Co-generation of synthesis gas and C2C hydrocarbons from methane and carbon dioxide in a hybrid catalytic-plasma reactor: A review

Transcription

1 Diponegoro University From the SelectedWorks of Istadi August, 2006 Co-generation of synthesis gas and C2C hydrocarbons from methane and carbon dioxide in a hybrid catalytic-plasma reactor: A review Istadi, Diponegoro University Nor Aishah Saidina Amin, Universiti Teknologi Malaysia Available at:

2 Fuel 85 (2006) Review Co-generation of synthesis gas and C 2C hydrocarbons from methane and carbon dioxide in a hybrid catalytic-plasma reactor: A review Istadi 1, Nor Aishah Saidina Amin * Chemical Reaction Engineering Group (CREG), Faculty of Chemical and Natural Resources Engineering, Universiti Teknologi Malaysia, UTM Skudai, Johor Bahru, Malaysia Received 27 May 2005; received in revised form 2 September 2005; accepted 4 September 2005 Available online 27 September Abstract The topics on conversion and utilization of methane and carbon dioxide are important issues in tackling the global warming effects from the two greenhouse gases. Several technologies including catalytic and plasma have been proposed to improve the process involving conversion and utilization of methane and carbon dioxide. In this paper, an overview of the basic principles, and the effects of CH 4 /CO 2 feed ratio, total feed flow rate, discharge power, catalyst, applied voltage, wall temperature, and system pressure in dielectric-barrier discharge (DBD) plasma reactor are addressed. The discharge power, discharge gap, applied voltage and CH 4 /CO 2 ratio in the feed showed the most significant effects on the reactor performance. Co-feeding carbon dioxide with the methane feed stream reduced coking and increased methane conversion. The H 2 /CO ratio in the products was significantly affected by CH 4 /CO 2 ratio. The synergism of the catalyst placed in the discharge gap and the plasma affected the products distribution significantly. Methane and carbon dioxide conversions were influenced significantly by discharge power and applied voltage. The drawbacks of DBD plasma application in the CH 4 CO 2 conversion should be taken into consideration before a new plausible reactor system can be implemented. q 2005 Elsevier Ltd. All rights reserved. Keywords: Dielectric-barrier discharge; Plasma discharge; Microdischarge; Plasma reactor; CH 4 CO 2 conversions 1. Introduction Mitigation of CO 2, one of the most important greenhouse gases, is the crucial agenda in global warming issues. Meanwhile, the direct conversion of methane to C 2C hydrocarbons and synthesis gas has a large implication towards the utilization of natural gas in the gas-based petrochemical and liquid fuel industries. The CH 4 /CO 2 ratio in Natuna s and Arun s natural gas compositions (28/71 and 75/15, respectively) should be strategically utilized for the production of synthesis gas, higher hydrocarbons, liquid fuels and other important chemicals. As a consequence, the conversion and utilization of methane and carbon dioxide are widely researched in the field of C 1 chemistry. Several technologies have been proposed to improve the efficiency of * Corresponding author. Tel.: C ; fax: C address: noraishah@fkkksa.utm.my (N.A.S. Amin). 1 Permanent address: Chemical Reaction Engineering & Catalysis (CREC) Group, Department of Chemical Engineering, Diponegoro University, Semarang, Indonesia. address: istadi@tekim.ft.undip.ac.id /$ - see front matter q 2005 Elsevier Ltd. All rights reserved. doi: /j.fuel methane and carbon dioxide utilization. In the past decades, from the perspective of catalytic chemistry, most efforts have focused on the utilization of CO 2 as a source of carbon. Only recently it has been proposed that CO 2 might also be utilized as an oxygen source or oxidant as it can be considered to be a non-traditional oxidant and oxygen transfer agent. The potentials of non-conventional DBD plasma reactor for converting the two greenhouse gases, methane and CO 2,to synthesis gas and higher hydrocarbons at low temperature and ambient pressure have also been recently reported [1 4]. A comprehensive review on recent development of plasma reactor technology for the co-generation of synthesis gas and C 2C hydrocarbons from methane and carbon dioxide is essential to address the features, drawbacks, challenges, and feasibility of this technology. Non-thermal plasma can be defined as gas consisting of electrons, highly excited atoms and molecules, ions, radicals, photons and neutral particles in which the electrons have a much higher energy than the neutral gas particles. Non-thermal plasma is also called non-equilibrium plasma due to the significant difference of temperature or kinetic energy between the electrons and the neutral particles [1 4]. The gas temperature can be within the range of room temperature,

3 578 Istadi, N.A.S. Amin / Fuel 85 (2006) while the electrons can reach temperatures of Kina dielectric-barrier discharge. The non-thermal plasma can be generated and maintained by electrical discharge [4 6]. The electrical discharge is a direct way to produce non-thermal plasma by applying a high voltage to a gas space and incurring gas breakdowns. The gas breakdowns generate electrons that are accelerated by an electric field forming non-thermal plasma. The electrical discharges can be realized in several ways depending on the types of voltage applied and reactor specification. In the plasma reactor, the energetic electrons collide with molecules in the gas, resulting in excitation, ionization, electron multiplication, and the formation of atoms and metastable compounds [2,4,6]. When the electric field in the discharge gap is high enough to cause breakdowns in most gases a large number of microdischarges are observed. The active atoms and metastable compounds subsequently collide with molecules, and reactions may occur. Pertaining to methane and carbon dioxide conversions, it is expected that the methane and carbon dioxide participate in the reactions and be converted into synthesis gases and higher hydrocarbons. In this paper, a review of DBD plasma reactor application for co-generation of C 2C hydrocarbons and synthesis gas in the reaction between methane and carbon dioxide is reported. The DBD plasma reactor application is also compared with the development of conventional catalyst and reactor technology for CH 4 and CO 2 conversions to C 2C hydrocarbons. The effect of plasma reactor process variables, such as CH 4 /CO 2 feed ratio, discharge power, system pressure, total feed flow rate, applied voltage, pulsed power, and the role of heterogeneous catalysis, are considered. In addition, the drawbacks and challenges of plasma-catalysis technologies in CH 4 CO 2 conversion are also addressed. 2. Methane and carbon dioxide conversions to C 2C hydrocarbons in conventional catalytic reactor The oxidative coupling of methane (OCM) is a promising and novel route for the conversion of natural gas to C 2 hydrocarbons in the presence of a basic catalyst within the temperature range of K [7]. The inevitable formation of carbon monoxide (CO) and CO 2, however, seems to be one of the most serious problems from the practical point of view. Recently, carbon dioxide was used to replace oxygen as an oxidant in oxidative coupling of methane by carbon dioxide (CO 2 OCM) to produce active oxygen species promoted CH 3 * radicals [8 13]. Carbon dioxide, as an oxidant, has several advantages over O 2. CO is the only by-product in this case. Unlike O 2,CO 2 does not induce gas phase radical reactions. In particular, the reactions of CH 4 and CO 2 to produce C 2 hydrocarbons are mainly be controlled by heterogeneous catalysis. It is thus expected that the development of active catalysts achieve high selectivity to C 2 hydrocarbons. Eqs. (1) and (2) are the two main reaction schemes for CO 2 OCM to produce C 2 hydrocarbons with carbon monoxide and water as the by-products. 2CH 4 CCO 2 #C 2 H 6 CCO CH 2 O (1) DG K Z 69:6 kj mol ÿ1 2CH 4 C2CO 2 #C 2 H 4 C2CO C2H 2 O (2) DG K Z 51:9 kj mol ÿ1 The equilibrium conversion of methane to C 2 H 6 and C 2 H 4 based on thermodynamic calculations was reported [14]. The conversion enhanced as the temperature and CH 4 /CO 2 ratio decreased. Lower CH 4 /CO 2 ratio increased the methane conversion to C 2 H 6 and C 2 H 4 and their yield exceeded 15 and 25%, respectively, at 1073 K for the reactant with a CH 4 / CO 2 ratio of 0.5. An efficient catalyst that is capable not only of activating both methane and CO 2 but also of producing C 2 hydrocarbons selectively is imminent. Direct methane and carbon dioxide conversion into C 2 hydrocarbons provides a novel route for the simultaneous activation and utilization of methane and carbon dioxide gases. Table 1 reveals the studies involving oxidative coupling of Table 1 Summary of catalysts system for CO 2 OCM to C 2 hydrocarbons using conventional fixed bed reactor Catalysts Temp. (K) CH 4 conv. (%) CO 2 /CH 4 ratio C 2 hydrocarbons Ref. Selectivity (%) Yield (%) La 2 O 3 ZnO [15] CaO Cr 2 O [16] CaO MnO [16] Na 2 WO 4 Mn/SiO [17] CaO CeO [14,18] CaO ZnO [19] SrO Cr 2 O [13] SrO ZnO [13] SrO CeO [13] SrO MnO [13] BaO ZnO [13] BaO CeO [13] BaO Cr 2 O [13] BaO MnO [13] MnO 2 SrCO [20]

4 Istadi, N.A.S. Amin / Fuel 85 (2006) methane to ethane and ethylene (C 2 hydrocarbons) [13 20]. Enhancement of C 2 hydrocarbons formation by CO 2 was first observed in the oxidative coupling of methane over a PbO/ MgO catalyst, but it could not be sustained in the absence of O 2 [21]. Asami et al. [10,11] systematically investigated the catalytic activities of more than 30 metal oxides for the conversion of methane by CO 2 in the absence of O 2 and found that lanthanide oxides generally showed higher activities. Among them, praseodymium oxide or terbium oxide exhibited relatively good catalytic performance with a C 2 yield of 1.5% and a selectivity of 50% at 1123 K. Wang et al. [22] investigated the conversion of methane to C 2 H 6 over praseodymium oxide and reported that the oxide could be effective in the presence of CO 2 at temperatures as low as K to form C 2 hydrocarbons. In addition, the binary oxides catalysts were also investigated by several research groups for the oxidative coupling of methane by carbon dioxide as presented in Table 1 [15,16,18 20]. However, over the binary oxide catalysts, the CO 2 OCM reaction still exhibited low performances at high temperature (about 1123 K), i.e. methane conversion up to 6.3%, C 2 hydrocarbons selectivity up to 91%, and C 2 hydrocarbons yield up to 4.5%. Pertaining to ternary metal oxides system, Na 2 WO 4 Mn/SiO 2 catalyst, which was a better catalyst for O 2 OCM, was investigated for the conversion of CH 4 and CO 2 to C 2 hydrocarbons [17]. A methane yield of about 4.5% and a C 2 hydrocarbons selectivity of 94% were obtained at 1093 K [17]. The reaction temperature was observed to favour methane conversion but it did not favour the selectivity to C 2 hydrocarbons. This is due to surface lattice oxygen, which is responsible for the selective oxidation of methane to C 2 hydrocarbons, whereas the bulk lattice oxygen is responsible for deep oxidation. The results of CeO 2 -based catalysts screening for CO 2 OCM process over binary and ternary metal oxides revealed that the 15 wt% CaO 5 wt% MnO/CeO 2 catalyst was the most potential [23]. Interestingly, the stability test showed that the 15 wt% CaO 5 wt% MnO/CeO 2 catalyst was stable with no obvious coking during 20 h of reaction time on stream. In another investigation, single- and multi-responses optimizations of CO 2 OCM reaction over CaO MnO/CeO 2 catalyst were developed to address the optimal process parameters and catalyst compositions [24,25]. Maximum C 2 hydrocarbons selectivity and yield of 76.6 and 3.7%, respectively, were achieved over 12.8% CaO 6.4% MnO/ CeO 2 catalyst at optimum reactor temperature being 1127 K and CH 4 /CO 2 feed ratio being 0.5. Thermodynamically, the equilibrium yields of ethane and ethylene at 1073 K are sufficiently high (13 and 57%, respectively) in the CO 2 OCM [12,26]. Another study [14] reported equilibrium conversion of methane to C 2 H 6 and C 2 H 4 from thermodynamic calculations as mentioned above. However, the experimental results on methane carbon dioxide conversion to C 2 and higher hydrocarbons are not laudable as the methane conversion and C 2 hydrocarbons yield is not higher than 7 and 6%, respectively. The major intricacy for direct methane conversion to higher hydrocarbons and oxygenates involve the intensive energy consumption requirement and also activating the stable C H bonds in the methane molecule using conventional catalysis [3,27]. Oxidative coupling of methane to ethane and ethylene requires a high reaction temperature due to the high activation dissociation energy of most C H bonds [28]. Due to the difficulty in obtaining high yield in CO 2 OCM, further improvements are required including the exploitation of some non-conventional technologies, for example, application of plasma reactor technologies. In the plasma reactor, the methane and carbon dioxide conversions proceed over different reaction pathways, where radicals and excited species play a dominant role. Therefore, the interaction between the gas phase and the catalyst surface in the combined catalytic-plasma can be significantly different from the conventional catalytic mechanisms. 3. Methane and carbon dioxide conversions in dielectricbarrier discharge plasma reactor 3.1. Principles of dielectric-barrier discharge plasma reactor The DBD plasma reactor shows potential for organic and inorganic chemical reactions. The potential is due to its nonequilibrium properties, low input power requirement, and its capacity to induce physical and chemical reactions within gases at relatively low gas temperature. The DBD is also called silent discharge or barrier discharge. The characteristics of non-thermal discharges are described in Table 2. The basic configurations of dielectric-barrier discharge are depicted in Fig. 1 [5,6]. The DBD is charged by a high voltage electrical power source (coded as AC in Fig. 1). The classical DBD configurations are planar or cylindrical with at least one electrode covered by dielectric materials such as glass, quartz and ceramics [5,6]. The dielectric, being an insulator, cannot pass a current. Its dielectric constant and thickness determine the amount of displacement current that can be passed through the dielectric. In most applications the dielectric limits the average current density in the gas space. The dielectric can also be placed between the electrodes to separate two gas layers. The roles of the dielectric are not only limited by the amount of charge and energy conveyed to an individual microdischarge, but it also distributes the microdischarges over the entire electrode area. An intact dielectric guarantees that no spark or arc can occur in the discharged gap. The total charge transferred in a microdischarge depends on the gas properties Table 2 Characteristic of non-thermal discharges [1,5,29] Parameter Glow discharge Corona discharge Barrier discharge Pressure (bar)!10 K2 1 1 Electric field (kv cm K1 ) , variable Electron energy (ev) , variable 1 10 Electron temperature (K) ,000 50,000 10, ,000 Electron density (cm K3 ) , variable Degree of ionization 10 K6 10 K5 Small, variable 10 K4

5 580 Istadi, N.A.S. Amin / Fuel 85 (2006) Fig. 1. Basic configurations of dielectric-barrier discharge (AC denotes a high voltage electrical source). and can be influenced by the discharge gap spacing and by the properties of the dielectric. An advantage of the DBD reactor over many other discharge types is that one can vary the average energy of the electrons generated by adjusting the product of gas density and gap width [5,6,30,31] Gas breakdown in dielectric-barrier discharge plasma reactor In order to initiate and maintain an electrical discharge process, a high voltage is applied across two electrodes separated by the dielectric materials and the discharge gap as demonstrated in Fig. 1. The voltage applied should be high enough to establish a high electric field that causes the gas breakdown. The electron temperature reaches thousands degree K, while the neutral gas can be in the range of room temperature [6,30]. The most important properties of DBD reactor are gas breakdown, which is a large number of independent current filaments of nanosecond duration, namely microdischarges [5,30,32]. Each microdischarge has an almost cylindrical plasma channel, typically of about 100 mm radius and spreads into a larger surface discharge at the dielectric surface [6,30,32]. There are three stages during the life cycle of such a filament: the electrical breakdown, transport of charge across the gas, and the excitation of atoms and molecules initiating the reaction kinetics. At atmospheric pressure electron densities of cm K3 and current densities in the range of A cm K2 are reached [30]. Research on DBD plasma reactor has focused on tailoring the microdischarge characteristics by making use of special gas properties, adjusting pressure and wall temperature, optimizing the electrode geometry, and the properties of the dielectric material [6,31,33]. The first step in the DBD plasma reaction is a dissociation of the initially molecular species by electron collisions. The ionic and excited atomic and molecular species initiate chemical reactions that finally result in the synthesis of a desired species [5,30]. Studies on electron impact dissociation of gases, like methane and carbon dioxide, is necessary in relation to utilization of both gases in the DBD reactor. The important factor influencing the DBD plasma applications is the efficiency of the initial dissociation process with respect to energy consumption. In CO 2 decomposition under favourable conditions, at most 40% of the electron energy can be utilized for the dissociation process [6,34]. Zhou et al. [35], Liu et al. [36,37] and Zhang et al. [38 40] suggested that the DBD was an efficient tool for converting CH 4 and CO 2 to synthesis gas at low temperature and ambient pressure. If a sufficient electric potential is applied across the gas, free electrons within the gas are accelerated to high energies and can collide with the gas molecules resulting in ionization. The ionization produces more number of free electrons that are also accelerated and repeat the process until an electron avalanche (sudden abundance) occurs between electrodes [2,30]. The accelerated electrons can gain considerable energy in the electric field and are capable of transferring their kinetic energy to the molecules of the feed gases through inelastic collisions. These collisions can increase the internal energy of the gas molecules and may result in excitation, dissociation, or ionization without significantly increasing the bulk gas temperature [2,4,30,34,39]. The increase in the internal energy of the gas molecules can overcome the reaction activation energy. The high energetic electrons collision does not involve significant heating to the bulk gas. Hence, the bulk gas temperature is not in equilibrium with the electrons and remains relatively low [3]. The electron energies within the system are affected by the electric field strength and interactions with the gas species present. The average electron field increment results in increasing the average electron energy. The electric field is also related to breakdown strength and is a function of the breakdown voltage, gas gap distance and system pressure. Increasing the system pressure or gas gap distance results in a decrease of the electric field and hence the average electron energy within a system Influence of dielectric material properties in DBD plasma reactor It is believed that the reactivity of DBD plasma reactor could be improved by increasing the permittivity of the dielectric barrier. The dielectric barrier with large dielectric constant and strength is extremely desirable in order to generate highly reactive DBD plasma [41]. As an example, ceramic and SrTiO 3 as a dielectric has a high dielectric constant and modest dielectric strength, while CaTiO 3 exhibited a low dielectric constant and high dielectric strength. Moreover, Ca 0.7 Sr 0.3 TiO 3 barrier was expected to offer high permittivity and high dielectric strength [41,42]. In the case of Ca 0.7 Sr 0.3 TiO 3 barrier, the arching voltage decreased and CO 2 conversion increased with increasing input frequency. In terms of using only alumina or silica glass as a barrier, the CO 2 conversion was much lower. These results might be attributed to the much plasma power generated by using the Ca 0.7 Sr 0.3- TiO 3 barrier than an alumina and silica glass [41,42]. The power of plasma seemed to be the most important parameter in governing the CO 2 decomposition because the plasma reaction was associated mainly with the energy provided, and may be

6 Istadi, N.A.S. Amin / Fuel 85 (2006) having synergistic effect with Ca content. The similar result was reported by Wen and Jiang [43] over g-al 2 O 3 catalyst. The essential effect of the dielectric properties on the DBD plasma generation can be elucidated by the following equations [5,41,42]. The dielectric capacity (C) is written as follows: C Z 3S (3) d If QZCV, the electric charge (Q) can be expressed: Q Z 3SV (4) d The electric current (I) and electric power (P) can be written: I Z Q t P Z IV (6) The electric power can also be written as follow: P Z 3SV2 (7) td In this relations, 3 is permittivity, S is area of electrode, d is distance between parallel-plate electrodes, V is input voltage, I is electric current and t is time. The reactivity of the DBD plasma is improved by increasing the permittivity of the dielectric barrier, which is more electrons with respect to the discharge physics. The important initiating reactions in the DBD plasma reactor are collisions of the energetic electrons and the reactive constituents. As an example, the energetic electron collides with CO 2 or CH 4 molecules and excites them to higher energy levels to dissociate or initiate the reactions: CO 2 Ce/CO 2 Ce/CO C1=2O 2 Ce (8) CO 2 Ce/CO 2 Ce/CO CO Ce/CO CO (9) CH 4 Ce/CH 4 Ce/CH 3 CH Ce (10) where CO 2 and CH 4 denotes the excited CO 2 molecules, respectively [1,41,43,44]. and CH 4 4. Various designs of plasma reactor for CH 4 and CO 2 conversions Various configurations of plasma reactor for methane and carbon dioxide conversions have been proposed. A pulsedcorona discharge (PCD) and a dielectric-barrier discharge (DBD) were the two of the most extensively investigated devices producing atmospheric non-equilibrium plasma. The catalytic corona discharge reactor was proposed for carbon dioxide reforming of methane reaction [44 46] and partial oxidation of methane [46] as depicted in Fig. 2(a). In the figure, the top wire electrode is a stainless steel rod concentric with the centre of the reactor tube, while the lower ground electrode is a circular disk positioned perpendicular to the reactor axis and 10 mm below the tip of the top wire electrode when the remote discharge is employed. The catalyst bed is about 8 mm deep, (5) thus the tip of the wire electrode is about 2 mm above the top of the catalyst bed for the remote discharge (Fig. 2(a), left part) and about 2 mm within the bed for the direct discharge (Fig. 2(b), right part). The plasma reactor is designed to the remote and the direct discharge promoted with OH groupscontented catalyst [45]. The streamer corona discharge is characterized by low bulk gas temperature and high electrons temperature. Similar design was presented by Caldwell et al. [2] in which the height of catalyst zone of 1 cm was placed on the discharge zone between the tip of high voltage wire and the plate of ground. The visible discharge of fine bluish filamentous sparks between the electrodes was appeared. Similar principles but different design was shown by Yao et al. [47] using pulsed corona discharge with co-axial cylindrical (CAC) reactor [47] as depicted in Fig. 2(b) and point-to-point (PTP) reactor [48] as demonstrated in Fig. 2(c). From Fig. 2(b), the CAC reactor is comprised mainly of a Pyrex tube, quartz tube, a stainless steel tube, and a stainless steel wire [47]. A quartz tube and two acrylic resin holders are used to keep the wire straight and on the axial of the Pyrex tube. The discharge occurs between the central wire and the inside surface of the stainless steel tube. Owing to the PTP reactor design as presented in Fig. 2(c), the reactor is simply composed of a Pyrex tube and two stainless steel electrodes of sharp terminals [48]. In the pulsed corona discharge, most of the energy is injected over the duration of the pulsed corona discharge and is used to heat the methane in the discharge channels to temperatures at which methane pyrolysis occurs. A different design of corona/spark discharge reactor was suggested by Kado et al. [49 51] as presented in Fig. 2(d). The flow type reactor (Fig. 2(d)) is composed of a quartz tube under the conditions of room temperature and atmospheric pressure without a catalyst. The electrode configuration is needle-toneedle type. The stainless steel rods are used as the electrodes [49 51]. The electrodes is centred and supported by the stainless steel mesh. DC negative high voltage is applied between the electrodes. An interesting reactor design was also proposed by Huang et al. [52] using Y-type reactor to generate plasmas according to reaction specifications as depicted in Fig. 3. In this design, the angle between the two arms is An input high voltage can be applied on either or both arms of the Y-type reactor. The inner electrodes for the reactor are stainless steel rods and the outer electrodes are aluminium foil wrapped around quartz tube. Plasmas are generated between the inner electrodes and inner walls of the quartz tube. Another plasma reactor design for methane and carbon dioxide conversions is DBD reactor. The DBD reactor is a nonthermal plasma phenomenon, which is very promising for synthesis gas and organic chemical productions because of its non-equilibrium properties, low input power requirement, and its capacity to induce physical and chemical reactions within gases at relatively low temperatures. Generally, the feed gas flow was subjected to the action of DBD in an annular gap formed between an outer stainless steel/aluminium tube as ground electrode, maintained at constant temperature, and an inner quartz tube as dielectric [26,27,29,36,37,49,53 55].

7 582 Istadi, N.A.S. Amin / Fuel 85 (2006) Fig. 2. Schematic of corona discharge reactor system. (a) Remote and direct discharge [45,46], (b) coaxial cylindrical (CAC) [28,47], (c) corona/spark discharge [49 51], (d) point-to-point (PTP) [48]. The various designs of DBD reactor for CH 4 and CO 2 conversions are shown in Fig. 4. From Fig. 4(a), the electrodes configuration is the coaxial type. Inside of the reactor, the high voltage electrode of copper is inserted and fixed at the centre. The outside of the reactor is covered with aluminium tape and grounded. The quartz tube plays a role of the dielectric material [49,56]. Fig. 4(b) presents the DBD plasma reactor with catalyst designed by Kim et al. [29]. In the figure, the dielectric material of the reactor is a quartz tube. Two stainless steel wires are installed in the quartz tube as an electrode. The outer surface of quartz tube is coated with silver as another electrode, and the length of reaction zone is 200 mm. An AC pulse power supply is used in this equipment. The catalyst is packed in the lower part of the DBD reactor, while the upper part is remained blank. Below the catalyst-packed volume, i.e. at the lower nonplasma zone, glass beads are packed. The lower and upper parts of Fig. 4(c) demonstrate the DBD reactor with and without after-glow zone, respectively [55]. The high voltage electrode is an aluminium foil attached to the inner surface of the quartz tube. A stainless steel tube around the quartz tube acts as ground electrode. The width of discharge gap is equal to the distance between the steel tube and the quartz tube. Pertaining to after-glow zone DBD reactor (Fig. 4(c), lower part), the high voltage electrode (aluminium foil) is divided into five parts with equal length [55]. A certain distance between every part Fig. 3. Y-type plasma reactor without catalyst [52].

8 Istadi, N.A.S. Amin / Fuel 85 (2006) Fig. 4. Various designs of DBD reactor. (a) DBD plasma without catalyst [49,56], (b) DBD plasma with catalyst [29], (c) DBD reactor without catalyst equipped with and without after-glow zone [55]: (1) high voltage electrode, (2) quartz tube, (3) aluminium foil, (4) grounded electrode, (d) annular DBD plasma reactor without catalyst [4]. serves as an after-glow zone. Another design of DBD plasma reactor was proposed by Larkin et al. [4] owing to annular reactor without catalyst as presented in Fig. 4(d). From the figure, the inner cylinder is made of glass and acts as the dielectric. The glass dielectric cylinder has a stainless steel metal foil on its inner wall that acted as an electrode. The outer cylinder is also made of stainless steel and its inner wall acted as the other electrode. The reactant gases enter through the top of the reactor and flow axially downward in a gas gap between two concentric cylinders. In particular, the DBD combines large volume excitation of glow discharge with high pressure characteristics of corona discharge. A dielectric layer covers at least one electrode in the silent discharge. The entire electrode area is effective for discharge reactions. A great advantage of DBD over other discharges is that the average energy of electrons can be influenced by changing either gas pressure or gap width. The DBD has also higher electron density than other kinds of plasma reactor. Acetylene was the main C 2 hydrocarbons products obtained from coupling of methane under pulse corona discharge plasma using CO 2 as an oxidant without catalyst [57,58]. However, methane conversion of up to 50% were achieved by using two silent discharge reactors in series without catalyst with a total residence time of 3 min at 13 kv and 25 ma [59]. 5. Effect of operating conditions on CH 4 and CO 2 conversions in DBD plasma reactor 5.1. Effect of CH 4 /CO 2 feed ratio As an oxidant, CO 2 is first decomposed to CO and O radical with a high activity. The dissociation reactions of CO 2 generate some oxygen species. Some excited atomic species such as metastable O( 1 D) are active species for the generation of methyl radicals from methane and are also active for methanol formation from methane [36 38,60,61]. The dissociation or dissociative attachment of CO 2 generates active species that assist plasma catalytic methane conversion. The methane conversion is deeply influenced by the new reactive oxygen species so that the conversion of methane is much higher in the presence of CO 2 than that without CO 2. The possible mechanism is that the reactive oxygen removes H atom in methane to generate the hydrocarbons radicals. The excited species reacts with the hydrocarbons radicals from methane in

9 584 Istadi, N.A.S. Amin / Fuel 85 (2006) the plasma zone. The co-feed of CH 4 and CO 2 promote the conversion of each other. It was found that drawback of DBD plasma methane conversion with pure methane feed was the formation of carbon deposits [36,60]. The feed of carbon dioxide could inhibit the deposition of carbon and thus sustained the operations of DBDs [36,60]. Oxygen was also very effective for low gas temperature plasma activation of methane by the interaction between the plasma and the catalyst. However, oxygen could induce an oxidation of hydrocarbons [60]. The distribution of hydrocarbon products changed significantly with the CH 4 /CO 2 feed ratio [26,36 38,55,57,60]. Enhancement of methane conversion in the DBD reactor might be due to the addition of CO 2 as co-feed [27]. In fact, the methane conversion decreased when the concentration of methane in the feed increased. The methane concentration in the feed was an important factor for determining the total amount of hydrocarbons produced. Lower amount of CO 2 in the feed led to the production of C 2 hydrocarbons [60]. The larger the CO 2 amount in the feed, the higher the CO selectivity and more oxygenates in the products, whereas the selectivities of all higher hydrocarbons decreased [26,27,36 38,55]. In addition to the higher hydrocarbons, a mixture of oxygenates including methanol, DME, formaldehyde, has also been produced. The conversion of methane decreased with increasing CH 4 /CO 2 ratio, while the CO 2 conversion increased. The yield of hydrogen and gaseous hydrocarbons (C 2 C 3 ) increased with increasing CH 4 /CO 2 ratio [38]. Meanwhile, the H 2 /CO ratio increased considerably with increasing methane concentration in the feed gas. It is possible to control the composition of synthesis gas by adjusting the molar ratio of feed gases [27,35 37,39,55]. Effect of CO 2 concentration in the feed on the direct higher hydrocarbons formation using the catalytic DBD over NaX zeolite was reported by Eliasson et al. [26]. The methane conversion increased rapidly from 27 to 46% when the CO 2 content in the feed rose from 0 to 17%. Further increasing the CO 2 content in the feed significantly increased the methane conversion but it decreased the selectivities of higher hydrocarbons. In the case of higher CO 2 concentration in the feed, more oxygenates were produced. The oxygenates include alcohols, acetone, and aldehydes. It was reported that for a cogeneration of syngas and higher hydrocarbons, the optimum CH 4 /CO 2 ratio in the feed would be in the range of 2 3 [26]. Non-catalytic methane and carbon dioxide reaction by varying methane composition in the feed over DBD plasma reactor was also reported at temperature 423 K, discharge power 200 W, feed flow rate 500 ml min K1 and atmospheric pressure [36]. The high CO yield of 91% was achieved at methane composition of 20% in the feed with C 2 selectivity being 8% and methane conversion being 28%. The CO yield decreased as the methane content in the feed increased. In order to obtain a high selectivity of light hydrocarbons with high methane conversion, a CH 4 /CO 2 ratio of about 2 (CH 4 content of 66.7%) was preferred. The maximum C 2 and C 3 hydrocarbons selectivities of 23.8 and 11.1%, respectively, were achieved in the DBD reactor without catalyst with discharge gap 1.8 mm, discharge power 100 W and flow rate of 60 ml min K1 as depicted in Fig. 2(d), but the methane conversion was smaller than other reactor types [37,53]. The reactor with the gap of 1.1 mm produced hydrocarbons higher than that of 1.8 mm. The maximum selectivity of oxygenates was reached with methane concentration of 67.4% in the feed with two major components of acetic acid and ethanol. An interesting result was the generation of acetic acid directly from methane and carbon dioxide as revealed in Eq. (11). CH 4 CCO 2 /CH 3 COOH DG Z 71:17 kj mol K1 (11) The above reaction is not feasible thermodynamically. Under the condition of non-equilibrium DBD, this reaction was favourable at a lower CH 4 /CO 2 feed ratio [55]. However, Bugaev et al. [62] reported that most products were formic acid with the feed mixture of methane and oxygen using DBD plasmas. It was suggested that carbon dioxide probably provided one of carbon atoms in the generation of acetic acids. The amount of liquid hydrocarbons was larger with the existence of after-glow zones as depicted in Fig. 2(d) than that without these zones when methane was predominant in the feed [55]. However, with increasing carbon dioxide in the feed, much more C x H y radicals would combine with the groups containing oxygen so that more oxygenates were generated [37,55]. Pertaining to DBD plasma without catalyst, Zhang et al. [57] found that maximum C 2 yield of 12.7% was achieved over corona discharge reactor with C 2 selectivity of 47.7% at 40% CO 2 content in the feed. In this result, C 2 H 2 was found to be relatively high amounts among the C 2 compounds. At the above conditions, C 2 H 2 content in the product was about 76%, while the C 2 H 4 and C 2 H 6 distributions were 13 and 11%, respectively [57]. The results indicate that C 2 H 2 is the main C 2 product obtained from coupling of methane under non-catalytic pulse corona plasma using CO 2 as an oxidant. Similar results were also achieved by Yao et al. [47] over the similar reactor system. The selectivity of acetylene production increased, but those of ethylene and ethane decreased with increasing pulse frequency. The selectivity of acetylene achieved a level of 85.1% at methane conversion of 23.5% [63]. The catalysts activities of the CH 4 and CO 2 reactions under corona plasma over various La 2 O 3 /g-al 2 O 3 catalysts were reported by Zhang et al. [57]. The 7% La 2 O 3 /g-al 2 O 3 catalyst exhibited higher C 2 selectivity of 72.8% than without catalyst and the methane conversion achieved 24.9%. The catalyst showed the highest selectivity to C 2 H 2 (54.5%), which was higher than C 2 H 4 and C 2 H 6 selectivity being 9.3 and 9.1%, respectively. The 0.01% Pd 5% La 2 O 3 /g-al 2 O 3 catalyst changed the distribution of C 2 products, where C 2 H 4 was the major C 2 products with C 2 H 4 selectivity (46%) higher than C 2 H 6 selectivity (17.9%). Similar results were reported by Liu et al. [60] over zeolite catalysts in a DC corona discharge plasma reactor. The promising result was demonstrated at lower gas temperature (308 K) that gave higher C 2 yield and selectivity (15.46 and 34.4%, respectively) with larger H 2 /CO ratio (4.12). Lower content of CO 2 in the feed led to the

10 Istadi, N.A.S. Amin / Fuel 85 (2006) production of C 2 hydrocarbons and hydrogen [60]. The conversion of CO 2 reached a maximum (40%) when the ratio of CH 4 /CO 2 was 1 over zeolite NaY in a DBD reactor [38]. At the ratio lower than 1, the methane conversion decreased with increasing feed molar ratio, while the conversion of CO 2 increased. In contrast, at CH 4 /CO 2 ratio above 1, the methane and CO 2 conversions decreased with increasing molar ratio. It can be concluded that increasing the molar ratio leads to an increased selectivity to gaseous hydrocarbons (C 2 4 ) and decreased selectivity to CO. The selectivity to C 5C reached a maximum (34.1%) when the molar ratio of methane and CO 2 was 1 at the wall temperature of 423 K, feed flow rate of 200 ml min K1, and discharge power of 500 W [38]. The promising results were also reported by Zhang et al. [39] over cylindrical DBD plasma reactor using HY zeolite catalyst at temperature of 423 K. The highest C 2 selectivity of 15.2% was achieved at CH 4 /CO 2 ratioz3 with C 2 H 6 is the major C 2 content and methane conversion of 52%. In addition at the similar conditions, the products of C 3 and C 4 hydrocarbons and CH 3 OH were also obtained with selectivity of 11, 52.1, and 0.17%, respectively. Synthesis gases were also produced with H 2 /CO ratio of 3.4 and CO selectivity of 21.7%. C 5C hydrocarbons was also produced at CH 4 /CO 2 ratio of 0.5 with C 5C and C 2 selectivity of 19.5 and 9.0%, respectively, while CO selectivity achieved 55.5% with H 2 /CO ratio of Effect of discharge power in DBD plasma reactor Various kinds of electric sources have been used in the plasma reactor, such as AC, DC, microwave, or pulse power supply operated at high voltage and high frequency AC power. However, Song et al. [54] reported that a high frequency pulse power was not widely used on atmospheric gas discharge because of its high price due to difficulty of manufacturing. Discharge voltage and current can be measured using an oscilloscope, while discharge power can be calculated by multiplying voltage by current as described in Eq. (12). The discharge power (P, in W) was calculated by integrating the single period power and multiplying by frequency as written in Eq. (12) [28,47,54] 0 1 ð t 2 B C P VðtÞIðtÞdtAf (12) t 1 where V(t) is voltage as function of time, I(t) denotes current as function of time, and f is frequency. Many authors have used voltage/charge Lissajous figures to determine the averaged discharge power [6,64]. Varying the discharge power affected predominantly on methane conversion and higher hydrocarbons (C 2 C 4 ) selectivity. The conversions of CH 4 and CO 2 increased with increasing discharge power in a catalytic DBD reactor [2,26, 38,39]. Similar trend was reported on the non-catalytic plasma reactors [2,36,47,54]. At low discharge power, CO 2 conversion was higher than methane conversion. It is suggested that CO 2 is easier to be dissociated than methane in the certain discharge power. The dissociation energy of CO 2 (5.5 ev) was lower than methane (10 ev). A rising time of pulse voltage also influenced the methane conversion and C 2 hydrocarbons selectivity. The rising time has remarkable effect on the selectivities of acetylene and ethane when the frequency of pulse was less than 2000 pulses per second (PPS) [65]. When the discharge power increased above 450 W, the methane conversion became higher than CO 2 conversion implying that more plasma species were generated at higher discharge power [26]. In addition to electron, species such as H, OH, O and O K could attack methane molecules to produce more methyl radicals [66]. In fact, increasing discharge power led to decreased selectivities of light hydrocarbons and methanol, and increased C 4 and C 5C selectivities. The highest C 2 and C 3 hydrocarbons selectivities of 36.4 and 18%, respectively, were achieved at discharge power 200 W using DBD reactor with zeolite catalyst [26]. As the power increased, light hydrocarbons (C 2 C 3 ) and methanol might be converted to other higher hydrocarbons (C 4 C 5C ) [26,38,39]. Similar reports have also shown that the low discharge power was beneficial to obtain higher selectivity to CO, methanol and gaseous hydrocarbons (C 2 C 3 ) in catalytic DBD reactor [38,39]. Actually, an increase in the discharge power from 200 to 500 W decreased the methanol selectivity from 0.34 to 0.19% and decreased most of gaseous hydrocarbons (C 2 C 3 ). The selectivity to C 4 hydrocarbons reached a maximum (25.3%) at discharge power of 400 W. It is suggested that higher input power greater than 300 W is necessary for generating higher selectivity to higher hydrocarbons (C 5C ) over DBD reactor with the zeolite catalysts. Eliasson et al. [26] concluded that higher input power should be favoured for efficient methane conversion. Pertaining to non-catalytic plasma reactor, Liu et al. [36] suggested that the conversion of CH 4 and CO 2 increased with increasing power, while the selectivity to light hydrocarbons (C 2 C 4 ) decreased. The increasing discharge power led to an increase of the H 2 /CO ratio in the syngas produced [36]. Since the selectivity of carbon monoxide almost keeps constant with the discharge power, the increasing H 2 /CO ratio implied a decreasing H/C in the hydrocarbons products. In a pulsed corona discharge reactor, the selectivity of acetylene increased with increasing voltage and tended to a level of around 85% at a voltage above 11 kv [47]. The high energy input could yield a high temperature discharge channel. The different trend was shown by Song et al. [54] in non-catalytic DBD reactor with pulsed power. The conversion increased with increasing frequency in both AC and pulse power supplies, but products selectivity and H 2 /CO ratio were constant [54]. The electrical efficiency decreased with increasing system power as more energy was required to convert the available methane when active oxygen becomes depleted. As the discharge power increased, the bulk gas temperature in the reaction zone increased. At higher temperature, the thermal dehydrogenation reactions of methane to form olefins and carbon became more predominant [2]. The acetylene fraction increased considerably with system power. In addition, the

11 586 Istadi, N.A.S. Amin / Fuel 85 (2006) coking formation also became a crucial problem at higher power density Effect of heterogeneous catalysis in DBD plasma reactor Interaction of gas discharge and catalyst in DBD plasma reactor In many heterogeneously catalyzed reactions, the reacting molecules have to be transformed into activated species. The role of the catalyst is often to promote the formation of such species. The non-equilibrium plasma can be used to modify the chemical reactivity. Most electrical energy in a plasma gas discharge goes into the production of energetic electrons rather than into gas heating. The gas discharges are plentiful source of active species for various oxidative or reductive reactions through electron-impact dissociation and ionization reactions. When the gas discharge is introduced into the catalyst layer, the catalyst changes the electronic state of the gas species [45,53, 67,68]. When a gas phase consisting of electrically neutral species, electrons, ions and other excited species flow through the catalyst bed, the catalyst particles become electrically charged. The charge on the catalyst surface, together with other effects of excited species in the gas discharge leads to variations of the electrostatic potential of the catalyst surface. The chemisorption and desorption performances of the catalyst therefore may be modified in the catalyst surface [67,69,70]. The effects of these modifications on methane conversion depend on the amount and concentration of surface charge and the species present on the catalyst surface [29]. The principle of combining a DBD with a heterogeneous catalyst was possible to activate the reactants in the discharge prior to the catalytic reaction, which should have a positive influence on the reaction conditions. Kraus et al. [71] proposed a numerical modelling of the gas phase reactions in the plasma to get an idea of the radical chemistry in the gas. They used the results of the modelling to deduce possible mechanisms for the gas phasesurface interaction. An important difference between regular catalysis and the non-equilibrium gas discharge promoted catalysis is the energy distribution between the products and the catalysts. In the regular catalysis, the temperature of gaseous products is the same as that of the catalyst. However, in the gas discharge promoted catalysis there is an unequal partitioning of the energy between the charged gas species and the catalyst with the charged gas molecules having most of energy. By acting as a source of charged species, the catalyst may enhance the nonequilibrium properties of the gas discharge. Liu et al. [45] reported that in the presence of catalyst, the corona discharge was stable at a lower gas temperature compared to the gas discharge in the absence of the solid catalyst. The gas phase reactions were very important particularly at low temperatures. Their study concluded that catalytic activity of the hybrid plasma catalysis was determined by the ability of the catalyst to influence the vibrational energy of the plasma at the interface of the plasma-solid catalyst. In the CH 4 CO 2 conversion the role of the catalyst is to adsorb methane- and CO 2 -based radicals and to desorb the product molecules [67], while in the absence of plasma the function of the catalyst is to adsorb the methane and CO 2 reactants, to dissociate the C H and CZO bonds and finally to desorb the products of the reaction Effect of hybrid catalytic-plasma DBD reactor The role and properties of catalyst in combination with the dielectric have been studied for the application of DBD plasma technology in catalytic CH 4 CO 2 reaction. Since most of the energetic electrons required to activate the feed gas are located in a discharge gap, special consideration must be taken in the design a reactor that maximized the contact time between the energetic electrons and the neutral feed gas species. The catalyst located in the discharge gap is an alternative to increase the time and area of contact in addition to other modification of electronic properties. The energetic electron determined the chemistry for the conversions of CH 4 and CO 2 [2,26,28,35,66]. The natures of dielectric and electrode surfaces are also an important factor for products distribution of CH 4 and CO 2 conversions using the DBD. The dielectric, as an insulator, cannot pass a DC current in the DBD plasma reactor. In most application, the dielectric limits the average current density in the gas space based on its dielectric constant and thickness in combination with the applied voltage. Preferred materials for dielectric are glass, silica glass, ceramic, polymer, and quartz [1,6]. The chemistry occurring within the reaction zone can be influenced by changing the electrical properties of a gas-phase system within DBD reactor [4,72]. Since the DBD allows the direct insertion of a dielectric material into the discharge gap, it has attracted considerable interest for combining the plasma-chemical activation of reactants with heterogeneous catalysis under atmospheric pressure and relatively low temperatures. In this system, the catalyst acts as a dielectric material. In the DBD most of the discharge energy is used to produce and accelerate electrons, which generate highly active species (metastable, radicals and ions) [26,27,34,39]. The combined action of catalysts and a non-equilibrium gas discharge led to an alternative method to syngas and hydrocarbons production from CH 4 and CO 2 [26]. Methane and carbon dioxide were chemically activated directly by electron collisions. Liu et al. [45] concluded that the electronic properties of catalysts have an important role in oxidative coupling of methane using DBD plasma reactor. The electronic properties and therefore catalytic properties can be expected to change if the catalyst is electrically charged. Plasma processes offer a unique way to induce gas phase reactions by electron collisions. However, the plasma processes seemed to be less selective than conventional catalytic processes. The conventional catalytic reactions on the other hand can give high selectivity, but they require a certain gas composition, an active catalyst, and high temperature condition (endothermic reaction). When the discharge gap was completely filled by the catalyst, the plasma caused ignition in the reactor and the reaction can proceed. In this case, significant overheating while operating the discharge may occur and it was more difficult to maintain a stable plasma condition [34]. In the catalysis promoted plasma, the catalyst