Introduction
Volatile organic compounds (VOCs) could be emitted from various sources, including industrial processes, construction materials, consumer products, and furniture (Jeong et al., Reference Jeong, Yoon, Kim, Jeong, Lee, Chan Kang and Lee2017; Dong et al., Reference Dong, Qu, Qin, Fu, Sun and Duan2019). Fugitive emissions of VOCs would cause damage to human health and the eco-environment (Pöschl and Shiraiwa, Reference Pöschl and Shiraiwa2015; Kamal et al., Reference Kamal, Razzak and Hossain2016; Ismail et al., Reference Ismail, Zahid, Ali, Bakhtiar, Ali, Khan and Zhu2023). The efficient control of VOC emissions is of great importance in maintaining a clean and healthy air environment. Over recent decades, many control technologies have been developed to diminish the emissions of VOCs, including physical adsorption, photocatalytic oxidation, plasma methods, catalytic oxidation, and others (Zhu et al., Reference Zhu, Zhou, Zhu, Xi and He2015; Ranjbaran et al., Reference Ranjbaran, Kamio and Matsuyama2018; Mansoori et al., Reference Mansoori, Ando and Higuchi2019; Deng et al., Reference Deng, Lei, Fu, Jin, Xu and Wang2022; Liu et al., Reference Liu, Zhang, Jiang, Liu, Cao and Ai2022). Of these VOC control techniques, catalytic oxidation has been recognized as the most promising method due to its high conversion efficiency, low reaction temperature, and environmentally friendly degradation characteristics (Liang et al., Reference Liang, Liu, Deng, Zhang, Hou, Zhao, Zhang, Zhang, Wei and Dai2019; Rochard et al., Reference Rochard, Olivet, Tannous, Poupin, Siffert and Cousin2021; Kim et al., Reference Kim, Im, Cho, Jang, Jang, Kim, Kim, Heo, Kim and Lee2022; Dong et al., Reference Dong, Chen, Ye, Zhang and Dai2023).
The core of catalytic oxidation depends on the development of efficient catalysts. To date, the catalysts used for VOC oxidation are mainly divided into noble metal-supported catalysts and metal oxide-based catalysts. Although noble-metal catalysts (e.g. Pt, Au, and Rh) have exhibited excellent catalytic performance in VOC oxidation at low temperatures, their wide applications are limited by their high cost, poor resources, and easy poisoning (Gaálová et al., Reference Gaálová, Topka, Kaluža, Soukup and Barbier2019; Zhang et al., Reference Zhang, Zou, Xu, Li, Zeng and Li2022). In comparison, transition metal oxide (TMO) catalysts (e.g. NiO, FeOx, CoOx, and MnOx) have been studied extensively and show comparatively high catalytic activity for VOC elimination (Liu et al., Reference Liu, Zhang, Li, Zhang, Wang, Zhan, Zheng and Jiang2020; Ilieva et al., Reference Ilieva, Petrova, Venezia, Anghel, State, Avdeev and Tabakova2021; Wang et al., Reference Wang, Wang, An, Shi, Shangguan, Hao, Xu, Tang, Abudula and Guan2021a; Yadav et al., Reference Yadav, Pophali, Verma and Kim2022). To improve the catalytic performance further, mixed metal oxide catalysts (e.g. Cu-Co, Fe-Mn, and Fe-Ni) were investigated, which exhibited better catalytic activity compared with mono-metallic oxide catalysts (Carrillo and Carriazo, Reference Carrillo and Carriazo2015; Chen et al., Reference Chen, Chen, Xu, Xu, Chen, Jia and Chen2017; Wang et al., Reference Wang, He, Li, Wang, Zhou, Xiao, Chen and Lu2021b). Among them, Fe-Mn oxide catalysts have attracted much research attention owing to their adsorptive capacity, excellent redox properties, and sintering resistance (Lashanizadegan et al., Reference Lashanizadegan, Mirzazadeh and Ahmadi2021; Shah et al., Reference Shah, Bailey, Morgan and Taylor2023), showing outstanding catalytic activity and stability in VOC oxidation (Wang et al., Reference Wang, Zhang and Guo2018; Liu et al., Reference Liu, Shen, Lu and Peng2023). These materials demonstrated great efficiency in VOC removal; however, the synthesis process was relatively complex and expensive to prepare in many cases.
Recently, natural mineral materials have gained attention as catalysts or supports in the field of environmental pollution treatment, due to their low cost, distinct structures, and physicochemical properties (Ardakani et al., Reference Ardakani, Mahabadi and Jafari2019; Rezaei et al., Reference Rezaei, Rezaei and Meshkani2019; Miao et al., Reference Miao, Chen, Shang, Liang and Ouyang2022; Miao et al., Reference Miao, Chen, Liang, Shang and Ouyang2023). Limonite is a mineral with nanostructure and relatively high specific surface area, showing great application prospects in environmental pollution control (Tsubouchi et al., Reference Tsubouchi, Hashimoto and Ohtsuka2015; Sahin et al., Reference Sahin, Tapadia and Sharma2016). For example, mechanically activated limonite exhibited excellent efficacy for the adsorption and stabilization of As in soil (Yan et al., Reference Yan, Shao, Wen and Shen2020), and the limonite was also proved to be a catalyst with great efficiency for methylene blue removal in the Fenton system (Toda et al., Reference Toda, Tanaka, Tsuda, Ban, Koveke, Koinuma and Ohira2014). Compared with the removal of heavy metals and organic pollutants in water or soil environments, the treatment of gaseous containments by limonite has been rarely studied. A previous study has shown that the Fe-Mn binary oxides prepared by thermal treatment of natural Mn-rich limonite displayed excellent performance for NO reduction in a reaction temperature range of 130–300°C, due to the large specific surface area and high redox ability (Zhang et al., Reference Zhang, Chen, Liu, Chen, Xu and Qing2018). In fact, these properties of Fe-Mn binary oxides might also be beneficial for the removal of VOCs. On the other hand, keeping in mind the need for catalysts with high activity, which are environmentally friendly, and economic for real-world VOC removal, the preparation of Fe-Mn binary oxides by thermal activation of natural Mn-rich limonite might be a good approach.
In this study, the Fe-Mn binary oxides were prepared by thermal treatment of Mn-rich limonite at different reaction temperatures and investigated for their catalytic activity in the oxidation of toluene. The effect of different calcination temperatures on the catalytic performance of the catalysts was investigated, and the reaction mechanism of toluene oxidation over the thermally treated Mn-rich limonite is discussed. The aims of this work are to (i) explore the feasibility of the thermally treated Mn-rich limonite for toluene oxidation; (ii) to understand the kinetics and mechanism of toluene oxidation over the catalysts; and (iii) to promote the application of the natural Mn-rich limonite in VOC removal. This work provides a reference for the development of low-cost and efficient mineral-based materials for the elimination of VOCs.
Materials and methods
Preparation of catalyst
The catalysts were prepared through the thermal treatment of Mn-rich limonite at different temperatures under air atmosphere (denoted as YL-X, X=300, 400, 500, and 600, respectively). YL-Raw represents the natural Mn-rich limonite. The detailed preparation of the catalysts is shown in the Supplementary material.
Catalyst characterization
The different catalysts were characterized by X-ray diffraction (XRD), X-ray fluorescence (XRF) analysis, Raman spectroscopy, H2 temperature-programmed reduction (H2-TPR), O2 temperature-programmed desorption (O-TPD), spin-trapping electron paramagnetic resonance (EPR), transmission electron microscopy (TEM), Brunauer-Emmett-Teller analysis (BET), X-ray photoelectron spectroscopy (XPS), and in situ diffuse reflectance infrared Fourier transform spectroscopy (in situ DRIFTS). The Supplementary material gives the details of the characterization.
Catalytic activity measurements
The catalytic activity of toluene oxidation was performed in a continuous-flow fixed-bed reactor (Fig. S1 in the Supplementary material). In addition, toluene conversion (η, %) and CO2 yield (YCO2, %) were used to evaluate the catalytic activity of the catalyst. Details of the catalytic performance evaluation tests are shown in the Supplementary material.
Results and Discussion
Catalytic performance
The catalytic activity for toluene oxidation of the catalysts was measured as a function of reaction temperature (Fig. 1). The reaction temperatures corresponding to 50% and 90% toluene conversion (T50 and T90) were used to evaluate the catalytic activity. Experimental measurements of catalytic activity (Fig. 1a) revealed that toluene conversion increased with an increase in reaction temperature, and the catalysts showed different catalytic activities in toluene oxidation. YL-Raw reached T50 and T90 values at 224 and 247°C, respectively. The catalytic activity of the catalysts changed significantly after thermal treatment; YL-300 showed the highest catalytic activity. The T50 and T90 values for YL-300 were 215 and 239°C, respectively, which were lower than those of the other catalysts. In addition, the catalytic activities of Fe2O3 and MnO2 (Fig. S2 in the Supplementary material) were also lower than that of YL-300. According to the values of T90 for toluene conversion, the catalytic activities of the catalysts were as follows: YL-600 (283°C) < YL-500 (267°C) < YL-400 (258°C) < YL-Raw (247°C) < YL-300 (239°C). The catalytic activities of the catalysts (Fig. 1a) gradually decreased as the thermal treatment temperature increased, indicating that proper thermal treatment has a positive effect on the toluene oxidation of the Mn-rich limonite catalyst.
Similar to the overall trend of toluene conversion, CO2 yield (Fig. 1b) showed an increase with the increasing reaction temperature over the catalysts. CO2 yield tests (Fig. 1b) revealed that YL-300 had superior catalytic activity in converting toluene to CO2 compared with the other catalysts, achieving complete oxidation of toluene to CO2 at 260°C. The T90 values of CO2 decreased in the sequence of YL-300 > YL-Raw > YL-400 > YL-500 > YL-600. It should be noted that the CO2 yield was lower than that of the toluene conversion at the same reaction temperature due to the incomplete conversion of the intermediate species (Chen et al., Reference Chen, Chen, Xu, Xu, Chen, Jia and Chen2017). Compared with the previously reported performances of the catalysts (Table 1), YL-300 also had a higher specific reaction rate and toluene conversion under similar reaction conditions, suggesting that Mn-rich limonite has great potential for the removal of VOCs.
Catalytic stability tests
Considering the practical applications, the catalytic stability of YL-300 was evaluated at 215°C (50% toluene conversion) and 239°C (90% toluene conversion), respectively. During the long-term stability experiment over 30 h (Fig. 2a), the toluene conversion of YL-300 at 215 and 239°C remained at 50 and 90%, respectively. Reusability tests (Fig. S3 in the Supplementary material) showed that YL-300 exhibited T50 and T90 values at 215 and 240°C after three runs. The effect of WHSV on the catalytic activity of YL-300 (Fig. S4a in the Supplementary material) showed that total toluene oxidation was achieved below 260°C even at 60,000 mL g–1 h–1). The toluene concentration effects test (Fig. S4b in the Supplementary material) showed that the temperature for total toluene oxidation at 1600 ppm was only about 10°C greater than that at 600 ppm. Catalytic stability tests (Fig. 2a; Figs S3 and S4 in the Supplementary material) confirmed the excellent catalytic stability of YL-300 for toluene oxidation further.
Taking into account that the water vapor in the actual exhaust gas would compete with the reactants during the adsorption process, different concentrations of water vapor were introduced into the reaction system to probe the effect of water vapor on the catalytic performance of YL-300 at 260°C. Water vapor tests (Fig. 2b) showed that the toluene conversion of YL-300 remained unchanged at 100% after reaction for 5 h in the presence of 1 and 5 vol.% of water vapor; however, the CO2 yield gradually decreased to 87% with the increase in water vapor from 1 to 5%. Water vapor tests (Fig. 2b) indicated that the existence of water vapor had a minimal effect on the toluene conversion to intermediate species within a certain range, but inhibited the complete conversion of intermediates (e.g. benzoate species) to CO2 (Liu et al., Reference Liu, Li, Zhang, Li, Zhou, Meng, Guo, Jia and Sun2019). The curves as a function of temperature under different water vapor conditions (Fig. S5 in the Supplementary material) showed that the water vapor had almost no effect on the toluene oxidation over YL-300, indicating the excellent property of YL-300 in water resistance. CO2 yield could rapidly return to the initial level after ceasing the introduction of water vapor, suggesting that the influence of water vapor could be completely reversible.
Kinetic measurements
To compare the catalytic activity of the catalysts for toluene oxidation, kinetic measurements were performed in the absence of external and internal diffusion limitations (Fig. S6 in the Supplementary material). The specific reaction rate and apparent activation energy (E a, kJ mol–1) for toluene oxidation was calculated using the following equations:
where r (mol s–1 m–2) is the specific reaction rate normalized by specific surface area, F (mL min–1) is the total flow rate, $ {Y}_{{\mathrm{CO}}_2} $ (%) corresponds to the conversion of toluene based on CO2 generation, $ {C}_{\mathrm{C}7\mathrm{H}8} $ (ppm) denotes the toluene inlet concentration in the mixed gases, m (g) is the catalyst weight, S BET (m2 g–1) is the specific surface area of the catalyst, T is the reaction temperature, and R represents the gas constant.
Specific reaction rate tests at 200°C (Fig. 3a) showed that YL-300 had the greatest toluene oxidation rate of 4.12×10−8 mol s–1 m-2, which was 1.25 and 1.83 times greater than that of YL-Raw (3.29×10−8 mol s–1 m–2) and YL-500 (2.25×10−8 mol s–1 m–2), respectively. Apparent activation energy tests for toluene oxidation (Fig. 3b) revealed that the E a (Fig. 3b) of YL-300 (40.4 kJ mol–1) was lower than that of YL-Raw (46.8 kJ mol–1) and YL-500 (50.8 kJ mol–1), which was consistent with the catalytic activity (Fig. 1). Kinetic measurements (Fig. 3) further demonstrated that an appropriate thermal treatment temperature for YL could effectively improve its catalytic activity for toluene oxidation.
Catalyst characterizations
Structural and crystal properties
The structural properties of the catalysts were identified by XRD analysis. The characteristic reflections observed at 21.3, 36.6, and 53.6°2θ for YL-Raw (Fig. 4a) corresponded to the crystal faces of (110), (111), and (221) in goethite (α-FeOOH, JCPDS-81-463) (Chen et al., Reference Chen, Chen, Xie, Xu, Liu and Zhou2018). The peaks located at 20.8, 26.6, 50.2, 59.9, and 68.2°2θ were assigned to quartz (JCPDS-85-795), further demonstrating that YL-Raw was mainly composed of goethite and quartz. No peaks of MnOx were found in the XRD patterns of YL-Raw, indicating that the MnOx in YL-Raw was amorphous or poorly crystallized (Mane et al., Reference Mane, Kim, Han, Kim, Soo Lee, Roh, Lee and Jeon2023). After thermal treatment at 300°C, some new diffraction peaks corresponding to hematite (α-Fe2O3, JCPDS-2-915) were observed in YL-300, which were attributed to the transformation of α-FeOOH into α-Fe2O3 through the dehydroxylation process (González et al., Reference González, Sagarzazu and Villalba2000). Compared with YL-300, the intensities of α-Fe2O3 increased in the cases of YL-400, YL-500, and YL-600, while the increase in crystallinity might be related to the decrease in catalytic activity.
Raman vibration spectra of YL-Raw (Fig. 4b) showed four peaks located at about 298, 498, 557, and 664 cm–1 were associated with the symmetric and asymmetric stretching vibration modes of Fe-OH or Fe-O in α-FeOOH, respectively (Thibeau et al., Reference Thibeau, Brown and Heidersbach1978; Liu et al., Reference Liu, Lu, Li, Zhang, Pan, Zhang, Li and Xiang2018). Two peaks were found at 405 and 219 cm–1 after thermal treatment, which were assigned to the Eg and A1g vibrations of the Fe-O bond in α-Fe2O3 (Oh et al., Reference Oh, Cook and Townsend1998). Two peaks related to the stretching vibrations of the Mn-O bond within the octahedron structure were observed at 259 and 618 cm–1 (White and Keramidas, Reference White and Keramidas1972). The Raman spectral band at 1137 cm–1 was related to the oxygen vacancies in the metal oxide catalysts (Spanier et al., Reference Spanier, Robinson, Zhang, Chan and Herman2001). Raman tests (Fig. 4b) revealed that YL-300 had a distinct peak at 1137 cm–1, indicating that an appropriate thermal treatment temperature was favorable for the formation of oxygen vacancies. The presence of oxygen vacancies facilitated the replenishment and activation of gas-phase oxygen and promoted the toluene oxidation activity, which was consistent with the catalytic activity of the catalysts (Fig. 1).
TEM images (Fig. 5) revealed the structures of the catalysts at different thermal treatment temperatures. Mn was found in EDS mapping (Fig. 5a–d) despite the phases of MnO2 being absent in XRD patterns due to the high distribution of Mn in limonite or the detection limit of XRD (Fig. 5b,c).
TEM images of YL-Raw (Fig. 5e) showed that YL-Raw had a compact needle-like structure with a diameter of around 4–8 nm. Selected area electron diffraction (SAED; Fig. 5f) and HRTEM (Fig. 5g) revealed that the YL-Raw sample consisted mainly of compact needle-like α-FeOOH and MnO2. The structure of YL-300 (Fig. 5h) became loose after calcination at 300°C, but still maintained the needle-like structure. Also, the specific surface area increased from 96.9 m2 g–1 (YL-Raw) to 115.8 m2 g–1 (YL-300) after calcination at 300°C. The crystal plane of α-Fe2O3 found in YL-300 (Fig. 5i,j) showed that α-FeOOH particles were transformed into α-Fe2O3 after thermal treatment. The structure of YL-500 particles remained unchanged after further calcination at 500°C (Fig. 5k), and the plane of α-Fe2O3 was also observed (Fig. 5l,m). The specific surface area of YL-500 (80.6 m2 g–1) decreased compared with YL-Raw and YL-300 due to the sintering of the particles at high calcination temperatures. Some defect pits with irregular lattice spacing (marked with orange dashed lines in Fig. 5g,j,m) were found in the samples, which might be associated with oxygen vacancies generated on the surface of α-FeOOH (Liu et al., Reference Liu, Lu, Li, Zhang, Pan, Zhang, Li and Xiang2018; Cao et al., Reference Cao, Zheng, Du, Shen, Zheng, Au and Jiang2019). Defect sites served as active reaction centers, facilitating the transformation of gas-phase oxygen into reactive oxygen species and thus increasing the toluene oxidation reaction rate (Yang et al., Reference Yang, Su, Xu, Yang, Peng and Li2020).
Surface properties
The surface chemical environment on the surfaces of the catalysts was investigated, and the ratios of Fe, Mn, and O species were calculated by their integrated areas (Fig. 6). The Fe 2p3/2 spectrum (Fig. 6a) showed two distinct peaks centered at about 710.5 and 712.5 eV, corresponding to Fe2+ and Fe3+ on the surface of the catalysts, respectively (Liang et al., Reference Liang, Wang, Wen, Liu, Zhang, Liu, Zhu and Long2022). The surface content ratio of Fe3+/(Fe2++Fe3+) of the catalysts decreased in the order of YL-300 (62.2%) > YL-Raw (59.1%) > YL-500 (53.6%). Previous studies have proved that a greater concentration of Fe3+ could improve the catalytic performance of the catalysts in the low-temperature oxidation of VOCs (Xue et al., Reference Xue, Li, Gao and Wang2020). The Mn 2p3/2 spectra (Fig. 6b) of the catalysts were classified into three peaks: Mn2+ (641.0 eV), Mn3+ (642.4 eV), and Mn4+ (643.7 eV) (Zhang et al., Reference Zhang, Chen, Liu, Chen, Xu and Qing2018). The formation of Mn3+ in the catalyst could facilitate the dissociation and activation of oxygen atoms and the generation of oxygen vacancies. More Mn3+ led to more oxygen vacancies and improved the catalytic performance for the oxidation of VOCs (Hong et al., Reference Hong, Zhu, Sun, Wang, Li and Shen2019; Yang et al., Reference Yang, Su, Xu, Yang, Peng and Li2020). YL-300 had the greatest ratio of Mn3+ (51.5%) compared with that of YL-Raw (42.0%) and YL-500 (40.6%), which was consistent with the catalytic performance in Fig. 1.
Figure 6c shows three distinct peaks at around 529.9, 531.5, and 532.5 eV, which correspond to the lattice oxygen (Olatt), surface adsorbed oxygen (Oads), and surface hydroxyl species (OOH), respectively (Patterson et al., Reference Patterson, Carver, Leyden and Hercules1976). In general, the amount of Oads represents the amount of surface oxygen vacancies on the catalyst surface and plays an important role in promoting the deep oxidation of VOCs (Hong et al., Reference Hong, Zhu, Sun, Wang, Li and Shen2019). The ratio of Oads/Ototal decreased in the order YL-300 (51.4%) > YL-Raw (46.0%) > YL-500 (35.8%), showing the same trend as the catalytic performance in the catalysts. The variation of Oads/Ototal (Fig. 6c) and the catalytic activity (Fig. 1) indicated that a proper thermal treatment temperature was favorable to promote the generation of surface oxygen vacancies, thus enhancing the catalytic performance of the catalysts. The amount of oxygen vacancy in the catalysts was also examined by electron paramagnetic resonance (EPR) spectroscopy (Fig. 6d). YL-300 exhibited the strongest oxygen vacancy signal at g = 2.003, which was higher than that of YL-Raw and YL-500. According to the Mars–van Krevelen mechanism, oxygen vacancies play a key role in toluene oxidation. The greater the concentration of oxygen vacancies, the easier the adsorption and activation of gas-phase oxygen to form adsorbed oxygen species (Oads) on the catalyst surface. EPR test (Fig. 6d) showed that the amount of oxygen vacancies was in the order of YL-300 > YL-Raw > YL-500. XPS and EPR tests of the catalysts (Fig. 6) further indicated that the proper thermal treatment promoted the generation of surface oxygen vacancies and facilitated the mobility of the oxygen species, thus improving the catalytic activity of the catalysts for toluene oxidation.
Redox properties of the catalysts
H2-TPR and O2-TPD experiments were conducted to evaluate the redox behavior of the catalysts. H2-TPR test of YL-Raw (Fig. 7) showed four peaks at 311, 340, 420, and 698°C, relating to the reduction of MnO2 to Mn2O3, Mn2O3 to Mn3O4, Fe2O3 to Fe3O4, and Fe3O4 to FeO, respectively (Durán et al., Reference Durán, Barbero, Cadús, Rojas, Centeno and Odriozola2009; Chen et al., Reference Chen, Chen, Xu, Xu, Chen, Jia and Chen2017; Qin et al., Reference Qin, Huang, Zhao, Wang and Han2019). These reduction peaks shifted to lower temperatures over YL-300 while shifting to higher temperatures over YL-500. Experimental measurements of H2-TPR (Fig. 7) demonstrated that the thermal treatment temperature would affect the reducibility of the catalysts. The reducibility of the catalysts decreased in the order YL-300 > YL-Raw > YL-500, consistent with their catalytic performance for toluene conversion.
The O2-TPD curve (Fig. 7b) showed three desorption regions: low (100–300°C), medium (300–550°C), and high (>550°C) temperature regions. These regions corresponded to different oxygen species: physical or chemically adsorbed oxygen species (OI), the oxygen adsorbed on surface vacancies or surficial lattice oxygen (OII), and bulk lattice oxygen species (OIII), respectively (Nie et al., Reference Nie, Mei, Xiong, Peng, Ren, Hernandez, DeLaRiva, Wang, Engelhard, Kovarik, Datye and Wang2017; Sun et al., Reference Sun, Xu, Bai, Li, Kang, Hu, Liao and He2022). The desorption peaks of YL-300 were lower than those of YL-Raw and YL-500, indicating its better oxygen migration ability, resulting in a faster reaction with adsorbed toluene on the catalyst surface. Previous studies have indicated that these oxygens of OII were generally adsorbed on the sites of oxygen vacancies and participated in the reactions for the VOCs deep oxidation (Hong et al., Reference Hong, Zhu, Sun, Wang, Li and Shen2019). Among these catalysts, YL-300 had the largest integrated area of OII compared with YL-Raw and YL-500, thus displaying the best catalytic activity for toluene oxidation. The H2-TPR and O2-TPD experiments (Fig. 7) showed that the oxygen migration ability and redox capacity of the catalysts were in the order of YL-300 > YL-Raw > YL-500, indicating that an appropriate thermal treatment could promote toluene oxidation by improving the redox properties.
Relationship between catalytic activity and catalyst properties
The thermal treatment temperature showed a significant effect on the properties of the catalyst, including specific surface area, oxygen species, and oxygen vacancies. These variations would further affect the catalytic performance of toluene oxidation. YL-300 had the largest specific surface area of 115.8 m2 g–1, which was higher than that of YL-Raw (96.9 m2 g–1) and YL-500 (80.6 m2 g–1), respectively. In general, a larger specific surface area corresponds to a lager contact area and more effective catalytic activity (Cheng et al., Reference Cheng, Song, Zhang, Qi, Liang, Cui, Guo, Jin, Li, Mao, Pang and Fan2023). The presence of abundant surface oxygen vacancies could contribute to the participation of more surface-adsorbed oxygen species in the toluene oxidation reaction (Li et al., Reference Li, Zhang, Wang, Huang, Peng and Li2018). The XPS results indicated the largest ratio of Oads/Ototal over YL-300. The linear relationship (Fig. 8) between T 90 and the specific surface area (R 2=0.86), surface Oads abundance (R 2=0.98), and surface oxygen vacancy abundance (R 2=0.92) of the catalysts was calculated. Therefore, the surface Oads should be the most significant factor for toluene oxidation, while the specific surface area might not be the key factor for the catalytic activity.
Toluene oxidation pathway and mechanism
In situ DRIFTS was used to investigate the generated intermediates and the reaction mechanism further during the toluene oxidation over YL-300. The peaks corresponding to various intermediate species related to toluene oxidation and their attribution were summarized (see Supplementary material, Table S1).
The typical set of temperature-dependent DRIFTS spectra for toluene oxidation over YL-300 within the range of 30–260°C was performed (Fig. 9a). The surface adsorption species changed significantly with the variations of reaction temperatures. For example, a strong band appearing at 1493 cm–1 upon the introduction of toluene into the system at 30°C should be ascribed to the υ(C=C) vibration of the aromatic ring in toluene, demonstrating the adsorption of toluene on the surface of YL-300 (Yang et al., Reference Yang, Su, Xu, Yang, Peng and Li2020). Two weak peaks at 1081 and 1178 cm–1 were associated with the stretching vibrations of υ(C-O) in benzyl alcohol (Mo et al., Reference Mo, Zhang, Li, Sun, Ren, Zou, Zhang, Lu, Fu, Mo, Wu, Huang and Ye2020). This indicated that the adsorbed toluene on the surface of YL-300 was further transformed into benzyl alcohol (C6H5-CH2O-) species at low temperatures by the oxidation of surface oxygen species (Dong et al., Reference Dong, Qu, Qin, Fu, Sun and Duan2019). The 1493 cm–1 band decreased significantly when the reaction temperature was increased to 120°C, while two bands at 1452 and 1594 cm–1 corresponding to the vibrations of the skeleton υ(C=C) in benzene ring of benzaldehyde were observed (Shen et al., Reference Shen, Deng, Han, Ren and Zhang2022; Shen et al., Reference Shen, Deng, Hu, Chen, Yang, Cheng and Zhang2023). This observation confirmed the further degradation of toluene and oxidation of benzyl alcohol groups at greater reaction temperatures. Two peaks at 1414 and 1543 cm–1 corresponded to the symmetric and anti-symmetric stretching vibrations of υ(C-O) in the carboxylate group, respectively, suggesting the formation of benzoate species on the catalyst surface (Shen et al., Reference Shen, Deng, Han, Ren and Zhang2022). A weak peak of the maleic anhydride species appeared at 1304 cm–1 when the temperature reached 175°C, indicating benzyl alcohol and benzoate would be further converted into anhydride species (Tang et al., Reference Tang, Zhao, Jiang, Huang, Zhang and Li2022).
To distinguish the change of the intermediate species during toluene oxidation, the integrated area of characteristic peaks representing the intermediates of benzyl alcohol (1178 cm–1), anhydride (1304 cm–1), and benzoate (1414 cm–1) were evaluated (Fig. 9b). Benzyl alcohol and benzoate species accumulated on the catalyst surface at lower reaction temperatures, and then, gradually converted into anhydride species at higher temperatures. The integrated area of benzoate species reached the maximum at 200°C, while the maximum integrated area of anhydride was presented at 240°C. In general, ring opening usually demands greater reaction temperatures. It can be deduced that the further conversion of benzoate species into anhydride species should be the rate-limiting step in the oxidation of toluene. The time-dependent DRIFTS spectra of toluene oxidation over YL-300 recorded at 260°C (Fig. 9c,d) showed that the anhydride did not accumulate further as the reaction time increased, demonstrating that the intermediates were continuously oxidized into H2O and CO2.
To explore the effect of oxygen species, the toluene adsorption process over YL-300 was exposed to 300 ppm toluene-N2 (Fig. 9e). The intensities of the peaks corresponding to various intermediate species and the peak areas increased with the reaction temperature under the N2 condition (Fig. 9f). It should be noted that these peaks of different intermediates had reduced intensities under air condition than those of the peaks under N2. The results indicated that the presence of surface lattice oxygen over YL-300 could be utilized and responsible for the oxidation of toluene to some extent, whereas these intermediates were difficult to completely oxidize into H2O and CO2 without the replenishment of gaseous oxygen. This was attributed to the fact that gaseous oxygen could be readily replenished on the catalyst surface and facilitated the regeneration of reactive oxygen species, thus improving the deep conversion of intermediates (Mo et al., Reference Mo, Zhang, Li, Sun, Ren, Zou, Zhang, Lu, Fu, Mo, Wu, Huang and Ye2020).
Based on the in situ DRIFTS results, the reaction mechanism for toluene oxidation over YL-300 (Fig. 10) involved the initial adsorption of toluene on the catalyst surface at low temperature, then the C-H bond of the methyl group was attacked by reactive oxygen species to generate benzyl (C6H5-CH2) and benzyl alcohol (C6H5-CH2O-). After that, the oxidation of benzyl alcohol continued with increasing temperature to form benzaldehyde, benzoate (C6H5-COO), and maleic anhydride, and finally toluene was converted into H2O and CO2. The rate-limiting step in toluene oxidation was determined to be the further oxidation of benzoate species accumulated on the YL-300 surface. Surface lattice oxygen in YL-300 could partially participate in toluene oxidation, but the intermediates formed during the oxidation by surface lattice oxygen species were difficult to further convert into CO2 and H2O. The oxygen adsorbed on the surface of YL-300 played a crucial role in toluene oxidation and could be replenished by activating the gas-phase oxygen through oxygen vacancies.
Conclusions
In this study, the Fe-Mn oxide catalysts with nanostructure were obtained successfully through the thermal treatment of natural Mn-rich limonite. The catalytic performance and reaction mechanism of the catalysts in toluene oxidation were investigated. The different thermal treatment temperatures induced the variations in specific surface areas and oxygen vacancies of the catalysts, which further affected the catalytic activities for toluene conversion. YL-300 exhibited the best catalytic activity with T 90 (90%) value of 239°C and 100% toluene conversion at 260°C. The existence of water vapor (5 vol.%) showed negligible effect on the toluene conversion of YL-300 at 260°C. The excellent catalytic activity of YL-300 was related to its higher surface Oads species and abundant oxygen vacancies. In situ DRIFTS results suggested that the conversion pathway for toluene oxidation over YL-300 was: toluene → benzyl species → benzyl alcohol → benzaldehyde → benzoate species→ maleic anhydride → H2O and CO2. The cleavage of the C=C bond in benzoate species should be regarded as the rate-limiting step. This work provides a reference for exploring the effect of thermal treatment temperature on the properties of natural materials, thus improving the catalytic activity and promoting the application of natural materials.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/cmn.2024.19.
Author contribution
All authors contributed to the study concept and design. Methodology, writing - original draft, data curation, and formal analysis were performed by Shiwei Dong. Supervision and funding acquisition were performed by Tianhu Chen. Yinsheng Zhang and Haibo Liu made the formal analysis and investigation. Minghao Ji participated in the process of formal analysis and validation. Chengzhu Zhu supported the resources. Conceptualization, writing - review & editing, and funding acquisition were performed by Xuehua Zou.
Acknowledgements
None.
Financial support
The work was supported by the National Natural Science Foundation of China (Nos. 42102029 and 41872040), the Anhui Provincial Natural Science Foundation (No. 2108085QD164), and the Fundamental Research Funds for the Central Universities (No. JZ2022HGTB0359).
Competing interest
The authors declare that they have no competing interests.
Data availability statement
Not applicable.