Performance and Mechanism of Chlorine Dioxide on BTEX Removal in Liquid and Indoor Air

With the development of the chemical industry, benzene, toluene, ethylbenzene, and xylene (BTEX) have gradually become the major indoor air pollutants. Various gas treatment techniques are widely used to prevent the physical and mental health hazards of BTEX in semi-enclosed spaces. Chlorine dioxide (ClO2) is an alternative to chlorine as a secondary disinfectant with a strong oxidation ability, a wide range of action, and no carcinogenic effects. In addition, ClO2 has a unique permeability which allows it to eliminate volatile contaminants from the source. However, little attention has been paid to the removal of BTEX by ClO2, due to the difficulty of removing BTEX in semi-enclosed areas and the lack of testing methods for the reaction intermediates. Therefore, this study explored the performance of ClO2 advanced oxidation technology on both liquid and gaseous benzene, toluene, o-xylene, and m-xylene. The results showed that ClO2 was efficient in the removal of BTEX. The byproducts were detected by gas chromatography-mass spectrometry (GC-MS) and the reaction mechanism was speculated using the ab initio molecular orbital calculations method. The results demonstrated that ClO2 could remove the BTEX from the water and the air without causing secondary pollution.


Introduction
As an important chemical raw material, benzene, toluene, ethylbenzene, and xylene (BTEX) are widely used in many industries, and their hazards are of increasing concern. Benzene was listed as a "known human carcinogen" (Group A) in the EPA's 1986 risk assessment guidelines and was classified as a known human carcinogen in 2005 [1]. Overexposure to benzene causes not only symptoms such as headaches, dizziness, drowsiness, confusion, tremors and loss of consciousness, moderate eye irritation, and skin irritation, but also increases the risk of leukemia [2,3]. Toluene not only causes photochemical haze and ozone depletion due to its high photochemical reactivity, but is also toxic, mutagenic, and carcinogenic within the human body [4,5]. The most commonly reported symptoms, after chronic inhalation exposure to xylenes, are depression of the visual and central nervous system, and respiratory irritation. Irritability, depression, insomnia, agitation, extreme tiredness, tremors, impaired concentration, and short-term memory issues may occur after long-term occupational exposure to xylene [6][7][8]. As typical volatile organic compounds (VOCs), BTEX causes severe air pollution in semi-enclosed areas such as supermarkets, offices, residences, workshops, etc. [9,10].
In recent years, extensive research has been carried out on treating gaseous BTEX. The adsorption method consists of removing BTEX pollution by adsorbing the contaminants onto the surface of the adsorbent. It has sufficient history in design and operation, strong applicability, and can transform pollutants into valuable products [11,12]. Shim et al., modified the surface-wrinkled silica nanoparticles (WSNs) with organosilane, to adsorb gaseous BTEX. At room temperature, 50 mg of hexadecyltrimethoxysilane (HDTMS)-WSN Figure 1 shows the removal effect of ClO 2 concentration on the liquid BTEX removal rate. Experimental results show that BTEX removal increases with increasing reaction time. In addition, as the concentration of ClO 2 increases, the BTEX removal rate revels the same trend of first increasing, and then decreasing. Figure 1a shows the removal rate of 10 mg/L liquid benzene by 30 mg/L-50 mg/L ClO 2 . The optimal degradation rate of benzene at 5 h is 47.22% at 45 mg/L ClO 2 , which is 10.29% higher than that of 40 mg/L ClO 2 . Figure 1b shows the removal rate of 10 mg/L liquid toluene by 12 mg/L-36 mg/L ClO 2 . The oxidation efficiency of ClO 2 on toluene is more pronounced than that of benzene. The highest removal rate for toluene was 42.77% at 30 mg/L ClO 2 for 5 h. ClO 2 concentration change has little effect on the toluene removal. When the concentration of ClO 2 increased to 36 mg/L, the removal rate of toluene decreased by 4.44%. Figure 1c shows the removal rate of 10 mg/L liquid o-xylene by 0.1 mg/L-5 mg/L ClO 2 . The optimal ClO 2 dosage required for o-xylene oxidation is only 0.5 mg/L, and the removal rate of o-xylene is 24.65%. Unlike toluene, ClO 2 concentration change has a significant effect on o-xylene removal. With the increase of the ClO 2 dosage from 0.5 mg/L to 5 mg/L, the removal rate of o-xylene gradually decreased. At a ClO 2 concentration of 5 mg/L, the o-xylene removal rate dropped to 0.05%. This suggests that the optimal concentration of ClO 2 for o-xylene is lower than that for benzene and toluene, due to the lower stability of o-xylene molecules. However, the low ClO 2 dosage may lead to premature oxidant depletion, reducing the o-xylene removal rate. Figure 1d shows the removal rate of 10 mg/L liquid benzene by 3 mg/L-7 mg/L ClO 2 . Compared to the o-xylene molecule, the m-xylene molecule exhibits a more significant symmetry and is therefore more stable than the o-xylene molecule. This results in a higher ClO 2 concentration required for the oxidation of m-xylene than o-xylene. The optimal ClO 2 dosage required for removing m-xylene is 5 mg/L, under which the removal rate of m-xylene can reach 41.24% within 5 h. Additionally, the effect of ClO 2 concentration on the removal rate of m-xylene was also significant. The removal rate of m-xylene decreased from 41.24% to 23.86% when ClO 2 concentration increased by only 1 mg/L. These results indicate that ClO 2 is effective in removing liquid BTEX as representative VOCs.  Figure 2 examined the removal rate of BTEX by optimal ClO2 dosage (45 mg/L, 30 mg/L, 0.5 mg/L, and 5 mg/L for benzene, toluene, o-xylene, and m-xylene, respectively) at 24 h. The experimental data show that the reaction rate increases slightly when using ClO2 to oxidize BTEX for more than 5 h. Compared with the result at 5 h, the removal rate of  2 examined the removal rate of BTEX by optimal ClO 2 dosage (45 mg/L, 30 mg/L, 0.5 mg/L, and 5 mg/L for benzene, toluene, o-xylene, and m-xylene, respectively) at 24 h. The experimental data show that the reaction rate increases slightly when using ClO 2 to oxidize BTEX for more than 5 h. Compared with the result at 5 h, the removal rate of benzene, toluene, o-xylene, and m-xylene at 10 h increased by 9.8%, 26.89%, 3.7%, and 7.01%, respectively. However, when the reaction continued for 24 h, the effect of ClO 2 removal on BTEX did not improve significantly, with removal rates increasing by 5.36%, 3.72%, 1.42%, and 2.65%, respectively. Thus, the optimal reaction time for oxidizing liquid BTEX with ClO 2 is 10 h.  Figure 2 examined the removal rate of BTEX by optimal ClO2 dosage (45 mg/L, 30 mg/L, 0.5 mg/L, and 5 mg/L for benzene, toluene, o-xylene, and m-xylene, respectively) at 24 h. The experimental data show that the reaction rate increases slightly when using ClO2 to oxidize BTEX for more than 5 h. Compared with the result at 5 h, the removal rate of benzene, toluene, o-xylene, and m-xylene at 10 h increased by 9.8%, 26.89%, 3.7%, and 7.01%, respectively. However, when the reaction continued for 24 h, the effect of ClO2 removal on BTEX did not improve significantly, with removal rates increasing by 5.36%, 3.72%, 1.42%, and 2.65%, respectively. Thus, the optimal reaction time for oxidizing liquid BTEX with ClO2 is 10 h.

Removal of Gaseous BTEX by ClO 2
The blank experiment was carried out under sealed and dark conditions, without adding ClO 2 . The results showed that the concentration of BTEX (4 mg/m 3 ) had not significantly changed in 10 h (the changing range was within 0.1%), and the influence on the experimental results was negligible. The blank experiment demonstrated the stability of o-xylene and the reliability of the reactor. Figure 3 shows the effect of ClO 2 concentration on the gaseous BTEX removal rate. Experimental data reveal that the removal rate of gaseous benzene revels a trend of first increasing and then decreasing with increasing ClO 2 concentration, which was the same as with liquid benzene. Figure 3a shows the removal rate of 4 mg/m 3 gaseous benzene by 10-50 mg/m 3 ClO 2 . The removal rate of benzene increases rapidly and then becomes stable after 2 h. When the ClO 2 dosage was 25 mg/m 3 , the removal effect of benzene reached a maximum of 81.75% in 10 h. Figure 3b shows the removal rate of 4 mg/m 3 gaseous toluene by 10-50 mg/m 3 ClO 2 . The figure demonstrates that the initial rate of the reaction was fast during the first 2 h, then decreased significantly and remained virtually constant. The highest toluene removal rate of 72.04% was achieved by 20 mg/m 3 ClO 2 in 10 h. Figure 3c shows the removal rate of 4 mg/m 3 gaseous o-xylene by 1-15 mg/m 3 ClO 2 . The initial reaction rate is relatively fast for the first 2 h, after which it decreases markedly. The highest o-xylene removal rate (64.06%) was observed with 8 mg/m 3 of ClO 2 and 10 h. Figure 3d shows the removal rate of 4 mg/m 3 gaseous m-xylene by 40-120 mg/m 3 ClO 2 . For the first 2 h, m-xylene reacts with ClO 2 at a relatively rapid rate. However, the removal rate of m-xylene within 2 h is about 15% lower than that of benzene, toluene, and o-xylene. After 2 h, the removal rate tends to rise slowly. The removal rate of m-xylene reached its highest level of 57.49% at a concentration of 80 mg/m 3 and 10 h of ClO 2 . action rate is relatively fast for the first 2 h, after which it decreases markedly. The highest o-xylene removal rate (64.06%) was observed with 8 mg/m 3 of ClO2 and 10 h. Figure 3d shows the removal rate of 4 mg/m 3 gaseous m-xylene by 40-120 mg/m 3 ClO2. For the first 2 h, m-xylene reacts with ClO2 at a relatively rapid rate. However, the removal rate of mxylene within 2 h is about 15% lower than that of benzene, toluene, and o-xylene. After 2 h, the removal rate tends to rise slowly. The removal rate of m-xylene reached its highest level of 57.49% at a concentration of 80 mg/m 3 and 10 h of ClO2.  In order to study the effect of ClO 2 removal on mixed BTEX gas, 10-90 mg/m 3 ClO 2 was used to oxidize 4 mg/m 3 each of benzene, toluene, o-xylene, and m-xylene as a gaseous mixture, as shown in Figure 4. From the data, it is evident that the BTEX removal increases with the reaction time. Similarly to the BTEX removal experiment studied above, the mixed gaseous BTEX reacts rapidly with ClO 2 within 2 h, followed by a significant slowing of the reaction rate after 2 h. This phenomenon shows that the optimal reaction time for oxidizing gaseous BTEX with ClO 2 is 2 h. In contrast to the results obtained using ClO 2 to oxidize a single gaseous BTEX, the optimal ClO 2 dosage and the optimal removal rate of mixed gaseous BTEX at 10 h change significantly, due to multiple BTEX interactions. Figure 5 shows the removal rate of a 4 mg/m 3 gaseous mixture of benzene, toluene, o-xylene, and m-xylene by 10-90 mg/m 3 ClO 2 in 10 h. It can be observed that the removal rates of benzene and toluene first increase and then decrease, with increasing ClO 2 concentration. Additionally, 10 mg/m 3 ClO 2 has a higher removal effect on o-xylene and m-xylene than 30 mg/m 3 . After the ClO 2 dosage exceeds 30 mg/m 3 , the removal rates of o-xylene and m-xylene first increase and then decrease with increasing ClO 2 concentration. When ClO 2 concentration was 50 mg/m 3 , the removal rates of benzene, toluene, and p-xylene reached the highest, which were 53.09%, 61.70%, and 61.27%, respectively. Under this condition, the removal rate of o-xylene was 58.54%, which was 1.08% lower than that of the optimal ClO 2 dosage (10 mg/m 3 ). Combined with the outcomes obtained from o-xylene oxidation by ClO 2 , this phenomenon is mainly due to the low optimal ClO 2 concentration required for o-xylene oxidation. The results showed that the optimal ClO 2 concentration of 4 mg/m 3 mixed BTEX was 50 mg/m 3 .
When ClO2 concentration was 50 mg/m 3 , the removal rates of benzene, toluene, and pxylene reached the highest, which were 53.09%, 61.70%, and 61.27%, respectively. Under this condition, the removal rate of o-xylene was 58.54%, which was 1.08% lower than that of the optimal ClO2 dosage (10 mg/m 3 ). Combined with the outcomes obtained from oxylene oxidation by ClO2, this phenomenon is mainly due to the low optimal ClO2 concentration required for o-xylene oxidation. The results showed that the optimal ClO2 concentration of 4 mg/m 3 mixed BTEX was 50 mg/m 3 .

BTEX Removal Mechanism
The above results confirmed that ClO2 has a strong ability to remove both liquid and gaseous BTEX. However, Ganiev et al., reported that ClO2 does not react directly with the simplest alkyl-substituted benzenes such as benzene, toluene, and xylene, which was inconsistent with the above data [40]. However, studies have revealed that not only ClO2, which has an oxidizing effect on contaminants, but also ·OH radicals, generated during the reaction, contribute significantly to the oxidation process together. Sun et al., performed the NO removal from fuel gas with ClO2 and found that ClO2 has an excellent oxidizing effect on NO. The research also demonstrated, through the electron spin resonance (ESR) and inhibitor (sodium formate and isopropanol), the presence of ·OH in the reaction process, and it has been concluded that ·OH had a significant effect on the removal of NO [43]. Marcon et al., investigated the mechanism of the formation of radicals during ClO2 decomposition. The results of electron paramagnetic resonance (EPR) spectroscopy revealed the presence of ·OH with ClO2. Only a small amount of ·OH was de-− −

BTEX Removal Mechanism
The above results confirmed that ClO 2 has a strong ability to remove both liquid and gaseous BTEX. However, Ganiev et al., reported that ClO 2 does not react directly with the simplest alkyl-substituted benzenes such as benzene, toluene, and xylene, which was inconsistent with the above data [40]. However, studies have revealed that not only ClO 2 , which has an oxidizing effect on contaminants, but also ·OH radicals, generated during the reaction, contribute significantly to the oxidation process together. Sun et al., performed the NO removal from fuel gas with ClO 2 and found that ClO 2 has an excellent oxidizing effect on NO. The research also demonstrated, through the electron spin resonance (ESR) and inhibitor (sodium formate and isopropanol), the presence of ·OH in the reaction process, and it has been concluded that ·OH had a significant effect on the removal of NO [43]. Marcon et al., investigated the mechanism of the formation of radicals during ClO 2 decomposition. The results of electron paramagnetic resonance (EPR) spectroscopy revealed the presence of ·OH with ClO 2 . Only a small amount of ·OH was detected with ClO − at the same condition, indicating that liquid ClO − reduced the production of ·OH. [44] According to research, ·OH is produced by the reaction of ClO 2 with H 2 O (Equations (1)- (7)). [44][45][46][47] Moreover, it is clear that ·OH is efficient in degrading benzene, toluene, and xylene [48][49][50][51][52]. Therefore, it could be speculated that the ·OH produced by ClO 2 with H 2 O plays a vital role in removing BTEX.
ClO 2 is soluble in water and exists in a stable dynamic equilibrium state. From a macroscopic point of view, ·OH, the intermediate product of the above disproportionation reaction, undergoes continuous production and disappearance. When adding BTEX to liquid and gaseous ClO 2 , ·OH (2.8 eV), which has a higher redox potential than ClO 2 (1.5 eV), will rapidly oxidize BTEX. The reaction not only removes BTEX but also considerably promotes the forward progress of Equation (1) [43]. Thus, a further increase in ClO 2 would reduce the speed of Equation (1) and promote the forward progress of Equations (2)-(4), resulting in a decrease in the ·OH concentration and a rapid reduction in the BTEX removal rate. The decrease in ·OH is one of the reasons why the BTEX removal rate decreases at high ClO 2 concentrations, which is in agreement with our experimental results in Figures 1 and 3. Additionally, the presence of ·OH in the oxidation system may promote the production of various free radicals, which will interact and further enhance the oxidizing ability of ClO 2 [43].
In order to analyze the oxidation byproducts of gaseous BTEX by ClO 2 , the gas chromatography-mass spectrometry (GC-MS) experiments were performed on gas samples from the reactor at 0 h, 0.5 h, 1 h, 4 h, and 10 h ( Figure 6, Table 1). Figure 6a indicates a trace of organic compounds, including dibutyl phthalate, n-hexadecane, n-heptadecane, n-heptadecyl acetate, and methyl palmitate were formed from 0.5-10 h during the oxidation of benzene. Figure 6b shows that n-hexadecane, n-heptadecane, n-heptadecyl acetate, and methyl palmitate were the main products from 0.5-10 h, during the oxidation of toluene. Figure 6c,d indicate that the concentrations of o-xylene and m-xylene decreased significantly with the increasing reaction time, while no representative intermediates were produced. It can be inferred from this phenomenon that most o-xylene and m-xylene are mineralized into CO 2 and H 2 O, and that some of the resulting intermediates have concentrations below the detection limit, in agreement with the results of Zou et al. [24]. tion of benzene. Figure 6b shows that n-hexadecane, n-heptadecane, n-heptadecyl acetate, and methyl palmitate were the main products from 0.5-10 h, during the oxidation of toluene. Figure 6c,d indicate that the concentrations of o-xylene and m-xylene decreased significantly with the increasing reaction time, while no representative intermediates were produced. It can be inferred from this phenomenon that most o-xylene and m-xylene are mineralized into CO2 and H2O, and that some of the resulting intermediates have concentrations below the detection limit, in agreement with the results of Zou et al. [24].                ·OH, were phenol, phenylcatechol, and 1,4-benzoquinone. In contrast, Jacoby et al., reported that the photocatalytic oxidation of benzene produced phenol, malic acid, hydroquinone, benzoic acid, and benzoquinone [53][54][55]. He et al., speculated that ·OH replaced an ·H molecule of benzene to produce phenol. After the ring-opening process, 1,3-butanediol, glycolic acid, and other small molecule alkanes were produced and finally mineralized into CO 2 and H 2 O [56]. Zhong et al., detected the presence of ethyl acetate, 2,5-cyclohexadiene1,4,diketone,2,6bis(1,1-dimethyl), butylated hydroxytoluene, and dibutyl phthalate from the GC-MS results of the photocatalytic degradation of gaseous benzene. In addition, it reveals that the long-chain alkanes produced by the ring-opening products further react with benzene and ·OH to produce dibutyl phthalate [57]. Based on the GC-MS results above, the oxidation mechanism of gaseous BTEX by ClO 2 was speculated and depicted in Scheme 1. The oxidation of benzene, via the attack of the six-membered ring with ·OH, produces phenol [48]. The phenol can be oxidized by ·OH to form intermediates such as phenol and benzoquinone. These byproducts further degrade into ring-opened species, such as acetaldehyde, 2-butanone, and other fragmented alkyl groups [56]. Most of these products will mineralize to form CO 2 and H 2 O, while some will produce species with higher carbon numbers, such as n-hexadecane, n-heptadecyl acetate, and methyl palmitate [57]. The oxidation of toluene is possible, due to the attack of the methyl with ·OH, which produces benzaldehyde and benzoic acid. The phenols then undergo a ring-opening reaction to produce CO 2 , H 2 O, and species with higher carbon numbers [58]. Another possible pathway for toluene is through the oxidation of ·OH, which leads to the break-off of the methyl group on the toluene to produce benzene, which is then further oxidized and decomposed [59]. There are two possible pathways for the degradation of o-xylene and m-xylene: (1) through the attack of the methyl with ·OH to form phenols such as o-quinone and m-quinone [60][61][62][63], and (2) through the break-off of the methyl group to produce toluene [64]. After the ring-opening process, most of the product will eventually mineralize into H 2 O and CO 2 [60,65,66]. ·OH attacks the six-membered ring of BTEX, which is a key factor in the decomposition and removal of BTEX in semi-enclosed spaces. Therefore, the translation states of the degradation pathway of benzene (Figure 7), the most stable compound of BTEX, with ·OH, was calculated to verify the reliability of the reaction pathway. The transition state and the relative thermodynamic properties of each reaction are computed as follows: in which G (X) and G ref are the Gibbs free energy of structure X, and the reference, respectively ( Figure 8 and Table 2). The first step of each reaction is considered as the reference.
The results show that, at 298.15 K and 1 atm, the initial relative Gibbs free energy of the ·OH-initiated reaction (AA* and BB*) is 96.24 KJ/mol and 110.32 KJ/mol, respectively, indicating that gaseous benzene will react with ·OH to produce catechol. Catechol is known to yield quinones in a water solution [57,67]. Morales-Roque et al., found that the degradation of catechol depends on the presence of ·O 2 − radicals that also exist in the disproportionation reaction of ClO 2 (Equation (7)) [68]. This feature of the reaction mechanism cannot be explained thoroughly, in view of the effect of ·OH, and requires a more precise investigation of the energy values of the reaction intermediates. This is the reason for the high initial relative Gibbs free energy of the reaction of catechol with ·OH (CD and DE, 214.75 KJ/mol and 209.07 KJ/mol, respectively). The research confirms that ·OH is able to react with BTEX in semi-enclosed areas. Furthermore, Ben Amor et al., have reported that o-quinone can undergo a ring-opening reaction under the oxidation of ClO 2 [69].
The ClO2 stock solution in this experiment was prepared daily, according to the standard method [70]. A total of 20 mL of sulfuric acid (10%) was slowly added into 0.013 g/mL of sodium chlorite solution, and the resulting gaseous ClO2 was discharged into a borosilicate glass collection bottle. The content of ClO2 was determined by an ultraviolet (UV) spectrophotometer (UV-8000S, Metash Co., Ltd, Shanghai, China) and the five-step
The ClO 2 stock solution in this experiment was prepared daily, according to the standard method [70]. A total of 20 mL of sulfuric acid (10%) was slowly added into 0.013 g/mL of sodium chlorite solution, and the resulting gaseous ClO 2 was discharged into a borosilicate glass collection bottle. The content of ClO 2 was determined by an ultraviolet (UV) spectrophotometer (UV-8000S, Metash Co., Ltd, Shanghai, China) and the five-step iodometric method. The characteristic UV absorption peak of ClO 2 is at 430 nm, and the ClO 2 was quantitative when using the external standard method. The prepared ClO 2 solution was placed in a sealed amber glass bottle and stored away from light.
Experiments were performed using a sealed reactor with a volume of 200 L (Figure 9). The reactants were added from the injection port (1). The reactor was equipped with two fans (2,3) to ensure the volatilization of BTEX. Samples were withdrawn from the sampling point (3) at the time for analysis. iodometric method. The characteristic UV absorption peak of ClO2 is at 430 nm, and the ClO2 was quantitative when using the external standard method. The prepared ClO2 solution was placed in a sealed amber glass bottle and stored away from light. Experiments were performed using a sealed reactor with a volume of 200 L ( Figure  9). The reactants were added from the injection port (1). The reactor was equipped with two fans (2,3) to ensure the volatilization of BTEX. Samples were withdrawn from the sampling point (3) at the time for analysis.

BTEX Removal Experiments
The liquid BTEX removal experiment was performed in a 250 mL sealed conical flask. 10 mg/L of BTEX solution and different concentrations of liquid ClO2 were added to the flask at room temperature. A magnetic stirrer ensured complete mixing. Samples were taken every 1 h and then analyzed immediately using high-performance liquid chromatography (HPLC). The BTEX concentration was calculated using the external standard method. Each process was repeated three times to ensure the reliability and repeatability of the results.
The decomposition ratio (η) of liquid BTEX was calculated as where C0 and Ct depict the concentration of liquid BTEX at 0 h and t h, respectively. The gaseous BTEX removal experiments were performed in the reactor mentioned above. The dosage of BTEX refers to the concentration of BTEX detected in the air of unventilated renovated houses. A total of 4 mg/m 3 of benzene, toluene, o-xylene, and mxylene was added to the reactor at room temperature, respectively. The fans were turned on to completely mix the gaseous BTEX. After an hour, a sample was taken as blank control to reduce experimental errors. The effects of ClO2 concentration and reaction time were investigated experimentally. The sampling and detection of benzene series are based on the standard HJ 584-2010: Ambient air-Determination of benzene and its analogies by activated charcoal adsorption carbon disulfide desorption and gas chromatography [71]. The products were collected by a 150 mg activated carbon tube (HPLC/ACS, Macklin Biochemical Co., Ltd, Shanghai, China) at a flow rate of 0.6 L/min, desorbed with 1 mL carbon disulfide solution for 1 h, and then analyzed using gas chromatography (GC) with a flame ionization detector (FID). BTEX was identified through its retention time and was quantitative, using the external standard method [72]. Each process was repeated three times to ensure the reliability and repeatability of the results.
The decomposition ratio (η) of gaseous BTEX was calculated as

BTEX Removal Experiments
The liquid BTEX removal experiment was performed in a 250 mL sealed conical flask. 10 mg/L of BTEX solution and different concentrations of liquid ClO 2 were added to the flask at room temperature. A magnetic stirrer ensured complete mixing. Samples were taken every 1 h and then analyzed immediately using high-performance liquid chromatography (HPLC). The BTEX concentration was calculated using the external standard method. Each process was repeated three times to ensure the reliability and repeatability of the results.
The decomposition ratio (η) of liquid BTEX was calculated as where C 0 and C t depict the concentration of liquid BTEX at 0 h and t h, respectively. The gaseous BTEX removal experiments were performed in the reactor mentioned above. The dosage of BTEX refers to the concentration of BTEX detected in the air of unventilated renovated houses. A total of 4 mg/m 3 of benzene, toluene, o-xylene, and m-xylene was added to the reactor at room temperature, respectively. The fans were turned on to completely mix the gaseous BTEX. After an hour, a sample was taken as blank control to reduce experimental errors. The effects of ClO 2 concentration and reaction time were investigated experimentally. The sampling and detection of benzene series are based on the standard HJ 584-2010: Ambient air-Determination of benzene and its analogies by activated charcoal adsorption carbon disulfide desorption and gas chromatography [71]. The products were collected by a 150 mg activated carbon tube (HPLC/ACS, Macklin Biochemical Co., Ltd, Shanghai, China) at a flow rate of 0.6 L/min, desorbed with 1 mL carbon disulfide solution for 1 h, and then analyzed using gas chromatography (GC) with a flame ionization detector (FID). BTEX was identified through its retention time and was quantitative, using the external standard method [72]. Each process was repeated three times to ensure the reliability and repeatability of the results.
The decomposition ratio (η) of gaseous BTEX was calculated as where C 0 and C t depict the concentration of gaseous BTEX at 0 h and t h, respectively.

Analytical Methods
All liquid BTEX samples were analyzed using a Thermo Scientific UltiMate 3000 HPLC with a UV detector. Separations were carried out using a thermostated (31.6 • C) 150 mm × 4.6 mm, 5 µm Thermo Hypurity C 18 column, with methanol and water (70:30) used as mobile phase, and the flow rate was 1.0 mL/min. The UV detection wavelength was set at 254 nm. All aqueous solvents were filtered through 0.45 µm membrane filters.
All gaseous BTEX samples were analyzed using a Techcomp GC-7900 with a PEG-20M column (30 m × 0.32 nm × 1.00 µm) and an FID detector. The nitrogen carrier gas and makeup gas flow rates were 2.6 mL/min and 30 mL/min, respectively. The injector temperature was 150 • C, and the detector temperature was 250 • C. The column temperature rose in line with the program with an initial temperature of 65 • C, maintained for 10 min, and rose at a rate of 5 • C/min to 90 • C and was held for 2 min.
All reaction products were analyzed using a Bruker EVOQ GC-TQ with HP-5MS column (30 m × 0.25 nm × 0.25 µm). The Helium carrier gas flow rate was 1 mL/min. The ion source temperature was 230 • C. The column temperature rose in line with the program, with an initial temperature of 55 • C maintained for 3.2 min, then rose at a rate of 15 • C/min to 110 • C, was held for 1 min, and then rose to 180 • C at 30 • C/min and was held for 4 min.

Calculation Methods
The thermodynamic properties of benzene and its byproducts have been computed using the Gaussian 09W program package [73]. The ab initio molecular orbital calculations (MP2) method, in combination with the 6-31G (2d, p) basis set, has been used to obtain the geometric optimization of all reactants, products, and transition states [74]. The vibrational frequencies were performed in the same level of theory for the optimized reactants, products, and transition states to characterize transition states (N imag = 1). The IRCs were calculated at a level of MP2/6-31G (2d,p) to ensure that the identified transition states smoothly connected the reactants and products [75]. To obtain more accurate energies, single point calculations were performed with the MP2 method with the larger def2-tzvpp basis set. Shermo software was used to calculate the thermodynamic variables at 1 atm and 298.15 K [76].

Conclusions
This paper verified that ClO 2 is a convenient and effective method to remove liquid and gaseous benzene, toluene, o-xylene, and m-xylene. For 4 mg/m 3 mixed BTEX gas, 50 mg/m 3 ClO 2 ensures the efficient removal of BTEX. With this ClO 2 dosage, the removal rates of benzene, toluene, o-xylene, and m-xylene reach 53.09%, 61.70%, 58.54%, and 61.27%, respectively. Moreover, GC-MS analysis detected small concentrations of n-hexadecane, n-heptadecane, n-heptadecyl acetate, and methyl palmitate in the oxidation products of benzene and toluene. No representative intermediates were detected in the GC-MS results of the oxidation products of o-xylene and m-xylene. It is suggested that the ·OH, produced by ClO 2 and H 2 O, played a significant role in the BTEX removal. Based on the results, the degradation mechanisms were speculated. This study illustrated the potential of ClO 2 for indoor air pollution removal, and provided an effective method to exenterate BTEX in semi-closed spaces and to improve the IAQ. Since high concentrations of ClO 2 can cause harm to humans, we recommend using ClO 2 at night or when the buildings are empty.