Plasma Chemistry and Plasma Processing, Vol. 24, No. 2, June 2004 (g 2004)
NOx Removal Characteristics in Plasma plus Catalyst Hybrid Process Y. H. Lee,1,4 J. W. Chung,2 Y. R. Choi,3 J. S. Chung,1 M. H. Cho,1 and W. Namkung1 Received January 14, 2003; revised July 31, 2003
NOx removal characteristics and NO conversion trends were investigated for plasma process, catalytic process, and plasma catalytic hybrid process. In the experiments, we studied effects of the flow rate and the carrier gas on the NO conversion in the plasma process, and effects of ammonia concentration and temperature on the NOx removal in the catalytic process. We also investigated the synergetic effect of a plasma-catalytic hybrid process. Dielectric barrier discharge was combined with V2O5–WO3=TiO2 catalyst for removing nitrogen oxides. The maximum conversions of nitrogen oxides were approximately 52, 80, and 98% at the temperature of 100, 200, and 300xC, respectively. The optimal energy density, ammonia concentration, and ratio of nitrogen oxides exist for the highest removal of nitrogen oxides in the plasma catalytic hybrid process. KEY WORDS: Plasma catalytic hybrid process; carrier gas; dielectric barrier discharge; V2O5–WO3=TiO2 catalyst; nitrogen oxides.
1. INTRODUCTION Globally, quantities of nitrogen oxides (NOx) are produced by manmade emissions. Nitrogen oxides are formed when fuel is burned at a high temperature, as in combustion processes. The primary sources of NOx are motor vehicles, electric utilities, and other industrial, commercial, and residential sources that burn fuels. It is highly toxic, causing serious lung damages with delayed effects. It also plays a major role in the atmospheric reactions that produce ground-level ozone or smog. Many countries have 1
School of Environmental Science and Engineering, Pohang Univerity of Science and Technology, Pohang, Kyungbuk 790-784, Republic of Korea. 2 Department of Environmental Engineering, Jinju National University, 150 Chilamdong, Jinju, Kyungnam 660-758, Republic of Korea. 3 Pohang Accelerator Laboratory, San-31 Hyoja-dong, Pohang, Kyungbuk 709-784, Republic of Korea. 4 To whom correspondence should be addressed. Telephone: + 82-54-279-2758; fax: + 82-54279-3099; e-mail:
[email protected] 137 0272-4324=04=0600-0137=0 g 2004 Plenum Publishing Corporation
Lee et al.
138
regulated emissions of NOx to very low levels, and the development of the NOx control technologies becomes an important issue. NOx control can be achieved by modification of operating conditions and tail-end control equipment. Recently, there are intensive investigations on the application of the popular non-thermal plasmas combined with a catalytic process for the treatment of NOx.(1–7) Non-thermal plasma gives an efficient means for oxidation of NO to NO2. Pre-converting NO to NO2 opens the opportunity for a wider range of the selective catalytic reduction (SCR) catalysts and improves the durability of these catalysts.(8) Dielectric barrier discharge (DBD) is one of non-thermal plasma techniques, and it offers the advantage to excite molecules for reaction processes on a low temperature level in the near-atmospheric pressure range. NOx is reduced at low temperature when NO2 is involved in the reaction mechanism. Therefore, if non-thermal plasma converts NO to NO2, the reduction rate of NOx will be greatly enhanced in the catalytic process even at a low temperature. In the experiments, dielectric barrier discharge process was combined with V2O5 catalyst supported on TiO2 with additives including WO3 for removal of NOx. In the DBD-ammonia SCR catalytic hybrid process, effects of flow rate, temperature, NO=NO2 ratio, ammonia concentration and energy density were investigated. 2. EXPERIMENTS 2.1. Plasma Reactor and Catalytic Process Figure 1 shows the plasma reactor used in the experiments. The reactor geometry was a concentric cylinder. The diameter of discharge electrode was 40 mm. The stainless steel discharge electrode has a smooth surface. Cylindrical dielectric barriers made of pyrex was wrapped with a copper film tape serving as a grounded electrode. Generally there are two kinds of DBD reactors. One is a single dielectric barrier discharge reactor that uses only one dielectric material around the inner or outer electrodes. The other is a double DBD reactor that uses a dielectric material around both of electrodes. We
Fig. 1. Dielectric barrier discharge reactor.
NOx Removal Characteristics
139
used a single DBD type reactor. The length of the grounded copper film tape was 400 mm. The inner diameter and thickness of the pyrex tube were 44 and 3 mm, respectively. Both ends of pyrex tube were blocked with two ceramic plates. Ammonia SCR catalyst was combined with a plasma process. The catalytic process consisted of V2O5–WO3=TiO2 catalyst with a volume of 110 cm3. The catalyst was placed at the downstream separated from the discharge region. NH3 was injected as an additive over the catalyst in a flow reactor system. 2.2. Experimental Setup The experimental setup is schematically shown in Fig. 2. In order to calculate energy dissipation in the reactor, the applied voltage and charge were measured. The high voltage applied to the reactor was measured by a Tektronix P6015A high voltage probe. The charge was determined by measuring the voltage across 1 mF capacitor with a Tektronix P5100 voltage probe. The dissipated power was determined by calculating the area of charge–voltage plot. Figure 3 shows the typical charge and voltage plot. In charge–voltage Lissajous figure, the area inside the parallelogram is equal to the energy dissipated during one cycle, and each slope of parallelogram gives specific information of dielectric barrier discharge.(9) Gas enters
Fig. 2. The experimental setup.
Lee et al.
140
Fig. 3. Charge–voltage Lissajous figure.
into the inlet of the DBD reactor, travels vertically along the discharge region, and the gas exited from the outlet of the reactor passed through the catalyst. In the DBD-ammonia SCR catalytic hybrid process, ammonia is injected into the outlet of the DBD reactor. A Thermo-Electron chemiluminescent NOx analyzer (model 42H) was used to measure the concentration of nitrogen oxides in the gas stream entering and exiting from the DBD reactor and catalyst zone. 2.3. Experimental Conditions Every experiment was carried out at atmospheric pressure with an inlet gas mixture of NO and NO2 in the air or argon. Initial concentration of NO was varied from 200 to 400 ppm, and that of NO2 was changed from 60 to 100 ppm. The temperatures of catalyst were changed from the room temperature to 300xC. AC voltage from 0 to 20 kV was applied to the DBD reactor, and the discharge frequency was set at 60 Hz throughout the experiments. In the DBD-ammonia SCR catalyst hybrid process, the concentration of ammonia injected into the catalyst was varied from 150 to 500 ppm. 3. RESULTS AND DISCUSSIONS 3.1. NOx Conversion Characteristics in DBD Process Figure 4 shows the conversion characteristics of NOx in the DBD reactor. They are plotted as a function of electrical energy density (J=L),
NOx Removal Characteristics
141
Fig. 4. The conversion characteristics of NOx in DBD reactor (NOx initial concentration of 466 ppm).
which is the electrical power delivered to the plasma reactor divided by the gas flow rate. This experiment was carried out at room temperature, and the gas flow rate was 10 L=min. Initial concentrations of NO and NO2 were 400 and 66 ppm, respectively. As shown in Fig. 4, the concentration of NO decreased with increasing energy density. As the concentration of NO decreased, that of NO2 increased. The total concentration of NOx, the sum of NO and NO2, was not changed significantly, which means most of NO was converted to NO2 in the plasma process. Approximately 50% of NO was converted to NO2 at the energy density of 50 J=L. In the plasma, oxidation is the dominant process for gases containing NO in the mixture of air (N2 and O2). The kinetic energy of electron is deposited primarily to nitrogen and oxygen, and N and O radicals are produced through electron impact dissociation as shown in reactions (1) and (2). In these reactions, N(4S), N(2D), O(3P) and O(1D) are the ground states and the electronically excited states of nitrogen and oxygen atoms, respectively. The dissociation energy of O2 is smaller than that of N2. So, at the lower average electron kinetic energy, around 3–6 eV,(10–12) the dissociation rate of O2 is much higher than that of N2.(13,14) The dissociation of O2 produces only oxidative radicals via reaction (1), and the ground state oxygen atom and ozone will convert NO to NO2 as shown in reactions (3)–(5). At a higher electron kinetic energy, N(4S) and N(2D) are produced by electron impact as shown in reaction (2).(15,16) N(2D) species can lead to produce NO in the presence of O2 via reaction (6).
Lee et al.
142
e + O2 fi e + O(3 P) + O(3 P, 1 D)
(1)
e + N2 fi e + N(4 S) + N(4 S, 2 D)
(2)
O(3 P) + NO + M fi NO2 + M
(3)
O(3 P) + O2 + M fi O3 + M
(4)
O3 + NO fi NO2 + O2
(5)
N(2 D) + O2 fi NO + O
(6)
The experimental results shown in Fig. 5 are well explained by reactions (1)–(6). Initial concentration of NOx in the mixture of air was 204 ppm in the experiment. The concentration of NO decreased rapidly as the energy density increased. As the concentration of NO decreased, that of NO2 increased in the low energy density regime. The total concentration of NOx was not changed significantly similar to the result as shown in Fig. 4. The concentration of NO dropped rather fast up to y120 J=L, and exhibits saturation till the NO concentration slowly rises again at the energy input of 260 J=L or more. It can be explained by reaction (6). It is believed that the increase of NO concentration beyond y300 J=L is caused by the background air. When air without NOx passes through the silent discharge reactor, NO2
Fig. 5. Typical NO conversion trend in air (NOx initial concentration of 204 ppm).
NOx Removal Characteristics
143
Fig. 6. The trend of NO2 production when air passed through plasma reactor (Blank test).
is produced during the discharge. Figure 6 shows the trend of NO2 production when air only passed through the DBD reactor. The concentration of NO2 increases very slowly as energy density increases. 3.2. Effects of Carrier Gas on NO Conversion in DBD Process As one can see in reactions (1)–(6), oxygen in the air mainly contributes to the conversion of NO to NO2. In the experiments, we have investigated the NO conversion trend without oxygen. To exclude the effect of oxygen, inert argon was used as a carrier gas instead of air. Figure 7 shows the comparing results of NO conversion trend between air and argon gas. The initial concentration of NOx was 300 ppm in the experiment. When NOx is mixed with air, most of NO was converted to NO2 as represented by squares in the figure. On the other hand, when argon is used as a carrier gas, the concentration of NO decreased without being converted to NO2 marked as circles. The increase of NO2 concentration was negligible in the plasma process. When the byproducts were measured with FTIR (Fourier transform infrared spectrometer), only a small NO2 peak was observed. We also measured the concentration of NO2 with a chemiluminescent NOx analyzer, and a few ppm of NO2 was detected. It is believed that most of NO converted to N2O, N2, O2, and radicals when argon is used as a carrier gas.
Lee et al.
144
Fig. 7. Effects of carrier gas on NO conversion.
3.3. Effects of Residence Time on NO Conversion in DBD Process As already mentioned, the goal of DBD process is to convert NO to NO2. We investigated the effect of residence time on NO conversion by changing flow rate. Figure 8 shows the effect of the residence time on the NO conversion in the dielectric barrier discharge reactor. The flow rate was changed from 5 to 25 L=min, which corresponds to the residence time of 0.43–2.2 s. The initial concentration of NO was 400 ppm. As shown in Fig. 8, the flow rate up to 25 L=min did not affect the NO conversion in the DBD reactor. The maximum flow of 25 L=min corresponds to the residence time of 0.43 s in the DBD reactor, and this is the current limitation in our experimental setup. We plan to upgrade our setup to investigate the existence of optimum condition. 3.4. Selective Catalytic Reduction of Nitric Oxide with Ammonia over a V2O5 /WO3 /TiO2 Catalyst In the experiments, we used V2O5 catalyst supported on TiO2 with additives including WO3. In ammonia SCR catalyst, reaction mechanisms are as follows: 4NO + 4NH3 + O2 fi 4N2 + 6H2 O
(7)
6NO2 + 8NH3 fi 7N2 + 12H2 O
(8)
NOx Removal Characteristics
145
Fig. 8. The effect of residence time on NO conversion in DBD reactor.
2NO2 + 2NH3 fi N2 + H2 O + NH4 NO3
(9)
NO + NO2 + 2NH3 fi 2N2 + 3H2 O
(10)
At a high temperature over 300xC, the temperature range of the full catalytic activity for the NO reduction, reaction (7) is dominant.(17) At a temperature below 200xC, the reduction of NO2 and the formation of ammonium nitrate become more important, that is to say, reactions (8) and (9) are dominant.(18) These two reactions would only explain the NOx removal of 50%. Thus an additional reaction must be considered,(19) which is explained by reaction (10). As one may see in reactions (8)–(10), NOx is reduced at low temperature when NO2 is involved in the reaction mechanism. Therefore, if NO is converted to NO2 in the DBD reactor, the reduction rate of NOx will be greatly enhanced in the catalytic process even at low temperature. We have tried to find out the optimum condition of catalytic process for the comparison of plasma plus catalyst hybrid process with catalytic process. To find the optimum condition, the initial concentrations of NO and NO2 were kept constant, and then the concentration of NH3 was increased slowly. Figure 9 shows the effects of NH3 on NOx conversion in NH3-SCR catalyst. As already described, the concentration of ammonia injected into the catalyst was varied from 150 to 500 ppm. This experiment was carried out at the temperature from room temperature to 300xC, and the gas flow rate was
146
Lee et al.
Fig. 9. The effect of ammonia concentration and flow rate on NOx removal in NH3-SCR catalyst.
varied from 10 to 15 L=min. The initial concentration of NOx was varied in the range 470–500 ppm. There was an optimum NH3 concentration for the highest removal of NOx at the temperature of 300xC. At the flow rate of 10 L=min, the concentration of NO and NO2 decreased as the injected ammonia concentration increased up to 300 ppm. However, even though the concentration of NO was not changed, the concentration of NO2 started to increase at the ammonia concentration of 300 ppm and above. As a result, the concentration of NOx (the sum of NO and NO2) increased. The result could be explained by ammonia slip. In the experiment, a chemiluminescent NOx analyzer was used to measure the concentration of nitrogen oxides. A chemiluminescent NOx analyzer reads the partial concentration of NH3 as that of NO2. Thus the increase of NOx at the high concentration of NH3 was caused by NH3 that does not react with NOx. The optimal concentrations of ammonia were approximately 300 and 350 ppm at the flow rate of 10 and 14.7 L=min, respectively. The removal rates of NOx were 94 and 84% at those flow rates. At the temperature of 200xC, the optimal concentrations of ammonia may exist in the range 100–200 ppm. Because of the limitation of mass flow controller, the concentration of ammonia was not controlled below 150 and 200 ppm at the flow rate of 14.7 and 10 L=min, respectively. The maximum NOx removal rates were 32 and 30% at the flow rates of 10 and 14.7 L=min, respectively.
NOx Removal Characteristics
147
Figure 9 also shows the effect of flow rate on NOx removal in NH3-SCR catalyst. As the flow rate increased, the removal rate of NOx decreased. NH3-SCR catalyst was not well activated at the temperature below 100xC. As a result, the removal rate of NOx was very low. 3.5. Removal Characteristics of NOx in the Plasma Process Combined with NH3-SCR Catalyst Figure 10 shows the comparing results of NO removal between the plasma catalytic hybrid process and catalytic process only. Initial concentrations of NO, NO2, and NOx were 400, 66, and 466 ppm, respectively. V2O5–WO3=TiO2 catalyst was placed downstream from the region in which the discharge occurs: 240 ppm of NH3 was injected into the inlet of catalytic process. It is known that V2O5–WO3=TiO2 catalyst is well activated over 300xC. During the experiments, catalyst temperature was maintained at 200xC. Therefore, the removal rate of NOx was approximately 30% when NO passed through the catalyst only, as represented by a dotted line in Fig. 10. On the other hand, the concentration of NOx was dramatically dropped when NO passed through the DBD reactor and catalyst at the same time. As already described, the DBD converts NO to NO2 at a low energy density. It means that the total concentration of NOx does not change. However, the concentration of NO and NO2 simultaneously decreased in the
Fig. 10. The conversion characteristics of NOx in the plasma catalytic hybrid process.
Lee et al.
148
plasma catalytic hybrid process. Approximately 80% of NOx was removed at the energy density of 40 J=L. As one may see in Fig. 10, there was an optimum energy density for reduction of NOx in the plasma catalytic hybrid process. At a low energy density, the concentration of NOx decreased with increasing energy density. The concentration of NOx dramatically dropped until 30 J=L, and then, that of NOx started to increase at a high energy density of 50 J=L and above. In our experimental conditions, the optimum energy density for the maximum NO conversion existed in the region of 40–50 J=L. The result is partially caused by the ratio of NO to NO2 and the conversion of NO to NO2. After the mixing gas passed through the DBD reactor only, the ratio of NO to NO2 was approximately 1: 1 at the energy density of 50 J=L. It was reported that the ratio of NO to NO2 could affect the removal rate of NOx.(20) Also, the conversion of NO to NO2 enhances NOx removal in NH3-SCR catalytic process as shown in reactions (8)–(10). The effect of NO=NO2 ratio is given in the last section of the paper. 3.5.1. Effects of Temperature and Energy Density on NOx Removal in the Plasma Catalytic Hybrid Process We have investigated the characteristics of NOx removal in the plasma catalytic hybrid process with changing temperature and energy density. The temperatures of catalyst were varied in the range from 100 to 300xC, and the concentration of ammonia injected into the catalyst was changed from 200 to 300 ppm. The gas flow rate was 10 L=min. The concentration of NOx was measured at the inlet and outlet of the catalyst. As shown in Table I, the concentration of NOx decreased when the mixing gas passed through the catalyst without plasma. However, the concentration of NOx was not changed without ammonia injection into the catalytic process, as shown in Table I. That is to say, NH3-SCR catalyst did not work without NH3 injection. The removal rate was highly affected by the temperature of catalytic process.
Table I. The Comparison of NOx Concentrations Before and After Catalytic Process After catalyst Before catalyst
Without NH3 injection
With NH3 injection
Temp.
NO (ppm)
NO2 (ppm)
NO (ppm)
NO2 (ppm)
NO (ppm)
NO2 (ppm)
100xC 200xC 300xC
396 400 398
77 66 63
396 400 398
77 66 63
369 219 26
56 114 5
NOx Removal Characteristics
149
Fig. 11. Effect of temperature and energy density on NOx conversion in the plasma catalytic hybrid process.
Figure 11 shows the effect of catalyst temperature and energy density on the NOx conversion in the plasma catalytic hybrid process. The NOx concentrations were measured at the outlet of catalytic process after ammonia was injected into the catalyst. That is to say, the concentration at the energy density of 0 J=L is the concentration measured after the catalytic treatment only in Fig. 11. At the temperature of 100xC and ammonia concentrations of 200 ppm, V2O5=WO3=TiO2 catalyst is not well activated. Generally, the typical operating temperatures of vanadia=titania catalysts are between 250 and 450xC, and that satisfies many industrial applications. Thus much of ammonia does not react with NOx at 100xC and emissions of ammonia reagent increases. Ammonia slip must be limited due to downstream impacts associated with corrosion and fouling. In the experiments, we tried to find out the optimum concentration of ammonia for minimizing NH3 slip. The concentration of ammonia was varied from 200 to 300 ppm at 100xC. As the concentration of ammonia decreased, NH3 slip also decreased in the experimental range. Due to the limitation of the mass flow controller, the concentration of ammonia cannot be reduced below 200 ppm. Further investigations are required to find the optimum concentration of ammonia for minimizing NH3 slip at the temperature of 100xC. The initial concentration of NOx was 473 ppm, and the concentration of NOx measured after catalytic treatment was 425 ppm as shown in Table I. The removal efficiency of NOx was approximately 10% in
150
Lee et al.
the catalytic process only. In the plasma catalytic hybrid process, the removal of NOx increased as the energy density at the catalyst temperature of 100xC and NH3 concentration of 200 ppm, as shown in Fig. 11. At the temperature of 100xC, the optimum energy density and ammonia concentration were 67 J=L and 200 ppm, respectively. The removal rate of NOx was 52% at those experimental conditions. Compared with the catalytic process, the removal efficiency of NOx increased more than 40% in the plasma catalytic hybrid process. At the temperature of 200xC, the initial concentration of NOx was 466 ppm, and the concentration of NOx measured after catalytic treatment was 333 ppm as shown in Table I. Approximately 30% of NOx was removed without discharge. There was also an optimum energy density for the highest removal of NOx at the temperature of 200xC. At the ammonia concentration of 240 ppm, the optimum energy density for the highest removal of NOx was existed in the range between 40 and 50 J=L. In the plasma catalytic hybrid process, the removal rate of NOx was approximately 80% at those experimental conditions. At the temperature of 300xC, the removal of NOx was insensitive to the energy density. The initial concentration of NOx was 461 ppm, and the concentration of NOx measured after catalytic treatment was 31 ppm as shown in Table I. Without discharge, the removal rate of NOx was 93% at the ammonia concentration of 300 ppm. As described earlier, V2O5–WO3= TiO2 catalyst is well activated at the temperature of 300xC. Therefore, if a proper amount of ammonia is injected into the catalyst, NOx is reduced to nitrogen and water with high efficiency. When excess ammonia (over 300 ppm) was injected to ensure a high NOx reduction, a small amount of ammonia slip enters the exhaust stream, and it is emitted. As ammonia itself is a pollutant, the SCR unit is normally set to minimize the amount of ammonia slip emitted. In the experiment, the optimum concentration of NH3 was approximately 300 ppm for the highest removal of NOx in the catalytic process. Even though NOx removal was not much affected by the energy density at the temperature of 300xC, the concentration of NOx dropped a little as the energy density increased. The optimum energy density for the highest NOx removal also existed at the temperature of 300xC. When the ammonia concentration of 300 ppm was injected into the plasma-catalytic hybrid process, 98% of NOx was removed at the energy density of 69 J=L. 3.5.2. The Effect of NO=NO2 Ratio on NOx Removal Rate in the Catalytic Process As shown in Figs. 4 and 5, the ratio of NO to NO2 is changed as energy density in the plasma process, and the change of NO=NO2 ratio in the
NOx Removal Characteristics
151
plasma process could affect the removal rate of NOx in the catalytic process. In the experiments, we have investigated that the effect of NO=NO2 concentration ratio on removal efficiency in the catalytic process with changing the ratio of NO to NO2, and the experimental result was compared with those obtained from plasma catalytic hybrid process. The temperatures of catalyst were varied in the range from 100 to 200xC, and the concentration of ammonia injected into the catalyst was changed from 200 to 240 ppm. Gas flow rate was 10 L=min and the initial concentration of NOx was approximately 500 ppm. The experiments were done with changing the ratio of NO to NO2. The ratios of NO to NO2 were varied from 3.4 :1 to 1: 4. Figure 12a shows the effect of NO=NO2 concentration ratio (366 : 136, 300: 201, 247 :248, 196 :301 and 98:397) on removal efficiency at the temperature of 100xC. In Fig. 12, parenthesized numbers mean the ratio of NO to NO2 injected into V2O5=WO3=TiO2 catalyst. 200 ppm of NH3 was injected into the catalyst, and the flow rate was 10 L=min. As shown in Fig. 12a, NOx removal efficiency increased as the ratio of NO2 increased, and it saturated after the ratio of 1:1. The optimum NO=NO2 ratio for the highest efficiency was 2 :3, and the removal efficiency of NOx was 34.6% in the catalytic process. This removal trend is similar to that obtained in the plasma catalytic process (Fig. 11). In the plasma catalytic hybrid process, the optimum energy density for the highest NOx removal was 67 J=L at the temperature of 100xC. At the energy density of 67 J=L, the ratio of NO to NO2 was approximately 2 :3 in the plasma process. The removal rate was 52% in the hybrid process as shown in Fig. 11. The difference between 34.6 and 52% is thought to be a synergetic effect of plasma catalytic process at the temperature of 100xC. Figure 12b shows the effect of NO=NO2 concentration ratio (385 :112, 300: 203, 252 :249, 202: 299 and 100 :401) on the removal efficiency at the temperature of 200xC. 240 ppm of NH3 was injected into the catalyst, and the gas flow rate was 10 L=min. The optimum ratio of NO to NO2 was 1: 1, and the highest removal efficiency of 48.1% was obtained at the ratio of 1: 1. The removal trend at 200xC is not so similar to that obtained in the plasma catalytic hybrid process. However, in the plasma catalytic hybrid process, the optimum energy density for the highest NOx removal was 50 J=L, and the ratio of NO to NO2 in the plasma reactor was approximately 1:1 at the energy density of 50 J=L. The removal rate was approximately 80% in the plasma catalytic hybrid process as shown in Fig. 10. The difference between 48.1 and 80% is also thought to be a synergetic effect of plasma catalytic process at the temperature of 200xC. As shown in Fig. 12, the ratio of NO to NO2 affects the removal efficiency in the catalytic process, and NO=NO2 concentration ratio is also related to the removal efficiency in the plasma catalytic process. So, if we find
Lee et al.
152
Fig. 12. Effect of NO=NO2 concentration ratio on NOx removal efficiency.
the optimum ratio of NO to NO2 in the plasma reactor with changing energy density, the removal efficiency of NOx can be greatly enhanced in the plasma catalytic hybrid process. In this paper, we considered the effect of NO=NO2 ratio on the removal efficiency of NOx. However, many excited species are produced in the plasma process, and they may also contribute to enhance the efficiency of catalyst.
NOx Removal Characteristics
153
Therefore, we plan to investigate the effect of excited species on the removal efficiency of NOx in the hybrid process. 4. CONCLUSIONS In this experiment, we have shown characteristics for the plasma process, the catalytic process, and the synergetic effect of the plasma-assisted catalytic process in reducing NOx. We found that the ammonia concentration, energy density, and temperature should be considered for obtaining the maximum NOx removal in plasma catalytic hybrid process. The experiments provide us following conclusions. 1. In the range of 5–25 L=min, the flow rate did not affect NO conversion in the dielectric barrier discharge reactor. The NO conversion strongly defends on the energy density. 2. Oxygen in the air mainly contributes to the conversion of NO to NO2 in the DBD process. 3. There was an optimum ammonia concentration for the highest reduction of NOx with ammonia over V2O5=WO3=TiO2 catalyst. 4. There was an optimum energy density for obtaining the maximum NO removal in the plasma catalytic hybrid process. 5. The ratio of NO to NO2 affected the NOx removal rate in the catalytic process and plasma catalytic hybrid process. 6. The synergetic effect was observed in the plasma-assisted catalytic process for reducing NOx. ACKNOWLEDGMENTS This work is partially supported by Envichem Co. Ltd., Environmental Plasma Co. Ltd., and Korea Institute of Environmental Science and Technology (KIEST).
REFERENCES 1. T. Oda, T. Kato, T. Takahashi, and K. Shimizu, J. Electrostat. 42, 151–157 (1997). 2. H. Miessner, K.-P. Francke, R. Rudolph, and Th. Hammer, Catal. Today 75, 325–330 (2002). 3. S. Yoon, A. G. Panov, R. G. Tonkyn, A. C. Ebeling, S. E. Barlow, and M. L. Balmer, Catal. Today 72, 243–250 (2002). 4. S. Yoon, A. G. Panov, R. G. Tonkyn, A. C. Ebeling, S. E. Barlow, and M. L. Balmer, Catal. Today 72, 251–257 (2002). 5. H. Miessner, K.-P. Francke, and R. Rudolph, Appl. Catal. B 36, 53–62 (2002). 6. B. S. Rajanikanth and S. Rout, Fuel Process. Technol. 74, 177–195 (2001).
154
Lee et al.
7. S. Bro¨er and T. Hammer, Appl. Catal. B 28, 101–111 (2000). 8. B. M. Penetrante, R. M. Brusasco, B. T. Merritt, W. J. Pitz, and G. E. Vogtlin, in Proc. Int. Fall Fuels and Lubricants Meeting (Society of Automotive Engineers, SP-1395), 1998, pp. 57–66. 9. U. Kogelschatz and B. Elliasson, in Handbook of Electrostatic Processes (J. S. Chang, A. J. Kelly, and J. M. Crowley, eds.), Marcel Dekker, New York, 1995, pp. 581–588. 10. B. M. Penetrante, M. C. Hsiao, B. T. Merritt, G. E. Vogtlin, P. H. Wallman, M. Neiger, O. Wolf, T. Hammer, and S. Broer, Appl. Phys. Lett. 68, 3719–3721 (1996). 11. B. M. Penetrante, A. M. C. Hsiao, B. T. Merritt, G. E. Vogtlin, P. H. Wallman, A. Kuthi, C. P. Burkhart, and J. R. Bayless, Appl. Phys. Lett. 67, 3096–3098 (1995). 12. B. M. Penetrante, M. C. Hsiao, B. T. Merritt, G. E. Vogtlin, and P. H. Wallman, IEEE Trans. Plasma Sci. 23, 679–687 (1995). 13. B. M. Penetrante, J. N. Bardsley, and M. C. Hsiao, Jap. J. Appl. Phys. 36, 5007–5017 (1997). 14. B. M. Penetrante and S. E. Schultheis (eds.), Plasma Chemistry and Power Consumption in Non-thermal DeNOx, Non-thermal Plasma Techniques for Pollution Control, Part A— Overview, Fundamentals and Supporting Technologies, Springer-Verlag, Berlin, 1993, pp. 65–89. 15. E. C. Zipf, P. J. Espy, and C. F. Boyle, J. Geophys. Res. 85, 687 (1980). 16. P. C. Cosby, J. Chem. Phys. 98, 9544–9553 (1993). 17. V. Tufano and M. Turco, Appl. Catal. B 2, 9–26 (1993). 18. C. U. I. Odenbrand, L. A. H. Andersson, J. G. M. Brandin, and S. T. Lundin, Appl. Catal. 27, 363–377 (1986). 19. Suggested by E. Jacobs, MAN Nutzfahrzeuge AG, Nurrnberg (1997). 20. T. Hammer and S. Broer (Siemens AG), in Proc. Int. Fall Fuels and Lubricants Meeting (Society of Automotive Engineers, SP-1395), 1998, pp. 7–12.
