Tailoring Pore Size in Bimetallic Nb-Mn/MCM-41 Catalysts for Enhanced Plasma-Driven Catalytic Oxidation of Toluene

Abstract

This study explored how pore size engineering in Nb-Mn/MCM-41 catalysts affects plasma-catalytic toluene oxidation. Adjusting the pore diameter (2.49–3.98 nm) modulated metal-support interactions and oxygen dynamics, with pore expansion to 3.73 nm (M3) optimizing the Mn⁴⁺/(Mn²⁺ + Mn³⁺) ratio (XPS: 36.8%), the amount of lattice oxygen species (O₂-TPD: 0.222 mmol/g), and crystallite size control (1.5 ± 0.2 nm, TEM). Smaller pores (M1: 2.49 nm) enhanced toluene adsorption but limited active site accessibility, while oversized pores (M4: 3.98 nm) reduced oxygen storage capacity (0.600→0.412 mmol/g). The Nb-Mn/M3 catalyst achieved superior performance with 96.8% toluene conversion, 55.0% CO₂ selectivity, and 85.4% carbon balance, while minimizing organic byproducts (GC-MS). Mechanistic studies revealed pore-mediated oxygen storage-transport cycles as critical for decoupling adsorption and oxidation steps. This study reveals fundamental mechanisms linking pore architecture to plasma-catalytic synergy in toluene oxidation, offering critical insights for the systematic design of energy-efficient, plasma-catalytic systems targeting industrial VOCs remediation.

1. Introduction

Volatile organic compounds (VOCs), particularly recalcitrant aromatic hydrocarbons such as toluene, present critical environmental and health challenges due to their carcinogenicity, widespread emissions from industrial processes, and role in ozone formation [1,2,3]. Conventional technologies like adsorption [4], thermal catalysis [5], and biodegradation [6] face limitations in treating low-concentration, dynamic toluene emissions under industrial conditions. Non-thermal plasma (NTP) technology has emerged as a promising solution for VOCs abatement owing to its rapid activation, adaptability to ambient conditions, and operational flexibility [7,8,9]. However, standalone NTP systems often exhibit incomplete toluene oxidation, low energy efficiency, and undesirable byproduct formation [10,11,12]. Integrating NTP with catalytic processes has proven advantageous, as catalysts enhance reactive species utilization, improve CO₂ selectivity, and reduce energy consumption [13,14,15].

The performance of plasma-catalytic systems is critically governed by the structural and physicochemical properties of catalysts, particularly their pore architecture. Pore size plays a dual role in regulating reactant adsorption, diffusion dynamics, and confinement effects that dictate plasma-catalyst interactions [16,17,18,19]. Smaller pores enhance adsorption capacity through elevated surface area and intensified van der Waals forces while simultaneously promoting the stabilization of reactive oxygen species (ROS) and localized electric field effects, thereby synergistically improving toluene degradation efficiency [20,21,22,23,24,25]. In contrast, larger pores favor plasma penetration and active species transport, mitigating pore-blocking effects and facilitating interfacial reactions [26,27,28,29]. Furthermore, pore geometry modulates metal-support interactions by altering the geometric and electronic states of confined active sites. Hierarchical mesoporous frameworks optimize metal oxide dispersion and redox properties, directly enhancing catalytic activity in VOCs oxidation [30,31,32,33]. Zhang et al. [34] demonstrated that such tailored pore structures enhanced MnO_x dispersion, surface oxygen mobility, and Mn⁴⁺ content, leading to superior catalytic activity and durability in toluene oxidation. Despite these insights, the influence of pore structure—particularly pore size—on plasma-catalytic performance remains inadequately explored. Current studies often focus on composite pore systems rather than isolating the role of pore dimensions, leaving critical gaps in understanding how specific pore sizes balance adsorption, plasma-catalyst synergy, and ROS utilization. This knowledge gap hinders the rational design of pore-optimized catalysts for energy-efficient toluene abatement under mixed-VOC conditions.

Our prior work identified mesoporous MCM-41 as an exceptional catalyst support in plasma-catalytic systems, outperforming ZSM-5 and 13X zeolites [35]. Unlike microporous ZSM-5 (pore size: 0.5–0.6 nm) and 13X (pore size: 0.7–0.8 nm), MCM-41’s ordered hexagonal mesopores (2–10 nm) enable superior mass transfer of toluene and intermediates, mitigate pore-blocking, and enhance accessibility to active sites. Additionally, MCM-41’s high specific surface area and tunable surface acidity optimize metal dispersion and redox cycle stability under plasma conditions [36]. The Nb-Mn/MCM-41 bimetallic catalyst demonstrated superior performance, leveraging high surface acidity, efficient ozone utilization, and synergistic plasma-catalytic interactions [37]. Building on these findings, this study systematically investigates Nb-Mn/MCM-41 catalysts with tailored pore sizes. Advanced characterization techniques, including N₂ physisorption, TEM, XPS, and O₂-TPD, were employed to correlate pore structure with metal oxide dispersion and electronic interactions. This work elucidates mechanistic relationships between pore architecture and plasma-driven toluene oxidation, advancing the design of energy-efficient VOC abatement technologies.

2. Results

2.1. Characterization of Catalysts

2.1.1. Structural Characterization

This study investigated the pore structure evolution in MCM-41 molecular sieves with engineered pore sizes and their niobium-manganese oxide-loaded derivatives through nitrogen physisorption analysis. Pore size distribution of supports and catalysts is shown in Figure 1. The MCM-41 series (M1–M4) exhibited a gradual increase in pore diameters from 2.49 to 3.98 nm, validating the achievement of targeted pore size regulation (Table S1). The introduction of metal oxides led to a consistent decrease in pore diameters, ranging from 2.48 to 3.68 nm, which was attributed to the adsorption of Nb-Mn oxide clusters. This decrease was accompanied by a reduction in both specific surface areas and pore volumes, from 1276 to 760 m²g⁻¹ and from 0.761 to 0.725 cm³g⁻¹, respectively.

However, Nb-Mn/M4 exhibited an anomalous increase in pore volume (1.202 cm³g⁻¹) compared to its support (0.725 cm³g⁻¹), which may arise from the unique interplay between the expanded pore architecture of M4 (3.98 nm) and the spatial distribution of metal oxides. The larger pore size of M4 likely facilitated the formation of well-dispersed Nb-Mn clusters within the mesochannels without significant pore blockage, while partial structural reorganization during calcination could contribute to localized pore expansion [38]. The non-linear distribution of surface areas indicated that factors such as structural defects or variations in wall thickness during pore expansion influenced the surface area [39]. The proportional decrease in surface area and pore volume suggested a uniform dispersion of metal oxides within the mesoporous channels [40]. The controlled deposition process preserved the hierarchical structure of the MCM-41 without significant agglomeration on the surface. The observed pore narrowing, with a difference of 0.3 nm between M4 and M1, indicated the size-selective accessibility of metal precursors within the pores.

The XRD patterns of Nb-Mn/MCM-41 catalysts with varied pore sizes are depicted in Figure 2. The prominent diffraction peak at 2θ = 33.12° was attributed to the crystalline form of Nb₂O₅, indicating that niobium was primarily present as Nb₂O₅ on the MCM-41 support. For manganese, several oxidation states were detected, including MnO at 2θ = 36.2°, Mn₂O₃ at 2θ = 38.2° and 55.4°, and MnO₂ at 2θ = 49.2° and 65.9°, as observed in the Nb-Mn/M1 sample [41,42]. Notably, the progressive broadening of MnO, Mn₂O₃, and MnO₂ reflections in Nb-Mn/M1 to Nb-Mn/M4 catalysts indicates a reduction in crystallite size with increasing MCM-41 pore diameter. This trend suggests that the larger pore dimensions of the molecular sieve facilitated better dispersion of the active metal species, thereby inhibiting crystallite growth. The size regulation was likely governed by the metal–support interaction that was dependent on the pore architecture, where the confined pore channels in smaller-pore MCM-41 (e.g., M1) limited metal precursor diffusion, promoting the formation of larger crystallites. In contrast, the wider pore systems (e.g., M4) enabled a more uniform distribution of metal oxides by mitigating mass transfer constraints, resulting in smaller crystallites. ICP-OES results (Table S1) confirmed a nominal Nb/Mn molar ratio of 1:1 across all catalysts (deviation < 3%), ruling out compositional variation as a cause of reflection broadening. The intensity-area invariance further supports that phase fractions remain consistent, with observed broadening attributed solely to crystallite size effects.

2.1.2. Surface and Redox Properties

(1)

TEM

The morphological characteristics of the Nb-Mn/M1, Nb-Mn/M2, Nb-Mn/M3, and Nb-Mn/M4 catalysts were characterized using TEM. Surface metal oxide nanoparticle size distributions were statistically analyzed using Image J software (version 1.8.0), as depicted in Figure 3. TEM images indicated uniform dispersion of metal oxide particles across all catalysts, with no significant agglomeration, reflecting high dispersion efficiency. Notably, the Nb-Mn/M3 catalyst displayed the best dispersion with an average particle size of approximately 1.5 ± 0.2 nm. In contrast, the Nb-Mn/M2 and Nb-Mn/M4 catalysts exhibited slightly larger average crystallite sizes of ~2.0 ± 0.3 nm, whereas the Nb-Mn/M1 catalyst had the largest average particle dimension at ~2.5 ± 0.4 nm. This trend of increasing crystallite size with decreasing pore diameter in the MCM-41 supports was consistent with the XRD observations, implying that limited pore geometries impede precursor mobility during synthesis, fostering crystallite growth [43]. The inverse relationship between the support’s pore size and the metal oxide particle dimensions further supports the pivotal role of pore architecture in governing nanoscale metal distribution.

(2)

XPS

The surface elemental composition and chemical states of the catalysts were meticulously analyzed using XPS, with illustrative spectra for the Mn 2p, Nb 3d, and O 1s regions depicted in Figure 4. Deconvolution of the Mn 2p spectra revealed three distinct oxidation states, with binding energies at 640.4 eV, 641.2 eV, and 642.8 eV corresponding to Mn²⁺, Mn³⁺, and Mn⁴⁺, respectively [44]. All four catalysts exhibited a coexistence of these manganese valence states (

Table 1. XPS results of different Nb-Mn/MCM-41.

Catalysts	Area (%)			Mn4+/	Area (%)		Olat/	Area (%)		Nb5+/
	Mn2+	Mn3+	Mn4+	(Mn2+ + Mn3+)	Oad	Olat	Oad	Nb5+	Nb4+	Nb4+
Nb-Mn/M1	56.6	27.5	15.9	18.9%	83.7	16.3	19.4%	73.7	26.3	2.80
Nb-Mn/M2	47.6	37.2	15.2	17.9%	78.4	21.6	27.5%	73.3	26.7	2.74
Nb-Mn/M3	20.5	52.6	26.9	36.8%	64.3	35.7	55.5%	84.0	16.0	5.25
Nb-Mn/M4	36.9	41.5	21.5	27.4%	76.5	23.5	30.7%	84.1	15.9	5.28

), likely influenced by the shared isoelectric point of the MCM-41 support. Notably, the Mn⁴⁺/(Mn²⁺ + Mn³⁺) ratio showed a systematic increase with increasing pore diameter, reaching a maximum of 36.8% in Nb-Mn/M3, indicating enhanced redox properties in the larger-pore catalysts. The Nb 3d spectra presented dual peaks at 205.2 eV (Nb⁴⁺) and 206.0 eV (Nb⁵⁺), confirming the coexistence of both oxidation states across all catalysts [37,45]. As shown in Table 1, the Nb⁵⁺/Nb⁴⁺ ratio increased from 2.80 in Nb-Mn/M1 to 5.28 in Nb-Mn/M4, with Nb⁵⁺ content rising from 73.7% to 84.1%. This mixed-valent niobium configuration exhibited dual functionality: Nb⁵⁺ promoted electron-deficient sites via strong Nb-O bonds. While Nb⁴⁺ promoted charge transfer between Mn centers and adsorbed intermediates [46]. Notably, the highest Nb⁵⁺/Nb⁴⁺ ratio (5.25) in Nb-Mn/M3 coincided with its optimal toluene degradation efficiency (96.8%), suggesting synergistic redox coupling between Nb and Mn. O 1s spectrum analysis identified two oxygen species: adsorbed oxygen (O_ad, 530.9 eV) and lattice oxygen (O_lat, 529.3 eV). As detailed in Table S2, the O_lat/O_ad ratio was found to be pore-dependent, with the highest values observed in Nb-Mn/M3 (55.5%). This trend was correlated with the observed Mn⁴⁺ enrichment, as higher-valent manganese species are more likely to stabilize lattice oxygen through stronger metal–oxygen bonds [47,48]. Collectively, these findings indicate that while the chemical nature of the supported manganese and niobium oxides is consistent across the pore-modified catalysts, their relative abundance is significantly regulated by the pore architecture [49]. The synergistic interplay between Nb⁵⁺-stabilized lattice framework and Mn⁴⁺-mediated ROS generation underscores the critical role of dual-metal valency in catalytic oxidation. The increased presence of Mn⁴⁺ and O_lat in the larger-pore systems (e.g., Nb-Mn/M3) suggests enhanced surface oxygen activation and redox cycling capacity, which are crucial factors influencing catalytic performance in oxidation reactions.

(3)

O₂-TPD

The catalytic properties of the Nb-Mn/M1, Nb-Mn/M2, Nb-Mn/M3, and Nb-Mn/M4 catalysts were characterized through O₂-TPD analysis. As depicted in Figure 5, all catalysts exhibited four distinct oxygen desorption peaks. Peak I (71–115 °C) was assigned to molecular oxygen (O₂) adsorbed physically on the catalyst surface, while peaks II (285–306 °C) and III (433–492 °C) were attributed to chemisorbed atomic oxygen species. Peak IV (629–685 °C) corresponded to the oxygen in the lattice of the metal oxides [50]. Quantitative analysis of the data presented in Table S2 revealed that Nb-Mn/M3 exhibited the most superior oxygen storage properties, with a total oxygen desorption amount of 0.600 mmol/g, which was approximately 46% higher than that of Nb-Mn/M1. Notably, the lattice oxygen content in Nb-Mn/M3 (0.222 mmol/g) was significantly higher than that of other catalysts. The optimized pore size in Nb-Mn/M3 (3.73 nm) improves the dispersion and accessibility of active sites (Table S1), despite its lower total surface area (981 m²/g) compared to Nb-Mn/M1 (1261 m²/g). This enhanced active site exposure facilitates oxygen adsorption and promotes the uniform distribution of oxygen species, thereby amplifying redox activity and catalytic performance.

2.2. Toluene Degradation

2.2.1. Effect of Plasma Discharge

In a low-temperature plasma-catalysis system, the zeolite or catalyst was packed within a DBD reactor. The formation of micro-discharge channels primarily depended on the equivalent capacitance of the dielectric layer and the gas breakdown threshold, thereby influencing reactive species such as radicals, excited-state molecules, atoms, and ions within the discharge region [51]. Therefore, in investigating the effect of zeolite catalysts with varying pore structures on toluene degradation, it was necessary to examine the discharge characteristics following the packing of different pore-sized zeolites and catalysts. Voltage–current waveforms (Figure S1) revealed that the current signal values were similar across different materials, indicating a weak correlation between the pore size of the zeolites (M1–M4) and their Nb-Mn-supported catalysts on the discharge characteristics of the DBD system.

2.2.2. Effect of Adsorption

This study investigated the impact of zeolite pore size on toluene adsorption and catalytic oxidation, revealing the critical role of porous material architecture in VOCs abatement. Adsorption breakthrough curves (Figure 6a) showed that toluene adsorption capacity on both supports and catalysts decreased in the order Nb-Mn/M1 > Nb-Mn/M2 > Nb-Mn/M3 > Nb-Mn/M4. This trend was inversely correlated with the zeolite pore size distribution (M4 > M3 > M2 > M1) presented in Table S1. This inverse relationship indicated a non-linear influence of microstructural properties on the adsorption process within the microporous regime. Thermodynamic analysis suggested that reduced pore size enhanced adsorbate-adsorbent interactions via a pore wall superposition effect. When the zeolite pore size approached the kinetic diameter of toluene, a micropore filling mechanism became dominant [52]. This resulted in overlapping van der Waals potentials from adjacent pore walls, thus increasing the depth of the adsorption potential well for toluene. Molecular dynamics simulations have shown that this superposition effect is maximized when the pore diameter is 1.5–2 times that of the adsorbate molecule [53]. This explained the prolonged breakthrough time observed for the M1 support, which possessed an optimized pore architecture that maximized the adsorption free energy.

To further explore the adsorption-catalysis synergy, we compared the performance of different zeolites and Mn-supported catalysts in a DBD reactor. As shown in Figure 6b, the toluene removal rate gradually saturated, reaching a steady-state value of 62.3%. This suggested a limited oxidation depth under plasma-only conditions. The incorporation of ZSM-5, 13X, and MCM-41 shortened the time to reach steady state but only marginally increased the final removal rate to 65.1–67.8%. This demonstrated that the zeolite adsorption capacity primarily influenced the reaction kinetics rather than the ultimate degradation efficiency.

2.2.3. Effect of Pore Structure

To systematically elucidate the influence mechanism of zeolite pore structure on the performance of the plasma-assisted catalytic system, we prepared a series of Nb-Mn catalysts supported on MCM-41 zeolites with controlled pore sizes. This approach established a multi-dimensional structure–activity relationship linking pore structural parameters, catalyst properties, and reaction performance. As shown in Figure 7, Nb-Mn catalysts supported on MCM-41 with varying pore sizes were packed into a DBD reactor. As the carrier pore size increased, the toluene removal rate initially increased and then decreased, reaching a maximum value of 96.8% with Nb-Mn/M3, a CO₂ selectivity of 55.0%, and a carbon balance of 85.4%. The results indicated a non-linear regulatory effect of pore size evolution on the catalytic system. When the pore size increased from 2.49 nm (M1) to 3.73 nm (M3), the toluene removal rate increased to 96.8%, and the CO₂ selectivity and carbon balance significantly increased by 26% and 39.1%, respectively. However, further increasing the pore size to 3.98 nm (M4) resulted in a performance decline.

The gas phase composition at the outlet of the NTP-catalyzed reactor containing Nb-Mn/MCM-41 was characterized using GC-MS analysis. Figure S2 displayed the chromatographic profiles of organic byproducts generated by different Nb-Mn/M catalysts (M1–M4), revealing distinct intermediate product distributions. Specifically, the Nb-Mn/M3 system produced propyl ester, glyceraldehyde, benzyl alcohol, and heptene dialdehyde as primary intermediates. Comparative analysis demonstrated catalyst-specific variations: Nb-Mn/M1 generated additional hydroxycyclohexane and benzaldehyde, while Nb-Mn/M2 exhibited additional hydroxycyclohexane and benzaldehyde formation but lacked glyceraldehyde detection. Notably, the Nb-Mn/M4 system produced an additional adipic acid byproduct. These findings corroborate the superior selectivity and enhanced carbon balance achieved by the Nb-Mn/M3 catalyst relative to its counterparts.

The long-term stability of the catalysts was evaluated through a 60-h continuous plasma-catalytic oxidation test (Figure S3). Both Nb-Mn/M3 and Nb-Mn/M1 maintained high toluene degradation efficiencies with minimal deactivation, highlighting their potential for industrial-scale applications.

3. Discussion

XRD and TEM characterization revealed that pore size expansion significantly influenced the diffusion kinetics of precursors within the channels. As the pore size increased from a suboptimal range (M1: 2.49 nm) to an optimized range (M3: 3.73 nm), the capillary forces of the impregnation solution within the mesoporous structure weakened substantially. This facilitated uniform distribution of metal precursors along the entire channel axis, effectively mitigating particle aggregation during the subsequent drying process. As a result, the MnO_x crystallite size decreased sharply from 2.5 ± 0.4 nm (Nb-Mn/M1) to 1.5 ± 0.2 nm (Nb-Mn/M3), while the BET surface area increased remarkably, thereby exposing significantly more active catalytic sites. However, when the pore size exceeded this ideal range, the excessive space dramatically reduced the zeolite’s ability to anchor metal species, leading to renewed sintering tendencies and partial loss of dispersion control. XRD/TEM directly resolve crystallite/pore structures; the inferred diffusion effects are based on structural trends: larger pores (M4) enable uniform metal dispersion (TEM, Figure 3), shortening diffusion paths, whereas confined pores (M1) restrict precursor mobility, aligning with crystallite size differences (XRD, Table S3).

XPS analysis revealed that pore size evolution modulated the valence state distribution of Mn by altering the strength of the metal–support interaction. When the pore size increased to 3.73 nm, the surface Mn⁴⁺/(Mn²⁺ + Mn³⁺) ratio reached 36.8% (Table 1). The high Mn⁴⁺ content not only stabilized the lattice oxygen framework, as evidenced by O₂-TPD analysis, but also significantly improved the generation efficiency of ROS by promoting plasma-activated pathways. Specifically, Mn⁴⁺ sites facilitated the stabilization of ${\mathsf{O}}_2^{-}$ intermediates and their subsequent dissociation into atomic oxygen radicals (O⁻), which are critical for C-H bond cleavage in toluene [54].

Concurrently, the dual oxidation states of Nb (Nb⁴⁺ and Nb⁵⁺) synergistically modulated catalytic behavior. The Nb⁵⁺/Nb⁴⁺ ratio increased from 2.8 (Nb-Mn/M1) to 5.28 (Nb-Mn/M4), with Nb⁵⁺% rising from 73.7% to 84.1% (Table 1). Higher Nb⁵⁺ content (e.g., 84.0% in Nb-Mn/M3) promoted electron-deficient sites via strong Nb-O bonds, while Nb⁴⁺ facilitated electron transfer between Mn⁴⁺ and adsorbed intermediates. This dual functionality of Nb stabilized Mn⁴⁺ via Nb-O-Mn linkages and amplified ROS generation, as evidenced by the optimized O_lat/O_ad ratio (55.5%) and toluene degradation efficiency (96.8%) in Nb-Mn/M3 [46]. Notably, the excessively large pore size (M4), while further improving mass transfer efficiency, resulted in a reduced surface area (1015→760 m²/g). This led to a decreased loading density of active components and an excessively fast desorption rate of intermediate products, reducing opportunities for secondary oxidation. Despite the highest Nb⁵⁺% in Nb-Mn/M4 (84.1%), the diminished Mn⁴⁺/(Mn²⁺ + Mn³⁺) ratio (27.4%) and lower active site density disrupted the Nb-Mn redox synergy, ultimately explaining the reduced carbon balance (Figure 7b). The XRD identifies bulk MnO, Mn₂O₃, and MnO₂ phases (Figure 2), and XPS reveals surface Mn³⁺/Mn⁴⁺ enrichment (Table 1) due to calcination-induced oxidation. The systematic increase in Mn⁴⁺/(Mn²⁺ + Mn³⁺) ratios with pore size (up to 36.8% in Nb-Mn/M3) aligns with enhanced catalytic redox activity. This apparent disparity arises from XPS’s surface sensitivity (~5 nm depth) versus XRD’s bulk averaging, highlighting complementary insights: surface Mn valency tuning (XPS) governs reactivity, while bulk phase uniformity (XRD) ensures structural consistency. The synergy between surface Mn⁴⁺ enrichment and pore-dependent redox enhancement underpins the catalytic mechanism.

O₂-TPD results showed that Nb-Mn/M3 exhibited the most superior oxygen storage capacity. Its total oxygen desorption amount reached 0.600 mmol/g (Table S2), approximately 46% higher than Nb-Mn/M1, with a lattice oxygen content of 0.222 mmol/g. The low-temperature plasma’s high-energy electrons selectively activated oxygen adsorbed on the catalyst surface (peaks I–II, <500 °C), yielding active species like superoxide radicals. These species promptly initiated the cleavage of C-H bonds within toluene molecules, resulting in a significant removal rate during the early stages of the reaction [55]. The higher total oxygen storage capacity of Nb-Mn/M3 (0.600 mmol/g) provided an ample supply of reactive oxygen for this stage. As the reaction proceeded, the surface-adsorbed oxygen was gradually consumed. At this point, the high lattice oxygen content of Nb-Mn/M3 (0.222 mmol/g) was activated through the plasma-induced local hotspot effect and migrated to the surface in the form of ${\mathsf{O}}_2^{-}$ to participate in the benzene ring opening and further oxidation of intermediate products. This mechanism effectively inhibited carbon deposition, resulting in a high carbon balance of 85.4%. In contrast, low lattice oxygen catalysts (such as M1) suffered from insufficient oxygen replenishment, leading to the accumulation of intermediate products within the channels, causing catalyst deactivation and a decrease in CO₂ selectivity.

Based on experimental observations and GC-MS analysis (Figure S2), the oxidation mechanism of toluene molecules over bimetallic catalysts was systematically elucidated. As illustrated in Figure 8, three fundamental models govern the plasma-catalytic degradation of toluene: the Mars-van Krevelen (MvK) mechanism, Langmuir–Hinshelwood (LH) mechanism, and Eley–Rideal (ER) mechanism [37,49]. In the MvK mechanism, adsorbed toluene and intermediate products (including benzaldehyde, benzyl alcohol, and hydroxycyclohexane) interact with ROS on the catalyst surface. These species undergo sequential oxidation through intermediates such as adipic acid, heptenedial, carboxypentadiene, and propyl esters, ultimately mineralizing into CO₂ and H₂O. The Nb-Mn/MCM-41 catalyst facilitates this process through its dual redox couples: the Nb⁴⁺/Nb⁵⁺ pair functions as a buffering reservoir for oxygen storage and transport, while the Mn³⁺/Mn⁴⁺ pair mediates oxygen transfer to the catalyst surface. While ozone formation in plasma-catalytic systems may compete with reactive oxygen species for active sites, the Nb-Mn/MCM-41 catalyst effectively mitigates this effect through Mn⁴⁺-mediated ozone decomposition into reactive O⁻ species and Nb⁴⁺/Nb⁵⁺ redox stabilization of oxygen species, minimizing gas-phase ozone accumulation while maintaining efficient O_lat replenishment. Depleted O_lat during toluene oxidation is replenished through ozone decomposition or molecular oxygen capture from the gas phase. The LH mechanism involves surface-adsorbed toluene and intermediates reacting directly with chemisorbed oxygen species, leading to complete oxidation products. The ER mechanism proposes gas-phase interactions between oxygen species and toluene molecules, generating similar intermediates that subsequently undergo mineralization.

GC-MS analysis of the plasma reactor effluent (Figure S2) demonstrated pore-mediated control over toluene degradation pathways. The Nb-Mn/M3 catalyst, exhibiting an optimized Mn⁴⁺/(Mn²⁺ + Mn³⁺) ratio (36.8%) and lattice oxygen abundance (O_lat/O_ad = 55.5%) within the tested catalyst series (Nb-Mn/M1 to Nb-Mn/M4, Table 1), was synthesized under consistent conditions (incipient wetness impregnation, 500 °C calcination) with a fixed Nb/Mn molar ratio of 1:1. It selectively generated oxidizable intermediates: propyl ester and glyceraldehyde. This aligns with its superior oxygen storage capacity (0.600 mmol/g total O₂ consumption) and the amount of lattice oxygen species (0.222 mmol/g) (Table S2), enabling rapid dehydrogenation of benzyl alcohol to heptenedial through MvK cycles. Therefore, the Nb-Mn/M3 catalyst exhibited superior toluene degradation efficacy in the plasma-catalytic system compared to other catalysts.

4. Experimental Instruments and Methods

4.1. Chemical Reagent

Surfactant templates, including decyltrimethylammonium bromide (C₁₀TAB, AR grade), dodecyltrimethylammonium bromide (C₁₂TAB, AR), and tetradecyltrimethylammonium bromide (C₁₄TAB, AR), were obtained from Shanghai Maclin Biochemical Technology Co., Ltd. (Shanghai, China). Cetyltrimethylammonium bromide (C₁₆TAB, AR), sodium hydroxide (NaOH, AR), sodium aluminate (NaAlO₂, AR), tetraethyl orthosilicate (TEOS, AR), and toluene (C₇H₈, AR) were procured from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Manganese acetate tetrahydrate (Mn(CH₃COO)₂·4H₂O, AR) and niobium oxalate hydrate (C₁₀H₅NbO₂₀·xH₂O, AR) were supplied by Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China) and Guangzhou Kexing Chemical Co., Ltd. (Guangzhou, China), respectively. Commercial molecular sieves (MCM-41; ZSM-5; 13X) were provided by Tianjin Nanhua Catalyst Co., Ltd., Tianjin, China. High-purity N₂ (≥99.999%) and O₂ (≥99.999%) gases were purchased from Nanjing TeZhong Gas Co., Ltd., Nanjing, China.

4.2. Catalyst Preparation

Mesoporous MCM-41 substrates with controlled pore dimensions were fabricated through alkaline-mediated hydrothermal crystallization. The preparation protocol involved sequential combination of alkyltrimethylammonium bromide (C_nTAB, n = 10, 12, 14, 16), sodium hydroxide solution, and deionized water in a fixed molar ratio (1 TEOS: 0.49 NaOH: 0.17 C_nTAB: 70 H₂O), followed by controlled addition of TEOS via peristaltic pump. The homogeneous mixture underwent continuous agitation (2 h) before hydrothermal aging in Teflon-lined autoclaves at 110 °C (24 h). Post-synthesis processing included vacuum filtration (0.22 μm membrane), repeated ethanol/H₂O washing cycles, and desiccation at 80 °C (12 h). Template removal was achieved through programmed calcination (2 °C/min ramp to 550 °C, 5 h dwell time), yielding four distinct MCM-41 substrates (M1–M4) corresponding to C_nTAB chain lengths.

The Nb-Mn/MCM-41 series was synthesized through incipient wetness impregnation, where calcined MCM-41 substrates (M1–M4) were saturated with aqueous solutions containing stoichiometric amounts of manganese acetate tetrahydrate and niobium oxalate hydrate. The impregnated materials underwent sequential drying protocols: ambient-air solvent evaporation (24 h) followed by forced convection drying at 90 °C (6 h). Final activation occurred via oxidative calcination (500 °C, 4 h, static air) in a muffle furnace, producing four catalyst variants designated as Nb-Mn/M1 through Nb-Mn/M4. The labels Nb-Mn/M1 to Nb-Mn/M4 denote Nb-Mn/MCM-41 catalysts with incrementally enlarged pore diameters (specific values in Table S1), synthesized using MCM-41 supports of distinct pore architectures. All samples maintain a nominal Nb/Mn molar ratio of 1:1, as verified by ICP-OES (Table S1). All the catalysts were maintained under desiccator storage prior to characterization.

4.3. Catalyst Characterization

Textural properties, including specific surface area, pore volume, and architecture, were analyzed by N₂ physisorption using a Micromeritics ASAP 2020 analyzer (Micromeritics, Norcross, GA, USA). Prior to measurements at 77.3 K, approximately 300 mg of each sample underwent degassing treatment at 300 °C for 6 h under vacuum conditions. The specific surface area and pore size distribution were calculated using the Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) pore distribution methods, respectively.

Crystalline phase identification was performed by X-ray diffraction (XRD) using an ARL X’TRA diffractometer (Thermo Scientific, Waltham, MA, USA) with Cu Kα1 radiation (λ = 0.15406 nm) operated at 30 kV and 30 mA. Diffraction patterns were recorded in the 2θ range of 5° to 90° with a scanning rate of 4° min⁻¹.

X-ray photoelectron spectroscopy (XPS) measurements were conducted on a PHI 5000 VersaProbe III spectrometer (ULVAC-PHI, Chigasaki, Japan) equipped with a monochromatic Al Kα X-ray source (300 W, 14 kV). All spectra were charge-referenced to the adventitious carbon C 1s peak at 284.6 eV and subsequently deconvoluted using the XPS Peak 4.1 software package with Shirley-type background subtraction.

Transmission electron microscopy (TEM) observations were carried out using a JEOL JEM-200CX instrument (JEOL, Tokyo, Japan) operated at 200 kV. Specimens were prepared by dispersing the powder samples in ethanol through ultrasonic treatment, followed by deposition onto carbon-coated copper grids and subsequent drying under infrared lamp illumination.

Oxygen temperature-programmed desorption (O₂-TPD) experiments were performed on a Micromeritics AutoChem II 2920 system (Micromeritics, Norcross, GA, USA) equipped with a thermal conductivity detector (TCD). The protocol included 100 mg of catalyst pretreated in ultrapure He (99.999%, 30 mL min⁻¹) at 150 °C for 2 h. Exposed to 2 vol% O₂/He mixture (50 mL min⁻¹) at 30 °C for 2 h after cooling to room temperature. He flowed (50 mL min⁻¹) for 0.5 h to stabilize the baseline. The temperature ramped from 30 °C to 800 °C at 10 °C min⁻¹ under He flow, with desorbed oxygen monitored by TCD.

4.4. Catalytic Performance Evaluation

The plasma-catalytic system (Figure 9) aligned with established methodologies [37] in reactor geometry and analytical infrastructure. A coaxial dielectric barrier discharge (DBD) configuration was implemented, featuring a quartz reactor with 8 mm internal and 10 mm external dimensions. The high-voltage electrode consisted of a concentric stainless steel rod (3 mm diameter), while a grounded stainless steel mesh enveloped the reactor exterior, powered by an AC plasma generator (Nanjing Sumant Plasma Technology Co., Nanjing, China). Temperature stabilization (30 °C) was ensured through a recirculating glycol jacket. Adsorption evaluations utilized catalyst-loaded beds exposed to synthetic air streams containing 300 ppm toluene, regulated by calibrated mass flow controllers. Real-time compositional tracking was achieved via an inline sampling loop interfaced with gas chromatography.

Plasma-catalytic evaluations employed 0.2 g of catalyst positioned within the discharge gap. Reactant streams (1 L/min, N₂:O₂ = 4:1) containing toluene vapor passed through the energized reactor under controlled potentials (0–30 kV). Electrical diagnostics incorporated voltage/current waveform acquisition using a high-frequency digital oscilloscope (Tektronix TBS1102B), implementing the voltage-charge (V-Q) Lissajous methodology for precise energy quantification. Post-reaction gas analysis combined methanizer-assisted GC-FID (toluene quantification, CO/CO₂ detection) and GC-MS for intermediate speciation. Performance metrics were derived through the following computational framework:

$$\mathrm{SIE} (\mathrm{J}/\mathrm{L})=60\times \mathrm{Discharge} \mathrm{power} (\mathrm{w})/\mathrm{Gas} \mathrm{flow} \mathrm{rate} \left(\mathrm{L}/\min \right)$$

(1)

$$\mathrm{Toluene} \mathrm{removal} \mathrm{efficiency} (\%)=({\mathrm{C}}_{\mathrm{inlet}}-{\mathrm{C}}_{\mathrm{outlet}}\left)\right(\mathrm{ppm})/{\mathrm{C}}_{\mathrm{inlet}}\left(\mathrm{ppm}\right)$$

(2)

$$\mathrm{CO} \mathrm{selectivity} (\%)={\mathrm{C}}_{\mathrm{co}}\left(\mathrm{ppm}\right)/\left({\mathrm{C}}_{\mathrm{co}}+{\mathrm{C}}_{\mathrm{c}{\mathrm{o}}_2}\right)\left(\mathrm{ppm}\right)$$

(3)

$${\mathrm{CO}}_2 \mathrm{selectivity} (\%)={\mathrm{C}}_{\mathrm{c}{\mathrm{o}}_2}\left(\mathrm{ppm}\right)/\left({\mathrm{C}}_{\mathrm{co}}+{\mathrm{C}}_{\mathrm{c}{\mathrm{o}}_2}\right)\left(\mathrm{ppm}\right)$$

(4)

$$\mathrm{Carbon} \mathrm{balance} (\%)=({\mathrm{C}}_{\mathrm{co}}+{ \mathrm{C}}_{\mathrm{c}{\mathrm{o}}_2}) (\mathrm{ppm})/7\left({\mathrm{C}}_{\mathrm{inlet}}-{\mathrm{C}}_{\mathrm{outlet}}\right)\left(\mathrm{ppm}\right)$$

(5)

where C_inlet and C_outlet are the concentrations of toluene, C_co2 and C_co are the concentrations of CO₂ and CO.

5. Conclusions

This study establishes pore architecture engineering as a critical design principle for optimizing plasma-catalytic systems in VOC abatement. The systematic modulation of MCM-41 pore dimensions (2.49–3.98 nm) revealed the following insights. A pore diameter of 3.73 nm maximized metal oxide dispersion (1.5 ± 0.2 nm crystallites) while enhancing the redox-active Mn⁴⁺/(Mn²⁺ + Mn³⁺) ratio (36.8%) and the amount of lattice oxygen species (0.222 mmol/g), synergistically boosting ROS generation. GC-MS analysis identifies evolution—smaller pores (M1: 2.49 nm) accumulate hydroxycyclohexane and benzaldehyde, while oversized pores (M4: 3.98 nm) promote adipic acid formation. The 3.73 nm pore (M3) optimizes intermediate selectivity toward oxidizable species (glyceraldehyde, heptenedial), enabling 85.4% carbon balance. The Nb-Mn/M3 catalyst minimizes refractory byproduct formation through pore-mediated ROS gradients, achieving 96.8% toluene conversion with 55.0% CO₂ selectivity. By correlating pore-structural features with plasma-enhanced oxidation pathways, our findings establish actionable guidelines for fine-tuning catalyst architectures to optimize energy utilization and degradation efficiency in next-generation VOC abatement technologies.