Manuscript

CCl4 Decomposition by Gliding Arc Plasma: Role of C2 compounds on products distribution Shortened title: CCl4 Decomposition by Gliding Arc (manuscript id: 130-06)

Antonius Indarto1,2,†, Dae Ryook Yang3, Jae-Wook Choi2, Hwaung Lee2, and Hyung Keun Song2 1

School of Environmental, Resources, and Development, Asian Institute of Technology, P.O. Box 4, Klong Luang, Pathumthani 12120, Thailand 2

Clean Technology Research Center, Korea Institute of Science & Technology P.O. Box 131, Cheongryang, Seoul 130-650, Korea 3

Department of Chemical and Biological Engineering, Korea University, Seoul 136-701, Korea

†

Correspondence address: e-mail: [email protected], tel:+82-10-2296-3748

1

Abstract- The goal of this work is to investigate the role of existing C2 compounds in the plasma reactions of carbon tetrachloride (CCl4) decomposition. The experiment of CCl4 decomposition was carried out by gliding arc plasma. The decomposition products were dominated by CO, CO2, and Cl2. The conversion of CCl4 into Cl2 and (CO+CO2) reaches ~50% and ~40%, respectively. Other chlorinated compounds were suspected to be produced, such as COCl, COCl2, and C2 compounds. In order to prove the existence of those compounds, for example chlorinated C2 compounds, a kinetic simulation was performed and crossed check with the experimental results to clarify the reactions mechanism.

Keywords: Plasma; Gliding Arc; CCl4 decomposition; Reaction mechanism.

2

INTRODUCTION

Emissions of chlorinated volatile organic compounds (CVOCs) from various industries, such as carbon tetrachloride (CCl4), create environmental problems (Butler, 2000; WMO-UNEP, 2002). Chemical degradation of CVOCs in our atmosphere will produce another toxic chloride compound, e.g. HCl, which is classified as the main component of acid rain (Sanhueza, 2001). Some studies report that CCl4 will produce very-active chlorine radicals by solar radiation reactions. These species will react and destruct ozone molecules in the stratosphere (Butler, 2000; US-EPA, 2002). Moreover, the most severe problem of the emission of CCl4 is due to toxic and carcinogenic to the human health (IARC, 1987). Those can be the reasons to find effective methods in reducing the emission of CVOCs. The most widely adopted technique for the treatment of CVOCs is thermal combustion or incineration (Cheremisinoff, 2002). The decomposition process is done by direct thermal reaction with oxidant, e.g. air and oxygen. Although this method is simple, it requires high burning temperature (between 800 and 1,100℃) to achieve the optimum decomposition rate. Taylor and Delinger’s paper reported that incomplete combustion condition could produce a large amount of complex chlorinated products (Taylor and Dellinger, 1988). To overcome these problems, many studies were carried out on the application of new technologies. Plasma-assisted technology is one of the promising technologies to decompose CVOCs. Low thermal plasma processes, such as RF plasma (Lee et al., 1996), surface discharge reactor (Oda et al., 2002), dielectric barrier discharge reactor

3

(Tonkyn et al., 1996), pulsed discharge reactor (Yamamoto et al., 1992), and capillarytube type discharge reactor (Kohno et al., 1998) have been investigated and developed. In this study, gliding arc plasma was used to decompose carbon tetrachloride. Previously, we successfully conducted experiments with gliding arc plasma to treat CH2Cl2 (Indarto et al., 2006a; 2006b), CHCl3 (Indarto et al., 2005b; 2006b), CO2 (Indarto et al., 2006c), and methane (Indarto et al., 2005b; 2006d). Compared with the above plasmas methods, there is a greater chance that gliding arc plasma can be applied in the chemical industry (Fridman et al., 1999; Fanmoe et al., 2003). This method allows the higher flame temperature and energy that are necessary to destroy the chemical bond of toxic compounds than other plasmas, such as corona, DBD, etc. In addition, gliding arc is applicable for high input flow rate conditions. Kinetic reaction models of CCl4 decomposition using electron beam in dry air (Penetrante et al., 1995; Koch et al., 1995; Nichipor et al., 2000) and liquid (Mak et al., 1997) have been investigated. The models suggested that the main products were dominated by chlorinated C1 compounds, such as: COCl2, CO, CO2, ClNO3, ClO3, and Cl2. Fragmentation of CCl4 was occurred by two reaction mechanisms (Penetrante et al., 1995). In the oxygen-rich condition, the atomic oxygen species, such as ground state of atomic oxygen O(3P) and excited atomic oxygen O(1D) have enough energy to destruct the bonds of CCl4. O( 3P ) + CCl 4 = ClO + CCl 3

(1)

O(1D) + CCl 4 = CCl 3 + ClO2

(2)

The second mechanism was occurred via reaction with secondary electron. Secondary

4

electrons will dissociate CCl4 into CCl3 and negative chlorine ion (Cl‾), e + CCl 4 = CCl 3 + Cl −

(3)

Radical ClO and CCl3 were suspected to be the most important intermediate species to determine the final products of CCl4 decomposition. The concentration of O and Cl could be also influenced by the radical reactions, both initiation and termination reactions. Nichipor’s (2000) investigation showed that CO, Cl2 and CO2 were formed by dissociation process between electron and COCl2: e + COCl 2 = CO + Cl 2

k = 10-7 cm3/mol

(4)

then, CO was oxidized by ClO and produce CO2: CO + ClO = CO2 + Cl

(5)

However, less paper discussed about the importance of chlorinated C2 in the decomposition reaction of CCl4. In this study, we investigate the role of chlorinated C2 compounds on the reactions mechanism and products selectivity, especially CO, CO2, and Cl2. Two different schemes of reaction mechanism are proposed and compared with experimental results. Some reaction mechanisms, obtained from microwave plasma and combustion process, were used to approach the real system of gliding arc plasma. The simulation results were compared to the experimental ones to satisfy the kinetic models.

EXPERIMENTAL SETUP

5

Figure 1

The schematic diagram of experimental setup is shown in Figure 1. Carbon tetrachloride and dry atmospheric air were used as the source gases. Each part of the system is described in detail in the following section. The reactor was made from a quartz tube with inner diameter of 45 mm and length of 300 mm. The upper and bottom part of the reactor were supplied with polytetrafluoroethylene (PTFE)

seals, the lower one holding two electrodes. The

electrode, made of stainless steel, has a triangular form with the height of 100 mm. The shortest gap between two electrodes was only 1.5 mm. The gas was injected between these electrodes through a capillary of inner diameter of 0.8 mm. A thermocouple, located 10 cm above the electrode, was installed to measure the outlet gas temperature. A high frequency alternating-current (AC) power supply (Auto electric A1831, Korea) was connected to the electrodes of gliding arc to generate plasmas. Figure 2 shows the typical waveform of voltage and current obtained in this experiment.

Figure 2

Liquid carbon tetrachloride (CCl4) was purchased from Kanto Chemical Co., Inc. The purity was more than 99.5%. The concentrations of CCl4 in the input gas were varied by 1, 3, 5, 8 vol% of total flow rate. Dry atmospheric pressure air was used as the carrier gas and controlled by a Mass Flow Controller (Tylan FC-280S, U.S.). The flow rates of air were 180, 240, and 300 L/hr. CCl4 was introduced to the reactor by passing a portion

6

of dry air through a bubbling tube of liquid CCl4, which was immersed in a water bath. Concentration of CCl4 was adjusted by controlling both temperature of water bath and injected gas flow rate to the bubbling tube. The feed line stream was covered by a heating band to avoid the condensation of CCl4. Mass spectroscopy and two Gas Chromatographers (GCs) were used for the qualitative and quantitative analysis of the reactants and products. The concentrations of CCl4 in the gas mixture before and after the reaction was determined by GC flame ionized detector (YoungLin M600D, Column: Bentone, Korea) and the concentrations of CO and CO2 were measured by GC thermal conductivity detector (YoungLin M600D, Column: SK Carbon, Korea). Species analysis of the decomposition products was done by a quadruple mass spectroscopy (Balzers QMS 200, U.S.) with Quadstar 421 software. Chlorine gas (Cl2) was determined by bubbling the output gas through 0.05 M of aqueous KI during measured-experiment time and followed by iodometric titration 0.05 M of Na2SO3 (Skoog et al., 2000). Selectivity of products and conversion of CCl4 were defined as: Selectivity of Cl 2 =

moles of Cl 2 formed × 100% 2 × moles of CCl 4 converted

(6)

Selectivity of CO =

moles of CO formed × 100% moles of CCl 4 converted

(7)

Selectivity of CO2 =

moles of CO2 formed × 100% moles of CCl 4 converted

Conversion of CCl 4 =

moles of CCl 4 consumed × 100% moles of CCl 4 introduced

7

(8)

(9)

Cl 2 / (CO + CO2 ) ratio =

moles of Cl 2 formed moles of CO formed + moles of CO2 formed

(10)

The consumed power for the process was calculated as the products of voltage and current waveform captured by oscilloscope (Agilent 54641A, U.S.). In this experiment, we limit the power from 286 to 304 Watt. Sample data were obtained after 30 minutes operation referring to the stable temperature condition of output line measured by thermocouple.

KINETIC STUDY AND MODEL Figure 3

Table 1

To investigate the reactions mechanism, a chemical kinetic reaction model of CCl4 decomposition was constructed. The gas-phase reaction model consists of the 21 elementary reactions that are proposed by literatures as the most significant pathway of reactions (Lou and Chang, 1997; Chang and Senkan, 1989; NASA, 1994; Penetrante et al., 1995, Koch et al., 1995, Nichipor et al., 2000). The rate constant of each reactions is expressed in a modified Arrhenius form, k j = A jT

βj

 Ej   exp −  RT 

(11)

8

The simulation was focused on the reactions of Cl2, CO, CO2 and chlorinated C1 production. Other products, such as chlorinated C2, were considered only as intermediate species because the concentrations of those components in the product line were small. Table 1 shows list of elementary chemical reactions of CCl4 decomposition. The model was coded in MATLAB program and numerically solved using shooting method (Walas, 1991; Chapra and Canale, 1990). The algorithm of the program is shown in Figure 3a. MATLAB modules of ode23s and fmins were utilized to solve simultaneously the 15 sets (15 species) of differential equation (shown in formula 12 and 13) and minimize the absolute error (MathWorks, 1992). m dxi = ∑ f j ( k j , xi (t ),..., x n (t ) ) dt j =1

(12)

xi (t 0 ) = xiinput ; xi (t ) = xiproduct

(13)

The concentration of Cl2 in the products was used as the basis of calculation (boundary condition) to calculate the Cl2 to (CO+CO2) ratio and converted CCl4. To verify the validity of the model, the simulation and experimental results were compared for each point. The absolute error was calculated as:

{ (

)

(

)

(

product sim product ∆ ε = abs xClsim2 − xClproduct + abs x (sim CO + CO2 ) − x ( CO + CO2 ) + abs x CCl 4 − xCCl 4 2

)}

2

(14)

The model stiffness was checked by varying input guess. Figure 3b shows that the model produces an almost similar absolute error although the input guess is varied from 3.9 to 6.5.

9

RESULTS AND DISCUSSION

CCl4 conversion

Figure 4

Figure 4a shows the effect of initial concentration of CCl4 on the conversion at frequency of 20 kHz. Conversion decreases gradually when the initial concentration of CCl4 is increased. The maximum conversion reaches 80% at the concentration of 1 vol% and flow rate of 180 L/hr. Increasing concentration of CCl 4 will decrease the ratio of high-energy energetic species (e.g. electron and radical atoms/molecules) to CCl4. It is because at the same flow rate and supplied energy, the amount of high-energy energetic species, both quality and quantity, can be assumed to be same. This condition will decrease the relative probability of single CCl4 molecule to collide with energetic species which able to destruct the C-Cl molecule bonds. However, although the collision probabilities are lower, the effectiveness of absolute collision at higher initial concentration of CCl4 is higher that that at lower concentration of CCl4. For example, at 1% and 3 % of CCl4 concentration, it can be calculated as 1 (% of input concentration) x 80 (% of conversion) and 3 x 73, respectively. This will lead to the ratio of 80:219 which shows that the process at 3% of CCl4 concentration produces 219/80 times more effective than that at 1% of CCl4 concentration. Analysis of QMS spectrum (Figure 8) shows that the main gaseous products are CO, CO2, and Cl2. Instead of those products, other chlorinated compounds are also detected

10

but the concentrations are small.

CO+CO2 selectivity

Increasing initial concentration of CCl4 will affect on lowering selectivity of CO+CO2. As shown in Figure 4b, the selectivity of CO+CO 2 decreases around 2 % when the concentration of CCl4 is increased from 1% to 8%. Our kinetic simulation calculates that higher initial concentration of CCl4 produces higher concentration of single chlorine species (Cl) in the reactor. The reactions will follow (Chang 1989): CCl 4 = CCl 3 + Cl

(15)

CCl 3 + O = COCl 2 + Cl

(16)

Cl atom will compete with carbon atom(C) to react with oxygen (O). Higher concentration of Cl makes the probability of C and O reaction to produce CO becomes smaller. Kinetic reaction studies of CO, CO2, and Cl2 formation in thermal oxidation show the similar reaction mechanisms as proposed by literatures (Chang and Senkan, 1989; Lou and Chang, 1997). The major pathways responsible for the formation of CO2 are: COCl + Cl = CO + Cl 2

(17)

COCl + M = CO + Cl + M

(18)

11

CO2 is mainly generated from the reaction of CO and ClO: CO + ClO = CO2 + Cl

(19)

and ClO is formed via: CCl 4 + O = CCl 3 + ClO

(20)

CCl 3 + O2 = COCl 2 + ClO

(21)

Koch et al. (1995) reported a similar radical mechanism, as mentioned above, using an electron beam process for CCl4 decomposition in the presence of O2. However, formula 19 is not counted as the main mechanism for Cl production because the concentration of CO2, measured in the output stream, was relatively low. The concentration of CO2 was 5-10 times lower than CO in all of experimental ranges. An interesting phenomenon was found in case of nitrogen. The concentration of nitrogen in the air was around 80%. Analysis of QMS spectrum shows that no significant different of the concentration before and after plasma reaction. It means that although having lower excitation level, N2 was acted as an inert gas in the gliding arc plasma. Products analysis by QMS also shows that there were no significant Ncontained compounds detected, except small concentration of N2O. In gliding arc plasma, it tent that oxygen gave an influence more than nitrogen. Our previous researches on oxygen showed that atomic oxygen, both ground-state and metastable form, could be more active than ozone (Indarto et al., 2005a; 2006c). The production of ozone was relatively small because in gliding arc plasma, the fragmentation reactions

12

were preferred than recombination reactions. The importance of ground-state atomic oxygen O(3P) and metastable atomic oxygen O(1D) in the reaction kinetic has been rigorously investigated (Davidson et al., 1978). Both O(3P) and O(1D) can be produced from dissociation reactions (Penetrante et al., 1995): e + O2 = e + O ( 3 P ) + O ( 3 P )

(22)

e +O 2 = e + O( 3P ) + O(1D)

(23)

In this experiment, the initial concentration of oxygen was around ~20% as we used atmospheric air as the dilution gas for CCl4. QMS spectrum analysis shows that in the output stream, the peak intensity of O2 in product stream was slightly lower than that in the feed stream. Moreover, the magnitude of decreasing peak intensity of oxygen was not change significantly when the initial concentration of CCl4 was varied. Based on this result, it can be assumed that the concentrations of oxygen atoms are similar for each process at the same power and flow rate. On the other hand, increasing concentration of CCl4 in the feed stream will increase the concentration of intermediate Cl atoms. It will effect on smaller probability of C and O reaction as the probability of Cl and O reaction is getting higher. This analogy can be the reason why the selectivity of (CO+CO2) decreases when the initial concentration of CCl4 is increased.

Cl2 selectivity

13

Different from previous phenomena of CO and CO2 production, the selectivity of Cl2 increases when the initial concentration of CCl4 is also increased (Figure 4c). The existence of Cl atom will affect on the termination reactions of intermediates chlorine molecules that produce Cl2. The major pathways of Cl2 formation can be: COCl 2 + Cl = COCl + Cl 2

(24)

COCl + Cl = CO + Cl 2

(25)

The existence of COCl2 and COCl were detected by our QMS (Figure 8) and it gives higher possibility of above reaction to be occurred. In the higher initial concentrations of CCl4, the concentration of Cl will increase and produce more Cl2 follows formula 24 and 25.

Simulation of Cl2, CO, and CO2

Figure 6

Figure 7

In above reaction mechanisms, it shows that only chlorinated C1 compounds is counted as the intermediate species in the production CO, CO2, and Cl2. In order to investigate the role of chlorinated C2 compounds in the decomposition reaction of CCl4, especially

14

for CO, CO2, and Cl2 production, we constructed two types of reactions mechanism based on the elementary reactions listed in table 1. The first scenario considers the formula (17) – (19) and (24) – (25) as the main reaction pathways to produce CO, CO2, and Cl2 and neglects the production of those compounds from the chlorinated C2 compounds, such as C2H4, C2H5, and C2H6 (Figure 4a). This reaction pathway is chosen to examine the dependence of Cl2, CO, and CO2 production from the proposed formulas, (17) - (19) and (24) – (25). The second scenario will consider the existence of higher carbon-chloride and calculate their effects on the decomposition reactions (Figure 4b). Figure 6 shows the party comparison plot of those two models. The horizontal axis shows the experimental result while the vertical axis shows the simulation calculation. The second model shows that the results of simulation are closer to the experiment results compared to the first model. The first scenario of the model gives 8-50% of error value and it is much higher than the second model which only produces error values of 2-10%. Based on these two models, neglecting the existence of chlorinated C2 compounds in the reactions mechanism of CO, CO2, and Cl2 production will produce lower selectivity of chlorine gas (Cl2) and higher selectivity of CO and CO2 than the experimental results. Direct conversion of chlorinated C2 compounds to Cl2 could be counted as a significant pathway of Cl2 production. Lou and Chang (1997) also identified the production of Cl 2 from chlorinated C2 compounds, e.g. C2Cl5. C 2 Cl 5 + Cl = CCl 3 + CCl 3

(26)

On the other hand, the existence of chlorinated C2 compounds will reduce the selectivity

15

of CO and CO2. Instead of oxidation reactions, couple reactions of CCl3 can produce chlorinated C2 compounds CCl 3 + CCl 3 = C 2 Cl 6

(27)

CCl 3 + CCl 3 = C 2 Cl 4 + Cl 2

(28)

Although the results of first model are not match exactly with experimental results, still, it can conclude that formula (17) – (19) have an important role on the production of CO, and CO2 and formula (24) – (25) on Cl2 production. Calculation of the first model shows that formula (17) – (19) and (24) - (25) can enhance more than 50% of the production of CO, CO2, and Cl2. The existence of COCl2 which is obtained using our QMS, shown in figure 8b, can be a good evidence for this case. COCl 2 is one of the most important intermediate species in Cl2 production (formula 24). Figure 7 shows the main routes of Cl2, CO, and CO2 formation from CCl4 decomposition by gliding arc plasma.

QMS spectra

Figure 8

QMS spectra of CCl4 decomposition diluted in atmospheric arc by gliding arc plasma are shown in Figure 8. Figure 8a shows the compounds spectrum of 1 vol% of CCl 4 diluted in 180 L/hr of atmospheric air. The main spectrums of CCl4 are m/z 83/85 ( CCl 2+ ) and m/z 117 ( CCl 3+ ). CO2 has its molecular ion at m/z 44 ( CO2+ ). CO molecular

16

+

+

ion spectrum is colliding with minor N2 at m/z 29 ( CO / N 2 ). The main N2 is at m/z 28 +

( N 2 ). However, due to small concentration of CO in the atmospheric air, CO compound can be neglected in this spectrum line.

The spectrum of the products of CCl4 decomposition is shown in Figure 8b. The peak intensity of m/z 83/85 is decreased as well as intensity of m/z 117 and a new spectrum + of Cl2 is appeared at m/z 71 ( Cl 2 ). Intensity of CO2 increases but the difference

between before and after plasma condition is slightly small. High magnification of spectrum intensity is occurred at m/z 29 of N2 and CO. Because N2 is classified as a stable compound and no available source of nitrogen (N) in this system to produce N2, it can be sure that the increment intensity of m/z 29 is caused by production of CO. At high initial concentration of CCl4, spectrogram of products produces a new spectrum at +

m/z 63 ( CClO ). Based on QMS library, it can be suspected to be COCl or COCl 2. Good evidence that COCl and COCl2 which have a significant role especially to the production of CO, CO2, and Cl2 existed in the reaction mechanisms. Another intermediate species that possibly exists is CCl2 at m/z 84 (CCl+).

CONCLUSION

The decomposition of CCl4 in gliding arc plasma as well as the kinetic study was investigated. The maximum conversion of CCl4 was 80% at 1 vol% of CCl4 and total

17

flow rate of 180 L/hr. CO, CO2, and Cl2 were identified as the main products of the decomposition process. CCl3, COCl2, and COCl were the important intermediate species on CO, CO2 and Cl2 production. The existence of chlorinated C2 compounds has to be counted also as the source of Cl2 production in the final products. However, formation of chlorinated C2 compounds can reduce the selectivity of CO and CO 2 due to coupling reactions of intermediate species, e.g. CCl3.

ACKNOWLEDGEMENTS

This study was supported by the National Research Laboratory Program of the Korea Minister of Science and Technology.

18

REFERENCES

Alberici, R. M. and Jardim, W. F. (1997). “Photocatalytic destruction of VOCs in the gas-phase using titanium dioxide.” Appl. Catal. B: Environ., 14, 55-68. Butler, J.H. (2000). “Better budgets for methyl halides?” Nature, 403, 260-261. Chang, W-D. and Senkan, S. (1989). “Detailed chemical kinetic modeling of fuel-rich trichloroethane/oxygen/argon flames.” Environ. Sci. Technol., 23, 442-450. Chapra, S. C. and Canale, R. P. (1990), Numerical methods for engineers, 2nd Ed., McGraw-Hill, Singapore. Cheremisinoff, N. P. (2000), Handbook of Hazardous Chemical Properties, ButterworthHeinemann, Boston. Davidson, T. J., Schiff, H. I., Brown, T. J., and Howard, C. J. (1978). “Temperature dependence of the rate constants for reactions of O(1D) atoms with a number of halocarbons.” J. Chem. Phys., 69, 4277-4279. Fanmoe, J., Kamgang, J. O., Moussa, D., and Brisset, J. L. (2003). “Application de l'arc glissant d'air humide au traitement des solvants industriels: cas du 1,1,1-trichloroéthane.” Phys. Chem. News, 14, 1-4. Fridman, A., Nester, S., Kennedy, L. A., Saveliev, A., and Mustaf-Yardimci, O. (1999). “Gliding arc gas discharge.” Prog. Energy Combust. Sci., 25, 211-231. Indarto, A., Choi, J. W., Lee, H., Song, H. K., (2005a). “Gliding arc processing for decomposition of chloroform.” Toxicol. Environ. Chem., 87(1-4), 509-519. Indarto, A., Choi, J. W., Lee, H., Song, H. K. (2005b). “Kinetic modeling of plasma methane conversion using gliding arc plasma.” J. Natur. Gas Chem., 14, 13-21. Indarto, A., Choi, J. W., Lee, H., and Song, H. K. (2006a). “Treatment of dichloromethane using gliding arc plasma.” Intl. J. Green. Energy, 3(3), 309-321. Indarto, A., Choi, J. W., Lee, H., and Song, H.K. (2006b). “Discharge characteristics of a gliding-arc plasma in chlorinated methanes diluted in atmospheric air.” Plasma Devices Operations, 14(1), 15-26. Indarto, A., Choi, J. W., Lee, H., and Song, H.K. (2006c). “Conversion of CO2 by gliding arc plasma.” Environ. Eng. Sci., 23(6), 1033-1043. Indarto, A., Choi, J. W., Lee, H., Song, H. K. (2006d). “Effect of additive gases on methane conversion using gliding arc discharge.” Energy, 31, 2650-2659. International Agency for Research on Cancers (IARC). (1987). Monographs on the evaluation of carcinogenic risk to humans, Supplements 7.

19

Koch, M., Cohn, D. R., Patrick, R. M., Schuetze, M. P., Bromberg, L., Reilly, D., Hadidi, K., Thomas, P., and Falkos, P. (1995). “Electron beam atmospheric pressure cold plasma decomposition of carbon tetrachloride and trichloroethylene.” Environ. Sci. Technol., 29, 2946-2952. Kohno H., Berezin, A. A., Chang, J. S., Tamura, M., Yamamoto, T., Shibuya, A., and Honda, S. (1998). “Destruction of volatile organic compounds used in a semiconductor industry by a capillary tube discharge reactor.” IEEE Trans. Ind. Applicat., 34(5), 953-966. Krawczyk, K. and Ulejczyk, B. (2003). “Decomposition of chloromethanes in gliding discharges.” Plasma Chem. Plasma Process., 23(2), 265-281. NASA Panel for Data Evaluation. (December 15, 1994). Chemical kinetics and photochemical data for use in stratospheric modeling, Evaluation Number 11. Lee, W. J., Chen, C. Y., Lin, W. C., Wang, Y. T., and Chin, C. J. (1996). “Phosgene formation from the decomposition of 1,1-C2H2Cl2 contained gas in an RF plasma reactor.” J. Hazard. Mat., 48, 51-67. Lou, J. C., and Chang, Y. S. (1997). “Thermal oxidation of chloroform.” Combust. Flame, 109, 188-197. Mak, F. T., Zele, S. R., Cooper, W. J., Kurucz, C. N., Waite, T. D., and Nickelsen, M. G.. (1997). “Kinetic modeling of carbon tetrachloride, chloroform and methylene chloride removal from aqueous solution using the electron beam process.” Wat. Res., 31(2), 219-228. Nichipor, H., Dashouk, E., Chmielewski, A. G., Zimek, Z., and Bulka, S. (2000). “A theoretical study on decomposition of carbon tetrachloride, trichloroethylene and ethyl chloride in dry air under the influence of an electron beam.” Rad. -Phys. Chem., 57, 519-525. Oda, T., Takahahshi T., and Yaaji, K. (2002). “Nonthermal plasma processing for dilute VOCs decomposition.” IEEE Trans. Ind. Applicat., 38(3), 873-878. Penetrante, B. M., Hsiao, M. C., Bardsley J. N., Merritt, B. T., Vogtlin, G. E., Wallman, P. H., Kuthi, A., Burkhart, C. P., and Bayless, J. R. (1995). “Electron beam and pulsed corona processing of carbon tetrachloride in atmospheric pressure gas streams.” Phys. Lett. A, 209, 69-77. Sanhueza, E. (2001). “Hydrochloric acid from chlorocarbons: a significant global source of background rain acidity.” Tellus, 53B, 122-132. WMO-UNEP. (2002). “Scientific assessment of ozone depletion: 2002.” Report No.47 Skoog, D.A., West, D. M., Holler, F. J., and Crouch, S.R. (2000). Analytical chemistry, an introduction. 7th Ed., Saunders College Publishing. Taylor, P. H., and Dellinger, B. (1988). “Thermal degradation characteristics of chloromethane mixtures.” Environ. Sci. Technol., 22, 438-447. The MathWorks, Inc. (1992), MATLAB, reference guide, The MathWorks, Inc., Massachusetts.

20

Tonkyn, R. G.., Barlow, S. E., and Orlando, T. M. (1996). “Destruction of carbon tetrachloride in a dielectric barrier/packed-bed corona reactor.” J. Appl. Phys., 80(9), 4877-4886. U.S. Environmental Protection Agency (EPA), Office of Atmospheric Programs. (April 2002). Greenhouse gases and global warming potential values, EPA 430-R-02-003. Walas, S. M. (1991), Modeling with differential equations in chemical engineering, Butterworth-Heinemann, Boston. Yamamoto, T., Ramanathan, K., Lawness, P. A., Ensor, D. S., Newsome, J. R., Plaks, N., and Ramsey, G. H. (1992). “Control of volatile organic compounds by an AC energized ferroelectric pellet reactor and a pulsed corona reactor.” IEEE Trans. Ind. Applicat., 28(3), 528-534.

21

v e n t to a tm o s p h e r e F ID G C B e n to n e

MFC- 1 TC D G C S K C a rb o n

C o m p re s s e d a ir

Q u a d ra p o le M ass S p e c t ro m e te r

MFC- 2

C C l4 w a t e rb a th K I s o l u t io n

P la s m a R e a c to r

Figure 1. Schematic diagram of experimental set up

22

6000

Voltage (V)

4000 2000 0

-2000 -4000 -6000 0.4

Current (A)

0.2

0.0

-0.2

-0.4

-0.6 -2e-5

-1e-5

0

1e-5

Cycle of time (s)

Figure 2. Voltage and current profile

23

2e-5

Start

Guess: Cl2 selectivity

Set constant: reaction formula new guess Calculate: CCl4 conversion, (CO+CO2) selectivity solve the ODEs Absolute Error: (Cl2)exp.- (Cl2)sim. (CO+CO2)exp.- (CO+CO2)sim. (CCl4)exp.- (CCl4)sim. unsatisfactory Result: (CCl4)sim. (CO+CO2)sim.

End

1.8

3.9

1.6

rε

4.55 1.4

4.55

1.2

5.85 6.5

1 0.8 0.6 0.4 0.2 0 0

5

10

15

20

25

30

Number of iteration steps

Figure 3. (a) The algorithm of simulation;

24

35

40

(b) Error sensitivity of model as function of input guesses variation

25

r5

CCl4

CCl3

r 8,r9

r9

r13

ClO

r1

r3 r4

r14

COCl2

r10

Cl2

r5 r14

CO2

r 10

CO

r11

r11 ,r12 ,r13

COCl

(a)

r1

C C l4 r9

r2

r5 r 8,r9 r13

C lO

C C l3 r3 r4

CO

C 2C l6

r6

r19 r21

r1 r14

C O C l2

r10

r5 r14

2

r10

CO

r7

r17

r11,r12,r 13

C l2

r 20 ,r 21 r19

r18

r18

r11

CO Cl

r15

r2

Figure 4. The model scenario of CCl4 decomposition;

(a) first scenario (b) second scenario

r16

C 2 C l4

(b)

26

C 2 C l5

r2

C C l2

100 90 80

Conversion (%)

70 60 50 40 30 20

180 L/min 240 L/min

10

300 L/min

0 0

1

2

3

4

5

6

7

8

9

CCl4 concentration (%)

(a)

100 90

180 N/min (exp.) 240 L/min (exp.)

(CO+CO2) selectivity (%)

80

300 L/min (exp.) simulation

70 60 50 40 30 20 10 0 0

1

2

3

4

5

CCl4 concentration (%)

(b)

27

6

7

8

9

100 90 80

Cl2 selectivity (%)

70 60 50 40 30 180 L/min (exp.) 20

240 L/min (exp.) 300 L/min (exp.)

10

simulation

0 0

1

2

3

4

5

6

7

8

9

CCl4 concentration (%)

(c)

Figure 5. Effects of initial CCl4 concentration on (a) CCl4 conversion, (b) (CO + CO2) selectivity and (c) Cl2 selectivity

28

Simulation Cl2, CO,CO2 selectivity (%)

100

80

60

40

180 Nl/hr1 180 Nl/hr2 240 Nl/hr1 240 Nl/hr2 300 Nl/hr1 300 Nl/hr2

20

0 0

20

40

60

80

100

Experimental Cl2, CO, CO2 selectivity (%)

Note: 1first model scenario, 2second model scenario

Figure 6. Parity plot of products selectivity comparison between experimental and simulation results.

29

C C l4

e

C C l3 O

O

C C l3 O

C lO

C O C l2

Cl

Cl

2

CO

CO

CO

Cl

CO Cl

C l2

2

Figure 7. Main routes of CO, CO2 and Cl2 production from CCl4 decomposition

30

(a) before

1.4e-9 1.2e-9

Intensity

1.0e-9 CO and N2 8.0e-10 6.0e-10 4.0e-10 2.0e-10

Cl2

CO2

CCl4 CCl4

0.0

(b) after

1.4e-9 1.2e-9 CO and N2

COCl, COCl2

Intensity

1.0e-9 8.0e-10

0

20

40

60

80

100

120

6.0e-10 4.0e-10 CO2

2.0e-10

CCl4

Cl2

0.0 0

20

40

60

80

100

120

140

160

180

200

AMU (m/z)

Note: - obtained at 1 vol% of CCl4 and total gas flow rate of 180 L/h

Figure 8. QMS spectra of CCl4 decomposition; (a) input stream (b) products stream

31

Table 1. Chemical kinetic for decomposition of CCl4

reaction1

log A

n

E

∆Hr(298 K)

Ref.

1.

CCl 2 + Cl 2 = CCl 3 + Cl

12.70

0

3.0

-7.8

[2]

2.

CCl 2 + O2 = ClO + COCl

13.00

0

1.0

-37.1

[2]

3.

CCl 3 + O = COCl 2 + Cl (*)

14.00

0

0

-102.8

[2]

4.

CCl 3 + O2 = COCl 2 + ClO (*)

13.00

0

28.0

-47.7

[2]

5.

CCl 3 + Cl 2 = CCl 4 + Cl

12.40

0

6.0

-11.4

[2]

6.

CCl 3 + CCl 3 = C 2 Cl 6

36.15

-7.48

6.7

-67.5

[2]

7.

CCl 3 + CCl 3 = C 2 Cl 4 + Cl 2

26.35

-4.43

9.0

-40.7

[2]

8.

CCl 4 = CCl 3 + Cl (*)

35.87

-6.52

75.4

70.8

[2]

9.

CCl 4 + O = CCl 3 + ClO (*)

10.40

0

2.3

5.0

[2]

10.

CO + ClO = CO2 + Cl

11.78

0

7.4

-62.2

[2]

32

11.

COCl + Cl = CO + Cl 2 (*)

12.

COCl + M = CO + Cl + M

13.

13.10

0.5

0.5

-52.4

[2]

14.30

0

6.5

7.1

[2]

COCl + O = CO + ClO

14.00

0

0

-58.7

[2]

14.

COCl 2 + Cl = COCl + Cl 2 (*)

13.50

0.5

20

18.5

[2]

15.

C 2 Cl 4 + O = COCl 2 + CCl 2

13.00

0

5.0

-54.3

[3]

16.

C 2 Cl 4 + Cl = C 2 Cl 5

35.42

-7.71

5.3

-17.8

[3]

17.

C 2 Cl 5 + Cl = CCl 3 + CCl 3

27.23

-4.01

12.1

-1.0

[3]

18.

C 2 Cl 5 + Cl = C 2 Cl 4 + Cl 2

27.10

-4.73

8.9

-41.7

[3]

19.

C 2 Cl 6 = Cl 2 + C 2 Cl 4

35.21

-6.53

63.2

26.8

[3]

20.

C 2 Cl 6 = Cl + C 2 Cl 5

36.13

-6.48

74.4

68.5

[3]

21.

C 2 Cl 6 + Cl = Cl 2 + C 2 Cl 5

13.80

0

18.3

(*)

9.0

[3]

n Note: 1 k = AT exp(− E / RT ) , in cm, kcal, s, and mole units; 2See reference: Lou and Chang,

1997; 3See reference: Chang and Senkan, 1989; (*) most significant reactions in this experiment.

33