Morphology dependence of Nb₂O₅-supported cobalt oxide in catalytic toluene oxidation†

Shuangju Li^a, Xueli Cheng^a, Wei Zhou^b, Junxiang Jiang^a, Chao Feng^a, Yuanshuai Liu*^a, Xuebing Li^c, Xiaodong Zhang^d and Zhong Wang*^a
aKey Laboratory of Biofuels, Qingdao Institute of Bioenergy and Bioprocess Technology, Chinese Academy of Sciences, Qingdao 266101, Shandong Province, PR China. E-mail: liuys@qibebt.ac.cn; wangzhong@qibebt.ac.cn
^bTongliao Technology Service Center of Ecological and Environment, Tongliao 028000, PR China
^cCollege of Materials Science and Engineering, Ningbo University of Technology, Ningbo 315211, PR China
^dEnvironment and Low-Carbon Research Center, School of Environment and Architecture, University of Shanghai for Science and Technology, Shanghai 200093, PR China

First published on 6th August 2024

---

---

1. Introduction

Volatile organic compounds (VOCs) serve as key precursors to O₃ and PM_2.5, posing significant threats to both ecosystems and human health.^1–4 Among the array of technologies aimed at addressing this issue, catalytic oxidation stands out as a particularly promising approach.^5–9 At the core of this approach lies the imperative quest for catalysts that are both cost-effective and exhibit highly activity. Among these, the transition metal Co₃O₄ catalyst, characterized by its abundance and affordability, has garnered substantial interest owing to its impressive performance in catalytic oxidation processes.^10–13 Recently, Xiao et al.¹⁴ reviewed the latest progress and advancements in the application of cobalt-based catalysts for the low-temperature catalytic oxidation of VOCs. They also highlighted that it is crucial to study non-noble composite cobalt-based catalysts for the construction of synergistic effects, active interfaces, and oxygen vacancies.

In our previous work,¹⁵ we reported the synthesis of CoO_x-based catalysts supported on various materials, including Nb₂O₅, SiO₂, and Al₂O₃. We found that the nature of the support significantly influenced the catalytic performance of the CoO_x-based catalysts. Notably, the Nb₂O₅ support tends to enhance the catalytic activity of CoO_x due to its unique electronic and structural properties. Furthermore, considering that shape control is an effective approach for altering the catalytic performance of supported catalysts, numerous researchers have employed this technique using Nb₂O₅ of various shapes to achieve adjustable catalytic activities in Nb₂O₅-supported catalysts.^16–18 For example, Kommula et al.¹⁹ synthesized diverse Nb₂O₅ nanostructures such as thick flakes, thin sheets, small spikes, and elongated spiky sea urchins and found that the electrochemical capacitance of suNb₂O₅-CNS (spiky sea urchin Nb₂O₅ grown in carbon nanospheres) composites display the highest capacitance of ≈32.5 F g⁻¹. Guo et al.²⁰ synthesized three Ru/Nb₂O₅ catalysts with different morphologies (layer, hollow, and flower-like Nb₂O₅) in the reductive amination of cyclopentanone, and the layer-like Nb₂O₅-loaded Ru catalyst achieved the best performance, which was related to good Ru dispersion over its surface. Despite the extensive research on the catalytic applications of Nb₂O₅, its unique morphological characteristics in the process of catalytic oxidation of toluene have not been fully reported.

Herein, we studied the performances of CoO_x supported on three types of Nb₂O₅ with different morphologies (rod, grid, and spherical) in the oxidation of toluene. The comprehensive structure–property relationship was investigated through multiple techniques to get a better understanding of the synergistic morphology effect.

2. Experimental section

2.1. Catalysts preparation

Rod-like Nb₂O₅ (labeled as Nb₂O₅-R) was synthesized by the hydrothermal method. Ammonium niobate (C₄H₄NNbO₉·xH₂O) was dissolved in 40 mL deionized water. Then, the solution was transferred into a 100 mL Teflon-lined autoclave, crystallized at 175 °C for three days, and cooled overnight. The obtained solid product was separated by centrifugation and washed three times with deionized water. After drying overnight at 80 °C, it was calcined at 400 °C (5 °C min⁻¹) for 4 h in a muffle furnace to finally yield the spherical Nb₂O₅ support.

Grid-like Nb₂O₅ (labeled as Nb₂O₅-G) was synthesized as follows. P123 (polyethylene oxide–polypropylene oxide–polyethylene oxide) was dissolved in ethanol, then NbCl₅ was added under stirring at room temperature and continually stirred for another 30 min. Afterwards, the deionized water was added. After the mixture was aged for two days at 40 °C, the product was calcined in a muffle furnace at 400 °C (5 °C min⁻¹) for 4 h to obtain the yellow grid Nb₂O₅ support.

For the synthesis of spherical Nb₂O₅ (labeled as Nb₂O₅-S), hydroxy niobate (H₅Nb₃O₁₀) and dihydrate oxalic acid (H₂C₂O₄·2H₂O) were dissolved in 20 mL deionized water and stirred at 80 °C for 10 min. Then, the mixture was transferred into a 100 mL Teflon-lined autoclave with 20 mL ethanol. Hydrothermal treatment was performed at 175 °C for 12 h. Afterward, the solid product was collected by centrifugation and washed three times with deionized water, followed by vacuum drying at 50 °C for 12 h. Finally, the obtained product was calcinated at 550 °C (2 °C min⁻¹) for 4 h, thus yielding the Nb₂O₅-S support.

The 10% CoO_x/Nb₂O₅ catalysts were prepared using incipient wetness impregnation (Co(NO₃)₂·6H₂O as the precursor) and designated. After being statically aged overnight at room temperature, the obtained materials were dried at 110 °C overnight and calcined at 400 °C (5 °C min⁻¹) for 4 h in a muffle furnace.

2.2. Characterization of the catalysts

The details of X-ray diffraction (XRD), transmission electron microscope (TEM), inductively coupled plasma optical emission (ICP-OES) spectrometry, N₂ adsorption/desorption measurements, X-ray photoelectron spectra (XPS), H₂ temperature-programmed reduction (H₂-TPR), and NH₃ temperature-programmed desorption (NH₃-TPD) are provided in the ESI.†

2.3. Activity measurement

The test for examining various catalysts' performance in toluene oxidation was conducted within a fixed-bed flow reactor setup. Normally, 0.2 g catalyst, with particle size between 60 and 80 mesh, was loaded into a quartz glass tube that had an internal diameter of 6 millimeters. The inlet gas mixture comprised 1000 ppm toluene, which was bubbled with helium in an ice-water bath, along with 20% oxygen balanced with helium. Throughout the process, the temperature was initially boosted to 100 °C under helium pretreatment for one hour before the reactant gas, flowing at a rate of 100 mL min⁻¹, and passed over the catalyst-packed bed. The temperature within the bed was meticulously monitored using a thermocouple and varied from 200 to 400 °C, with each set point maintained for 40 minutes. The outlet gases were analyzed via online mass spectrometry using an Agilent 7890A instrument. Toluene conversion (X_C7H8) and TOF_Co served as metrics to assess the efficiency of the catalysts, and the calculation formula were as follows.

In the above equations, [C₇H₈]_in and [C₇H₈]_out represent the concentrations of toluene in the inlet and outlet, respectively. V_gas is the total molar flow rate.

2.4. DFT calculation details

DFT calculation were carried out by the Vienna ab initio simulation package (VASP). The definition and calculation details are presented in the ESI.†

3. Results and discussion

3.1. Catalytic performance for toluene oxidation

The catalytic performances of the prepared CoO_x/Nb₂O₅ catalysts for toluene oxidation were assessed by conducting experiments and long-term stability tests. As shown in Fig. 1a, pure spherical-like Nb₂O₅ had no toluene oxidation activity in the temperature range of 260–400 °C, while the toluene could be fully oxidized after CoO_x loading. To be specific, the temperatures of toluene conversion at 10%, 50% and 90% (T₁₀, T₅₀ and T₉₀) are summarized in Table 1. Among them, the CoO_x/Nb₂O₅-S catalyst with spherical morphology showed the best catalytic activity. The T₉₀ of all the catalysts in the reaction followed the sequence: CoO_x/Nb₂O₅-G (392 °C) < CoO_x/Nb₂O₅-R (385 °C) < CoO_x/Nb₂O₅-S (339 °C). Meanwhile, the turnover frequency (TOF) of these CoO_x/Nb₂O₅ catalysts at 283 °C was also calculated according to the above equation. It was found that the sequence of TOF values (Table 1) were consistent with the catalytic performance. Furthermore, from the linear Arrhenius plot for catalytic toluene oxidation (Fig. 1b), it can be estimated that CoO_x/Nb₂O₅-S had a relatively lower apparent activation energy compared to CoO_x/Nb₂O₅-R and CoO_x/Nb₂O₅-G, which were 92, 106, and 113 kJ mol⁻¹, respectively. The difference in the apparent activation energy correlated well with the overall low-temperature catalytic activities in the reaction, further substantiating the positive impact of the spherical Nb₂O₅ support's morphology on toluene activation and oxidation. Furthermore, Table 2 also summarized the results of previous studies where different CoO_x-based catalysts were used for the oxidation of toluene. The CoO_x/Nb₂O₅-S catalyst evaluated in our study exhibited improved performance in the catalytic oxidation of toluene.

---

---

Simultaneously, the thermal stability and water resistance of CoO_x/Nb₂O₅-S catalyst were evaluated. As seen in Fig. 2a, there was no indication of deactivation after five cycles using the CoO_x/Nb₂O₅-S catalyst; toluene could still be completely oxidized at 340 °C. Moreover, a long-term stability test over the CoO_x/Nb₂O₅-S catalyst was conducted at 336 °C. As seen in Fig. 2b, the catalyst continued to react for 120 h, maintaining a consistent level of toluene conversion. XRD was used to analyze the catalyst after long-term stability test; as depicted in Fig. 2c, there was no variations in the intensity of the diffraction peak. This indicated the structural stability of the catalyst following the long-term stability test. Furthermore, the CoO_x/Nb₂O₅-S catalyst was studied before and after the stability test using ICP-OES. It was found that the content of Nb and Co after the reaction was 53.63% and 8.93%, respectively, which was not significantly different from their content before the reaction (54.68% and 9.04%). This indicated that no apparent leaching of Nb or Co was observed. In addition, the effects of dry and humid conditions on the oxidation of toluene by CoO_x/Nb₂O₅-S were examined at 330 °C (Fig. 2d). In dry conditions, the toluene conversion rate was 63%, which decreased to about 37% upon exposure to 1% water vapor. However, when the water vapor was removed, the catalyst's activity rapidly returned to 68% and even exceeded the initial rate. With 3% water vapor, the toluene conversion rate reduced to 31%, but it rebounded to approximately 72% once the water vapor was eliminated. Notably, in humid environments, the competition between toluene and water for adsorption onto the catalyst surface led to a decrease in the available active sites, and consequently, a reduced rate of toluene oxidation.^25,26

3.2. Structure and morphology analysis

Fig. 3 shows the XRD patterns of the synthesized Nb₂O₅ supports and the corresponding supported cobalt catalysts. Patterns of Nb₂O₅-R showed the diffraction peaks at 22.7° and 46.2° belonging to the (001) and (002) crystal planes of Nb₂O₅ (JCPD: 30-0837), respectively.¹⁷ Nb₂O₅-G support had no characteristic diffraction peaks, which indicated its amorphous structure.²⁷ The diffraction peaks of the Nb₂O₅-S support at 2theta angles of 22.6°, 28.3°, 28.9°, 36.5°, 36.9°, 46.1° and 50.9° were ascribed to the (001), (180), (200), (181), (201), (002) and (380) planes of Nb₂O₅ (JCPDS: 07-0061) with hexagonal phase, respectively. It was worth noting that the diffraction peak intensity of all Nb₂O₅ supports decreases after CoO_x loading. In addition, the minor diffraction peaks at 36.8°, assignable to Co₃O₄ (311) crystalline phases, were observed over CoO_x/Nb₂O₅-R and CoO_x/Nb₂O₅-G catalysts, probably due to the good dispersion or small particle size of Co over niobium-based supports. However, both Co₃O₄ and Nb₂O₅-S had the same diffraction peak at 36.8°, making it hard to distinguish.

The morphology of the as-synthesized Nb₂O₅ supports and CoO_x-loaded Nb₂O₅ were studied by TEM, as shown in Fig. 4. Fig. 4a displays that the Nb₂O₅-R synthesized by the hydrothermal method existed in the form of nanorods with a length of 40–100 nm, growing along the (001) direction, which was evidenced by the lattice fringe with an interplanar spacing of 0.39 nm. Obviously, Nb₂O₅-G and Nb₂O₅-S presented a typical irregular grid (Fig. 4b) and spherical (Fig. 4c) morphology, respectively. It was worth noting that the morphologies of the supports were essentially retained after the loading of CoO_x (Fig. 4d–f). However, it is hard to distinguish the Co₃O₄ phase and Nb₂O₅ phase in TEM due to the similar electron density.²⁸ CoO_x/Nb₂O₅-S in Fig. 4i displays the lattice spacing of 0.314 nm corresponding to the (180) facet of Nb₂O₅, which was in agreement with the XRD results.

The HAADF-STEM images and element mappings of the CoO_x/Nb₂O₅-S catalyst before and after the stability test had also been investigated (Fig. 5). Through Fig. 5a and e, it can be seen that there was sintering after the stability test, where the boundaries between the particles were blurred, and the gaps between the particles were reduced. The EDS mapping results indicated that the distribution of elements Nb, Co, and O was relatively even before and after the stability test, without any significant migration or clustering.

The nitrogen adsorption–desorption isotherm was used to further characterize the porosity properties of Nb₂O₅ supports and CoO_x/Nb₂O₅ catalysts. As seen from Fig. 6, all the prepared Nb₂O₅ supports presented type IV isotherms, in which Nb₂O₅-R and Nb₂O₅-S presented typical H3 hysteresis loops at a relative pressure (p/p₀) of 0.7–0.95, indicating their irregular mesoporous structures. Additionally, Fig. 6b proved its board pore size distribution, mainly concentrated at 2–20 nm and 10–30 nm, respectively. Differently, Nb₂O₅-G possessed a H1 hysteresis loop in the range of 0.5 < p/p₀ < 0.9 with an obviously saturated adsorption platform, which was attributed to its mesopores with uniform pore size distribution of 2–10 nm. Moreover, the nitrogen adsorption–desorption isotherms of CoO_x-loaded catalysts were consistent with those of the supports, which indicated that the loading of CoO_x would not change the pore structure of the supports. Furthermore, the specific surface area and pore volume of Nb₂O₅-R and Nb₂O₅-G decreased obviously after CoO_x loading (Table 3), which indicated that CoO_x was highly dispersed on the supports.

---

3.3. Surface chemical properties analysis

Fig. 7 shows the XPS spectra of Co 2p, O 1s and Nb 3d of CoO_x/Nb₂O₅ catalysts. In Fig. 7a, the Co 2p spectra over CoO_x/Nb₂O₅-R, CoO_x/Nb₂O₅-G, and CoO_x/Nb₂O₅-S is composed two counterparts, Co 2p_3/2 and Co 2p_1/2, at 780.0 ± 0.2 and 795.2 ± 0.2 eV, respectively.^29–31 The spin–orbit splitting energy (ΔE) was 15.2 eV. Generally, the ΔE for Co²⁺ is 16.0 eV, while that of Co³⁺ is 15.0 eV. Therefore, this indicated that CoO_x existed in the mixture of Co³⁺ and Co²⁺. At the same time, Co 2p_3/2 were divided into two contributions (779.7 ± 0.1 eV, 781.1 ± 0.1 eV), typically attributed to Co³⁺, Co²⁺, respectively, and a shake-up satellite peak at 786.6 ± 0.2 eV.^32,33 Furthermore, the corresponding Co³⁺/Co²⁺ was calculated according to the fitted peak area, in which the order of Co³⁺/Co²⁺ was CoO_x/Nb₂O₅-S > CoO_x/Nb₂O₅-R > CoO_x/Nb₂O₅-G (Table 4). Notably, this order was consistent with the activity sequence of toluene oxidation. Usually, Co³⁺ species acted as active sites for toluene oxidation and can participate actively in the catalytic process.^34,35

---

The O 1s spectrum of CoO_x/Nb₂O₅ series catalysts was deconvolved (Fig. 7b), and the peaks with binding energies of 529.8 ± 0.1 eV, 531.0 ± 0.1 eV and 533 eV were attributed to surface lattice oxygen (O_latt), surface adsorbed oxygen (O_ads) and chemisorbed oxygen (O_w), respectively.^36,37 According to the ratio of each peak area, the proportion of O_ads/O (O_latt + O_ads + O_w) in each catalyst was obtained (Table 4), and the order was CoO_x/Nb₂O₅-S > CoO_x/Nb₂O₅-G > CoO_x/Nb₂O₅-R. Usually, higher O_ads/O molar ratios indicated more oxygen vacancies, which can promote the transformation of gaseous oxygen to surface-adsorbed oxygen.^38,39

In Fig. 7c, the Nb 3d spectrum consisted of two counterparts at 207.0 eV and 209.7 eV, corresponding to 3d_5/2 and 3d_3/2 of Nb⁵⁺ in pure Nb₂O₅, respectively, with a spin–orbit splitting energy (ΔE) of 2.7 eV. It was worth noting that the BE values of Nb 3d shifted to higher values after CoO_x loading, which suggested a strong interaction between the Nb and Co species. Furthermore, the ΔE for CoO_x/Nb₂O₅ catalysts remained at 2.7 eV, which indicated that the valence of Nb⁵⁺ remained the same after CoO_x loading.

H₂-TPR measurements can characterize the adsorption and dissociation of hydrogen molecules by chemical combination with active oxygen species, thereby providing information on the availability of surface-active oxygen and the abundance of oxygen vacancies on metals. CoO_x/Nb₂O₅-R, CoO_x/Nb₂O₅-G, and CoO_x/Nb₂O₅-S presented typically two reduction regions. Usually, the reduction region at low temperature was attributed to the reduction of Co³⁺ to Co²⁺, while the reduction peak at a higher temperature was attributed to the reduction of Co²⁺ to metallic Co. As seen from Fig. 8, the maximum temperature of the reduction peak in the high-temperature region of the CoO_x/Nb₂O₅ catalysts followed the order CoO_x/Nb₂O₅-R > CoO_x/Nb₂O₅-G > CoO_x/Nb₂O₅-S. Xiao et al.⁴⁰ attributed the reduction peak at the reduction temperature of 450–700 °C to the cobalt species, which had strong interaction with the carrier Nb₂O₅ in the CoO_x/Nb₂O₅ catalyst. Mejía et al.²⁸ prepared Nb₂O₅-supported CoO_x catalyst from niobic acid with a reduction peak greater than 450 °C, which was attributed to the strong interaction between CoO_x and niobic acid, resulting in the formation of Co–Nb oxide species without Co₃O₄ during impregnation and calcination. Thus, the high-temperature region at 380 to 580 °C of the CoO_x/Nb₂O₅ catalysts may also be caused by the strong interaction between Nb₂O₅ and CoO_x. This was also confirmed by the XPS results. Moreover, the CoO_x/Nb₂O₅-S catalyst showed a weak reduction peak at 154 °C, which was attributed to the reduction of surface-active oxygen species.⁴¹ Moreover, the reduction peak at 700–800 °C was assigned to the reduction of Nb₂O₅. Accordingly, the CoO_x/Nb₂O₅-S catalyst demonstrated superior low-temperature reducibility and oxygen mobility, which were consistent with its best catalytic activity for oxidizing toluene.

Generally, the acidic sites on the surface of the catalyst can promote the absorption of toluene and the cleavage of the C–H bond. NH₃-TPD was conducted to study the surface acidity of the CoO_x/Nb₂O₅ catalysts (Fig. 8b). Only weak desorption peaks of NH₃ were detected in a wide temperature range (100–300 °C) over the CoO_x/Nb₂O₅ catalysts, which were attributed to the weak/moderate acidic sites. Additionally, the NH₃ desorption peak on the CoO_x/Nb₂O₅-S catalyst was the weakest among all the CoO_x/Nb₂O₅ catalysts. Therefore, the acidity in the current catalyst system might not have a substantial influence on toluene oxidation, which was also confirmed by the other researchers.^41,42

3.4. DFT calculation

The influence of the morphology effect was further verified by DFT calculation. Two-layer slab model surfaces of Nb₂O₅ (001) and (180) substitution was built to calculate the adsorption energies and Gibbs free energies. Two p (3 × 3) unit cell expansions were used to model the surface of Nb₂O₅ (001) and (180). By observing the difference charge density plot (Fig. 9), the O in CoO_x interacted with Nb after CoO_x loading, which resulted in the transformation of electrons from the Nb surface to CoO_x. Accordingly, the rich electron CoO_x was beneficial for accelerating the catalytic reaction.

Furthermore, the density of states (DOS) of CoO_x/Nb₂O₅-S (180) and CoO_x/Nb₂O₅-R (001) were analysed (Fig. 10a and b); it was found that CoO_x/Nb₂O₅-S (180) had a richer state density near the Fermi level, which was favourable for electronic transmission.^43,44 The project density of states (PDOS) of Co in CoO_x/Nb₂O₅-S (180) presented a lower d-band center position (−0.79 eV) (Fig. 10c), confirming that the CoO_x in CoO_x/Nb₂O₅-S (180) had more electrons and better catalytic performance. The Bader charge analysis quantified the average number of valence electrons on the CoO_x/Nb₂O₅-S (180) and CoO_x/Nb₂O₅-R (001) surfaces (Table S1†). It revealed that there were more valence electrons on the CoO_x/Nb₂O₅-S (180) surface, with an average of 7.823, compared to the CoO_x/Nb₂O₅-R (001) surface (7.667). This indicated that more electrons could be transferred from the Nb₂O₅-S (180) surface to the CoO_x phase, leading to a stronger interaction between them.

The adsorption state of toluene on the surface was also calculated (Fig. 11); the toluene adsorption energy for CoO_x/Nb₂O₅-S (180) was −0.65 eV, which was lower than that of CoO_x/Nb₂O₅-R (001) (−0.23 eV), implying that toluene was easily adsorbed on CoO_x/Nb₂O₅-S (180). The shortened distance between toluene and CoO_x in CoO_x/Nb₂O₅-S (180) also corroborated stronger adsorption (Fig. S1, Table S2†). After adsorption on the CoO_x/Nb₂O₅-S (180) surface, toluene presented greater stretching and twisting, indicating its easier activation and decomposition.

Additionally, the adsorption and activation characteristics of O₂ were also proved by DFT calculation. The oxygen vacancy formation energy (E_v) was calculated, and the E_v values of CoO_x/Nb₂O₅-S (180) and CoO_x/Nb₂O₅-R (001) were 0.54 eV and 0.82 eV (Fig. 12), respectively, indicating that the structure of CoO_x/Nb₂O₅-S (180) was conductive to surface oxygen migration.⁴⁵ Meanwhile, owing to the electron-rich characteristic of the CoO_x surface on CoO_x/Nb₂O₅-S (180), CoO_x/Nb₂O₅-S (180) exhibited lower oxygen adsorption energy (−0.13 eV) and oxygen atom dissociation energy (0.78 eV). The enhanced oxygen adsorption and dissociation capacity were beneficial for the catalytic oxidation of toluene.

4. Conclusion

A series of CoO_x supported on rod-like, grid-like and spherical Nb₂O₅ catalysts were synthesized and evaluated for toluene oxidation. It was found that CoO_x/Nb₂O₅-S displayed the best catalytic activity (T₉₀ = 339 °C) and stability. The characterization and DFT calculation results indicated that 180 facets exposed on Nb₂O₅-S were favourable for toluene adsorption, and the higher content of Co³⁺ and surface adsorbed oxygen promoted the adsorption and activation of oxygen, resulting in superior toluene oxidation performance.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article and its ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The work described above was supported by the Shandong Provincial Natural Science Foundation (ZR2023MB032, ZR2023QB198), Qingdao New Energy Shandong Laboratory Open Project (QNESL OP202310) and Qingdao Postdoctoral Program Funding (QDBSH20220202039, QDBSH20230102032).

References

1. X. Zhou, X. Zhou, C. Wang and H. Zhou, Chemosphere, 2023, 313, 137489 CrossRef CAS PubMed .
2. J. Ma and L. Li, J. Environ. Sci., 2024, 135, 570–584 CrossRef CAS PubMed .
3. J. Liu, S. Wang, J. Xuan, B. Shan, H. Luo, L. Deng, P. Yang and C. Qi, Tungsten, 2022, 4, 38–51 CrossRef .
4. B. Gao, X. Feng, Y. Zhang, Z. Zhou, J. Wei, R. Qiao, F. Bi, N. Liu and X. Zhang, Chem. Eng. J., 2024, 484, 149604 CrossRef CAS .
5. F. Bi, J. Wei, B. Gao, N. Liu, J. Xu, B. Liu, Y. Huang and X. Zhang, ACS EST Eng., 2024, 4, 1346–1355 CrossRef CAS .
6. Y. Su, K. Fu, C. Pang, Y. Zheng, C. Song, N. Ji, D. Ma, X. Lu, C. Liu, R. Han and Q. Liu, Environ. Sci. Technol., 2022, 56, 9854–9871 CrossRef CAS PubMed .
7. C. He, J. Cheng, X. Zhang, M. Douthwaite, S. Pattisson and Z. Hao, Chem. Rev., 2019, 119, 4471–4568 CrossRef CAS PubMed .
8. C. Feng, C. Chen, J. Wang, S. Li, F. Liu, Y. Pan, Y. Lu, Y. Liu, X. Li, Y. Liu, R. Zhang and Z. Wang, Surf. Interfaces, 2024, 46, 104011 CrossRef CAS .
9. S. Ma, Z. Zhou, Y. Zhang, R. Rao, H. Han, J. Liang, Z. Zhao, F. Bi, N. Liu and X. Zhang, Sep. Purif. Technol., 2024, 339, 126636 CrossRef CAS .
10. T. Xiang, D. Dai, X. Li, D. Liu, C. Feng, P. Dai, L. Li, X. Gu and Y. Liu, Appl. Catal., B, 2024, 348, 123841 CrossRef CAS .
11. Q. Ren, X. Zhao, J. Zhong, J. Zhang, J. Tian, D. Yan, P. Liu, M. Fu, L. Chen, J. Wu and D. Ye, J. Catal., 2023, 418, 130–140 CrossRef CAS .
12. Y. Jian, M. Tian, C. He, J. Xiong, Z. Jiang, H. Jin, L. Zheng, R. Albiali and J. Shi, Appl. Catal., B, 2021, 283, 119657 CrossRef CAS .
13. Z. Lu, L. Guo, Q. Shen, F. Bi, C. Li and X. Zhang, Sep. Purif. Technol., 2024, 340, 126772 CrossRef CAS .
14. M. Xiao, T. Zhao, Y. Li, B. Zhu, T. Yu, W. Liu, M. Zhao and B. Cui, ChemCatChem, 2024, e202301524 CrossRef CAS .
15. S. Li, L. Xu, C. Feng, Q. Li, J. Jiang, D. Liang, P. Zhang, C. Zhang, X. Li and Z. Wang, J. Mol. Liq., 2023, 389, 122835 CrossRef CAS .
16. Y. Xin, L. Dong, Y. Guo, X. Liu, Y. Hu and Y. Wang, J. Catal., 2019, 375, 202–212 CrossRef CAS .
17. T. Murayama, J. Chen, J. Hirata, K. Matsumoto and W. Ueda, Catal. Sci. Technol., 2014, 4, 4250 RSC .
18. K. Su, H. Liu, Z. Gao, P. Fornasiero and F. Wang, Adv. Sci., 2021, 8, 2003156 CrossRef CAS PubMed .
19. B. Kommula, R. Prabhu, H. Kopperi, V. Bhat, G. Hegde and N. John, Part. Part. Syst. Charact., 2023, 2300118 Search PubMed .
20. W. Guo, T. Tong, X. Liu, Y. Guo and Y. Wang, ChemCatChem, 2019, 11, 4130–4138 CrossRef CAS .
21. J. Zhao, Z. Tang, F. Dong and J. Zhang, Mol. Catal., 2019, 463, 77–86 CrossRef CAS .
22. T. Xue, R. Li, Y. Gao and Q. Wang, Chem. Eng. J., 2020, 384, 118749 CrossRef .
23. C. Gennequin, S. Siffert, R. Cousin and A. Aboukais, Top. Catal., 2009, 52, 482–491 CrossRef CAS .
24. M. Castano, R. Molina and S. Moreno, Appl. Catal., A, 2015, 492, 48–59 CrossRef CAS .
25. F. Bi, J. Wei, S. Ma, Q. Zhao, J. Zhang, R. Qiao, J. Xu, B. Liu, Y. Huang and X. Zhang, J. Colloid Interface Sci., 2024, 665, 898–910 CrossRef CAS PubMed .
26. L. Zhao, Z. Zhang, Y. Li, X. Leng, T. Zhang, F. Yuan, X. Niu and Y. Zhu, Appl. Catal., B, 2019, 245, 502–512 CrossRef CAS .
27. E. Heracleous and A. Lemonidou, J. Catal., 2006, 237, 162–174 CrossRef CAS .
28. C. Mejía, J. Otter, J. Weber and K. Jong, Appl. Catal., A, 2017, 548, 143–149 CrossRef .
29. T. Li, J. Zhao, Y. Quan, D. Luo, C. Miao and J. Ren, Environ. Sci. Pollut. Res., 2021, 28, 37592–37602 CrossRef CAS PubMed .
30. J. Yao, F. Dong, H. Feng and Z. Tang, ACS Appl. Nano Mater., 2021, 4, 9322–9332 CrossRef CAS .
31. A. Rastegarpanah, J. Deng, Y. Liu, L. Jing, W. Pei, J. Wang and H. Dai, J. Environ. Sci., 2025, 147, 617–629 CrossRef PubMed .
32. J. Yao, F. Dong, H. Feng and Z. Tang, ACS Appl. Nano Mater., 2021, 4, 9322–9332 CrossRef CAS .
33. X. Li, X. Liu, C. Liu, J. Zeng and X. Qi, Tungsten, 2023, 5(01), 100–108 CrossRef .
34. Y. Ma, L. Wang, J. Ma, H. Wang, C. Zhang, H. Deng and H. He, ACS Catal., 2021, 11, 6614–6625 CrossRef CAS .
35. H. Yi, Z. Yang, X. Tang, S. Zhao, F. Gao, J. Wang, Y. Huang, Y. Ma, C. Chu, Q. Li and J. Xu, Chem. Eng. J., 2018, 333, 554–563 CrossRef CAS .
36. C. Huang, Q. Ji, H. Zhang, Y. Wang, S. Wang, X. Liu, Y. Guo and C. Zhang, J. Colloid Interface Sci., 2022, 606, 654–665 CrossRef CAS .
37. Y. Zheng, Y. Su, C. Pang, L. Yang, C. Song, N. Ji, D. Ma, X. Lu, R. Han and Q. Liu, Environ. Sci. Technol., 2022, 56, 1905–1916 CrossRef CAS PubMed .
38. F. Bi, X. Feng, Z. Zhou, Y. Zhang, J. Wei, L. Yuan, B. Liu, Y. Huang and X. Zhang, Chem. Eng. J., 2024, 485, 149776 CrossRef CAS .
39. J. Zhang, L. Wang, X. Hu, Q. Shao, X. Xu and C. Long, J. Cleaner Prod., 2021, 318, 128564 CrossRef CAS .
40. G. Xiao, Y. Han, F. Jing, M. Chen, S. Chen, F. Zhao, S. Xiao, Y. Zhang, J. Li and J. Hong, Ind. Eng. Chem. Res., 2020, 59, 17315–17327 CrossRef CAS .
41. Y. Wang, C. Wang, K. Zeng, S. Wang, H. Zhang, X. Li, Z. Wang and C. Zhang, Appl. Surf. Sci., 2022, 576, 151797 CrossRef CAS .
42. Z. Huang, J. Ding, X. Yang, H. Liu, P. Song, Y. Guo, Y. Guo, L. Wang and W. Zhan, Environ. Sci. Technol., 2022, 56, 17278–17287 CrossRef CAS PubMed .
43. Y. Xie, S. Wang, P. Jin, H. Mao, X. Zheng, Z. Wang and C. Zhang, Surf. Interfaces, 2024, 45, 103885 CrossRef CAS .
44. C. Feng, Q. Gao, G. Xiong, Y. Chen, Y. Pan, Z. Fei, Y. Li, Y. Lu, C. Liu and Y. Liu, Appl. Catal., B, 2022, 304, 121005 CrossRef CAS .
45. C. Feng, C. Chen, G. Xiong, D. Yang, Z. Wang, Y. Pan, Z. Fei, Y. Lu, Y. Liu, R. Zhang and X. Li, Appl. Catal., B, 2023, 328, 122528 CrossRef CAS .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4cy00596a

This journal is © The Royal Society of Chemistry 2024