Advances in catalytic conversion of methane and carbon dioxide to highly valuable products

Two typical processes have been developed for the conversion of methane and carbon dioxide to higher valuable products: dry reforming of methane (DRM) and CO2‐oxidative coupling of methane (CO2‐OCM). Numerous articles reviewed progresses in either DRM or CO2‐OCM, but no one covers both processes. In this review article, we systematically evaluated progresses in both DRM and CO2‐OCM processes for conversion of methane and CO2 to highly valuable products. Critical issues, which are carbon deposition and high energy cost for DRM and the contradiction between large activity and high selectivity for CO2‐OCM, were emphasized. Strategies to develop effective catalysts were evaluated, including the enhancement of metal‐support interactions, the introduction of promotors, the formation of solid solutions, and the construction of core‐shell structures. Plasma and photocatalysis were discussed as new promising technologies for DRM and CO2‐OCM.


| INTRODUCTION
Methane and carbon dioxide, which are greenhouse gases, are the main components of natural gas. Methane is the most stable hydrocarbon due to its strong C-H bond (434 kJ/mol) and thus its activation needs a high temperature. There are various techniques to convert methane to more valuable chemicals and fuels, [1][2][3][4][5] which can be classified into three types of routes [6][7][8]  As an indirect approach for highly valuable hydrocarbon production from CH 4 , the first route can be followed by Fischer-Tropsch reaction (Equation 7). Although such a multi-step process has high product yield, the syngas production is expensive due to its high capital costs and it is therefore only economically viable if it is conducted on a large scale. The latter two ways can convert methane directly to other hydrocarbons.
(1) CH 4 + H 2 O → CO + 3H 2 ΔH 0 298K = +206 kJ mol −1 (2) CH 4 + CO 2 → 2CO + 2H 2 ΔH 0 298K = +247 kJ mol −1 Carbon dioxide, which is a linear molecule with a C-O bond strength of 532 kJ mol −1 , is very stable. Furthermore, the large positive value (394.6 kJ mol −1 ) of CO 2 formation Gibbs free energy contributes to its high inertness. Due to its being the most oxidized state of carbon, reduction is the only possible route for the conversion of CO 2 via one of the following four main methodologies [9][10][11][12] : (1) Using high-energy starting materials such as hydrogen, unsaturated compounds, small-membered ring compounds, and organometallics; (2) Choosing oxidized low-energy synthetic targets such as organic carbonates; (3) Shifting the equilibrium to the product side by removing a particular compound; and (4) Supplying physical energy such as light or electricity. So far, the catalytic hydrogenation of carbon dioxide to methanol and other oxygenates using hydrogen has attracted the most attention (Equation 8).
Conversion of CH 4 or CO 2 individually possesses relatively lower activation energy compared with activation of both CH 4 and CO 2 in the same reaction. However, considering methane and carbon dioxide co-exist in natural gas, the conversions of both gases have significant implication toward the utilization of natural gas. 13 The well-investigated reaction to convert these two reactants is CO 2 reforming of CH 4 (dry reforming of methane, DRM), which was thoroughly explored by Fischer and Tropsch in 1928 14 and was investigated as early as 1888. 6 Dry reforming of methane yielded a lower syngas ratio (H 2 / CO 1:1) (Equation 2), which is suitable for production of oxygenated chemicals 15 from Fischer-Tropsch synthesis. However, the activation of both C-H bond in CH 4 and C-O bond in CO 2 , which requires efficient catalysts, is a great challenge due to their high stabilities. [16][17][18][19] Catalysts used for DRM can be divided into two groups: earth-abundant transition metals and noble metals. 20 The majority of catalysts are based on Ni due to its high activity and low cost, 21 whereas these catalysts usually undergo deactivation processes mainly due to carbon deposition 22 ; Noble metals [23][24][25][26] have demonstrated much more resistance to carbon deposition than Ni catalysts, but are generally uneconomical. Furthermore, those catalysts could also be exploited to convert raw biogas (which mainly contains CH 4 and CO 2 ) into syngas. 27,28 However, more attention is needed to develop sulfur-resistant catalysts, because H 2 S is the main impurity in biogas. 29,30 Oxidative coupling of methane by CO 2 (CO 2 -OCM) is a direct process to convert both CH 4 and CO 2 to higher value hydrocarbons, particularly ethane and ethylene (Equations 9 and 10): In 1982, Keller et al. 31 published the first report on the production of C 2 hydrocarbons from CH 4 with oxygen over catalysts (metal oxides supported on Al 2 O 3 ) at atmospheric pressure and temperature range of 500-1000°C. O 2 as an oxidant unavoidably induces some gas-phase radical reactions, which complicate the optimization of catalyst and cause the limitation in C 2 yield due to the unavoidable, consecutive oxidation of the produced hydrocarbons to CO and finally CO 2 . 32 Replacing of oxygen with other oxidant is considered as a possible approach to prevent the sequential reaction of C 2 products in the gas phase. [33][34][35] Carbon dioxide, which is a soft oxidant to provide extra carbon atom for the methane conversion, has been explored as an oxidant for CO 2 -OCM reaction. [36][37][38][39][40] Early in 1988, Aika et al. found that oxidative coupling of CH 4 by O 2 is promoted by CO 2 as a reactant. They attributed it to decrease of free energy by conversion CO 2 to CO at 800°C. However, CO 2 -OCM reaction (Equations 9 and 10) is not thermodynamically favorable even at temperatures as high as 900°C. 41 One way to overcome this is to introduce the solid coreactants that undergo solid-phase transformations during CO 2 -OCM. It was reported that the selectivity of the methane oxidation is determined by the ability of the oxygen active species on the catalyst surface to make discrimination between a C-H bond in methane and a weaker C-H bond in the product(s). 42 The lattice oxygen of catalysts participates in the conversion of methane, and then the reduced catalyst needs to be reoxidized for maintaining the activity of the catalyst. Furthermore, CO 2 chemisorption requires suitable sites to form active oxygen species for recovering the reduced sites and activating methane. Therefore, it is important to develop efficient catalysts that should be capable not only of activating both CH 4 and CO 2 , but also of producing C 2 hydrocarbons selectively. 43 As shown in Table 1, one can see that multi-component catalysts with alkali metal oxide are more effective than mono-component ones. Comparing with conversional O 2 -OCM whose maximum C 2 yield ranges from 16% to 27% with selectivity 72%-82%, CO 2 -OCM exhibits a lower maximum C 2 yield (6%) but comparable selectivity. 7,42 Furthermore, to achieve higher methane conversion at low temperatures, nonconventional catalytic systems have also been reported, such as catalytic reactions in an electric field, [44][45][46] catalytic reactions with discharge, [47][48][49] and photocatalytic reactions. [50][51][52] In this article, we will review the progress in both DRM and CO 2 -OCM, with emphasis on (1) how to inhibit carbon deposition over catalysts for DRM and (2) how to increase the C 2 -yield of catalysts for CO 2 -OCM. Plasma and photocatalytic technologies, which were exploited as prospecting routines to reduce reaction temperature and increase activity and selectivity of catalysts for the two processes, will also be discussed.

METHANE
Deactivation of catalysts by coke formation, which origins from the CH 4 dissociation and/or CO 2 disproportionation, is a serious problem in DRM. Thermodynamic considerations 16,53 suggest operation at CO 2 /CH 4 ratios higher than 1 and high temperature to minimize carbon formation for DRM. However, from industrial viewpoint, it is desirable to operate at lower temperature and with a CO 2 /CH 4 ratio near unity. Such an operation requires a catalyst that kinetically inhibits the carbon formation under conditions that are thermodynamically favorable for carbon deposition. Carbon deposition in DRM occurs via two main possible pathways, methane decomposition (Equation 11), and Boudouard reaction (ie, CO disproportionation) (Equation 12), The first reaction, which is endothermic, is favored at high temperature, whereas the second one prefers lower temperature due to its exothermic feature. Since a high temperature is required for methane reforming (>800°C), Boudouard reaction would not be significant. 54 Therefore, the dissociation of methane, which may occur via four steps (CH 4 →*CH 3 →*CH 2 →*CH), constitutes the main within 5-10 minutes after introduction of feed gas, CO 2 as main product, H 2 O, and C 2 as minor ones; in steady state, CO and C 2 as main products with C 2 yield of 4%.

| 7
CAI And HU contribution to carbon deposition in DRM. [55][56][57][58] Carbon formation by CH 4 decomposition is a structure-sensitive reaction. 55,56 For example, the Ni (100) and Ni (110) surfaces are more active for the decomposition of CH 4 to carbon than the Ni (111) surface. Another factor affecting carbon deposition is the surface acidity of catalysts. 59 Carbon formation can be diminished or even suppressed when the active metal is supported on a metal oxide carrier with Lewis basicity. The strategy to inhibit carbon deposition is to tune surface structure of a catalyst and to decrease its surface acidity. 20 It was revealed that noble metal catalysts are more resistant to carbon deposition than non-noble metal catalysts, [60][61][62][63][64][65][66][67] namely, the amount of carbon deposited on metal catalysts decreased in the order Ni>>Rh>Ir=Ru>Pt≈Pd at 500°C and Ni>Pd=Rh>Ir>Pt>>Ru at 650°C. 68 Supports are often exploited to modify the surface structure and acidity of catalysts, leading to the increase in the resistance to carbon deposition. 20 The support with high Lewis basicity, such as La 2 O 3 and CeO 2 , can increase the ability of catalysts to chemisorb CO 2 , which can react with carbon to form CO, decreasing carbon formation. The interaction of metal and support would affect dispersion and size of metal particles. The higher dispersion and smaller size of catalyst particles can inhibit carbon deposition.

| Noble metal-based catalysts
Various supported noble metal catalysts have been explored for DRM with an activity order: Rh∼Ru>Ir>Pt>Pd for DRM. [69][70][71] It is generally recognized that CH 4 is activated on the metal of a catalysts and CO 2 on its support. However, Bitter et al. 72 observed that CH 4 activation occurs at the Pt-ZrO 2 interfacial sites instead of Pt alone. The acidic/basic property of support has direct effect on CO 2 activation process: CO 2 activates by formation of formates with the surface hydroxyls on acidic support (like Al 2 O 3 ) and on basic supports (like La 2 O 3 , CeO 2 ) by forming oxycarbonates. [73][74][75] The metal particle size or metal dispersion can also be affected by support. For example, Yokota et al. 76 showed that Rh dispersion increases in the order: TiO 2 <MgO<SiO 2 <MCM-41 < γ-Al 2 O 3 . Whang 77 found that a very small amount of Ru (0.13 wt%) supported on ZrO 2 -SiO 2 are very active and stable for DRM ( Figure 1). This happened because the strong interaction between Ru and ZrO 2 resulted in the highly dispersed Ru with small particle size of 1.4 nm. Different from other oxide supports, TiO 2 possesses reducibility. As a result, TiO 2 can maintain the metallic nature of Rh, but it masks the catalytically active metal due to partial reduction of TiO 2 to TiO X , which then forms a layer over Rh in a Rh/TiO 2 catalyst. 78,79 Noble metals on other types of supports have also been explored for DRM, including crystalline oxides (Pyrochlores, perovskite), 80-87 mesoporous-structure silicate, 88 ZnLaAlO 4 ,89 and NiO/MgO solid solution. 90 In addition, DRM performance over Pd, Ru, or Ag metal catalyst could be enhanced using membrane reactors, which can separate the products 8 | CAI And HU CO and H 2 from the reaction system to drive the reaction toward formation of CO and H 2 91-95 at a lower temperature.
Carbon deposition on noble metal catalysts was evaluated. Barbier et al. 96,97 found that the carbon deposition per Pt atom decreased with increasing Pt dispersion, which was supported by work of Spivey and his coworkers. 98 This can be explained by particle size effect of Pt, namely, the increase in Pt dispersion on a support could result in the small Pt particles, on which the assembly of carbon atoms to coke becomes difficult. The promoters also have a significant effect on carbon deposition. [99][100][101][102][103][104] For example, it was reported that the bimetallic Pt-Au/SiO 2 , Pt-Sn/SiO 2 , and Pt-Sn/ZrO 2 catalysts exhibited less carbon deposition during DRM than the corresponding monometallic platinum catalysts, probably because of the formation of alloys. 99 Substituting a noble metal into the crystalline structure of a thermally stable oxide (such as perovskites and pyrochlores with empirical formula of ABO 3 and A 2 B 2 O 7 ) was explored to improve catalytic performance for DRM. [80][81][82][83][84][85][86][87] In this strategy, the B sites of ABO 3 or A 2 B 2 O 7 , which are normally occupied by a transition metal, can be substituted by noble metal isomorphically, leading to a more efficient catalysts. Recently, Spivey and his coworkers explored noble metal-substituted lanthanum zirconate pyrochlore catalysts. [105][106][107][108][109][110][111] For example, they reported that a 1% Ru-substituted lanthanum strontium zirconate pyrochlore, La1 .97 Sr 0.03 Ru 0.05 Zr 1.95 O 7 (LSRuZ), exhibited higher resistance to carbon formation than 0.5% Ru/ Al 2 O 3 catalyst under all tested conditions. 110 Furthermore, they evaluated two pyrochlores (LRuZ and LPtZ), which were synthesized by isomorphically partially substituting Zr in the B-site of La 2 Zr 2 O 7 with Ru (2.00 wt%) and Pt (3.78 wt%), respectively. 106 It was found that LRuZ exhibited much lower activation energies than LPtZ. In addition, they demonstrated that Rh is even better than Ru and Pt to substitute B sites in both perovskites and pyrochlores, leading to excellent activity ( Figure 2) and high resistance to carbon formation for DRM. [108][109][110][111] This indicates that the three substitutes follows the order: Rh>Ru>Pt. Over the noble metal-substituted La 2 Zr 2 O 7 catalysts, noble metal is responsible for CH 4 activation and dissociation, while CO 2 activation takes place on La site by the formation of three polymorphs of La 2 O 2 CO 3 , the oxidation of surface could be occurring at the Rh-La interface. 109,111 Carbon formation decreases with increasing Rh loading. 106

| Supported Ni-based catalysts
Alumina is one of the most commonly used supports for nickel catalysts. 112 However, these acidic sites caused carbon formation, leading to deactivation. Therefore, various methods were employed to improve Ni/Al 2 O 3 . Firstly, smaller Ni particles over Al 2 O 3 were prepared to inhibit carbon deposition, 113,118-123 because carbon deposition occurs more easily on larger nickel particles than smaller ones. 124,125 For example, Shang 118 applied atomic layer deposition (ALD) to synthesize Ni/Al 2 O 3 with | 9 CAI And HU high dispersion of Ni, resulting in the stable conversion of 93% at 850°C. Chen and Ren 120 found that the carbon deposition over Ni/Al 2 O 3 was markedly suppressed when NiO and Al 2 O 3 have strong interaction, which increases the difficulty of reduction of Ni 2+ to Ni 0 . As a result, small nickel crystallites on the catalyst surface, which are smaller than the size necessary for carbon deposition, 121 were formed. Liu and coworkers reported that the glow discharge plasma-treated Ni/Al 2 O 3 catalyst exhibited an excellent anticoke property for DRM. 126 This happened because the plasma treated catalyst contains high concentration of close packed plane with improved Ni dispersion and enhanced Ni-alumina interaction. The change in Al 2 O 3 pore structure can also tune the catalytic properties of Ni/Al 2 O 3 , namely, 25 wt% Ni supported on mesoporous nanocrystalline Al 2 O 3 exhibited good catalytic performance in CH 4 dry reforming. 127 Secondly, various types of promoters were employed to modify Ni/Al 2 O 3 catalysts, including alkali, alkaline earth, transition metal, and rare earth metal oxides. The basic oxides of Na, K, Mg, and Ca were explored to decrease the acidity of Ni/Al 2 O 3 . 112,[128][129][130][131][132][133][134] The surface of nickel catalyst incorporating basic metal oxide is favored by CO 2 adsorption, leading to the increase in CO 2 adsorption that can enhance its reaction with carbon and thus inhibits carbon deposition. The addition of a second metal, such as Co, Cu, and Sn, can form less C-sensitive alloys. 133,135-141 Choi et al. examined the effect of Co, Cu, Zr, and Mn as promoters for the Ni/Al 2 O 3 catalyst. 135 They found that, in comparison with unmodified Ni/Al 2 O 3 catalyst, those modified with Co, Cu, and Zr exhibited slightly improved activities, whereas the Mn-promoted catalyst provided a remarkable reduction in coke deposition with only a small decrease in catalytic activity. Seok et al. demonstrated that a partial coverage of the nickel surface of Ni/Al 2 O 3 by MnO X promoted the adsorption of CO 2 , leading to the decrease in the carbon deposition. 136 Ni/Al 2 O 3 catalyst can also be improved by tin and ceria, leading to enhanced conversion and stability. 141 Noble metals (such as Pt, Ru, and Pd) can remarkably enhance both activity and stability due to their excellent ability to maintain Ni reduced. 139,140,142,143 Furthermore, small amount of rare earth metals in Ni/Al 2 O 3 exhibited promising promoting behaviors. For example, Slagtern et al. evaluated Ni/Ln/Al 2 O 3 (Ln=rare earth mixture) catalysts at 800°C and 1 atm, showing that the catalyst with a rare earth content of 10.7 wt% Ln is more active and stable than the unpromoted catalyst. [144][145][146][147][148][149][150][151][152][153] Promoting Al 2 O 3 by Mg can also improve Ni/Al 2 O 3 catalysts. As reported by Damyanova et al., 154 Ni/MgAl 2 O 4 catalyst showed much better performance for DRM than Ni/Al 2 O 3 . Furthermore, NiFe/ Mg x Al y O z catalyst is even better than Ni/ Mg x Al y O z (Figure 3). 155 The promoting effect of Fe was attributed to its dynamic redox reaction under dry reforming conditions, ensuring a close proximity of the carbon removal (FeO) and methane activation (Ni) sites. In addition, rare earth elements (Sc, Y, Ce, and Pr) exhibited promoting effect on Ni/ Mg x Al y O z catalysts, leading to the improved catalytic stability and coke resistance due to the enhanced surface basicity, abundant oxygen vacancies, superior redox properties, and highly dispersed Ni particles. 156 Besides Al 2 O 3 , other supports were also explored. As an inert support, SiO 2 has a weak interaction with Ni, leading to relatively weak metal-support interaction and are less stable and less active compared to mildly acidic (Al 2 O 3 ) and basic (La 2 O 3 , CeO 2 ) supports. [72][73][74][75] However, the high dispersion of Ni nanoparticles in nanochannels of ceriummodified silica aerogels or in mesoporous silica can improve the catalytic activity and thermal stability. 178,179 Ni/La 2 O 3 catalysts were widely evaluated for DRM. [180][181][182][183][184][185][186][187][188][189][190][191] Zhang   exhibited a higher activity and higher stability than Ni/Al 2 O 3 and Ni/CaO catalysts for DRM. The rate of reaction on Ni/ La 2 O 3 catalyst increased significantly with time on stream during the initial 2-5 hours of reaction, and then tended to remain unchanged for 100 operating hours. The high steady state reaction rate was attributed to the basic sites on La 2 O 3 which assisted in the activation of CO 2 and oxidation of surface carbon. The preparation method also affects the Ni/ La 2 O 3 catalysts. 182 4 and 75% conversion of CO 2 at 700°C even after 50 hours. 185 Zirconia-and CeO 2 -supported nickel catalysts were investigated for the CO 2 reforming reaction with emphasis on the stability of the catalysts. [192][193][194][195] It was found that a lower nickel loading (<2 wt%) is benefit for stability of Ni/ZrO 2 catalysts at temperatures between 720 and 780°C. The stabilities of Ni/ZrO 2 catalysts were also dependent on preparation methods. 196 Ni/ZrO 2 catalyst prepared from large Zr(OH) 4 particles deactivated rapidly. In contrast, a catalyst with high metal loading of nickel (27 wt%), obtained by impregnating ultrafine Zr(OH) 4 particles (6 nm) with nickel nitrate, exhibited high and stable activity for DRM without deactivation. The activity of this catalyst at 757°C with a CH 4 /CO 2 =1:1 molar feed rate of 24 000 mL (g catalyst) −1 h −1 did not change for 600 hours, but exhibited oscillations in the CH 4 conversion between 80 and 85%. The incorporation of Ni particles within the mesoporous zirconia was used to suppress the sintering of Ni particles and to enhance the metal-support interaction because of the formation of additional interfaces, resulting in excellent stability over 80 hours. 197,198 ZrO 2 with different morphologies may possess different lattice oxygen mobility and interactions between support and active metals, which can also affect the catalytic behavior of Ni/ZrO 2 . 199,200 CeO 2 nanorods (NR) and nanopolyhedra (NP) were compared as supports for Ni catalysts. Du et al. showed that the Ni/CeO 2 -NR catalysts possess larger catalytic activity and higher coke resistance for DRM than Ni/CeO 2 -NP. 195 This is because the predominantly exposed planes are the unusually reactive {110} and {100} planes for the CeO 2 -NR, but the stable {111} one for the CeO 2 -NP. The {110} and {100} planes with Ni particles could generate the strong metal-support interaction (SMSI), which can inhibit the sintering of Ni particles and thus reduce the deactivation.
Zhang and his coworkers exploited 2D layer materials as support for Ni-based catalysts. 201,202 They reported that Ni nanoparticles embedded on vacancy defects of hexagonal boron nitride nanosheets (Ni/h-BNNS) can optimize catalytic performance by taming two-dimensional (2D) interfacial electronic effects. Surface engineering for defects of Ni/h-BNNS catalyst can strongly influence metal-support interaction via electron donor/acceptor mechanisms and favor the adsorption and catalytic activation of CH 4 and CO 2 , leading to superior catalytic performance during 120-h durability test [ Figure 4]. This work highlights promotional mechanism of defect-modified interface. This mechanism was further supported by nickel nanoparticles on defected nanosheets of halloysites. 203 The catalyst exhibited good coke and sintering resistance performance in DRM, which was reflected by negligible loss of activity after a 20 hours. This was attributed to the strong interaction between the Ni nanoparticles and the support.

| 11
CAI And HU ZSM-5 and carbon were used as supports of Ni-based catalysts for DRM by Fan and his coworkers. 204,206 It was found that the addition of Ce to Ni/ZSM-5 could not only promote CH 4 decomposition for H 2 production, but also the gasification of deposited carbon with CO 2 . Furthermore, the dispersion of Ni particles could be improved by Ce. They also evaluated carbon as support for Ni catalysts. 205,206 The char was synthesized by pyrolysis of a long-flame coal at 1000°C and used as a support for Ni-based catalysts. 205 It was showed that Cr could play a role of promotor in Ni/Char catalysts due to its enhancement for CO 2 adsorption. Furthermore, the promotion effect is dependent on the introduction approach of Cr, namely, the catalyst prepared by co-impregnation of Ni and Cr exhibited larger activity than that by sequential impregnation.
Mesoporous MCM-41 was also explored as a support by Damyanova et al. 207 They found that the addition of a small amount of Pd to Ni-containing catalysts, which generated easily reducible nanostructured NiO particles, increased the activity and stability of the Ni-based catalysts. Furthermore, the PdNi catalyst with Ni/Si ratio of 0.3 reached the best performance. Hydroxyapatite-supported bimetallic Ni-Co catalysts were also evaluated for DRM. The selectivities of CO and H 2 over the catalyst reached 80%-90% with high stability. 208

| Core-shell structured Nibased catalysts
CeO 2 was explored as a support for Ni catalysts for DRM because of its excellent redox properties and the oxygen mobility, 209 revealing that the interaction of CeO 2 with Ni particles plays an important role. Li et al. found that the reduction of NiO/CeO 2 by H 2 in a temperature range of 500-700°C generated a strong binding between Ni and CeO 2 , which can inhibit the sintering of Ni particles (8.7-9.4 nm) ( Figure 5). 210 This happened because the high-temperature (≥600°C) reduction induced the encapsulation of Ni nanoparticles by a thin layer of reduced ceria support. The encapsulation effect can be exploited to improve the catalytic activity of Ni/CeO 2 and inhibits carbon deposition in DRM.
A NiCe@m-SiO 2 yolk-shell framework catalyst, in which the CeO 2 -modified Ni nanoparticles are the core and the mesoporous SiO 2 is the shell, was developed. 174 The yolkshell framework catalyst showed high catalytic activity and stability for DRM. Its special structure and large surface area might enhance its activity, and the confinement effect of the yolk-shell framework could contribute to the stabilization of the Ni nanoparticles. The modification of the catalyst by CeO 2 , which increased the active oxygen species and improved the dispersion of Ni nanoparticles, could also enhance the catalytic activity and suppressed the carbon deposition. Furthermore, NiCo@SiO 2 core-shell catalyst (ie, single NiCo alloy nanoparticle encapsulated by SiO 2 shell) was also synthesized with microemulsion technology. 211 The catalyst exhibited larger activity and selectivity and higher stability than the Ni@SiO 2 , the Co@SiO 2 core-shell catalyst, or the NiCo/SiO 2 supported catalyst ( Figure 6). The encapsulation of metal nanoparticles by SiO 2 shell could effectively inhibit the agglomeration of active sites. 211,212 Meanwhile, the enhanced activity of NiCo alloy catalyst could also diminish the surface carbon deposition.
The combination between SRM and DRM as the steam-CO 2 bireforming was employed to inhibit carbon deposition by the oxidation of C species with H 2 O. 55 Furthermore, the variation of CO 2 /H 2 O ratio can tune the produced H 2 /CO ratio from 0.91 to 3.0, covering a wide range of syngas composition for various downstream hydrocarbon synthesis. So far, efforts for the bireforming have been mainly focused on Ni-based catalysts. [214][215][216]

| Supported co-based catalysts
Co supported on metal oxides were explored as catalysts for DRM. 217,218 It was found that 20 wt% Co/ CeO 2 exhibited high conversions up to 87.6% for CO 2 and 79.5% for CH 4 at 750°C. 217 Furthermore, the combination of Co with Ni over Al x Mg y O z enhanced its activity and stability, 218 which is due to the high metal dispersion and the strong metal-support interaction as well as the synergic effect between Co and Ni.

| MgO-based solid-solution catalysts
MgO, which possesses high thermal stability and low cost, is an excellent catalyst support. The very high melting point (2852°C) of MgO can maintain its relatively large surface area at high temperatures compared to most oxides used as catalyst supports. Furthermore, basic metal oxide MgO has the same crystal structure as NiO (and CoO). As a result, the combination of MgO and NiO (or CoO) can easily form a solid-solution catalyst with a basic surface, [219][220][221] which is helpful in inhibiting carbon deposition. The basic surface increases CO 2 adsorption and thus reduces or inhibits carbon deposition. The reduction of NiO (or CoO) in the NiO-MgO (or CoO-MgO) solid solution is much more difficult than that of pure NiO (or CoO), which contributes to the formation of very small Ni particles to inhibit carbon deposition. 221 It is generally recognized that the reduction of a metal oxide is determined by its metal-oxygen bond strength. However, we demonstrated that the reduction of a metal oxide is strongly dependent on both metal-oxygen bond strength of the metal oxide and the metal-metal bond strength of its metal product. 222 Furthermore, it is revealed that a critical factor to control the reduction of NiO (or CoO) in the solid solution is the isolation effect that NiO (or CoO) is isolated by MgO, which inhibits the metal-metal bond formation during the reduction. 222 Several groups reported the excellent results of CO 2 reforming of methane in the presence of NiO-MgO and CoO-MgO solid-solution catalysts. In 1995, we invented a highly efficient NiO/MgO solid-solution catalyst for DRM, which was prepared by impregnation and was calcined at 800°C and atmosphere pressure. 223 The reduced solid-solution catalyst exhibited almost 100% conversion of CO 2 , >91% conversion of CH 4 , and >95% selectivity to CO and H 2 at 790°C and a very high space velocity of 60 000 mL (g catalyst) −1 h −1 with a CH 4 /CO 2 molar ratio of 1.0 (Figure 7). To the best of our knowledge, this would be the best performance among all reported results for DRM. In contrast to MgO, the other alkaline earth oxides (such as CaO, SrO, and BaO) were found to be poor supports for NiO probably due to sintering. The excellent catalytic performance of NiO/MgO was attributed to the formation of a solid solution. 222,223 Furthermore, the conversion, selectivity, and stability characteristic of NiO/ MgO solid-solution catalysts were found to be dependent on their composition, preparation conditions, and even the properties of MgO. 24,[224][225][226][227] For example, we found that high and stable CO yields (>95%) occurred with NiO/MgO catalysts having NiO contents between 9.2 and 28.6 wt%. 24 No activity was observed for a NiO content of 4.8 wt%. At the high NiO content of 50 wt%, the CO yield decreased from 91% to 53% after 40 hours, and the catalyst became black, because of carbon deposition. In contrast, the other NiO/MgO solidsolution catalyst maintained their initial color, and no carbon deposition was detected by TEM even after 120 hours of reaction. 219,220 This happened because too small amount of NiO in the NiO/MgO catalyst provided too-small numbers

OF METHANE
Conventional oxidative coupling of methane has been realized by adding oxygen as oxidant, which is called OCM. The OCM follows complicated homogeneous (gas-phase) and heterogeneous (surface-catalyzed) reaction pathways. In the presence of oxygen, solid oxide catalysts can form active surface oxygen species that selectively abstract one hydrogen from methane to release free methyl radicals(CH 3 ·)in the gas phase, which subsequently couples to form ethane. The reaction is followed by the dehydrogenation of ethane to form ethylene or by the irreversible formation of oxidation products (CO or CO 2 ). 7 Furthermore, it was reported that electrophilic lattice oxygen species and the facile filling of surface lattice oxygen vacancies by gas-phase oxygen are key factors to design efficient catalysts for the oxidative coupling of methane. 239 Aika and Nishiyama are the first researchers to introduce CO 2 into the catalytic oxidative coupling of CH 4 . 41 They combined Equation 5 with Equation 12 to decrease free energy loss and utilize the oxygen of CO 2 , leading to negative ∆G of the total reaction and 6.1% of theoretical yield for C 2 hydrocarbons. The following mechanism was proposed for the CO 2 -promoted CH 4 coupling with O 2 (Equations 13-16): Effective surface activation of CO 2 plays an important role in coupling of CH 4 . Isotope and kinetic evaluations revealed that the oxygen atoms of CO 2 participate the CH 4 coupling through the reverse shift reaction of CO 2 , and C 2 hydrocarbons can further react with the oxygen from CO 2 under conditions of high conversion. 240,241 If oxygen is replaced with CO 2 to realize CO 2 -oxidative coupling of methane (CO 2 -OCM), the conversion mechanism of methane involves following steps 242,243 : 1. oxygen-assisted breakage of a C-H bond of CH 4 on catalyst surface; 2. heterogeneous decomposition of CO 2 to CO and oxygen active species on catalyst surface; 3. homogeneous recombination of CH 3 · radicals released from the surface; 4. homogeneous oxidative or radical dehydrogenation of C 2 H 6 to C 2 H 4 .
In this mechanism, (a) and (b) are surface reactions for activation of CH 4 and CO 2 , in which produced O radical intermediate can attack CH 4 molecular to attain CH 3 · intermediate. Two CH 3 · radicals can combine to form C 2 H 6 (reaction c), followed by dehydrogenation to C 2 H 4 (reaction d) or adding another CH 3 · to generate a higher hydrocarbon. However, CH 4 activation may also include lattice oxygen of catalyst. 52,244 Therefore, CO 2 -OCM would have two kinds of oxidant: lattice oxygen and oxygen active species from CO 2 decomposition, both of them are involved in the surface reaction associated with active sites of catalyst. It was shown that the selectivity of the methane oxidation reaction is determined by the ability of the oxygen active species on the catalyst surface to discriminate between a C-H bond in methane and a weaker C-H bond in the product(s). 245 Those are strongly dependent on the properties of catalysts, which has promoted intensive research to explore efficient catalysts for CO 2 -OCM (Table 1).

| Mono-component catalysts for CO 2 -OCM
It is generally recognized that the activation of CO 2 requires electron transfer from the catalyst probably through an anionic CO 2precursor. 13 Although CO 2 is easily adsorbed on alkaline earth metals (such as CaO), it is unable to activate CO 2 over CaO because CaO cannot donate electrons. 246 However, CO 2 can be activated by ZnO due to its defects, which are generated by reducing Zn 2+ with electrons from CH 3 formed by splitting of methane. The existence of defect sites such as Zn 1+ or oxygen vacancy centers has been reported for ZnO, 247 which may be responsible for CO 2 activation. Nevertheless, C 2 selectivity is less than 5% and thus its yield is very low over ZnO alone. 248 This happened probably because ZnO is not efficient for CH 4 splitting. Asami et al. evaluated various unsupported metal oxides for C 2 production by CH 4 and CO 2 reaction at 850°C. 38 As shown in Figure 8, C 2 yield and selectivity vary with different metal oxide. Rare earth metal (yttrium, lanthanum, and samarium) catalysts exhibited high selectivity of about 30%. Although C 2 yield over manganese is comparable to that over yttrium, its C 2 selectivity is considerably lower than that of yttrium. They also examined catalytic effectiveness of other unsupported metal oxides (including alkaline earth, rare earth, and transition metals) for C 2 formation from CO 2 -OCM. [249][250][251][252] For CO 2 -OCM over Pr oxide catalyst, the unstable lattice oxygen may participate in C 2 formation through a redox mechanism. 252

| Multi-component catalysts
Various metal oxides were combined as binary catalysts to improve catalytic performance for CO 2 -OCM. The binary oxides of rare earth and alkaline earth metals constitute the best type of catalysts. CaO/CeO 2 mixed oxides, which form solid solutions with highly mobile lattice oxygen, are applied as model catalysts to evaluate the role of oxygen mobility in C 2 production. 253 In the CaO/CeO 2 catalyst, oxygen ion conductivity can be tuned by changing the CaO content, whereas electron conductivity is negligible compared to anion conductivity. 253 Although lower CH 4 conversion and C 2 selectivity were observed over CaO alone, the introduction of CaO to CeO 2 can remarkably increase C 2 selectivity. 254 Wang's group reported high C 2 yield of more than 5% with C 2 selectivity of 60%~70% for CaO-CeO 2 catalysts. 254 It was found that CO and H 2 are the main products in the conversion of CH 4 with the lattice oxygen of CaO-CeO 2 without CO 2 , while the presence of CO 2 induced the formation of C 2 hydrocarbons that increased with increasing partial pressure of CO 2 . They proposed that CO 2 adsorbs on Ca 2+ sites and subsequently yield active oxygen species on neighboring Ce 3+ or other metal sites, which work as an oxidant to convert CH 4 to C 2 hydrocarbons. The similar results were obtained by Litawa et al. 255 The synergistic effect of catalytic basicity and reducibility in CaO-MnO/CeO 2 on CO 2 -OCM were evaluated by Amin's group with CO 2 -TPD and H 2 -TPR. 40,242,256,257 It was demonstrated that the introduction of CaO into MnO/CeO 2 could increase the basicity, leading to enhancement in CO 2 adsorption. The proper amount of CO 2 adsorbed on the CaO-MnO/ CeO 2 catalyst may inhibit the redox reaction involving the lattice oxygen and promote the defect sites or oxygen vacancy on Mn 2.7+ (mainly due to Mn 3 O 4 ) and/or Ce 3+ . In this process, basic Ca 2+ sites are responsible for absorbing CO 2 , and then the Ce 3+ and Mn 2.7+ sites activate CO 2 to generate CO and active oxygen species (possibly O − ). The oxygen species can convert CH 4 to CH 3 · radical. Furthermore, they created Response Surface Methodology (RSM) to optimize process parameters and catalyst compositions for CO 2 -OCM, leading to the maximum selectivity of 82.62% for C 2 hydrocarbons over the CaO-MnO/CeO 2 (with 8.2 wt% CaO and 6.8 wt% MnO) at 1.9 ratio of CO 2 /CH 4 and reactor temperature of 807°C. 257 The solid-solution feature of CeO 2 /CaO catalyst may play a role in the above activation process. CaO can partially dissolve in CeO 2 to form a fluorite-type solid solution, creating a synergistic effect between CaO and CeO 2 on the C 2 selectivity. 43 Although the increase of P(CO 2 ) did not affect the C 2 selectivity over individual CaO or CeO 2 , F I G U R E 8 C 2 yield and selectivity with different metal oxide.
Reprinted with permission from [38]. Copyright © 1995 Published by Elsevier B.V. the selectivity increased with increasing P(CO 2 ) over CaO-CeO 2 catalyst. This indicates that CO 2 plays a crucial role in the selective formation of C 2 hydrocarbons over the catalyst. The formation of solid solution, which possesses neighboring Ca 2+ and Ce 3+ sites, is efficient for CO 2 adsorption and its subsequent activation. Furthermore, the incorporation of bivalent Ca 2+ cation into CeO 2 lattice generates defect sites, which promote redox reactions between Ce 4+ and Ce 3+ . This may be the reason why the formation of CaO-CeO 2 solid solution can increase in the rate of CH 4 conversion.
For CO 2 -OCM, Zn 2+ in ZnO can accept electrons from CH 3 formed by splitting of methane and the subsequent reduction of adsorbed oxygen to O 2 lattice oxygen by electron transfer from Zn 1+ or Zn. The existence of defect sites in ZnO (such as Zn 1+ or oxygen vacancy centers) may be responsible for CO 2 activation. 247 However, C 2 selectivity is less than 5% over ZnO alone. 248 To improve catalytic performance, ZnO 2 was combined with other metal oxides for more efficient catalysts. Chen's group prepared ZnO/La 2 O 3 by impregnating ZnO with lanthanum nitrate solution and evaluated its catalytic performance with a fixed-bed reactor. The ZnO/La 2 O 3 catalysts exhibited a high C 2 selectivity of 90%, which is larger than that of pure ZnO (7.6%) or La 2 O 3 (56.8%). 37 The interaction between ZnO and La 2 O 3 was proposed to generate synergic sites at La 2 O 3 -ZnO interfacial area, which are responsible for converting CH 4 and CO 2 into C 2 hydrocarbon. Furthermore, Grzybek et al. demonstrated that doping ZnO into NaOH-CaO catalyst can increase C 2 selectivity. 258 This was supported by experimental results from Wang's group, 248 namely, CaO/ZnO with a Ca/Zn ratio of 0.5 exhibited the highest C 2 selectivity (80%) and yield (4.3%) for CO 2 -OCM, which are higher than those of CaO/ CeO 2 . In the absence of CO 2 , ZnO was reduced to Zn after the reaction, which was proved by XRD patterns of deposit downstream on the wall of the reactor. The main products are CO and H 2 . In contrast, when CO 2 existed, XRD diffraction peaks remained almost unchanged and no deposit was detected. This indicates that the catalyst reduced by CH 4 is instantly reoxidized by CO 2 to form CO and the active oxygen species. Zinc sites associated with the reduced state or oxygen vacancies connected to the zinc sites are responsible for CO 2 activation. Raouf et al. employed CaO-ZnO catalyst as the secondary bed to combine the main bed of Li/MgO, leading to the increase in selectivity and yield of C 2 hydrocarbons at relatively low temperature. 259 The combination between CeO 2 and ZnO without alkaline earth metal oxide exhibited promising catalytic performance for CO 2 -OCM. He et al. prepared CeO 2 /ZnO nanocatalyst by a novel approach, in which homogeneous precipitation was combined with microemulsion. 260 CO 2 -TPD exhibited a larger peak of CO 2 desorption, implying that the nanocatalyst had a higher density of basic sites. ZnO could modulate the intensity of the surface basicity through adjusting the intensity of the strong basic sites and weak basic sites. The conversion of methane over the CeO 2 /ZnO nanocatalyst showed 83.6% selectivity and 4.76% yield for C 2 , which are higher than those over the CeO 2 /ZnO catalysts prepared by conventional method. Furthermore, the nanostructured CeO 2 /ZnO catalyst also exhibited a high stability. 261 Oxide catalysts of Mn and alkaline earth metals (Ca, Sr, and Ba) were explored for the coupling of CH 4 to C 2 hydrocarbons (C 2 H 6 and C 2 H 4 ) using CO 2 as oxidant. 40,262 The binary oxides showed different behaviors in different range of reaction temperature, which may be caused by different phase composition. The Ca-Mn oxide catalyst exhibited similar catalytic performances to those of other Ca-containing binary oxides at 840°C. However, reducing reaction temperature to 825°C caused rapid decrease in CH 4 conversion and C 2 selectivity. This was attributed to the phase transformation of Ca 0.48 Mn 0.52 O to CaCO 3 . Furthermore, in the temperature range of 700-900°C, the Sr-Mn oxide catalyst showed the highest C 2 selectivity, the Ca-Mn oxide catalyst the lowest, and the Ba-Mn oxide catalyst in between. The Sr-Mn oxide catalyst provided a high C 2 yield of 6.3% with C 2 selectivity of 64% at 900°C. It was found that SrMnO 2.5 and BaMnO 2.5 are the main phase compositions for Sr-Mn and Ba-Mn oxide catalysts in the reaction, respectively. These species with Mn 3+ probably catalyze the activation of CH 4 . In addition, MnO-SrCO 3 was prepared by simultaneously adding solutions of Mn(NO 3 ) 2 and Sr(NO 3 ) 2 with aqueous solution of K 2 CO 3 . 40 The catalyst showed the maximum C 2 yield of 5.1% with a C 2 selectivity of 68.1% at 900°C. The excellent performance of the MnO-SrCO 3 catalyst for the CH 4 -CO 2 reaction could be attributed to both the dissociation of SrCO 3 to form CO 2 and the formation of an Mn 3+ /Mn 2+ couple to activate CO 2 ·and thus to activate CH 4 . A possible mechanism for CH 4 and CO 2 to produce C 2 hydrocarbons over Mn-SrCO 3 was proposed by as follows:

CONVERSION OF CH 4 AND CO 2
Nonthermal plasma (NTP) is considered as a promising alternative process for the conversion of CH 4 and CO 2 due to its ability to initiate reactions at near ambient temperature. 263 NTP configurations, which have been exploited for dry reforming CAI And HU reaction, include gliding arc discharges, dielectric barrier discharge (DBD), corona discharges, and glow discharges. [264][265][266][267][268][269][270] The most promising one is DBD due to its high electron density and ability to produce highly reactive species at room temperature. Furthermore, to improve the selectivity to desired syngas, DBD would be combined with heterogeneous catalysts, such as Pd (Cu, Ni, Co, Mn, or Ag) supported on γ-Al 2 O 3 . [271][272][273][274][275] So far, most of the catalysts used for plasma-assisted DRM are Ni-based catalysts, especially Ni/Al 2 O 3 . However, the conversions of CH 4 and CO 2 in plasma-assisted DRM are usually lower than those from conventional DRM. [275][276][277][278] This means that the coupling of the plasma with the catalyst did not generate a plasma-catalytic synergy in NTP process. Nevertheless, Tu and Whitehead observed the plasma-catalyst synergetic effect on DRM, which is dependent on packing methods. 279 They found that, when the 10 wt% Ni/-Al 2 O 3 catalyst in flake form calcined at 300°C was partially packed in the plasma, the synergy of plasma-catalysis was clearly observed, leading to the doubled CH 4 conversion and H 2 yield (17.5%). This synergistic effect also contributes to a significant enhancement in the energy efficiency of feed gas conversion. This synergistic effect from the combination of low temperature plasma and solid catalyst can be attributed to both strong plasma-catalyst interactions and high activity of the Ni catalyst calcined at low temperature (300°C). A higher conversion was also obtained by plasma with NiO/Al 2 O 3 (or NiO-MnO 2 /Al 2 O 3 ) catalyst compared to plasma alone, whereas the best syngas ratio was achieved with plasma alone. 31 The increase in CO 2 and CH 4 conversions were obtained by plasma over Ni, Co, or Cu catalyst. 280 Furthermore, the plasma-assisted DRM is able to achieve coke free. 281 This may be attributed to highly reactive radicals which can easily react with carbon atoms to inhibit coke formation.
As shown above, previous work on DRM with NTPs mainly focused on syngas production, whereas very limited efforts were devoted to CO 2 -OCM for the synthesis of liquid fuels and chemicals. 282,283 This is because plasmacatalytic CO 2 -OCM systems possess low selectivity with various products, including syngas, higher hydrocarbons, and oxygenates. The use of NaX zeolite could inhibits the formation of the undesired solid carbonaceous species and increases the selectivity of hydrocarbons (ethane, ethylene, acetylene, propane, propene, butane et al. with syngas). 47 Higher CH 4 conversion and C 2 selectivity could be obtained by Pd/Al 2 O 3 catalyst. Furthermore, Na-ZSM-5 exhibited a selectivity for production of liquid aromatic hydrocarbons from CH 4 -CO 2 . The product distribution in solid catalystplasma system can also be adjusted by introducing third gas (such as O 2 and Ar) into CH 4 /CO 2 . 284 Very recently, Tu and coworkers reported the synthesis of liquid fuels and chemicals from CO 2 and CH 4 at room temperature and atmospheric pressure using a novel DBD plasma reactor with a water electrode. 279 For the DBD plasma process without a catalyst, a total liquid selectivity of 59.1% was achieved with selectivities of 33.7, 11.9, 11.9, and 1.6% for acetic acid, ethanol, methanol, and acetone, respectively (in Figure 9A). The selectivity was only about 20.0% for CO ( Figure 10B), and the conversion is 18.3% for CH 4 and 15.4% for CO 2 ( Figure 9C). Furthermore, catalysts were exploited for the plasma process to tune the production of different oxygenates under ambient conditions. Cu/γ-Al 2 O 3 catalyst packed in the DBD reactor enhanced the selectivity for acetic acid to 40.2%, which is higher than those of the plasma-only mode (about 34%) and the plasma reaction using γ-Al 2 O 3 only (20.2%). The major product was acetic acid regardless of the catalyst used, followed by methanol and ethanol ( Figure 9A). Formaldehyde (HCHO) was formed only when the noble metal catalysts were used in the plasma reaction, and the highest selectivity for HCHO was obtained with Pt/γ-Al 2 O 3 . The plasma processes with and without catalysts exhibited similar gaseous product distributions with H 2 , CO, and C 2 H 6 as major gaseous products ( Figure 9B). However, combining the plasma with the catalysts resulted in 10%-20% increase for H 2 selectivity, slight increase for C 2 H 6 , and negligible effect on CO production with exception for Cu/γ-Al 2 O 3 . Compared to the plasma-only mode, the conversion of CO 2 and CH 4 slightly decreased with packing catalysts. This phenomenon was attributed to the change in discharge behavior induced by the catalyst, which had a negative effect on the reaction.
The density functional theory (DFT) was used to evaluate the synthesis reaction mechanism of hydrocarbons from CH 4 and CO 2 under cold plasmas. 285 It was proposed that the main obstacle for the synthesis is the dissociation of CH 4 and CO 2 , the cold plasma can supply the necessary energy for this dissociation through the use of its energetic electrons without extra heating gas. The electrons, which are the main species to initiate the discharge reactions, can activate CH 4 directly into methyl radical CH 3 · without assistance of CO 2 (Equation 23). The methyl radical, which is the key species for the formation of hydrocarbons, can be further dissociated into CH 2 , CH, and even carbon. Dissociation of CO 2 can also be enhanced by electron attack (Equations 24 and 25), and oxygen species can terminate activating CH 4 . There are two possible ways for CO 2 dissociation, namely, one is via CO 2 - The conversion of CH 4 and CO 2 at low temperature is thermodynamically unfavorable. Furthermore, large energy requirement of dry reforming of methane (DRM) has obstructed its application. Solar energy would be a solution for the energy issue. Therefore, photocatalysis was exploited to break this thermodynamic limitation and to solve the issue of large energy requirement. 52

ANALYSIS
The economic analysis coupled with process simulation is called techno-economic analysis, and Aspen Plus is usually selected for simulating the reforming process. 295 Furthermore, the techno-economic analysis of DRM is often compared with mature steam reforming of methane (SRM) that has been commercialized for several decades. 296,297 Comparing to DRM, SRM needs two reformers and extra desulfurizer unit, leading to its higher cost. However, if biogas or landhill gas is used as CH 4 and CO 2 sources for DRM, desulfurization is also needed. 298 The furnace or combustion of natural gas can be used as heating units. 298,299 There are two types of reformer reactors: the conventional packed-bed (PBR) reactor and the membrane reactor (MR). The latter reactor can produce syngas with lower cost due to the absence of compressor and pressure swing adsorption for gas purification. 298 The running cost and the price of unit product can be valued by simulating process. For synthesis of methanol via syngas, DRM cost (1066 ∕MTmethanol)islessthanthatofSRM(1087 / MT methanol), indicating that DRM is competitive to SRM. For a smaller production scale, the total cost of DRM is around 51533 $/year with production of 7 m 3 /h H 2 using a conventional packed-bed reactor. 298 However, as an emerging technology for direct conversion of CH 4 and CO 2 to highly valuable hydrocarbons, CO 2 -OCM has not yet been evaluated by techno-economic analysis.

| CONCLUSION AND PROSPECTS
We discussed two routes to convert CH 4 and CO 2 into highervalued products: Dry reforming of methane (DRM) for syngas production and the CO 2 oxidative coupling of methane (CO 2 -OCM) for the formation of hydrocarbons. For DRM, there are two types of catalysts: noble metal and non-noble metals. Noble metal catalysts (such as Pt, Pd, Rh, and Ru) possess excellent catalytic activity and selectivity as well as high stability with less carbon deposition, but they are expensive with limited source. Non-noble metals (such as Ni and Co) also exhibited excellent activity and selectivity for DRM to produce syngas at low cost. However, those non-noble metal catalysts suffered deactivation due to carbon deposition. Therefore, intensive efforts were made to inhibit the carbon deposition for the non-noble metal catalysts by employing three strategies: (1) increasing the interactions between metals and supports, (2) introducing promoters, (3) forming solid solutions to create isolation effect, and (4) generating core-shell structures. The target was successfully achieved by several types of catalysts, among which NiO/MgO solid-solution catalysts would be the best.
Three types of catalysts were explored for the production of C 2 hydrocarbons from CO 2 -OCM: (1) the binary oxides of rare earth and alkaline earth metals, such as CaO/CeO 2 catalysts, (2) ZnO-based catalysts, such as La 2 O 3 /ZnO and CaO/ ZnO, and (3) the binary oxides of Mn and alkaline earth metals (Ca, Sr, and Ba), such as Ca/Mn oxide catalysts and Sr/ Mn oxide catalysts. Among them, the combination between the rare earth metal oxides and the alkaline earth metal oxides is the best. Furthermore, the increased attention was paid to the activation of CO 2 on the catalyst surface, because it plays a key role to increase C 2 selectivity. Therefore, most catalysts contain the component of alkali metal oxides to increase the basicity of catalysts for CO 2 adsorption. Future research is necessary to design and synthesize highly efficient catalysts, which must possess unique ability to activate the C-H bond of methane without breaking the weak C-H bond of hydrocarbon products.
The critical issue for DRM, which has limited its commercialization, is large energy requirement. Solar energy is considered as its solution, which requires an efficient PDRM. However, most catalysts explored for PDRM are not efficient, because they can use only UV light (about 4% of the total solar energy). Recently, a progress was made by combining Pt/blackTiO 2 catalyst with a light-diffuse-reflection-surface of a SiO 2 substrate, which created an efficient visible light PDRM at relatively low temperature (much lower than the temperatures used in conventional DRM). This would open a new direction for the development of efficient DRM approaches. The increase in photocatalyst stability for PDRM would attract great attention in the future research.
Several catalysts were developed for photocatalytic CO 2 oxidative coupling of methane (CO 2 -OCM) for hydrocarbons. However, their efficiency is quite low due to their feasibility only for UV light. The future research is required for the development of advanced catalysts, which can efficiently use visible light for photocatalytic CO 2 oxidative coupling of methane.
Plasma was exploited for the conversion of CH 4 and CO 2 at low temperature, and a broad distribution of products was obtained, including syngas and various hydrocarbons. Although the plasma-assisted processes could provide a high conversion rate, but a very low selectivity. This issue will stimulate future research to develop selectivity-controlled plasma processes for the conversion of CH 4 and CO 2 .