Catalytic Combustion of Propane over Ce-Doped Lanthanum Borate Loaded with Various 3d Transition Metals

Abstract

Ce-doped LaBO₃ (Ce_0.05La_0.95BO₃) and a corresponding incorporation with 3d transition metals (TMs) were prepared and evaluated for eliminating propane. Our results showed the catalytic activity toward propane combustion has a close relationship with the loaded TMs, which promoted oxygen vacancies density and further enhanced the reduction and acidity of this material. This eventually led to 90% propane conversion at 718 K for a Cu-loaded Ce_0.05La_0.95BO₃ catalyst. During 10 h of catalytic propane oxidation, the propane-elimination rate was maintained very well, with no degradation of the catalyst.

1. Introduction

Volatile organic compounds (VOCs) are a class of chemicals that are confirmed to have short- and long-term adverse environmental and health effects [1,2,3,4]. Propane is a typical hydrocarbon in VOCs and is difficult to remove due to the high C–H bond energy [5,6,7,8,9]. There are many methods for VOC treatments, such as thermal incineration [10], solvent absorption [11], catalytic upgrading [12,13], and condensation [14]. Compared to these methods above, catalytic combustion is a promising technology, because it provides the potential to totally destroy pollutants to CO₂ and H₂O [15,16,17]. Thus, the development of novel catalysts with high-performance catalytic combustion is of technological importance in eco-chemical engineering and heterogeneous catalysis.

So far, various catalyst systems have been introduced for the total oxidation of propane, and a high efficiency in propane removal can be achieved with these heterogeneous catalysts. They can be divided into noble metal and non-noble metal catalysts. Although catalysts based on noble metals always exhibit a high efficiency in propane removal, limited natural resources and a low poisoning resistance hinder their further development [18,19]. Transition metals (TM) were applied in the solid catalysts to facilitate propane combustion because of their high recycle-stability and low cost [20,21,22]. However, they have the disadvantage of insufficient catalytic efficiency. Adopting appropriate support is an alternative method to solve the above problems, which can provide a high dispersion of the active centers, leading to a satisfying activity [23,24].

In recent years, rare-earth elements (REEs) have attracted research interest due to their lanthanide contraction and partially fulled 4f orbitals [25,26]. These characteristics enable high performance in the catalytic combustion of propane when applying rare-earth elements in the catalysts, such as perovskite materials, which are famous for their vast structural tunability and satisfying thermal stability. Moreover, as supports, perovskite materials are beneficial for the dispersion of the active components, which facilitates the combustion process [27,28,29]. In particular, La-based perovskite catalysts (ABO₃, lanthanum located at the A-position) have been widely used for the catalytic combustion of VOCs [30]. For example, Chuanhui Zhang et al. found that Ni-doped LaMnO₃ showed a satisfying redox ability and abundant oxygen species, thus leading to high activity toward propane conversion (the temperature for 90% conversion of propane, T90 = 587 K) [31]. Lu Dai et al. reported that the increased content of Co³⁺ on the surface of particle-aggregated LaCoO₃ perovskite could promote active oxygenated species on the surface of materials, leading to satisfying catalytic efficiency toward CO and C₃H₈ combustion [32]. Wenjun Zhu and colleagues also reported similar reaction results: low amounts of cobalt substitution enhanced the specific surface area of LaMnO₃ (LaCo_0.2Mn_0.8O₃) and exhibited a lower T90 at 628 K than that of LaMnO₃ (653 K) [33]. Thus, the selection of B-site elements, such as different transition metals (TM), can affect the catalytic activity of La-based perovskite. In addition, TMs that involve La-based perovskite possess satisfying redox properties due to their variable valences, which are favorable for catalytic combustion [33,34].

Recently, boron has been attractive due to its electron deficiency and Lewis acidity, which facilitate various reactions such as propane dehydrogenation and the ring-opening processes of epoxides [35,36,37,38,39]. LaBO₃, with boron as the B-site, belongs to the ABO₃-type perovskite (Figure 1), which has been successfully used as a lubricant additive and luminescent material [40,41]. In the monoclinic structure of this rare-earth borate, the nine-coordinated La alternately arranged with [BO₃] via oxygen atoms, forming internal lattice interstices. It is favorable for the uniform dispersion of metals. Combined with 3d TMs’ loading and further modified with Ce atoms (which are always involved to boost the ability to activate reactant molecules and enhance the stability of solid catalysts), LaBO₃ should show satisfactory catalytic efficiency toward propane combustion [42]. Herein, TM-modified Ce_0.05La_0.95BO₃ (TMs/Ce_0.05La_0.95BO₃) catalysts were synthesized using a citrate method. The catalytic activities were evaluated with propane combustion. Multiple characterization techniques, such as XRD, UV–Vis DRS, XPS, SEM, nonaqueous potentiometric titrations, NH₃-TPD, and H₂-TPR, were employed to analyze their physical and chemical properties. The effect of the reductive properties and acidic nature of TMs/Ce_0.05La_0.95BO₃ on the total oxidation of propane were systematically studied.

2. Results and Discussion

2.1. Catalyst Characterization

These materials, i.e., LaBO₃, Ce_0.05La_0.95BO₃, and TMs/Ce_0.05La_0.95BO₃, display a similar plate-like morphology to the one characterized by SEM (Figure 2c–f and Figure S1). XRD patterns of these synthesized materials are shown in Figure 2a. They all show diffraction peaks at 22.8°, 28.7°, 29.5°, 35.3°, 43.2°, 46.4°, and 47.3°, which are characteristic Bragg diffractions for LaBO₃ material in the monoclinic space group (PDF No. 73–1149) [43]. In addition, the diffraction peaks of Ce_0.05La_0.95BO₃ and TMs/Ce_0.05La_0.95BO₃ agreed well with the pure LaBO₃, and there was the absence of any extra peaks of impurities. This indicates that the bulk structure of LaBO₃ shows no obvious changes after these modifications. In addition, the crystal phases belonging to these TMs or their oxides cannot be found in the spectra. Systematic IR tests were also performed to understand the physico-chemical properties of LaBO₃, Ce_0.05La_0.95BO₃, and TMs/Ce_0.05La_0.95BO₃. As shown in Figure 3, the absorption peaks at 1232 cm⁻¹ and 949 cm⁻¹ correspond to the antisymmetric-vibrational and symmetric-vibrational absorptions of the B–O bond in the [BO₃] group. The strong absorption peak around 744 cm⁻¹ corresponds to the out-of-plane bending vibration of the B–O bond, and the absorption peaks at 630 cm⁻¹ and 586 cm⁻¹ are associated with the in-plane bending vibration of this bond. The absorption peaks near 1637 cm⁻¹ and 2700–3700 cm⁻¹ are related to the stretching vibrations of the [–OH] group. No characteristic peaks corresponding to the [BO₄] group are observed at 1018 cm⁻¹ and 853 cm⁻¹, indicating that the boron atoms in the skeleton are exclusively in a planar triangular [BO₃]. Thus, the data above suggest that Ce and TMs loading did not affect the bulk structure of LaBO₃ nor the uniform dispersion for these TMs over Ce_0.05La_0.95BO₃ [44].

UV–Vis diffuse reflectance spectroscopy was employed to obtain information about the coordination and oxidation states of the metal ions on the material surface. Figure 2b shows the UV–Vis diffuse reflectance spectra of LaBO₃, Ce_0.05La_0.95BO₃, and TMs/Ce_0.05La_0.95BO₃. LaBO₃ shows no absorption band in the 200–800 nm range. In contrast, three features around 225, 275, and 350 nm were found in the spectrum of Ce_0.05La_0.95BO₃. These peaks correspond to three sub-energy levels split from the 5d energy level of cuprous species due to its orthogonal crystal environment [45]. This is reported to be able to change the surface composition of oxygenated species in materials due to the charge compensation mechanism, thus regulating catalyst activity [46]. For TMs/Ce_0.05La_0.95BO₃ materials, there are merely no new peaks detected in the UV–Vis spectra, suggesting that the surface coordination and oxidation states of the metal ions in the host matrix are not affected by the incorporation of TMs.

The above analysis indicates Ce and TMs have been effectively doped into the LaBO₃ host lattice. The loaded TMs may occupy the vacancy sites between La (Ce) ions and the [BO₃] groups, which is suggested to facilitate the formation of the TM–REE solid solution, such as Cu–Ce, thus improving the reducibility of the catalyst [47]. This will be verified in the following parts.

2.2. Variations over Surface Chemistry

To understand the variations over the surface chemistry of these TM-loaded rare metal borates, Cu/Ce_0.05La_0.95BO₃, Ni/Ce_0.05La_0.95BO₃, Co/Ce_0.05La_0.95BO₃, and V/Ce_0.05La_0.95BO₃ were selected for systematic XPS measurements. Figure S2 exhibits the XPS survey spectrum of these materials, and the peaks of O, B, Ce, La, and the incorporated TM elements over the surfaces of these four materials can be clearly observed. The existence of the C 1s peak, 284.8 eV, could be attributed to the widespread presence of carbon in the environment. Multiplex high-resolution scans over the O 1s, B 1s, Ce 3d, and La 3d spectral regions are shown in Figure 4a–d, respectively. Figure 4a demonstrates that three different oxygen species presented in the materials, i.e., lattice oxygen (O_latt), chemisorbed oxygen (O_ads), and surface hydroxyl oxygen (O_hyd), were located around 529.9 eV, 531.3 eV, and 532.9 eV, respectively [48]. In the B 1s spectra (Figure 4b), the bonding energy (BE) at 191.4 eV agreed well with that of [BO₃], while the other, around 196.3 eV, can be attributed to B-O-La in the LaBO₃ lattice [49,50]. The peaks belonging to the boron species of Cu/Ce_0.05La_0.95BO₃ shift to the lower BE (191.3 eV and 196.2 eV), in comparison to those of the other three materials. This suggests that the electronic structure of boron was more sensitive to Cu incorporation. A similar phenomenon was reported previously. For instance, Yao and coworkers showed that a CuO species wrapped over CeO₂-TiO₂ nanotubes can improve the electron structure of the materials and the equilibrium of a Ce⁴⁺/Ce³⁺ ion pair in structure, thus finally achieving 90% conversion of butane at 573 K [51]. Peiwe Wu and coworkers also reveal that Cu nanoparticles loaded on g-BN material could significantly improve the charge transfer between copper and boron species, leading to a high activity toward the oxidation of aromatic sulfur compounds [52]. This charge transfer was suggested to relate with the lowest resistivity of copper atoms compared with other TMs, which ensures a well-conducted electronic transfer between Cu and its neighboring atoms [53]. In Figure 4d, the peaks are observed at 852.5 eV and 835.8 eV, which can be attributed to two spin orbits, La 3d3/2 and La 3d5/2, respectively. The other peaks, located at 856.1 eV and 839.3 eV, are La 3d satellite peaks, indicating the presence of La³⁺ [54]. As for the Ce species in these four materials, Ce⁴⁺ and Ce³⁺ were found to coexist in these materials, as shown in the high-resolution XPS scan of Ce 3d (Figure 4c). Thus, introducing this atom can result in a Ce³⁺/Ce⁴⁺ redox couple, which was suggested to contribute to the evolution of lattice oxygen to electrophilic oxygen species [55] and lead to a high efficiency of propane elimination. Similarly, as shown in Figure 5a–d, Cu, Ni, Co, and V also exhibit two different oxidation states on the surface of Cu/Ce_0.05La_0.95BO₃, Ni/Ce_0.05La_0.95BO₃, Co/Ce_0.05La_0.95BO₃, and V/Ce_0.05La_0.95BO₃, respectively. These analyses above suggest a strong electronic interaction over these elements on the surface of materials.

Table 1. The surface composition of surface elements in Cu/Ce_0.05La_0.95BO₃, Ni/Ce_0.05La_0.95BO₃, Co/Ce_0.05La_0.95BO₃, and V/Ce_0.05La_0.95BO₃.

Sample	Atomic Ratios (%)
	Olatt/(Olatt + Oads + Ohyd)	Oads/(Olatt + Oads + Ohyd)	Ohyd/(Olatt + Oads + Ohyd)	Ce3+/(Ce3+ + Ce4+)	Ce4+/(Ce3+ + Ce4+)
Cu/Ce0.05La0.95BO3	12.9	74.1	13.0	52.6	47.4
Ni/Ce0.05La0.95BO3	16.3	71.8	11.9	49.7	50.3
Co/Ce0.05La0.95BO3	16.7	65.7	17.6	47.0	53.0
V/Ce0.05La0.95BO3	19.2	66.0	14.8	45.9	54.1

shows that the relative element composition ratio was determined by multiplex high-resolution scans over the O 1s and Ce 3d spectral regions. Our results show that the contents of these two atoms in those four materials were affected by the loaded TM. Cu/Ce_0.05La_0.95BO₃ shows a higher ratio of Ce³⁺/(Ce³⁺ + Ce⁴⁺), 52.6%, compared with the ratios of the Ni-, Co-, and V-loaded samples, i.e., 49.7%, 47.0%, and 45.9%, respectively. This ratio might reflect oxygen vacancy density on the material surface [56,57]. In addition, the content of the O_ads, an indicator of oxygen vacancy density, in these samples also displays similar variations [58]. The O_ads content of the Cu/Ce_0.05La_0.95BO₃ catalyst (74.1%) is higher than in the other samples. The high density of oxygen vacancy is claimed to be conducive to molecular oxygen adsorption and activation, thus accelerating the charge transfer and facilitating propane combustion [59]. In addition, the terminal unsaturated metal cations on the surface of the material can function as acidic sites to facilitate the reaction process, and these cations are supposed to be formed as the rupture of the metal–oxygen bond over the material surface [60]. In other words, the high density of oxygen vacancy can dramatically promote the breaking of the metal–oxygen bond, thus enhancing the surface acidity of these TMs/Ce_0.05La_0.95BO₃ samples. Thus, this rare-earth borate modified with copper should show a higher activity toward propane total oxidation compared with other synthesized materials, which will be tested later.

A material’s acidity is claimed to contribute to its catalysis activity for propane total oxidation, due to the promotion toward the rate-controlling step, i.e., the C–H bond activation and cleavage of propane [60,61,62]. Thus, the acidic properties of these samples were investigated using nonaqueous potentiometric titrations with n-butylamine in acetonitrile. Quantitative analysis (

Table 2. Potentiometric titration employing n-butylamine for LaBO₃, Ce_0.05La_0.95BO₃, and TMs/Ce_0.05La_0.95BO₃.

Sample	Acid Amount (mmol·g−1)
V/Ce0.05La0.95BO3	0.040
Cr/Ce0.05La0.95BO3	0.097
Mn/Ce0.05La0.95BO3	0.108
Fe/Ce0.05La0.95BO3	0.103
Co/Ce0.05La0.95BO3	0.063
Ni/Ce0.05La0.95BO3	0.213
Zn/Ce0.05La0.95BO3	0.088
Cu/Ce0.05La0.95BO3	0.241
Ce0.05La0.95BO3	0.089
LaBO3	0.037

, Table S1, and Figure S3) showed the numbers for the acid sites of the samples, ranked in decreasing order: Cu/Ce_0.05La_0.95BO₃ (0.241 mmol/g) > Ni/Ce_0.05La_0.95BO₃ (0.213 mmol/g) > Mn/Ce_0.05La_0.95BO₃ (0.108 mmol/g) > Fe/Ce_0.05La_0.95BO₃ (0.103 mmol/g) > Cr/Ce_0.05La_0.95BO₃ (0.097 mmol/g) > Ce_0.05La_0.95BO₃ (0.089 mmol/g) > Zn/Ce_0.05La_0.95BO₃ (0.088 mmol/g) > Co/Ce_0.05La_0.95BO₃ (0.063 mmol/g) > V/Ce_0.05La_0.95BO₃ (0.040 mmol/g) > LaBO₃ (0.037 mmol/g). These data suggest that Cu/Ce_0.05La_0.95BO₃ possesses a higher number of acid sites than the others, which means that it may exhibit a better activity for promoting propane combustion. Temperature-programmed desorption (TPD) experiments, with NH₃ as the probe molecular, were employed to further understand the acidic properties of these materials. In general, the temperature of NH₃ desorption was related to the acid strength [63]. It was reported that weak and moderate acidic sites, i.e., the temperature of NH₃ desorption below 773 K, were good for facilitating the propane total oxidation [60,61]. As shown in Figure 6, NH₃ desorption bands were found above 773 K for Cu/Ce_0.05La_0.95BO₃, V/Ce_0.05La_0.95BO₃, Co/Ce_0.05La_0.95BO₃, Ni/Ce_0.05La_0.95BO₃, Ce_0.05La_0.95BO₃, and LaBO₃, which suggested strong acidic centers in these materials [64]. Note that the temperature of NH₃ desorption for Cu/Ce_0.05La_0.95BO₃ was 790 K and lower than other catalysts, indicating its highest activity toward C–H bonds’ activation. These data were consistent with the hypothesis provided above, that the enhanced surface acidity of the TMs/Ce_0.05La_0.95BO₃ samples should be originated from the formation of oxygen vacancies.

The reducibility of Ni/Ce_0.05La_0.95BO₃, Co/Ce_0.05La_0.95BO₃, V/Ce_0.05La_0.95BO₃, and Cu/Ce_0.05La_0.95BO₃ was evaluated by H₂-TPR, and the relevant TPR profiles are shown in Figure 7. Our results show that these four materials have a similar reduction peak around 1030 K, which is attributed to the reduction in bulk lattice oxygen in LaBO₃ [65]. While the reduction behaviors of these materials vary with the different loaded TMs. Specifically, these samples exhibit another reduction peak at ~428 K, 588 K, 872 K, and 839 K, respectively, presenting the reduction behaviors in CuO_x, NiO_x, CoO_x, and VO_x on the surface of Ce_0.05La_0.95BO₃ [66,67,68,69]. The temperature of the CuO_x in Ce_0.05La_0.95BO₃ was much lower than that of the other three materials, suggesting a better oxygen mobility inside the material and the satisfying catalytic efficiency for propane total oxidation according to the Mars–van Krevelen mechanism [70].

2.3. Catalyst Performance

The catalytic performances for these TMs/Ce_0.05La_0.95BO₃ materials toward propane combustion are shown in Figure 8a,b and

Table 3. T10, T50, and T90 obtained with LaBO₃, Ce_0.05La_0.95BO₃, and TMs/Ce_0.05La_0.95BO₃ in propane combustion.

Catalysts	T10(K)	T50(K)	T90(K)
V/Ce0.05La0.95BO3	604	---	---
Cr/Ce0.05La0.95BO3	473	---	---
Mn/Ce0.05La0.95BO3	555	708	769
Fe/Ce0.05La0.95BO3	523	715	---
Co/Ce0.05La0.95BO3	475	700	---
Ni/Ce0.05La0.95BO3	589	702	765
Zn/Ce0.05La0.95BO3	456	764	---
Cu/Ce0.05La0.95BO3	473	641	718
Ce0.05La0.95BO3	515	---	---
LaBO3	691	---	---

. Our result shows that propane cannot be converted over LaBO₃ and Ce_0.05La_0.95BO₃ until the reaction temperature reaches 773 K. Meanwhile, the T10 of Ce_0.05La_0.95BO₃, 515 K, is remarkably lower than that of LaBO₃. In addition, propane conversion dropped when Ce-loading was less than or exceeded 5%, i.e., 2% and 10%. Thus, replacing 5% of La with Ce was demonstrated to be the most effective for facilitating propane combustion. In addition, the catalytic activity could not be further improved when V, Ni, Mn, and Fe were loaded on Ce_0.05La_0.95BO₃. In contrast, the promotion of propane elimination was observed over Co-, Cr-, Cu-, and Zn-modified Ce_0.05La_0.95BO₃. On the basis of T50 (the temperature for 50% conversion of propane), the order of the catalytic activity for the propane degradation is as follows: Cu/Ce_0.05La_0.95BO₃ (641 K) > Co/Ce_0.05La_0.95BO₃ (700 K) > Ni/Ce_0.05La_0.95BO₃ (702 K) > Mn/Ce_0.05La_0.95BO₃ (708 K) > Zn/Ce_0.05La_0.95BO₃ (764 K). As the reaction temperature increases, Cu/Ce_0.05La_0.95BO₃ shows the highest activity, giving a 90% conversion of propane at WHSV mL·g_cat⁻¹·h⁻¹ at about 718 K. In addition, our results showed that Cu/Ce_0.05La_0.95BO₃ exhibited the highest stability for the catalytic combustion of propane in comparison to Ce_0.05La_0.95BO₃, V/Ce_0.05La_0.95BO₃, and Co/Ce_0.05La_0.95BO₃ (see Figure S4 and

Table 4. The deactivation rates of LaBO₃, Ce_0.05La_0.95BO₃, V/Ce_0.05La_0.95BO₃, Co/Ce_0.05La_0.95BO₃, Ni/Ce_0.05La_0.95BO₃, and Cu/Ce_0.05La_0.95BO₃.

Sample	Deactivation Rate (h−1)
V/Ce0.05La0.95BO3	0.046
Co/Ce0.05La0.95BO3	0.052
Ni/Ce0.05La0.95BO3	0.013
Cu/Ce0.05La0.95BO3	0.013
Ce0.05La0.95BO3	0.021
LaBO3	0.611

). This indicates that Cu-loading has a strong interaction with Ce_0.05La_0.95BO₃, and maintains the catalysis activity toward propane combustion. The selectivity of CO and CO₂ from the propane combustion catalyzed by the Cu/Ce_0.05La_0.95BO₃ catalyst is presented in Figure S5. This shows that the CO₂ selectivity of this material was higher than 90%, compared with that of the Cu-based catalysts and La-based catalysts reported in recent years (

Table 5. The CO₂ selectivity of the propane combustion catalyzed by Cu/Ce_0.05La_0.95BO₃ and other reported catalysts.

Catalyst	T50(K)	Reaction Temperature (K)	Selectivity to CO2 (%)	Ref.
Cu/Ce0.05La0.95BO3	641	673	91	This work
NiCeOx (Ni/Ce = 1:1)	528	501	87.8	[71]
Pt/LaCoO3	599	723	97	[72]
CuAl2O4	796	--	--	[73]
CuFe2O4	698	--	--	[73]
LaCuMnO6-CIT	753	--	--	[74]

) [71,72,73,74], which suggested the Cu/Ce_0.05La_0.95BO₃ catalyst has a high potential for industrial application due to its enhanced activity, high selectivity for CO₂, and satisfying catalysis stability in propane total oxidation.

Metal borates are always synthesized at high-temperature regions (>600 K), so most of them are bulk materials, have a dense framework, and are rarely applied into heterogeneous catalysis, such as catalytic VOCs elimination. However, these materials can be potential candidates for catalyst carriers, due to their high thermal stability and strong interaction with metal species. For instance, R. Abbas-Ghaleb et al. prepared an Al₁₈B₄O₃₃ oxide supported with Pd, which showed the superior stability for catalytic methane combustion in a high-temperature region, i.e., 868 K to 1173 K, owing to the strong interaction between the PdO species and this aluminum borate [75]. Hinokuma and coworkers found that the high efficiency for NH₃ removal catalyzed by CuO_x-loaded Al₂₀B₄O₃₆ at a low temperature was closely related to the highly dispersed cuprous on Al₂₀B₄O₃₆ [76].

For the kinetics study, our results show that the propane total oxidation catalyzed by these materials obeys first-order reaction kinetics, which was consistent with previous studies [77]. With the same reaction conditions, 1000 ppm propane and WHSV = 20,000 mL·g_cat⁻¹·h⁻¹, the apparent activation energies (E_a) for Cu/Ce_0.05La_0.95BO₃, Ni/Ce_0.05La_0.95BO₃, Co/Ce_0.05La_0.95BO₃, and V/Ce_0.05La_0.95BO₃ are 35.20, 44.68, 100.51, and 106.29 kJ/mol, respectively. Cu/Ce_0.05La_0.95BO₃ possesses the lowest E_a value and the best catalytic activity, suggesting that propane is easier decomposed over this sample. This is consistent with catalytic evaluation results.

Overall, Ce and these 3d TMs were suggested to exist in the lattice of LaBO₃. The redox pair of Ce³⁺ and Ce⁴⁺ was involved if Ce atoms were loaded into LaBO₃ material, which was suggested to contribute to the transition from lattice oxygen to electrophilic oxygen species. This allows the T10 of Ce_0.05La_0.95BO₃ to reduce to 515 K, which is much lower than that of LaBO₃ (691 K). When 3d TMs were further introduced, the density of oxygen vacancy over these materials was improved. This led to systematic variations over the surface acidity and better reductive properties, which were characterized by n-butylamine potentiometric titration and H₂-TPR experiments. These changes can facilitate the C–H bond activation of propane and promote oxygen mobility inside the catalyst, which eventually leads to a 90% propane conversion at 718 K for the Cu-loaded Ce_0.05La_0.95BO₃ catalyst.

3. Materials and Methods

3.1. Catalyst Synthesis

3.1.1. Preparation of Ce_0.05La_0.95BO₃ Support

Ce_xLa_1−xBO₃ was prepared by the citrate method. Taking Ce_0.05La_0.95BO₃ as an example, 1.3 mmol La(NO₃)₃·6H₂O and 0.1 mmol Ce(NO₃)₃·6H₂O were dissolved in DI water, followed by adding appropriate volumes of concentrated nitric acid. Then, 2.2 mmol of H₃BO₃ and 1.9 mmol of citric acid were added to the solution and stirred at 353 K to remove the excess water. The resulting gel precursor was further dried at 453 K and then preheated at 773 K for 10 h with 1 K/min. This mixture was finally calculated at 923 K to obtain the target material.

3.1.2. Preparation of Ce_0.05La_0.95BO₃ Supported with 1 wt% Metal Oxide

This step was carried out with a wet impregnation method [78]. It was reported that TMs with low concentration (<1.5 wt%) could ensure the strong interaction between TMs and the support and further enable the atomic dispersions of metals [79]. This has shown high activity toward various reaction processes, such as CO oxidation, cyclohexane oxidation, and NH₃–SCR reaction [79,80,81]. Thus, 1 wt% loading for these TMs, i.e., Cu, Ni, Co, V, Cr, Mn, Fe, and Zn, was selected to prepare catalysts to enhance the dispersion over Ce_0.05La_0.95BO₃. Taking 1wt% copper-loaded Ce_0.05La_0.95BO₃ as an example, 0.0049 g (2.028 × 10⁻² mmol) Cu (NO₃)₂·3H₂O and 0.1900 g Ce_0.05La_0.95BO₃ were dispersed in DI water and then stirred for 2h. This mixture was heated gently with continuous stirring to remove the excess water. The resulting gel precursor was dried in an oven at 453 K and then heated at 773 K for 10 h to obtain the final material. Other samples loaded with different transition metals (Cu, NI, Co, V, Cr, Mn, Fe, and Zn) were prepared in a similar procedure. For simplicity, all these as-synthesized materials were denoted as Cu/Ce_0.05La_0.95BO₃, Ni/Ce_0.05La_0.95BO₃, Co/Ce_0.05La_0.95BO₃, V/Ce_0.05La_0.95BO₃, Cr/Ce_0.05La_0.95BO₃, Mn/Ce_0.05La_0.95BO₃, Fe/Ce_0.05La_0.95BO₃, and Zn/Ce_0.05La_0.95BO₃.

3.2. Catalyst Characterization

Powder X-ray diffraction (XRD) was performed on Shimadzu Lab XRD-6100 with Cu Kα radiation (λ = 1.5406 Å). The operation voltage, current, and scan speed rate were 40 kv, 30 mA, and 4°/min, respectively. X-ray photoelectron spectroscopy (XPS) test was carried out with Physical Electronics Quantum 2000, equipped with monochromatic Al-Kα source (Kα = 1486.6 eV) at 300 W. The powder material was very thinly pasted on the carbon conductive tape for test preparing. Binding energy was corrected against the C 1s line at 284.8 eV. The UV–Vis diffuse reflectance spectroscopy (UV–Vis DRS) was acquired on a Shimadzu Co. UV-2550 (Japan) spectrophotometer in the wavelength range of 200–800 nm, and a white powder BaSO₄ was used as a reference. Scanning electron microscopy (SEM) was employed on ZEISS MERLIN Compact to observe the morphologies of the as-prepared materials at an accelerating voltage of 10 kV. IR spectra were recorded in KBr disks with a Shimadzu IRprestige-21 FTIR spectrophotometer.

N-butylamine potential titration method was adopted to measure the acidic properties of samples on 904/906 Titrando (Metrohm, Switzerland). For instance, a mass of 50 mg solid was suspended in 50 mL of acetonitrile. Later, this suspension was titrated with a solution of n-butylamine in acetonitrile (0.1 mmol/L) until a constant potential was reached.

NH₃ temperature-programmed desorption (NH₃-TPD) experiment was performed on Auto Chem II 2920 equipped with a thermal conductivity detector (TCD). Before the test, the sample was pretreated in 50 mL/min of He flow at 673 K for 40 min. After cooling to ambient temperature, it was heated to 1173 K in a flow of 6.57% NH₃/He (50 mL/min). The ammonia desorption was monitored by a thermal conductivity detector (TCD).

H₂ temperature-programmed reduction (H₂-TPR) experiment was performed on Auto Chem II 2920 equipped with a thermal conductivity detector (TCD). Before the test, the sample was pretreated in 30 mL/min of He flow at 573 K for 1 h. After cooling to ambient temperature, it was heated to 1073 K in a flow of 10% H₂/Ar (30 mL/min). The hydrogen consumption was monitored by a thermal conductivity detector (TCD).

3.3. Catalytic Evaluations and Kinetic Measurements

The catalytic activities of these samples toward propane total oxidation were measured in a quartz reactor tube (i.d. = 60 mm) under atmospheric pressure. Next, 150 mg of catalyst diluted with 600 mg of quartz sand was placed in the reactor. The reaction gas, composed of C₃H₈ (0.1 vol%), air (80%), and balance Ar, flows through the catalyst in 50 mL/min (weight hourly space velocity (WHSV) = 20,000 mL·g⁻¹_catalyst·h⁻¹). The reaction results were analyzed online by a gas chromatograph (Techcomp GC 7900) equipped with FID detector, while CO and CO₂ concentrations were measured by another GC equipped with TCD detector. The catalytic activities of TMs/Ce_0.05La_0.95BO₃ samples (TM = Cu, Ni, Co, V, Cr, Mn, Fe, and Zn) were measured in the temperature range of 373–773 K. The propane conversion (X_propane) was calculated by the following equation:

$$X_{\mathit{propane}}=\frac{{\left[C_3{H}_8\right]}_{in}-{\left[C_3{H}_8\right]}_{out}}{{\left[C_3{H}_8\right]}_{in}}\times 100\%$$

where the [C₃H₈]_in and [C₃H₈]_out represent the inlet and outlet concentrations of propane, respectively.

The selectivity of CO and CO₂ was defined as follows:

$$X_{{\mathit{CO}}_2}=\frac{{\left[C{O}_2\right]}_{out}}{3\times ({\left[C_3{H}_8\right]}_{in}-{\left[C_3{H}_8\right]}_{out})}\times 100\%$$

$$X_{\mathit{CO}}=\frac{{\left[CO\right]}_{out}}{3\times ({\left[C_3{H}_8\right]}_{in}-{\left[C_3{H}_8\right]}_{out})}\times 100\%$$

where [CO₂]_out and [CO]_out are the CO₂ and CO concentrations in the outlet gas.

The deactivation rate constant (k_d, h⁻¹) was employed to evaluate the catalyst stability and was calculated as follows [82]:

$${\mathrm{k}}_{\mathrm{d}}=\frac{\ln \left(\frac{1-{\left[C_3{H}_8\right]}_{\mathit{out}}}{{\left[C_3{H}_8\right]}_{\mathit{out}}}\right)-\ln \left(\frac{1-{\left[C_3{H}_8\right]}_{\mathit{in}}}{{\left[C_3{H}_8\right]}_{\mathit{in}}}\right)}{\mathrm{t}}$$

(1)

where the t represents the duration hours in the experiment.

The kinetics parameters for C₃H₈ total oxidation were measured in a fixed-bed reactor operated in a differential mode with C₃H₈ conversion below 15%. The effect of the products on the reaction rate (r) can be ignored, and the empirical kinetics expression of the reaction rate can be described as the following equation:

$$r=Aexp\left(-\frac{Ea}{RT}\right)P_{C3H8}^{\alpha }{P}_{O2}^{\beta }$$

(2)

where A is the pre-exponential factor, and E_a is the apparent activation energy (kJ/mol).

4. Conclusions

In the present work, Ce_0.05La_0.95BO₃ was prepared by the citrate method, and different kinds of transition metals were loaded over this material using the wet impregnation method to obtain different TMs/Ce_0.05La_0.95BO₃ catalysts. The TMs’ loading can promote oxygen vacancy formation, which results in an enhancement in the number of acid sites and reducibility of the catalyst. These can remarkably increase the catalytic activity toward propane combustion compared with other materials, i.e., Cu/Ce_0.05La_0.95BO₃ enables the complete propane elimination at 773 K. This material also exhibited favorable thermal stability during 10 h of continuous testing. Generally, the catalytic activity of metal borates toward propane combustion has not been extensively studied; our work on the TMs/Ce_0.05La_0.95BO₃ materials suggests the potential functionality of VOCs removal for rare-earth borates.