- Email: [email protected]
Oxidation of toluene by dielectric barrier discharge with photo-catalytic electrode
Chemical Engineering Journal 284 (2016) 166–173
Contents lists available at ScienceDirect
Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej
Oxidation of toluene by dielectric barrier discharge with photo-catalytic electrode Jie Chen ⇑, Zhengmiao Xie, Junhong Tang, Jie Zhou, Xianting Lu, Hongting Zhao Institute of Environment Science & Engineering, Hangzhou Dianzi University, Hangzhou 310018, PR China
h i g h l i g h t s A DBD reactor with photo-catalytic electrode was adopted to remove toluene. Photo-catalytic electrode improved toluene conversion dramatically. The reactor had good performance in mineralization rate and CO2 selectivity.
a r t i c l e
i n f o
Article history: Received 23 March 2015 Received in revised form 2 September 2015 Accepted 3 September 2015 Available online 8 September 2015 Keywords: Toluene Dielectric barrier discharge Photo-catalytic electrode Mineralization rate CO2 selectivity
a b s t r a c t A specially designed photo-catalytic electrode-incorporated-dielectric barrier discharge (DBD) reactor has been used to investigate the interaction between photo-catalytic electrode and DBD. Toluene conversion by oxidation was enhanced dramatically due to the synergy between DBD and photo-catalytic electrode. The photo-catalytic electrode could be fully activated in the non-thermal plasma at high peak voltages without extra UV light source. Peak voltage, initial concentration, residence time and humidity were important factors that influenced the toluene conversion. 160 mg/m3 toluene conversion reached 89.9% at a peak voltage of 40 kV and with a residence time of 4 s in the reactor photo-catalytic electrode. Photo-catalytic electrode could improve DBD system’s water-resistance by turning more water molecules into OH radicals. And it also made the reactor have better performance in mineralization rate and CO2 selectivity. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Volatile organic compounds (VOCs) are considered to be primary toxic air pollutants due to their detrimental effects on both human health and environment [1]. Toluene is a typical toxic VOC. It is an important organic chemical material with a pungent odor [2,3]. It is also often used as ingredient for paints and diluents. The inhalation of toluene can cause human nerve system damage [4]. Traditional VOCs control methods, such as wet scrubbing, active carbon adsorption, ozone oxidation and bio-filtration, are limited technically and economically [5,6]. Non-thermal plasma (NTP) technique application in VOCs control has been receiving increasing attention because of good economy and easy operation [7,8]. Many studies on removal of VOCs by NTP have been performed under different conditions such as target molecules, gas composition, reactor type and geometry, type of power supply, etc. [9–20]. For the practical application, it is also important to control the product of the ⇑ Corresponding author. Tel.: +86 571 86919158; fax: +86 571 86919185. E-mail address: [email protected] (J. Chen). http://dx.doi.org/10.1016/j.cej.2015.09.006 1385-8947/Ó 2015 Elsevier B.V. All rights reserved.
VOCs after treatment by NTP. However, some studies reported poor mineralization efficiencies of VOCs decomposed by NTP alone. Wallis et al. [21] measured no CO2 during the destruction of 500 ppm dichloromethane when using only plasma. Kim et al. [22] found the CO2 selectivity was 26.7% when benzene was decomposed by pulsed corona. Combining NTP with other pollution control techniques to overcome the drawbacks was becoming a new focus of researches [23–29]. Karuppiah et al. [30] oxidized diluted benzene by a dielectric barrier discharge (DBD) reactor with sintered metal fiber (SMF) electrode modified with transition metal oxides. For 50 ppm of benzene, the reactor with TiO2/MnOx/SMF showed 100% conversion at 230 J/L. Thevenet et al. [31] thought the interaction between the plasma and the material porosity (silica fibers) clearly enhances the elimination of acetylene at atmospheric pressure. And the mineralization into CO and CO2 could be improved by additional irradiation of the photo-catalyst. Subrahmanyam et al. [32,33] used a DBD reactor with SMF electrode modified with Ti, Mn and Co to oxidize VOCs (toluene, isopropanol and trichloroethylene). With CoOx and MnOx/SMF catalytic electrodes, 100% conversion of toluene was achieved at 235 J/L. However, catalyst
J. Chen et al. / Chemical Engineering Journal 284 (2016) 166–173
167
was prone to being poisoned. And the research on stabilization of SMF electrode modified with catalyst was rare. In this study, toluene was decomposed by a DBD reactor. SMF coated with TiO2 was used as discharging electrode. The synergy between DBD and photo-catalytic electrode was explored. Special consideration was given to some important parameters that may affect toluene conversion, including peak voltage, initial concentration, residence time and humidity. The mineralization rate and CO2 selectivity of toluene decomposed in various reactors were studied. A long-term experiment (25 d) was operated to investigate the stability of the photo-catalytic electrode.
specified impurities of H2O. The purity of toluene (Huadong Medicine Limited CO. LTD) was 99.9%. The concentration of anatase nano-TiO2 solution (VK-TG02, Wangjing Material Co. Ltd., China) was 10%. SMF (average pore radius of 5 lm, porosity of 65% and diameter of 100 lm) was supplied by Xi’an Filter Metal Materials Co. Ltd., China. And the TiO2/SMF-n electrode was prepared by a modified anodic oxidation method [24]. UV light was applied by a 125 W high-pressure mercury lamp (Shanghai Yamingdianqi Zhaoming Dianqi Limited CO. LTD) with wave length of 365 nm.
2. Experimental
The following standard procedure was developed to achieve stable and reproducible conditions for the gas flow rate and composition. The reactor was first flushed with air at 1 L/min for a few minutes. Toluene-containing gas was then allowed in at an appropriate flow rate for 30 min, at which point the toluene concentration measured at the reactor outlet was the same as that at its entry. The desired flow rate for the specific experiment was then set and the high voltage was applied. After a stabilization time of about 5 residence times to obtain a steady concentration of the residual odor at the reactor outlet, two consecutive samples of the treated gas were allowed to GC analysis. After a group of experiments was completed, the gas flow was stopped and the high voltage source was switched off. The apparatus was finally cleaned by flushing air at 1 L/min for 15 min. In experiments with humidified gas, preconditioning of the reactor was achieved by flushing humidified air through it for 30 min. Then, the toluenecontaining gas, humidified as described above, was introduced into the system. All the experiments above were achieved at ambient temperature.
2.1. Experimental setup and reactor Fig. 1 showed a schematic diagram of the experimental setup. Dry air from a gas cylinder was allowed into a buffer tank at a controlled flow rate through tubes. A fraction of the gas flow went through a toluene generator to carry out toluene with an appropriate level. For experiments with humid air, a portion of the air flow went through a bubbler filled with deionized water to an appropriate level. Then all the streams were mixed in the buffer tank, where humidity was measured by a hygrometer (Rotronic A1H). At last, the mixed stream was introduced to the DBD reactor. Samples of the treated gas for on-line chemical analyses were allowed into a gas chromatography (GC) from the sampling ports located at the inlet and the outlet of the reactor. The DBD reactor was a 24 mm (inner diameter) 200 mm quartz cylinder with a SMF wire of 1.5 mm in diameter fixed along its axis. The SMF wire served as the support of TiO2 and also an energized electrode. The ground electrode was a 150 mm copper film which was embedded outside of quartz cylinder. An AC high voltage source with frequency of 20 kHz was applied in the experiment. The voltage and current waveforms were measured by using a four channel Tecktronix TDS 2014B 350 MHz digital storage oscilloscope capable of sampling 1 GS/s (Giga Sample per second), a Texas HVP-3020 high voltage passive probe and a CT4 TCP202 current probe. 2.2. Materials Air (Jingong Co., Ltd.) was a synthetic mixture (80% nitrogen and 20% oxygen) from pure liquid nitrogen and pure liquid oxygen with
2.3. Experimental procedure
2.4. Chemical analyses The concentration of toluene was measured by the gas chromatograph GC7890II (Tianmei corporation) which was equipped with 6-port gas sample valve. Ozone formed in the DBD reactor was measured by an UV absorption ozone monitor (AP1-450 NEMA). The formation of CO and CO2 was monitored with an infrared gas analyzer (Siemens ultramat 23). A Fourier transform infrared (FTIR) (Ncolet6700) was used for gas qualitative analysis. OH radical in the DBD reactor was detected by salicylic acid method [34]. A film, which was impregnated with salicylic acid,
Fig. 1. Schematic diagram of the experimental setup. (1) Gas cylinder; (2) mass flow control; (3) buffer tank; (4) water bath tank; (5) toluene generator; (6) water vapor generator; (7) hygrometer; (8) sampling of inlet gas; (9) sampling of outlet gas; (10) DBD reactor; (11) UV lamp; (12) high voltage source; (13) digital storage oscilloscope; (14) current probe; (15) high voltage probe; (16) gas chromatograph; (17) computer.
168
J. Chen et al. / Chemical Engineering Journal 284 (2016) 166–173
was used to detect OH radical in plasma reactor [34]. Salicylic acid reacts with OH radical and produces 2,5-dihydroxybenzoic acid (2,5-DHBA). Then, a high performance liquid chromatography (HPLC) was carried out to detect the concentration of 2,5-DHBA. Therefore, OH radical in NTP reaction could be calculated. One piece of filter paper was cut to a calculated size (300 mm 45 mm), then a solution, which was made by 0.3 g salicylic acid dissolved in 10 mL anhydrous ethanol (99.9%), was dripped equably by a injector to this filter paper and dried. The prepared sampling film was puckered and then placed in plasma reactor. The collecting time was set as 3 h. A shorter or longer time made the results below the minimum detectable limits or inaccurate. After the sampling film was taken out and was cut to fragments, all these fragments were dipped in water (15 mL). After that, an ultrasonic cleaner was used to accelerate dissolution of the products. Then the concentration of 2,5-DHBA was detected by a HPLC, equipped with an auto sampler (model G1329A) and an ultraviolet spectroscopic photometer (300 nm) [34]. The concentration of OH radical could be calculated as the following equation
C OH
C 2;5-DHBA V L N ¼ F 2;5-DHBA F g t g ð1 aÞ
ð1Þ
where, C2,5-DHBA is the concentration of 2,5-DHBA (mol/L); VL is the volume of absorbing liquid (L); N is Avogadro constant (6.02 l023); F2,5-DHBA is the producing ratio of 2,5-DHBA (in this experiment, almost all salicylic acid that reacted with OH radical produced 2,5-DHBA, therefore, F2,5-DHBA was assumed to be 1); Fg is gas flow rate (mL/min); t is sampling time (min); g is pick-up efficiency (it was assumed to be 1 in this experiment because of almost all 2,5-DHBA picked up from the film); a is the product losing ratio (it was assumed to be 0). 3. Results and discussion 3.1. Decomposition of toluene in various reactors To understand the interaction between NTP and photo-catalytic electrode, toluene was decomposed in four types of experiments. Fig. 2 showed the evolution of toluene conversion as a function of peak voltages. The main observation was that the toluene conversion increased with the increase of peak voltage in all four types
of experiments. When peak voltage increased from 10 to 40 kV, the toluene conversion was promoted from 5.2% to 89.9% in the reactor with photo-catalytic electrode (SMF + TiO2). Similar effect could be observed in other reactors. Higher peak voltages produced more high-energy electrons, active radicals, ions and other excited species, which was favorable for the toluene removal process [7]. As seen from Fig. 2, reactors with Ni–Cr alloy electrode (diameter of 1.5 mm) and SMF electrode (diameter of 1.5 mm) performed similarly in toluene decomposition with conversions of 51.3% and 53.7% at 35 kV. It suggested SMF was suitable for discharge electrode as Ni–Cr alloy did. When TiO2 was coated onto SMF (Fig. 3), toluene conversion was promoted dramatically at a fixed peak voltages. The toluene conversion reached 78.4% in the reactor with photo-catalytic electrode at 35 kV. It was because that TiO2 could be activated by high-energy electrons, excited molecules or radicals [35]. And NTP could produce UV light due to excited nitrogen molecules as follows
N2 þ e ! N2 þ e
ð2Þ
N2 ! N2 þ hm
ð3Þ
Consequently, oxidation reactions, which promoted the toluene conversion, were triggered by formation of electron–hole pairs (h+ and e) on the catalysts surface [23,36]
TiO2 þ hm ! h þ e þ
þ
H2 O þ h ! OH þ Hþ
ð5Þ
Besides, the conversion disparity between the reactor with SMF and the reactor with SMF + TiO2 became larger when the peak voltage increased. The conversion in the reactor with SMF + TiO2 was larger than that in the reactor with SMF by 13.1% at 20 kV, which increased to 28.6% at 40 kV. NTP could not produce enough highenergy electrons, radicals and UV light (active species) to activate TiO2 at low peak voltages. At high peak voltages, photo-catalytic electrode was much more active due to more active species in reactor. In order to determine if plasma UV and excited species activated sufficiently photo-catalytic TiO2 on electrode, experiments were carried out by introducing an extra UV-lamp. The lamp was placed 80 mm above the reactor with photo-catalytic electrode. This process was presented in Fig. 2 as SMF + TiO2 + UV. The presence of extra UV light source modified the toluene conversion at low peak voltages. At a peak voltage of 10 kV, the toluene conversion in the reactor with SMF + TiO2 was 5.2%. After turning on the UV lamp, the conversion reached 18.9% at the same peak voltage. The conversion disparity was 13.7%. However, the conversion disparity reduced to 2.9% at 40 kV. It was discussed above that NTP could not activate photo-catalytic electrode at low peak voltage with little active species. Thus, an extra UV light souse was favorable for photo-catalytic electrode at low peak voltage. And photocatalytic electrode was activated by active species in reactor at high peak voltages, at which the effect of the extra UV light source weakened. As seen from the results presented in the Fig. 2, it was believed that the photo-catalytic electrode could be activated in the NTP at high peak voltages in our research system. Synergistic effect (hsg) was defined as following
hsg ¼ h3 ðh1 þ h2 Þ
Fig. 2. Comparison of toluene conversion in various reactors. (Initiate concentration: 160 mg/m3; relative humidity: 0%; residence time: 4 s; TiO2/SMF-n: 5 wt%.)
ð4Þ
ð6Þ
where h3 was the toluene conversion rate of reactor with photocatalytic electrode and UV lamp (%); h1 was the toluene conversion rate of reactor with SMF electrode (%); h2 was the toluene conversion rate of reactor with photo-catalytic electrode (at a peak voltage of 0 kV) and UV lamp (%).
J. Chen et al. / Chemical Engineering Journal 284 (2016) 166–173
169
Fig. 3. SEM images of SMF + TiO2 at 2000 (a) and 5000 (b). (TiO2/SMF-n: 5 wt%.)
Synergistic effect (hsg) between DBD and photo-catalytic electrode on removal of toluene was studied as a function of peak voltage varied in the range of 10–40 kV. Fig. 4 showed that hsg between DBD and photo-catalytic electrode on removal of toluene increased with increase of peak voltage at low peak voltages. The highest hsg of 25.5% was obtained at 30 kV. When the peak voltage was higher than 30 kV, hsg dropped to 20%. It could be explained by that as discussed above. photo-catalytic electrode was activated by active species in reactor at high peak voltages. Energy deposition (ED) is defined as follows [37]
I ¼ Cm 1 P¼ T ED ¼
Z
T 0
P Q
dV m dt Cm VIdz ¼ T
ð7Þ Z
T 0
dV m V ¼ f Cm dt
Z VdV m
ð8Þ
ð9Þ
where I was the instantaneous current (A); Cm was the known capacitor (F); Um was the instantaneous voltage of the capacitor (V); P was the discharge power (W); T was the alternating current cycle (s); U was the instantaneous voltage of the DBD reactor (V); f was the frequency (Hz); ED was the energy deposition (J/L); Q was the gas flow rate (L/s);
Fig. 4. The synergistic effect between DBD and photo-catalytic electrode on removal of toluene.
Fig. 5 showed the variation of energy deposition as a function of applied voltage for two reactors. With the increasing of peak voltage, energy deposition of two kinds of reactor increased. The energy deposition of the reactor with SMF + TiO2 was larger than that of reactor with SMF. When peak voltage was 60 kV, the energy deposition of the reactors with SMF + TiO2 and SMF were 154 and 121 J/L. It was believed that the nano-TiO2 coated on the SMF was polarized. And an intense electric field was formed on the surface of the TiO2, resulting in partial discharge. Therefore, the energy deposition of the reactor with SMF + TiO2 was promoted. 3.2. Photo-catalytic electrode with different amount of TiO2 loaded on Oxidation of toluene in DBD with different amount of TiO2 loading on SMF (3–5%) was shown in Fig. 6. The toluene conversion increased with the increase of the amount of TiO2 loading on SMF at a fixed peak voltage. At 35 kV, the toluene conversions with TiO2/SMF-3, TiO2/SMF-4 and TiO2/SMF-5 were 65.8%, 72.6% and 78.4%, respectively. It implied that the amount of TiO2 loaded on photo-catalytic electrode might not be sufficient for the system. More TiO2 (more than 5 wt%) loaded on photo-catalytic electrode might further promote toluene conversion. However, as seen from Table 1, some of the TiO2 loaded on TiO2/SMF-6 was detached in the ultrasonic test. The ultrasonic test was performed to evaluate the adherence of TiO2/SMF-n. TiO2/SMF-n was placed in an ultrasound cleaner full of ethanol for a 45 min. Judging from the results of removal experiments and detachment experiments the
Fig. 5. Variation of energy deposition as function of peak voltage. (TiO2/SMF-n: 5 wt %; flow rate: 1000 mL/min.)
170
J. Chen et al. / Chemical Engineering Journal 284 (2016) 166–173
Fig. 7. Effect of initiate concentration on conversion of toluene. (Residence time: 4 s; relative humidity: 0%; TiO2/SMF-n: 5 wt%.) Fig. 6. Oxidation of toluene in DBD with TiO2/SMF-n. (Initiate concentration: 160 mg/m3; relative humidity: 0%; residence time: 4 s.)
optimum amount of TiO2 loaded on photo-catalytic electrode was 5 wt%. 3.3. Effect of initial concentration and residence time on removal of toluene Removal of toluene was investigated in the DBD reactor with photo-catalytic electrode under various initial concentration and residence time. The initial concentrations of toluene were 95, 160 and 280 mg/m3. And the residence time of toluene in reactor were 1, 2, 4 and 6 s, which corresponded to the flow rate of 4, 2, 1 and 0.7 L/min, respectively. Fig. 7 showed that the initial toluene concentration was an important factor that influenced the toluene removal process. At a fixed peak voltage, the removal rate of toluene decreased with an increase in the initial concentration, while the amount of toluene removed increased. At 35 kV, the toluene removal rate of 95, 160 and 280 mg/m3 were 92.6%, 78.4% and 51.8%, respectively. And the amount of toluene removed increased from 42.6 to 70.22 mg/h. The phenomenon could be explained by that with the same amount of active species produced in the processed stream, the competition of high-energy electrons and active species among toluene molecules became fiercer for higher toluene concentration. It caused the reduction of toluene removal rate. On the other hand, more toluene molecules were introduced into the reactor with higher toluene initial concentration. Active species in the reactor were much more prone to collide with toluene molecules leading to high energy yield. Thus, the amount of toluene removed increased. Fig. 8 showed the relationship between the toluene removal rate and the residence time. The removal rate of toluene increased sharply with an increase of residence time at a fixed peak voltage.
For a peak voltage of 30 kV, the toluene removal rate reached to 78.3% when the residence time was 6 s. And it dropped to 38.9% when the residence time was 1 s. Most of the reactive species in DBD had a short lifetime and could only exist with permanent energy supply. And toluene would decompose deficiently without sufficient time to react with high energy electrons and reactive species. 3.4. Effect of humidity on decomposition of toluene The effect of humidity on toluene decomposition was examined by passing dry and humidified air (0–80%) through the DBD reactor with/without TiO2. The room and experimental system temperature were control at 298 K. Fig. 9 showed that humidity affected toluene decomposition differently in DBD system with/without TiO2. For a peak voltage of 30 kV in DBD system without TiO2, the toluene removal rate was 27% in dry air. And the toluene removal rate decreased to 20% as the humidity increased to 80% relative humidity. This indicated that high humidity had an inhibition to toluene decomposition in DBD system without TiO2. Water molecules had electronegative characteristics [35]. High humidity
Table 1 Detachment experiment of TiO2/SMF-n.
TiO2/SMF-4 TiO2/SMF-5 TiO2/SMF-6
Mb
Ma
2.2042 2.221 2.2472
2.2039 2.2207 2.2295
Mb is the mass of TiO2/SMF-n before ultrasonic test. Ma is the mass of TiO2/SMF-n after ultrasonic test.
Fig. 8. Effect of residence time on conversion of toluene. (Initiate concentration: 160 mg/m3; relative humidity: 0%; TiO2/SMF-n: 5 wt%.)
J. Chen et al. / Chemical Engineering Journal 284 (2016) 166–173 TiO2
H2 O þ hm ! H þ OH
Fig. 9. The effect of humidity on conversion of toluene. (Initiate concentration: 160 mg/m3; residence time: 4 s; TiO2/SMF-n: 5 wt%.)
in the discharging zone limited the electron density and quenched the activated chemical species, resulting in low toluene removal rate. However, a promoting effect of humidity on toluene decomposition was observed in DBD system with TiO2. For a peak voltage of 30 kV, the conversion of toluene was 55.6% in dry air. The conversion of toluene was promoted to 69.5% as the humidity increases to 20%. With the increase of humidity (20–80%), the conversions of toluene all keep above 64%. Similar results were obtained for other peak voltages. In order to investigate the role of humidity in the DBD system with/without TiO2, OH radical density in the reactor was measured. Fig. 10 showed the OH density as a function of humidity in the DBD system. In the system without TiO2, OH density increased with an increase of humidity at lower humidity (620%). And OH production was saturated as humidity increases at a higher humidity. In the system with TiO2, the OH radical density was higher than that in the system without TiO2. H2O molecules collided with high energy electrons and form OH radicals in the DBD system
H2 O þ e ! H þ OH þ e
ð10Þ
when TiO2 was introduced onto the electrode, OH radicals also could be produced by following reaction:
Fig. 10. OH radical density as a function of relative humidity in the DBD system. (Balance gas: dry air; residence time: 4 s; TiO2/SMF-n: 5 wt%; peak voltage: 37.5 kV.)
171
ð11Þ
OH was extraordinarily active and takes part in toluene decomposition processes [19]. What’s interesting was that the toluene conversion rate did not increase when the humidity increases from 20% to 80%. In the same time there was a 50% increase of the concentration of OH. It implied that next to OH formation, also other processes played an important role towards toluene decomposition. Ozone played an important role in plasma chemical process. In order to investigate the role of ozone in our case, ozone was injected downstream of the reactor without discharge. Ozone was produced by an ICAN CFY-3 ozone generator. The initial O3 flow rate and concentration were 48 mL/min and 30 g/m3, respectively. Then the O3 flow was mixed with total stream of 1000 mL/min to get the O3 concentration of 1450 mg/m3. When the initial concentration of toluene was 160 mg/m3, the conversation of toluene reached to 28%. Thus, the experimental results indicated that the toluene could react with ozone. On the other hand, humidity also strongly affected forming ozone in plasma system. Water molecules might compete with oxygen for high energy electrons. The ozone concentration in the plasma outlet gas stream was 1542 mg/m3 in dry air at 37.5 kV. The ozone produced in the plasma decreased at higher humidity. The ozone concentration monitored was 1164 mg/m3 at 40% humidity and 980 mg/m3 at 80% humidity. Since more ozone in reactor could promote the decomposition of toluene, high humidity, resulting in inhibiting ozone production, leaded to an adverse effect on toluene decomposition. It could be seen that different factors codetermined the decomposition of toluene under humid conditions. Generally, photocatalytic electrode could improve DBD system’s water-resistance by turning more water molecules into OH radicals. 3.5. Product analysis Judging from the gas analysis using FTIR toluene mostly decomposed into CO2 and CO in DBD reactor with SMF + TiO2. Formic acid (HCOOH) and formaldehyde (HCOH) were also detected in the outlet gas at low peak voltages from the FTIR analysis. The amount of HCOOH and HCOH decreased with raising peak voltages. To evaluate the mineralization rate of the toluene removal system, the MR was defined as follows
MR ð%Þ ¼
½CO2 þ ½CO 7ð½Toluenein ½Tolueneout Þ
ð12Þ
where the MR (%) is mineralization rate; [Toluene]in (mg/m3) and [Toluene]out (mg/m3) indicate inlet and outlet concentration of Toluene, respectively; [CO2] (mg/m3) and [CO] (mg/m3) indicate outlet concentration of CO2 and CO, respectively. Fig. 11 showed the mineralization rate of three reactors as a function of peak voltage. Mineralization rate were low at 20 kV and increased with increasing peak voltage for all three reactors. In the reactor with SMF, mineralization rate was 38.6% at 20 kV and increased to 79.4% at 40 kV. The reactor with SMF + TiO2 and the reactor with SMF + TiO2 + UV revealed better mineralization rate. Mineralization rate of the reactor with SMF + TiO2 and the reactor with SMF + TiO2 + UV were 92.4% and 93.2%, respectively. The results suggested that toluene mostly converted to other byproducts except for CO and CO2 in the reactor with SMF. While good mineralization rate in the reactor with SMF + TiO2 and the reactor with SMF + TiO2 + UV also indicated that the formation of other reaction byproducts such as aerosol and smaller organic compounds was negligible. It was desirable to achieve deep oxidation to CO2 rather than CO in the decomposition of VOCs. CO2 selectivity was defined as follows
172
J. Chen et al. / Chemical Engineering Journal 284 (2016) 166–173
Fig. 11. Comparison of mineralization rate. (Balance gas: dry air; initiate concentration: 160 mg/m3; residence time: 4 s; TiO2/SMF-n: 5 wt%; experiment temperature: 298 K.)
SCO2
½CO2 ð%Þ ¼ 100% 7ð½Toluenein ½Tolueneout Þ
Fig. 13. Comparison of the evolution of the toluene conversion in different reactors. (Initial concentration of toluene: 160 mg/m3; residence time: 4 s; peak voltage: 40 kV; humidity: 0%.)
3.6. Long-term operation
ð13Þ
where the SCO2 (%) was CO2 selectivity. As seen from Fig. 12, CO2 selectivity was enhanced with the increase of peak voltage for all three reactors. The CO2 selectivity of reactor with SMF ranged from 18.1% to 50.2%, which was much lower than that of the other two reactors. It indicated that SMF + TiO2 electrode was effective in increasing the CO2 selectivity as well as enhancing the toluene decomposition. The reactor with SMF + TiO2 and the reactor with SMF + TiO2 + UV nearly got the same high CO2 selectivity at 40 kV. They were all above 90%. It implied that NTP could activate TiO2 at high peak voltages. While the CO2 selectivity of reactor with SMF + TiO2 + UV was higher than that of reactor without UV source by 18%. It was believed that NTP could not produce enough active species to activate TiO2 at low peak voltages, resulting in TiO2 not working on SMF.
Fig. 13 showed the results of long-term operation experiments of two reactors. Every reactor ran 8 h/d with a fixed peak voltage of 40 kV in the experiment. Gas samples were measured before and after reactors every hour. As seen from the Fig. 12, the reactor with SMF + TiO2 exhibited higher toluene conversion rate and performance stabilization. The reactor with SMF + TiO2 got a toluene conversion rate of 89.9% at the first day. And the conversion remained 89.8% at the 25th day. While the conversion of the reactor with SMF dropped from 61.3% to 54.2% during the long-term operation (25 d). After long-term operation, the SMF discharge electrode in the reactor was covered with brown oily materials which were polymerized by intermediates. The polymer attached on the electrode weakened discharge, resulting in toluene conversion reduction. And yet the SMF + TiO2 discharge electrode was not covered with any polymer. It was because that irradiated SMF–TiO2 had high energy content than the band-gap. And it produced electron–hole pairs which led to redox reactions between separated electron–hole pairs and trapped VOCs. Thus, VOC molecules and intermediates colliding with SMF–TiO2 electrode would decompose completely, avoiding damaging to discharge electrode. 4. Conclusion A DBD reactor with photo-catalytic electrode was applied for removal of toluene. Toluene conversion was enhanced dramatically due to the photo-catalytic electrode. At a peak voltage of 40 kV, the toluene conversion of 89.9% was obtained in the novel reactor. Peak voltage, initial concentration and residence time were important factors that influenced the toluene conversion. Photocatalytic electrode improved DBD reactor’s water-resistance. And the reactor with photo-catalytic electrode had better performance in mineralization rate and CO2 selectivity. Acknowledgments
Fig. 12. The CO2 selectivity of three different reactors. (Balance gas: dry air; initiate concentration: 160 mg/m3; residence time: 4 s; TiO2/SMF-n: 5 wt%; experiment temperature: 298 K.)
We gratefully acknowledge the financial support of the Natural Science Foundation of Zhejiang province (Project No. LQ13B070 006), the National Natural Science Foundation of China (Project No. 21406044, No. 21307023, No. 41373121), the Innovative Team Foundation of Hangzhou Dianzi University (Project No. ZX130203318006).
J. Chen et al. / Chemical Engineering Journal 284 (2016) 166–173
References [1] H.Z. Streicher, P.A. Gabow, A.H. Moss, D. Kono, W.D. Kaehny, Syndromes of toluene sniffing in adults, Ann. Intern. Med. 94 (1981) 758–762. [2] M. Lu, R. Huang, J. Wu, M. Fu, L. Chen, D. Ye, On the performance and mechanisms of toluene removal by FeOx/SBA-15-assisted non-thermal plasma at atmospheric pressure and room temperature, Catal. Today 242 (2015) 274– 286. [3] T. Masato, H. Manabu, A. Masakazu, Efficient removal of toluene and benzene in gas phase by the TiO2/Y-zeolite hybrid photocatalyst, J. Hazard. Mater. 237 (2012) 133–139. [4] M.J. Munoz-Batista, M. Fernandez-Garcia, A. Kubacka, Promotion of CeO2–TiO2 photoactivity by g-C3N4: ultraviolet and visible light elimination of toluene, Appl. Catal. B 164 (2015) 61–270. [5] J. Chen, J. Yang, H. Pan, Q. Su, Y. Liu, Y. Shi, Abatement of malodorants from pesticide factory in dielectric barrier discharges, J. Hazard. Mater. 177 (2010) 908–913. [6] W. Liang, J. Li, J. Li, Y. Jin, Abatement of toluene from gas streams via ferroelectric packed bed dielectric barrier discharge plasma, J. Hazard. Mater. 170 (2009) 633–638. [7] T. Yamamoto, K. Ramanathan, P.A. Lawless, D.S. Ensor, J.R. Newsome, N. Plaksn, G.H. Ramsey, Control of volatile organic compounds by an AC energized ferroelectric pellet reactor and a pulsed corona reactor, IEEE Trans. Ind. Appl. 28 (1992) 528–534. [8] J. Chen, Q. Su, H. Pan, J. Wei, X. Zhang, Y. Shi, Influence of balance gas mixture on decomposition of dimethyl sulfide in a wire-cylinder pulse corona reactor, Chemosphere 75 (2009) 261–265. [9] Y.F. Guo, D.Q. Ye, Y.F. Tian, K.F. Chen, Humidity effect on toluene decomposition in a wire-plate dielectric barrier discharge reactor, Plasma Chem. Plasma Process. 26 (2006) 237–249. [10] S.B. Han, T. Oda, Decomposition mechanism of trichloroethylene based on byproduct distribution in the hybrid barrier discharge plasma process, Plasma Sources Sci. Technol. 16 (2007) 413–421. [11] K. Hensel, S. Katsura, A. Mizuno, DC microdischarges inside porous ceramics, IEEE Trans. Plasma Sci. 33 (2005) 574–575. [12] C. Hibert, I. Gaurand, O. Motret, J.M. Pouvesle, [OH(X)] measurements by resonant absorption spectroscopy in a pulsed dielectric barrier discharge, J. Appl. Phys. 85 (1999) 7070–7075. [13] J. Jarrige, P. Vervisch, Decomposition of gaseous sulfide compounds in air by pulsed corona discharge, Plasma Chem. Plasma Process. 27 (2007) 241–255. [14] H.H. Kim, Nonthermal plasma processing for air-pollution control: a historical review, current issues and future prospects, Plasma Process. Polym. 1 (2004) 91–110. [15] H.M. Lee, M.B. Chang, Abatement of gas-phase p-xylene via dielectric barrier discharges, Plasma Chem. Plasma Process. 23 (2003) 541–558. [16] E. Marotta, A. Callea, M. Rea, C. Paradisi, DC corona electric discharges for air pollution control. Part 1. Efficiency and products of hydrocarbon processing, Environ. Sci. Technol. 41 (2007) 5862–5868. [17] T. Oda, T. Takahashi, K. Tada, Decomposition of dilute trichloroethylene by nonthermal plasma, IEEE Trans. Ind. Appl. 35 (1999) 373–379. [18] R. Ono, T. Oda, Measurement of hydroxyl radicals in pulsed corona discharge, J. Electrostat. 55 (2002) 333–342. [19] K. Urashima, J.S. Chang, Removal of volatile organic compounds from air streams and industrial flue gases by non-thermal plasma technology, Trans. Dielectr. Electr. Insul. 7 (2000) 602–614.
173
[20] G.P. Van-Durme, B.F. McNamara, C.G. McGinley, Bench-scale removal of odor and volatile organic compounds at a composting facility, Water Environ. Technol. 64 (1992) 19–27. [21] A.E. Wallis, J.C. Whitehead, K. Zhang, The effect of temperature on the removal of DCM using non-thermal, atmospheric-pressure plasma-assisted catalysis, Appl. Catal. B 74 (2007) 111–116. [22] H.H. Kim, H. Kobara, A. Ogata, S. Futamura, Comparative assessment of different nonthermal plasma reactors on energy efficiency and aerosol formation from the decomposition of gas-phase benzene, IEEE Trans. Ind. Appl. 41 (2005) 206–214. [23] J. Van-Durme, J. Dewulf, C. Leys, H. Van-Langenhove, Conbining non-thermal plasma with heterogeneous catalysis in waste gas treatment: a review, Appl. Catal. B 78 (2008) 324–333. [24] Z. Ye, C. Wang, Z. Shao, Q. Ye, Y. He, Y. Shi, A novel dielectric barrier discharge reactor with photocatalytic electrode based on sintered metal fibers for abatement of xylene, J. Hazard. Mater. 241 (2012) 216–223. [25] L. Huang, L. Xia, X. Ge, H. Jing, W. Dong, H. Hou, Removal of H2S from gas stream using combined plasma photolysis technique at atmospheric pressure, Chemosphere 88 (2012) 229–234. [26] F. Thevenet, O. Guaitella, E. Puzenat, C. Guillard, A. Rousseau, Influence of water vapour on plasma/photocatalytic oxidation efficiency of acetylene, Appl. Catal. B 84 (2008) 813–820. [27] O. Guaitella, F. Thevenet, E. Puzenat, C. Guillard, A. Rousseau, C2H2 oxidation by plasma/TiO2 combination: influence of the porosity and photocatalytic mechanisms under plasma exposure, Appl. Catal. B 80 (2008) 296–305. [28] M. Ksibi, S. Rossignol, J.M. Tatibouët, C. Trapalis, Synthesis and solid characterization of nitrogen and sulfur-doped TiO2 photocatalysts active under near visible light, Mater. Lett. 62 (2008) 4204–4206. [29] J. Chen, Z.M. Xie, Removal of H2S in a novel dielectric barrier discharge reactor with photocatalytic electrode and activated carbon fiber, J. Hazard. Mater. 261 (2013) 38–43. [30] J. Karuppiah, E. Lingareddy, L. Sivachandiran, R. Karvembu, Ch. Subrahmanyam, Nonthermal plasma assisted photocatalytic oxidation of dilute benzene, J. Chem. Sci. 124 (2012) 841–845. [31] F. Thevenet, O. Guaitella, C. Guillard, E. Puzenat, G. Stancu, J. Roepcke, A. Rousseau, Comparison of the plasma-photocatalyst synergy at low and atmospheric pressure, Int. J. Plasma Environ. Sci. Technol. 1 (2007) 52–56. [32] Ch. Subrahmanyam, M. Magureanu, D. Laub, A. Renken, L. Kiwi-Minsker, Nonthermal plasma abatement of trichloroethylene enhanced by photocatalysis, J. Phys. Chem. C 111 (2007) 4315–4318. [33] Ch. Subrahmanyam, Catalytic non-thermal plasma reactor for total oxidation of volatile organic compounds, Indian J. Chem. 48 (2009) 1062–1068. [34] Y.F. Guo, X.B. Liao, D.Q. Ye, Detection of hydroxyl radical in plasma reaction on toluene removal, J. Environ. Sci. 20 (2008) 1429–1432. [35] H.H. Kim, Y.H. Lee, A. Ogata, S. Futamura, Plasma-driven catalyst processing packed with photocatalyst for gas-phase benzene decomposition, Catal. Commun. 4 (2003) 347–351. [36] A.L. Linsebigler, G.Q. Lu, J.T. Yates, Photocatalysis on TiO2 surfaces: principles, mechanisms, and selected results, Chem. Rev. 95 (1995) 735–758. [37] D. Evans, L.A. Rosocha, G.K. Anderson, J.J. Coogan, M.J. Kushner, Plasma remediation of trichloroethylene in silent discharge plasmas, J. Appl. Phys. 74 (1993) 5378–5386.
Contents lists available at ScienceDirect
Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej
Oxidation of toluene by dielectric barrier discharge with photo-catalytic electrode Jie Chen ⇑, Zhengmiao Xie, Junhong Tang, Jie Zhou, Xianting Lu, Hongting Zhao Institute of Environment Science & Engineering, Hangzhou Dianzi University, Hangzhou 310018, PR China
h i g h l i g h t s A DBD reactor with photo-catalytic electrode was adopted to remove toluene. Photo-catalytic electrode improved toluene conversion dramatically. The reactor had good performance in mineralization rate and CO2 selectivity.
a r t i c l e
i n f o
Article history: Received 23 March 2015 Received in revised form 2 September 2015 Accepted 3 September 2015 Available online 8 September 2015 Keywords: Toluene Dielectric barrier discharge Photo-catalytic electrode Mineralization rate CO2 selectivity
a b s t r a c t A specially designed photo-catalytic electrode-incorporated-dielectric barrier discharge (DBD) reactor has been used to investigate the interaction between photo-catalytic electrode and DBD. Toluene conversion by oxidation was enhanced dramatically due to the synergy between DBD and photo-catalytic electrode. The photo-catalytic electrode could be fully activated in the non-thermal plasma at high peak voltages without extra UV light source. Peak voltage, initial concentration, residence time and humidity were important factors that influenced the toluene conversion. 160 mg/m3 toluene conversion reached 89.9% at a peak voltage of 40 kV and with a residence time of 4 s in the reactor photo-catalytic electrode. Photo-catalytic electrode could improve DBD system’s water-resistance by turning more water molecules into OH radicals. And it also made the reactor have better performance in mineralization rate and CO2 selectivity. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Volatile organic compounds (VOCs) are considered to be primary toxic air pollutants due to their detrimental effects on both human health and environment [1]. Toluene is a typical toxic VOC. It is an important organic chemical material with a pungent odor [2,3]. It is also often used as ingredient for paints and diluents. The inhalation of toluene can cause human nerve system damage [4]. Traditional VOCs control methods, such as wet scrubbing, active carbon adsorption, ozone oxidation and bio-filtration, are limited technically and economically [5,6]. Non-thermal plasma (NTP) technique application in VOCs control has been receiving increasing attention because of good economy and easy operation [7,8]. Many studies on removal of VOCs by NTP have been performed under different conditions such as target molecules, gas composition, reactor type and geometry, type of power supply, etc. [9–20]. For the practical application, it is also important to control the product of the ⇑ Corresponding author. Tel.: +86 571 86919158; fax: +86 571 86919185. E-mail address: [email protected] (J. Chen). http://dx.doi.org/10.1016/j.cej.2015.09.006 1385-8947/Ó 2015 Elsevier B.V. All rights reserved.
VOCs after treatment by NTP. However, some studies reported poor mineralization efficiencies of VOCs decomposed by NTP alone. Wallis et al. [21] measured no CO2 during the destruction of 500 ppm dichloromethane when using only plasma. Kim et al. [22] found the CO2 selectivity was 26.7% when benzene was decomposed by pulsed corona. Combining NTP with other pollution control techniques to overcome the drawbacks was becoming a new focus of researches [23–29]. Karuppiah et al. [30] oxidized diluted benzene by a dielectric barrier discharge (DBD) reactor with sintered metal fiber (SMF) electrode modified with transition metal oxides. For 50 ppm of benzene, the reactor with TiO2/MnOx/SMF showed 100% conversion at 230 J/L. Thevenet et al. [31] thought the interaction between the plasma and the material porosity (silica fibers) clearly enhances the elimination of acetylene at atmospheric pressure. And the mineralization into CO and CO2 could be improved by additional irradiation of the photo-catalyst. Subrahmanyam et al. [32,33] used a DBD reactor with SMF electrode modified with Ti, Mn and Co to oxidize VOCs (toluene, isopropanol and trichloroethylene). With CoOx and MnOx/SMF catalytic electrodes, 100% conversion of toluene was achieved at 235 J/L. However, catalyst
J. Chen et al. / Chemical Engineering Journal 284 (2016) 166–173
167
was prone to being poisoned. And the research on stabilization of SMF electrode modified with catalyst was rare. In this study, toluene was decomposed by a DBD reactor. SMF coated with TiO2 was used as discharging electrode. The synergy between DBD and photo-catalytic electrode was explored. Special consideration was given to some important parameters that may affect toluene conversion, including peak voltage, initial concentration, residence time and humidity. The mineralization rate and CO2 selectivity of toluene decomposed in various reactors were studied. A long-term experiment (25 d) was operated to investigate the stability of the photo-catalytic electrode.
specified impurities of H2O. The purity of toluene (Huadong Medicine Limited CO. LTD) was 99.9%. The concentration of anatase nano-TiO2 solution (VK-TG02, Wangjing Material Co. Ltd., China) was 10%. SMF (average pore radius of 5 lm, porosity of 65% and diameter of 100 lm) was supplied by Xi’an Filter Metal Materials Co. Ltd., China. And the TiO2/SMF-n electrode was prepared by a modified anodic oxidation method [24]. UV light was applied by a 125 W high-pressure mercury lamp (Shanghai Yamingdianqi Zhaoming Dianqi Limited CO. LTD) with wave length of 365 nm.
2. Experimental
The following standard procedure was developed to achieve stable and reproducible conditions for the gas flow rate and composition. The reactor was first flushed with air at 1 L/min for a few minutes. Toluene-containing gas was then allowed in at an appropriate flow rate for 30 min, at which point the toluene concentration measured at the reactor outlet was the same as that at its entry. The desired flow rate for the specific experiment was then set and the high voltage was applied. After a stabilization time of about 5 residence times to obtain a steady concentration of the residual odor at the reactor outlet, two consecutive samples of the treated gas were allowed to GC analysis. After a group of experiments was completed, the gas flow was stopped and the high voltage source was switched off. The apparatus was finally cleaned by flushing air at 1 L/min for 15 min. In experiments with humidified gas, preconditioning of the reactor was achieved by flushing humidified air through it for 30 min. Then, the toluenecontaining gas, humidified as described above, was introduced into the system. All the experiments above were achieved at ambient temperature.
2.1. Experimental setup and reactor Fig. 1 showed a schematic diagram of the experimental setup. Dry air from a gas cylinder was allowed into a buffer tank at a controlled flow rate through tubes. A fraction of the gas flow went through a toluene generator to carry out toluene with an appropriate level. For experiments with humid air, a portion of the air flow went through a bubbler filled with deionized water to an appropriate level. Then all the streams were mixed in the buffer tank, where humidity was measured by a hygrometer (Rotronic A1H). At last, the mixed stream was introduced to the DBD reactor. Samples of the treated gas for on-line chemical analyses were allowed into a gas chromatography (GC) from the sampling ports located at the inlet and the outlet of the reactor. The DBD reactor was a 24 mm (inner diameter) 200 mm quartz cylinder with a SMF wire of 1.5 mm in diameter fixed along its axis. The SMF wire served as the support of TiO2 and also an energized electrode. The ground electrode was a 150 mm copper film which was embedded outside of quartz cylinder. An AC high voltage source with frequency of 20 kHz was applied in the experiment. The voltage and current waveforms were measured by using a four channel Tecktronix TDS 2014B 350 MHz digital storage oscilloscope capable of sampling 1 GS/s (Giga Sample per second), a Texas HVP-3020 high voltage passive probe and a CT4 TCP202 current probe. 2.2. Materials Air (Jingong Co., Ltd.) was a synthetic mixture (80% nitrogen and 20% oxygen) from pure liquid nitrogen and pure liquid oxygen with
2.3. Experimental procedure
2.4. Chemical analyses The concentration of toluene was measured by the gas chromatograph GC7890II (Tianmei corporation) which was equipped with 6-port gas sample valve. Ozone formed in the DBD reactor was measured by an UV absorption ozone monitor (AP1-450 NEMA). The formation of CO and CO2 was monitored with an infrared gas analyzer (Siemens ultramat 23). A Fourier transform infrared (FTIR) (Ncolet6700) was used for gas qualitative analysis. OH radical in the DBD reactor was detected by salicylic acid method [34]. A film, which was impregnated with salicylic acid,
Fig. 1. Schematic diagram of the experimental setup. (1) Gas cylinder; (2) mass flow control; (3) buffer tank; (4) water bath tank; (5) toluene generator; (6) water vapor generator; (7) hygrometer; (8) sampling of inlet gas; (9) sampling of outlet gas; (10) DBD reactor; (11) UV lamp; (12) high voltage source; (13) digital storage oscilloscope; (14) current probe; (15) high voltage probe; (16) gas chromatograph; (17) computer.
168
J. Chen et al. / Chemical Engineering Journal 284 (2016) 166–173
was used to detect OH radical in plasma reactor [34]. Salicylic acid reacts with OH radical and produces 2,5-dihydroxybenzoic acid (2,5-DHBA). Then, a high performance liquid chromatography (HPLC) was carried out to detect the concentration of 2,5-DHBA. Therefore, OH radical in NTP reaction could be calculated. One piece of filter paper was cut to a calculated size (300 mm 45 mm), then a solution, which was made by 0.3 g salicylic acid dissolved in 10 mL anhydrous ethanol (99.9%), was dripped equably by a injector to this filter paper and dried. The prepared sampling film was puckered and then placed in plasma reactor. The collecting time was set as 3 h. A shorter or longer time made the results below the minimum detectable limits or inaccurate. After the sampling film was taken out and was cut to fragments, all these fragments were dipped in water (15 mL). After that, an ultrasonic cleaner was used to accelerate dissolution of the products. Then the concentration of 2,5-DHBA was detected by a HPLC, equipped with an auto sampler (model G1329A) and an ultraviolet spectroscopic photometer (300 nm) [34]. The concentration of OH radical could be calculated as the following equation
C OH
C 2;5-DHBA V L N ¼ F 2;5-DHBA F g t g ð1 aÞ
ð1Þ
where, C2,5-DHBA is the concentration of 2,5-DHBA (mol/L); VL is the volume of absorbing liquid (L); N is Avogadro constant (6.02 l023); F2,5-DHBA is the producing ratio of 2,5-DHBA (in this experiment, almost all salicylic acid that reacted with OH radical produced 2,5-DHBA, therefore, F2,5-DHBA was assumed to be 1); Fg is gas flow rate (mL/min); t is sampling time (min); g is pick-up efficiency (it was assumed to be 1 in this experiment because of almost all 2,5-DHBA picked up from the film); a is the product losing ratio (it was assumed to be 0). 3. Results and discussion 3.1. Decomposition of toluene in various reactors To understand the interaction between NTP and photo-catalytic electrode, toluene was decomposed in four types of experiments. Fig. 2 showed the evolution of toluene conversion as a function of peak voltages. The main observation was that the toluene conversion increased with the increase of peak voltage in all four types
of experiments. When peak voltage increased from 10 to 40 kV, the toluene conversion was promoted from 5.2% to 89.9% in the reactor with photo-catalytic electrode (SMF + TiO2). Similar effect could be observed in other reactors. Higher peak voltages produced more high-energy electrons, active radicals, ions and other excited species, which was favorable for the toluene removal process [7]. As seen from Fig. 2, reactors with Ni–Cr alloy electrode (diameter of 1.5 mm) and SMF electrode (diameter of 1.5 mm) performed similarly in toluene decomposition with conversions of 51.3% and 53.7% at 35 kV. It suggested SMF was suitable for discharge electrode as Ni–Cr alloy did. When TiO2 was coated onto SMF (Fig. 3), toluene conversion was promoted dramatically at a fixed peak voltages. The toluene conversion reached 78.4% in the reactor with photo-catalytic electrode at 35 kV. It was because that TiO2 could be activated by high-energy electrons, excited molecules or radicals [35]. And NTP could produce UV light due to excited nitrogen molecules as follows
N2 þ e ! N2 þ e
ð2Þ
N2 ! N2 þ hm
ð3Þ
Consequently, oxidation reactions, which promoted the toluene conversion, were triggered by formation of electron–hole pairs (h+ and e) on the catalysts surface [23,36]
TiO2 þ hm ! h þ e þ
þ
H2 O þ h ! OH þ Hþ
ð5Þ
Besides, the conversion disparity between the reactor with SMF and the reactor with SMF + TiO2 became larger when the peak voltage increased. The conversion in the reactor with SMF + TiO2 was larger than that in the reactor with SMF by 13.1% at 20 kV, which increased to 28.6% at 40 kV. NTP could not produce enough highenergy electrons, radicals and UV light (active species) to activate TiO2 at low peak voltages. At high peak voltages, photo-catalytic electrode was much more active due to more active species in reactor. In order to determine if plasma UV and excited species activated sufficiently photo-catalytic TiO2 on electrode, experiments were carried out by introducing an extra UV-lamp. The lamp was placed 80 mm above the reactor with photo-catalytic electrode. This process was presented in Fig. 2 as SMF + TiO2 + UV. The presence of extra UV light source modified the toluene conversion at low peak voltages. At a peak voltage of 10 kV, the toluene conversion in the reactor with SMF + TiO2 was 5.2%. After turning on the UV lamp, the conversion reached 18.9% at the same peak voltage. The conversion disparity was 13.7%. However, the conversion disparity reduced to 2.9% at 40 kV. It was discussed above that NTP could not activate photo-catalytic electrode at low peak voltage with little active species. Thus, an extra UV light souse was favorable for photo-catalytic electrode at low peak voltage. And photocatalytic electrode was activated by active species in reactor at high peak voltages, at which the effect of the extra UV light source weakened. As seen from the results presented in the Fig. 2, it was believed that the photo-catalytic electrode could be activated in the NTP at high peak voltages in our research system. Synergistic effect (hsg) was defined as following
hsg ¼ h3 ðh1 þ h2 Þ
Fig. 2. Comparison of toluene conversion in various reactors. (Initiate concentration: 160 mg/m3; relative humidity: 0%; residence time: 4 s; TiO2/SMF-n: 5 wt%.)
ð4Þ
ð6Þ
where h3 was the toluene conversion rate of reactor with photocatalytic electrode and UV lamp (%); h1 was the toluene conversion rate of reactor with SMF electrode (%); h2 was the toluene conversion rate of reactor with photo-catalytic electrode (at a peak voltage of 0 kV) and UV lamp (%).
J. Chen et al. / Chemical Engineering Journal 284 (2016) 166–173
169
Fig. 3. SEM images of SMF + TiO2 at 2000 (a) and 5000 (b). (TiO2/SMF-n: 5 wt%.)
Synergistic effect (hsg) between DBD and photo-catalytic electrode on removal of toluene was studied as a function of peak voltage varied in the range of 10–40 kV. Fig. 4 showed that hsg between DBD and photo-catalytic electrode on removal of toluene increased with increase of peak voltage at low peak voltages. The highest hsg of 25.5% was obtained at 30 kV. When the peak voltage was higher than 30 kV, hsg dropped to 20%. It could be explained by that as discussed above. photo-catalytic electrode was activated by active species in reactor at high peak voltages. Energy deposition (ED) is defined as follows [37]
I ¼ Cm 1 P¼ T ED ¼
Z
T 0
P Q
dV m dt Cm VIdz ¼ T
ð7Þ Z
T 0
dV m V ¼ f Cm dt
Z VdV m
ð8Þ
ð9Þ
where I was the instantaneous current (A); Cm was the known capacitor (F); Um was the instantaneous voltage of the capacitor (V); P was the discharge power (W); T was the alternating current cycle (s); U was the instantaneous voltage of the DBD reactor (V); f was the frequency (Hz); ED was the energy deposition (J/L); Q was the gas flow rate (L/s);
Fig. 4. The synergistic effect between DBD and photo-catalytic electrode on removal of toluene.
Fig. 5 showed the variation of energy deposition as a function of applied voltage for two reactors. With the increasing of peak voltage, energy deposition of two kinds of reactor increased. The energy deposition of the reactor with SMF + TiO2 was larger than that of reactor with SMF. When peak voltage was 60 kV, the energy deposition of the reactors with SMF + TiO2 and SMF were 154 and 121 J/L. It was believed that the nano-TiO2 coated on the SMF was polarized. And an intense electric field was formed on the surface of the TiO2, resulting in partial discharge. Therefore, the energy deposition of the reactor with SMF + TiO2 was promoted. 3.2. Photo-catalytic electrode with different amount of TiO2 loaded on Oxidation of toluene in DBD with different amount of TiO2 loading on SMF (3–5%) was shown in Fig. 6. The toluene conversion increased with the increase of the amount of TiO2 loading on SMF at a fixed peak voltage. At 35 kV, the toluene conversions with TiO2/SMF-3, TiO2/SMF-4 and TiO2/SMF-5 were 65.8%, 72.6% and 78.4%, respectively. It implied that the amount of TiO2 loaded on photo-catalytic electrode might not be sufficient for the system. More TiO2 (more than 5 wt%) loaded on photo-catalytic electrode might further promote toluene conversion. However, as seen from Table 1, some of the TiO2 loaded on TiO2/SMF-6 was detached in the ultrasonic test. The ultrasonic test was performed to evaluate the adherence of TiO2/SMF-n. TiO2/SMF-n was placed in an ultrasound cleaner full of ethanol for a 45 min. Judging from the results of removal experiments and detachment experiments the
Fig. 5. Variation of energy deposition as function of peak voltage. (TiO2/SMF-n: 5 wt %; flow rate: 1000 mL/min.)
170
J. Chen et al. / Chemical Engineering Journal 284 (2016) 166–173
Fig. 7. Effect of initiate concentration on conversion of toluene. (Residence time: 4 s; relative humidity: 0%; TiO2/SMF-n: 5 wt%.) Fig. 6. Oxidation of toluene in DBD with TiO2/SMF-n. (Initiate concentration: 160 mg/m3; relative humidity: 0%; residence time: 4 s.)
optimum amount of TiO2 loaded on photo-catalytic electrode was 5 wt%. 3.3. Effect of initial concentration and residence time on removal of toluene Removal of toluene was investigated in the DBD reactor with photo-catalytic electrode under various initial concentration and residence time. The initial concentrations of toluene were 95, 160 and 280 mg/m3. And the residence time of toluene in reactor were 1, 2, 4 and 6 s, which corresponded to the flow rate of 4, 2, 1 and 0.7 L/min, respectively. Fig. 7 showed that the initial toluene concentration was an important factor that influenced the toluene removal process. At a fixed peak voltage, the removal rate of toluene decreased with an increase in the initial concentration, while the amount of toluene removed increased. At 35 kV, the toluene removal rate of 95, 160 and 280 mg/m3 were 92.6%, 78.4% and 51.8%, respectively. And the amount of toluene removed increased from 42.6 to 70.22 mg/h. The phenomenon could be explained by that with the same amount of active species produced in the processed stream, the competition of high-energy electrons and active species among toluene molecules became fiercer for higher toluene concentration. It caused the reduction of toluene removal rate. On the other hand, more toluene molecules were introduced into the reactor with higher toluene initial concentration. Active species in the reactor were much more prone to collide with toluene molecules leading to high energy yield. Thus, the amount of toluene removed increased. Fig. 8 showed the relationship between the toluene removal rate and the residence time. The removal rate of toluene increased sharply with an increase of residence time at a fixed peak voltage.
For a peak voltage of 30 kV, the toluene removal rate reached to 78.3% when the residence time was 6 s. And it dropped to 38.9% when the residence time was 1 s. Most of the reactive species in DBD had a short lifetime and could only exist with permanent energy supply. And toluene would decompose deficiently without sufficient time to react with high energy electrons and reactive species. 3.4. Effect of humidity on decomposition of toluene The effect of humidity on toluene decomposition was examined by passing dry and humidified air (0–80%) through the DBD reactor with/without TiO2. The room and experimental system temperature were control at 298 K. Fig. 9 showed that humidity affected toluene decomposition differently in DBD system with/without TiO2. For a peak voltage of 30 kV in DBD system without TiO2, the toluene removal rate was 27% in dry air. And the toluene removal rate decreased to 20% as the humidity increased to 80% relative humidity. This indicated that high humidity had an inhibition to toluene decomposition in DBD system without TiO2. Water molecules had electronegative characteristics [35]. High humidity
Table 1 Detachment experiment of TiO2/SMF-n.
TiO2/SMF-4 TiO2/SMF-5 TiO2/SMF-6
Mb
Ma
2.2042 2.221 2.2472
2.2039 2.2207 2.2295
Mb is the mass of TiO2/SMF-n before ultrasonic test. Ma is the mass of TiO2/SMF-n after ultrasonic test.
Fig. 8. Effect of residence time on conversion of toluene. (Initiate concentration: 160 mg/m3; relative humidity: 0%; TiO2/SMF-n: 5 wt%.)
J. Chen et al. / Chemical Engineering Journal 284 (2016) 166–173 TiO2
H2 O þ hm ! H þ OH
Fig. 9. The effect of humidity on conversion of toluene. (Initiate concentration: 160 mg/m3; residence time: 4 s; TiO2/SMF-n: 5 wt%.)
in the discharging zone limited the electron density and quenched the activated chemical species, resulting in low toluene removal rate. However, a promoting effect of humidity on toluene decomposition was observed in DBD system with TiO2. For a peak voltage of 30 kV, the conversion of toluene was 55.6% in dry air. The conversion of toluene was promoted to 69.5% as the humidity increases to 20%. With the increase of humidity (20–80%), the conversions of toluene all keep above 64%. Similar results were obtained for other peak voltages. In order to investigate the role of humidity in the DBD system with/without TiO2, OH radical density in the reactor was measured. Fig. 10 showed the OH density as a function of humidity in the DBD system. In the system without TiO2, OH density increased with an increase of humidity at lower humidity (620%). And OH production was saturated as humidity increases at a higher humidity. In the system with TiO2, the OH radical density was higher than that in the system without TiO2. H2O molecules collided with high energy electrons and form OH radicals in the DBD system
H2 O þ e ! H þ OH þ e
ð10Þ
when TiO2 was introduced onto the electrode, OH radicals also could be produced by following reaction:
Fig. 10. OH radical density as a function of relative humidity in the DBD system. (Balance gas: dry air; residence time: 4 s; TiO2/SMF-n: 5 wt%; peak voltage: 37.5 kV.)
171
ð11Þ
OH was extraordinarily active and takes part in toluene decomposition processes [19]. What’s interesting was that the toluene conversion rate did not increase when the humidity increases from 20% to 80%. In the same time there was a 50% increase of the concentration of OH. It implied that next to OH formation, also other processes played an important role towards toluene decomposition. Ozone played an important role in plasma chemical process. In order to investigate the role of ozone in our case, ozone was injected downstream of the reactor without discharge. Ozone was produced by an ICAN CFY-3 ozone generator. The initial O3 flow rate and concentration were 48 mL/min and 30 g/m3, respectively. Then the O3 flow was mixed with total stream of 1000 mL/min to get the O3 concentration of 1450 mg/m3. When the initial concentration of toluene was 160 mg/m3, the conversation of toluene reached to 28%. Thus, the experimental results indicated that the toluene could react with ozone. On the other hand, humidity also strongly affected forming ozone in plasma system. Water molecules might compete with oxygen for high energy electrons. The ozone concentration in the plasma outlet gas stream was 1542 mg/m3 in dry air at 37.5 kV. The ozone produced in the plasma decreased at higher humidity. The ozone concentration monitored was 1164 mg/m3 at 40% humidity and 980 mg/m3 at 80% humidity. Since more ozone in reactor could promote the decomposition of toluene, high humidity, resulting in inhibiting ozone production, leaded to an adverse effect on toluene decomposition. It could be seen that different factors codetermined the decomposition of toluene under humid conditions. Generally, photocatalytic electrode could improve DBD system’s water-resistance by turning more water molecules into OH radicals. 3.5. Product analysis Judging from the gas analysis using FTIR toluene mostly decomposed into CO2 and CO in DBD reactor with SMF + TiO2. Formic acid (HCOOH) and formaldehyde (HCOH) were also detected in the outlet gas at low peak voltages from the FTIR analysis. The amount of HCOOH and HCOH decreased with raising peak voltages. To evaluate the mineralization rate of the toluene removal system, the MR was defined as follows
MR ð%Þ ¼
½CO2 þ ½CO 7ð½Toluenein ½Tolueneout Þ
ð12Þ
where the MR (%) is mineralization rate; [Toluene]in (mg/m3) and [Toluene]out (mg/m3) indicate inlet and outlet concentration of Toluene, respectively; [CO2] (mg/m3) and [CO] (mg/m3) indicate outlet concentration of CO2 and CO, respectively. Fig. 11 showed the mineralization rate of three reactors as a function of peak voltage. Mineralization rate were low at 20 kV and increased with increasing peak voltage for all three reactors. In the reactor with SMF, mineralization rate was 38.6% at 20 kV and increased to 79.4% at 40 kV. The reactor with SMF + TiO2 and the reactor with SMF + TiO2 + UV revealed better mineralization rate. Mineralization rate of the reactor with SMF + TiO2 and the reactor with SMF + TiO2 + UV were 92.4% and 93.2%, respectively. The results suggested that toluene mostly converted to other byproducts except for CO and CO2 in the reactor with SMF. While good mineralization rate in the reactor with SMF + TiO2 and the reactor with SMF + TiO2 + UV also indicated that the formation of other reaction byproducts such as aerosol and smaller organic compounds was negligible. It was desirable to achieve deep oxidation to CO2 rather than CO in the decomposition of VOCs. CO2 selectivity was defined as follows
172
J. Chen et al. / Chemical Engineering Journal 284 (2016) 166–173
Fig. 11. Comparison of mineralization rate. (Balance gas: dry air; initiate concentration: 160 mg/m3; residence time: 4 s; TiO2/SMF-n: 5 wt%; experiment temperature: 298 K.)
SCO2
½CO2 ð%Þ ¼ 100% 7ð½Toluenein ½Tolueneout Þ
Fig. 13. Comparison of the evolution of the toluene conversion in different reactors. (Initial concentration of toluene: 160 mg/m3; residence time: 4 s; peak voltage: 40 kV; humidity: 0%.)
3.6. Long-term operation
ð13Þ
where the SCO2 (%) was CO2 selectivity. As seen from Fig. 12, CO2 selectivity was enhanced with the increase of peak voltage for all three reactors. The CO2 selectivity of reactor with SMF ranged from 18.1% to 50.2%, which was much lower than that of the other two reactors. It indicated that SMF + TiO2 electrode was effective in increasing the CO2 selectivity as well as enhancing the toluene decomposition. The reactor with SMF + TiO2 and the reactor with SMF + TiO2 + UV nearly got the same high CO2 selectivity at 40 kV. They were all above 90%. It implied that NTP could activate TiO2 at high peak voltages. While the CO2 selectivity of reactor with SMF + TiO2 + UV was higher than that of reactor without UV source by 18%. It was believed that NTP could not produce enough active species to activate TiO2 at low peak voltages, resulting in TiO2 not working on SMF.
Fig. 13 showed the results of long-term operation experiments of two reactors. Every reactor ran 8 h/d with a fixed peak voltage of 40 kV in the experiment. Gas samples were measured before and after reactors every hour. As seen from the Fig. 12, the reactor with SMF + TiO2 exhibited higher toluene conversion rate and performance stabilization. The reactor with SMF + TiO2 got a toluene conversion rate of 89.9% at the first day. And the conversion remained 89.8% at the 25th day. While the conversion of the reactor with SMF dropped from 61.3% to 54.2% during the long-term operation (25 d). After long-term operation, the SMF discharge electrode in the reactor was covered with brown oily materials which were polymerized by intermediates. The polymer attached on the electrode weakened discharge, resulting in toluene conversion reduction. And yet the SMF + TiO2 discharge electrode was not covered with any polymer. It was because that irradiated SMF–TiO2 had high energy content than the band-gap. And it produced electron–hole pairs which led to redox reactions between separated electron–hole pairs and trapped VOCs. Thus, VOC molecules and intermediates colliding with SMF–TiO2 electrode would decompose completely, avoiding damaging to discharge electrode. 4. Conclusion A DBD reactor with photo-catalytic electrode was applied for removal of toluene. Toluene conversion was enhanced dramatically due to the photo-catalytic electrode. At a peak voltage of 40 kV, the toluene conversion of 89.9% was obtained in the novel reactor. Peak voltage, initial concentration and residence time were important factors that influenced the toluene conversion. Photocatalytic electrode improved DBD reactor’s water-resistance. And the reactor with photo-catalytic electrode had better performance in mineralization rate and CO2 selectivity. Acknowledgments
Fig. 12. The CO2 selectivity of three different reactors. (Balance gas: dry air; initiate concentration: 160 mg/m3; residence time: 4 s; TiO2/SMF-n: 5 wt%; experiment temperature: 298 K.)
We gratefully acknowledge the financial support of the Natural Science Foundation of Zhejiang province (Project No. LQ13B070 006), the National Natural Science Foundation of China (Project No. 21406044, No. 21307023, No. 41373121), the Innovative Team Foundation of Hangzhou Dianzi University (Project No. ZX130203318006).
J. Chen et al. / Chemical Engineering Journal 284 (2016) 166–173
References [1] H.Z. Streicher, P.A. Gabow, A.H. Moss, D. Kono, W.D. Kaehny, Syndromes of toluene sniffing in adults, Ann. Intern. Med. 94 (1981) 758–762. [2] M. Lu, R. Huang, J. Wu, M. Fu, L. Chen, D. Ye, On the performance and mechanisms of toluene removal by FeOx/SBA-15-assisted non-thermal plasma at atmospheric pressure and room temperature, Catal. Today 242 (2015) 274– 286. [3] T. Masato, H. Manabu, A. Masakazu, Efficient removal of toluene and benzene in gas phase by the TiO2/Y-zeolite hybrid photocatalyst, J. Hazard. Mater. 237 (2012) 133–139. [4] M.J. Munoz-Batista, M. Fernandez-Garcia, A. Kubacka, Promotion of CeO2–TiO2 photoactivity by g-C3N4: ultraviolet and visible light elimination of toluene, Appl. Catal. B 164 (2015) 61–270. [5] J. Chen, J. Yang, H. Pan, Q. Su, Y. Liu, Y. Shi, Abatement of malodorants from pesticide factory in dielectric barrier discharges, J. Hazard. Mater. 177 (2010) 908–913. [6] W. Liang, J. Li, J. Li, Y. Jin, Abatement of toluene from gas streams via ferroelectric packed bed dielectric barrier discharge plasma, J. Hazard. Mater. 170 (2009) 633–638. [7] T. Yamamoto, K. Ramanathan, P.A. Lawless, D.S. Ensor, J.R. Newsome, N. Plaksn, G.H. Ramsey, Control of volatile organic compounds by an AC energized ferroelectric pellet reactor and a pulsed corona reactor, IEEE Trans. Ind. Appl. 28 (1992) 528–534. [8] J. Chen, Q. Su, H. Pan, J. Wei, X. Zhang, Y. Shi, Influence of balance gas mixture on decomposition of dimethyl sulfide in a wire-cylinder pulse corona reactor, Chemosphere 75 (2009) 261–265. [9] Y.F. Guo, D.Q. Ye, Y.F. Tian, K.F. Chen, Humidity effect on toluene decomposition in a wire-plate dielectric barrier discharge reactor, Plasma Chem. Plasma Process. 26 (2006) 237–249. [10] S.B. Han, T. Oda, Decomposition mechanism of trichloroethylene based on byproduct distribution in the hybrid barrier discharge plasma process, Plasma Sources Sci. Technol. 16 (2007) 413–421. [11] K. Hensel, S. Katsura, A. Mizuno, DC microdischarges inside porous ceramics, IEEE Trans. Plasma Sci. 33 (2005) 574–575. [12] C. Hibert, I. Gaurand, O. Motret, J.M. Pouvesle, [OH(X)] measurements by resonant absorption spectroscopy in a pulsed dielectric barrier discharge, J. Appl. Phys. 85 (1999) 7070–7075. [13] J. Jarrige, P. Vervisch, Decomposition of gaseous sulfide compounds in air by pulsed corona discharge, Plasma Chem. Plasma Process. 27 (2007) 241–255. [14] H.H. Kim, Nonthermal plasma processing for air-pollution control: a historical review, current issues and future prospects, Plasma Process. Polym. 1 (2004) 91–110. [15] H.M. Lee, M.B. Chang, Abatement of gas-phase p-xylene via dielectric barrier discharges, Plasma Chem. Plasma Process. 23 (2003) 541–558. [16] E. Marotta, A. Callea, M. Rea, C. Paradisi, DC corona electric discharges for air pollution control. Part 1. Efficiency and products of hydrocarbon processing, Environ. Sci. Technol. 41 (2007) 5862–5868. [17] T. Oda, T. Takahashi, K. Tada, Decomposition of dilute trichloroethylene by nonthermal plasma, IEEE Trans. Ind. Appl. 35 (1999) 373–379. [18] R. Ono, T. Oda, Measurement of hydroxyl radicals in pulsed corona discharge, J. Electrostat. 55 (2002) 333–342. [19] K. Urashima, J.S. Chang, Removal of volatile organic compounds from air streams and industrial flue gases by non-thermal plasma technology, Trans. Dielectr. Electr. Insul. 7 (2000) 602–614.
173
[20] G.P. Van-Durme, B.F. McNamara, C.G. McGinley, Bench-scale removal of odor and volatile organic compounds at a composting facility, Water Environ. Technol. 64 (1992) 19–27. [21] A.E. Wallis, J.C. Whitehead, K. Zhang, The effect of temperature on the removal of DCM using non-thermal, atmospheric-pressure plasma-assisted catalysis, Appl. Catal. B 74 (2007) 111–116. [22] H.H. Kim, H. Kobara, A. Ogata, S. Futamura, Comparative assessment of different nonthermal plasma reactors on energy efficiency and aerosol formation from the decomposition of gas-phase benzene, IEEE Trans. Ind. Appl. 41 (2005) 206–214. [23] J. Van-Durme, J. Dewulf, C. Leys, H. Van-Langenhove, Conbining non-thermal plasma with heterogeneous catalysis in waste gas treatment: a review, Appl. Catal. B 78 (2008) 324–333. [24] Z. Ye, C. Wang, Z. Shao, Q. Ye, Y. He, Y. Shi, A novel dielectric barrier discharge reactor with photocatalytic electrode based on sintered metal fibers for abatement of xylene, J. Hazard. Mater. 241 (2012) 216–223. [25] L. Huang, L. Xia, X. Ge, H. Jing, W. Dong, H. Hou, Removal of H2S from gas stream using combined plasma photolysis technique at atmospheric pressure, Chemosphere 88 (2012) 229–234. [26] F. Thevenet, O. Guaitella, E. Puzenat, C. Guillard, A. Rousseau, Influence of water vapour on plasma/photocatalytic oxidation efficiency of acetylene, Appl. Catal. B 84 (2008) 813–820. [27] O. Guaitella, F. Thevenet, E. Puzenat, C. Guillard, A. Rousseau, C2H2 oxidation by plasma/TiO2 combination: influence of the porosity and photocatalytic mechanisms under plasma exposure, Appl. Catal. B 80 (2008) 296–305. [28] M. Ksibi, S. Rossignol, J.M. Tatibouët, C. Trapalis, Synthesis and solid characterization of nitrogen and sulfur-doped TiO2 photocatalysts active under near visible light, Mater. Lett. 62 (2008) 4204–4206. [29] J. Chen, Z.M. Xie, Removal of H2S in a novel dielectric barrier discharge reactor with photocatalytic electrode and activated carbon fiber, J. Hazard. Mater. 261 (2013) 38–43. [30] J. Karuppiah, E. Lingareddy, L. Sivachandiran, R. Karvembu, Ch. Subrahmanyam, Nonthermal plasma assisted photocatalytic oxidation of dilute benzene, J. Chem. Sci. 124 (2012) 841–845. [31] F. Thevenet, O. Guaitella, C. Guillard, E. Puzenat, G. Stancu, J. Roepcke, A. Rousseau, Comparison of the plasma-photocatalyst synergy at low and atmospheric pressure, Int. J. Plasma Environ. Sci. Technol. 1 (2007) 52–56. [32] Ch. Subrahmanyam, M. Magureanu, D. Laub, A. Renken, L. Kiwi-Minsker, Nonthermal plasma abatement of trichloroethylene enhanced by photocatalysis, J. Phys. Chem. C 111 (2007) 4315–4318. [33] Ch. Subrahmanyam, Catalytic non-thermal plasma reactor for total oxidation of volatile organic compounds, Indian J. Chem. 48 (2009) 1062–1068. [34] Y.F. Guo, X.B. Liao, D.Q. Ye, Detection of hydroxyl radical in plasma reaction on toluene removal, J. Environ. Sci. 20 (2008) 1429–1432. [35] H.H. Kim, Y.H. Lee, A. Ogata, S. Futamura, Plasma-driven catalyst processing packed with photocatalyst for gas-phase benzene decomposition, Catal. Commun. 4 (2003) 347–351. [36] A.L. Linsebigler, G.Q. Lu, J.T. Yates, Photocatalysis on TiO2 surfaces: principles, mechanisms, and selected results, Chem. Rev. 95 (1995) 735–758. [37] D. Evans, L.A. Rosocha, G.K. Anderson, J.J. Coogan, M.J. Kushner, Plasma remediation of trichloroethylene in silent discharge plasmas, J. Appl. Phys. 74 (1993) 5378–5386.