This action might not be possible to undo. Are you sure you want to continue?
Conversion of CO2 by Gliding Arc Plasma
Antonius Indarto1, Jae-Wook Choi, Hwaung Lee, Hyung Keun Song
Korea Institute of Science & Technology, Clean Technology Research Center, P.O. Box 131, Cheongryang, Seoul 130-650, Korea
Correspondece author: Plasma-Catalyst Chemical Process Laboratory, Korea Institute of Science and Technology (KIST), PO Box 131, Cheongryang, Seoul 130-650 Phone number: 82-10-2296-3748 Email: email@example.com
Abstract- Decomposition of carbon dioxide (CO2) by gliding arc plasma was examined. The plasma device consisted of two or four triangles stainless steel plates. The gas enters through a nozzle tube from the upper part of cylinder reactor and exits at the bottom part of the reactor. Some additional gases, such as N2, O2, air, and H2O, have been used to study the dilution effects on CO2 conversion. A simple kinetic reaction model was made to investigate the pathways of plasma reaction. Experiment results indicate that the conversion of pure CO2 reaches 18% at the total gas flow rate of 0.85 L/min and produces CO and O2 as the final products. The effects of additional gases injection will be described detail in this paper.
Keywords: Plasma, Gliding Arc, CO2, kinetic reaction, conversion
The conversion of priceless material, such as carbon dioxide (CO2), into valuable gases is still becoming a challenge now (Woolf, 2004; Yamaji, 1997). CO 2 is a wellknown material as main source of green house gas that contributes and responsible on the climate change (Shah and Singh, 1998; US EPA, 2002). Concerning this situation, Kyoto protocol obligates industrialized countries to cut their greenhouse gases emission by an average 5.2% between 2008 and 2012. Nowadays, around 2 × 1015 g per annum of CO2 has been released to the atmosphere from many sources. This situation gives a clear description for the need of finding effective methods to reduce CO2 emission. Industrial sector has been suspected as the main contributor on emitting numerous amount of CO2 to the ambient air (Kiani et al., 2004; Yabe, 2004). However, the main problem to reduce the concentration of CO2 from industrial gas waste is the strong chemical bond of CO2. Based on thermodynamic calculation, the chemical structure of CO2 starts broken at 1500oC and it will be completely broken at temperature >5000oC. It means that large amount of energy has to be supplied to the system to achieve a required process temperature. Some research studies related to the CO2 reaction have been reported scientifically. Ren et al. (2002) tried to convert CO2 into CO and O2 using combustion based powergeneration cycle process. Another high thermal approach has been done also by using membrane-equipped glass furnace (Fan et al., 2003). More advance investigation was done by Clavenna and Schmidt (1972) which explains rigorously the CO decomposition on polychristalline W. Instead of breaking the C-O bond, some
experiments of CO2 reduction were done by chemical process. Shin et al. (2002) used (NixCu1-x)Fe2O4 solid solution as the media for absorbing CO2 existed in the air chamber. Heating rate program was applied to optimize the absorption process of CO2 and CO flow was kept constant at 60 ml/min. In recent years, some studies tend to convey the application of new technologies to reduce the emission of CO2 rather than using conventional (old fashion) systems. But, direct conversion process is still preferred as the main choice. Plasma-assisted methods, such as RF plasma (Savinov et al., 1999), corona (Wen and Jiang, 2001; Maezono and Chang, 1990), dielectric barrier discharge (Li et al., 2004), glow discharge (Wang et al., 1999; Buser and Sullivan, 1970), and thermal plasma ( Kobayashi et al., 2002) have been develop. In this study, the gliding arc plasma was performed as one of the promising media to decompose CO2 emission. In our previous experiment, methane conversion (Indarto et al., 2005) and decomposition of chloromethane has been done successfully using the gliding arc plasma. Under gaseous state, direct thermal of CO2 decomposition goes to produce CO as the possible target reaction. 2CO2 2CO + O2 ∆H = 552 kJ/mol (1)
Reaction is occurred in highly exothermic reaction. Tschauner et al. (2001) by has proved the decomposing CO2 in high pressure system will produce oxygen and diamond. CO2 C + O2 ∆H = 393.5 kJ/mol (2)
He showed the breaking process of the bonds was occurred via a new phase state of CO2. In case of plasmachemical process, Hoag et al. (1973) proves that the reaction 1 is the main pathway reaction of CO2 decomposition. This result has been supported by
Li et al. (2004) which produced CO as the main product in the DBD reaction. Following Maezono and Chang (1990) report, discharge current has an important factor that should be considered to increase the conversion of CO2. Andrev et al. (2004) has showed that the effective way to control the CO2 reduction is mainly caused by electric field strength (E) and neutral gas plasma particle (N). E/N is an important factor for influencing the vibrational temperature. Different basic variables of plasma operation, such as gas flow rate, initial concentration, power voltage, rod metal, and barrier material were investigated also (Wang et al., 1999). However, no one has been trying to investigate the effect of carrier gas on the direct conversion of CO2. This paper presents the effect of different carrier gas on CO2 conversion. Simulation of pure CO2 conversion is also done to know the kinetic pathway of basic CO2 reaction conversion mechanism. Application of gliding arc to decompose CO2 is another novelty of this paper.
2. Experimental Setup
The schematic diagram of experimental setup is shown in Figure 1. CO2 was used as the main input gas with purity 99% and controlled with a Mass Flow Controller (Tylan, FC-280S). The total flow rate was varied from 0.85 to 2 L/min. The composition of the outlet mixture was analyzed before and after plasma operation. The reactor was made of a quartz-glass tube of inner diameter of 45 mm and length of 250 mm. The upper part supplied with a teflon seal comprised two 150 mm in length of triangle electrodes made from stainless steel. The separation of the electrodes in the
narrowest section was only 1 mm. The gas mixture was introduced between the electrodes through a capillary with inner diameter of 0.3 mm. A high frequency AC power supply (Auto electric, A1831) was connected to the gliding arc electrodes. The maximum voltage was 10 kV and current was 100mA. Figure 2 shows typical waveforms of voltage and discharge current used in this experiment.
Concentration of CO2, also CO and O2, in the gas mixture before and after the reaction ware measured by GC-TCD (YoungLin M600D, Column: SK Carbon). To evaluate the performance of system, selectivity and conversion were used and defined as: Conversion of CO2 = moles of CO2 consumed × 100% moles of CO2 introduced
Selectivity of CO =
moles of CO formed × 100% 2 × moles of CO2 converted moles of O2 formed × 100% moles of CO2 converted moles of H 2 formed × 100% moles of H 2 O converted
Selectivity of O2 =
Selectivity of H 2 =
The power supplied was calculated as a product of voltage and current by oscilloscope (Agilent 54641A).
3. Result and Discussion 3.1. Pure CO2 conversion
The experiment series were started by varying the input flow rates of pure CO 2 to the reactor. Different flow rates of CO2 gas will effect on the residence time of gas particle in the plasma reactor. The flow rates ware set from 0.85 L/min into maximum of 1.45 L/min while the power frequency was kept constant at 20 kHz. Figure 3 shows the changes of CO2 conversion according to the various flow rates. The conversion of CO2 slightly decreases from 16.5% to 15% when the gas flow increases. This conversion result is significantly higher than when it is conducted by Corona discharge (Maezono and Chang, 1990). Although the conversion percentage relatively same compared to the DBD discharge (Wang et al., 1999), this system can handle higher input flow rates (~40× higher). However, increment of flow rate reduces the residence time of CO2 gas in the reactor. It reduces the chance and time of CO2 gas to collide with electrons and other high energy state species which have enough energy to break the carbon-oxygen bond.
3.2. Product selectivity and simulation
A simple plasma kinetic model has been developed to investigate the plasma reaction process. The model was structured by multiple radical reactions and it was coded in Matlab program. To solve a set of differential equations problem simultaneously, Matlab module ‘ode23’, has been used as a solver tool (Mathworks, 1994). Special
attention is given on the reaction of CO2, CO, and O2 production because in all range of experiment conditions, those molecules are produced significantly (as shown in figure 4). Figure 5 and 6 show the plasma reaction diagram of CO 2 decomposition and CO reaction. The list of kinetic reactions is collected from literatures (Maezono and Chang, 1990; Atkinson et al., 2004; NASA, 1994). The basic kinetic reaction follows the Arrhenius rule: dC A = r = k ⋅CA dt E k = A ⋅ exp A R ⋅T (7)
and the applied reaction temperature is taken from Fridman et al. (1999) paper.
Following the kinetic calculation, reaction of CO2 conversion is mostly initiated by collision between CO2 molecule and electron (e). Previously, Elliasson (1991) has reviewed the importance of fast electron to initiate the chemical reaction in nonequilibrium plasma. This idea is supported by Buser and Sulivant (1970) by listing the importance of electron in the dissociation reaction (e + kinetic energy) + CO2 CO2* + e (direct ionization) CO2* CO + O (10) (9)
However, CO production is not only produced by excited neutral CO2 molecule but
also electron can involve in the reaction and attach to the CO2 molecule to produce intermediate (CO2-)* species CO2 + e + ∆E (CO2-)* CO + O(11)
Another reaction mechanism that has a possibility to be occurred is a dissociative ionization (Maezono and Chang, 1990) CO2 + e CO+ + O + 2e (12)
In our model simulation, it shows that reaction of CO 2 molecule with e can be separated into two kinds of reaction: (i) direct reaction which produced CO and O2, CO2 + e C+ +O2 + 2e CO + O + e (13) (14)
and (ii) intermediate reaction which produces high energetic intermediate species and ions. Among the radical species and ions, three species (O, O 2+, and CO+) have an important role to control way of reaction. Except direct reaction e with CO2, O2 can be produced by O3 reaction, O + O3 + M O + O3 2 O2 + M 2 O2 (15) (16)
or recombinant process, O2+ + CO2 + e CO2 + O2 (17)
CO+ has significant role in CO production by, CO+ + O CO+ + CO2 CO + O+ CO + CO2+ (18) (19)
Figure 4 shows the selectivity comparison of CO and O2 between experiment and
model calculation. The selectivity of CO reaches ∼30% and O2 reaches ∼35%. Good achievement that calculation of oxygen atom balance of the experiment is close to 1. It means that total input oxygen equals to the total oxygen in the output stream. Unfortunately, atomic C balance is not calculated because of our instrument limitation. The amount of carbon solid that comes from the plasma (reaction 2) can not be detected quantitatively. Although it is not completely analyzed, it can be summarized that most plasma reaction produced O2 as the main product and more than 60% of CO2 is converted into CO. This is a good result because O2 is more valuable and useful gas compared to CO2. Although CO is classified as toxic gas, this gas is more reactive than CO2 and nowadays, it can be utilized as a part of synthesis gas and other polymer product, such as polycarbonates (Goyal et al., 2000).
3.4. Effect of Additional gases
Three different gases have been used and mixed to study their effects on CO2 conversion. Figure 7 shows the input concentration vs. CO2 conversion at the fixed gas flow rate of 2 L/min. For all range of initial CO2 concentration, only N2 gas gives positive effect on CO2 conversion. Other gases, such as (atmospheric) air and O2, produce an opposite result. Increasing N2 content, from 75% to 95%, increases the rate of CO2 conversion from 19% to 35%. In our previous work, we found that at low nitrogen concentration, N 2 molecule was not involved in plasma-chemical reactions, but at higher concentration,
it has a high possibility to contribute in the reaction mechanism by excitation of N2 molecules (Savinov et al., 2003). This excitation of N2 in the plasma zone into higher excitation level, both neutral and ion, will help increasing the conversion of methane. For example, exited ground level of N2 (A3∑u+) has a 9.7 eV energy to be released before it returns to the ground state (Goldman and Amouroux, 1981) although it is still debated whether vibrational excited state (X1∑g+) (Kenty, 1962; Milne, 1970; Sadowski et al., 1972; Haug etl al., 1974) or A3∑u+ state collision (Cann et al., 1972). More over, increasing the ratio of N2, the rate of energy transferred from the electrons to N2 vibration increases and it produces a greater probability for energy transfer by colliding vibrationally excited N2 and CO2 (Bardelay and Pinson, 1979). In the other hand, this phenomena is not occurred in O 2 case. Although O2 has an excitation level, the potential energy is relatively small to be transferred, only 0.98eV from a1∆g and 1.63eV from b1∑g+ to the ground state (Goldman and Amouroux, 1981). In term of photoionization, unfortunately, O2 requires more input energy compared to CO2 and CO (Alagia et al., 2001). Comparison the bond energy factor shows that O2 is much difficult to be broken than CO2. Those will effect on the reducing number of ion and active species in the plasma system. Reducing conversion target material which is caused by increasing O2 content was found also in chloromethane decomposition (Snyder, 1998). However, the existence of exited O2 both in neutral or ion condition, possibly produce the reverse reaction of CO2 from C C + (O2)* CO2 C + (O2-)* CO2 + e C + (O2+)* + e CO2 or from CO by exited single O or radical (O∙) (20) (21) (22)
CO + O* CO2
Using atmospheric air as the gas mixture, CO2 conversion trend is existed between N2 and O2 result. The effect of some existing impurities can be negligible due to small concentration. Moreover, it can be said that the phenomena will be combination of N 2 and O2. When it is compared with pure CO2 conversion, the overall conversion is lower. It means that existing O2 molecule gives more dominant physical effect compared to N2 molecule although the concentration of N2 higher than O2. The existence of high concentration O2 molecule exhibits the production of active species. Some references suggest that in few concentration, O2 molecule gives some positive effects ( Jamshidi, 1978), but unfortunately, in this experiment that range is not experienced.
The product selectivity has been measured also. Figure 8 shows the effect of CO 2 content on the product selectivity of CO. The general trend for all case is that the selectivity of CO approached to the 35% with increasing CO2 content. At the same gas flow rate, 2 L/min, the converted pure CO2 experiment into CO reaches 35%. However, good analogy here can be used to explain these results. In case of CO 2 conversion with O2, the existence of O from O2 can help to increase the selectivity of CO when it collides with C (Goldman and Amouroux, 1981) e + O2 O+ + O + 2e e + O2 O- + O (dissociative ionization) (dissociative attachement) (24) (25)
e + O2 O- + O+ + e e + O2 2O + e and C + <O> CO
(disoociative attachement) (dissociation)
<…> means that it can be in active states: excited neutral/ion or radical. In case of N2, the existence N in any form and energy level can reduce the probability of C to be CO. And also, N can react with single O or double O atoms and converted into NO, N2O, and NO2, although the selectivity is quite less (Suib et al., 1998).
3.5. Effect of water
The experiment on water (H2O) injection and its effect on CO2 conversion were done. The purpose of this experiment is to react between CO 2 and H2O to produce fuel products such as H2 and CH4 which is still challenging (Anpo and Chiba, 1992). Figure 9a shows the existence water in all range concentration at 2 L/min has decreased the conversion of CO2. In our observation, moisture water makes the gliding plasma production becomes unstable and the instability increases with increasing water content. This can be the reason why the conversion gets decreased. Generally, in all range of water content, the conversion is less than 10%. In order to increase the conversion rate, additive gas N2 is added which has been proved to give a positive effect as experimentally before. Figure 9b shows by adding N2, the conversion has increased and higher than the conversion at pure CO 2 injection.
The highest conversion is occurred at H2O:CO2:N2=0.05:0.055:0.94, or at the highest concentration of N2 and lowest concentration of H2O. When this result is compared to figure 7, although the concentration of N2 is equal, reducing effect which was caused by water is much more dominant. For example, in figure 7, at 95% of N 2 concentration, the conversion reaches 35% while in figure 9b, at similar condition, the conversion is only 29%. In gliding arc plasma system, the property of water might be same with atmospheric air. When N2 gas is changed with air (figure 9b), the conversion is similar with result in figure 9a.
The possibility of H2O converted into H2 is also investigated. The mixing ratio is exactly same with the previous system (figure 9b) where N2 is injected. Without N2 injection, the conversion of water is small and H2 product can not detected well in our GC system. Figure 10 shows the selectivity of H2. Unfortunately, the detected H2 is small, not exceed than 6% in all range of experiment.
The performance of CO2 decomposition in gliding arc plasma at atmospheric pressure was studied. Some additional gases were used to investigate the conversion efficiency. Proved by the similarity between experimental and simulation result, the conversion
reaction was initiated by electron. The existence of excited N2 level gives a positive effect while O2 and air produce an opposite effect which might be caused by reverse reaction C and CO to CO2. The conversion of CO2 reaches 35% at N2 concentration 95%, higher that pure CO2 injection conversion which is only 15-18%. The plasma reaction produces CO and O2 as the main products. Existing water in the plasma reaction decreases the CO2 conversion and the selectivity of H2 from water is less than 6%. Some process modification should be done, for example catalyst addition, to enhance better experimental results.
Acknowledgment This study was supported by the National Research Laboratory program of the Korea Minister of Science and Technology.
Alagia, M., Coreno, M., de Simone, M., Richterc, R., Stranges, S., 2001. High resolution K-edge spectroscopy of oxygen transient species: the metastable O2 a1∆g molecule and the O (3P) atom. J. Elec. Spectro. Related Phe. 114–116, 85–92. Andreev, S.N., Zakharov, V.V., Ochkin, V.N., Savinov, S.Yu., 2004. Plasma-chemical CO2 decomposition in a non-self-sustained discharge with a controlled electronic component of plasma. Spectro. Acta Part A. 60, 3361–3369. Anpo, M., Chiba, K., 1992. Photocatalytic reduction of CO2 on anchored titanium oxide catalyst, J. Mol. Catal. 74, 207. Atkinson, R., Baulch, D.L., Cox, R.A., et. al., 2004. Evaluated kinetic and photochemical data for atmospheric chemistry: volume i-gas phase reaction of Ox, HOx, NOx and SOx species. Atmos. Chem. Phys. 4, 1461-1738. Bardelay, J., Pinson, P., 1979. Calculation of vibrational excitation in CO2, CO, N2, He discharges. Phys. Letters. 71A(5,6), 427-430. Buser, R.G., Sullivan, J.J., 1970. Initial process in CO2 glow discharges. J. Appl. Phys., 41(2), 472479. Cann, R.C., Kaufman, F., Biondi. M.A., 1972. Interferometric study of the chemiluminescent excitation of sodium by active nitrogen. Chem. Phys. Letters. 16, 380-384. Clavenna, L.R., Schmidt, L.D., 1972. Decomposition of CO2 on (100) W. Surf. Scie. 33, 11-26. Eliasson, B., Nonequilibrium volume plasma chemical processing, 1991. IEEE Trans. Plasma Sci. 19(6), 1063-1077. Fan, Y., Ren, J-Y., Onstot, W., Pasale, J., Tsotsis, T.T., Egolfopoulos, F.N., 2003. Reactor and technical feasibility aspects of a CO2 decomposition-based power generation cycle, utilizing a hightemperature membrane reactor. Ind. Eng. Chem. Res. 42, 2618-2626. Fridman, A., Nester, S., Kennedy, L.A., Saveliev, A., Yardimci, O.M., 1999. Gliding arc gas discharge. Prog. Energy Combust. Sci. 25, 211. Goldman, A., Amouroux, J., 1981. Plasma Chemistry. In: Kunhardt, E.E. and Luessen, L.H., (ed.). Proceedings Electrical Break and Discharges in Gases. Plenum Press, New York, pp 293-346. Goyal, M., Nagahata, R., Sugiyama, J., Asai, M., Ueda, M., Takeuchi, K., 2000. Pd catalyzed polycarbonate synthesis from bisphenol A and CO: control of polymer chain—end structure. Polymer. 41, 2289–2293. Haug. R., Rappeneckjzr, G., Schmidt, C., 1974. Collisional ionization of alkali atoms by vibrationally excited N2. Chem. Phys. 5, 255-264. Hoag, E., Pease, H., Staal, J., Zar, J., 1973. Performance characteristics of a 10-kW industrial CO2 laser system. IEEE J. Quant. Electron. 9(6), 652-652. Indarto, A., Choi, J-W., Lee, H., Song, H.K., 2005. Kinetic modeling of plasma methane conversion using gliding arc. J. Nat. Gas Chem., 14, 13-21. Jamshidi, E., 1978. The effect of a small amount of O2 on vibrational excitation of N2. J. Photochem, 9, 16
1–9. Kenty, C., 1962. Evidence for a long-lived metastable state of N2 at about 8 eV. J. Chem. Phys. 37, 1567-1568.Kiani, B., Hamamoto, Y., Akisawa, A., Kashiwagi, T., 2004. CO2 mitigating effects by waste heat utilization from industry sector to metropolitan areas. Energy. 29, 2061–2075. Kobayashi, A., Osaki, K., Yamabe, C., 2002. Treatment of CO2 gas by high-energy type plasma 65, 475-479. Li, R., Yamaguchi, Y., Yin. S., Tang, Q., Sato, T., 2004. Influence of dielectric barrier materials to the behavior of dielectric discharge plasma for CO2 decomposition. Sol. State Ion. 172, 235-238. Maezono, I., Chang, J-S., 1990. Reduction of CO2 from combustion gases by DC corona torches. IEEE Trans. Ind. Appl. 26(4), 651-655. Mathworks, 1994. MATLAB Reference Guide. 3rd printing. The Mathworks Inc, Massacuset. Milne, E.L., 1970. Sodium-atom excitation in nitrogen afterglows. J. Chem. Phys. 52, 5360-5365. NASA Panel for Data Evaluation. 1994. Chemical kinetics and photochemical data for use in stratospheric modeling. Evaluation Number 11. Ren, J-Y., Tsotsis, T.T., Egolfopoulos, F.N., 2002. Basic aspects of combustion stability and pollutant emissions of a CO2 decomposition-based power-generation cycle. Ind. Eng. Chem. Res. 41, 45434549. Sadowski, C.M., Schiff, H.I., Chow, G.K., 1972. Electronic excitation of Na by vibrationally excited N2. J. Photochem. 1, 23-38. Savinov, S.Y., Lee, H., Song, H.K., Na, B-K., 1999. Decomposition of methane and carbon dioxide in a radio-frequency discharge. Ind. Eng. Chem. Res. 38, 2540-2547. Savinov, S.Y., Lee, H., Song, H.K., Na, B-K., 2003. The effect of vibrational of molecules of plasmachemical reaction involving methane and nitrogen. Plasma Chem. Plasma Process. 23 (1), 159-173. Shah, J.J., Singh, H.B., 1998. Distribution of volatile organic chemicals in outdor and indoor air. Environ. Sci. Technol. 22, 1381. Shin, H.C., Oh, J.H., Lee, J.C., Han, S.H., Choi, S.C., 2002. The carbon dioxide decomposition reaction with (NixCu1-x)Fe2O4 solid solution. Phys. Stat. Sol. (a). 189(3), 741–745. Snyder, H.R., 1998. Effect of air and oxygen content on the dielectric barrier discharge decomposition of chlorobenzene. IEEE Trans. Plasma Sci. 26(6), 1695-1699. Suib, S.L., Brock, S.L., Marquez, M., Luo, J., Matsumoto, H., Hayashir, Y., 1998. Efficient Catalytic Plasma Activation of CO2, NO, and H2O. J. Phys. Chem. B. 102, 9661-9666. Tschauner, O., Mao, H-K., Hemley, R.J., 2001. New transformations of CO2 at high pressures and temperatures. Phys. Rev. Let. 87(7). U.S. Environmental Protection Agency, 2002. Greenhouse gases and global warming potential values. Wang, J-Y., Xia, G-G., Huang A., Suib, S.L., Hayashi, Y., Matsumoto, H., 1999. CO 2 decomposition using glow discharge plasmas. J. Catal. 185, 152-159. Wen, Y., Jiang, X., 2001. Decomposition of CO2 using pulsed corona discharges combined with 17
catalyst. Environ. Sci. Technol. 21(4), 665-678. Woolf, A., 2004. Challenge and promise: the future of poison control services. Toxicol. 198, 285-289. Yabe, N., 2004. An analysis of CO2 emissions of Japanese industries during the period between 1985 and 1995. Energy Policy. 32, 595–610. Yamaji, K., 1997. A study of the role of end-of-pipe technologies in reducing CO2 Emissions. Waste Manag. 17(5/6), 295-302.
This action might not be possible to undo. Are you sure you want to continue?