Complete mineralization of acetic acid using catalytic ozonation with a high energy eciency enhanced by MnO2/ γ -Al2O3

The complete mineralization of acetic acid in a biodegradation process is difficult due to the α -position methyl on the carboxyl group of acetic acid. This study explores the complete oxidation of acetic acid by catalytic ozonation. Metal oxides of MnO 2 , Co 3 O 4 , Fe 3 O 4 , and CeO 2 loaded on γ -Al 2 O 3 power were used as the catalysts. The experimental results showed that MnO 2 / γ -Al 2 O 3 catalyst had the best mineralization performance for acetic acid. Typically, the mineralization of acetic acid is as high as 88.4% after 300 min ozonation of 100 mL of 1.0 g L ‒ 1 acetic acid catalysed by 3.0 g 1.0wt.% MnO 2 /γ -Al 2 O 3 catalyst powder with an energy efficiency of 15 g kWh ‒ 1 . However, without a catalyst, the mineralization of acetic acid is only 33.2% with an energy efficiency of 5.1 g kWh − 1 . The effects of MnO 2 loading, catalyst dosage, acetic acid concentration, O 3 concentration, ozonation temperature, and initial pH value of the acetic acid solution were systematically investigated. Radical quenchers and in-situ DRIFTS analyses indicated that •OH radical and reactive oxygen species on catalyst surface played an important role in the ozonation of acetic acid to CO 2 and H 2 O. The mechanism of acetic acid oxidation on MnO 2 /γ -Al 2 O 3 is proposed. two ion and two columns: a 2-m N column with a methanizer prior to the FID to analyse CO and CO 2 ; and a capillary column (SH-Stabilwax-DA, Shimadzu, Japan) to analyse the acetic acid concentration in the water liquid solution. The liquid samples without the catalyst powder were injected directly into the injection port of the GC. For the liquid samples with the catalyst powder, the catalyst powder was separated by centrifugation, and the supernatant liquid was used for the GC analyses. The O 3 concentrations in the gases from the water saturator and the ozonation reactor were measured using an O 3 meter, and two ice baths were used to remove the water in the gas before entering the O 3 meter (BMOZ-200T, By-products from acetic acid oxidation were analysed using high-performance liquid chromatography (HPLC, Agilent USA) equipped with an Eclipse Plus C18 column and an (UV) detector. The mobile buffers of A (KH 2 PO 4 , 0.015 mol L − 1 and B 0.5 mL min − 1 were used. and 25 °C ozonation temperature. Energy efficiency (e) and O 3 concentration and O 3 concentration drop (f) versus ozonation time at various pulse frequencies; experimental condition: 1.0 g L 1 acetic acid concentration, 3.42 pH, 3.0 g MnO 2 /γ -Al 2 O 3 catalyst dosage, and 25 °C ozonation temperature. Energy efficeicncy as a function of ozonation time (g) and relation of energy effciciency and ozonation temperature experimental condition: 1.0 3.42 initial


Introduction
The production of acetic acid is common in chemical industries such as petrochemicals and wood pulp mills, and these industries usually produce wastewater containing acetic acid (Wisniewski and Pierzchalska 2005;Shan et al. 2017). In addition, acetic acid has always been a by-product in some conventional wastewater treatment processes of macromolecular organic matter, since it is difficult to further oxidize the α-position methyl group on the carboxyl group of acetic acid to CO2 and H2O (Centi et al. 2000). It was found that by-products, such as acetic acid, were formed when O3 was used for 2propanol oxidation (Reisz et al. 2019). Zhong et al. (2018) reported several short-chain acids, such as acetic acid and formic acid, as major by-products that formed during the ozonation degradation of humic acids. Acetic acid-contaminated water not only causes great harm if ingested by livestock and to crop irrigation but also adversely affects the respiratory system and sensory organs of human beings (Sedjame et al. 2018;Zhao et al. 2018). Therefore, the effective degradation of acetic acid is of great significance for the complete mineralization of organic compounds in wastewater.
Several physical approaches, such as adsorption and membrane separation methods, have been investigated for the removal of acetic acid. Borhan et al. (2016) used activated carbon with zinc chloride to remove acetic acid using the adsorption method. Lee (2013) developed an emulsion membrane system for the selective removal of acetic acid from simulated hemicellulose hydrolysates. Acetic acid decreases the pH of the waste water. The neutralization with alkali would lead to an enormous load of total dissolved solid in the treated effluent for biodegradation. An early literature reported that 1.0 kg of NaOH or 0.62 kg of CaCO3 needed to neutralize 500 g of acetic acid in the wastewater for further biodegradation (Jin 1989). Biodegradation, on the other hand, while not impossible, will be a very slow process (Sinha and Annachhatre 2007).
Advanced oxidation processes have been widely used to remove organic matter from aqueous environments (Giardina et al. 2019). In addition, reactive species, such as ·OH and ·O, can effectively break down organic matter into small molecules, such as CO2 and H2O (Miklos et al. 2018). However, these methods have disadvantages, such as the large scale of the removal unit, where secondary contamination would occur because the contaminant was not fully degraded (Mu et al. 2019). Cihanoğlu et al. (2017) combined ultrasound with a catalyst to oxidize acetic acid, and the chemical oxygen demand (COD) degradation was only 25.5%. Sannino et al. (2011) studied the heterogeneous photo-Fenton oxidation of acetic acid on LaFeO3, LaMnO3, and LaFeO3 chalcogenides with total organic compound removals of 60%, 54%, and 100% after 5-h oxidation, respectively. Experimental results showed that the presence of SO2 4− /ZrO2−Fe2O3 can greatly improve the oxidative efficiency of H2O2/O3 for acetic acid ozonation in the pH range from 0 to 5.0 (Wang et al. 2016). Although the ozonation process can be conducted at ambient temperature and atmospheric pressure and is a cleaner wastewater treatment process that does not produce sludge, the reaction rate should be further improved (Malik et al. 2020).
Furthermore, the high energy consumption of O3 generation makes the ozonation process costly (Nur et al. 2016).
Catalytic ozonation has been proven to be an effective method to effectively degrade organic matter and to produce more reactive oxygen species, and this can greatly reduce the ozonation process cost.
Transitional metal oxides have exhibited superior O3 decomposition ability. It was deduced that the increased degradation ability was due to the formation of more reactive hydroxyl radicals (·OH) with higher redox potentials with the help of specific catalysts, which can then be used to further degrade the target pollutants (Chen et al. 2016;Sui et al. 2010). It has been found that nano-CeO2 greatly improved the oxidation efficiency of the H2O2/O3 system and promoted the degradation of acetic acid small molecules . Peng et al. showed that the O3 oxidation efficiency of succinic acid reached 100% in combination with a Ni/Al2O3 catalyst (Peng et al. 2018). Nevertheless, little attention has been paid so far to the effective degradation of acetic acid by catalytic ozonation, and therefore there is a great need to improve the degradation of acetic acid.
In this work, the best catalyst for the mineralization of acetic acid is firstly screened from the metal oxides of MnO2, Co3O4, Fe3O4, and CeO2 loaded on γ-Al2O3. The influencing factors (such as metal oxide loading, catalyst dosage, acetic acid concentration, O3 concentration, pH, and ozonation temperature) are also investigated. Tert-butanol (TBA) and p-benzoquinone (PBQ) are used to explore reactive species for acetic acid oxidation. The gas-liquid-solid system is demonstrated by applying in-situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS), and the mechanism of acetic acid ozonation on MnO2/γ-Al2O3 catalyst is revealed.

Catalyst preparation
Nitrates of manganese, cobalt, iron, and cerium were used as the precursors of MnO2, Co3O4, Fe3O4, and CeO2, respectively. The nitrates in a water solution were loaded on γ-Al2O3 powder (as support, 2000 mesh, 99%) using the sequential impregnation method. The impregnated powder was dried at 100 o C for 5 h and calcined in air at 500 o C for 3 h to obtain the MnO2/γ-Al2O3, Co3O4/γ-Al2O3, Fe3O4/γ-Al2O3, and CeO2/γ-Al2O3 catalysts. The loading amounts of the metal oxide were adjusted by changing the amounts of nitrate precursors.

Catalyst characterization
The crystalline phases of the catalysts were determined using X-ray diffraction (XRD, Rigaku Ultima IV powder diffractometer, Japan) with a Cu Ka radiation. The morphological properties of the catalysts were measured using scanning electron microscopy (SEM, Supra 55, Zeiss, Germany) with energy-dispersive X-ray spectroscopy (EDS). The surface elements on MnO2/γ-Al2O3 were analysed using X-ray photoelectron spectroscopy (XPS, ESCALAB 250Xi+, Thermo, USA).

Ozonation of acetic acid
The ozonation of acetic acid was investigated using the experimental system in Fig. 1a. The system was primarily composed of a dielectric barrier discharge (DBD) reactor and an ozonation reactor. Pure oxygen (99.999%, 100 mL min −1 ) was supplied to the DBD reactor using a mass flow controller (D07, Sevenstars, Beijing, China). A pulse power supply (M10K-08, Suzhou Allftek, China) was used to drive the DBD reactor to generate O3. The temperature of the ozonation reactor was controlled using a water bath and an electric heater. The ozonation reactor contained 100 mL of an acetic acid solution with or without catalyst. The acetic acid solution was stirred with a magnetic stirrer to promote the suspension of the catalyst powder and the contact of acetic acid with the catalyst powder and O3 gas bubbles from a bubbler head (BSJ1348-1/8, Langfang Shengguang Filter, China).
The acetic acid degradation was studied by investigating the loadings of metal oxide on the γ-Al2O3 powder, the catalyst dosage, the acetic acid concentration, the O3 concentration, pH, and the oxidation temperature on the acetic acid degradation.
The pH value of the acetic acid solution was modified using sodium hydroxide and measured using a pH meter (AZ 86505, Hengxin, China). The O3 oxidation mechanism of acetic acid was investigated by adding OH quencher tert-butanol (TBA, 362 mM) and the superoxide radical (·O − 2 ) quencher pbenzoquinone (PBQ, 1.1 mM) to the acetic acid solution.

Analytical methods
The concentrations of acetic acid and oxidation products (CO and CO2) in the gases from the outlet of the ozonation reactor were on-line analysed using a gas chromatograph (GC) (GC2014, Shimadzu, Japan) equipped with two flame ion detectors (FIDs) and two columns: a 2-m Porapark N column (Dalian Institute of Chemical, China) with a methanizer prior to the FID to analyse CO and CO2; and a capillary column (SH-Stabilwax-DA, Shimadzu, Japan) to analyse the acetic acid concentration in the water liquid solution. The liquid samples without the catalyst powder were injected directly into the injection port of the GC. For the liquid samples with the catalyst powder, the catalyst powder was separated by centrifugation, and the supernatant liquid was used for the GC analyses. The O3 concentrations in the gases from the water saturator and the ozonation reactor were measured using an O3 meter, and two ice baths were used to remove the water in the gas before entering the O3 meter (BMOZ-200T, Weifang Shenxing, China). By-products from acetic acid oxidation were analysed using high-performance liquid chromatography (HPLC, Agilent 1260, USA) equipped with an Eclipse Plus C18 column and an ultraviolet (UV) detector. The mobile buffers of A (KH2PO4, 0.015 mol L −1 ) and B (Acetonitrile with 0.5 mL min −1 ) were used.
The catalytic ozonation of acetic acid on the MnO2/γ-Al2O3 catalysts of 0.5wt.%, 1.0wt.%, and 10wt.% MnO2 loadings was studied using an in-situ DRIFTS system (Fig. 1b). The system primarily consisted of an acetic acid/water bubbler, an O3 generator, an in-situ cell (HVC-DRP-5, Harrick, USA), and the DRIFTS system (Nicolet iS50, Thermo Scientific, USA). Gases containing acetic acid and water were obtained by bubbling an acetic acid solution (100 mL, acetic acid concentration 1.0 g L −1 ), and this was supplied to the in-situ cell to simulate the reactions within acetic acid, water, and O3 on MnO2/γ-Al2O3 catalyst. O3 with a concentration 37.5 g Nm −3 was obtained from the O3 generator. The temperature of the in-situ cell was 25 °C. The DRIFTS spectra were processed using OMNIC software and converted using the Kubelka-Munk function.
Where C0 is the initial concentration of acetic acid in g L -1 , and Ct is the concentration of acetic acid in g L -1 at reaction time t (min).
Mineralization (Yi+1) and energy efficiency (ηi) (g kWh -1 ) were defined as follows: Eq. 4 was used to calculate the amount of CO2 and CO generated from acetic acid oxidation.
Where m0 is the initial amount of acetic acid in g. mg is the amount of acetic acid in g and calculated from the average concentration of [CO2] and [CO] at ti+1 and ti time (min) with the difference in oxidation time (ti+1−ti). P is the discharge power in W, 60 is the conversion factor between hour and minute, and 1000 is the conversion factor between W and kW. F is the flow rate of the bubbling gas, 0.100 L min -1 ; 22.4 is the molar volume (L) of a gas at standard state and 60.5 is the molecular weight of acetic acid.
The discharge power (P) was calculated from waveforms of the pulse voltage supplied from the pulse power supply, using the voltage probe (P6015A, Tektronix, USA) and current probe (CP8030H, Cybertek, China) and oscilloscope (MDO 3022, Tektronix, USA).

Catalytic activity investigation
The MnO2/γ-Al2O3, Co3O4/γ-Al2O3, Fe2O3/γ-Al2O3, and CeO2/γ-Al2O3 catalysts were used to screen for the best metal oxide for acetic acid degradation. The degradation of acetic acid as a function of ozonation time is shown in Fig. 2a. The acetic acid was not obviously degraded by O2 bubbling only, suggesting that O2 had difficulty oxidizing acetic acid. The degradation of acetic acid was 39.9% with O3 bubbling without a catalyst at a 300 min ozonation time. The degradation of acetic acid with O3 bubbling and MnO2/γ-Al2O3 increased rapidly to 49.2% at 20 min and gradually to 88.4% at 300 min. It was obvious that the degradation of acetic acid was enhanced by the application of the MnO2/γ-Al2O3 catalyst in comparison with the result of O3 bubbling without a catalyst. The rapid increase in degradation during the first 20 min of ozonation time was primarily due to the adsorption of acetic acid on the MnO2/γ-Al2O3 surface. Similar adsorption effects were found on the surfaces of Co3O4/γ-Al2O3, Fe2O3/γ-Al2O3, and CeO2/γ-Al2O3. The adsorption effect and acetic acid degradation had the same order of MnO2/γ-Al2O3>Co3O4/γ-Al2O3>Fe2O3/γ-Al2O3>CeO2/γ-Al2O3. This order did not change when the metal oxide loading increased from 1.0wt.% (Fig. 2a) to 10wt.% (Fig. 2b).
The mineralization of acetic acid is important to explore the ability of the catalyst that promotes acetic acid oxidation to CO and CO2. The mineralization of acetic acid is linear to ozonation time, and the slopes of the mineralization of acetic acid using MnO2/γ-Al2O3 (both 1.0wt.% and 10wt.%) were the largest (Figs. 2c and 2d). This demonstrated that the MnO2/γ-Al2O3 was the best catalyst for the mineralization of acetic acid. The energy efficiency using MnO2/γ-Al2O3 was approximately 14.9 g kWh −1 and also the highest in comparison with the other three catalysts (Figs. 2e and 2f). Since MnO2/γ-Al2O3 was the best catalyst for acetic acid degradation, the primary factors influencing the catalytic performance, such as MnO2 loading, catalyst dosage, acetic acid concentration, O3 concentration, pH, and ozonation temperature, were further investigated.
The effect of MnO2 loading on the degradation of acetic acid was investigated in detail, and the results are shown in Fig. 3. The acetic acid concentration decreased from 1.0 g L −1 to a level of 0.6 g L −1 at 20 min and gradually to approximately 0.1 g L −1 at 300 min. The highest degradation of acetic acid was 88.4% at 300 min of ozonation time (Fig. 3b). Figs. 3c and 3d indicate that the CO2 was the primary product from acetic acid ozonation since the CO2 concentration was approximately 6.5-43 times as many as CO. The mineralization increased linearly to a maximum level of 88.9% (Fig. 3e). The energy efficiency using the 1.0wt.% MnO2/γ-Al2O3 catalyst was the highest at 15.3 g kWh −1 (Fig. 3f). Fig. 3g illustrates the relationships of the average energy efficiencies between 20 and 300 min as a function of MnO2 loading. It is obvious that the 1.0wt.% MnO2/γ-Al2O3 catalyst had the maximum average energy efficiency of 14.9 g kWh −1 and was the best catalyst for acetic acid degradation. Fig. 3h shows O3 Figs. 4a and 4b. The highest energy efficiency was also found when using 3.0 g of the 1.0wt.% MnO2/γ-Al2O3 catalyst (Fig. 4a). It was evident that the higher energy efficiency was obtained since a generally higher O3 consumption occurred when using 3.0 g of the 1.0wt.% MnO2/γ-Al2O3 catalyst (Fig.   4b). The distinct O3 concentration drop within the first 100 min was deduced to be the adsorption effect of the MnO2/γ-Al2O3 catalyst. In addition to the dosage of the catalyst, the concentration of acetic acid also had a certain effect on its degradation. Fig. 4c illustrates the influence of initial acetic acid concentration on energy efficiency at various ozonation time. The highest and lowest energy efficiencies were obtained when the initial acetic acid concentrations were 1.0 g L −1 and 0.5 g L −1 , respectively. Since the ozonation of acetic acid related with the adsorption of acetic acid and O3 on catalyst surface, the presence of the highest initial concentration of acetic acid is possibly due to the optimum adsorption of acetic acid and O3 on catalyst surface under the experimental condition. Fig. 4d shows the mineralization of acetic acid at 300 min ozonation time as a function of initial acetic acid concentration. The mineralization reached 89%, 84%, 67%, and 50%, when the initial acetic acid concentration was 0.5, 1.0, 1.5, and 2.0 g L −1 , respectively.
The mineralization decreased with increasing the initial acetic acid concentration, however, the amount of acetic acid mineralized increased. This finding implied that in order to get a large amount of mineralized acetic acid, a high initial acetic acid concentration is required; however, in order to get the highest energy efficiency, the ozonation should be carried out with optimized initial acetic acid concentration.
O3 played a vital role in the degradation of acetic acid. The O3 concentration was concisely adjusted by changing the frequency of the voltage pulses to the DBD reactor, and the O3 concentrations were 20.7, 35.6, 47.0, and 55.0 g Nm −3 when the frequencies were 50, 100, 150, and 200 Hz, respectively (Fig. 4f).
The maximum energy efficiency (25.5 g kWh −1 ) at 20 min was obtained when the frequency was 50 Hz.
However, the energy efficiency decreased with increasing frequency (Fig. 4e). It was found that the O3 concentration drop was within 4.0-6.0 g/Nm 3 after 140 min of ozonation time (Fig. 4f), although the O3 concentration increased with the frequency. This indicated that the amount of O3 used for acetic acid was limited, and most of the O3 did not take part in acetic acid ozonation and flowed away from the ozonation reactor.
The acetic acid degradation was characterized at different ozonation temperatures with or without a catalyst. The degradation, mineralization, and energy efficiency generally increased with an increase in the ozonation temperature (Figs. 4g and 4h), indicating that the ozonation reaction was temperature sensitive. The energy efficiency increased from 8.81 g kWh −1 at 0 °C to 17.9 g kWh −1 at 40°C and slightly to 18.4 g kWh −1 at 70 °C (Fig. 4g). The energy efficiency without a catalyst increased from 4.40 g kWh −1 at 0 °C to the maximum level of 8.46 g kWh −1 at 55 °C and decreased to 6.74 g kWh −1 at 70 °C (Fig.   S2d). The highest degradation, mineralization, and energy efficiencies were achieved, and they were 95.2%, 96.6%, and 18.4 g kWh −1 , respectively, at 70 °C. The O3 concentration drop was enhanced in the presence of the 1.0wt.% MnO2/γ-Al2O3 catalyst within the ozonation temperature range (Fig. S3).
The initial pH value of the acetic acid solution was adjusted with 0.1 M NaOH. Fig. S4 shows the energy efficiencies at different pH values as a function of ozonation time. When the pH was 3.42 (the original pH of acetic acid solution), the energy efficiency was at the maximum level of 14.9 g kWh −1 , higher than that when the pH value was 7.06 or 11.26. It has been reported that •OH radical has a higher oxidation ability than other types of reactive oxygen species (such as O − 2 ), especially under acidic conditions, the similar pH effect was reported by Sahni et al. (2005) during the degradation of polychlorinated biphenyls using liquid-phase discharge plasma. Therefore, it was deduced that the •OH radicals played a major role in acetic acid ozonation, and more details are discussed in next section.

Characteristics of the MnO2/γ-Al2O3 catalysts
The crystalline structures of γ-Al2O3 and MnO2/γ-Al2O3 with 0.5wt.%, 1.0wt.%, and 10wt.% MnO2 loadings were characterized, and the XRD patterns are shown in Fig. 5a. The peaks ascribed to γ-Al2O3 were clearly observed, indicating that the crystalline structure was well reserved after the addition of  (224) planes of Mn3O4 (PDF#24-0734), respectively. When the MnO2 loadings were 0.5wt.% and 1.0wt.%, no characteristic peaks were found, indicating MnO2 was amorphous or highly dispersed on the surface of γ-Al2O3. In addition, it was noted that as the MnO2 addition increased to 10wt.% (Fig. 5iv), the peaks belonging to the MnO2 phase were present, while the Mn3O4 phase could not be found. This suggested that MnO2, rather than Mn3O4, was the primary phase on the γ-Al2O3 support. The SEM images showed that the Mn and O elements were uniformly dispersed (Fig. 5b), which was consistent with the result of XRD.
The surface chemical states of the 1.0wt.% MnO2/γ-Al2O3 catalysts before and after acetic acid ozonation were characterized using XPS. As shown in Fig. 5c, the Mn 2p3/2 spectra were deconvoluted into three peaks at 640.6, 641.7, and 643.0 eV, which were ascribed to the Mn 2+ , Mn 3+ , and Mn 4+ species, respectively (Gao et al. 2020). The peak areas of the Mn 4+ species accounted for 43.8% and 48.1% before and after acetic acid ozonation. A high Mn 4+ ratio for the manganese-based catalysts is typically strongly linked to a superior catalytic activity (Liu et al. 2021;Saputra et al. 2014), and this also corresponded to the excellent catalytic ozonation performance of the MnO2/γ-Al2O3 catalyst. In addition, it was noted that the ratio of Mn 2+ decreased from 27.5% to 16.2% after the catalytic ozonation process (Table 1), while the ratio of Mn 3+ and Mn 4+ increased. This was possibly due to a portion of Mn 2+ being oxidized by O3.
The O 1s spectra before and after ozonation were deconvoluted into the peaks at 531.7-531.8 eV that were assigned to surface chemisorbed oxygen (Oads) and at 530.8-530.9 eV that were ascribed to lattice oxygen (Olatt) (Fig. 5d) (Garcia et al. 2013;Jiang et al. 2019). Due to the high reaction activity, surface chemisorbed oxygen played an important role in a series of organic substances oxidation reactions.
Obviously, the Oads species accounted for a much higher ratio than the Olatt species at the surface of the 1.0wt.% MnO2/γ-Al2O3 catalyst. The peak area ratio of Oads/(Oads+ Olatt) decreased from 77.9% before ozonation to 74.4% after ozonation (Table 1)

Mechanism of acetic acid ozonation over the MnO2/γ-Al2O3 catalyst
To study the mechanism of acetic acid ozonation on the surface of the MnO2/γ-Al2O3 catalyst, insitu DRIFTS was used to identify the functional groups on the catalyst surface. An acetic acid/water bubbler and O3 generator were used to feed the in-situ cell with a gas mixture of acetic acid, water, and O3 and to simulate the gas-liquid-solid reaction on the MnO2/γ-Al2O3 catalyst surface. Figs. 6a, 6b, and 6c show the DRIFT spectra of acetic acid ozonation within different durations on 0.5wt.%, 1.0wt.%, and 10wt.% MnO2/γ-Al2O3 catalysts at room temperature (25°C). The absorption in the υOH region (3200 cm -1 ) is attributed to the surface hydroxyl group, and the absorption peak at 3460 cm -1 was attributed to H2O (Zaki and Knözinger 1987). This corresponded to the reactive oxygen species of the ·OH group investigated above. The absorption peak at 1345 cm -1 is caused by the δCH3 vibration of the CH3C=O fraction, while the absorption peak at 1640 cm -1 is caused by the δ(H2O) vibration in liquid water phase on MnO2/γ-Al2O3 surface. The absorption peak at 1563 cm -1 is attributed to υCOO from acetic acid (Nakamoto 1970). The absorption peak at 982 cm -1 is due to superoxide O 2− 2 on the metal oxide (M 2+ -O

2−
2 ) (Li et al. 1990). The peak heights of the primary functional groups M 2+ -O 2− 2 at 982 cm -1 and COO at 1563 cm -1 are illustrated in Figs. 6d and 6e. As shown in Fig. 6d, the peak height of the reactive species M 2+ -O 2− 2 generated on the 1.0wt.% MnO2/γ-Al2O3 after 80 min was greater than that on the 0.5wt.% and 10wt.% MnO2/γ-Al2O3. It happened to verify once again that superoxide radicals are involved in the degradation process of acetic acid. In addition, the COO peak height on the 0.5wt.% MnO2/γ-Al2O3 increased with time, while that of the 1.0wt.% MnO2/γ-Al2O3 was relatively stable at a low level (Fig. 6e). This suggested that the COO species accumulated on the 0.5wt.% MnO2/γ-Al2O3 surface and could not be further oxidized due to a lack of reactive oxygen species, leading to an undesirable catalytic ozonation performance. The M 2+ -O 2− 2 peak was ascribed to an oxygen molecule adsorbed on the surface of MnO2 that was associated with oxygen vacancies and would play an important role in catalytic oxidation reactions of acetic acid (Sun et al. 2020). Thus, the peak height ratio of COO to M 2+ -O 2− 2 (COO/O 2− 2 ) is an index to show the ability of a catalyst to inhibit the accumulation of COO (acetic acid). Fig. 6f shows that the 1.0wt.% MnO2/γ-Al2O3 had the lowest COO/O 2− 2 ratio. This fact implies that the best performance of the 1.0wt.% MnO2/γ-Al2O3 was enhanced by the surface O 2− 2 superoxide. O 2− 2 (d) and COO (e); and the peak height ratio of COO/O 2− 2 (f) at various ozonation times To determine the main active species during acetic acid ozonation, TBA and PBQ radical quenchers were added to the acetic acid solution, and the results are shown in Fig. 7. It was noted that when the hydroxyl radical (·OH) quencher TBA (362 mM) was added, the degradation of acetic acid decreased from 88.1% to 15.7%. When the superoxide radical (O − 2 ) quencher PBQ (1.1 mmol/L) was added, the degradation decreased from 88.1% to 63.2%. This finding suggests that the hydroxyl radical (·OH) and superoxide radical (O − 2 ) species contributed to acetic acid degradation, and the ·OH species played a more important role than the O − 2 ·species. This result was consistent with the results in the section (Effect of pH) where the high energy efficiency was obtained at a low pH.

Fig. 7
Effect of the 1.0wt.% MnO2/γ-Al2O3 on the degradation rate of acetic acid after adding the free radical scavenger. Experimental condition: 1.0 g L −1 acetic acid concentration, 3.42 initial pH, 3.0 g 1.0wt.% MnO2/γ-Al2O3 dosage, 30.0-40.0 g Nm −3 O3 concentration. 25 °C ozonation temperature, 362 mM TBA, and 1.1 mM PBQ The acetic acid solution after ozonation in the presence of MnO2/γ-Al2O3 was analyzed using the HPLC. At 120 min of ozonation time, only oxalic acid could be found as the by-product of acetic acid oxidation (Fig. S5). The acetic acid ozonation was related to the adsorption of O3 and acetic acid on the surface of the MnO2/γ-Al2O3 catalyst. This result was similar to that of the experimental and theoretical research on the decomposition of acetic acid by pulsed DBD plasma reported by Matsui et al. (2011). It should be noted that the adsorption configuration of acetic acid on metal oxides is in a bidentate state (Almutairi et al. 2018).
The breakings of the C-H and C-C bands are the primary steps in acetic acid oxidation that produce oxalic acid and formic acid (Partenheimer et al. 2011). By combining the results of this study with the research on catalytic O3 decomposition Zhu et al. 2017) and acetic acid oxidation (Almutairi et al. 2018;Samskruthi et al. 2021) over an MnO2 catalyst, the mechanism of acetic acid catalytic ozonation is illustrated in Fig. 8. Eqs. 5-6 and 8-9 are O3-related surface reactions to produce reactive species Osurf such as surface M-O and M 2+ -O 2− 2 (Anpo et al. 2021;Che and Sojka 2001). Osurf is then supplied to acetic acid oxidation. Furthermore, O3 reacts with adsorbed water to form surface OH (Eqs. 8 and 9) and ·OH radicals (Eq. 10) (Zhu et al. 2017), which also contributes to acetic acid oxidation (Eq. 11). The surface oxygen species from the M 2+ -O 2− 2 group and ·OH radicals played an important role in the oxidation of acetic acid into CO2 and H2O.

Fig. 8 Mechanism of acetic acid ozonation over MnO2/γ-Al2O3
Kinetic study of acetic acid ozonation over the MnO2/γ-Al2O3 catalyst Catalytic ozonation can be described as a first order reaction (Cihanoğlu et al. 2015). According to the experimental results of acetic acid ozonation on 1.0wt.% MnO2/γ-Al2O3 catalyst at different ozonation temperatures (Fig. 4), we fitted the degradation (X) data using the first-order reaction. As shown in Fig.   9, ln(1−X) was linear to ozonation time t, and the linear coefficient R 2 were between 0.95−0.99. This finding implied that the acetic acid ozonation is exactly first order reaction (Eq. 13).
The activation energy (Ea) of acetic acid ozonation on 1.0wt.% MnO2/γ-Al2O3 catalyst was calculated to be 16.1 kJ mol -1 using Arrhenius equation. This activation energy level is close to 15.9 kJ mol -1 for acetic acid ozonation without a catalyst. This fact suggested that the acetic acid ozonation with or without catalyst may be controlled by the same kinetic step, perhaps the transformation of O3 from gas phase to liquid or on to catalyst surface.

Conclusions
The mineralization of acetic acid by catalytic ozonation was studied, the primary results are summarized as follows.
(2) Using 3.0 g of the 1.0wt.% Mn/γ-Al2O3 catalyst at 25°C to treat 100 mL of acetic acid at a concentration of 1.0 g L −1 , the degradation and mineralization of acetic acid reached as high as 88.4% at 300 min, with an average energy efficiency of approximately 14.9 g kWh −1 . However, without a catalyst, the degradation and mineralization of acetic acid were only 45.3% and 33.2%, respectively, and the energy efficiency was only 5.1 g kWh −1 .
(3) •OH radicals and surface-active oxygen species (such as O and Mn-O 2− 2 ) due to the O3 reaction played an important role in acetic acid ozonation to CO2 and H2O.
(4) The reaction temperature had a positive effect on improving the degradation, mineralization, and energy efficiency.
(5) A low pH favored acetic acid ozonation since •OH radicals could be effectively produced from O3 and H2O.