Mn–C_a bond lengths shortened to 2.12 Å and 2.18 Å, indicating a strengthened interaction between toluene and the Mn sites adjacent to Ru.

In summary, the introduction of Ru into CoMn₂O₄ led to the formation of new Ru–O–Mn and Ru–O–Co structures. These structural modifications resulted in a moderate enhancement in oxidizing ability, a marginal increase of surface-adsorbed toluene and lattice oxygen content, and a higher proportion of highly valent surface species (namely Co³⁺, Mn⁴⁺, and Ru⁴⁺). Consequently, despite a slight reduction in surface adsorbed oxygen, Ru doping significantly enhanced the toluene oxidation activity of CoMn₂O₄.

3. Materials and Methods

3.1. Catalyst Preparation

CoMn₂O₄ spinel was synthesized via a sol–gel method. Stoichiometric amounts of cobalt nitrate and manganous nitrate (provided by Sinopharm Group Chemical Reagent Co., Ltd., Shanghai, China, analytical reagent) were dissolved in ultrapure water and then added into a citric acid solution under stirring for 0.5 h. The pH of the mixture was adjusted to 9.0 using ammonia solution, followed by stirring at 80 °C for 10 h. The resulting product was dried at 105 °C for 10 h and subsequently calcined at 400 °C for 3 h to obtain CoMn₂O₄, which was designated as CoMn₂O₄ (fresh).

Ru-CoMn₂O₄ was prepared by a reduction method. Specifically, 0.5 g of CoMn₂O₄ (fresh) was dispersed in 25 mL of ultrapure water, after which 0.0039 g of ruthenium chloride (corresponding to 0.38 wt% Ru loading, provided by Sinopharm Group Chemical Reagent Co., Ltd., analytical reagent) was added to the suspension and stirred for 1 h. Then, NaBH₄ and NaOH (with molar ratios of NaBH₄/Ru = 2/1 and NaBH₄/NaOH = 1/5, provided by Sinopharm Group Chemical Reagent Co., Ltd., analytical reagent) were introduced into the suspension under stirring for another hour. Finally, the product was washed once with ultrapure water and once with ethanol, and dried at 105 °C for 12 h to obtain Ru-CoMn₂O₄. For comparison, Ru/CoMn₂O₄ was also synthesized via a conventional impregnation method using ruthenium chloride as the precursor and CoMn₂O₄ (fresh) as the support, with the same Ru loading of 0.38 wt%. Additionally, to rule out potential interference from chemical reagents, a control sample of CoMn₂O₄ (fresh) was treated following the same procedure used for Ru-CoMn₂O₄ synthesis but without the addition of ruthenium chloride. This sample was denoted as CoMn₂O₄ for subsequent experiments and analyses.

3.2. Performance Assessment

The performance for toluene oxidation was evaluated in a packed-bed reactor operated between 160–260 °C. A catalyst weight of 200 mg and a total gas flow rate of 200 mL min⁻¹ were used, resulting in a weight hourly space velocity (WHSV) of 60,000 cm³ g⁻¹ h⁻¹. The simulated flue gas consisted of 500 ppm toluene, 5% O₂, 5% H₂O (when introduced), balanced with N₂. The outlet concentrations of toluene and CO₂ were continuously measured via an infrared gas analyzer (Thermo Scientific IGS Analyzer, Waltham, MA, USA). Catalytic activity was evaluated based on two indicators: toluene conversion and CO₂ selectivity, which were calculated using the following equations:

$$\mathrm{toluene} \mathrm{conversion} ={\displaystyle \frac{{[\mathrm{toluene}]}_{\mathrm{in}}-{[\mathrm{toluene}]}_{\mathrm{out}}}{{[\mathrm{toluene}]}_{\mathrm{in}}}}\times 100\%$$

(10)

$${\mathrm{CO}}_2 \mathrm{selectivity}={\displaystyle \frac{{[{\mathrm{CO}}_2]}_{\mathrm{out}}}{7({[\mathrm{toluene}]}_{\mathrm{in}}-{[\mathrm{toluene}]}_{\mathrm{out}})}}\times 100\%$$

(11)

where [toluene]_in denotes the toluene concentration at the reactor inlet, and [toluene]_out and [CO₂]_out represent the concentrations of toluene and CO₂ at the reactor outlet, respectively.

3.3. Characterization

X-ray diffraction (XRD) pattern, X-ray photoelectron spectroscopy (XPS), temperature-programmed reduction of H₂ (H₂-TPR), and high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) and energy-dispersive X-ray (EDX) spectroscopy were measured using a Bruker-AXS D8 ADVANCE diffractometer, a Thermo Scientific K-Alpha spectrometer, a Micromeritics AutoChem II 2920 chemical adsorption analyzer, and a Bruker Quantax XFlash 7 X-ray energy spectrometer, respectively.

Temperature-programmed desorption of toluene (toluene-TPD) was also carried out in a packed-bed reactor. The procedure involved the following steps: First, the CoMn₂O₄ or Ru-CoMn₂O₄ samples were purged by a 5% O₂/N₂ mixture at 300 °C for 1 h and then cooled to 50 °C. Afterwards, the samples were exposed to 500 ppm toluene for 0.5 h. Lastly, the temperature was increased at a rate of 10 °C min⁻¹ in N₂.

In situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) was conducted on a Thermo Scientific Nicolet iS50 (Thermo Fisher Scientific, Waltham, MA, USA) Fourier transform infrared spectrometer equipped with a mercury cadmium telluride (MCT) detector.

3.4. Density Functional Theory (DFT) Calculation

First-principles calculations were performed using the Vienna Ab initio Simulation Package (VASP 5.4.4). The most stable CoMn₂O₄ (211) surface and a typical Ru₄ cluster were employed to construct the slab models of CoMn₂O₄ and Ru-CoMn₂O₄, respectively. Further computational details and specific structural information were provided in the Supporting Information.

4. Conclusions

This study demonstrated that Ru doping effectively enhanced the toluene oxidation performance of CoMn₂O₄. The optimized Ru-CoMn₂O₄ catalyst achieved 90% toluene conversion at 224 °C with high CO₂ selectivity under a high WHSV of 60,000 cm³ g⁻¹ h⁻¹. A combination of mechanistic analyses revealed that the reaction proceeded primarily via Mars–van Krevelen (MvK) and Eley–Rideal (E–R) mechanisms, in which the abundance of highly valent metal species (Co³⁺, Mn⁴⁺, and Ru⁴⁺), surface lattice oxygen activity, and presence of surface adsorbed oxygen and toluene molecules collectively dictated the catalytic behavior. The doping of Ru into the CoMn₂O₄ spinel structure generated new Ru–O–Mn and Ru–O–Co linkages. This structural modification resulted in a moderately enhanced oxidizing capability and simultaneously increased the concentrations of highly valent surface species, surface adsorbed toluene, and surface lattice oxygen. Despite a minor decrease in surface adsorbed oxygen, Ru doping significantly improved the overall catalytic activity of CoMn₂O₄ for toluene oxidation. These findings not only clarified the promotional role of Ru in spinel-structured catalysts but also validated CoMn₂O₄-based materials as efficient systems for the catalytic oxidation of VOCs.