Sepiolite-Supported Manganese Oxide as an Efficient Catalyst for Formaldehyde Oxidation: Performance and Mechanism

The current study involved the preparation of a number of MnOx/Sep catalysts using the impregnation (MnOx/Sep-I), hydrothermal (MnOx/Sep-H), and precipitation (MnOx/Sep-P) methods. The MnOx/Sep catalysts that were produced were examined for their ability to catalytically oxidize formaldehyde (HCHO). Through the use of several technologies, including N2 adsorption–desorption, XRD, FTIR, TEM, H2-TPR, O2-TPD, CO2-TPD, and XPS, the function of MnOx in HCHO elimination was examined. The MnOx/Sep-H combination was shown to have superior catalytic activities, outstanding cycle stability, and long-term activity. It was also able to perform complete HCHO conversion at 85 °C with a high GHSV of 6000 mL/(g·h) and 50% humidity. Large specific surface area and pore size, a widely dispersed active component, a high percentage of Mn3+ species, and lattice oxygen concentration all suggested a potential reaction route for HCHO oxidation. This research produced a low-cost, highly effective catalyst for HCHO purification in indoor or industrial air environments.


Introduction
Formaldehyde (HCHO) is a gaseous indoor pollutant and possible human carcinogen that is released by paints, plastic cements, construction and furnishing materials, and other consumer goods [1][2][3][4].Therefore, HCHO removal technologies have been developed to lessen HCHO pollution indoors.Catalytic oxidation is an economical and environmentally benign method of directly breaking down HCHO into CO 2 and H 2 O [5][6][7][8].Noble metal catalysts (Pt, Au, Pd, Ag, and Ir) can fully catalyze HCHO at low or ever room temperature [9][10][11][12][13][14].However, the practical use of catalysts was limited by the high production costs associated with the usage of precious metals.Many non-noble metal oxide catalysts (such as Ce, Mn, Co, Fe, and Cu) have been produced to date for the complete oxidation of HCHO at low temperature [15][16][17][18][19][20][21].Among them, manganese-based catalysts with various supports, plentiful valence states, structural flexibility, and functional design have garnered the most interest because of their superior catalytic activity at lower temperatures [6,[22][23][24][25][26][27].But the low catalytic activity remains a barrier that these catalysts urgently need to overcome.
Other techniques include supported-manganese-based catalysts, in which manganese oxides are typically placed on high-specific-surface-area carrier materials, like zeolite [28][29][30], activated carbon (AC) [31,32], and MOF [33].These carrier materials are primarily used to increase the dispersion degree of the active components and the super capture ability of HCHO and serve as a platform for designing and synthesizing manganese oxides with richer oxygen vacancies or metal defects that can activate surface oxygen species and H 2 O to degrade the captured HCHO into harmless CO 2 and H 2 O.To the best of our knowledge, we reviewed the pertinent papers and discovered that the natural clay mineral sepiolite, with its large specific surface area, stability, and abundance of hydroxyl species, is a perfect carrier material that can avoid an aggregation of active components and capture ability of HCHO [20,34].In Hunan province, China, sepiolite (Sep) is a naturally occurring phyllosilicate fibrous clay mineral with large reserves and inexpensive pricing (0.4-0.6 $/kg) [35].It is composed of continuous tetragonally coordinated silicon layers that reverse their orientation every few tetrahedral units and discontinuous octahedrally coordinated magnesium layers.Sepiolite's unique structure accounts for its homogeneous pore size, elevated porosity, large surface area, and profusion of silanol groups [36][37][38][39].Although there have been reports of sepiolite in HCHO degradation, most of them have focused on photocatalysis [40][41][42], with thermocatalysis receiving even fewer reports [20,34].Our research on HCHO oxidation via the use of sepiolite and silver has shown that the catalysts have unique adsorption characteristics and are capable of fully decomposing HCHO at 52 • C into CO 2 and H 2 O [43].However, the usage of silver species carries a price risk for the catalyst.
On this basis, based on the effective catalytic ability of manganese and the unique properties of sepiolite, we synthesized three sepiolite-manganese catalysts by impregnation (MnO x /Sep-I), precipitation (MnO x /Sep-P), and hydrothermal methods (MnO x /Sep-H).We then assessed the catalytic activities of these catalysts for the total degradation of HCHO.We found that at lower temperatures, the MnO x /Sep-H catalyst exhibited better catalytic efficacy because of the higher ratio of Mn 3+ /Mn 4+ , larger pore size and abundant surfaceactive oxygen species.Moreover, the technologies of XRD, FTIR, TPD/TPR, XPS, and N 2 adsorption-desorption were used to demonstrate the important effect that vacancies play on surface oxygen species and H 2 O activation for HCHO oxidation.The potential mechanism for the HCHO oxidation process was finally put forth.

Structural Characterization
The XRD patterns for MnO x /Sep-P, MnO x /Sep-I, and MnO x /Sep-H are shown in Figure 1.The findings showed that the (110), (010), (101), and (010) planes of in Sep were responsible for the diffraction reflections at 2θ = 9.5 • , 19.8 • , 20.7 • , and 32.5 (JCPDS 75-1621) [43,44].Furthermore, the distinctive diffraction reflections of the Mn 3 O 4 structure (JCPDS, No. 80-0382) were found at 28.9 • , 32.4 • , and 36.1 • .These peaks might be attributed to the (112), (103), and (211) crystal facets of Mn 3 O 4 , respectively [7,25].It appears that the structure of Sep remained unaffected by the addition of manganese species.The fact that the Mn 3 O 4 structure diffraction reflections for MnO x /Sep-P and MnO x /Sep-I were not readily visible suggests that the Mn 3 O 4 structures were evenly distributed across the sepiolite surface.However, the Mn 3 O 4 structure's diffraction reflections were seen in MnO x /Sep-H, indicating that it might produce a Mn 3 O 4 structure.The existence of Mn 3 O 4 might promote the formation of Mn-vacancies (Mn v ) or manganese defects [25,26].The formation of a Mn v is conducive to the adsorption and activation of oxygen molecules, producing the reactive oxygen species [45][46][47].Several investigations have demonstrated that the formation of Mn-vacancies or manganese defects in catalysts is advantageous for absorbing oxygen and water molecules, creating surface-active oxygen species, and accelerating the catalytic oxidation of HCHO [17,23,25].
In Figure 2, the catalysts' FTIR spectra are displayed.For Sep, the -OH groups are represented by the bands near 3420 cm −1 , the hydroxyl groups in crystalline water and surface adsorption water are represented by the bands at 3637 cm −1 and 3637 cm −1 , and the -OH vibration of H 2 O is responsible for the peaks at 1640 cm −1 [36,38,43].Furthermore, the vibrations of Si-O in the tetrahedral silica sheet are responsible for the peaks at 1030 and 795 cm −1 [36,39].After loading the manganese species, the intensity of vibration of the hydroxyl groups and other group structure did not significantly change for MnO x /Sep-P, MnO x /Sep-I, and MnO x /Sep-H.The results indicated that surface hydroxyl species are abundant, which facilitates HCHO adsorption and may enhance the catalyst's catalytic activity [48].Furthermore, for MnO x /Sep-H, additional vibration peaks emerged at 603 cm −1 , which could be related to the vibration of the Mn-O bond [49].Manganese oxides may have been more firmly bound to Sep as shown by the formation of the Mn-O bond in Mn 3 O 4 , which could have aided in the creation of Mn-vacancies (Mn v ).This finding was in line with the XRD results as well, which provided more evidence that Mn-vacancies could exist in the MnO x /Sep-H catalyst.In Figure 2, the catalysts' FTIR spectra are displayed.For Sep, the -OH groups are represented by the bands near 3420 cm −1 , the hydroxyl groups in crystalline water and surface adsorption water are represented by the bands at 3637 cm −1 and 3637 cm −1 , and the -OH vibration of H2O is responsible for the peaks at 1640 cm −1 [36,38,43].Furthermore, the vibrations of Si-O in the tetrahedral silica sheet are responsible for the peaks at 1030 and 795 cm −1 [36,39].After loading the manganese species, the intensity of vibration of the hydroxyl groups and other group structure did not significantly change for MnOx/Sep-P, MnOx/Sep-I, and MnOx/Sep-H.The results indicated that surface hydroxyl species are abundant, which facilitates HCHO adsorption and may enhance the catalyst's catalytic activity [48].Furthermore, for MnOx/Sep-H, additional vibration peaks emerged at 603 cm −1 , which could be related to the vibration of the Mn-O bond [49].Manganese oxides may have been more firmly bound to Sep as shown by the formation of the Mn-O bond in Mn3O4, which could have aided in the creation of Mn-vacancies (Mnv).This finding was in line with the XRD results as well, which provided more evidence that Mn-vacancies could exist in the MnOx/Sep-H catalyst.In Figure 2, the catalysts' FTIR spectra are displayed.For Sep, the -OH groups are represented by the bands near 3420 cm −1 , the hydroxyl groups in crystalline water and surface adsorption water are represented by the bands at 3637 cm −1 and 3637 cm −1 , and the -OH vibration of H2O is responsible for the peaks at 1640 cm −1 [36,38,43].Furthermore, the vibrations of Si-O in the tetrahedral silica sheet are responsible for the peaks at 1030 and 795 cm −1 [36,39].After loading the manganese species, the intensity of vibration of the hydroxyl groups and other group structure did not significantly change for MnOx/Sep-P, MnOx/Sep-I, and MnOx/Sep-H.The results indicated that surface hydroxyl species are abundant, which facilitates HCHO adsorption and may enhance the catalyst's catalytic activity [48].Furthermore, for MnOx/Sep-H, additional vibration peaks emerged at 603 cm −1 , which could be related to the vibration of the Mn-O bond [49].Manganese oxides may have been more firmly bound to Sep as shown by the formation of the Mn-O bond in Mn3O4, which could have aided in the creation of Mn-vacancies (Mnv).This finding was in line with the XRD results as well, which provided more evidence that Mn-vacancies could exist in the MnOx/Sep-H catalyst.Based on the previous literature [50,51], the N 2 adsorption-desorption curve and pore size distribution patterns of Sep, MnO x /Sep-P, MnO x /Sep-I, and MnO x /Sep-H are displayed in Figure 3.Each sample showed a type IV isotherm and a type H4 hysteresis loop, indicating that they were all made of mesoporous material [36,43].Also, the physical parameters of the mesoporous materials and surface element composition of the samples are displayed in Table 1.The pore diameter was between 10 and 50 nm.The S BET , V pore , and D pore of Sep were 325.16 m 2 /g, 0.32 cm 3 /g, and 3.82 nm.The partial coverage of the Sep surface pores by Mn species causes the S BET to fall to 229.45, 136.13, and 176.77m 2 /g after adding manganese species.However, the V pore for the samples were barely altered.
In contrast to Sep (3.82 nm), the D pore of MnO x /Sep-P and MnO x /Sep-I decreased from 3.82 nm to 3.41 nm and 3.42 nm, respectively.But the D pore of MnO x /Sep-H (3.82 nm) remained relatively unchanged, indicating that the manganese species did not block the pores of sepiolite in MnO x /Sep-H catalysts.HCHO adsorption might be favored by a bigger pore size; hence, it further promotes the catalytic activity of the catalyst.
lite in MnOx/Sep-H catalysts.HCHO adsorption might be favored by a bigger pore size; hence, it further promotes the catalytic activity of the catalyst.Figure 4 display the H2-TPR profiles of the catalysts as prepared.The purpose of the experiment was to assess the reactivity and reducibility of various oxygen species in catalysts.Each sample curve had two distinct reduction peaks, as Figure 4 illustrates.These peaks corresponded to the reduction of bigger-particle manganese oxides (MnO2→Mn2O3→Mn3O4→MnO) and surface-active oxygen [30,52].Theoretically, significant reducibility and high reactivity of the oxygen species are indicated by a lower reduction temperature.Since the peak of MnOx/Sep-H's reduction temperature (202 °C) is lower than that of MnOx/Sep-P (230 °C) and MnOx/Sep-I (290 °C), it may be concluded that MnOx/Sep-H's surface-adsorbed oxygen showed the best reducibility [48].Furthermore, for MnOx/Sep-H, MnOx/Sep-P, and MnOx/Sep-I, the reduction temperatures were 680 °C,  The S BET , V pore , and D pore calculated by the BET method.b Surface element molar ratio calculated by peak areas of XPS.c The AOS values calculated by peak areas of XPS.
Figure 4 display the H 2 -TPR profiles of the catalysts as prepared.The purpose of the experiment was to assess the reactivity and reducibility of various oxygen species in catalysts.Each sample curve had two distinct reduction peaks, as Figure 4 illustrates.These peaks corresponded to the reduction of bigger-particle manganese oxides (MnO 2 →Mn 2 O 3 →Mn 3 O 4 →MnO) and surface-active oxygen [30,52].Theoretically, significant reducibility and high reactivity of the oxygen species are indicated by a lower reduction temperature.Since the peak of MnO x /Sep-H's reduction temperature (202 • C) is lower than that of MnO x /Sep-P (230 • C) and MnO x /Sep-I (290 • C), it may be concluded that MnO x /Sep-H's surface-adsorbed oxygen showed the best reducibility [48].Furthermore, for MnO x /Sep-H, MnO x /Sep-P, and MnO x /Sep-I, the reduction temperatures were 680 • C, 695 • C, and 683 • C for bigger-particle manganese oxides, and the specific gravity of hydrogen consumption was determined to be 1.8:1.6:1,suggesting that MnO x /Sep-H has more lattice oxygen species, which may also promote catalytic activity.
To learn more about the desorption patterns and chemical reactions of various oxygen species, O 2 -TPD was carried out.Two peaks were seen at 100-300 • C and 600-800 • C in the O 2 -TPD profiles of the MnO x /Sep-H, MnO x /Sep-P, and MnO x /Sep-I catalysts in Figure 5.The adsorbed oxygen species (O 2 − , O 2− or O − ) at the surface was the first desorption peak and is active for the full oxidation of HCHO.The bulk lattice oxygen (O 2− ) is responsible for other desorption peaks.The desorption temperatures of MnO x /Sep-H, MnO x /Sep-P, and MnO x /Sep-I were 137 • C, 141 • C, and 170 • C, respectively, for surfaceadsorbed oxygen species.Furthermore, the specific gravity of oxygen desorption for MnO x /Sep-H, MnO x /Sep-P, and MnO x /Sep-I was determined to be 1.4:1.3:1.2.According to these findings, MnO x /Sep-H, MnO x /Sep-P, and MnO x /Sep-I has a higher concentration of activated oxygen species than the others, which is advantageous for the oxidation of HCHO [48].
Molecules 2024, 29, x FOR PEER REVIEW 5 of 16 695 °C, and 683 °C for bigger-particle manganese oxides, and the specific gravity of hydrogen consumption was determined to be 1.8:1.6:1,suggesting that MnOx/Sep-H has more lattice oxygen species, which may also promote catalytic activity.Furthermore, the basic sites in Figure 6 were identified using CO 2 -TPD profiles.MnO x /Sep-H had its first peak at 201 • C, while MnO x /Sep-P and MnO x /Sep-I had their first peaks at 204 • C and 212 • C, respectively.The reason for this was that compared to the surfaces of MnO x /Sep-P and MnO x /Sep-I, the basic sites on the MnO x /Sep-H surface were weaker.Furthermore, the specific gravity of CO 2 desorption for MnO x /Sep-H, MnO x /Sep-P, and MnO x /Sep-I was determined to be 1.8:1.7:1.2.According to these findings, MnO x /Sep-H was favorable for CO 2 desorption, which can promote the desorption of CO 2 products and increase catalytic activity [43].The CO 2 desorption peaks of MnO x /Sep-H were pushed to lower temperatures in the high-temperature range (500-800 • C) in comparison to those of MnO x /Sep-P and MnO x /Sep-I.Therefore, the MnO x /Sep-H had more basic sites that were advantageous for the adsorption of HCHO on its surface, and its higher catalytic activity was due to its decreased CO 2 desorption.
surfaces of MnOx/Sep-P and MnOx/Sep-I, the basic sites on the MnOx/Sep-H surface were weaker.Furthermore, the specific gravity of CO2 desorption for MnOx/Sep-H, MnOx/Sep-P, and MnOx/Sep-I was determined to be 1.8:1.7:1.2.According to these findings, MnOx/Sep-H was favorable for CO2 desorption, which can promote the desorption of CO2 products and increase catalytic activity [43].The CO2 desorption peaks of MnOx/Sep-H were pushed to lower temperatures in the high-temperature range (500-800 °C) in comparison to those of MnOx/Sep-P and MnOx/Sep-I.Therefore, the MnOx/Sep-H had more basic sites that were advantageous for the adsorption of HCHO on its surface, and its higher catalytic activity was due to its decreased CO2 desorption.[38,39].These visualizations suggest a robust interaction between the active ingredients and the carrier, which is advantageous for the dispersion of active components and the formation of new bonds.
Three different types of oxygen species were found in the core level O 1s spectra (Figure 7B [17,23,54].For MnOx/Sep-H and MnOx/Sep-I, the corresponding AOS values were found to be 2.31 and 2.65, respectively.The AOS of the surface Mn for MnOx/Sep-I was close to the typical oxidation state of Mn in Mn3O4 (2.67), whereas MnOx/Sep-H catalysts had a lower AOS of surface Mn (2.31), suggesting that it may form the Mn-vacancies or O-vacancies.The Mn 2p3/2 spectra, which were fitted into two peaks (Figure 7C) at around ∼642.1 eV and ∼643.2 eV, which can be attributed to Mn 3+ and Mn 4+ species, respectively [21,25,26,48].
According to the XPS-41 software's fitting results, MnOx/Sep-P, MnOx/Sep-H, and MnOx/Sep-I have Mn 3+ /Mn 4+ fractions of 1.15:1, 1.67:1, and 1:1, respectively.MnOx/Sep-H has a high percentage of Mn 3+ , and the surface Mn 3+ promotes the development of surface oxygen or Mn-vacancies (Ov or Mnv), which makes it easier to activate oxygen to surfaceactive species and improves oxidation capacity.This explains why the MnOx/Sep-H catalysts have higher activity in the oxidation of HCHO. Figure 7C,D display high-resolution XPS spectra of Mn 3s and Mn 2p.The formula for calculating the average oxidation state (AOS) of the surface Mn in MnO x /Sep-H and MnO x /Sep-I is AOS = 8.956 − 1.126∆E, where ∆E is the splitting width of Mn 3s [17,23,54].For MnO x /Sep-H and MnO x /Sep-I, the corresponding AOS values were found to be 2.31 and 2.65, respectively.The AOS of the surface Mn for MnO x /Sep-I was close to the typical oxidation state of Mn in Mn 3 O 4 (2.67), whereas MnO x /Sep-H catalysts had a lower AOS of surface Mn (2.31), suggesting that it may form the Mn-vacancies or O-vacancies.The Mn 2p3/2 spectra, which were fitted into two peaks (Figure 7C) at around ∼642.1 eV and ∼643.2 eV, which can be attributed to Mn 3+ and Mn 4+ species, respectively [21,25,26,48].According to the XPS-41 software's fitting results, MnO x /Sep-P, MnO x /Sep-H, and MnO x /Sep-I have Mn 3+ /Mn 4+ fractions of 1.15:1, 1.67:1, and 1:1, respectively.MnO x /Sep-H has a high percentage of Mn 3+ , and the surface Mn 3+ promotes the development of surface oxygen or Mn-vacancies (O v or Mn v ), which makes it easier to activate oxygen to surface-active species and improves oxidation capacity.This explains why the MnO x /Sep-H catalysts have higher activity in the oxidation of HCHO.
The microstructure of the MnO x /Sep-H catalyst was obtained through TEM micrographs, as depicted in Figure 8.The sepiolite had a fibrous structure resembling a needle, and manganese nanoparticles were evenly dispersed on the surface of the sepiolite.The average size of these particles was less than 10 nm.Moreover, some of manganese nanoparticles are clearly loaded onto the sepiolite fiber boundaries, which may suggest a strong interaction between surface nanoparticles and sepiolite.The microstructures of various Mn 3 O 4 samples were observed by TEM.It is evident that micro-conjunctive nanoparticles make up the macroscopic morphology of MnO x /Sep-H.The lattice spacing of 0.49 nm observed on MnO x /Sep-H is associated with the (101) facets of Mn 3 O 4 , as demonstrated by high-resolution transmission electron microscopy (HRTEM) pictures [25].Using element distribution micrographs from the HAADF-STEM of MnO x /Sep-H, the elements C, Si, Mg, O, and Mn were identified and constructed the catalyst.average size of these particles was less than 10 nm.Moreover, some of manganese nanoparticles are clearly loaded onto the sepiolite fiber boundaries, which may suggest a strong interaction between surface nanoparticles and sepiolite.The microstructures of various Mn3O4 samples were observed by TEM.It is evident that micro-conjunctive nanoparticles make up the macroscopic morphology of MnOx/Sep-H.The lattice spacing of 0.49 nm observed on MnOx/Sep-H is associated with the (101) facets of Mn3O4, as demonstrated by high-resolution transmission electron microscopy (HRTEM) pictures [25].Using element distribution micrographs from the HAADF-STEM of MnOx/Sep-H, the elements C, Si, Mg, O, and Mn were identified and constructed the catalyst.(f-j) corresponding to C, Si, Mg, O, and Mn, (k) spectra from area of MnOx/Sep-H.

Catalytic Performance
Figure 9A displays the HCHO conversion over MnOx/Sep-H, MnOx/Sep-P, and MnOx/Sep-I.These catalysts have the ability to fully oxidize HCHO at varying temperatures, indicating that the manganese active components supported by sepiolite exhibit activity with different methods of preparation.The phenomenon in question could potentially be attributed to the combined action of activating manganese active components and sepiolite's efficient uptake of HCHO.However, the differences in MnOx/Sep-H, MnOx/Sep-P, and MnOx/Sep-I catalytic performance are clearly visible.With reaction conditions of 100 ppm of HCHO, 50% reaction humidity, and 6,000 mL/(g•h), the MnOx/Sep-H catalyst can totally oxidize HCHO to CO2 at 85 °C; however, it is required at 95 °C and 125 °C over MnOx/Sep-P and MnOx/Sep-I with the same reaction circumstances, respectively.MnOx/Sep-H clearly demonstrates superior catalytic performance.The Mn3O4 structure was created by manganese species, as shown by XRD, FTIR, and XPS data.Manganese species have the potential to produce Mn-vacancies (Mnv), which is favorable for the adsorption and activation of oxygen molecules and generates reactive oxygen species (ROS) that enhance catalytic performance.Furthermore, MnOx/Sep-H has the biggest pore size and least amount of variation in comparison to sepiolite among the three catalysts, although it has the smallest specific surface area.A larger pore size may be beneficial for capturing and storing HCHO.Additionally, MnOx/Sep-H demonstrated superior oxygen activation and product desorption abilities in H2-TPR, O2-TPD, and CO2-TPD, all of which aided in the promotion of HCHO oxidation.Figure 10 demonstrates the stability of oxidative degradation of 100 ppm HCHO over MnO x /Sep-H at 85 • C, 25% relative humidity, and 6000 mL/(g•h) GHSV.This indicates that the catalyst sustains the activity even when the reaction period reaches 105 h.Nevertheless, the catalytic activity gradually decreases as the reaction goes on.As the reaction goes on for more than 160 h, the catalytic activity drops to roughly 70%.This could be caused by the active components' decreased capacity to activate molecular oxygen and the sepiolite's ability to capture HCHO, as well as the continual accumulation of intermediate and target products that occupy the active sites.
Figure 9B-D illustrate the effects of the HCHO inlet concentration, reaction humidity, and space velocity (GHSV) on HCHO catalytic oxidation over MnOx/Sep-H.According to the results, at 65 °C, 80 °C, 85 °C, 105 °C, and 130 °C, with HCHO concentrations of 10, 50, 100, 150, and 200, and under reaction conditions of 6,000 mL/(g•h) and 50% reaction humidity, MnOx/Sep-H can totally oxidize HCHO.And under reaction conditions of GHSV of 6,000 mL/(g•h) and 100 ppm of HCHO, MnOx/Sep-H can oxidize HCHO entirely at 75 °C, 85 °C, 105 °C, and 150 °C at reaction humidities of 25%, 50%, 75%, and 90%.Furthermore, the HCHO may be fully oxidized at 50 °C, which exhibits the highest catalytic performance when compared to MnOx/Sep-H at the lowest temperatures of 65 °C, 85 °C, 105 °C, and 130 °C under reaction conditions of 50% reaction humidity and 100 ppm HCHO.And the oxidative degradation temperature increased with the increment in GHSV.
Figure 10 demonstrates the stability of oxidative degradation of 100 ppm HCHO over MnOx/Sep-H at 85 °C, 25% relative humidity, and 6000 mL/(g•h) GHSV.This indicates that the catalyst sustains the activity even when the reaction period reaches 105 h.Nevertheless, the catalytic activity gradually decreases as the reaction goes on.As the reaction goes on for more than 160 h, the catalytic activity drops to roughly 70%.This could be caused by the active components' decreased capacity to activate molecular oxygen and the sepiolite's ability to capture HCHO, as well as the continual accumulation of intermediate and target products that occupy the active sites.

Reaction Mechanism of HCHO Oxidation over MnOx/Sep-H
Based on the acquired data and earlier studies [43], a potential mechanism of the catalytic oxidation of HCHO on the MnOx/Sep-H catalyst is shown in Figure 11.Firstly, it involved the adsorption of HCHO and O2 onto the catalyst surface through the hydroxyl groups and d-orbital vacancies of Mn at step I, respectively.Secondly, the adsorbed O2 split to active oxygen radicals (O − ) on Mn-vacancies, simultaneously converting Mn 3+ to Mn 4+ at step II.In this stage, oxygen radicals (O − ) attacked the C=O bond of adsorbed HCHO, converting it to HCOOH (formate).Thirdly, HCOOH was further oxidized to CO2 and H2O by another oxygen radical (O − ) at step III and step IV which has the ability to attack the C-O bond of formate species.Finally, the Mn 4+ converts to Mn 3+ , and the occupied hydroxyl groups and d-orbital vacancies of Mn were also released at step V, providing exceptional stability for recycling.Therefore, with the exception of physisorption on sepiolite nanofibers and oxidation on MnOx active sites, the catalytic oxidation of HCHO

Reaction Mechanism of HCHO Oxidation over MnO x /Sep-H
Based on the acquired data and earlier studies [43], a potential mechanism of the catalytic oxidation of HCHO on the MnO x /Sep-H catalyst is shown in Figure 11.Firstly, it involved the adsorption of HCHO and O 2 onto the catalyst surface through the hydroxyl groups and d-orbital vacancies of Mn at step I, respectively.Secondly, the adsorbed O 2 split to active oxygen radicals (O − ) on Mn-vacancies, simultaneously converting Mn 3+ to Mn 4+ at step II.In this stage, oxygen radicals (O − ) attacked the C=O bond of adsorbed HCHO, converting it to HCOOH (formate).Thirdly, HCOOH was further oxidized to CO 2 and H 2 O by another oxygen radical (O − ) at step III and step IV which has the ability to attack the C-O bond of formate species.Finally, the Mn 4+ converts to Mn 3+ , and the occupied hydroxyl groups and d-orbital vacancies of Mn were also released at step V, providing exceptional stability for recycling.Therefore, with the exception of physisorption on sepiolite nanofibers and oxidation on MnO x active sites, the catalytic oxidation of HCHO on MnO x /Sep-H catalysts is therefore greatly enhanced by the synergistic impact of conversion between Mn 3+ to Mn 4+ and hydroxyl groups.
on MnOx/Sep-H catalysts is therefore greatly enhanced by the synergistic impact of conversion between Mn 3+ to Mn 4+ and hydroxyl groups.

Synthesis of MnOx/Sepolite-I Nanoparticles by Impregnation Method
The MnO2/Sepolite-I was obtained by the impregnation method as follows.First, sepiolite fibers were dispersed in deionized aqueous solution at a mass ratio of 1:40, and the mixture was stirred at room temperature for 24 h.Then, a certain amount of potassium permanganate was dispersed in the aqueous solution and fully dissolved, and then, it was slowly added dropwise to the sepiolite fiber aqueous suspension at a rate of 2 s/drop and stirred under reflux for 6 h at 105 °C.Subsequently, a quantitative amount of L-ascorbic acid was dissolved and slowly added dropwise to the above mixture at a speed of 2 s/drop and stirred for 6 h at 105 °C under reflux.After the reaction, the mixture was centrifuged to retain the solid; then, the solid was washed three times with absolute ethanol and water and dried overnight at 80 °C.Lastly, the dried solid was ground into powder and roasted at 350 °C for 5 h to obtain MnO2/Sepolite by the impregnation method.The sample was labeled as MnO2/Sepolite-I.

Synthesis of MnOx/Sepolite-P Nanoparticles by Precipitation Method
The synthesis of MnO2/Sepolite-P nanoparticles by the precipitation method are as follows.The Sep was dispersed in a deionized aqueous solution following the method described above.A certain amount of potassium permanganate solution, L-ascorbic acid solution, and 5M NaOH solution were prepared.Both solutions are added dropwise to the sepiolite dispersion at the same time.And the pH alkalinity was adjusted to ensure that the solution was stable (pH = 10).It was stirred for 12 h at 105 °C under reflux.After the reaction, the mixture was centrifuged and washed three times with absolute ethanol and water, dried overnight at 105 °C, and finally ground into powder, and the powder

Synthesis of MnO x /Sepolite-I Nanoparticles by Impregnation Method
The MnO 2 /Sepolite-I was obtained by the impregnation method as follows.First, sepiolite fibers were dispersed in deionized aqueous solution at a mass ratio of 1:40, and the mixture was stirred at room temperature for 24 h.Then, a certain amount of potassium permanganate was dispersed in the aqueous solution and fully dissolved, and then, it was slowly added dropwise to the sepiolite fiber aqueous suspension at a rate of 2 s/drop and stirred under reflux for 6 h at 105 • C. Subsequently, a quantitative amount of L-ascorbic acid was dissolved and slowly added dropwise to the above mixture at a speed of 2 s/drop and stirred for 6 h at 105 • C under reflux.After the reaction, the mixture was centrifuged to retain the solid; then, the solid was washed three times with absolute ethanol and water and dried overnight at 80 • C. Lastly, the dried solid was ground into powder and roasted at 350 • C for 5 h to obtain MnO 2 /Sepolite by the impregnation method.The sample was labeled as MnO 2 /Sepolite-I.

Synthesis of MnO x /Sepolite-P Nanoparticles by Precipitation Method
The synthesis of MnO 2 /Sepolite-P nanoparticles by the precipitation method are as follows.The Sep was dispersed in a deionized aqueous solution following the method described above.A certain amount of potassium permanganate solution, L-ascorbic acid solution, and 5M NaOH solution were prepared.Both solutions are added dropwise to the sepiolite dispersion at the same time.And the pH alkalinity was adjusted to ensure that the solution was stable (pH = 10).It was stirred for 12 h at 105 • C under reflux.After the reaction, the mixture was centrifuged and washed three times with absolute ethanol and water, dried overnight at 105 • C, and finally ground into powder, and the powder was roasted at 350 • C for 5 h to obtain MnO 2 /Sepolite by the precipitation method.The sample was labeled as MnO 2 /Sepolite-P.

Synthesis of MnO x /Sepolite-H Nanoparticles by Hydrothermal Method
The synthesis of MnO x /Sepiolite-H nanoparticles by the hydrothermal method was similar to the initial process of the impregnation method.The potassium permanganate solution was slowly added dropwise to the sepiolite fiber aqueous suspension at a rate of 2 s/drop and stirred at room temperature.Subsequently, the L-ascorbic acid was slowly added dropwise to the above mixture at a speed of 2 s/drop and stirred for 1 h at room temperature.And then, the resulting solution was transferred to a Teflon-lined autoclave (100 mL in capacity) and maintained at 140 • C for 12 h.After cooling, the resulting precipitate was isolated by filtration and washed with deionized water and ethanol three times.The sample was then dried at 105 • C overnight.The powder was roasted at 350 • C for 5 h to obtain MnO 2 /Sepolite by the hydrothermal method.The sample was labeled as MnO x /Sep-H.

Catalysts Characterization
X-ray diffraction (XRD) patterns of three samples were recorded on an X-ray diffractometer (Ultima IV, CuKα radiation 40 kV/40 mA, the step of 0.1 • /s, 2θ = 5-90 • ).Fouriertransform infrared spectra (FTIR) of samples were collected on a Nicolet 6700 spectrometer (400 to 4000 cm −1 , 1.0 cm −1 interval).The N 2 adsorption-desorption isotherms, surface area and pore size distribution of as-prepared samples were characterized on a Quantachrome NOVA-2200e based on Brunauer-Emmett-Teller (BET) model [50,51].Transmission electron microscope (TEM) and elemental mapping micrographs were obtained through a transmission electron microscope (Talos F200X (FEI)).The samples were dispersed in alcohol with ultrasonic treatment for 20 min, and then uniformly dispersed in copper mesh for drying.X-ray photoelectron spectroscopy (XPS) measurements of samples was tested on instrument (Thermo Escalab 250Xi, Thermo Scientific, Waltham, MA, USA) equipped with an Al Kα X-ray source to determine the elemental composition and chemical states of the elements.The XPS spectra were deconvoluted by using a commercially available data-fitting program (XPSPEAK-41 software 2.0, Informer Technologies, Inc., Los Angeles, CA, USA) after a Shirley background subtraction procedure, and the binding energies were calibrated by referencing the C 1s at 284.8 eV.
H 2 temperature-programmed reduction (H 2 -TPR) and CO 2 temperature-programmed desorption (CO 2 -TPD) profiles were obtained from an Auto Chem II 2920 instrument equipped with a thermal conductivity detector (TCD).For H 2 -TPR, 0.1 g catalysts were heated to 200 • C with N 2 gas (30 mL/min, 10 • C/min) for 30 min.After it cooled to room temperature, the catalyst was heated in pure H 2 from 20 • C to 900 • C at a heating rate of 10 • C/min.For CO 2 -TPD, 0.1 g catalysts were heated to 200 • C with N 2 gas (30 mL/min, 10 • C/min) for 30 min.After it cooled to room temperature, the sample was cleaned with pure CO 2 or NH 3 (50 mL/min) at room temperature, and then cleaned with 50 mL/min N 2 for 30 min to remove the physically adsorbed CO 2 or NH 3 .Once the baseline was stable, the temperature rose from room temperature to 900 • C at a rate of 10 • C/min.
The O 2 -TPD analysis of samples was measured on Auto Chem II 2920, 0.1 g catalysts were placed in a u-shaped tube and heated from room temperature to 200 • C and pretreated with He (50 mL/min) for 1 h to remove the surface-adsorbed water and then cooled to room temperature.The samples were then cleaned with 50 mL/min O 2 at room temperature for 1 h and then cleaned with 50 mL/min He for 30 min to remove the physically adsorbed oxygen.Once the baseline was stable, the temperature rose from room temperature to 900 • C with a heating rate of 10 • C/min.

Catalyst Evaluation
The catalytic activities of samples for HCHO oxidation were performed in a continuousflow fixed-bed reactor [43].0.5 g catalysts (40-60 mesh) were loaded in a quartz reactor (6 mm).Gaseous HCHO was generated by purging paraformaldehyde with air.Another stream of air mixed with the gaseous HCHO.The HCHO concentration was controlled by adjusting the flow rate by a mass flow controller (GFC17A, Aalborg, New York, NY, USA.) and the temperature of incubator.The test route of the catalytic activities is shown in Scheme 1.

Figure 1 .
Figure 1.The XRD patterns of Sep, MnO x /Sep-P, MnO x /Sep-I, and MnO x /Sep-H.

Figure 3 .
Figure 3.The N2 adsorption-desorption curve and pore size distribution patterns of catalysts.(A) the N2 adsorption-desorption curve; (B) the pore size distribution curve.

Figure 3 .
Figure 3.The N 2 adsorption-desorption curve and pore size distribution patterns of catalysts.(A) the N 2 adsorption-desorption curve; (B) the pore size distribution curve.

Figure 4 .
Figure 4.The H2-TPR profiles of MnOx/Sep-H, MnOx/Sep-P, and MnOx/Sep-I.To learn more about the desorption patterns and chemical reactions of various oxygen species, O2-TPD was carried out.Two peaks were seen at 100-300 °C and 600-800 °C in the O2-TPD profiles of the MnOx/Sep-H, MnOx/Sep-P, and MnOx/Sep-I catalysts in Figure 5.The adsorbed oxygen species (O2 − , O 2− or O − ) at the surface was the first desorption peak and is active for the full oxidation of HCHO.The bulk lattice oxygen (O 2− ) is responsible for other desorption peaks.The desorption temperatures of MnOx/Sep-H, MnOx/Sep-P, and MnOx/Sep-I were 137 °C, 141 °C, and 170 °C, respectively, for surface-adsorbed oxygen species.Furthermore, the specific gravity of oxygen desorption for MnOx/Sep-H, MnOx/Sep-P, and MnOx/Sep-I was determined to be 1.4:1.3:1.2.According to these findings, MnOx/Sep-H, MnOx/Sep-P, and MnOx/Sep-I has a higher concentration of activated oxygen species than the others, which is advantageous for the oxidation of HCHO[48].

Figure 4 .
Figure 4.The H 2 -TPR profiles of MnO x /Sep-H, MnO x /Sep-P, and MnO x /Sep-I.

Figure 5 .
Figure 5.The O 2 -TPD profiles of MnO x /Sep-H, MnO x /Sep-P, and MnO x /Sep-I.

Figure 6 .
Figure 6.The CO2-TPD profiles of MnOx/Sep-H, MnOx/Sep-P, and MnOx/Sep-I.The chemical composition of MnOx/Sep-H, MnOx/Sep-P, and MnOx/Sep-I is studied by XPS (Figure7).In Figure7A, the wide-scan spectra indicates that Si, O, C, and Mn elements exist in three catalysts.And the C 1s and Si 2p high-resolution XPS spectra are shown in FiguresS1 and S2(Supplementary Materials).Comparing MnOx/Sep-P and MnOx/Sep-I, the shift of binding energy position appeared in MnOx/Sep-H, indicating a stronger interaction between the nanoparticles and the carrier.In addition, the position near 103.9 eV and 103.2 eV correspond to SiO2 and Si-O bonds, but the proportion of Si-O bonds in MnOx/Sep-H significantly increases, suggesting that there may be a Si-O-metal structure on the surface of MnOx/Sep-H, such as Si-O-Mn bonds[38,39].These visualizations suggest a robust interaction between the active ingredients and the carrier, which is advantageous for the dispersion of active components and the formation of new bonds.Three different types of oxygen species were found in the core level O 1s spectra (Figure7B) for MnOx/Sep-H, MnOx/Sep-P, and MnOx/Sep-I.These species include the following: (i) OL or Oads (530.0-530.8eV); (ii) surface hydroxyl species (OOH, 532.6-532.8eV); and (iii) surface water (OH2O, 534.5-534.9eV)[26,48,53].Furthermore, according to the XPS-41 software's fitting results, the proportions of OL or Oads surface OOH and OH2O in MnOx/Sep-P were 6.7%, 89.6%, and 3.8%, were 10%, 85.6%, and 4.3% in MnOx/Sep-H, and were 16.4%, 80%, and 3.7% in MnOx/Sep-I, respectively.More surface hydroxyl groups

Figure 6 .
Figure 6.The CO 2 -TPD profiles of MnO x /Sep-H, MnO x /Sep-P, and MnO x /Sep-I.The chemical composition of MnO x /Sep-H, MnO x /Sep-P, and MnO x /Sep-I is studied by XPS (Figure7).In Figure7A, the wide-scan spectra indicates that Si, O, C, and Mn elements exist in three catalysts.And the C 1s and Si 2p high-resolution XPS spectra are shown in FiguresS1 and S2(Supplementary Materials).Comparing MnO x /Sep-P and MnO x /Sep-I, the shift of binding energy position appeared in MnO x /Sep-H, indicating a stronger interaction between the nanoparticles and the carrier.In addition, the position near 103.9 eV and 103.2 eV correspond to SiO 2 and Si-O bonds, but the proportion of Si-O bonds in MnO x /Sep-H significantly increases, suggesting that there may be a Si-O-metal structure on the surface of MnO x /Sep-H, such as Si-O-Mn bonds[38,39].These visualizations suggest a robust interaction between the active ingredients and the carrier, which is advantageous for the dispersion of active components and the formation of new bonds.Three different types of oxygen species were found in the core level O 1s spectra (Figure7B) for MnO x /Sep-H, MnO x /Sep-P, and MnO x /Sep-I.These species include the following: (i) O L or O ads (530.0-530.8eV); (ii) surface hydroxyl species (O OH , 532.6-532.8eV); and (iii) surface water (O H2O , 534.5-534.9eV)[26,48,53].Furthermore, according to the XPS-41 software's fitting results, the proportions of O L or O ads surface O OH and O H2O in MnOx/Sep-P were 6.7%, 89.6%, and 3.8%, were 10%, 85.6%, and 4.3% in MnO x /Sep-H, and were 16.4%, 80%, and 3.7% in MnO x /Sep-I, respectively.More surface hydroxyl groups were present in three catalytic materials, which was advantageous for the adsorption of HCHO.However, MnO x /Sep-I and MnO x /Sep-H have greater amounts of O L or O ads than MnO x /Sep-P, suggesting that these two catalysts may have more reactive oxygen species and the capacity to activate oxygen, which would aid in the conversion of adsorbed HCHO.On the other hand, MnO x /Sep-H exhibits positive-shift binding energies (about 0.5 eV) of the O L or O ads in contrast to MnO x /Sep-P and MnO x /Sep-I, suggesting the presence of interactions among Sep, O species, and Mn species.In conclusion, the abundance of surface O OH and O L or O ads along with the robust interaction between the Sep, O species, and Mn species all work together to enhance the degradation of HCHO by MnO x /Sep-H catalysts.
) for MnO x /Sep-H, MnO x /Sep-P, and MnO x /Sep-I.These species include the following: (i) O L or O ads (530.0-530.8eV); (ii) surface hydroxyl species (O OH , 532.6-532.8eV); and (iii) surface water (O H2O , 534.5-534.9eV)[26,48,53].Furthermore, according to the XPS-41 software's fitting results, the proportions of O L or O ads surface O OH and O H2O in MnOx/Sep-P were 6.7%, 89.6%, and 3.8%, were 10%, 85.6%, and 4.3% in MnO x /Sep-H, and were 16.4%, 80%, and 3.7% in MnO x /Sep-I, respectively.More surface hydroxyl groups were present in three catalytic materials, which was advantageous for the adsorption of HCHO.However, MnO x /Sep-I and MnO x /Sep-H have greater amounts of O L or O ads than MnO x /Sep-P, suggesting that these two catalysts may have more reactive oxygen species and the capacity to activate oxygen, which would aid in the conversion of adsorbed HCHO.On the other hand, MnO x /Sep-H exhibits positive-shift binding energies (about 0.5 eV) of the O L or O ads in contrast to MnO x /Sep-P and MnO x /Sep-I, suggesting the presence of interactions among Sep, O species, and Mn species.In conclusion, the abundance of surface O OH and O L or O ads along with the robust interaction between the Sep, O species, and Mn species all work together to enhance the degradation of HCHO by MnO x /Sep-H catalysts.HCHO.On the other hand, MnOx/Sep-H exhibits positive-shift binding energies (about 0.5 eV) of the OL or Oads in contrast to MnOx/Sep-P and MnOx/Sep-I, suggesting the presence of interactions among Sep, O species, and Mn species.In conclusion, the abundance of surface OOH and OL or Oads along with the robust interaction between the Sep, O species, and Mn species all work together to enhance the degradation of HCHO by MnOx/Sep-H catalysts.

Figure
Figure7C,D display high-resolution XPS spectra of Mn 3s and Mn 2p.The formula for calculating the average oxidation state (AOS) of the surface Mn in MnOx/Sep-H and MnOx/Sep-I is AOS=8.956-1.126ΔE,where ΔE is the splitting width of Mn 3s[17,23,54].For MnOx/Sep-H and MnOx/Sep-I, the corresponding AOS values were found to be 2.31 and 2.65, respectively.The AOS of the surface Mn for MnOx/Sep-I was close to the typical oxidation state of Mn in Mn3O4 (2.67), whereas MnOx/Sep-H catalysts had a lower AOS of surface Mn (2.31), suggesting that it may form the Mn-vacancies or O-vacancies.The Mn 2p3/2 spectra, which were fitted into two peaks (Figure7C) at around ∼642.1 eV and ∼643.2 eV, which can be attributed to Mn 3+ and Mn 4+ species, respectively[21,25,26,48].

Figure 8 .
Figure 8. TEM, STEM, and element distribution micrographs of selected samples.(a-d) TEM micrograph of MnO x /Sep-H, (e) HAADF-STEM of MnO x /Sep-H with element distribution micrographs (f-j) corresponding to C, Si, Mg, O, and Mn, (k) spectra from area of MnO x /Sep-H.

Figure
Figure 9A displays the HCHO conversion over MnO x /Sep-H, MnO x /Sep-P, and MnO x /Sep-I.These catalysts have the ability to fully oxidize HCHO at varying temperatures, indicating that the manganese active components supported by sepiolite exhibit activity with different methods of preparation.The phenomenon in question could poten-

Figure 9 .
Figure 9.The performance of catalytic oxidation of HCHO.(A) The catalytic activity of MnO x /Sep-H, MnO x /Sep-P, and MnO x /Sep-I (reaction conditions: 100 ppm of HCHO, WHSV 6000 mL/(g•h)); (B) Effects of different HCHO concentrations on catalytic oxidation of HCHO over MnO x /Sep-H (reaction conditions: RH of 50% and GHSV 6000 mL/(g•h)); (C) Effects of different reaction humidities on catalytic oxidation of HCHO over MnO x /Sep-H (reaction conditions: 100 ppm of HCHO and GHSV 6000 mL/(g•h)); (D).Effects of different GHSV on MnO x /Sep-H catalyst activity.(reaction conditions: 100 ppm of HCHO and RH of 50%).

Figure 10 .
Figure 10.The stability of oxidative degradation of 100 ppm HCHO over MnO x /Sep-H at 85 • C, GHSV of 6000 mL/(g•h) and RH of 25%.

Figure 11 .
Figure 11.Schematic of proposed mechanism for HCHO removal process over MnOx/Sep-H catalyst.HCHO.

Figure 11 .
Figure 11.Schematic of proposed mechanism for HCHO removal process over MnO x /Sep-H catalyst.HCHO.

Table 1 .
Physical parameters of the mesoporous materials and surface element composition of the samples.

SBET (m 2 /g) a Vpore (cm 3 /g) a Dpore (nm) a Mn 3+ /Mn 4+ b OL or Oads b AOS Values c
The SBET, Vpore, and Dpore calculated by the BET method.bSurfaceelement molar ratio calculated by peak areas of XPS.cThe AOS values calculated by peak areas of XPS. a

Table 1 .
Physical parameters of the mesoporous materials and surface element composition of the samples.

a Mn 3+ /Mn 4+ b O L or O ads b AOS Values c
a