Structural Effect of Cu-Mn/Al₂O₃ Catalysts on Enhancing Toluene Combustion Performance: Molecular Structure of Polyols and Hydrothermal Treatment

Abstract

This study presents a series of Cu-Mn/Al₂O₃ catalysts prepared by the polyol method to improve the toluene combustion process. The catalytic activity evaluation results showed that the different polyols have a great influence on catalyst activity, in which the catalyst prepared with glycerol through a hydrothermal reaction at 90 °C displayed the highest catalytic activity. The lowest T90 and T50 values could be achieved by CMA-GL-90 with 260 and 237 °C, respectively. Moreover, the XRD and BET results showed that the hydrothermal treatment was more favorable with Cu-Mn crystal formation, and an abundance of mesopores remained in all catalysts with a high specific surface area from 94.37 to 123.03 m²·g⁻¹. The morphology analysis results by SEM and TEM indicated that employing glycerol coupled with hydrothermal treatment at 90 °C could enhance the formation of CuMn₂O₄ spinel. The toluene catalytic combustion mechanism of Cu-Mn/Al₂O₃ catalysts was discussed based on XPS and H₂-TPR, and a high atomic ratio of Mn³⁺ could be obtained with 51.03%, and the ratio of O_ads/O_latt also increased to 2.85 in CMA-GL-90. The increase in Mn³⁺ species and oxygen vacancies on the surface of catalysts exhibited excellent activity and stability for toluene combustion. These findings offer valuable insights for optimizing the design and application of Cu-Mn/Al₂O₃ catalysts in addressing the catalytic oxidation reactions of organic volatile compounds.

1. Introduction

Volatile organic compounds (VOCs) as important precursors to creating smog have been referred to as an important task for atmospheric governance in the world [1,2,3]. China, as the largest CO₂ emitter, has issued relevant VOC control policies, and mitigating VOCs was included in China’s “14th Five-Year Plan” as a ”short-board” for atmospheric pollution control [4]. Therefore, more effective control measures for VOCs are needed. Commonly, the control strategies of VOCs can be summarized as two primary pathways: recovery and elimination. The recovery pathway includes absorption, condensation, membrane separation methods, and so on [5,6,7]. However, the VOCs’ compositions are complex, and emission sources are scattered and varied, limiting the application scenarios of the recovery pathway. On the other hand, the elimination pathway includes catalytic combustion, thermal oxidation, plasma treatment, and so on [8,9]. Among them, catalytic combustion is regarded as the most economic and effective method to remove VOCs; it can convert VOCs into low-harm compounds at a low temperature and time cost.

Catalysts are the key point of catalytic combustion, which can reduce the combustion temperature below 500 °C by the catalytic oxidation reaction [10]. Hence, the high activity and thermal stability of catalysts are both required during the catalytic combustion of VOCs. Noble metals (Pt, Pd, Rh, Au) are the most effective catalysts and are widely used in industrial processes [11]. With the wider industrial implementation of catalytic combustion, the quantity demand for noble metals increases. However, their scarce reserves are bound to affect the overall cost of the catalytic combustion system. Therefore, the development of a low-cost and highly effective alternative is the key technical bottleneck. Recently, earth-abundant metal (EAM) catalysts have been broadly studied, such as Cu [12], Cr [13], Fe [14], Mn [15], and so on. Compared with the noble catalysts, the catalytic activity of EAM catalysts has a gap, but this difference can be compensated by several supports, such as molecular sieves [16], alumina [17], mesoporous ceramics [18], and so on.

Recently, copper–manganese (Cu-Mn) oxides have received considerable attention due to their exceptional low-temperature activity. Cocuzza et al. examined a series of Cu-Mn oxides to oxidize VOCs, and the complete converted temperature of propylene and toluene was 190 and 250 °C with a low proportion of CuO, respectively [12]. Actually, the complete oxidation temperature of toluene in the noble metal catalyst system was also nearly 250 °C. Obviously, the Cu-Mn catalyst has potential as a substitute with comparable catalytic performance to noble metals like Pt. According to the literature, the electronic and geometric effects between Cu and Mn can affect the activity of the catalytic oxidation reaction. Among them, the phase transition from amorphous to crystalline will affect the catalytic activity strongly. The amorphous phase is more effective in releasing oxygen from Cu-Mn oxides, which determines the rate of redox reactions for VOCs [19]. However, in the crystalline phase, the surface of the catalyst is rich in Cu and Mn ions, which can reduce the activity of the catalyst. On the other hand, the surface properties are also important in tuning the combustion efficiency. Therefore, the structural engineering of Cu-Mn oxide catalysts is very important for the catalytic reaction.

A brief look at the recent literature shows that various chemical synthesis methods have been used to prepare Cu-Mn oxide catalysts, such as sol–gel [20], co-precipitation [21], the hydrothermal method [22], impregnation [23], and so on. However, these synthesis methods face some weak points, for example, the final particles tend to agglomerate or sinter under the high temperature, causing a reduction in the catalytic performance. Chuang et al. compared the preparation methods of activated-carbon-supported copper catalysts to remove NO and found that the activity of catalysts prepared by the polyol process was higher than that of those prepared by the impregnation method because a higher dispersion of active sites was obtained by the polyol process [24]. The polyol process, an instant chemical reduction technique, is a simple, fast, and cost-effective method to control the size and morphology of metals oxides. The polyol can limit the growth, tune the morphology, and prevent the agglomeration of particles through a space-filling, structure-directing, and template role. For example, Liu et al. successfully obtained manganese oxide nanoplates with different shapes by using ethylene glycol as a solution [25]. Diethylene glycol and glycerol are also widely used as the solvent [26]. Li et al. tried to use alcohols (glycerol and ethylene glycol)/water as a solvent to prepare porous Mn-Cu oxides to catalyze CO, and the complete conversion of CO could be achieved at 45 °C [27]. Moreover, Lu et al. employed the polyol method to immobilize Cu, Co, Fe, and Ni on activated carbon (AC), testing these catalysts in the oxidation of benzene, toluene, and xylene [28]. Their study revealed well-dispersed metal nanoparticles on the AC support, demonstrating the catalysts’ remarkable catalytic activity. In fact, comprehensive studies about the preparation of Cu-Mn oxide catalysts by the polyol process are scarce, and evaluations of their catalytic activity for VOC removal are rarer.

In spite of the literature mentioned above, we attempt to reveal the effects of various polyols on the crystallization of Cu-Mn alloys and their catalytic efficiency in the catalytical combustion of VOCs. In this study, polyols were chosen with different numbers of active hydroxyl groups and physicochemical properties to form complexes with metal ions, including glycerol, ethylene glycol, 1,3-propanediol, and 1,4-butanediol. The Cu-Mn precursors were assembled onto γ-Al₂O₃ supports by an aging process at room temperature or hydrothermal treatment under 90 °C to enhance the crystal process of Cu-Mn oxides. Toluene was used as the target compound to evaluate the effect of preparation conditions on the Cu-Mn catalyst activity, and the lower conversion temperature can be achieved with Cu-Mn catalysts prepared with glycerol and hydrothermal treatment. The textural and physicochemical properties of each catalyst were measured and reflected that the generation of Cu-Mn spinel could enhance the toluene combustion activity. In general, this finding provides new insight into the polyol choice during the preparation of Cu-Mn catalysts for VOC catalytic combustion.

2. Results and Discussions

2.1. Catalytic Activity of Toluene Combustion

The catalytic activity of Cu-Mn/Al₂O₃ catalysts prepared with different types of polyols for toluene combustion was studied and the results are shown in Figure 1. The results show that different polyol has a great influence on the catalytic activity for toluene combustion and the catalytic activity order for the four catalysts prepared through room temperature aging is as follows: CMA-GL-RT > CMA-BDO-RT > CMA-PDO-RT > CMA-EG-RT. Especially, the T90 and T50 values of CMA-GL-RT can reach 280 and 257 °C, respectively, as shown in Figure 1a. A similar tendency can also be found for the four catalysts prepared through hydrothermal reaction at 90 °C, and their catalytic activity is higher than that prepared by room temperature aging, as shown in Figure 1b. To be specific, the lowest T90 and T50 values can be achieved by CMA-GL-90 at 260 and 237 °C, respectively. The conversion temperature of toluene by CMA-GL-90 was competitive. For example, Kim et al. evaluated different manganese oxide catalysts for the catalytic combustion of toluene, and the T90 values were 270 to 375 °C [29]. Moreover, the catalytic activities of the optimal catalysts were compared by the conversion curve and reaction rate to further compare the performance difference between the Cu-Mn catalysts with different treatments. As shown in the results in Figure 1c,d, the conversion rate of toluene with both catalysts increases with temperature increase, and the conversion rate of CMA-GL-90 was obviously higher than that of CMA-GL-RT in the range between 180 and 280 °C. Especially, the reaction rates at 240, 250, and 260 °C of CMA-GL-90 are 3.711 × 10⁻⁸, 5.137 × 10⁻⁸, and 1.113 × 10⁻⁷ mol∙m⁻²∙s⁻¹, respectively, and that of CMA-GL-RT are 1.374 × 10⁻⁸, 1.940 × 10⁻⁸, and 3.715 × 10⁻⁸ mol∙m⁻²∙s⁻¹, respectively. The maximum difference was close to an order of magnitude. Moreover, compared with other studies, the reaction rates of CMA-GL-90 are considerable. For example, Yang et al. [30] reported that the reaction rate of CuMn₂O₄ with alkali treatment was 7.05 × 10⁻⁸ mol∙m⁻²∙s⁻¹ for toluene conversion at 260 °C. Hence, the hydrothermal treatment can improve the catalytic performance for toluene combustion. The hydrothermal treatment method is more conducive to the formation of CuMn₂O₄ spinel species, which could be more effectively oxidizing toluene, and more discussion about the texture and structure are conducted in the next section to explain those differences.

2.2. Textural and Structural Characterization

The effect of different polyols and hydrothermal treatment on the crystal structure of Cu-Mn/Al₂O₃ catalysts was studied by XRD, as shown in Figure 2. In general, the main diffraction peaks corresponded to γ-Al₂O₃ (2θ = 37.6°, 45.8°, and 66.8°, JCPDS # 1-1303) and CuMn₂O₄ (2θ = 18.5°, 30.4°, 35.9°, and 57.7°, JCPDS # 74-2422), indicating that the spinel structure of CuMn₂O₄ is the main crystal in Cu-Mn/Al₂O₃. Moreover, some weak peaks can be observed in some catalysts belonging to CuO (35.4°, 35.5°, and 38.7°, JCPDS # 48-1548) and MnO₂ (18.1°, 28.8°, 37.5°, and 49.9°, JCPDS # 44-0141).

Figure 2a shows the XRD pattern of Cu-Mn/Al₂O₃ prepared through aging at room temperature. In comparison to the other three catalysts, the intensity of diffraction peaks attributed to the CuMn₂O₄ spinel phase in the CMA-GL-RT catalyst is more pronounced, while the intensity of diffraction peaks associated with the CuO and MnO₂ phase is relatively weaker. This difference may be attributed to the distinct molecular structure of glycerol as compared to other polyols. In the chemical microenvironment created by glycerol, the interaction between Cu and Mn species is intensified, facilitating the formation of CuMn₂O₄ spinel species. Figure 2b illustrates the XRD pattern of Cu-Mn/Al₂O₃ treated by the hydrothermal reaction. Compared with another catalyst, the diffraction peak intensity of CuMn₂O₄ spinel phase in the CMA-GL-90 catalyst remains the most prominent, confirming the effect of glycerol for the formation of spinel structure again. Moreover, after being treated by the hydrothermal reaction, the diffraction peak intensity of the CuMn₂O₄ spinel phase increases, while the diffraction peak intensity of the CuO and MnO₂ phases is notably weakened. Those phenomena can be attributed to the substantial strengthening of the interaction between copper and manganese oxide resulting from the hydrothermal crystallization at 90 °C and the structure-directing effect of polyols. Particularly, glycerol exhibited the most pronounced enhancement in the formation of CuMn₂O₄ spinel.

Figure 3 shows the N₂ adsorption–desorption isotherms of all the Cu-Mn/Al₂O₃ catalysts. Based on the IUPAC classification, all catalysts display the IV-type isotherms with the H3-type hysteresis loops, corresponding to the capillary condensation of N₂, which indicates the existence of the mesoporous structure [31].

Table 1. Texture properties of Cu-Mn/Al₂O₃ catalysts, derived from N₂ physisorption and XPS survey spectra.

Catalysts	S aBET (m2·g−1)	V bBJH (mL·g−1)	rave (nm)	Al at% c	O at% c	Mn at % c	Cu at % c
γ-Al2O3	257.08	0.44	15.40	-	-	-	-
CMA-GL-RT	94.37	0.28	2.67	38.78	57.87	1.52	1.82
CMA-GL-90	95.99	0.31	2.63	38.43	57.91	1.72	1.94
CMA-BDO-RT	106.11	0.31	2.91	37.26	58.64	2.51	1.58
CMA-BDO-90	123.03	0.37	2.78	39.19	57.76	1.51	1.53
CMA-PDO-RT	106.04	0.31	2.87	38.07	59.05	1.47	1.41
CMA-PDO-90	108.25	0.30	2.96	37.59	58.94	1.75	1.71
CMA-EG-RT	106.71	0.28	2.72	37.55	59.46	1.58	1.41
CMA-EG-90	107.81	0.31	2.86	38.28	58.33	1.89	1.50

^a Specific surface area calculated by BET method. ^b Total pore volume based on the BJH method. ^c Atomic concentration derived from XPS survey spectra.

shows the textural properties of Cu-Mn/Al₂O₃ catalysts based on the results of the N₂ adsorption–desorption test and XPS survey spectra. In comparison to the γ-Al₂O₃ support, the Cu-Mn/Al₂O₃ catalyst exhibits a significant reduction in specific surface area, void volume, and average pore diameter. This reduction is primarily attributed to the occupation and blockage of most of the pores in γ-Al₂O₃ by Cu-Mn oxide. Nevertheless, all Cu-Mn/Al₂O₃ catalysts retain an abundance of mesopores, which facilitate the diffusion of toluene into the catalyst and its complete reaction with Cu-Mn oxides, as demonstrated by previous research [32]. Moreover, compared with other catalysts, CMA-GL-RT and CMA-GL-90 have a smaller specific surface area and pore size. This discrepancy may be attributed to the distinct spatial structure of glycerol in contrast to other polyol molecules, and the complexation of metal ions with glycerol molecules enhances the interaction between Cu and Mn, resulting in a closer arrangement of Cu-Mn oxide molecules. Furthermore, the CMA-BDO-90 catalyst exhibits the highest specific surface area. This is due to the larger molecular weight and higher viscosity of 1,4-butanediol, which increases the spacing between the supports during hydrothermal crystallization at 90 °C. Moreover, the atomic concentration of each sample is also listed in

Table 2. The atomic ratios of the catalysts.

Catalyst	Surface Mn/Cu	O1s	Mn2p			Cu2p
		Oads/Olatt	Mn2+/%	Mn3+/%	Mn4+/%	Cu2+/Cu+
CMA-GL-RT	0.83	0.67	80.41	5.18	14.41	5.67
CMA-GL-90	0.88	2.85	37.12	51.03	11.85	0.96
CMA-BDO-RT	1.58	-	54.38	15.74	29.89	5.67
CMA-BDO-90	0.98	0.49	51.91	20.02	28.07	0.27
CMA-PDO-RT	1.04	-	63.21	18.85	17.94	2.33
CMA-PDO-90	1.02	1.70	12.16	60.55	27.29	0.28
CMA-EG-RT	1.12	0.64	72.54	22.57	4.89	0.82
CMA-EG-90	1.27	1.08	49.72	28.15	22.12	0.56

, as derived from XPS survey spectra. Apart from CMA-GL samples, the other catalysts have a similar tendency of atomic concentration, namely that the concentration of the Mn atom is higher than that of the Cu atom, which means that some Mn oxides could be generated rather than CuMn₂O₄. Inversely, the atomic concentration of Mn is lower than that of Cu in the catalysts prepared in GL, indicating that CuMn₂O₄ is more likely to be generated, and a similar tendency can be found in [12].

The morphology and crystal state changes of all catalysts were observed by SEM and TEM. Figure 4a–d shows the SEM image of Cu-Mn/Al₂O₃ catalysts prepared by aging at room temperature. It can be found that Cu-Mn oxide particles dispersed on the surface of the layered γ-Al₂O₃ support, but their morphology is obviously different with different polyols. CMA-GL-RT displays a better crystal structure of Cu-Mn oxide and obvious spinel particles can be observed dispersing uniformly on the surface of the γ-Al₂O₃ support. The spinel structure can also be found in CMA-BDO-RT, but most of those Cu-Mn oxides are dispersed on the support as clusters. Moreover, when PDO and EG were used in the preparation of catalysts, the crystal structure of Cu-Mn oxides underwent conversion. For CMA-PDO-RT, most Cu-Mn oxides present as a layer structure, and for CMA-EG-RT, a cotton-like structure was observed. After being treated by hydrothermal reaction, the crystal structure and distribution of Cu-Mn oxides particles apparently change, as shown in Figure 4e–h. With the same magnification times, there are significantly fewer larger particles in CMA-GL-90, implying more CuMn₂O₄ spinel generated, in agreement with XRD results. Figure 4f reveals the increase in Cu-Mn oxide particles with good dispersion on the surface of the CMA-BDO-90 catalyst. This observation aligns with the significant increase in specific surface area for CMA-BDO-90 measured by N₂ adsorption–desorption measurements. Moreover, for CMA-PDO-90 and CMA-EG-90, more Cu-Mn oxide particles can be observed with clearer crystal features after being treated with a hydrothermal reaction.

More morphology and crystal information was provided by TEM, as shown in Figure 5 and Figure 6. Corresponding to SEM results, the different crystal features can be found with different polyols. For GL and BDO as polyols, most of the Cu-Mn particles exhibit polyhedral structures, as shown in Figure 5a,b, which may correspond to the spinel CuMn₂O₄. However, when PDO and EG were used as polyols, the elongated structures from manganese oxide and pebble-like structures from copper oxides became rich in those catalysts, as shown in Figure 5c,d, so the obvious spinel structure cannot be found in the SEM image of CMA-PDO-RT and CMA-EG-RT, indicating that the copper hardly affects the formation of nanowires from manganese oxide with PDO or EG [12]. It could be explained by the different spaces of the hydroxyl groups from different polyols. For PDO and EG, the space of the hydroxyl group of each molecule is limited, which could complex the single metal ions and format single metal oxides. In contrast, there are rich spaces between hydroxyl groups in BDO and a rich amount of hydroxyl groups in GL, so more metal ions could be bound by one polyol molecule and become prone to forming polymetallic oxides. Hence, BDO and GL are more conducive to generating the CuMn₂O₄ spinel phase to enhance the whole catalytic combustion reaction for toluene. Moreover, after the hydrothermal reaction, the polyhedral structures increase obviously, especially in CMA-BDO-90. It indicates again the enhancement effect of the hydrothermal reaction for the generation of spinel crystal. While the nanowire structures still exist as shown in Figure 5g,h, the amorphous domains observed in the sample aged at room temperature convert to crystal phase after the hydrothermal reaction, and this change is visible is Figure 5d,h. Overall, GL and BDO, with more spaces to bond metal ions, can enhance the formation of Cu-Mn oxide species, so those catalysts possess better performance for toluene combustion. Moreover, the hydrothermal reaction could significantly improve the crystallization of Cu-Mn oxide species, so the conversion performance of toluene was improved in all catalysts.

To further confirm the existence of CuMn₂O₄, the HRTEM and SEAD results of CMA-GL-RT and CMA-GL-90 were analyzed by Digital Micrograph 3.5 software, as shown in Figure 6. It can be found that both catalysts display a single crystal structure with an interlayer spacing of 0.47 nm, corresponding to the (1 1 1) plane of CuMn₂O₄ (JCPDS # 74-2422) [33]. Moreover, the interplanar distances are calculated from the SAED pattern of both catalysts, and narrower interlayer spacing can be found to be 0.249, 0.150, and 0.127 nm in CMA-GL-RT, and 0.251, 0.150, and 0.127 nm in CMA-GL-90, corresponding to (3 1 1), (4 4 0), and (5 3 3) reflection planes for CuMn₂O₄ (JCPDS # 74-2422). In conclusion, CuMn₂O₄ is the main crystal structure in Cu-Mn catalysts prepared with GL.

2.3. XPS and H₂-TPR Analysis

XPS analysis was used to investigate the chemical state of Mn, Cu, and O species in different catalysts. The Mn and Cu 2p_3/2XPS spectra are shown in Figure 7 and Figure 8, and Mn/Cu ratios are reported in Table 3 for each catalyst. The Mn 2p_3/2 spectra can be deconvoluted into three different peaks, ascribed to Mn²⁺ (640.5–640.9 eV), Mn³⁺ (641.8–642.0 eV), and Mn⁴⁺ (643.2–644.5 eV) [34]. The Cu 2p_3/2 XPS spectra of all catalysts can be deconvoluted into two peaks of Cu²⁺ (933.6~934.2 eV) and Cu⁺ (932.5~933.0 eV) [35]. It is worth noting a negative shift of the Cu 2p_3/2 peak at 930.5 eV for the CMA-BDO-RT. Combined with the Mn/Cu ratios in Table 3, this phenomenon could be explained by the fact that the amount of Cu in the CMA-BDO-RT is relatively less, so the Cu⁺ species tend to occupy the tetrahedral sites in the spinel structure [19]. Moreover, as shown in Table 2, the Mn/Cu atomic ratio almost all increased for the catalysts treated by the hydrothermal reaction, reflecting the fact that the hydrothermal process is favorable for the crystal generation of Cu-Mn oxides.

The proportion of each species is summarized in Table 3 and there is a similar tendency of each catalyst for those species. Firstly, Mn²⁺ is the dominant species in manganite oxide for catalyst aged at room temperature, and the proportion of Mn²⁺ decreased in the catalyst after being treated by the hydrothermal reaction. It means that more CuMn₂O₄ spinel forms after the treatment of the hydrothermal reaction, which agrees with the result of XRD analysis. Moreover, the proportion of Mn³⁺ also increased and that of Mn⁴⁺ decreased after being treated by the hydrothermal reaction, and it is the most significant in catalysts prepared in glycerol. As reported in the literature, Mn³⁺ is the main active species for catalytic combustion and Mn⁴⁺ is known as a deactivated species for catalytic oxidation [12]. In agreement with this point, the catalytic performance is the best for CMA-GL-90. Moreover, there is also a certain tendency for the redox species of Cu and the ratios of Cu²⁺/Cu⁺ to decrease in all catalysts after treatment by the hydrothermal treatment, which suggests that CuMn₂O₄ spinel phase forms by the redox cycle of Cu²⁺ + Mn³⁺→Cu⁺ + Mn⁴⁺.

According to previous studies, the formation of surface oxygen vacancies in the CuMn₂O₄ spinel can promote the oxidation of the toluene molecule by generating the reactive O species, which is the key factor for the enhancement of the catalytic combustion of toluene [30]. Hence, the surface O species are analyzed based on O1s XPS spectra, as shown in Figure 9. Three peaks can be deconvoluted, including surface oxygen species O_sur (OH⁻) in 532.5–532.9 eV, the adsorbed oxygen O_ads 531.2–531.6 eV, and the lattice oxygen O_latt in 530.4–530.9 eV, respectively. Only CMA-PDO-RT and CMA-BDO-RT have strong peaks of O_sur, which indicates that OH⁻ on the surface of catalysts could not be removed completely during the preparation process due to the weaker binding effect from BDO and PDO to metal ions. However, after hydrothermal treatment, the peaks of O_sur disappear in all catalysts, indicating that hydrothermal treatment is more beneficial to form Cu-Mn crystals from the precursor, despite the limitation of the bonding effect from polyols. Moreover, the ratios of O_ads/O_latt are also summarized in Table 2. The value of O_ads/O_latt can indirectly reflect the amount of oxygen vacancies, because O_ads is from O₂ adsorption on oxygen vacancies. Obviously, after the hydrothermal treatment, the values of O_ads/O_latt increase for all samples, which represents the abundance of oxygen vacancies on Cu-Mn catalysts. Among them, the values of O_ads/O_latt in CMA-GL-90 are highest at 2.85, and the catalytic performance is the best with CMA-GL-90.

Moreover, the H₂-TPR test was conducted to delve further into the variations of copper and manganese oxides, and their interactions and the results are presented in Figure 10. It is evident that the reduction peak observed in the CMA-EG-RT catalyst at approximately 244 °C is attributed to the reduction of Cu²⁺ within CuO species. Additionally, the shoulder peak around 253.8 °C is attributed to the reduction of Cu²⁺ and Mn³⁺ in CuMn₂O₄ spinel species. The reduction peak in the CMA-GL-RT catalyst is primarily associated with CuMn₂O₄ spinel species [36]. The reduction peaks in the CMA-BDO-RT and CMA-PDO-RT catalysts shift to lower temperatures relative to the CMA-GL-RT catalyst. This shift is due to the presence of a small amount of CuO species, and the high reducibility of Cu species allows for a reduction in the temperature required for the reduction of Cu-Mn oxide [27]. Consequently, the order of oxidation performance among the four catalysts is CMA-GL-RT > CMA-BDO-RT > CMA-PDO-RT > CMA-EG-RT. In Figure 10b, the reduction peaks in all four catalysts are primarily associated with the reduction of Cu²⁺ and Mn³⁺ within CuMn₂O₄ spinel. Notably, the reduction peaks in the CMA-EG-RT and CMA-PDO-RT catalysts have shifted towards lower-temperature regions compared to the other two catalysts. This shift is once again attributed to the presence of a small amount of CuO species in the catalyst. Furthermore, when compared to catalysts prepared through aging at room temperature, those prepared through hydrothermal crystallization at 90 °C exhibit stronger oxidation performance. The order of oxidation performance among the four catalysts is CMA-GL-90 > CMA-BDO-90 > CMA-PDO-90 > CMA-EG-90.

2.4. Catalytic Mechanism

Based on the above-mentioned results, a catalytic combustion mechanism for toluene is proposed with the Cu-Mn catalysts. As shown in Scheme 1, the glycerol with three hydroxy groups can chelate Cu and Mn ions, building a tighter micro-environment for the formation of Cu-Mn complex. Then, the hydrothermal treatment at 90 ℃ can enhance the Cu-Mn crystal formation in the Cu-Mn precursor. Owing to the chelation effect of glycerol and the Cu-Mn crystal formation during hydrothermal treatment, there is more CuMn₂O₄ spinel structure generation in Cu-Mn oxide catalysts after calcination. During the catalytic combustion process, firstly, toluene molecules in the gas stream were adsorbed on the surface of the oxidation state active site of the Cu-Mn catalyst to form adsorbed toluene species. Secondly, O₂ molecules in the gas stream were adsorbed on the surface of the catalyst, resulting from oxygen vacancies from the CuMn₂O₄ spinel structure, and then formed surface active O species, which can undergo redox reaction with absorbed toluene species to generate CO₂ and H₂O. Meanwhile, the reduced state active states formed on the surface of the catalyst to prepare for the next redox reaction. Moreover, Cu²⁺ and Mn³⁺ on the CuMn₂O₄ spinel structure also can provide more surface Lewis acidic sites to enhance the catalytic activity, which is considered the other important factor favoring the catalytic oxidation of toluene [37].

2.5. Catalytic Stability

To investigate the impact of toluene concentration on catalyst stability and catalytic performance, the CMA-GL-90 and CMA-PDO-90 catalysts were selected for online toluene conversion experiments at a specific temperature (300 °C). As shown in Figure 11, the relationship between toluene conversion and reaction time (300 h) is presented. Both the CMA-GL-90 and CMA-PDO-90 catalysts exhibited remarkable catalytic stability. The CMA-PDO-90 catalyst achieved approximately 97% toluene conversion at a toluene concentration of 500 ppm, and toluene conversion reached 100% when the toluene concentration was increased to 1000 ppm. This observation suggests that with increasing toluene concentration, more heat is released from the reaction, which is advantageous for the deep oxidation of toluene. The toluene conversion for the CMA-GL-90 catalyst remained at 100% over the course of 300 h of continuous catalytic combustion reaction. This indicates that the CMA-GL-90 catalyst offers superior stability and a higher conversion rate in catalyzing the combustion reaction of toluene.

Moreover, Table 3 shows the comparison of the toluene conversion performance of CMA-GL-90 with reported catalysts in the other studies. It can be found that the values of T₅₀ and T₉₀ are both considerable. Especially for [38,39], those cases were both conducted with the catalyst with a low concentration of Cu-Mn oxide loading on the γ-Al₂O₃ carrier, and the value T₉₀ (260 °C) of CMA-GL-90 is far lower than a similar catalyst, which indicates the application potential of the preparation of Cu-Mn oxide catalyst with polyol methods for catalytic combustion.

Table 3. Summary of previous research results on toluene combustion by Cu-Mn catalysts.

Catalyst	GHSV	T50 (°C)	T90 (°C)	Ref.
CMA-GL-90	10,000 h−1	237	260	In this study
Cu0.5Mn7.5/Al	120,000 h−1	280	332	[38]
Cu1.5Mn1.5O4	46,154 mL/(g∙h)	239	-	[20]
Cu-MnOx/γ-Al2O3	20,000 h−1	279	310	[39]
Cu-Mn/Al	71,000 mL/(g∙h)	282	301	[40]

Table 3. Summary of previous research results on toluene combustion by Cu-Mn catalysts.

Catalyst	GHSV	T50 (°C)	T90 (°C)	Ref.
CMA-GL-90	10,000 h−1	237	260	In this study
Cu0.5Mn7.5/Al	120,000 h−1	280	332	[38]
Cu1.5Mn1.5O4	46,154 mL/(g∙h)	239	-	[20]
Cu-MnOx/γ-Al2O3	20,000 h−1	279	310	[39]
Cu-Mn/Al	71,000 mL/(g∙h)	282	301	[40]

3. Experimental Section

3.1. Catalyst Preparation

Firstly, certain proportions of Cu (NO₃)₂·3H₂O (purity > 99.9%) and Mn (NO₃)₂ (purity > 99.9%) were dissolved in the glycerol (GL, purity > 99%) aqueous solution and stirred at room temperature for 2.5 h. The molar ratio of Cu: Mn: -OH: H₂O was kept at 1: 2: 24: 12. After the colloid was obtained, we continued to stir at −10 °C for 0.5 h. Then, the pH of the colloid was adjusted to 8.5 under stirring by 1 M NaOH, and the slurry was further aged for 1 h under −10 °C with gentle stirring. Once the temperature of the slurry returned to room temperature, Al₂O₃ powders were added under vigorous stirring for 2 h. The total weight amount of Cu and Mn relative to Al₂O₃ was controlled at 8 wt.%. The resulting mixture was transferred to a 250 mL autoclave for a hydrothermal reaction at 90 °C, lasting 2 h. Following the hydrothermal reaction, the product was filtered and the solid was washed with distilled water until the pH reached 7. Subsequently, the filter cakes were dried at 85 °C and calcined in air at 550 °C for 5 h. The obtained catalyst was denoted as CMA-GL-90.

For comparison, the hydrothermal reaction at 90 °C was substituted with aging at room temperature, and the obtained catalyst was designated as CMA-GL-RT.

Moreover, various other polyols, such as ethylene glycol (EG, purity > 99%), 1,3-propanediol (PDO, purity > 99%), and 1,4-butylene glycol (BDO, purity > 99%), were utilized in this study, each following a similar procedure. The obtained products were named CMA-EG-90, CMA-EG-RT, CMA-PDO-90, CMA-PDO-RT, CMA-BDO-90, and CMA-BDO-RT, respectively. The information about the prepared catalysts is listed in

Table 4. The preparation and chemical composition of different catalysts.

Name	Polyols	Hydrothermal Reaction	Molecular Structure
CMA-GL-90	Glycerol (GL)	+
CMA-GL-RT	Glycerol (GL)	−
CMA-BDO-90	1,4-butylene glycol (BDO)	+
CMA-BDO-RT	1,4-butylene glycol (BDO)	−
CMA-PDO-90	1,3-propanediol (PDO)	+
CMA-PDO-RT	1,3-propanediol (PDO)	−
CMA-EG-90	Ethylene glycol (EG)	+
CMA-EG-RT	Ethylene glycol (EG)	−

.

3.2. Characterization

X-ray diffraction (XRD) patterns of the catalysts were acquired using a DX-2700 powder diffractometer (Dandong Fangyuan Co., Dandong, China). The instrument was operated at 40 kV and 30 mA, with a step size of 0.05 degrees, and Cu Kα radiation (λ = 0.15406 nm) was employed. The determination of the BET surface area and pore volume of the samples was conducted using the BET and BJH methods, respectively. This analysis was based on nitrogen (N₂) adsorption–desorption isotherms recorded at −196 °C. The measurements were performed on an SSA-4200 Micromeritics instrument (Beijing Builder Co., Beijing, China).

H₂-TPR profiles of catalysts (0.04 g) were obtained with a TP-5080 adsorption instrument (Tianjin Xianquan Co., Tianjin, China) with a 5% H₂-N₂ mixture gas (30 mL/min) in the temperature range 20–800 °C with a heating rate of 10 °C/min; the H₂ consumption was monitored with a thermal conductivity detector (TCD).

The catalysts’ morphology was examined using an Ultra 55 scanning electron microscope (SEM, Carl Zeiss AG., Oberkochen, Germany) operated at an accelerating voltage of 5.00 kV. Additionally, transmission electron microscopy (TEM) analysis was carried out on a JEM 3200 transmission electron microscope (JEOL Ltd., Beijing, China), featuring a field emission gun operating at 200 kV. This enabled an in-depth investigation into the dimensions and structural characteristics of the catalysts.

The surface of catalyst samples was qualitatively analyzed using the PHI 5300X X-ray photoelectron spectroscopy (XPS, ULVAC-PHI, INC., Tokyo, Japan). AlKα radiation (energy: 1486.6 eV) was employed as the excitation source, with a beam diameter of d = 400 μm. The sample chamber maintained a vacuum pressure of 10-8-10-9 torr throughout the analysis. C1s at 284.8 eV was utilized as the energy calibration reference point.

3.3. Catalytic Evaluation Test

Toluene combustion experiments were conducted in a fixed-bed reactor with a diameter (Φ) of 15.0 mm, operating at atmospheric pressure. A total of 5 mL of catalyst particles (ranging in size from 20 to 60 mesh, approximately 4.56 g in weight) were sandwiched between layers of quartz sand. The temperature of the catalytic bed was increased with a heating furnace with a temperature collector (Chengdu Zhongkepurui Purification Equipment Co., Chengdu, China) and the reactor temperature was monitored using a K-type thermocouple. Prior to initiating the catalytic reaction, the catalyst underwent an activation process in which it was exposed to a flow of air at a rate of 60 mL/min, maintained at 400 °C for a duration of 3–4 h. Following activation, the catalyst was allowed to cool to the desired reaction temperature. The air flow rate was then adjusted to 833 mL/min, corresponding to a Gas Hourly Space Velocity (GHSV) of 10,000 h⁻¹. Toluene was introduced into the reactor through a bubbler, with a portion of air introduced at 0 °C to facilitate this process. The reaction products were continuously monitored using an online Gas Chromatograph (GC) equipped with a Flame Ionization Detector (FID) and an HP-INNOWAX capillary column. Toluene conversion was calculated based on the toluene concentrations at the reactor inlet and outlet. The range of tested reaction temperatures extended from 50 °C to 350 °C. The catalytic data reported in this study represent the mean values obtained from three parallel tests.

The toluene conversion rate (X_toluene, %) is calculated as follows:

$${\mathrm{X}}_{\mathrm{t}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{e}\mathrm{n}\mathrm{e}}=100\times \frac{{\left[\mathrm{t}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{e}\mathrm{n}\mathrm{e}\right]}_{\mathrm{i}\mathrm{n}}-{\left[\mathrm{t}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{e}\mathrm{n}\mathrm{e}\right]}_{\mathrm{o}\mathrm{u}\mathrm{t}}}{{\left[\mathrm{t}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{e}\mathrm{n}\mathrm{e}\right]}_{\mathrm{i}\mathrm{n}}}$$

where ${\left[\mathrm{t}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{e}\mathrm{n}\mathrm{e}\right]}_{\mathrm{i}\mathrm{n}}$ and ${\left[\mathrm{t}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{e}\mathrm{n}\mathrm{e}\right]}_{\mathrm{o}\mathrm{u}\mathrm{t}}$ are the inlet and outlet concentration of toluene, respectively.

The surface area normalized reaction rate (r, mol∙m⁻² ∙s⁻¹) is calculated as follows:

$$\mathrm{r}=\frac{{\left[\mathrm{t}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{e}\mathrm{n}\mathrm{e}\right]}_{\mathrm{i}\mathrm{n}}·\mathrm{F}}{{\mathrm{m}}_{\mathrm{c}\mathrm{a}\mathrm{t}}·{\mathrm{S}}_{\mathrm{B}\mathrm{E}\mathrm{T}}}·\mathrm{l}\mathrm{n}(1-{\mathrm{X}}_{\mathrm{t}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{e}\mathrm{n}\mathrm{e}})$$

where F represents the flow rate (mol/s), m_cat represents the mass of the catalyst (g), and S_BET is the surface area of the catalyst (m²/g), respectively.

4. Conclusions

In this study, we synthesized a series of Cu-Mn/Al₂O₃ catalysts using the polyol method. Performance evaluation of these eight Cu-Mn/Al₂O₃ catalysts revealed that, among the four different polyols employed, glycerol, with its unique trihydroxyl structure, demonstrated outstanding characteristics. Glycerol enhanced the interaction between Cu and Mn species, facilitating the formation of a higher proportion of CuMn₂O₄ species. Furthermore, the 90 °C hydrothermal crystallization was found to be highly effective in improving the dispersion of Cu-Mn composite oxides and their interaction. When compared to room temperature aging, hydrothermal crystallization at 90 °C significantly enhanced the dispersion of Cu-Mn oxides on the catalyst’s surface, resulting in an ultimate enhancement of its catalytic activity. It is worth noting that CuMn₂O₄ spinel played a pivotal role in catalyzing toluene combustion. These results affirm that when glycerol is employed in the synthesis of Cu-Mn/Al₂O₃ catalysts via hydrothermal crystallization at 90 °C, the catalyst’s structure and performance remain excellent, demonstrating remarkable catalytic activity in toluene removal. This suggests that non-precious metals can be harnessed to craft cost-effective, highly active, and stable catalysts for efficient VOC removal at lower temperatures. However, this study was only focused on the decomposition of toluene, and the removal of halogenated VOC is also important for industrial VOC control, so the safe and effective conversion of halogenated VOCs will be a future topic of ours to study more deeply.