Chlorinated volatile organic compounds (CVOCs) are widely used as solvents and intermediates in the chemical industry. Consequently, the emission of CVOCs has caused great concern owing to environmental protections [1]. CVOCs are the most harmful organic pollutants because of their acute toxicity and high environmental stability [2, 3]. Vinyl chloride (VC) is the main raw material used in the industrial production of polyvinyl chloride (PVC). The concentration of VC in the exhaust is 1%-2%, but only 5 ppm of VC is permitted in the US according to increasing stringent environment limitations [4]. Controlling the emission of VC has become more and more important.

Catalytic oxidation, with low levels of energy consumption and high purification efficiency, is a promising technology for CVOCs emission control compared with direct incineration. The transition metal oxide catalysts, particularly perovskite-type oxides (ABO₃), are potential catalysts in CVOCs removal because of the wide range of resources and their good thermal stability, especially lanthanum manganese perovskite oxide (LaMnO₃) [5].

Texture property and chemical state play important roles in the activity of LaMnO₃. High surface area mesoporous LaCoO₃ oxides with well crystallized perovskite framework shows relatively high catalytic activity [6]. The three dimensionally ordered macroporous (3DOM) of LaMnO₃ has shown to be superior to the bulk counterpart [7], especially in solid-solid-gas reactions [8]. Meanwhile, porous spherical LaMnO₃ and cubic LaMnO₃ nanoparticles are good candidate catalytic materials [9]. Grain boundaries on polycrystalline epitaxial thin films of LaSrMnO₃ may not only facilitate fast oxygen diffusion but also fast oxygen exchange kinetics [10].

Substitution of metal ions in ABO₃ or preparing nonstoichiometric ABO₃ are important ways to change the composition and symmetry while still in its native form. The T₉₀ of the A-substituted (La_0.8Ce_0.2MnO₃) or B-substituted LaMnO₃ (LaB_0.2Mn_0.8O₃) shifted to low temperature during the combustion of vinyl chloride [11, 12], and this was also confirmed by the A-substituted LaMnO₃ (La_1-xAl_xMnO₃)in 1, 2-dichloroethane oxidation [13]. The enhanced oxygen mobility of nonstoichiometric La_0.8MnO₃ promoted the removal of Cl species on the catalyst surface [14].

The texture property and chemical state of LaMnO₃closely depend on the preparation method. In this work, LaMnO₃ perovskite oxides were prepared by various methods [15-17], and catalytic oxidation of vinyl chloride was used as a model reaction. N₂ adsorption-desorption, X-ray diffraction (XRD), Raman spectroscopy (Raman), inductively coupled-plasma (ICP), O₂ temperature programmed desorption (O₂-TPD), H₂ temperature programmed surface reaction (H₂-TPR), and X-ray photoelectron spectroscopy (XPS) were used to characterize the physicochemical properties of the samples. The correlation between the activity and physicochemical properties of the samples was investigated.

Metal nitrates were used as precursors to prepare LaMnO₃ perovskite-type oxides. An aqueous mixture of citric acid and metal nitrates of La and Mn was heated to a sol at 80 ℃, dried at 120 ℃ for 12 h, and calcined at 700 ℃ for 3 h to obtain LaMnO₃, which was denoted as LMO-SG. A NaOH and Na₂CO₃ solution mix was added dropwise to the aqueous La and Mn nitrates and kept the pH value of 10. The precipitate aged at room temperature for 4 h and was then filtered and washed. After drying at 120 ℃ for 12 h, the precipitate was calcined at 700 ℃ for 3 h and denoted as LMO-CP. The sample prepared using the mesoporous silica KIT-6 as a template was named as LMO-HT. For the hydrothermal treatment, a mixture of KOH, La, and Mn nitrate solution was kept into a teflon-lined autoclave at 200 ℃ for 24 h; the product was washed and dried at 120 ℃ for 12 h, calcined at 700 ℃ for 3 h, and denoted as LMO-HM. A solution of KMnO₄ was added dropwise to aqueous MnSO₄; after the KMnO₄ solution was added, the solution continued to stir for 1 h and was then filtered and washed. After vacuum drying at 100 ℃ for 12 h, the product was denoted as MnO₂. A solution of Na₂CO₃ was added dropwise to the aqueous Mn nitrates, and the pH value remained at 10. The precipitate aged at room temperature for 4 h and was then filtered and washed. After drying at 120 ℃ for 3 h, it was calcined at 650 ℃ for 3 h and denoted as Mn₂O₃.

Nitrogen adsorption-desorption at low temperature (-196 ℃) was performed on a NOVA 4200e surface area and porosity analyzer. The specific surface area (SSA) was obtained by the Brunauer-Emmett-Teller (BET) method.

The powder XRD measurements were performed on a Bruker AXS D8 Focus diffractometer with Cu K_α radiation (40 kV, 40 mA, λ = 0.154 nm) at a scanning rate of 6°/min. The average crystallite sizes of the catalysts were evaluated using the Scherrer equation.

Inductively coupled-plasma atomic emission spectroscopy (ICP-AES) was carried out on a Varian 710-ES instrument to determine the chemical compositions.

O₂-TPD and H₂-TPR were carried out on commercial equipment (Penxiang Co., Tianjing, China). A 50-mg sample was saturated with O₂ for 40 min at 30 ℃ and then purged by He to remove the unstable adsorbed O₂ and residual O₂ in the pipeline. The signal of desorbed oxygen (m/z = 32) was recorded by online mass spectrometry when heating the sample from 30 to 600 ℃ at a ramp of 10 ℃/min. In H₂-TPR experiments, a sample was heated in a flow of 5% H₂-95% N₂ from ambient temperature to 450 ℃ at a rate of 10 ℃/min. The amount of H₂ consumption was measured by a thermal conductivity detector (TCD) with the H₂ consumption with CuO as the reference for quantitative analysis.

The XPS spectra were acquired with an AXIS Ultra DLD spectrometer using Mg-K_α (hν = 1253.6 eV) radiation. Charged samples were avoided by setting the binding energy of adventitious carbon (C 1s) to 284.8 eV.

Raman spectra were recorded with a thin wafer on a Renishaw spectrometer using a 514.5 nm Ar⁺ laser as the excitation source at room temperature. The laser beam intensity and the spectrum slit width were 3 mW and 2 cm^-1, respectively.

The activity test was carried out in a continuous flow fixed-bed quartz reactor with 6-mm I.D. The catalyst (40-60 mesh) was packed in the reactor, and the catalytic activity measurement was carried out in the temperature range from room temperature to 500 ℃ at 1% VC-99% N₂ and controlled by a mass flowmeter that was diluted by air to reach a VC concentration of 1000 ppm at GHSV of 15000 h^-1. The conversion of VC was obtained by an on-line gas chromatograph (Minrui 2060) equipped with a flame ionization detector (FID). A quadrupole mass spectrometer (INFICON IPC400) was used to detect other by-products.

Catalytic behaviors of the samples for catalytic combustion of VC were tested and shown in Fig. 1. T₅₀ and T₉₀ (the temperatures at 50% and 90% VC conversion, respectively) for all of the catalysts were between 150-240 ℃ and 218-281 ℃, respectively. A mass spectrometer was used to identify the by-products. Only HCl, Cl₂, CO₂, and H₂O were detected, and the other chloric-hydrocarbons were not found. It was shown that the preparation had obvious effects on VC removal. The T₅₀ of the samples shifted to high temperature according to the following sequence: LMO-SG < LMO-CP < LMO-HT < LMO-HM. LMO-SG exhibited the highest activity with T₉₀ of 182 ℃, which was almost 100 ℃ lower than that of LMO-HM.

The activity of Mn oxides (MnO₂ and Mn₂O₃) was also tested, and the results are shown in Fig. 1. Compared with Mn₂O₃, MnO₂ possessed excellent performance during VC combustion. VC was completed converted over MnO₂ at 190 ℃, whereas complete conversion over Mn₂O₃ occurred at 410 ℃. The aforementioned result indicated that higher Mn valent values led to higher activity. Unfortunately, MnO₂ had poor stability shown in the insert of Fig. 1; T₅₀ of the second run was 80 ℃ higher than that of the first run. Unlike MnO₂, there was no obvious temperature difference in T₅₀ of the two runs of LMO-SG.

The XRD patterns of the samples are shown in Fig. 2. Only single phases were found in the XRD patterns, except for that of LMO-HM. The characteristic diffraction peaks of La(OH)₃ species were detected in LMO-HM at 2θ of 27.3°, 28.0°, and 39.5°, which were ascribed to La(OH)₃ species. The presence of this hydroxide attributed to the hydroxylation of the lanthanum present in the sample upon exposure to atmospheric humidity [18].

The characteristic diffraction peaks at 22.9° (101), 32.7° (121), 40.2° (220), and 52.6° (103), which correlated with LaMnO_3.15, were found in LMO-SG, LMO-HM, and LMO-HT. The main peaks of sample LMO-CP at 23.1° (012), 32.8° (104), 40.3° (202), 52.7° (122) indicated that it had a LaMnO_3.26 (JCPDS#50-0299) perovskite structure. In addition, the average crystallite sizes of the samples calculated by the Scherrer equation were in the range of 15.7 to 21.2 nm.

Specific surface areas and metal concentrations are summarized in Table 1. According to ICP analysis, the molar ratios of La and Mn atoms for all samples were nearly equal to the theoretical values. Preparation methods had obvious effects on the surface areas. LMO-HM had the lowest surface area, which was only 15.7 m²/g, while the surface area of LMO-HT was almost 3 times higher than that of LMO-HM.

Raman is very sensitive and used to investigate the structure of oxides and the chemical environment. Two peaks were determined from Raman analysis (Fig. 3); the intensive one located at 665 cm^-1 related to the extension of Mn-O in MnO₆ units [19], and the other peak at 485 cm^-1reflected the extension and compression of Mn-O bond pairs. A more intensive peak at 665 cm^-1 was observed with LMO-HM compared with the other samples.

The peak at 485 cm^-1 was linked with Jahn Teller distortion [20], which correlated with the ratio of Mn⁴⁺/Mn³⁺ on the catalyst surface. The coordination oxygen atoms of Mn³⁺O₆ can form a Jahn-Teller distorted octahedral, whereas Mn⁴⁺ ions in Mn⁴⁺O₆ units have almost undistorted coordination [21]. The more severe the Jahn-Teller distortion, the higher concentration of Mn³⁺ presented in LaMnO₃.

O₂-TPD was performed to investigate the adsorption amount and the rate of oxygen activation on the sample surface. Oxygen desorption profiles below 600 ℃ are displayed in Fig. 4, as the catalytic combustion of vinyl chloride usually occurred below 500 ℃.

Desorption peaks in the range of 150-450 ℃ were ascribed to the weaker molecular physisorbed and/or chemisorbed oxygen [22], Oα, on the sample surface. Oα are regarded as the most active species in the reaction of catalytic oxidation. Meanwhile, the amount of Oα was also used to characterize the concentration of oxygen vacancy on the surface or subsurface, which played an important role in exchange and transformation between gas oxygen and lattice oxygen [23]. To compare the relative amount of the O₂ desorption amount, the O₂ desorption peak area of LMO-SG was set as 1.00 as shown in Table 2. The results suggested that the amount of adsorbed O₂ species was related to the preparation method.

H₂-TPR profiles of the samples are shown in Fig. 5. The H₂ consumption amount, summarized in Table 2, was calculated using CuO as a reference. A small peak and a broad peak with a shoulder were found in the range of 100-250 ℃ and 250-500 ℃, respectively. Three peaks were obtained by deconvolution; α was ascribed to removing physically adsorbed oxygen and/or the ordinarily chemically adsorbed oxygen, β was attributed to the nonstoichiometric excess oxygen accommodated within the lattice, and γ was due to the reduction of Mn⁴⁺ to Mn³⁺ or the single-electron reduction of Mn³⁺ located in a coordination-unsaturated microenvironment [24].

Compared with the other samples, the reduction temperature of LMO-SG was shifted to a lower temperature. The reduction α of LMO-SG was only 232 ℃, which was 22 ℃ lower than that of LMO-HM. The temperature difference for the reduction of β between LMO-SG and LMO-HM was larger, approximately 40 ℃. There was not much difference in the total H₂ uptake between LMO-SG and LMO-CP; while the total H₂ uptake of LMO-HM decreased up to 20% compared with LMO-SG. The H₂ consumption of α and β was consistent with the following sequence: LMO-SG > LMO-CP > LMO-HT > LMO-HM.

Mn 2p and O 1s XPS spectra are illustrated in Fig. 6, and the integration of the corresponding peaks allowed us to determine the atomic ratios of surface species (Table 3).

A broad and asymmetrical Mn 2p_3/2 peak at 642 eV and Mn 2p_1/2 peak at 652 eV were observed in Fig. 6(a). By deconvoluting the Mn 2p_3/2 and Mn 2p_1/2 peaks, it was found that Mn species were mainly in the form of Mn⁴⁺ and Mn³⁺[25-27]. The surface La/Mn ratio of all samples was higher than the theoretic La/Mn atomic ratio, indicating a La surface enrichment. The extent of La enrichment on surface correlated with the preparation method. Despite the surface La enrichment in LMO-SG, LMO-CP, and LMO-HT, XRD analysis did not reveal the presence of diffraction peaks attributed to lanthanum oxide (La₂O₃) or lanthanum oxycarbonate ((LaO)₂CO₃); this could be due to the low quantity of those species on the sample surface. The ratio of La/Mn on LMO-HM was two times higher than that of the theoretic La/Mn atomic ratio, indicating the formation of new La species. The surface enrichment of lanthanum was ascribed to lanthanum (oxy/hydroxy) carbonate [28]. The detection of La(OH)₃ by XRD on LMO-HM confirmed the abovementioned speculation. The highest ratio of Mn⁴⁺/Mn was found on LMO-SG, which meant that there were more Mn⁴⁺ species on the surface of LMO-SG.

In Fig. 6(b), the asymmetrical O 1s spectra of each sample was de-convoluted to three peaks: the first peak at 529.4-529.7 eV corresponded to lattice oxygen O_latt (O^2-) in the oxide network; the second peak at 531.1-531.4 eV was ascribed to the surface-adsorbed O_ads(O^-, O₂^-or O₂^2-) from hydroxyl or carbonate groups, and the third peak at 533 eV was due to the adsorbed molecular water [29]. The oxygen species and the distribution varied with the manner by which the samples were prepared.

The LMO-SG had the highest O_ads/O_total molar ratio among all the samples, while the LMO-HM had the lowest ratio. The ranking in terms of O_ads/O_total molar ratio was in agreement with the results of H₂-TPR and O₂-TPD experiments. Combined with the analysis of Mn species, it was found the molar ratio of O_ads/O_total followed the same sequence as that of Mn⁴⁺/Mn³⁺. Particularly, the concentration of surface oxygen vacancies (λ = 3.15 -O_latt/(La + Mn)) were calculated using the XPS results (Table 3). The oxygen vacancies existed on the perovskite corroborating with the O₂-TPD results. Meanwhile, the λ value was in line with the catalytic activity. This indicated that surface oxygen vacancies played an important role in oxidation reaction since they were responsible for the adsorption desorption properties of the gas phase, and they facilitated the diffusion of lattice oxygen from the bulk to the surface.

The preparation method decided the morphology, redox properties, and chemical states of the catalysts; compared with the first factor, the last two factors played a more important role in deciding the catalytic activity of LaMnO₃ perovskite-type materials in the oxidation reaction. The nature of redox usually dominated the active oxygen species and oxygen activation, while the chemical state of the active species was consistent with the adsorption and activation of reactant.

LaMnO₃ perovskite oxides were synthesized by citrate sol-gel, co-precipitation, hard template, and hydrothermal methods, and catalytic performances were evaluated for VC oxidation. The poor activity of LMO-HM was assigned to the presence of La(OH)₃ on the surface. Although LMO-HT, prepared by using KIT-6 as template, had the largest specific surface area, it still led to fewer active oxygen species on the surface. In the process of preparing sample by citrate sol-gel, citric acid acted as a complexing agent, which increased the solubility of metal ions and helped maintain homogeneity by preventing selective precipitation, leading to the smallest obtained crystallite size. LMO-SG exhibited the highest activity with T₉₀ of 182 ℃.

From H₂-TPR profiles (Fig. 5) and H₂ uptake (Table 2), the total H₂ consumption in the low temperature range for the reduction of the adsorbed oxygen species varied with the catalyst preparation. From the O₂-TPD study (Fig. 4) and XPS analysis (Table 3), more desorbed oxygen species indicated that more oxygen vacancies were generated and participated in the oxygen migration process. The amount of adsorbed oxygen species and the rate of the oxygen activation followed the same sequence as that of the catalytic activity. The VC combustion occurred at low temperature ( < 300 ℃), so adsorbed oxygen was the dominant oxygen species participating in this reaction, namely the reaction followed the suprafacial mechanism.

Raman spectra (Fig. 3) showed the variance in the distribution of Mn³⁺ on all the samples, and XPS (Fig. 6) characterization also confirmed that the molar ratio of Mn⁴⁺/Mn³⁺ on the surface increased in the same sequences as the amount of oxygen vacancies. The first step of the reaction is adsorption; a higher amount of adsorption sites was on the surface, thereby activating a higher amount of reactant. The activity tests of MnO₂ and Mn₂O₃ confirmed that the high valence of Mn provided the active sites for VC adsorption and activation (Fig. 1). However, LaMnO₃ exhibited better stability than MnO_2, indicating that LaMnO₃ was a good catalyst candidate for future practical application.

LaMnO₃ perovskite oxides were synthesized by citrate sol-gel, co-precipitation, hard template, and hydrothermal methods, and the catalytic performances were evaluated for VC catalytic oxidation. LaMnO₃-SG prepared using the citrate sol-gel method exhibited the best performance in catalytic combustion of VC with T₉₀ of 182 ℃ and stability. The reaction followed the suprafacial mechanism, and no co-relation was found between surface area and activity. The ratio of Mn⁴⁺/Mn³⁺ and its redox ability differed based on catalyst preparation. The enrichment of Mn⁴⁺ benefited the adsorption and activation of VC, and the presence of oxygen vacancies facilitated the adsorption of oxygen. The improved redox capability of Mn⁴⁺/Mn³⁺ guaranteed the supply of active oxygen species. More available adsorbed oxygen and Mn⁴⁺ species on the surface were mainly responsible for the remarkable enhancement of the catalytic activity.