Download File

Journal: Plasma Devices and Operations

Article ID GPDO 149366

TO: CORRESPONDING AUTHOR AUTHOR QUERIES - TO BE ANSWERED BY THE AUTHOR The following queries have arisen during the typesetting of your manuscript. Please answer the queries.

No queries.

Production Editorial Department, Taylor & Francis Ltd. 4 Park Square, Milton Park, Abingdon OX14 4RN Telephone: +44 (0) 1235 828600 Facsimile: +44 (0) 1235 829000

Plasma Devices and Operations Vol. 00, No. 00, Month 2006, 1–14

Treatment of CCl4 and CHCl3 emission in a gliding-arc plasma ANTONIUS INDARTO*, JAE-WOOK CHOI, HWAUNG LEE and HYUNG KEUN SONG Korea Institute of Science and Technology, Clean Technology Research Center, PO Box 131, Cheongryang, Seoul 130-650, South Korea (Received 24 October 2004) The decomposition of the chlorinated hydrocarbons CCl4 and CHCl3 in a gliding-arc plasma was examined. The effects of initial concentrations, total gas flow rates and power consumption were investigated. The conversion of the hydrocarbons mentioned above was relatively high. It could reach 80% for CCl4 and 97% for CHCl3 . In atmospheric air as a carrier gas, the reaction was exothermic, and the main products were CO2 , CO and Cl2 . The transformation into CCl4 was also detected for the decomposition reaction of CHCl3 . The conversion of these compounds increased with increasing frequency of power supplied and decreasing total gas flow rate. Keywords: Plasma; Gliding arc; CCl4 ; CHCl3 ; Decomposition reaction

1.

Introduction

In various industrial processes, chlorinated volatile organic compounds (CVOCs) are released into the atmosphere with flue gases. Chlorinated hydrocarbons have gained widespread acceptance as solvents and chemical intermediates in the manufacture of herbicides and plastics [1]. Unfortunately, these compounds are toxic and harmful to the environment [2–4], and a serious effort has been made to reduce their emissions. The current technology for the elimination of these emissions is high-temperature (above 800 ◦ C) incineration [5]. This method allows the influent CVOCs to be combusted in air. This method is very simple, but a large amount of complex chlorinated products can be emitted owing to incomplete combustion [6]. Catalytic oxidation, which employs metal and metal oxide catalysts, has also been investigated [7]. It has been reported that this method of conversion is very effective and the product selectivity is very good, but the catalyst applied was easily deactivated by impurities and solid products. Also, this method requires the use of a heater to achieve the necessary temperature for catalyst activation and a satisfactory rate of reaction. One more limitation on the use of this method at industrial plants is the very low input flow rate. *Corresponding author. Email: [email protected]

Plasma Devices and Operations ISSN 1051-9998 print/ISSN 1029-4929 online © 2006 Taylor & Francis http://www.tandf.co.uk/journals DOI: 10.1080/10519990500493833

Techset Composition Ltd, Salisbury

GPDO149366.TeX

Page#:

14

Printed: 6/1/2006

2

A. Indarto et al.

To solve the problem of the elimination of CVOC emissions, many studies have been carried out to find new advanced technologies. Plasma-assisted technologies are promising and effective technologies for destroying CVOCs. Plasma-assisted technologies, such as electron beams [8], radio-frequency plasma [9], surface discharge reactors [10], dielectric barrier discharge reactors [11], pulsed discharge reactors [12] and discharge reactors of capillary-tube type [13] have been studied and developed. In this study, a gliding-arc plasma was used to decompose chloroform (CHCl3 ) and carbon tetrachloride (CCl4 ). Compared with the above plasma-assisted methods, there is a greater chance that a gliding-arc plasma can be applied in the chemical industry. This method allowed the higher temperature and energy that are necessary to destroy toxic input materials to be attained [14]. The possibility of treating a very high input flow rate is one of the advantages of this method.

2.

Experimental set-up

The schematic diagram of the experimental set-up is shown in figure 1. CCl4 , CHCl3 and atmospheric air were used as the source gases. Each part of the system is described in detail in the following section. 2.1 Plasma reactor and power system applied The reactor was made from a quartz-glass tube of inner diameter 45 mm and length 300 mm. The upper part and bottom of the reactor are supplied with a Teflon seal consisting of two electrodes made of stainless steel. The electrodes were of length 150 mm. The separation of electrodes in the narrowest section was 1.5 mm. Gas mixture was injected between these electrodes through a capillary of inner diameter 0.8 mm. A thermocouple, located 10 cm above the electrode, was provided to measure the outlet gas temperature. A high-frequency alternatingcurrent (AC) power supply (Auto electric A1831) with a maximum voltage of 10 kV and a maximum current of 100 mA was connected to the electrode of a gliding arc to generate

Figure 1. Schematic diagram of the experimental set-up: FID GC, flame ionization detector–gas chromatograph; TCD GC, thermal conductivity detector–gas chromatograph.

Treatment of CCl4 and CHCl3

3

a plasma. Figure 2 shows the typical waveforms of voltage and discharge current obtained in this experiment. 2.2

Input gas

The following chlorinated hydrocarbons were used as initial materials: (i) chloroform: CHCl3 ; molecular weight, 119.38; purity, 99.0%; purchased from the Junsei Chemical Co., Ltd; concentration, 1, 3, 5 and 8 vol.%; (ii) carbon tetrachloride: CCl4 ; molecular weight, 153.82; purity, 99.5%; purchased from the Kanto Chemical Co., Inc.; concentration, 1, 3, 5 and 8 vol.%.

Figure 2. Voltage and current profiles.

4

A. Indarto et al.

Atmospheric air was used as the carrier gas and controlled by a Tylan FC-280S mass flow controller. The total flow rates of mixed gas (air + chlorinated compounds) were 180, 240 and 300 N l h−1 . The composition of the outlet mixture was analysed before and after plasma operation. 2.3 Measuring system The composition of outlet gas was examined by a quadrupole mass spectroscope (Balzers QMS 200), with Quadstar 421 software. This software was used to analyse reactants and products qualitatively and quantitatively. Two gas chromatographs were used to measure the amount of products. The contents of CCl4 and CHCl3 in the gas mixture before and after the reaction were determined with a flame ionization detector–gas chromatograph (YoungLin M600D; column, Bentone). A thermal conductivity detector–gas chromatograph (YoungLin M600D; column, SK carbon) was used to determine whether CO and CO2 were available in the outlet gas mixture. Chlorine gas (Cl2 ) was determined by bubbling the gas reacted through 0.05 M aqueous potassium iodide (KI) with subsequent iodometric titration of 0.05 M Na2 SO3 . The performance of the reactor is described in terms of selectivity and the conversion of reaction that are formulated as follows: for the decomposition of CCl4 , moles of Cl2 formed × 100%, 2 × moles of CCl4 converted moles of CO formed selectivity of CO = × 100%, moles of CCl4 converted moles of CO2 formed selectivity of CO2 = × 100%, moles of CCl4 converted moles of CCl4 consumed conversion of CCl4 = × 100%; moles of CCl4 introduced selectivity of Cl2 =

(1) (2) (3) (4)

for the decomposition of CHCl3 , selectivity of Cl2 = selectivity of CO = selectivity of CO2 = selectivity of CCl4 = conversion of CHCl3 =

2 × moles of Cl2 formed × 100%, 3 × moles of CHCl3 converted moles of CO formed × 100%, moles of CHCl3 converted moles of CO2 formed × 100%, moles of CHCl3 converted moles of CCl4 formed × 100%, moles of CHCl3 converted moles of CHCl3 consumed × 100%. moles of CHCl3 introduced

(5) (6) (7) (8) (9)

These parameters were used to study the effects of the initial concentration of chloromethane, total gas flow rates and power frequency. The discharge power was calculated as a product of voltage and current recorded by a digital oscilloscope (Agilent 54641A):
discharge power = [V (t)I (t)] dt × frequency W. (10) In this study all experimental data were taken 30 min after the start of the gliding arc referred to the onset outlet temperature of the bulk gas measured by the thermocouple.

Treatment of CCl4 and CHCl3

3. 3.1

5

Results and discussion Effects of the initial concentration and total gas flow rate

Figure 3(a) shows the influences of initial concentration and total gas flow rate on the degree of conversion of chlorinated hydrocarbons injected. The conversion gradually decreased when the initial concentration increased. This conversion also decreased with increase in the total gas flow rate. The maximum conversion of CCl4 could reach 80% at a concentration of 1% and a total air flow rate of 180 N l h−1 . Under the same conditions, the conversion of CHCl3 could

Figure 3. Influences of the initial concentration and total gas flow rate on the conversion of the injected chlorinated hydrocarbons (a) CCl4 and (b) CHCl3 :
, 180 N l h−1 ; •, 240 N l h−1 ; , 300 N l h−1 .

6

A. Indarto et al. Table 1. Effect of the initial concentration and total gas flow rate on the product distribution of CCl4 decomposition. Concentration (vol.%)

Selectivity

Flow rate (N l h−1 )

Power (W)

CO + CO2

Cl2

180 180 180 180 240 240 240 240 300 300 300

242.6 240.5 239.3 229.6 236.8 234.2 229.4 219.6 227.7 225.6 218.2

48.7 32.4 28.9 25.4 48.0 49.0 41.0 41.7 26.7 25.2 18.2

78.9 73.0 68.0 65.0 63.3 55.1 48.4 45.6 51.6 48.0 42.0

1 3 5 8 1 3 5 8 1 3 5

reach a higher value (97%). Although conversion decreased, the removal efficiency increased with increase in the initial concentration. It was supposed that the Cl released during the decomposition process had a positive effect on decomposition. Besides the influence on conversion, different types of compound, initial concentration and total gas flow rate exert influences on the discharge power consumed. The stability of compounds under plasma treatment plays an important role in the destruction reaction. Tables 1 and 2 show that more energy is required to destroy CCl4 than to destroy CHCl3 . The chemical bond and stability of CCl4 structure were stronger than those of CHCl3 [1]. A stronger stability of a compound requires a higher energy to initiate the arc plasma and is responsible for a lower value of conversion (as can be seen from the comparison between the data for the conversion of CCl4 and CHCl3 in figure 3). This phenomenon showed the same tendency when chlorinated hydrocarbons are treated by combustion [12]. The total gas flow rate also has a significant effect on the power consumed. With increasing total gas flow rate, the total power decreased. This phenomenon can be interpreted in the context of the Paschen law [15]. According to the results of quadrupole mass spectra measurements and gas chromatography analysis, CO, CO2 and Cl2 dominated the gas products. The amounts of other C–Cl compounds were relatively small. In the case of CCl4 , the selectivity of the final product, Cl2 gas, was high. It could reach from 42% up to 78.9%. Although not as high as in the case of CCl4 , the Table 2. Effect of the initial concentration and total gas flow rate on the product distribution of CHCl3 decomposition. Concentration (vol.%) 1 3 5 8 1 3 5 8 1 3 5

Selectivity

Flow rate (N l h−1 )

Power (W)

CO + CO2

CCl4

Cl2

180 180 180 180 240 240 240 240 300 300 300

234.5 232.1 226.8 212.8 222.1 216.6 205.1 175.0 220.0 212.6 197.4

31.1 40.3 42.5 48.6 31.1 34.4 41.0 43.8 22.2 27.4 24.6

6.4 9.7 10.9 11.8 9.4 10.0 10.2 11.8 5.3 7.9 10.0

33.0 34.9 33.8 42.7 38.8 48.3 47.4 50.5 43.3 47.5

Treatment of CCl4 and CHCl3

7

product selectivity of Cl2 on the decomposition of CHCl3 was also high, from 33% up to 50.5%. The increase in the initial concentration of CCl4 leads to reduced product selectivity, thus allowing a variation in the product distribution. This phenomenon did not occur in the case of CHCl3 , when the selectivity of Cl2 gas was relatively constant. In the case of CO and CO2 production, the selectivity has the same trend as that of Cl2 , especially in the case of CCl4 decomposition. It could be assumed to be an effect of the C vacancy due to the loss of Cl species. When CCl4 was destroyed to give two Cl2 molecules, the C radical remained as an intermediate species. This heightens the chance that C radicals (C vacancies) collide with other compounds, thus raising the possibility that C radicals collide with oxygen and CO or CO2 compared with the production of other compounds. CO and CO2 have stable structures, and the concentration of oxygen gas (O2 ) in the air stream is high (about 20%). The processing of the quadrupole mass spectra of O2 has shown that the concentration of O2 in the output product (after treatment with plasma) is 5–10% lower than its concentration in the inlet stream (before treatment with plasma). One possible reason for this fact is the consumption of O2 in the reactions of CO and CO2 formation. The ratio of CO2 produced to CO present in the final product stream was about (1/10)–(1/5). On decomposition of CHCl3 , CCl4 was one of the products formed. The selectivity did not exceed 12%. During the experiment, traces of other compounds, such as COCl2 and HCl, were also detected. Liquid products were also produced that might be considered as remaining species of C and Cl. 3.2 Effect of the frequency As was mentioned above, the discharge power supplied to the plasma system was of considerable importance for the decomposition reaction. This factor was also important in governing the product distribution. Of the three variables that exert influences on the product distribution in this experiment, only the power frequency could be adjusted manually. The voltage and current were set automatically after the start-up of the gliding arc. The effect of the power frequency is shown in figure 4. The maximum power frequency was 20 kHz. As can be seen from figure 4, conversion increased with increase in the frequency. For CCl4 , the rate of conversion reached 4% per 1 kHz increment in frequency, and it was slightly lower for CHCl3 . The increase in the frequency leads to a higher discharge power supplied. Figure 5 shows the increase in the discharge power due to the higher power frequency. The change in frequency affects the AC waveform and results in a larger number of peaks per cycle and higher energy supplied to the system. The change in the power waveform due to the variation in the power frequency applied is shown in figure 6. Figure 7 shows that, by varying the frequency or discharge power, we obtain different production selectivities of Cl2 at the same initial concentration and flow rate. This testifies to the fact that the power frequency and discharge power are the parameters influencing the reaction pathway. With increasing power frequency, the selectivity of Cl2 also increased. It can be supposed that, at higher energies, the plasma reactor is likely to produce Cl2 gas as the final product. 3.3 Quadrupole mass spectroscopy The quadrupole mass spectra recorded in the conversion of CCl4 at the 180 N l h−1 feed stream containing atmospheric air with 1% of CCl4 are shown in figure 8. The basic spectra of CCl4 are characterized by the lines m/z = 83/85 (CCl+ 2 ), and m/z = + 117 (CCl+ ) (see figure 8(a)). CO has lines at m/z = 44 (CO ), and the CO spectrum coincides 2 3 2

8

A. Indarto et al.

with the minor N2 spectra at m/z = 29 (CO+ –N+ 2 ). The major N2 lines are located at m/z = 28 (N+ ). As the concentration of CO in air is quite low, its lines can be indistinguishable in 2 spectroscopic measurements. After treatment with plasma, the peak of CCl+ 2 (m/z = 83/85) vanished simultaneously with the decrease in the intensity of the line m/z = 117, and a new peak, m/z = 71, appeared, which was identified as Cl2 (Cl+ 2 ) (see figure 8(b)). The intensity of the CO2 line also increased, but the difference between the cases before and after discharge was very small. As N2 is a relatively stable gas and there is no another source of N, it can be said that the increasing

Figure 4. Effect of the frequency of input power on the conversion of CCl4 and CHCl3 :
, 180 N l h−1 ; •, 240 N l h−1 ; , 300 N l h−1 . Data were taken at 1 vol.% CCl4 and 1 vol.% CHCl3 .

Treatment of CCl4 and CHCl3

9

Figure 5. Effect of the applied power frequency on the consumed discharge power. Data were taken at 1 vol.% CCl4 and 1 vol.% CHCl3 .

intensity of the line m/z = 28 can be attributed to the production of CO during discharge burning. The mass spectroscopy of CHCl3 before and after treatment with plasma is shown in figure 9. Figure 9(a) shows the lines of components at the 240 N l h−1 feed stream containing compressed air mixed with 1% CHCl3 . The basic lines of CHCl3 were m/z = 82/84/86 (CCl+ 2 ).

Figure 6. Effect of the applied power frequency on the power-profile waveform. Data were taken at 5 vol.% CHCl3 and a total gas flow rate of 180 N l min−1 .

10

A. Indarto et al.

Figure 7. Effect of the applied power frequency on Cl2 gas production selectivity. Data were taken at 1 vol.% CCl4 and 1 vol.% CHCl3 .

CO2 has spectra at m/z = 44 (CO+ 2 ), but this spectrum coincides with the major spectra of N2 O (N2 O+ ) that can be present in gas products. The CO line was at m/z = 29 (CO+ ), but it also coincided with the minor spectra of N2 , m/z = 29 (N+ 2 ). It was difficult to distinguish these spectra using quadrupole mass spectroscopy. In this experiment, CO and CO2 species were analysed by gas chromatography. When the plasma burnt, the peak intensity of m/z = 83/84/86 decreased, and some new peaks appeared: m/z = 63/65, which is known to be COCl2 (COCl+ ), m/z = 69/71, which is known to be Cl2 (Cl+ 2 ), and m/z = 117, which is known to be CCl4 (CCl+ ) (see figure 9(b)). 3 3.4

Reaction pathway

The kinetic reaction of CO, CO2 and Cl2 formation from chlorinated hydrocarbons has been studied and reported [5, 8, 16–20]. The major pathways responsible for the formation of Cl2 were COCl2 + Cl = COCl + Cl2 ,

(11)

CHCl3 + Cl = CHCl2 + Cl2 ,

(12)

COCl + Cl = CO + Cl2 .

(13)

However, the third reaction (equation (13)) was also one of the primary reaction paths to form CO. Another path was COCl + M = CO + Cl + M. (14) It was seen that CO was initially formed through COCl2 . CO2 was formed mainly from CO: CO + ClO = CO2 + Cl,

(15)

CO + O + M = CO2 + M,

(16)

Treatment of CCl4 and CHCl3

11

Figure 8. Quadrupole mass spectrogram of CCl4 decomposition (a) before and (b) after treatment with a gliding plasma (taken at an applied frequency of 20 kHz, 1 vol.% CCl4 and a total gas flow rate of 180 N l h−1 ).

in which ClO was formed via CCl4 + O = CCl3 + ClO,

(17)

CCl3 + O2 = COCl2 + ClO.

(18)

This means that O2 and the O radical make a significant contribution to this reaction. The production of O radical is the result of a dissociation reaction [21]: e + O2 = e + O(3 P) + O(3 P),

(19)

e + O2 = e + O( P) + O( D).

(20)

3

1

12

A. Indarto et al.

The Cl radical in this formula can be obtained from CCl4 = CCl3 + Cl, CCl3 + O = COCl2 + Cl, CCl3 + Cl2 = CCl4 + Cl.

(21) (22) (23)

The reverse reaction can proceed in compliance with equation (21), as the main path to produce CCl4 is the decomposition of CHCl3 : CCl3 + Cl = CCl4 .

Figure 9. Quadrupole mass spectrogram of CHCl3 decomposition (a) before and (b) after treatment with a gliding plasma (taken at an applied frequency of 20 kHz, 1 vol.% CHCl3 and a total gas flow rate of 180 N l h−1 ).

Treatment of CCl4 and CHCl3

4.

13

Conclusions

The decomposition of CCl4 and CHCl3 in a gliding-arc plasma at atmospheric pressure have been studied at different initial concentrations, total gas flow rates and input power frequencies. A gliding-arc plasma can generate electrons and ions effective enough to decompose molecules. The maximum conversion of CCl4 was 80% and that of CHCl3 was 97% for the stream of input gas containing 1% v/v and with a total air flow rate of 180 N l h−1 . The major final products were CO, CO2 and Cl2 . The selectivity of Cl2 was relatively high and amounted to 78.9% for CCl4 and 50.5% for CHCl3 . The conversion of CHCl3 into CCl4 was found, but it did not exceed 12%. COCl2 plays an important role as an intermediate species to produce Cl2 and CO. Mostly CO2 was produced by CO reaction, from which it can be concluded that a gliding-arc plasma was very effective for the decomposition of CCl4 and CHCl3 . To remove some traces of unwanted compounds, such as COCl2 , in the final product, further development is needed.

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

References [1] Kirk-Othmer, Encyclopedia of Chemical Technology, Vol. 5, 3rd edition (Wiley, New York, 1978), pp. 668–713. [2] M.E. Meek, R. Beauchamp, G. Long et al. Chloroform: exposure estimation, hazard characterization, and exposure-response analysis. J. Toxicol. Environ. Health, Part B, 5, 283 (2002). [3] Overall Evaluations of Carcinogenicity to Humans, IARC Monographs, Supplements 7 (International Agency for Research on Cancer, Lyon, 1987). [4] Greenhouse gases and global warming potential values. Report, US Environmental Protection Agency, April 2002. [5] L.C. Lou and Y.S. Chang. Thermal oxidation of chloroform. Combust. Flame, 109, 188 (1997). [6] P.H. Taylor and B. Dellinger. Thermal degradation characteristics of chloromethane mixtures. Environ. Sci. Technol., 22, 438 (1988). [7] R.M. Alberici and W.F. Jardim. Photocatalytic destruction of VOCs in the gas-phase using titanium dioxide. Appl. Catal. B: Environ., 14, 55 (1997). [8] L. Prager, H. Langguth, S. Rummel et al. Electron beam degradation of chlorinated hydrocarbons in air. Radiat. Phys. Chem., 46, 1137 (1995). [9] W.J. Lee, C.Y. Chen, W.C. Lin et al. Phosgene formation from the decomposition of 1,1-C2 H2 Cl2 containing gas in an RF plasma reactor. J. Hazardous Mater., 48, 51 (1996). [10] T. Oda, T. Takahahshi and K. Yaaji. Nonthermal plasma processing for dilute VOCs decomposition. IEEE Trans. Industry Applic., 38, 873 (2002). [11] R.G. Tonkyn, S.E. Barlow and T.M. Orlando. Destruction of carbon tetrachloride in a dielectric barrier/packedbed corona reactor. J. Appl. Phys., 80, 4877 (1996). [12] T. Yamamoto, K. Ramanathan, P.A. Lawness et al.Control of volatile organic compounds by an AC energized ferroelectric pellet reactor and a pulsed corona reactor. IEEE Trans. Industry Applit., 28, 528 (1992). [13] H. Kohno, A.A. Berezin, J.S. Chang et al. Destruction of volatile organic compounds used in a semiconductor industry by a capillary tube discharge reactor. IEEE Trans. Industry Applic., 34, 953 (1998). [14] A. Czernichowski. Gliding arc: applications to engineering and environment control. Pure Appl. Chem., 66, 1301 (1994). [15] J.D. Cobine, Gaseous Conductor Theory and Engineering Application (Dover Publications, New York, 1958), pp. 160–177. [16] M. Koch, D.R. Cohn, R.M. Patrick et al. Electron beam atmospheric pressure cold plasma decomposition of carbon tetrachloride and trichloroethylene. Environ. Sci. Technol., 29, 2946 (1995). [17] H. Nichipor, E. Dashouk, A.G. Chmielewski et al. A theoretical study on decomposition of carbon tetrachloride, trichloroethylene and ethyl chloride in dry air under the influence of an electron beam. Radiat. Phys. Chem., 57, 519 (2000).

14

A. Indarto et al.

[18] W.-D. Chang and S. Senkan. Detailed chemical kinetic modeling of fuel-rich trichloroethane/oxygen/argon flames. Environ. Sci. Technol., 23, 442 (1989). [19] C. Feiyan, S.O. Pehkonen and M.B. Ray. Kinetics and mechanisms of UV-photodegradation of chlorinated organics in the gas phase. Water Res., 36, 4203 (2002). [20] Y.-P. Wu and Y.-S. Won. Pyrolysis of chloromethanes. Combust. Flame, 122, 312 (2000). [21] B.M. Penetrante, M.C. Hsiao, J.N. Bardsley et al. Electron beam and pulsed corona processing of carbon tetrachloride in atmospheric pressure gas streams. Phys. Lett. A, 209, 69 (1995).