Research on the Mechanism of Synergistic Treatment of VOCs–O3 by Low Temperature Plasma Catalysis Technology

In this research, xylene was utilized as a simulated gas, γ-Al2O3 pellets were selected as catalyst carriers, and FeOx, MnOx, CeOx, and CuOx were used as active components to analyze the synergistic treatment of VOCs–O3 by low-temperature plasma combined with supported catalysts. Different metal oxides and other factors influence the synergistic treatment of VOCs–O3. The results showed that the catalytic effect of Fe–Mn/γ-Al2O3 prepared by the equivalent volumes of consecutive impregnation method was better than that of Fe–Mn/γ-Al2O3 prepared by co-impregnation method. When combined with low temperature plasma technology, high-energy electron collision reaction and oxidation reaction between free radicals occurs, which played a synergistic role in the degradation of VOCs–O3. The total removal rate of xylene was 94.88%, and the depletion rate of ozone was 84.1%.


Introduction
In recent years, with the strengthening of global supervision on environmental pollution, particulate matter pollution has been improved significantly.However, with the concentration of Ozone (O 3 ) increasing, which become another air pollutant at present and need urgent action.The international community pointed out about the promotion of ozone pollution control and coordinated emission reduction of volatile organic pollutants (VOCs).Ozone in the troposphere is the main product of the photochemical reaction, which is derived from the secondary pollution process of VOCs and nitrogen oxides (NO x ) under sunlight [1].In the past, researchers lacked awareness of ozone pollution in VOCs treatment, the misuses of low-temperature plasma technology and photocatalytic oxidation technology led to ozone pollution.The effect of these technologies on VOCs is very obvious, but ozone pollution shows a rapid increase and spread.The relationship between ozone and its precursors is very complex and difficult to control, which is the main reason for ozone pollution [2].Therefore, it is necessary to work hard on reducing ozone precursors in order to suppress the generation of ozone.VOCs are important precursors for the formation of ozone pollution.Ozone pollution is a serious problem, and VOCs are the difficulty and key to ozone control.In order to meet increasingly stringent ozone emission standards and to reduce the harmful effects of VOCs and ozone.Its particularly important to find a method that can deal with VOC and ozone pollution effectively.
Low temperature plasma catalytic technology is currently one of the most promising technologies, but the use of this technology to remove VOCs is often associated with the generation of ozone.Ozone is not only a by-product of the low-temperature plasma catalytic system, but also a by-product of the active oxidising substances produced, which are mainly derived from the discharge process of the air plasma [3].The utilization rate of ozone in a single low temperature plasma system is very low, and the emission concentration of ozone in the system is high, ranging from 100 to 1000 ppm [4].In order to reduce ozone emissions, the purpose of suppressing the production of ozone and improving the utilisation rate of ozone is generally achieved by using a catalyst.The reaction mechanism of low-temperature plasma is divided into electron avalanche process, particle collision process and energy conversion process [5].The technology has high removal efficiency for the target pollutants, but its disadvantages are reflected in low energy utilization, incomplete degradation of the target pollutants, and easy generation of secondary pollutants such as ozone, aerosols, and nitrogen oxides [6].It is difficult to apply on a large scale.Consequently, the combination of low-temperature plasma technology.Catalytic technology is a very promising method [7], which provides conditions for reactions that are difficult to occur at room temperature [8].It can effectively degrade various VOCs (such as containing aldehydes and benzene series, etc.) [9].Fan et al. [10] used low-temperature plasma to treat p-xylene, ethylene, and trichloroethylene, and found that the treatment process was often accompanied by the production of ozone and NO 2 .Aerts et al. [11] found that when DBD was used to treat high-concentration ethylene, under the condition of energy density of 2.4 J/cm 3 , the amount of ozone generated was as high as 3043 ppm, a large amount of NO 2 , N 2 O, and HNO 2 were also generated.Liu Yanghaichao et al. [12] found that the MnO x / Al 2 O 3 catalyst contributed to the decomposition of ozone when using dielectric barrier discharge (DBD) to degrade chlorobenzene.They discussed the decomposition mechanism of chlorobenzene in the coupled reactor.Most researches at home and abroad use lowtemperature plasma technology to treat VOCs, which will generate ozone, and there is little research on how to reduce the generation of ozone.Therefore, it is required to study the VOCs treatment while considering the O 3 reduction.The synergistic control technology of VOCs and ozone is still lagging behind [13].The relationship between catalyst activity and its own properties has not been clearly elucidated.Especially the mechanism research of ozone is insufficient.In order to better solve the above problems, it is necessary to conduct research on the mechanism of co-processing VOCs-O 3 with low temperature plasma technology.Based on the research results of the previous group, this study combined the plasma discharge technology with the catalytic technology under the optimal operating parameters of the DBD reactor (oxygen content, discharge power, xylene inlet concentration).Clearly described the experimental procedures and evaluation methods.Selection of the optimal catalyst and in-depth study of the effect of low-temperature plasma catalysis on the synergistic removal of VOCs-O 3 .Therefore, combined with the national concern about O 3 , it is very meaningful to study the use of low-temperature plasma to remove VOCs while paying attention to O 3 emissions and synergistic treatment of VOCs-O 3 , which is a strong guide to the development of realistic environmental protection.
VOCs are the main cause of ozone pollution, and ozone pollution is the result of excessive VOCs emissions.In this paper, low-temperature plasma discharge technology combined with catalytic technology for the synergistic treatment of VOCs and O 3 with xylenes as the target pollutant was explored in depth and mechanically.Supported bimetallic catalysts were prepared separately with different metal oxides as an active component.The synergistic treatment effect of low temperature plasma combined with M x O y /γ-Al 2 O 3 metal oxide supported catalysts on VOCs-O 3 was studied, and two optimal supported single metal catalysts were screened out.Two different loading methods were used to make bimetallic catalysts.The oxide-supported catalyst X-Y/γ-Al 2 O 3 (X and Y are FeO x , MnO x , CeO x , CuO x , respectively) was used to explore its co-processing performance for VOCs-O 3 , and its reaction mechanism was analyzed by characterization of the catalyst.Reducing ozone generation and controlling the emission of VOCs had far-reaching guiding significance for the development of environmental protection.Reducing ozone production and controlling VOCs emission has far-reaching guiding significance for the development of environmental protection, and opens a new idea of collaborative air pollution management.

Catalyst Preparation
In Fig. 1, copper nitrate, cerium nitrate, iron nitrate and manganese nitrate were selected from the impregnation solution.Al 2 O 3 has the advantage of a large specific surface area and a porous structure, which allows the active components to be uniformly dispersed.This material is resistant to high temperatures and corrosion, and thus has been the most studied in related experiments [14].The selected transition metal oxides usually have special energy band structure and rich adjustable valence states.Not only are they long-lived and easily regenerated, but they are also highly malleable, and the designed dimensions and morphologies can be easily obtained [15].M x O y /γ-Al 2 O 3 supported catalysts were prepared by an equal volume impregnation method.After soaking activated alumina in ionized water for several hours, the catalyst was activated in an oven at 105 °C for 3 h.The catalyst was immersed in an equal volume of impregnating solution, placed in an ultrasonic cleaner for 30 min to make the mixture homogeneous and removed, and then left to stand at room temperature for 24 h.The catalyst was then dried in an oven at 80 °C for 6 h.The dried metal oxide-loaded catalysts were activated by roasting in a muffle furnace at 450 °C for 4 h.The catalysts were then placed in a muffle furnace at 450 °C for 4 h to activate.The X@Y/γ-Al 2 O 3 catalyst was obtained by the co-impregnation method, the X-Y/γ-Al 2 O 3 catalyst was obtained by consecutive impregnation method.
Depending on the location of the catalyst, low-temperature plasma-catalysis technologies are divided into Plasma-Driven Catalysis (PDC) and Plasma-Enhanced Catalysis (PEC) [16].As shown in Fig. 2 PDC is where plasma action and catalytic reactions occur in the same space and take place simultaneously.The interaction of the plasma with the catalytic reaction allows the active particles to react directly on the catalyst, which has the advantage of enhancing the discharge and prolonging the contact time and reaction time of the organics on the catalyst surface [17].PEC has the advantage of degrading VOCs while effectively reducing the generation of by-products O 3 and NO x due to the posterior placement of the catalyst [18].PEC lacks the synergistic reinforcement of the plasma and catalyst compared to the PDC.It is lower than that of the one-stage system [19] (Fig. 2).

Experimental Device
The flow chart was divided into three parts: gas distribution system, low temperature plasma reaction system and monitoring system, as shown in Fig. 3.The power supply of the low temperature plasma reaction system was the CTP-2000K experimental power supply of Nanjing Suman.The power was 500 W, the output voltage was 0-30 kV, and the input voltage is 220V.This power supply consisted of power supply host and regulator, output frequency between 5 and 20kHz, center frequency was 10 kHz.Based on the results of a large number of previous studies by the group, the experimental conditions were set as follows: the discharge power was 70 W, the oxygen flux was 15%, the initial concentration was 600 ppm, with a total inlet flow rate of 200 ml/min, a nitrogen flow rate of 160 ml/min 1 3 [20].The gas distribution system was run for 2 h before the experiment started, the temperature of the water bath was set to 25 °C, and the temperature of the heating tape was set to 27 °C, so that the xylene output concentration could be maintained at a relatively stable state.
The removal rate of xylene was calculated by the formula: In this formula: [xylene] inlet representd the xylene concentration before the reaction; [xylene] outlet represents the xylene concentration after the reaction.The consumption and consumption rate of ozone was calculated by the following formula: In the formula: [Ozone] blank represents the amount of ozone generated after only air passes through the low temperature plasma; [Ozone] xylene represents the amount of ozone produced by passing the gas mixture through the low temperature plasma.
The ideal end products of xylene degradation by low-temperature plasma co-catalyst are H 2 O and CO 2 , but various types of intermediate products and by-products would exist in the C CO represented represented the concentration of CO, in mg/m 3 ; C CO 2 represented the concentration of CO 2 , in mg/m 3 ; [xylene] inlet represented the concentration of pollutants ( 4)

Characterization
XRD The X-ray photoelectron spectrometer used in this experiment was produced by Thermo Fisher Scientific and its model was 250xi.Jade software was commonly used to process and analyze the XRD data, and the spectrum of the catalyst was corresponding to the PDF standard card, and its material composition was qualitatively analyzed.
BET The experiment used a fully automatic specific surface area and micropore pore analyzer with the JW-BK122W instrument model.Using nitrogen as the analytical gas and measuring the saturation pressure, and the specific surface area of the catalyst sample was calculated using the equation [21].
FT-IR FT-IR could determine and identify the nature and content of compounds according to the generated spectrum.Fourier infrared spectrometer with model VERTEX 70 is used.
XPS Thermo Fisher 250Xi, Thermo Fisher K-Alpha, USA.The chemical composition and relative content of the composite catalyst surface were qualitatively and semi-quantitatively analyzed based on the measurement of the deorbital photoelectrons combined with the Auger electron energy distribution [22].

Results and Discussion
The Effect of Supported Metal Oxide Catalysts on VOCs-O 3 Removal 8% FeO x , MnO x , CeO x and CuO x were loaded onto the spare γ-Al 2 O 3 by the equivalent volume impregnation method, respectively.The catalysts were placed into the reaction system to study the effect of different metal oxides on the synergistic treatment of VOCs-O 3 .Figure 4a showed the addition of the supported catalyst greatly improved the xylene removal rate.The xylene removal rate could reach more than 75% within 38th minute.The removal rate of xylene was up to 88.5%.Combined with Fig. 4b, it could be seen that the FeO x /γ-Al 2 O 3 catalyst has the best removal effect of p-xylene, and the total removal rate reached 84.63%.Compared with the single plasma removal of xylene, the total efficiency was increased by 20.22%.The low-temperature activity of the catalyst was enhanced by doping with metal oxides, resulting in an increase in the active oxygen content, which was consistent with the XPS characterization results.Due to the strong mobility of adsorbed oxygen, it played an important role in the performance and activity of the catalyst.The electrons in the electron orbits of transition metal elements transite to generate electron holes, which was very beneficial to the formation of the catalyst surface.Various unsaturated chemical bonds generate adsorbed oxygen, enhancing its ability to destabilize xylene chemical bonds [23].
In the Fig. 4c, the green area represented the amount of ozone generated when xylene was not introduced into the reaction system and the catalyst carrier was not loaded with active components.The red area was the amount of ozone generated when the catalyst was loaded with active components and no xylene was introduced into the system.Therefore the distance between the green and blue box lines represented the actual consumption of ozone after loading with active components.After the addition of the single metal oxide supported catalyst, the ozone content produced by the addition of xylene and the absence of xylene was significantly reduced.Combined with Fig. 4d, the single metal oxide supported catalyst had a significant inhibitory effect on the generation of ozone, making the ozone consumption rate to reach more than 75%, and the MnO x /γ-Al 2 O 3 catalyst made the ozone consumption rate to reach at the highest level.It reached 84.68%, and its ozone consumption was 351.77 ppm.
Among transition metal oxides, p-type semiconductor metal oxides had higher ozone decomposition activity than n-type semiconductor metal oxides, which might be due to the fact that p-type semiconductor metal oxides had higher ozonolysis activity than n-type semiconductor metal oxides.It was easier to accept electrons, hence, MnO x had the highest ozone depletion rate and consumption.Secondly, the increase of metal oxide resulted to the reduction of the activation energy of the whole reaction system, accelerates the xylene decomposition process and greatly increases the consumption of ozone.Therefore, the ozone consumption rate greatly improved.In summary, the removal efficiency of xylene from high to low was FeO The ozone depletion rates from high to low was: The single metal oxide catalysts FeO x /γ-Al 2 O 3 and MnO x /γ-Al 2 O 3 with the best xylene removal effect and ozone inhibition effect were screened out.The bimetallic oxide supported catalyst was prepared by the equivalent volume co-impregnation and the equivalent volume consecutive impregnation method, and the synergistic treatment effect of p-xylene and ozone was studied.The catalyst prepared by the equal volume co-impregnation method was named Fe@Mn/γ-Al 2 O 3 , and the catalyst prepared by the equivalent volume consecutive impregnation method was named Fe-Mn/γ-Al 2 O 3 .Figure 4e showed the removal rate of xylene could be further improved by the bimetallic oxide supported catalyst, among which the Fe-Mn/γ-Al 2 O 3 catalyst had the highest removal efficiency of xylene.Figure 4f showed the total removal rate reached 94.88%, and the multi-component composite metal oxide catalyst had an inhibitory effect on the generation of ozone.Combining Fig. 4g and Fig. 4h, it was found that the Fe-Mn/γ-Al 2 O 3 catalyst could make the ozone consumption reached 363.08 ppm, and the ozone consumption rate reached 84.1%, which was 42% higher than that of the single low-temperature plasma system.

COx Selectivity and by-Product Analysis
The products of low temperature plasma catalytic degradation of xylene were analyzed by GC-MS, which were mainly divided into two categories: one category was the products containing benzene rings, such as benzene, benzaldehyde and methyl benzophenone.Another class of substances was the small molecular products formed after the cleavage of the benzene ring, including alkanes, aldehydes, ketones and some nitrogen-containing substances.Some of these by-products were deposited on the catalyst surface and the reactor wall, and were yellow oily substances with unpleasant like bitter almonds, and some of them exist in the reactor exhaust in the form of aerosols.In the process of measuring CO x selectivity, in order to ensure the accuracy of the experiment, it was necessary to wait for the removal effect of low-temperature plasma to stabilize p-xylene, and then measure the concentrations of CO and CO 2 and calculate their selectivity.The completeness of xylene degradation could be evaluated by calculating the selectivity of carbon oxides.The results were shown in Fig. 5.
After adding the catalyst carrier, the selectivity of COx was increased about 18%, reaching 71% removal.The green part in the figure was approximately an isosceles triangle, indicating that the change of CO residual was not obvious, indicating that the change of CO 2 dominates.The whole pattern distribution extended to the left, indicating that the Fe-Mn/γ-Al 2 O 3 catalyst played a major role in the influence of CO x .With the introduction of the catalyst, a large amount of strong oxidizing substances, such as ozone and active oxygen atoms, were generated in the system, which made CO more easily oxidized to CO 2 .Accoringly, so the remaining amount of CO in the exhaust gas remained basically unchanged.The introduction of iron and manganese active substances increased the decomposition of ozone in the gas phase, produced more O atoms and other active particles, and increased the deep oxidation process of xylene.Although the γ-Al 2 O 3 carrier could play the above role, its effect was weak, and because it would adsorb some COx and convert it into carbonate substances, these substances cannot be completely converted into gaseous CO x .Thus, Fe-Mn/γ-The Al 2 O 3 catalyst could increase the interfacial electron transfer during the reaction and generate more active species, which played a key role in the increase of CO and CO 2 selectivity.

BET
The activity of the catalyst was mainly affected by the chemical structure of the catalyst [24].From Table 1, it could be seen that the specific surface area order of the seven catalysts was: γ- The results showed that: (1) the specific surface area of the catalyst after loading the active components was reduced by about 35% compared with that without loading; (2) the pore volume decreased to a certain extent; (3) the average pore size of the adsorption was enlarged to varying degrees.The above three points could indicate that the active components of the catalyst had entered the pore structure of the carrier.
It was found by the data that γ-Al 2 O 3 surface area decreased after metal loading.It was found from the data that the surface area of γ-Al 2 O 3 decreased after metal loading.γ-Al 2 O 3 was used as a carrier.The richness of the carrier pore structure was only for the better impregnation of the transition metal into the pore volume, and eventually a more uniform pore site was formed.When the pore site was formed, the active component of the catalyst occupied the site, thus the surface area decreased, and the total volume of the micropores decreased [20].The loading of the metal had basically no effect on the adsorption pore size of the material.The slight increase may be attributed to the long time of calcination during the preparation process The process of roasting played an activating role for the catalyst.
The catalyst with better catalytic effect had more decrease in the specific surface area.Since most of the catalyst surface area was the inner surface area.The active center was often distributed on the inner surface, xylene must diffused through the pores of the catalyst to reach the active center on the inner surface of the catalyst before it was adsorbed.This diffusion process was closely related to the pore structure of the catalyst.The smaller the specific surface area of the catalyst, the larger the average pore size.Indicating that the higher the dispersion of active components, the smaller the chance of simple compounding of electrons and holes, the more intense the diffusion processed, and the higher the catalytic activity of the catalyst.The specific surface area of the catalyst continues to decreased after the reaction because the catalyst has been deactivated and poisoned.The specific surface area of the Fe-Mn/γ-Al 2 O 3 catalyst adsorbed with xylene changes from 134.69257 m 2 /g to 116.3728m 2 /g.The specific surface area of the Fe@Mn/γ-Al 2 O 3 catalyst adsorbed with xylene changes from 165.71477 m 2 /g to 140.5239m 2 /g [25].
The catalyst made by co-impregnation method had the smallest specific surface area after reaction.Analyzing the reasons, the prolonged exposure to plasma during plasmacatalyzed xylene may produce entropic oxides.Entropic oxides provide additional spacing and rugosity to the surface [26], elevating the reaction yield and allowing the catalyst to exhibit superior catalytic effect.
Figure 6c, d showed that the Fe@Mn/γ-Al 2 O 3 catalyst prepared by the consecutive impregnation volume co-impregnation method had weak diffraction peak intensity and broad peak shape, indicating that the equivalent volume consecutive impregnation method was adopted.The crystallinity and grain size of the prepared catalyst Fe-Mn/γ-Al 2 O 3 were better than those of Fe@Mn/γ-Al 2 O 3 .Before and after the reaction, the Fe-Mn/γ-Al 2 O 3 diffraction peaks changed little, and the main crystal structure did not change significantly.Because the Fe-Mn/γ-Al 2 O 3 reaction process had a good stability of active components, high electron transfer rate, and the removal effect of VOCs-O 3 was more obvious.By comparing the same catalyst, Fe@Mn/γ-Al 2 O 3 catalyst existed FeMnO 3 before the reaction, presumably because the iron-manganese compound similar to perovskite would be formed after co-impregnation and roasting in a muffle furnace.Because the ozone in the low temperature plasma further oxidizes it deeply, it was oxidized to Fe 2 O 3 and MnO x , which consumes the active oxygen in the system.It was precisely because of the consumption of ozone and other substances in this process that the catalytic effect of Fe@Mn/γ-Al 2 O 3 catalyst was slightly lower than that of Fe-Mn/γ-Al 2 O 3 catalyst.

FT-IR
Figure 7 showed that the position of CO 2 absorption peak at 2300-2380 cm −1 indicated that a large amount of xylene was converted into CO 2 during the reaction.
The absorption peak of C-N appeared in the region of 1020-1360 cm −1 , indicating a large difference in the influence of the adjacent substituent.γ-Al 2 O 3 had a higher absorption peak than the other two catalysts due to the lone pair of electron pairs of nitrogen conjugated to the aromatic ring when not loaded with metal, which made C-N double bonded in nature.The absorption peak of the C-N stretching vibration aliphatic amine appeared near 1030-1230 cm −1 , which was due to the influence of the C-N stretching vibration by the adjacent substituent, where the unshared electron pair on the nitrogen atom was conjugated to the P-Π of the carbonyl group, which shifted the C-N stretching vibration to a higher wave number.The absorption peaks appeared from 1100 to 1700 cm −1 indicated the presence of derivatives with benzene rings in the products.These included C=O group Fig. 7 The supported catalyst on γ-Al 2 O 3 carrier after the reaction (1700-1680 cm −1 ), -CH 3 (1365-1385 cm −1 ) and benzene ring C=C and skeleton vibration absorption peaks (1450-1600 cm −1 ).Therefore, the presence of aromatic compounds could be confirmed, which indicated the presence of undegraded xylene or other intermediates contained benzene rings on the catalyst surface.The absorption peaks at 1450-1600 cm −1 and 3200-3600 cm −1 were caused by the -OH group in the organic by-products.The former was a sharp absorption peak, caused by the bending vibration of the intermolecular hydrogen bond O-H.The latter was a broad absorption peak, caused by the stretching vibration of O-H, and the hydroxyl compound was very significant in terms of the conjoining phenomenon.The location of the CO 2 absorption peak at 2300-2380 cm −1 indicated a large conversion of xylene to CO 2 during the reaction, which was in general agreement with the CO 2 peaks measured by Eduard [27] et al. at 2361 cm −1 and 2338 cm −1 .The absorption peak of the Fe-Mn/γ-Al 2 O 3 catalyst was significantly higher than those of the other two catalysts, indicating that the Fe-Mn/γ-Al 2 O 3 catalyst absorbs IR light more strongly at this location, where the electrons in the atoms could be excited more and the end product CO 2 was higher than that of the Fe-@Mn/γ-Al 2 O 3 catalyst.

XPS
Figure 8 showed the results of fitting the peaks in the spectra of O1s, Fe3d and Mn2p. 2− and O − , that had strong mobility and played an important role in the performance and activity of the catalyst.Whereas surface chemisorbed oxygen had faster and more efficient electron mobility than surface lattice oxygen, it played a key role in the degradation of p-xylene, and its content was listed in Table 2.When Fe-Mn/γ-Al 2 O catalyst participated in the reaction, the content of surface chemisorbed oxygen increased to 58.03%.This was because the lattice oxygen in metal oxides generally did not directly participate in the reaction of xylene, and part of the lattice oxygen is trapping electrons.It was then converted into adsorbed oxygen and underwent electrophilic addition with xylene or its intermediates at the electron-deficient place to form oxygen-containing or promyogenic compounds, which then underwent cleavage and oxidation reactions.When the catalyst participated in the redox reaction, the lattice Oxygen reduced highvalence metal ions to low-valence states, which were themselves oxidized to chemisorbed oxygen, resulting in an increase in the content of chemisorbed oxygen and a decrease in lattice oxygen.
Figure 8c showed that the two characteristic peaks were at the binding energies of 706-719 eV and 719-734 eV, which were assigned to the Fe2p 3/2 and Fe2p 1/2 spin orbits respectively.The characteristic peaks of Fe2p 3/2 and Fe2p 1/2 were sub-peak fitting, and the semi-quantitative analysis of the content of Fe 3+ , Fe 2+ and Fe on the catalyst surface was carried out.When Fe-When the Mn/γ-Al 2 O catalyst participates in the reaction, the high-energy electrons ionized could be captured by Fe 3+ to form Fe 2+ , and Fe 3+ and Mn 3+ would also cooperate to promote the conversion of Fe 3+ to Fe 2+ .In the presence of O 2− , the electrons on the 2p orbital of Fe 2+ were transferred to O 2− , while the electrons on the 2p orbital of O 2− were transferred from O 2− to Fe 3+ .The double exchange between Fe 2+ and Fe 3+ would cause the charge imbalance on the catalyst surface and increased the number of holes.The charge imbalance on the surface increased the number of holes and provides more oxygen vacancies for chemisorbed oxygen, which ultimately improved the xylene degradation efficiency.
Figure 8d showed that the binding energies of the splitting peaks generated by the Mn2p3/2 and Mn2p1/2 spin orbits correspond to 637-645 eV and 650-658 eV, and the characteristic peaks of Mn 3+ and Mn 4+ correspond to the Mn2p3/2 spin orbitals.The positions were at 641.5 eV and 643.0 eV, respectively.The content of Mn 4+ in the catalyst after the reaction decreased from 58.29 to 41.97%, while the content of Mn 3+ increased from 41.71 to 54.86%.Because Mn 4+ could promote the transfer of oxygen atoms, and ozone molecules would eventually form reactive oxygen species O 2 2− , O − , etc. through electrons provided by Fe 2+ and Mn 3+ .The transfer of electrons between Fe and Mn lead to the enhancement of the oxidative ability of ozone, which was conducive to the conversion of ozone into active oxygen by obtaining electrons, and finally improved the degradation rate of xylene.Comprehensive analysis showed that the activity of the catalyst was closely related to the density of oxygen vacancies on its surface.Usually, the ozone adsorbed on the oxygen vacancies participates in the cycle of electrons in the oxygen vacancies, which lead to the reduction of low valence state of Mn 4+ and Fe 3+ in the catalyst.This cycle could enhance the gas adsorption capacity and convert ozone and xylene more completely during the plasma catalysis process.Combined with the results of this study, it could be demonstrated that there was interfacial electron transfer in the reaction process.In conclusion, Mn 4+ , Fe 3+ and O lat played an important role in the co-processing of VOCs-O 3 with low temperature plasma.

Gas Phase Reaction
Figure 9 showed that the gas-phase reaction was divided into the collision reaction process of high-energy electrons and the oxidation reaction process between free radicals [28].
The excitation and dissociation of xylene molecules depended on the energy of highenergy electrons and the energy of intramolecular chemical bonds.Comprehensive research results showed that the basis of low temperature plasma technology to degrade xylene is the collision reaction between high-energy electrons and gas molecules [29].Figure 9a showed the electron orbital arrangement of ozone.Its structure is unstable and has the characteristics of electrophilic, nucleophilic and dipolar, hence its properties are active.Its reactivity is strong [30].When the electron energy is higher than the chemical bond Fig. 9 Gas phase reaction process and mechanism diagram energy of the pollutant, the direct collision of the high-energy electrons with the pollutant leads to the breaking of the chemical bond.A series of small molecular substances are formed finally.The reactions after high-energy electrons collide with xylene are shown in Table 3.
Xylenes initially form unsaturated methylene or unsaturated groups due to the dehydrogenation of a methyl group on xylene.The reason is the energy relationship between the bond energy and the high-energy electrons.It is found that the C-H bond on the methyl group of the phenyl ring has a lower energy and is most likely to break after direct collision with the high energy electrons.The second step is the cleavage of the C-C bond between the methyl and benzene rings, the products of which are methyl and phenyl.The third step is the cleavage of the C-H bond on the benzene ring to generate the reaction of C 6 H 3 From an energy point of view, the enthalpy change of the six-membered ring cracking reaction is greater than zero.Compared to the oxidation reaction of most hydrocarbons with an enthalpy change of less than zero, the cracking and ring opening reaction becomes more difficult.This process has therefore been identified as the rate-limiting step in the degradation of xylene.In the above reaction process, the cleavage of the benzene ring of xylene will produce unstable intermediates that will increase the chaos of the entire reaction system.These intermediates will use energy to combine with the active particles and react to form a new, more stable substance.But it made the ring-opening reaction of xylene more difficult.The fourth step relies on pure electron impact to cause the ring-opening reaction.The ring-opening reaction relies mainly on the oxidation and decomposition of ozone and various active particle free radicals in the reaction system.
The collision of high-energy electrons with the background gas can generate a series of free radicals (such as O 3 , O • and OH • , etc.).These radicals will further react with the xylene molecule and its intermediates.Finally it is completely oxidatively decomposed into water and carbon dioxide [31].The generation process of a series of free radicals such as ozone, O • and OH • is mainly divided into three steps, as shown in Fig. 9b.The plasma will generate highenergy electrons during the discharge process.This is due to the emergence of free electrons in the gas.The free electrons can obtain energy from the applied electric field, and collide with a gas molecule.Then transmit the energy to the gas molecules, resulting the outer shell electrons of the gas molecules to become free from the shackles of the core, leading in positively charged ions and free electrons [32].Oxygen molecules undergo collision, dissociation, ionization and recombination effects under the impact of high-speed moving electrons and under the action of the photoelectric effect, thereby forming oxygen plasma.There are many kinds of particles in oxygen plasma, mainly positive ions, negative ions, O  9c.The formation process of these particles is mainly affected by the chemical composition of the gas, but also by the physical discharge process [33].The reaction of the chemical part mainly depended on the average energy, electron density, etc.The main reaction formula is: The above reactions are dominant in the oxygen plasma reaction, and the reaction process accounts for 80% of the energy consumed by the plasma.When the oxygen plasma is formed, the oxygen atoms into an atomic reaction, thereby producing ozone [34].The main synthesis reaction formula of ozone is: The formation of ozone also involves the reaction of ion molecules, the dissociative adsorption and desorption of oxygen atoms, and the reaction of nitrogen plasma in the process of plasma formation.The generated free radical particles continue to undergo redox reactions with xylene molecules (excited state) or some decomposed small molecular substances.Through a series of coupling and synergistic pathways, they can finally obtain less or no pollution to the environment.Small molecule gases that are easier to handle, such as CO 2 , CO, H 2 O, etc.

Surface Reaction
The reaction mechanism on the surface of Fe-Mn/γ-Al 2 O 3 catalyst was shown in Fig. 10.
Ozone is adsorbed on the surface of the catalyst and undergoes a redox reaction with the oxides of the active components Fe and Mn indicated by the catalyst, and the active oxygen is converted to lattice oxygen.Fe 2+ plays a key role in the exchange of O species between Fe 2+ /Mn 4+ and Fe 2+ /Fe 3+ in Fe-Mn/γ-Al 2 O 3 catalyst.In Fe-Mn/γ-Al 2 O 3 catalysts, the participation of Fe can enable Mn 4+ to induce the rapid migration of oxygen vacancies and enhance the redox process between Mn 4+ and Mn 3+ .MnO x , a representative substance in P-type semiconductors, has a strong ozone inhibitory effect.Anions such as O 2− and O 2− 2 in the system can combine with MnO x under the action of Coulomb force to enhance the degradation of VOCs.Efficiency while suppressing the generation of ozone.Among them, the control of ozone by Mn is expected to occur in reactions 12-13; and if Fe is involved in the decomposition of O 3 , reactions 14-16 are expected.This is the key reaction process for Fe-Mn/γ-Al 2 O 3 catalysts to control ozone production.The xylene, intermediates and active components in the final mixed gas are adsorbed on the catalyst surface and then further oxidised by the active oxygen on the catalyst surface.
The interaction of FeO x and MnO x in the Fe-Mn/γ-Al 2 O 3 catalyst plays a key role in the movement of oxygen, resulting in the mutual conversion between adsorbed oxygen, lattice oxygen and active oxygen.The Fe-Mn/γ-Al 2 O 3 catalyst can play a synergistic effect.

Conclusion
In the process of co-processing VOCs-O 3 by low temperature plasma combined with supported catalysts, the effect of different supported metal oxide catalysts on the degradation of xylene and the inhibition of O 3 by the system was investigated.
(1) Using the consecutive volume impregnation method, the removal efficiency of xylene was found to be from high to the low order was as follows: were used for joint low-temperature plasma synergistic treatment of VOCs-O 3 using consecutive impregnation.The catalytic effect of the catalyst prepared by the method was better than that of the co-impregnation method.In the synergistic catalytic process, the total removal rate of xylene could reach up to 94.88%, and the ozone consumption rate can reach up to 84.1%.
(3) Low temperature plasma catalytic technology could enhance the generation of active groups.The electron transfer between bimetal oxide supported catalysts and active groups can greatly improve the degradation rate and reduce the energy consumption of the reaction.The production of by-product ozone and intermediate product benzene series VOCs is also significantly suppressed when energetic electrons collide with the background gas, and the final degradation rate of VOCs is significantly increased.

Fig. 2
Fig. 2 Low temperature plasma.a PDC b PEC

Fig. 4
Fig. 4 Effects of different catalysts on VOCs-O 3 .a removal rate with time, b xylene removal rate, c ozone content, d ozone depletion and monometallic oxide catalyst depletion rate, e removal rate with time, f removal rate, g ozone content, h ozone depletion and bimetallic oxide catalyst depletion rate

Fig. 5
Fig. 5 Effect of catalyst on CO x selectivity

Fig. 6
Fig. 6 XRD patterns a Before and b after the reaction of different catalysts; c Before and after the reaction of Fe@Mn/γ-Al 2 O 3 catalyst; d Fe-Mn/γ-Al 2 O 3 catalyst

• 2CH 3
and H • .The probability of the fourth step occurring is relatively low because the large π bonds between the carbon atoms of the benzene ring have high energy and are difficult to open due to the collision of high energy electrons.However, the methyl-benzene ring C-C bonds, methyl C-H bonds and benzene ring C-H bonds are relatively easy to open.
Fig. 1 Preparation process of M x O y /γ-Al 2 O 3 supported catalyst hexahydrate (

Table 1
Specific surface area, pore volume and pore size of different catalysts

Table 2
Semi-quantitative results based on peak splitting

Table 3
Different chemical bond energies in xylene and reactions upon breakage