Approach to the Characterization of Monolithic Catalysts Based on La Perovskite-like Oxides and Their Application for VOC Oxidation under Simulated Indoor Environment Conditions

: Catalysts are very important in controlling the pollutant emissions and are used for hundreds of chemical processes. Currently, noble metal-based catalysts are being replaced for other kinds of materials. In this study, three lanthanum-based perovskite-like oxides were synthesized (LaCo, LaCoMn, and LaMn) by the glycine-combustion method. The powder catalysts obtained were supported onto cordierite ceramic monoliths using an optimized washcoating methodology to obtain the subsequent monolithic catalysts (LaCo-S, LaCoMn-S, and LaMn-S). Sample characterization conﬁrmed the formation of the perovskite-like phase in the powder materials as well as the presence of the perovskite phase after supporting it onto the monolithic structure. The XPS analysis showed a general decrease in lattice oxygen species for monolithic catalysts, mainly caused by the colloidal silica used as a binder agent during the washcoating process. Additionally, some variations in the oxidation state distribution for elements in Co-containing systems suggest a stronger interaction between cordierite and such catalysts. The catalytic activity results indicated that powder and monolithic catalysts were active for single-component VOC oxidation in the following order: 2-propanol > n-hexane ∼ = mixture > toluene, and there was no evidence of loss of catalytic activity after supporting the catalysts. However, LaMn-S had a better catalytic performance for all VOC tested under dry conditions, achieving oxidation temperatures between 230–420 ◦ C. The oxidation efﬁciency for the VOC mixture was strongly affected by the presence of moisture linking the oxidation efﬁciency at wet conditions to the VOC chemical nature. Additionally, for higher VOC concentrations, the catalyst efﬁciency decreased due to the limited number of active sites.


Introduction
The term volatile organic compounds (VOC) includes a set of hydrocarbons with a high vapor pressure that can become dangerous to human health and the environment when their emission is not effectively controlled. Some health impacts of VOC include headaches, eyes, nose and throat irritation, damage to the central nervous system, and cancer, whereas some of them are corrosive for the environment and effective as greenhouse gases [1]. Among the VOC control techniques, catalytic oxidation is one of the most common alternatives for their removal at low concentrations because of its low energy consumption [2,3]. Nowadays, noble elements as Pt and Pd still prevail for the oxidation of air pollutants because of their good efficiency at low temperatures for several VOC groups [4], but their industrial application is still limited by their high cost and sensitivity to moisture or to be poisoned by chlorinated or sulfur compounds present in polluted gases [5]. Several studies have proposed the VOC oxidation through cheaper and more available catalysts based on oxides from transition metals [4,6], whose catalytic activity is mainly attributed to their redox properties and thermal stability, although they also have some limitations.

Synthesis of Powder and Monolithic Catalysts
The average yield of self-combustion synthesis was 73%, considering losses caused by reaction aggressiveness. In this study, the washcoating methodology was chosen over direct synthesis for monolithic catalyst preparation since it is expected to have higher catalyst load and higher catalytic activity [24,25,37,38]. Additionally, sonication was used during the washcoating methodology to improve the catalyst dispersion onto the monolithic structure). The catalyst load was calculated after one and two immersion cycles, and to estimate the binder contribution (colloidal silica), some tests were performed using a slurry without perovskite. Average load percentages are shown in Table 1. Assuming that binder contribution cannot be higher than 3.0%, a perovskite load around 5% (mass fraction) can be estimated, equivalent to~290 mg of perovskite per monolith. In most studies, catalysts based on noble metals such as Pt or Pd usually do not exceed 1% [39][40][41] considering their high competitiveness, whereas perovskite-like loads can reach around 30% or even more with better profitability [23,24]. However, the catalyst load stays in the background for processes with minimal diffusional limitations (high flows) where the amount of available surface is more important [42]. This study aimed to handle high gas flows, so increasing the catalyst load above 5% was considered unnecessary. After performing stability tests by thermal stress at 700 • C for 2 h and mechanical stress in an acetone ultrasound bath at 70 watts, perovskite losses below 5% were found.

Physicochemical Characterization of Powder and Monolithic Catalysts 2.2.1. X-ray Diffraction (XRD) Analysis
Diffractograms ( Figure 1) were analyzed by comparing them with the ICSD database. The large peak at nearly 38 • confirmed the formation of the perovskite-like crystalline phase for all of the powder catalysts, whereas secondary peaks corresponding to additional phases from metal oxides such as MnO x , LaO x , and CoO x were not observed [43]. LaCo and LaCoMn matched with the LaCoO 2.9 phase (ICSD code: 153991 and 153998 respectively), and LaMn with LaMnO 3.29 (ICSD Code: 167062), all of them with a hexagonal crystal system and space group R-3c. For better comparison, diffractograms for the obtained materials and their ICSD match were plotted in Figure S1. The crystallite size (τ hkl ), calculated according to the Scherrer equation: τ hkl = kλ β hkl cosθ hkl (where: τ hkl is the size of the crystalline domain, k is a factor shape, λ is the experimental X-ray wavelength, β is the line broadening at half the maximum intensity or FWHM, θ is the Bragg angle, and hkl are the Miller's indices for the selected diffracted signal), and unit-cell parameters are listed in Table 2. Since overlapping of 110 and 104 signals occurred, a signal deconvolution was conducted using OriginPro 9.0.0 software previous to the crystallite size analysis (see Figure S2). The peak shifting observed for all signals (for example, in the inset of Figure 1) corresponded with the incorporation of different transition elements inside each perovskite-like structure due to a change in the d-spacing, which depends on the amount of doping [44,45]. Previous work from Liang et al. [44] claimed that all diffraction angles (2θ) for Mn-substituted LaCoO 3 decreased for higher amounts of Mn incorporated.

rated.
The order of crystallite sizes obtained in this study, LaCo > LaCoMn > LaMn, suggests that the partial substitution of Co by Mn decreased the crystallite size and increased the density of structural defects [46] The perovskite phase on the monolithic catalysts were not seen by XRD due to signal overlapping caused by cordierite peaks (Figure 1, "LaCoMn-S" and "cordierite" ICSD code: 91815).

Raman Spectroscopy Analyses
As a complement to XRD, the Raman analysis suggests that the cordierite (main component of the monolith) did not show any interference with the signals coming from the perovskite at the conditions established in this work. Zhang et al. [47] performed a similar analysis for monolithic catalysts, observing a very low influence of cordierite when mixed oxides were analyzed. Figure 2 shows similar peaks for powder and supported catalysts  The order of crystallite sizes obtained in this study, LaCo > LaCoMn > LaMn, suggests that the partial substitution of Co by Mn decreased the crystallite size and increased the density of structural defects [46] The perovskite phase on the monolithic catalysts were not seen by XRD due to signal overlapping caused by cordierite peaks (Figure 1, "LaCoMn-S" and "cordierite" ICSD code: 91815).

Raman Spectroscopy Analyses
As a complement to XRD, the Raman analysis suggests that the cordierite (main component of the monolith) did not show any interference with the signals coming from the perovskite at the conditions established in this work. Zhang et al. [47] performed a similar analysis for monolithic catalysts, observing a very low influence of cordierite when mixed oxides were analyzed. Figure 2 shows similar peaks for powder and supported catalysts ruling out phase changes due to the washcoating process. The Raman active modes for perovskites were identified and labeled as La-V: lanthanum vibration, O-R: oxygen rotations, O-V: oxygen vibrations and JT: Jahn-Teller distortions that could also overlap other vibrational modes, as reported in [48,49]. lanthanum-oxygen dodecahedral vibrations (~157 cm ) and cobalt-oxygen octahedral in LaCo (~196 cm −1 , 416 cm −1 , and 480 cm −1 ) decreased due to Mn incorporation (see the enlarge region between 150-300 cm −1 for LaCoMn). The Jahn-Teller effect (bending and stretching) was stronger for LaMn (two broad peaks) than for LaCo where only one broad peak was observed, whereas for LaCoMn, an additional peak at ~540 cm −1 , whose frequency and intensity were quite different from that for LaMn (~520 cm −1 ), was attributed to the incorporation of Mn into the LaCo structure.  For the LaMn catalyst, the signal at 249 cm −1 is assigned to oxygen rotation of the MnO 6 octahedral, and the two bands near 520 cm −1 and 640 cm −1 are related to the Jahn-Teller octahedral distortions [49], which generate Raman-allowed modes from bending and stretching modes, respectively. Although Raman measurements were not performed below room temperature, it might be possible to see some additional peaks around 300 cm −1 for vibrations of apex oxygen as in [49]. Some room temperature measurements, lattice dynamical calculations, and assignment of Raman modes of LaMnO 3 were previously conducted by Iliev et al. [50] and are in good agreement with our results.
For LaCo, the main peaks correspond to La vibrations (157 cm −1 ), rotation of octahedral oxygen (196 cm −1 ), oxygen deformation vibrations (416 cm −1 and 480 cm −1 ), and Jahn-Teller distortions (640 cm −1 ). The LaCoMn spectrum differs from both LaCo and LaMn because of the partial substitution of Co by Mn. Gnezdilov et al. [49] studied such a substitution effect on the rotation, bending, and stretching-like vibrations of the BO 6 octahedral of perovskites. Through similar analysis, the contribution from peaks associated with lanthanum-oxygen dodecahedral vibrations (~157 cm −1 ) and cobalt-oxygen octahedral in LaCo (~196 cm −1 , 416 cm −1 , and 480 cm −1 ) decreased due to Mn incorporation (see the enlarge region between 150-300 cm −1 for LaCoMn). The Jahn-Teller effect (bending and stretching) was stronger for LaMn (two broad peaks) than for LaCo where only one broad peak was observed, whereas for LaCoMn, an additional peak at~540 cm −1 , whose frequency and intensity were quite different from that for LaMn (~520 cm −1 ), was attributed to the incorporation of Mn into the LaCo structure.

Scanning Electron Microscopy (SEM) and Energy Dispersive Spectroscopy (EDS) Analyses
SEM micrographs revealed the structure of a raw monolith with perfect square channel geometry and non-uniform rough surface (Figure 3a,b). However, after the washcoating process, the channel corners became rounded and the roughness of the surface turned more homogeneous and foamy because of the catalyst accumulation (Figure 3c,d). EDS analyses for monolithic catalysts showed a homogeneous coating not only on the wall surface but inside their macropore structure, since the cordierite (related to Al) and catalyst (related to Mn and Co) regions distinguished by color were equivalent (see Figure 4). SEM micrographs revealed the structure of a raw monolith with perfect square channel geometry and non-uniform rough surface (Figure 3a,b). However, after the washcoating process, the channel corners became rounded and the roughness of the surface turned more homogeneous and foamy because of the catalyst accumulation (Figure 3c,d). EDS analyses for monolithic catalysts showed a homogeneous coating not only on the wall surface but inside their macropore structure, since the cordierite (related to Al) and catalyst (related to Mn and Co) regions distinguished by color were equivalent (see Figure 4).

X-Ray Photoelectronic Spectroscopy (XPS) Analyses
General XPS spectra for monolithic catalysts agreed with the good catalyst deposition observed by SEM and EDS analysis since signals from cordierite (Al, Fe, Mg) were missing. XPS surveys for powder and monolithic catalysts were similar, and LaMn/LaMn-S systems are shown in Figure S3. Monolithic catalysts presented a Si 2s peak associated with colloidal silica used as a binder agent. Quantitative analysis for all the catalysts showed that LaCo/LaCo-S and LaCoMn/LaCoMn-S presented high lanthanum segregation and shifting of binding energies for metallic species after the washcoating process, whereas powder LaMn presented Mn segregation, but LaMn-S did not, and there was no SEM micrographs revealed the structure of a raw monolith with perfect square channel geometry and non-uniform rough surface (Figure 3a,b). However, after the washcoating process, the channel corners became rounded and the roughness of the surface turned more homogeneous and foamy because of the catalyst accumulation (Figure 3c,d). EDS analyses for monolithic catalysts showed a homogeneous coating not only on the wall surface but inside their macropore structure, since the cordierite (related to Al) and catalyst (related to Mn and Co) regions distinguished by color were equivalent (see Figure 4).

X-Ray Photoelectronic Spectroscopy (XPS) Analyses
General XPS spectra for monolithic catalysts agreed with the good catalyst deposition observed by SEM and EDS analysis since signals from cordierite (Al, Fe, Mg) were missing. XPS surveys for powder and monolithic catalysts were similar, and LaMn/LaMn-S systems are shown in Figure S3. Monolithic catalysts presented a Si 2s peak associated with colloidal silica used as a binder agent. Quantitative analysis for all the catalysts showed that LaCo/LaCo-S and LaCoMn/LaCoMn-S presented high lanthanum segregation and shifting of binding energies for metallic species after the washcoating process, whereas powder LaMn presented Mn segregation, but LaMn-S did not, and there was no

X-ray Photoelectronic Spectroscopy (XPS) Analyses
General XPS spectra for monolithic catalysts agreed with the good catalyst deposition observed by SEM and EDS analysis since signals from cordierite (Al, Fe, Mg) were missing. XPS surveys for powder and monolithic catalysts were similar, and LaMn/LaMn-S systems are shown in Figure S3. Monolithic catalysts presented a Si 2s peak associated with colloidal silica used as a binder agent. Quantitative analysis for all the catalysts showed that LaCo/LaCo-S and LaCoMn/LaCoMn-S presented high lanthanum segregation and shifting of binding energies for metallic species after the washcoating process, whereas powder LaMn presented Mn segregation, but LaMn-S did not, and there was no shifting after washcoating. The shifting is usually associated with changes in the chemical environment or oxidation states for some elements [51]. The general increase in oxygen species after the washcoating process was attributed to colloidal silica, and the binding energies and atomic concentrations of the species are reported in Table 3. High-resolution spectra obtained for O 1s, La 3d, Co 2p, and Mn 2p are shown in Figure 5, and the signals were identified and labeled according to the literature for perovskites [44,[52][53][54][55]. Similar changes were observed in the distribution of O, La, Co, and Mn species for all systems. For instance, in LaCo/LaCo-S systems (Figure 5a), the lattice oxygen species (OI: O 2− ) decreased, and other oxygen species (OII: O 2 2− , and OIII: O 2 − ) increased because of the presence of colloidal silica after the washcoating process [44]. Although additional crystallization is not expected by calcination after washcoating, the process as a whole can still promote the formation of highly stable species such as hydroxide and carbonates, for instance.
From Figure 5b, the ill-defined doublets for La 3d in powder catalysts (e.g., LaCoMn) suggest the presence of at least two species or chemical states for La (La I: La 3+ from perovskite, and La II: La 3+ from hydroxide or carbonated species); each one associated with a signal plus a satellite peak for each spin-orbit component (La 3d 3/2, and La 3d 5/2) [52,54]. However, the monolithic catalysts (e.g., LaCoMn-S) showed some well-defined doublets, suggesting the presence of lanthanum hydroxide (or La II species) on the surface of the catalytic system. A similar result was obtained for LaCo-S, whereas for LaMn-S, the dominant La species was the perovskite (or La I) (see Table S3). Figure 5c shows the Co signals for the powder and monolithic catalysts (e.g., LaCo and LaCo-S). Two chemical states were identified associated to the spin-orbit components (Co 2p 1 2 , and Co 2p 3/2) of Co 2+ and Co 3+ [55]. The presence of Co 2+ would justify the surface segregation of La. For powder catalysts, the sharp peak at~780 eV and the satellite peak at~790 eV suggest the presence of the Co 3 O 4 phase [52,54,55].
Similarly, two chemical states associated with spin-orbit components (Mn 2p 1/2, and Mn 2p 3/2) of Mn 3+ and Mn 4+ were identified (e.g., LaMn and LaMn-S) [51] (see Figure 5d). According to Ferrel-Álvarez et al. [51], the presence of Mn 4+ species helps to maintain the electro-neutrality in the perovskite structure and it may also be responsible for Mn segregation observed over the powder catalyst (Table 3). Meanwhile, for supported catalysts, it was observed that the Mn 3+ content only increased for LaCoMn-S since for LaMn/LaMn-S systems, the Mn 3+ did not change significantly.
The significant changes in species distribution for LaCo/LaCo-S and LaCoMn/LaCoMn-S may be related to higher interaction between Co-containing catalysts and cordierite monoliths, as suggested by Guiotto et al. [24]. Therefore, the LaMn catalyst was less affected by the washcoating procedure to anchor the catalyst and showed better incorporation of the species in the perovskite-like structure. Supplementary Materials Tables S2-S5 summarize the highresolution signals for O, La, Co, and Mn.

Catalytic Activity Results
As shown in Figure 6, powder catalysts were tested for the oxidation of n-hexane (expected to be the most difficult to oxidize due to the lack of organic functional groups) using a high space velocity (97,000 h −1 ) to obtain a reference of the catalytic activity under rough conditions. The results obtained indicated that the total oxidation of n-hexane in the catalytic systems was approximately 360 °C lower than the thermal oxidation of nhexane that took place around 800 °C. However, it is important to note that although Cocontaining catalysts presented lower ignition temperatures, the LaMn catalyst completed the oxidation reaction of n-hexane much faster than the other catalysts at a temperature around 440 °C.

Catalytic Activity Results
As shown in Figure 6, powder catalysts were tested for the oxidation of n-hexane (expected to be the most difficult to oxidize due to the lack of organic functional groups) using a high space velocity (97,000 h −1 ) to obtain a reference of the catalytic activity under rough conditions. The results obtained indicated that the total oxidation of n-hexane in the catalytic systems was approximately 360 • C lower than the thermal oxidation of n-hexane that took place around 800 • C. However, it is important to note that although Co-containing catalysts presented lower ignition temperatures, the LaMn catalyst completed the oxidation reaction of n-hexane much faster than the other catalysts at a temperature around 440 • C. Oxidation profiles of single-component VOCs over different monolithic catalysts are shown in Figure 7. A slight comparison between n-hexane oxidation using powder (Figure 6) and monolithic catalysts revealed that the Co-containing perovskites lost their catalytic activity at lower temperatures after supporting it over the cordierite (see also Figure  S4 for full comparison), possibly due to their strong chemical interaction as it was evidenced by XPS. On the other hand, LaMn-S had a better oxidation performance independent of the VOC type used, and the temperatures for total oxidation followed the order: 2propanol < n-hexane < toluene, which is partially in agreement with the relative destructibility of VOCs reported by Tichenor [56]. Several authors have reported that the catalytic destructibility was associated with the presence of functional groups and weak C-H bonds for VOCs [57].  Oxidation profiles of single-component VOCs over different monolithic catalysts are shown in Figure 7. A slight comparison between n-hexane oxidation using powder ( Figure 6) and monolithic catalysts revealed that the Co-containing perovskites lost their catalytic activity at lower temperatures after supporting it over the cordierite (see also Figure S4 for full comparison), possibly due to their strong chemical interaction as it was evidenced by XPS. On the other hand, LaMn-S had a better oxidation performance independent of the VOC type used, and the temperatures for total oxidation followed the order: 2-propanol < n-hexane < toluene, which is partially in agreement with the relative destructibility of VOCs reported by Tichenor [56]. Several authors have reported that the catalytic destructibility was associated with the presence of functional groups and weak C-H bonds for VOCs [57]. Oxidation profiles of single-component VOCs over different monolithic catalysts are shown in Figure 7. A slight comparison between n-hexane oxidation using powder (Figure 6) and monolithic catalysts revealed that the Co-containing perovskites lost their catalytic activity at lower temperatures after supporting it over the cordierite (see also Figure  S4 for full comparison), possibly due to their strong chemical interaction as it was evidenced by XPS. On the other hand, LaMn-S had a better oxidation performance independent of the VOC type used, and the temperatures for total oxidation followed the order: 2propanol < n-hexane < toluene, which is partially in agreement with the relative destructibility of VOCs reported by Tichenor [56]. Several authors have reported that the catalytic destructibility was associated with the presence of functional groups and weak C-H bonds for VOCs [57].  For the VOC mixture, although once again the LaMn-S catalyst showed the best oxidation performance among the catalysts tested ( Figure 8); the shoulder seen in the thermal oxidation profile of the VOC mixture corresponds to the partial decomposition of 2-propanol starting at 400 • C according to the following reaction CH 3 CH 2 CH 2 OH + O 2 → CO 2 + H 2 O + CH 3 CH 3 . This shoulder was also seen in the catalytic decomposition of the VOC mixture but its occurrence is faster and shifted toward lower temperatures. In general, the first section of the conversion profile (below 40% selectivity toward CO 2 ) corresponds to the oxidation of 2-propanol since it was the easiest VOC to be oxidized, while the second section, it was attributed to the combined n-hexane and toluene oxidation. In the latter case, the oxidation proceeds in a homogenous way due to synergistic effects, as reported in [22,56]; however, there are also reports for the negative effects for the oxidation of VOC mixtures [8,57,58]. On the other hand, by including the VOC mixture oxidation results, the following catalytic activity order toward VOC oxidation can be drawn as: 2-propanol > n-hexane ∼ = mixture > toluene (Figure 9).
dation performance among the catalysts tested ( Figure 8); the shoulder seen in the thermal oxidation profile of the VOC mixture corresponds to the partial decomposition of 2-propanol starting at 400 °C according to the following reaction CH3CH2CH2OH + O2 → CO2 + H2O + CH3CH3. This shoulder was also seen in the catalytic decomposition of the VOC mixture but its occurrence is faster and shifted toward lower temperatures. In general, the first section of the conversion profile (below 40% selectivity toward CO2) corresponds to the oxidation of 2-propanol since it was the easiest VOC to be oxidized, while the second section, it was attributed to the combined n-hexane and toluene oxidation. In the latter case, the oxidation proceeds in a homogenous way due to synergistic effects, as reported in [22,56]; however, there are also reports for the negative effects for the oxidation of VOC mixtures [8,57,58]. On the other hand, by including the VOC mixture oxidation results, the following catalytic activity order toward VOC oxidation can be drawn as: 2-propanol > n-hexane ≅ mixture > toluene (Figure 9).
The influence of water was tested for the oxidation of toluene and VOC mixture. Figure 10 shows that the total oxidation temperature for the VOC mixture was negatively affected by the presence of water compared to toluene. Papaefthimiou et al. [18] suggested that the inhibition effects were stronger for molecules resembling water, as was the case of the 2-propanol in our VOC mixture, indicating that the VOC chemical nature had greater influence than the water adsorption on the active sites of the catalyst surface, as other authors have stated [8].  The influence of water was tested for the oxidation of toluene and VOC mixture. Figure 10 shows that the total oxidation temperature for the VOC mixture was negatively affected by the presence of water compared to toluene. Papaefthimiou et al. [18] suggested that the inhibition effects were stronger for molecules resembling water, as was the case of the 2-propanol in our VOC mixture, indicating that the VOC chemical nature had greater influence than the water adsorption on the active sites of the catalyst surface, as other authors have stated [8]. Catalysts 2022, 12, x FOR PEER REVIEW 11 of 17 Figure 9. Temperatures for total oxidation of various VOCs using three different monolithic catalysts. To obtain a better approach to the influence of VOC concentration, water level, and temperature at conditions more similar to polluted indoor environments, results for two sets of experiments during the oxidation of toluene and the VOC mixture using the LaMn-S system are shown in Figure 11. For this, a higher space velocity was used (13,400 h −1 ).  To obtain a better approach to the influence of VOC concentration, water level, and temperature at conditions more similar to polluted indoor environments, results for two sets of experiments during the oxidation of toluene and the VOC mixture using the LaMn-S system are shown in Figure 11. For this, a higher space velocity was used (13,400 h −1 ). To obtain a better approach to the influence of VOC concentration, water level, and temperature at conditions more similar to polluted indoor environments, results for two sets of experiments during the oxidation of toluene and the VOC mixture using the LaMn-S system are shown in Figure 11. For this, a higher space velocity was used (13,400 h −1 ). Under dry conditions, ([VOC]/[H 2 O] = x/0, where x = 400 µL·L −1 , 800 µL·L −1 , and 1200 µL·L −1 ), the catalyst efficiency decreased when VOC concentration increased. This result disagrees with some studies [56] expecting that higher VOC concentration facilitates oxidation by the greater release of heat, but most studies highlight that such behavior is strongly dependent on the VOC type and catalyst composition [58]. For wet condition tests, a fixed VOC concentration of 1200 µL·L −1 was used by varying the amount of water, ([VOC]/[H 2 O] = 1200/x, where x = 5%, 7%, and 10%, in volumetric fraction). For both the toluene and VOC mixture, the performance decreased dramatically as the moisture level increased, although the VOC mixture continued to be more strongly affected. However, it should be considered that the amount of water tested here was still higher compared to those evaluated in most of the literature [59,60], even for noble metal-based catalysts, where, for example, Kim et al. [61] found that the presence of 0.6% of water affected a Pt-based catalytic system, increasing around 50°C the oxidation temperature for 50% of toluene conversion (from 230°C to 280°C, approximately), although the temperature for total oxidation was similar. It is known that carbonate and hydroxyl groups are easily formed on the surface of La-containing perovskites [62], as was also observed here by XPS analysis (Figure 5b). However, a particularly high degree of hydroxylation has been correlated with an increase in hydrophilic properties and a lower catalytic activity for perovskites [63]. The wet conditions evaluated in this study could promote the hydroxylation of the catalyst to obtain a more hydrophilic surface and increase the water competitive adsorption, which can explain the loss of catalytic efficiency observed. , the catalyst efficiency decreased when VOC concentration increased. This result disagrees with some studies [56] expecting that higher VOC concentration facilitates oxidation by the greater release of heat, but most studies highlight that such behavior is strongly dependent on the VOC type and catalyst composition [58]. For wet condition tests, a fixed VOC concentration of 1200 μL•L −1 was used by varying the amount of water, ([VOC]/[H2O] = 1200/x, where x = 5%, 7%, and 10%, in volumetric fraction). For both the toluene and VOC mixture, the performance decreased dramatically as the moisture level increased, although the VOC mixture continued to be more strongly affected. However, it should be considered that the amount of water tested here was still higher compared to those evaluated in most of the literature [59,60], even for noble metal-based catalysts, where, for example, Kim et al. [61] found that the presence of 0.6% of water affected a Ptbased catalytic system, increasing around 50 ℃ the oxidation temperature for 50% of toluene conversion (from 230 ℃ to 280 ℃, approximately), although the temperature for total oxidation was similar. It is known that carbonate and hydroxyl groups are easily formed on the surface of La-containing perovskites [62], as was also observed here by XPS analysis (Figure 5b). However, a particularly high degree of hydroxylation has been correlated with an increase in hydrophilic properties and a lower catalytic activity for perovskites [63]. The wet conditions evaluated in this study could promote the hydroxylation of the catalyst to obtain a more hydrophilic surface and increase the water competitive adsorption, which can explain the loss of catalytic efficiency observed. Red arrows (dotted lines) in Figure 11 indicate that for all temperatures tested, there was no catalyst deactivation when it was switched from wet to dry conditions or vice versa during the same test since the catalytic efficiencies were quite similar to the initial values. When passing from wet to dry conditions, it was clear that the water competitive adsorption disappeared, but achieving the same catalytic efficiency indicates that the possible water-promoted hydroxylation was reversible once the water feed was turned off [64]. Additionally, the blue arrows (dashed lines) in Figure 11 indicate that the perovskitelike catalyst used here possesses good thermal stability since its catalytic activity toward VOC oxidation was not reduced after being exposed at higher temperature conditions. Red arrows (dotted lines) in Figure 11 indicate that for all temperatures tested, there was no catalyst deactivation when it was switched from wet to dry conditions or vice versa during the same test since the catalytic efficiencies were quite similar to the initial values. When passing from wet to dry conditions, it was clear that the water competitive adsorption disappeared, but achieving the same catalytic efficiency indicates that the possible water-promoted hydroxylation was reversible once the water feed was turned off [64]. Additionally, the blue arrows (dashed lines) in Figure 11 indicate that the perovskite-like catalyst used here possesses good thermal stability since its catalytic activity toward VOC oxidation was not reduced after being exposed at higher temperature conditions. Such thermal stability was associated with the nature of the perovskite-like [42,65] and their preparation conditions [9,66].

Synthesis of Powder and Monolithic Catalysts
Three different perovskite-like catalysts were synthesized by the self-combustion method (LaCo, LaCoMn, and LaMn) using the respective metal nitrates of La, Co, and Mn as precursors and glycine as the combustion agent (molar ratios s La:Co = 1:1, La:Co:Mn = 1:0.75:0.25, La:Mn = 1:1, and Gly/NO 3 − = 1). The reactants were mixed by magnetic stirring in a glass beaker, adding a low quantity of water to homogenize. The temperature was set at 80 • C to evaporate most of the water and then raised to 400 • C to induce the combustion. The obtained materials were calcined at 700 • C using a heating rate of 10 • C·min −1 , and then manually milled and supported on cordierite monoliths by using the washcoating methodology (immersion of the monolith in a suspension/slurry of the catalysts) [24,25]. Monoliths with a diameter of 2.1 cm, length of 4.0 cm, and cell density of 200 cpsi (cells per square inch) were used, and slurries were prepared with a mass fraction of 5% of the catalyst, 3% of colloidal silica as the binder agent, and a viscous mixture of glycerol-water (1:1) to improve the dispersion of the catalyst. The slurries were magnetically stirred for 30 min, and then sonicated for 10 min. During the immersion of monoliths, the slurry was kept in sonication to ease the flow of the catalyst through the monolithic channels and to guarantee a good dispersion. The monoliths were vertically immersed into the suspension for 2 min, then the excess slurry was drained by soft blowing of the channels with compressed air. Immersion stages were similar to those proposed by Dwyer et al. [67]. The thermal treatment to fix the catalysts started drying at 100 • C for 1 h and then calcined at 600 • C for 2 h using a heating rate of 5 • C·min −1 . The monolithic catalysts were labeled as LaCo-S, LaCoMn-S, and LaMn-S.

Physicochemical Characterization of Powder and Monolithic Catalysts
Powder and monolithic catalysts were characterized by X-ray diffraction using a diffractometer Panalytical Empyream Serie 2 with radiation of Co kα = 1.78901 Å, operated at 40 kV and 40 mA. The X-ray patterns were recorded in the 2θ range of 10 • -90 • with a pass of 0.026 • . Raman spectra of both powdered and monolithic catalyst were recorded using a Labram HR (Horiba Jobin Yvon) system, equipped with an objective Olympus of 50X working in the visible range with a source of laser excitation of He/Ne emitting radiation of 632.81 nm at 10 mW. Dispersed light was analyzed by a spectrograph with a holographic grid of 600 mm −1 , slit width of 600 µm, and open confocal hole of 800 µm. Exposure time, acquisition time, and scan number were 60 s, 30 s, and 10, respectively. The spectral window was analyzed between 100 cm −1 and 900 cm −1 , where the most important and more intense Raman signals for perovskite-like oxides have been reported [68][69][70]. Scanning electron microscopy (SEM) was carried out using a JEOL JSM-7100 system equipped with a field emission gun (FEG) and an auxiliary detector of retro-scattered electrons operating with an acceleration voltage of 15 kV. Surface analysis of the catalysts was performed with an X-ray photoelectronic spectrometer Specs, with a PHOIBOS 150 1D-DLD analyzer, and a monochromatic source of Al-kα (1486.7 eV, 13 kV, 100 W) with a pass energy of 100 eV for the general spectra and 30 eV for high-resolution spectra. For monolithic catalysts, the flood gun at 7 eV and 60 mA was used.

Evaluation of Catalytic Activity
The catalysts were evaluated in the oxidation of individual hexane (Hex), toluene (Tol), and 2-propanol (2P), and for an equal mixture of them. The water effect was evaluated for the most efficient catalyst at dry conditions, adjusting the moisture levels at 5%, 7%, and 10% (in volumetric fraction). As a reactor, a vertical quartz tube placed inside of an electrically heated ceramic oven was used, and the reaction temperature was measured using a K-type thermocouple in direct contact with the catalytic bed. VOC concentration was adjusted controlling a bubbling airflow inside the liquid VOC [71].
For powder catalysts, 100 mg was supported on quartz wool and placed inside the quartz tube reactor (1.0 cm of inner diameter). Monolithic catalysts were put inside a quartz tube with a 2.1 cm inner diameter. For single-compound VOCs as well as for the mixture, the total concentration was initially fixed to 1000 µL·L −1 in air balance. The total flow rates were adjusted to 600 cm 3 ·min −1 and 1500 cm 3 ·min −1 for individual and multi-component VOCs, respectively, due to limitations in low flow control. Oxidation experiments were performed at dynamic conditions using a heating rate of 5 • C·min −1 , which started at~100 • C and ended at~850 • C. The selectivity toward CO 2 (M/Z = 44) was chosen as the measure of reaction efficiency since in the complete VOC oxidation, the formation of by-products is not expected. CO 2 signal was monitored by mass spectrometry using a ThermoStar TM Gas Analysis System GSD 301 T Pfeiffer Vacuum mass spectrometer, operated at a vacuum pressure of 2 × 10 −6 mbar-6 × 10 −6 mbar. In the last experiment, simultaneous effects of moisture level and VOC concentration were evaluated at stationary temperatures of 350 • C, 400 • C, 450 • C, and 500 • C. For each temperature, the VOC concentration varied from 400 µL·L −1 to 1200 µL·L −1 ; and for a fixed VOC concentration of 1200 µL·L −1 , three water levels of 5%, 7%, and 10% (volumetric fraction) were evaluated. In all cases, a total flow of 3000 cm 3 ·min −1 was used.

Conclusions
Perovskite-like catalysts obtained through the self-combustion method were supported onto cordierite monoliths using an optimized washcoating methodology without leading to crystalline phase changes. Similarly, changes in morphology and roughness of the monolith's channels accompanied by a homogeneous catalyst distribution on the cordierite surface were obtained. From a chemical standpoint, the monolithic catalysts presented an increase in oxygen species other than those coming from the perovskite lattice due to the contribution of colloidal silica used as a binder agent. Additionally, the Co-containing monolithic systems produced less active Co 3+ species, suggesting a strong interaction between perovskite and cordierite, whereas the LaMn-S system showed a high Mn content and low La segregation on the surface, which could explain its better catalytic activity.
All supported catalysts showed activity for the oxidation of n-hexane, toluene, and 2-propanol, decreasing the required temperature for total oxidation by more than 360 • C compared with the thermal process. The catalytic system LaMn-S had the better performance for total oxidation of VOC in all the experiments. For dry systems, the oxidation temperature followed the order: 2-propanol < n-hexane ∼ = mixture < toluene. In the presence of moisture, the VOC nature had a deep impact on the oxidation efficiency, more negatively affecting the VOC mixture, possibly due to the presence of a water resembling compound such as 2-propanol. Furthermore, the number of active sites on the catalyst also played an important role, since catalytic efficiency decreased for higher VOC concentrations. However, LaMn-S showed excellent thermal and chemical stability. The ease of synthesis, the good efficiency, and the thermal and chemical stability are some of the properties promoting monolithic catalysts based on perovskite-like structures such as LaMn for the removal of several VOC groups and the substitution of conventional catalysts in industrial applications.

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/catal12020168/s1, Table S1: Review about oxide-type materials used as catalysts for VOC oxidation. Figure S1: Comparison of diffractograms from the materials obtained and their matched ICSD reference profiles. Figure S2: Deconvolution of peaks around 38 • for analysis of crystallite size. Figure S3: General XPS survey for powder (bottom) and monolithic La-Mn based catalysts (upper). Table S2: High-resolution XPS for O 1s species in powder and monolithic catalysts. Table S3: High-resolution XPS for La 3d species in powder and monolithic catalysts. Table S4: Highresolution XPS for Co 2p species in powder and monolithic catalysts. Table S5: High-resolution XPS for Mn 2p species in powder and monolithic catalysts. Figure S4: Comparative plot of n-hexane oxidation using powder catalysts and the oxidation of other VOC (n-hexane, 2-propanol, and toluene) using monolithic catalysts. ExcelFile: Whole_References_Analysis.