Revision

Hydrogen production from methane in a dielectric barrier discharge using oxide zinc and chromium as catalyst Shortened title: H2 production from CH4 in a DBD with Zn-Cr oxide catalyst

Antonius Indarto1,*

1

Clean Technology Research Center, Korea Institute of Science and Technology, PO Box 131, Cheongryang, Seoul, Korea

1

Address: Plasma-Catalyst Process Laboratory, Korea Institute of Science & Technology, P.O. Box 131, Cheongryang, Seoul 130-650, Korea Phone:

+82-10-2296-3748

Email:

[email protected]

2

Abstract- The hydrogen fuel cell is a promising option as a future energy resource; however, the nature of the gas is such that the conversion process of other fuels to hydrogen on board is necessary. Among the raw fuel resources, methane could be the best candidate as it is plentiful. In this experiment, the possibility of producing hydrogen with less carbon formation from methane by a dielectric barrier discharge (DBD) was investigated. Without the addition of a catalyst, the formation of hydrogen reached between 30% and 35% at methane residence time of 0.22 min and supplied powers in the range of 60W to 130W. The hydrogen selectivity increased at higher supplied power, but the process efficiency, defined as a ratio of the produced hydrogen to the supplied power, decreased slightly. In order to boost the hydrogen production with less carbon formation, a mixed oxide catalyst of zinc and chromium was added to the reactor. It was shown that the production of hydrogen was ca. 40% higher than the non catalytic plasma process. Keywords: Plasma, dielectric barrier discharge, methane, hydrogen production

3

1. Introduction The development of fuel-efficient engines that produce fewer pollutants has been a major objective for many years. Regulatory requirements have been and will be of major importance in defining the standards (Cooper, 1994). Increasingly stringent legislation directs attention at factors that reduce polluting emissions, such as a cold-start process and lowering sulfur levels in fuel (Matsumoto, 2000). Among many selective compound candidates, hydrogen and methanol have been considered as green fuel starting materials. Compared to a hydrogen-based process, the kinetics of methanol conversion was relatively slower and catalyst deactivation occurred simultaneously. As a result, hydrogen is preferred (Cameron, 1999; Ralph and Hards, 1998; Hoogers and Thompsett, 1999), but presents obvious problems with generation, storage, and distribution of the gas. Pressurized vessels or metal hydrides of hydrogen in a vehicle are relatively dangerous because hydrogen is categorized as a highly flammable material. Currently, attention has been focused on the design and operation of compact and efficient devices designed to generate hydrogen on board a vehicle. Jamal and Wyszynski (1994) reviewed many possible onboard hydrogen generators and the alternative fuels for spark ignition engines. Hydrogen production from methane (CH4) and methanol has attracted the consideration of many experts. Although methanol has more advantages than methane, e.g. easier storage and safety, along with the byproduct of carbon monoxide being annoying for both environmental and human health (Velu et al., 1999; Liu et al., 2004). On the other hand, nowadays, methane for fuel is gaining popularity and widespread use, especially in the transportation sector. In order to produce hydrogen from methane, a methane cracking process is necessary. However, the conventional thermal method requires a high temperature condition to achieve high methane conversion (Indarto et al., 2006a) and can be very costly for the onboard process. In particular, decomposition of methane using plasma could be a candidate for this purpose as the existence of high-energy electrons were able to decompose methane conversion at lower temperature. Yao et al. (2001) investigated the methane conversion using non-thermal plasma and showed that the methane conversion could be dramatically improved through this method. As high hydrogen production is the goal of the process, process modification, e.g. by the addition of a catalyst, is necessary to change the nature of the product distribution. In thermal-based processes, the catalytic conversion of methane to hydrogen is usually carried out by the catalyst of Ni or Pt (Peña et al., 1996; Twigg, 1989; Rostrup-Nielsen, 1984). Our previous research using Pt/γ-Al2O3 catalyst to convert CH4 in a DBD showed a negative effect on the production of hydrogen, as higher hydrocarbon transformation was the more dominant phenomena (Kim et al., 4

2004). Better results employing Ru and Rh catalysts were obtained in a catalytic thermal process (Rostrup-Nielsen and Hanses, 1993). That result was supported by our investigation that Ru and Rh were also active metal catalysts for converting methane to hydrogen in a non-thermal plasma (Indarto et. al., 2007), but, as the price in the market is relatively expensive, this material has been rarely used in the industry. Bridger et al. (1970) developed a relatively cheaper mixture catalyst based on Zn and Cr oxide metal for hydrogen production. Using a mixture of CH4 and CO2 as the reactants, the production of hydrogen yielded 35% at a relatively low temperature (ca. 100oC). Davis et al. (1976) thoroughly investigated the activity of Zn catalyst for H2 production. Recent experiments by Liu et al. (2004) showed that the existence of ZnO-based catalyst increased the production of hydrogen from methanol. The activity of ZnO-Cr2O3 catalyst at low temperature condition has been investigated by Ohta et al. (2004) for the dehydrogenation of isobutane. However, no publication was found for the use of mixed oxide of Zn-Cr catalyst for methane conversion to produce hydrogen both by thermal and plasma methods. In this research, the direct methane conversion to hydrogen was investigated using a dielectric barrier discharge (DBD). DBD is a widely used plasma technique for many applications because it is an easily installed and low-cost of operation instrument. In accordance with the above discussion, a catalyst based on Zn and Cr oxide metal was made and employed in order to boost the production of hydrogen. In order to increase the lifetime of the catalyst, the research would like to suppress the production of carbon. The deposition of carbon by covering the surface of catalyst is a major reason of the catalyst deactivation (Venugopal et al., 2007). 2. Experimental setup Figure 1

The schematic diagram of the experimental setup is shown in Fig. 1. The reactor was a quartz tube with an inside diameter of 6 mm and length of 20 cm. A thin silver film, serving as the outer electrode, coated the outer wall of the reactor. The inner electrode was two stainless wires (∅ = 0.2 mm) located in the center of the reactor. The plasma was generated by a high-voltage alternating current (AC) generator (Auto electric, model A1831) that has maximum voltage of 10.0 kV and maximum frequency of 20 kHz. In this experiment, the power frequency was maintained fixed at 20 kHz, while the output supplied power to the reactor was manipulated by varying the voltages and currents. The measurement of applied power was done by a watt-meter (Metex, model M-3860M). All experiments were carried out under atmospheric pressure and ambient 5

temperature. The flow rates of methane gas were controlled by a mass flow controller. The output of the reactor was connected to a gas chromatograph (YoungLin model M600D, column: 30 ft of Hayesep D) equipped with two detectors (thermal conductivity and flame ionized detector). Considering the volume expansion and/or compression of the product stream, the flow difference between the input and output line was measured by a bubble flow meter (Hewlett Packard) referring to the condition before and after plasma reaction. The formulation of methane conversion and product selectivity follows: moles of CH 4 converted Conversion of CH 4 = × 100 (1) moles of initial CH 4 Selectivity of C m H n =

n moles of C m H n produced × × 100 2 moles of CH 4 converted

(2)

The atom balance of carbon and hydrogen was calculated in each experiment to gain the quantitative satisfaction detection of products by our analysis instruments. In order to increase the production of hydrogen, a mixed oxide catalyst of zinc (Zn) and chromium (Cr) was added on the plasma system. The catalyst was made by liquid mixing of Zn(NO3)2.6H2O and Cr(NO3)3.9H2O precursor with distillated water. The detailed procedure of the catalyst preparation was described clearly in the patent paper of Bridger et al. (1970). Before it was used, the dried gel catalyst was crushed and sieved until it reached uniform size of ca. 60-90 mesh. The fresh catalyst was then calcined at 300oC for 2hr in atmospheric air. The XRD spectrum of the catalyst after the calcinations is presented in Fig. 2, which shows the dominant ZnO and ZnCr2O4 crystal phase. Simard et al. (1995) mentioned that those two phases were more active compared to single ZnC2O4 or Cr2O3 for the transformation of synthetic gas. Those crystal phases could be active for the case of H2 production from methane. Only 0.5 gram of catalyst was added in the reactor and put ~1.5-2 cm before the end of the plasma zone in order to avoid the product’s decomposition by plasma. In this experiment, the catalyst treatments, e.g. calcinations and regeneration, were used as the variable to investigate the activity of the catalyst and its durability. 3. Results and discussion 3.1. Non-catalytic plasma reaction The conversion of methane using non-thermal plasma devices, especially dielectric barrier discharge (DBD), has been commonly used. Numerous reports and papers have been published with differing results and conclusions as many aspects and variables exist in a DBD (Yao et al., 2001; Yang, 2003; Kim et al., 2004; Indarto et al., 2005; 2006a; 2006b). Although plasma reaction is known as one of the most difficult reactions to determine the kinetics, the tendency to follow a certain pattern 6

could be studied (Indarto et al., 2007a). The pathway pattern can easily change under different conditions as the reactions are very sensitive to the population of active species, ions, and electrons. This is the reason the optimization of the parameters is necessary to obtain the optimum conditions for the best results. Figure 3

Supplied power from an electrical generator to the reactor is one of the most important parameters in the plasma chemical reaction to activate or decompose the methane molecule. The magnitude of the external supplied power is related to the intensity of the internal electric field inside the reactor. An increase of power could produce more energy to activate molecules to higher energy levels and result in more energetic electrons. In this condition, the possibility of breaking the C-H bond in the molecule of methane could be initiated. Figure 3 shows the effect of supplied power variation on the methane conversion and product selectivity at the fixed input flow rate of 15 ml/min or residence time of 0.6 min. Methane conversion was raised from 30% to 45% when the power was increased from 60 W to 130 W. It can be deduced, trivially, that supplying more energy will increase the molecules’ instability inside the reactor and cause more conversion of methane. Increasing energy will also increase the density of electron and radical species inside the plasma zone. The gaseous products of non-catalytic plasma of methane conversion were dominated by H2, C2 (mostly C2H6), C3, and small amount of C4H10. Liquid product was not detected in all ranges of experiment, and a small amount of solid carbon was found deposited on the reactor wall and the surface of inner electrodes. The production of hydrogen was relatively stable in the range between 30% and 35%. Figure 3b shows that the maximum production of C3 hydrocarbons occurred at a supplied power of 90 W or at the minimum C4 hydrocarbon production. On the other hand, production of C4 reached the highest value at higher supplied power (130 W) and suppressed the production of C3. Similar to the trend of H2, the production of C2 compounds was not affected much by supplied power variation. A previous paper by Kim et al. (2004) reported the similar situations when methane was treated by a DBD. Figure 4

The effect of the residence time of methane in the reactor was also examined. Figure 4 shows the results on methane conversion (Fig. 4a) and the selectivity of products (Fig. 4b) vs. residence time variation at a fixed supplied power of 80W. Different from the previous parameter, the effect of residence time gives more clear results in terms of product distribution. Methane conversion was significantly increased with an increase in residence time, as shown in Fig. 4a. Longer periods of methane inside the reactor will create more opportunities for methane molecules to 7

collide with other energetic species, e.g. electrons or radicals. Figure 4b shows that the production of H2 was clearly increased when the residence time was increased. As the production of H2 increased, it reduced the selectivity of other products containing an H atom in their molecules. Our previous research on gliding arc plasma also produced the similar results (Indarto et al., 2005; 2006b). At a longer residence time, produced higher hydrocarbon (CxHy) from adduct reactions will be easily decomposed in the presence of a single atom of hydrogen and transform into hydrogen molecules (H2) and other molecules (CxHy-1) by a dehydrogenation reaction. Ab-initio1 calculation of Gibbs energy (ΔGT=298K) shows the products of dehydrogenation reactions are more stable than the reactants, which means that H2 production is preferable. Figure 5 shows the Gibbs energy diagram of dehydrogenation reactions sequence occurring in C2 hydrocarbons. Although, at longer residence times, the selectivity of hydrogen reaches the highest value, the power efficiency to H2 production (calculated as ratio of H2 production to the total supplied power) decreased. In this condition, the process could be said to be not very efficient and rather costly to apply. The maximum selectivity of C2 and C3 hydrocarbons was 35% and 20% respectively, both occurring at a residence time of 0.22 min. Under similar conditions, the production of C4 also reached maximum ~15% total selectivity. At a very short residence time, it showed that the probability of higher hydrocarbon formation was greater and that the dehydrogenation reactions could be avoided. Eliasson et al. (2000) reported that methane chain reaction could occur following: •CH3 + •CH3 → C2H6 (3) C2H6 + e → •C2H5 + •H + e (4) •C2H5 + •C2H5 → C4H10 (5) As all of the above reactions have almost zero reaction energy barriers (~0 kcal/mol), a longer residence time is not necessary to form higher hydrocarbons. Moreover, the longer the higher hydrocarbons stay in the plasma zone, the greater the probability of being attacked by hydrogen or electrons. 3.2. Plasma with oxide of Zn and Cr catalyst Figure 6

The use of a catalyst is necessary in order to control the chemical reactions of plasma and increase the product yield. Some papers have reported some results related to the use of catalysts (usually metal oxide catalyst) for hydrogen production. Most authors preferred Pt, Ru, and Rh catalysts, owing to their good activity and stability. Meanwhile, Ni-based catalysts are commercially more interesting but suffer 1 Ab-inito calculation was done by using Gaussian 03 software (Frisch et al., 2004) with DFT (UB3LYP) method and basis set of 6-31G(d). 8

the disadvantage of a high rate of deactivation. Moreover, exploration to find the best catalyst is still challenging, especially as far as the perspective of reasonable price and good availability. In this experiment, a catalyst based on Zn and Cr oxide will be used combined with plasma to convert methane and produce H2 as the product. Low production of carbon is also necessary to prevent the catalyst from fast deactivation. In order to find the best region where the catalyst was able to show the activity significantly, we did a set of ‘screening’ experiments. As the reaction temperature was relatively low (ca. room temperature), some catalysts did not show any activity (Indarto et al., 2007b). In this case, increasing the supplied power to increase the density of active species could be a way of activating the surface property of the catalyst. At relatively low supplied powers or higher input flow rates, the product distribution was not much different compared to the distribution from the noncatalytic plasma process. The existence of radicals and/or ions from methane fragmentation was low due to the low conversion of methane. The optimal point of hydrogen production occurred at a flow rate of 30 ml/min and power of 80 W. Before the catalyst was used, we treated it with two different activation methods. The first catalyst was calcined at 300oC using atmospheric air for 2 h, noted as 1a, and the second one was not calcined, noted as 3a. After around 2 hours and 40 minutes operation, all catalysts were taken out of the reactor and dried for one night to remove any possible moisture (probably liquid hydrocarbons) attached to the surface of the catalyst. The dried catalysts were then re-used for the similar experiment. They were noted as 1b (for the catalyst coming from 1a) and 3b (for the dried catalyst coming from 3a). However, a half portion of the dried catalyst from 1b was reactivated by calcination at 300oC for 2 h, noted as 2a. After 2a was used in the experiment for around 2 hr and 40 min, the catalyst was then dried for one night and used again for the similar plasma reaction, and noted as 2b. This method is proposed to know the effect of the calcinations to the catalyst activity and to investigate the durability of the catalyst, i.e. whether or not it is re-useable. As shown in Figures 6a-b, the existence of the catalyst, especially the calcined catalyst, increased the conversion of methane. The methane conversion of the catalytic plasma process was higher ~50% than non-catalytic plasma process at the beginning of the operation but decreased slightly as the operating time increased (1a). We found that the catalyst began losing its activity after around 8hr of operation. After that time, the H2 selectivity and methane conversion were similar to the non-catalytic plasma process. Figure 6a shows that the activity the catalyst could not be regenerated by only by drying the catalyst, as the conversion of methane was similar to the noncatalytic plasma process but it could slightly recover the activity of H2 selectivity. 9

This means that the moisture or liquids which possibly existed on the surface were not affecting the catalyst activity or that the moisture or liquid was not present during the process. However, the catalytic activity significantly reappeared when the used catalyst was recalcined. The activity of 2a catalyst was nearly similar to the activity of the 1a catalyst for both methane conversion and H2 selectivity. It is good news that the catalyst can be re-used with almost similar activity by re-calcination. It is necessary to calcine the catalyst to oxidize the adsorbed carbon by oxygen. Moreover, in another point of view, the calcination could be an important factor, as shown by the different trend of 1a (catalyst with calcinations) and 3a (catalyst without calcinations). The activity of uncalcined catalyst was shown to be very low as compared to the calcined catalyst, and this could refer to the crystal phases present in the catalyst. The effect of re-calcination on the activity of converting methane is still being studied further by analyzing the XRD, TEM, and SEM picture. The above phenomenon also occurred for the H2 selectivity. The addition of the catalyst increased the H2 selectivity ~40% for the first time operation of the catalyst. Different from the catalyst activity in the methane conversion, the selectivity of H2 production was relatively stable both for 1a and 2a catalyst. For the 3a catalyst, the H2 selectivity was also stable but in lower values (ca. 30%). After the first hour of operation of the “b” catalyst (both 1b and 2b), the production of H2 was slightly lower than the “a” catalyst (i.e. 1a and 2a), and the production decreased rapidly after 2h operation with the “b” catalyst. The interesting point, similar to the case of methane conversion, is the catalyst activity can be regenerated by calcination. One can conclude that the catalytic plasma process is more efficient as compared to the non-catalytic process for H2 production as the catalytic plasma process produces higher H2 when compared to the non-catalytic process at similar supplied power to the reactor. However, Ohta et al. (2004) and Venugopal et al. (2007) mentioned the problem of the ZnO-Cr2O3 catalyst being easily deactivated by coke deposition. In the plasma process, nano-sized carbon can be formed in nonthermal plasma by methane fragmentation (Indarto, 2006c). Calcination of the catalyst using air could remove the carbon by oxidizing it to form CO or CO2. Other hydrocarbon product selectivity is shown in Fig. 6c. The phenomenon is similar to the case of residence time effect in non-catalytic experiments. When the H2 selectivity rose, it sacrificed the other hydrocarbon products by suppressing their selectivity. It is proposed that the existence of a catalyst will ease the H 2 production reactions, by adsorbing hydrogen produced from dehydrogenation of hydrocarbons (shown in Fig. 5). Although it is still ambiguous, the existence of Cr on the surface of catalyst seemed able to affect the kinetics of the H2 reaction (Kato et al., 2006). This finding also mentioned that Cr could adsorb more hydrogen atoms than Co and Ni and 10

possibly perform H2 formation on the surface of the catalyst. Higher distributed Cr, acting as active site, would increase the production of H2, and ZnO could disperse the Cr during the catalyst formation (Thomas and Thomas, 2006). 4. Conclusions The production of hydrogen from methane by a dielectric barrier discharge with mixed oxide catalyst of zinc and chromium was investigated. Non-catalytic plasma reactions resulted in hydrogen with selectivity of ca. 30-35%. The selectivity of hydrogen increased with increasing the power supplied to the reactor. The addition of calcined mixed Zn-Cr oxide catalyst increased the hydrogen production 40% higher than the non-catalytic reaction. The existence of a catalyst also increased the methane conversion, but the conversion decreased slightly over a longer operation time. Carbon deposition could be proposed as the main reason for catalyst deactivation, but catalyst activity can be regenerated by catalyst recalcination. Acknowledgments The first author thanks to the Korea Institute of Science and Technology (KIST) and the Korea University (KU) for the financial of the study. The first author also would like to express his appreciation to the Università degli studi di Torino for the support during study period in Turin, Italy.

11

References Bitter, J. H., K. Seshan and J. A. Lercher, “Mono and Bifunctional Pathways of CO 2/CH4 Reforming Over Pt and Rh Based Catalyst,” J. Catal., 176, 93 (1998). Bridger, G. W., D. O. Hughes, and P. W. Young, “Process for Producing Carbon Dioxide and Hydrogen from Carbon Monoxide and Steam using Mixed Oxide of Copper, Zinc and Chromium as Catalyst and Process for Producing Said Catalyst,” U.S. Patent, 3, 514, 261 (1970). Cameron, D.S., “The sixth grove fuel cell symposium,” Plat. Metals Rev., 43, 149 (1999). Cooper, B.J., “Challenges in emission control catalysis for the next decade,” Plat. Metal Rev., 38, 2 (1994). Davies, P., and A. J. Hall, “Process for Forming Hydrogen and Carbon Dioxide Using a Catalyst Consisting Essentially of Oxide of Copper, Zinc and Alumina or Magnesium,” U.S. Patent, 3, 961, 037 (1976). Eliasson, B., C. J. Liu and U. Kogelschatz, “Direct Conversion of Methane and Carbon Dioxide to Higher Hydrocarbons Using Catalytic Dielectric-Barrier Discharges with Zeolites,” Ind. Eng. Chem. Res., 39, 1221 (2000). Frisch, M. J., G. W. Trucks, H. B. Schegel, et al., Gaussian 03 Revision C.02, Gaussian Inc., Wallingford CT (2004). Hoogers, G. and D. Thompsett, “Releasing the Potential of Clean Power,” Chem. Ind., 20, 796 (1999). Indarto, A., J. W. Choi, H. Lee and H. K. Song, “Kinetic Modeling of Plasma Methane Conversion using Gliding Arc Plasma,” J. Natur. Gas Chem., 14, 13 (2005). Indarto, A., J. W. Choi, H. Lee and H. K. Song, “Methane conversion using dielectric barrier discharge: comparison with thermal process and catalyst effects,” J. Natur. Gas Chem., 15,

12

87 (2006a). Indarto, A., J. W. Choi, H. Lee and H. K. Song, “Effect of additive gases on methane conversion using gliding arc discharge,” Energy, 31, 2650 (2006b). Indarto, A., “Chemical Reactions by Gliding Arc Plasma,” Thesis, Division of Environmental and Process, Korea Institute of Science and Technology, Seoul, Korea (2006c). Indarto, A., N. Coowanitwong, J. W. Choi, H. Lee, and H. K. Song, “Kinetic modeling of plasma methane conversion in a dielectric barrier discharge,” Fuel Process. Technol., accepted (2007a). Indarto, A., J. W. Choi, H. Lee and H. K. Song, “A Brief Catalyst Study on Direct Methane Conversion Using a Dielectric Barrier Discharge,” J. Chin. Chem. Soc., accepted (2007b). Jamal, Y. and M. L. Wyszynski, “On-board Generation of Hydrogen Rich Fuels-A Review,” Int. J. Hydrogen Energy, 19, 557 (1994). Kato, S., R. Suzuki, Y. Kawamoto and H. Uchida, “Reactivities of H 2, O2 and H2O with the Cr surface between 77K and 298K,” J. Alloy. & Compound., 413, 214 (2006). Kim, S. S., H. Lee, B. K. Na and H. K. Song, “Plasma-Assisted Reduction of Supported Metal Catalyst Using Atmospheric Dielectric-Barrier Discharge,” Catal. Today, 89, 193 (2004). Liu, S., K. Takahashi, K. Uematsu and M. Ayabe, “Hydrogen Production by Oxidative Methanol Reforming on Pd/ZnO Catalyst: Effects of the Addition of a Third Metal Component,” Appl. Catal. A: Gen., 277, 256 (2004). Matsumoto, S., Y. Ikeda, H. Suzuki, M. Ogai, N. Miyoshi, “NOx storage-reduction catalyst for automotive exhaust with improved against sulfur poisoning,” Appl. Catal. B: Environ., 25, 115 (2000). Ohta, M., Y. Ikeda and A. Igarashi, “Preparation and Characterization of Pt/ZnO-Cr2O3 13

Catalyst for Low-Temperature Dehydrogenation of Isobutane,” Appl. Catal. A: Gen., 258, 153 (2004). Peña, M. A., J. P. Gómez and J. L. G. Fierro, “New Catalyst Routes for Syngas and Hydrogen Production,” Appl. Catal. A: Gen., 144, 7 (1996). Ralph, T. R. and G. A. Hards, “Powering the Cars and Homes of Tomorrow,” Chem. Ind., 9, 327 (1998). Rostrup-Nielsen, J. R. and J. H. B. Hansen, “CO 2-Reforming of Methane over Transition Metals,” J. Catal., 144, 38 (1993). Rostrup-Nielsen, J. R. Catalytic Steam Reforming; Danish Technical Press, Denmark (1984). Simard, F., U. A. Sedran, J. Sepúlveda, N. S. Fígoli and H. I. de Lasa, “ZnO-Cr 2O3 + ZSM-5 Catalyst with very low Zn/Cr ratio for the Transformation of Synthesis Gas to Hydrocarbons,” Appl. Catal. A: Gen., 125, 81 (1995) Thomas, J. M. and W. J. Thomas, Principles and Practice of Heterogeneous Catalysis, VCH Publishers, Weinheim, pp. 515–521. (1996). Twigg, M. V., Catalyst Handbook, Wolf Scientific Text, London, U.K. (1989). Velu, S., K. Suzuki and T. Osaki, “Selective Production of Hydrogen by Partial Oxidation of Methanol over Catalyst Derived from CuZnAl-layered Double hydroxides,” Catal. Lett., 62, 159 (1999). Venugopal A., S. N. Kumar, J. Ashok, D. H. Prasad, V. D. Kumari, K.B.S. Prasad, and M. Subrahmanyam, “Hydrogen Production by Catalytic Decomposition of Methane over Ni/SiO2,” Int. J. Hydrogen Energy, 32, 1782 (2007). Yang, Y., “Direct Non-Oxidative Methane Conversion by Non-Thermal Plasma: Experimental Study,” Plasma Chem. Plasma Process, 23 (2003). Yao S., A. Nakayama and E. Suzuki, “Methane Conversion Using a High-Frequency Pulsed 14

Plasma: Important Features,” AIChE J., 47, 413 (2001).

15

AC g e n e ra to r

M e th a n e M FC

p la s m a zone

c a ta ly s t

R e a c to r Figure 1. Experimental setup

16

B u b b le flo w m e te r

G as C h ro m a to g ra p h y

ZnO Z n C r2O C r2O

Figure 2. XRD spectra of calcined catalyst

17

3

4

70

Conversion (%)

60 50 40 30 20 10 0 60

80

100

120

140

S u p p lie d p o w e r ( W ) (a)

1.0

Selectivity

0.8

unkno w n n - C 4H 10 i- C 4 H 1 0 C3 C 2H 6 C 2H 2 + C 2H H2

0.6 0.4 0.2 0.0 60

70

80

90

100

110

120

130

S u p p lie d p o w e r ( W ) (b)

Figure 3. Effects of supplied power variation on (a) methane conversion and (b) products distribution

18

2

100

Conversion (%)

80

60

40

20

0 0.2

0.4

0.6

0.8

1

1.2

1.6

1.4

1.8

R e s id e n c e t im e ( m in ) (a)

1 .0

Selectivity

0 .8 0 .6 0 .4

unknow n C4 C3

0 .2

C H

2 2

0 .0 0 .3

0 .5

0 .7

0 .9

1 .1

1 .3

1 .5

1 .7

R e s id e n c e t im e ( m in ) (b)

Figure 4. Effects of methane residence time on (a) methane conversion and (b) products distribution

19

1 .9

20

∆G (kcal/mol)

0

C 2H

6

+ H --> C 2H

5

+ H

C 2H

5

+ H --> C 2H

2

- 20 - 40

C 2H

- 60

4

4

+ H

2

+ H --> C 2H

3

+ H

2

- 80 - 100

C 2H

3

+ H --> C 2H

2

+ H

2

- 120 - 140

Figure 5. The Gibbs energy different (ΔG) profile of dehydrogenation of C2 hydrocarbons in the present of hydrogen atom

20

60

H2 selectivity (%)

50 40 b la n k

30 1a 1b 2a 2b 3a 3b

20 10 0

20

40

60

80

100

120

140

160

180

S a m p lin g tim e ( m in .) (a)

CH4 conversion (%)

60 50 40 b la n k

30 1a 1b 2a 2b 3a 3b

20 10 0 0

20

40

60

80

100

120

S a m p l in g t im e ( m in . ) (b)

21

140

160

180

30

Selectivity (%)

25 20 15

C

3

10

C 2H

6

C 2H

2

+ C 2H

4

5

i- C 4 H

10

+ n -C 4H

10

0 0

20

40

60

80

100

120

140

S a m p l in g t im e ( m in . ) (c)

Figure 6. Effect of the catalyst on (a) methane conversion, (b) hydrogen selectivity, and (c) hydrocarbon selectivity. Data was obtained at gas flow rate of 30ml/min and supplied power of 80W.

22