Indirect coal to liquid technologies

Applied Catalysis A: General 476 (2014) 158–174

Contents lists available at ScienceDirect

Applied Catalysis A: General journal homepage: www.elsevier.com/locate/apcata

Review

Indirect coal to liquid technologies Erlei Jin a , Yulong Zhang b , Leilei He a , H. Gordon Harris a , Botao Teng a , Maohong Fan a,c,∗ a

Department of Chemical and Petroleum Engineering, University of Wyoming, Laramie, WY 82071, United States Western Research Institute, Laramie, WY 82070, United States c School of Energy Resources, University of Wyoming, Laramie, WY 82071, United States b

a r t i c l e

i n f o

Article history: Received 13 December 2013 Received in revised form 24 February 2014 Accepted 26 February 2014 Available online 6 March 2014 Keywords: Coal liquefaction Fischer–Tropsch MTO Ethylene glycol Syngas

a b s t r a c t Indirect coal liquefaction has enormous potential applications. Increasingly, new synthetic technologies have been concentrating in this area, and a number of new large-scale indirect coal liquefaction plants have been set up during very recent years. Further, a large volume of papers on indirect coal liquefaction have been published over the last two decades, including those on Fischer–Tropsch synthesis, syngas to ethylene glycol, syngas to methanol, dimethyl ether as well as methanol to olefins. In this review, the recent literature of indirect liquefaction, including Fischer–Tropsch and syngas to chemicals, are summarized, with an emphasis on the reaction mechanisms, conditions and novel catalysts. © 2014 Elsevier B.V. All rights reserved.

Contents 1. 2.

3. 4. 5. 6.

7.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The development of catalysts for the Fischer–Tropsch process (FT process) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Selectivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Deactivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methanol to olefin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Syngas to methanol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Syngas to dimethyl ether . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Synthesis of ethylene glycol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1. Synthesis of dialkyl oxalate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2. DMO hydrogenation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Coal is a material that some people like because of their needs while others hate due to the various emissions resulting from its combustion [1–7]. To overcome the environmental challenges as associated with the conventional utilization approaches, people are

∗ Corresponding author at: Department of Chemical and Petroleum Engineering, University of Wyoming, Laramie, WY 82071, United States. Tel.: +1 3077665633. E-mail address: [email protected] (M. Fan). http://dx.doi.org/10.1016/j.apcata.2014.02.035 0926-860X/© 2014 Elsevier B.V. All rights reserved.

158 159 160 161 162 163 165 166 166 166 169 172 173 173

increasingly interested in using alternative approaches including gasification and liquefaction. Liquefaction can be direct or indirect. Indirect coal liquefaction (ICL) processes mainly include two important steps. In the first step, the coal is gasified and converted into hydrogen and carbon monoxide, also called as syngas. In the second step, the syngas is further synthesized into liquid fuel. Coal is the most abundant energy reserve in the world. According to statistics of the International Energy Agency (IEA), of the top 10 coal producers in 2011, China has the highest coal production – 3576 metric tons (Mt) (46%), whereas the United States produces 1004 Mt (13%). Meanwhile, world crude oil demand in

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

2012 was approximately 92.0 million barrels per day (mmb/d), slightly higher than the 2011 demand of 91.9 mmb/d; global crude oil production in 2013 (from January to May) was 75.87 mmb/d. Global demand for crude oil continues to rise, which means that improvements in technologies to produce liquid fuels from other sources would be highly beneficial. In addition, because of the abundance and low-price of coal, many countries still use it in large tonnages in traditional way. However, it should not be ignored that emissions of SOX /NOX , Hg, CO2 from coal combustion cause environmental problems. Researchers have been making great effort in recent years to mitigate these environmental concerns, and great progress has been achieved worldwide. Coal energy resources have been developed and commercialized through alternative utilization technologies, such as pyrolysis, gasification and liquefaction. Among these technologies, indirect liquefaction promises to be one of the most effective approaches to convert coal to fuel liquid. Synthetic fuels derived via indirect liquefaction can outperform fuels directly derived from crude oil or from direct liquefaction, with regard to air pollution, and greenhouse gas emissions and other environmental constraints. In contrast to direct liquefaction, two steps have to be developed in order to make indirect liquefaction possible. The first step is to break down the carbon-based raw material to form syngas. The second step is to catalytically produce hydrocarbon fuels and/or chemicals from syngas. Indirect liquefaction can be classified into two principal areas: (1) conversion of syngas to light hydrocarbon fuels via Fischer–Tropsch synthesis (FTS) and (2) conversion of syngas to oxygenates such as methanol, dimethyl ether (DME), ethylene glycol (EG) and so on. FTS, a gas to liquid technology, is one of the most important processes, which produces synthetic fuel and lubrication oil, mainly from coal, natural gas or biomass resources. Following its invention by Fischer and Tropsch in the 1920s, research has made great strides in adjusting and refining the process. The development of FTS has been greatly influenced by fluctuations in the price of global crude oil. In recent decade, due to global energy-deficiency and the demand for green energy, FTS has received wide recognition. Based on coal gasification to syngas technology, the integrated gasification combined cycle (IGCC) process has also attracted extensive attention, due to its high efficiency and favorable environmental performance. As important chemical intermediates and peak shaving fuels, methanol and DME are the top-priority products of the IGCC process. As a primary part of indirect liquefaction, coal to EG has also been attracting extensive attention in both academic and business circles in the past decades. Since indirect liquefaction has enormous potential applications, more and more new synthetic technologies have been concentrated in this area, and research on FTS [8–12], syngas to EG [13–25], syngas to DME [18,26–28] and methanol [29–32], as well as methanol to olefins (MTO) [33–38], have resulted in a large number of publications during the past couple of decades. However, most of the papers and reviews mainly focused on one specific detail of the range of subjects relevant to indirect liquefaction. In this review, the recent developments of indirect liquefaction, including FTS, syngas to EG, DME, methanol and MTO are summarized, with an emphasis on the reaction mechanisms, conditions and novel catalysts.

2. The development of catalysts for the Fischer–Tropsch process (FT process) FTS is one of the most important synthetic processes. This process, which provides an effective gas to liquid technology, produces a broad range of hydrocarbon products, which are converted to synthetic lubrication oil and synthetic fuel in subsequent refining process, mainly from coal, natural gas resources or biomass.

159

FTS is a collection of chemical reactions, and mainly produces synthetic linear hydrocarbons (alkane and alkene). It also comes with production of oxygenates and utilizes the water–gas shift reaction (WGS). The products from FTS have many outstanding, advantageous properties. FTS produces sulfur free, nitrogen free and few aromatic hydrocarbon components, and thus is which are environmentally friendly. FTS is also an important path for industrial materials manufacture, due to its production of chemicals with high value, especially waxes and light olefins [39]. FTS has a long history. After it was firstly invented by Fischer and Tropsch in the 1920s, researchers working on FTS have made great strides in adjusting and refining the process. Many prominent large-scale coal to liquid companies have been established, including Sasol South Africa plant (the world’s largest oil-from-coal plant) and Sasol Qatar plant [40]. The development of FTS has been greatly influenced as a result of the fluctuation in the price of global crude oil. In recently decades, due to global the energy-deficiency and the demand of green energy, FTS has been widely recognized as an alternative path to liquid fuels. Currently there are two FT operating modes. The high-temperature (HTFT, 300–350 ◦ C) process with iron-based catalysts is used for the production of gasoline and linear low molecular mass olefins [41]. The low-temperature (LTFT, 200–240 ◦ C) process, with either iron or cobalt catalysts, is used for the production of high molecular mass linear waxes. Only the metals Fe, Ni, Co and Ru have the required FT activity for commercial application. On a relative basis, taking the price of scrap iron as 1.0, the approximate cost of Ni is 250, of Co is 1000 and of Ru is 50,000 [42]. However, Ni produces too much CH4 while Ru has really high price and the amount available is insufficient for large-scale application. So, only Fe and Co can be used as viable catalysts. Various types of reactors have been developed for the FTS process, such as fixed-bed (FBR), slurry bubble column reactor (SBCR, or CSTR in bench scale testing) and fluidized-bed reactor [43]. However, the selection of type of product in FTS is still the one of the most important issues. Chain growth in FTS follows the principles of stepwise polymerization and the product distribution of hydrocarbons follows an Anderson–Schulz–Flory distribution [44]. But the selectivity for methane and heavy hydrocarbon is higher than the selectivity for gasoline and diesel (C5–11 and C12–20 ). For maximum gasoline production the best option is high capacity fixed fluidized bed (FFB) reactors operating at about 340 ◦ C, with an iron catalyst. This produces about 40% straight run gasoline. Twenty percent of the FT product is propene and butene [45]. These can be oligomerized to gasoline and because the oligomers are highly branched it has a high octane value. The straight run gasoline, however, has a low octane value because of its high linearity and low aromatic content. The C5 /C6 cut needs to be hydrogenated and isomerized and the C7 –C10 cut requires severe platinum reforming to increase the octane value of these two cuts [46]. Mild hydrocracking of wax was investigated at the Sasol R&D division during the 1970s [47]. The product heavier than diesel was recycled to extinction. The overall yields were about 80% diesel, 15% naphtha and 5% C1 –C4 gas. When the decision to construct the third Sasol plant was made, the wax hydrocracking proposal was rejected because at that time making gasoline was the more economic option, and the straight duplication of the second plant resulted in huge savings in time and capital. Also at that stage, the FT slurry reactors had not yet been developed. About 20 years later the same concept of wax hydrocracking was implemented at the Shell Bintulu plant where multi-tubular FT reactors are used and currently Sasol/Chevron are designing a slurry FT plant with wax hydrocracking in Nigeria [48]. The high-temperature fluidized bed FT reactors with iron catalyst are ideal for the production of large amounts of linear-olefins. As petrochemicals they sell at much higher prices than fuels. The olefin content of the C3 , C5 –C12 and C13 –C18 cuts are typically

160

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

Fig. 2. The reaction energy diagram of CO insertion into adsorbed CH2 and subsequent CO dissociation [11].

Fig. 1. FT stepwise growth process. Note that no specific chemical mechanism is implied in the sequence presented [48].

85%, 70% and 60%, respectively [49]. Ethylene goes to the production of polyethylene, polyvinylchloride, etc. and propylene to polypropylene, acrylonitrile, etc. The extracted and purified C5 –C8 linear-olefins are used as co-monomers in polyethylene production. The longer chain olefins can be converted to linear alcohols by hydroformylation [50]. The only required purification of the narrow feed cuts is the removal of the acids. The hydroformylation was investigated at the Sasol R&D laboratories in the early 1990s. The alcohols are used in the production of biodegradable detergents. Their selling prices are about six times higher than that of fuel. The LTFT processes produce predominantly longer chain linear paraffins. After mild hydro-treatment to convert olefins and oxygenates to paraffins the linear oils and various grades of linear waxes are sold at high prices [51]. 2.1. Selectivity FT synthesis always produces a wide range of olefins, paraffins and oxygenated products such as alcohols, aldehydes, acids and ketones regardless of reaction conditions. Variables such as temperature, feed gas composition, pressure, catalyst type and promoters will influence the selectivity of products. Fig. 1 illustrates the relationship between the CH4 selectivity and that of some selected hydrocarbon cuts [48]. The explanation of these interrelationships lies in the stepwise growth process occurring on the catalyst surface. The CH2 is formed firstly by the hydrogenation of CO and plays the role a monomer in a stepwise chain propagation reaction. At each stage of growth the adsorbed hydrocarbon species have two possibilities of, (i) being hydrogenated to form the products or, (ii) adding another monomer to continue the chain growth. If the chain growth (˛) is independent of the chain length then it is really easy to calculate the product distribution with various values of ˛.

However, the stepwise growth process illustrated in Fig. 1 might not represent the actual FT mechanism. A lot of other mechanisms have been proposed and this matter still remains controversial [52–55]. Zhuo et al. [11,56] have computationally studied reaction intermediates and activation energies of the corresponding elementary reaction steps on the surface of Co catalyst (Fig. 2). Fig. 2 shows that formyl intermediate (HCO) is really unstable and the chain growth step may occur through CO insertion. More possibilities have been researched includes: (i) whether the CO molecule first dissociate into C and O be the hydrogenation of C to CH2 monomers or not, (ii) whether CO hydrogenated to “CHO” or “HCOH” then insert into the growing chain or not and (iii) whether CO hydrogenated after it inserted into the chain or not [52]. To improve the selectivity of FT process, better catalyst and optimized process conditions are the two important aspects. Generally, increasing of the operating temperature results in a shift in selectivity toward lower carbon number products and to more methane. At the same time, the degree of branching increases, and the amount of secondary products formed such as ketones and aromatics, also increases [42]. These shifts are proportional with thermodynamic expectations and the relative stability of the products. For Co catalyst, the CH4 selectivity rises more rapidly with increasing temperature than it does with Fe catalysts due to Co is a more active hydrogenating catalyst. Promoters also play important roles in varying the selectivity. The main function of alkali metals is through an electronic effect, which means these promoters may change the electronic properties of Fe-based catalysts, as well as modifying the adsorption pattern of reactants (H2 and CO) on the active sites. Alkali on iron catalysts could increase 1-alkene selectivity, reaction rate, growth probability of hydrocarbon chains and also decrease the yield of CH4 . Alkali promoters in iron-based catalysts could also enhance CO conversion with addition of external water, and cause an increase of both the average carbon number of synthesized hydrocarbons and the 1-alkene selectivity [57,58]. Many researchers have found the impact of Group I alkali metals (Li, Na, K, Rb and Cs) modified iron catalysts through different synthesis conditions and at the same carbon monoxide conversion levels [59,60]. Promoters have fewer effects on Co catalysts. The most used promoters for Co are noble metals, transition metals and rare earth oxides. Many studies indicate that introducing of proper promoters many enhance the adsorption of CO as well as the chain growth [61–63].

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

161

2.2. Catalysts Compared with other catalysts, Fe-based catalysts have lower cost and could increase the production of large amounts of olefins, paraffins and oxygenates and work over a wide range of temperature [49]. In addition, high temperature FT synthesis based on iron-based catalysts allows using a synthesis gas of hydrogendeficient (low H2 /CO ratio) which could be produced from coal gasification, due to their significant strong activity of the WGS reaction [64,65]. In order to obtain excellent performances of ironbased catalysts, which means improving their stability, activity and selectivity properties, a series of promoters/modifiers have been introduced to the catalysts. These include electronic promoters (e.g. K and Ru) and structural promoters, such as silica and alumina [64,66]. Typical promoters are alkali metals, alkaline earth metals, rare earth metals (mainly the Lanthanoids) and transition metals. Ngantsoue-Hoc et al. reported that with the following synthetic conditions: (i) ratio of syngas H2 /CO of 0.67, (ii) a pressure of 175 psig, and (iii) reaction temperature 543 K, Li has the same effectiveness as K at low conversion levels and the lowest promotion at high conversion levels, compared with other alkali metals promoters (K, Na, Rb and Cs), due to the presence of water in the reactor. They found that catalytic activity of these promoters is as shown in Fig. 3, with same space-velocities of 10 nl/h g [67]. It is well known that K has a significant effect of promotion on the FTS performance of catalysts. K as promoter could largely retard the reduction process and suppress H2 adsorption, while enhancing the CO adsorption and boosting the carburization [59]. Like other alkali metals, K can bring to the catalysts high surface basicity, while the surface basicity is suppressed by the acidic sites of structural promoters. Due to different interactions between K and structural promoters (mainly the SiO2 and Al2 O3 ), K modified iron based catalysts have varying performance for the production of synthetic liquid oil during FT process. Zhang et al. [59] studied the FT synthesis performance of K promoted iron-based catalysts on dif-

Fig. 3. CO conversion vs. alkalinity [67].

ferent acidic structural promoters (Al2 O3 , SiO2 and ZSM-5), they found that the addition of potassium have an obvious influence on the formation of iron carbides, and Fe/K–Al2 O3 has the highest CO convention while Fe/K–ZSM-5, which shows the highest H2 conversion relatively, compared with other catalysts at the same reaction conditions. Addition of alkali metals (or oxides) has some disadvantages. Alkali metals could easily escape from catalyst surfaces due to their lower melting points and high mobility [68]. This would result in the alkali metals modified catalysts having poor stabilities during the FT process. Compared with alkali metals, alkaline-earth metals have similar alkaline properties while their oxides have some attributes (such as higher melting points) similar to silica and alumina [64,66]. Li et al. [66] reported alkaline-earth metals could strengthen Fe O bonds of ␣-Fe2 O3 in fresh catalysts, and meanwhile decrease H2 adsorption but improve the adsorption of CO. They also found alkaline-earth metals played an important role in shifting the FTS product distribution to heavy hydrocarbons.

Fig. 4. The influence of cobalt particle size on (a) activity normalized to the cobalt loading; (b) surface-specific activities; (c) methane selectivity and (d) hydrogenation [72].

162

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

Fig. 5. TEM micrographs of Ru particles of 4Ru/SBA-16-4CA (a), 4Ru/SBA-16-20CA (b), and 4Ru/SBA-16-40CA (c), and corresponding size distributions estimated from TEM photographs. Insets are images of Ru particles after removal of the SBA-16 [73].

Zhang et al. used copper as a promoter of Fe catalyst [69]. Cu improves the rate of catalyst activation and shortens the induction period, whereas the addition of Cu has no apparent influence on the steady-state activity of the catalyst. Promotion of Cu strongly affects hydrocarbon selectivity. The product distribution shifts to heavy hydrocarbons, and the olefin/paraffin ratio is enhanced on the catalyst due to the indirect enhancement of surface basicity by the copper promotion effect. Wang et al. [70] found that the use of precipitation method in the Fe/Mn catalyst can significantly improve the selectivity of alkene (CO conversion was 96.3%, selectivity of C2–4 up to 52.1%). They suggested that samples with high yield of light alkenes have better reduction and carburization properties due to the modification of adding manganese by coprecipitation, which facilitates the oxidized active sites to be regenerated in the syngas atmosphere. Therefore, the catalyst surface could have a sufficient amount of active sites to adsorb and decompose CO. Alkaline earth metals, magnesium, was also developed as the Fe catalysts promoter. Yang et al. [65] added Mg in the Fe/Cu/K/SiO2 catalyst system, which resulted in the selectivity of CH4 only 8% while the selectivity of olefin 83%. Cobalt-based catalysts are only used in the LTFT process, as at the higher temperatures excess CH4 is produced. Because of the high price of Co it is desirable to minimize the amount used, but maximize the available surface area of the metal. To achieve this, Co is dispersed on high area stable supports such as Al2 O3 , SiO2 or TiO2 . Typically the cobalt metal loadings vary from 10 to 30 g per 100 g of support [71]. The cobalt particle size will also influence its selectivity and activity. Fig. 4 shows that both selectivity and activity changed for catalysts with smaller particles [72]. At 35 bar, the turnover frequency TOF decreased from 23 × 10−3 to 1.4 × 10−3 s−1 , while the C5+ selectivity decreased from 85 to 51 wt% when the cobalt particle size was reduced from 6 to 2.6 nm. When cobalt particles size larger than 6 nm, there was no much difference found.

The cobalt catalysts are usually promoted with a small amount of noble metals, transition metals and rare earth oxides, which will increase both the reaction rate and selectivity. Ru was reported to promote cobalt oxide nanoparticles on ␥-Al2 O3 supported catalysts [61]. The catalysts with 0.05% amount of Ru showed excellent results, however, when the concentration of Ru increased (larger than 0.1%), the CO conversion decreased. When Ru used as catalyst, its particle size would influence the both the activity and selectivity. Chen et al. [73] found the relationship between the Ru size and its reaction activity. Fig. 5 shows TEM images of different Ru particles on the surface of the support. The selectivity to C5+ increased, while the selectivity to methane and C2–4 light hydrocarbons decrease with Ru average particles size from 2.0 nm to 9.3 nm. The Ru catalyst confined in the SBA-16 with average nanoparticle size of 5.3 nm gives the best activity. Martinez et al. [74] reported Re as a promoter for Co catalyst. Although the activity of Co will not change too much, trace mount of Re helped the reduction of Co and increased CO conversion. Re also decreased the selectivity of CH4 because it could increase the density of Co on the support surface. Storsaeter et al. [75] also found that 0.5% of Re could accelerate the reaction rate as well as enhance the alkane selectivity which could reach up to 91.4%. Pt was reported as the Co catalysts promoter, which could improve the CO conversion but decrease the alkane selectivity. Rare earth oxides and transition metals were also reported as Co catalysts promoters. They will either improve the dispersion of Co or influence its electronic properties to change the reaction activity and selectivity. 2.3. Deactivation It is well known that iron will be oxidized at much lower H2 O/H2 ratios than that of cobalt so that the iron surface will be more easily occupied by oxygen atoms which results in a low conversion. Therefore, for iron-based catalysts, to achieve high conversions

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

(higher than 90%) two stage operations, together with gas recycling, are needed, which increases both capital and operating costs [42]. Deactivation of cobalt based catalysts due to oxidation during FT synthesis has been studied extensively [76–80]. Some researchers reported that oxidation was an issue; while others claimed that there was no oxidation occurred during FT synthesis. By taking the surface energy contribution of cobalt metal and cobalt oxides into account, van Steen et al. [81] reported that oxidation of small cobalt crystallites (<4.5 nm) is possible during FT synthesis. The reasons for deactivation of iron based FT catalysts include poisoning (particularly, sulphur under commercial application conditions), sintering (loss of surface area), oxidation, coking, fouling, attrition and so on. The deactivation mechanism would be different with different catalysts and reactors. A lot of excellent review articles focused on the deactivation mechanisms have been published during the past dozens of years [82–87]. Poisoning, sintering, oxidation, coking are due to the chemical failure while fouling and attrition are due to mechanical failure. Sulphur is the main poison agent for FT reactions. Fischer recommended a maximum sulphur content in the feed gas of 2–4 ppm while Dry recommended a maximum sulphur content of about 0.2 ppm for commercial considerations [88]. Sulphur poisoning will not affect the products selectivities for iron based catalysts, however, when severe poisoning occurs, the selectivity of acids will be increased. Sintering is related to the partial pressure of water, while the formation of carbonaceous species has been shown to play only a minor role in the deactivation process [89]. Reoxidation of the reduced precipitated iron based catalysts will decrease the surface area as well as change the selectivity. Carbon deposition and coke formation on metal catalysts from carbon monoxide and hydrocarbons are another main reason for deactivation of iron based FT catalysts. Carbon is typically a product of CO disproportionation while coke is produced by decomposition or condensation of hydrocarbons on catalyst surfaces and typically consists of polymerized heavy hydrocarbons. Different kinds of polymeric carbon and coke which vary in morphology and reactivity are formed in these reactions and encapsulate the metal surfaces causing the deactivation [83]. Fouling is due to the physical deposition of species from the fluid phase onto the catalyst surface, which results in activity loss due to blockage of sites and/or pores. Although cobalt based catalysts are more resistant to oxidation, they are also poisoned by sulphur. Except for this, the oxidation of smaller cobalt particles on the surface or inside the support, and the reaction between cobalt and the support which formed noncatalytical species are the main reasons of cobalt based catalysts deactivation. Furthermore, formation of carbon deposits, aggregation of cobalt particles and the loss of catalyst by formation of volatile carbonyl may also cause catalysts deactivation. Some metal oxide promoters, such as ThO2 , ZrO2 and La2 O3 , were reported to improve stability and activity of cobalt catalysts [90–92]. Those promoters could enhance the dispersion of catalysts or form active sites on the support promoting CO dissociation, thus improve the activity.

3. Methanol to olefin The conversion of methanol to olefins tends to be an interesting and promising way of converting methane to chemicals. Solid acids can effectively catalyze this reaction to form hydrocarbons. However, the selective production of light olefins, especially ethylene and propylene, is a challenge to catalysis. Before the year of 1990, most of the literatures discussed the zeolite ZSM-5 as an appropriate catalyst, which possesses a ten-member ring, interconnected channel system and yields branched hydrocarbons and aromatics as by-products [93]. ZSM-5, medium-pore zeolite composed

163

˚ and sinusoidal (5.1 A˚ × 5.5 A) ˚ of intersecting straight (5.3 A˚ × 5.6 A) 10-ring channels, was originally developed as a methanol-togasoline (MTG) catalyst by Mobil, then has also been studied as a potential MTO catalyst [94]. It was believed to be one of the most suitable catalysts for industrial application of MTO process due to its high resistance to deactivation by coke deposition [95]. However, ZSM-5 does not have high enough selectivity for MTO due to its broad pores. Then, researchers have focused on small pore silicoaluminophosphate (SAPO) molecular sieves, mainly on SAPO34, which gives a narrow range product distribution from C1 to C5 hydrocarbons. The small pore limitations of SAPO molecular sieves result in a rather high ethylene selectivity [96]. This progress greatly promoted the industrialization process of MTO. SAPO-34, a microporous silicoaluminophosphate with chabazite structure (CHA) discovered by Lok et al. [97], is now demonstrated as one of the best catalysts for an MTO reaction. The small pore of SAPO molecular sieves result in a high ethylene selectivity, however, for the same reason it is difficult to apply conventional modification methods, such as ion-exchange and impregnation, to change the properties of the catalyst. Modification in the synthesis of the molecular sieve seems to be critical for improvement of catalytic performance. Therefore, many efforts have been focused on improving its catalytic performance by various physicochemical modifications. The adjustments of shape selectivity as well as acidity are two key methods which can modify the reactive performance of such acidic crystalline material [98]. Yang et al. [99] investigated the size effect of SAPO-34 on the catalytic performance of MTO reactions. Four SAPO-34 catalysts with different sizes from 20-nm to 8 ␮m have been studied and the nanosized catalysts, especially the sheetlike SAPO-34 catalyst with 20 nm thickness, exhibited the longest catalyst lifetime and lowest coking rate in MTO reactions. Dai et al. [100] also reported the size effect of SAPO-34 on the methanol conversion and product selectivity. Fig. 6 shows SEM images of SAPO-34 with different sizes. It was obvious that the crystal size had a great impact on the lifetime and selectivity of the MTO catalysts. For SAPO-34-S (2.5 ␮m), a methanol conversion of 100% and a selectivity to light olefins of more than 90% were obtained up to 10 h while these two numbers decreased to 40% and less than 5%, respectively, within 30 min for SAPO-34-L (20 ␮m). Wu et al. [101] reported the influence of morphology of SAPO-34 on the catalytic performance. SAPO-34 catalysts with sheet-like and cubic morphology were synthesized. The sheet-like catalyst exhibited a longer stability life and much higher light olefins selectivity than the cubic catalyst. Hierarchical cross-like SAPO-34 catalysts with enriched multi-porosity (mesopore range: 10–50 nm) were developed and achieved dramatic light olefins selectivity (96%) and slow deactivation rates [102]. This may due to the large number of mesopores of the catalysts and so the reactants and the products can rapidly diffuse into and out of the thin nanocrystals, which greatly enhances the intracrystalline diffusion. Due to the fast deactivation of this catalyst, acidity adjustment was considered to be an efficient method for decreasing the rate of deactivation of SAPO-34. The acidity of SAPO molecular sieves could be varied through isomorphous substitution of cations, cation exchange, and adjustment of the Si/Al ratio [103]. The acidity adjustment through the Si/Al ratio of SAPO may be achieved in an easier and less expensive manner. Zhang et al. [104] reported that improved light olefins yield through increasing the ratio of Si/Al and introducing cobalt into catalysts in the meantime. Light olefins yield was increased to some extent and smaller amounts of benzenes were produced using Co-SAPO-34 over SAPO-34 and therefore slow down the production of coke. Ni and Mg were also used to modify SAPO-34 thus improved its performance for conversion of methanol to olefin [105]. The incorporation of Ni and Mg in

164

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

Fig. 6. Upper panel: SEM images of the SAPO-34-S (left), SAPO-34-M (middle), and SAPO-34-L (right) materials. Lower panel: Methanol conversion and product selectivities during the MTO conversion over the SAPO-34 catalysts under study at 673 K [100].

the framework of ASPO-34 exhibited higher light olefin production as well as longer lifetime. Fig. 7 shows the yield of products over SAPO-34, NiAPSO-34 and MgAPSO-34 catalysts at 400 ◦ C. The yield of light olefins decreased in the order of NiAPSO-34 > MgAPSO34 > SAPO-34. The mechanism for methanol conversion to olefins over zeolite is pretty complicated. Several classical mechanisms for MTO were shown in Fig. 8 [106]. Those mechanisms have been studied for many years. The first proposal envisioned that methanol might be activated on the catalyst to form a carbenium ion, CH3 + , or carbene,:CH2 as shown in Fig. 8a and b. Several papers have reported the similar mechanism [35,36,107]. Another possibility of the mechanism is that new species were formed through radical routes (Fig. 8c). CH2 free radical might be initiated first then formed carbon–carbon bond. Methanol can react with zeolite acid sites to form framework bound methoxy (methoxonium) species

Fig. 7. Yield of products over SAPO-34, NiAPSO-34 and MgAPSO-34 catalysts at 400 ◦ C [105].

Fig. 8. Several of the classical “direct” mechanisms for the conversion of methanol/dimethyl ether to olefins (or a direct precursor) shown in highly abbreviated or conflated form: (a) A pathway showing a carbenium ion alkylating dimethyl ether to form a carbonium ion. (b) One of several proposed carbene pathways. (c) An abbreviation of one of several free radical routes. (d) An alkoxy chain growth process occurring on a framework site. (e) One of the proposed mechanisms featuring CO, showing a role for transition-metal impurities. (f) An abbreviated oxonium-ylide route [106].

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

165

[108], but the alkoxy chain growth process in Fig. 8d has never been observed. Since CO is always observed in MTO products and iron is always an impurity in many catalysts, CO and H2 may react on iron or other transition metal based catalysts to make linear alkanes (Fig. 8e). However, NMR studies showed that the reaction rate was invariant with respect to added CO and CO was considered to be either an intermediate or a cocatalyst in methanol conversion [109]. Oxonium-ylide route (Fig. 8f) has also been studied and detected by NMR and no evidence showed that more oxonium could accelerate the MTO reaction [110].

4. Syngas to methanol The production of clean liquid fuels from coal-based syngas is one of the key steps for future clean coal utilization. Based on coal gasification to syngas technology, the integrated gasification combined cycle (IGCC) process attracts extensive attention due to its high efficiency and friendly environmental performance [30,33,111–114]. As an important chemical intermediate a fuel, methanol is the top-priority product of the IGCC process [115,116]. Originally methanol was produced by distilling off the small amount of methanol in wood. Alwin Mittasch, who worked in Badische Anilin und Soda Fabrik, and following his discovery of the Fe/K/Al2 O3 catalyst for ammonia synthesis, found that methanol could be produced from CO/H2 mixtures over a ZnO catalyst [117]. Imperial Chemical Industries (ICI) made a significant progress in 1996. They patented a co-precipitation method for the preparation of a stable, highly active Cu/ZnO/Al2 O3 catalyst for the synthesis of methanol with syngas [118]. This new catalyst lowered the operating temperature from 400 to 230 ◦ C and the operating pressure from 200 bar to as low as 50 bar. However, at that time, CO and H2 were convinced of the reactants and CO2 was not used as the reactant. After failure of separating CO2 from CO/H2 mixture, CO2 was accidentally added into the mixture and used as a reactant, thus a large progress in methanol synthesis was achieved. At present, synthesis of methanol from syngas is carried out on traditional CuO–ZnO–Al2 O3 (CZA) catalysts. The conversion of syngas to methanol is dependent upon the copper metal surface area. Cu based catalysts can promote the reaction rates of both methanol synthesis and water gas shift reaction. The particle size of copper and its dispersion is dependent upon preparation conditions, such as Cu/Zn molar ratio, type of precipitant and calcination temperature. Thus, huge efforts have been carried out to accurately control particle size to within a narrow size distribution [26,119,120]. There are many methods for deposition of nanosized materials over a support, including precipitation, nitrate combustion, sol–gel and sonochemical techniques. As the original methanol synthesis catalyst, the activity of ZnO was not high. Cu metal was found little or no activity neither. However, Cu/ZnO combination catalysts were found synergistic effect and showed high catalytic activities for methanol synthesis. Klier and co-workers have done a serial of pioneering work on the Cu/ZnO combination catalysts [121–125]. After the detection of different mole ratio of Cu in Cu/ZnO catalysts and Cu/ZnO/Al2 O3 catalysts, they found that the role of the CO2 in the CO/CO2 /H2 feed was an oxidant, which maintained the active center in the Cu+ state. Furthermore, they also made the conclusions that it was the CO2 molecule of the CO/CO2 /H2 feed which was hydrogenated to methanol and Cu metal was the active center. Besides the traditional methanol synthesis, methanol synthesis has been explored as a reaction of CO2 abatement. It has been reported that methanol can be synthesized directly from CO2 hydrogenation while H2 could be obtained from renewable sources. CO2 was reported to be reformed from coke oven gas to save energy and decrease CO2 emission (Fig. 9) [126]. By using this technology,

Fig. 9. CO2 partial recycling scheme by means of the CO2 reforming of coke oven gas for methanol production [126].

not only can partial recycle of CO2 , but also a syngas with a H2 /CO ratio close to 2, which is the best ratio for methanol synthesis, can be provided under stoichiometric conditions of CH4 and CO2 . The stable intermediate on the reaction coordinate from CO2 to CH3 OH was found to be a formate [127]. This was discovered by the use of temperature programmed reaction spectroscopy (TPRS). Since the adsorbed formate species is the highest temperature desorbing species on Cu and is formed both on CH3 OH formation and decomposition and it is the most stable and long-lived intermediate in CH3 OH synthesis on Cu, it is a most plausible intermediate in methanol synthesis. It is not clear how the adsorbed H atoms react with the adsorbed HCO3 to form the formate. It could be that the initial interaction is to produce an adsorbed bicarbonate HCO3 species. Even if the bicarbonate were formed, the exact mechanism of the decomposition of an adsorbed HCO3 to HCO2 and O is also not clear. The deactivation of Cu/ZnO/Al2 O3 catalyst during methanol synthesis and regeneration of deactivated catalyst, have been investigated [128,129]. Table 1 shows the catalyst deactivation rates upon exposure to contaminants in syngas during methanol synthesis [130]. Phosphine was found to poison the catalyst and introduction of 1.91 ppmv PH3 into a syngas feed resulted in a deactivation rate of 0.256%/h immediately. All four sulfur-containing contaminants, COS, CS2 , thiophene and CH3 SCN were found to be methanol catalyst poisons. Thus, the poisoning effects of these species continued even after the contaminant flow was stopped. Chlorine is a virulent poison to copper catalysts and should be excluded completely. Hydrogen chloride (HCl) reacted with the active copper to produce low melting copper chloride which caused quick sintering, thus reducing the copper surface area and killing the catalyst completely [131]. Even traces of chloride would remarkably accelerate sintering. The presence of chlorine caused an immediate decrease in catalytic activity which continued even after the HCl source has been removed. Long-term (120 h) methanol synthesis experiments were conducted to observe the progress of catalyst deactivation in the following environments: (i) typical CO-rich syngas (CO, H2 , CO2 , CH4 ), (ii) feed gas mixture without CO, and (iii) feed gas mixture without CO2 [132]. It was observed that the drop in catalytic activity was most serious in the CO free feed gas, as compared with the results of the other two. It is evident that the re-dispersion of the copper phase in the catalyst occurs during the oxidation–reduction cycle and as a result, the copper crystallite size is reduced. On the other hand, the cyclic oxidation–reduction treatment was not

166

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

Table 1 Methanol synthesis catalyst deactivation rates of exposure to contaminants in syngas. Syngas composed H2 (68.2%), CO (22.8%), CO2 (4.7%) and N2 (5.3%) with 6000 GHSV. P = 750 psig, T = 250 ◦ C [130]. Test designation

1 2 3 4 5 6 7 a b

Species

PH3 COS CS2 Thiophene CH3 SCN CH3 Cl CH3 F

Feed concentration (ppmv)

1.91 2.75 2.07 1.61 2.14 2.01 2.55

Deactivation rate (%h)

Contaminant concentration on spent catalyst (ppmw)

Initial, clean syngas

Contaminant/syngas

Final, clean syngas

0.0096 0.0717 0.0660 0.0803 0.0374 0.0311 0.0853

0.256 0.5714 1.330 1.000 0.674 0.657 0.272

∼0 0.1899 0.999 0.0988 0.323 0.169b 0.0498

Experimental 1580 3196 4510 – 1930 2650 344

Calculateda 1628 3320 3851 – 2080 2505 1637

The elemental loading was calculated from contaminant feed concentration, gas flow rates and exposure time. Initial value after returning to clean syngas.

effective in regenerating the catalyst aged in CO2 -free feed gas, where the size distribution did not change much. The production of methanol from CO/CO2 /H2 mixtures over Cu/ZnO/Al2 O3 catalysts is a well-established and, to some extent, a mature process. Roughly 40 million tons/year are produced via this process [133]. However, the mechanism of synthesis of methanol is not without on-going controversies. These include: (i) whether, at steady state, the Cu component of the catalyst is partially covered with O atoms as well as with formate species, (ii) if the reaction has an induction period and, if so, whether the induction period is brought about by a reconstruction of the Cu, and (iii) if the reaction is structure sensitive, or if all of the Cu surface is active, deriving from the formate intermediate being mobile on the Cu.

5. Syngas to dimethyl ether DME is gaining increasing worldwide interests due to its extensive application such as chemical feedstock, cooking fuel, spray-can propellant and diesel fuel oil [28,134–138]. DME can be manufactured in large quantities from coal, natural gas, biomass and municipal solid waste. DME synthesis is generally considered to proceed through three consecutive reactions: (i) methanol synthesis, (ii) methanol dehydration, and (iii) water–gas shift reaction. Presently there are two routes for the production of DME: indirect and direct methods [27]. In the indirect method, methanol is first synthesized on a metallic catalyst and subsequently dehydrated on an acid catalyst in second reactor, while in the direct method, both catalysts (i.e., a bi-functional catalyst) are used in one reactor to perform these reactions simultaneously. The main advantage of the single-step process is abatement of the thermodynamic restriction of methanol synthesis. The drawbacks of direct method are catalyst deactivation by carbon deposition and the hydrothermal effect of water produced as well as the high separation costs in this method. Actually, DME was as an intermediary for production of gasoline through the Topsøe Integrated Gasoline Synthesis (TIGAS) process when initial research was focused on using it. MeOH synthesis was combined with the DME synthesis and integrated into the gasoline synthesis loop in the TIGAS process by Haldor Topsoe [139]. However, it took many years of extensive studies to fully understand the detailed mechanisms. Methanol is first dehydrated to dimethylether (DME). Then an equilibrium mixture of methanol, DME and water is converted to light olefins (C2 –C4 ). Typically, two types of catalysts, CuO–ZnO–Al2 O3 and HZSM5 are used for direct synthesis of DME [140–143]. Fig. 10 shows the experimental reactor configuration for testing catalytic performance of CuO–ZnO–Al2 O3 /HZSM-5 nanocatalysts for direct conversion of syngas to DME [144]. Their mixing state should play an important role in catalytic activity. Finely dispersed copper can

be obtained through the choice of a suitable preparation method. During recent decades, DME synthesis has been explored in detail, and varieties of catalysts and synthesis methods have been the subject of numerous studies. For instance, Yang et al. [145] investigated a millimeter-sized zeolite capsule catalyst possessing a special core shell structure with a co-precipitated catalyst as the core. They prepared two different zeolite capsule catalysts by using H-ZSM5 zeolite (CZA-Z) and Silicalite-1 zeolite (CZA-S) separately. Fig. 11 shows the corresponding product distribution patterns of these two catalysts and their mixture (CZA-M). The selectivity for DME on this zeolite capsule catalyst strikingly exceeded that of the hybrid catalyst prepared by the traditional mixing method, while maintaining the near-zero formation of the unexpected alkanes byproduct. Nie et al. [146] reported special core–shell structured bi-functional catalysts which were prepared by means of simple homogeneous precipitation through urea hydrolysis. Compositions of these catalysts were an HZSM-5 zeolite core enwrapped by one layer of a CuO–ZnO shell. They established that this method offers welldefined structure, while maximizing contact between methanol synthesis sites with methanol dehydration sites and high Cu dispersion. Using an inert liquid as a heat sink for highly exothermic reactions offers another way in syngas processing [129]. Heat generated by the exothermic reactions is readily accommodated by the inert liquid medium such as Witco-40. The single-stage, liquid phase DME synthesis process has also been reported [147–150]. Si–Al based catalysts have been reported for synthesis of DME using the complete liquid-phase technology [147]. (C2 H5 O)4 Si was used and different ratio of Cu(NO3 )2 , Zn(NO3 )2 and Al(NO3 )2 were added. The addition of Si was proved that could modulate the pore structure and surface acidity to promote the increase in activity and selectivity to DME. Gao et al. [148,149] reported a serial of CuZnAl slurry catalysts. Although the CO conversion in this process was lower than the traditional methods, these catalysts could enhance the selectivity of DME to 93.08%. Other than moderation of the temperature, the synergistic effect occurring together in the liquid-phase technology could yield higher quantities of DME than that of from sequential processing.

6. Synthesis of ethylene glycol 6.1. Synthesis of dialkyl oxalate EG is a crucial chemical raw material with a global demand of around 25 million tons each year, which is mostly produced ethylene through traditional petrochemical technology. The cost of this production is high due to the continuous increasing price of natural gas and crude oil, and dwindling sources of petroleum. Coal to

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

167

Fig. 10. Experimental setup for testing catalytic performance of CuO–ZnO–Al2 O3 /HZSM-5 nanocatalysts synthesized via different methods toward direct conversion of syngas to DME [144].

ethylene glycol, as a potentially more green and economic technology, has been attracting extensive attention in both academic and business circles in the past decades. Although it is challenging to achieve high industrial production levels, due primarily to achieving good performance of the catalysts, this technology has been scaled-up to industrial levels of production in China and Europe. Until now, China leads the word in this area and successfully built

the world’s first annual 200 thousand tons coal to ethylene glycol production plant in 2009 [16]. The reactions employed are: Coal to ethylene glycol contains three main steps: (1) separation of carbon monoxide (CO) from coal-derived synthesis gas, (2) reaction of CO with methyl nitrite to form dimethyl oxalate (DMO), and (3) hydrogenation of DMO to ethylene glycol. Scheme 1 [20] shows the reaction mechanism for indirect liquefaction of coal to

Fig. 11. Products distribution of the CZA catalyst, zeolite capsule catalyst CZA-Z and CZA-S, and the hybrid catalyst CZA-M [145].

168

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

Scheme 1. The reaction mechanism of carbon monoxide (CO) to ethylene glycol (EG) [20].

ethylene glycol. Step (2) is the critical step in the scheme, since DMO is required for hydrogenation to EG. Esterification between oxalic acid and alcohol has been employed as a traditional way of synthesizing oxalic ester, in which inorganic acids are used as catalysts and the water generated from the reaction can be removed by using a low-boiling point solvent [22]. However, traditional processes of producing oxalic ester have several problems, such as severe pollution, high energy consumption and high upfront costs. To overcome these problems, oxidative coupling reaction of CO and alkyl nitrite, forming dialkyl oxalate, has attracted increasing research interest in the past decades. A variety of catalyst supports have been reported for the dialkyl oxalate, especially for DMO synthesis, such as activated carbon, silica, activated alumina, ␣-Al2 O3 , silica alumina and so on. Table 2 compares the effect of different solid supports for DMO synthesis [151]. Pd/␣-Al2 O3 has been proven to be the best catalytic system, with good activity, selectivity and stability. Therefore, the catalyst structure [14,152], support [24,153,154], operating conditions [22,155] and process, kinetics and mechanism [23,156–158] of Pd/␣-Al2 O3 catalytic system have been widely studied. Various supported palladium catalysts for gas-phase synthesis of DMO have been investigated, and the results have demonstrated that higher conversion and selectivity are realized on Pd/␣-Al2 O3 compared to Pd on active carbon and ␥-Al2 O3 . To further enhance the activity and stability of Pd/␣-Al2 O3 , several metals such as Fe, Ni and Ce were reported as promoters. CeO2 was reported as a promoter and CeO2 /␣-Al2 O3 catalyst showed rather high catalytic activity compared to Pd/␣-Al2 O3 catalyst (without CeO2 ) for the synthesis of dimethyl oxalate from CO and CH3 ONO [159]. After analysis via X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS) and in situ Fourier transform infrared spectroscopy (FTIR), it was established that promoter CeO2 not only Table 2 Effect of different solid supports on DMO synthesis. Reaction conditions: 110 ◦ C, atmospheric pressure, GHSV 2000 h−1 . DMC: dimethyl carbonate; MF: methyl formate [151]. Support

Activated carbon Silica Activated alumina ␣-Alumina NaY zeolite Si–Al–O (silica alumina)

Space time yield (g/l-cat. h) DMO

DMC

MF

286 255 114 489 58 112

12 13 7 11 3 11

2 4 3 1 4 5

improves the dispersion of palladium on the surface of the catalyst, but also enriches the palladium concentration on the catalyst surface. Also, it was found that the intensity of CO bands on Pd–CeO2 /␣-Al2 O3 is stronger than those on Pd/␣-Al2 O3 [157]. A novel Pd–Fe/␣-Al2 O3 /cordierite monolithic catalyst prepared for the synthesis of diethyl oxalate (DEO) from CO and ethylnitrite used Ferric chloride as a promoter [22]. A monolithic catalyst consists of a honeycomb-like ceramic structure, wash-coat with active component, and promoters. Compared with the traditional Pd–Fe/␣-Al2 O3 catalysts, the amount of Pd can be greatly reduced (7.2 times) while retaining the same catalytic activity. The high efficiency of this catalyst was attributed to its honeycomb structure, which facilitates high dispersion of Pd. Although Pd/␣-Al2 O3 showed higher conversion and selectivity, new carbon nanofibers have also been reported as supports, and resulted in a significant improvement in the gas-phase synthesis of dimethyl oxalate [24]. Carbon nanofibers have been synthesized by disproportionation of CO in the presence of hydrogen on 20 wt% Ni/␥-Al2 O3 and 20 wt% NiFe (1:1)/␥-Al2 O3 with relatively high yields at 600 ◦ C. The deposition precipitation method yields highly dispersed Ni and Ni-Fe particles on ␥-Al2 O3 . Palladium particles are homogeneously deposited on the outer edges of the carbon nanofibers by the standard incipient wetness impregnation method, and also on ␣-alumina and ␣-Al2 O3 supports. The diameter of the palladium particles of the carbon nanofiber supported catalyst is mainly less than 5 nm (Fig. 12), which is much smaller than those on ␣-alumina, and thus resulted in higher activity for DMO synthesis, compared with ␣-Al2 O3 supports. The relatively high Pd loading (around 2 wt%) is always an issue for industrial application of CO oxidative coupling to DMO, which will greatly increase the cost of production. Therefore, the design of low Pd loaded catalysts with high performance is important to industry. A Pd/␣-Al2 O3 nanocatalyst with ultra-low Pd loading that exhibits high activity and stability for CO oxidative coupling to DMO has been developed [16]. This catalyst was prepared by a Cu2+ assisted in situ reduction method at room temperature, which significantly increased the dispersion and the specific surface area of active component Pd, and also decreased the ensemble size of Pd nanoparticles dispersed over the Pd/␣-Al2 O3 (Fig. 13). The average size of Pd nanoparticles is 2.7 nm, and the Pd loading could be as low as 0.13 wt%. Due to its facile synthesis and low Pd loading, this highly efficient and long-lived nanocatalyst may have good prospects for industrial application. The mechanism of the formation of di-alkyl oxalate via oxidative coupling reaction has been studied since the 1980s. Waller [160] proposed a speculative mechanism for the formation of dialkyl oxalate involving the oxidative addition of alkyl nitrite, followed by an intramolecular coupling reaction. They suggested that the oxidative addition of alkyl nitrite (RONO) to Pd (0) would generate an intermediate ON–Pd–OR, and CO insertion in ON–Pd–OR would produce ON–Pd–COOR. Repeating the formation of the ON–Pd–COOR leads to a Pd(COOR)2 intermediate which yields the dialkyl oxalate. They also proposed that a double carbonylation pathway that would lead to a second CO insertion on ON–Pd–COOR, which would give ON–Pd–CO(COOR) as side product. Uchiumi et al. [151] found that the catalytically active species for the formation of oxalate was Pd (0) while the formation of alkyl carbonate was catalyzed by Pd (2+). The addition of RONO oxidizes Pd (0) to Pd (2+). RONO dissociative adsorption provides alkoxyl palladium (Pd–OR) for DMO formation, because alcohol has no effect on the rate of the DMO formation. Gao et al. [161] studied a coupling reaction of CO and ethyl nitrite (EN) to form diethyl oxalate by using XPS and in situ DRIFTS (Diffuse Reflectance Infrared Fourier Transform Spectroscopy). They identified this reaction to be a redox process in which palladium, the active component on the catalyst surface, undergoes a cycle of oxidation of Pd (0) to Pd (2+) and reduction

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

169

Fig. 12. (a, top left) TEM image of the fresh carbon nanofibers produced by disproportionation of CO on 20% Ni/Al2 O3 (CNF-00) at the conditions CO:H2 ) 4:1, Ptot) 1 bar, and T) 600 ◦ C. (b, top right) TEM image of the carbon nanofibers after treatment in HNO3 solution produced by disproportionation of CO on 20% Ni/Al2 O3 (CNF-00-HNO3 ) at the conditions CO:H2 ) 4:1, Ptot) 1 bar, and T) 600 ◦ C. (c, bottom left) TEM images of CNF-00-HNO3 supported palladium particles: 1 wt% Pd/CNF-00-air. (d, bottom right) TEM images of CNF-05-HNO3 supported palladium particles: 1 wt% Pd/CNF-05-air. Remodeled from Ref. [24].

of Pd (2+) to Pd (0). Two kinds of Pd represent two intermediates. The first one is inferred to be ethyloxy palladium complex and the second, more unstable one, to be carboethyoxypalladium complex. It is reported that enhancement of CO adsorption could promote catalyst performance. However, there are two types of CO adsorbed on Pd, i.e., linearly bonded and bridge bonded CO [20]. The ratio of bridge bonded CO to linearly bonded CO could change with changing Pd cluster size [15,22]. Investigating which kind of adsorbed CO participates in the reaction is significant to preparation of catalysts. The mechanism of CO coupling reaction to form DMO has been elucidated by in situ-FTIR under unsteady state conditions (Fig. 14) [23]. The coupling reaction is initiated by adsorption of reactants to form ON–Pd–OMe and bridge bonded CO. Once the ON–Pd–OMe is formed, bridge bonded CO inserts on the intermediate to form ON–Pd–COOMe. Repeating the formation of the ON–Pd–COOMe leads to a Pd(COOMe)2 intermediate which, upon reductive elimination, yields DMO. ON–Pd–OMe and ON–Pd–COOMe were observed in FTIR spectra, but ON–Pd–CO(CO2 Me) was not observed. Two intermediates, ONPd-COOMe and On-Pd-OMe, were found in the reaction and both linearly and bridge bonded CO were found generated on the surface of the catalysts. However, only bridge bonded CO took part in the reaction, and low coverage of bridge bonded CO on the catalysts surface resulted in the formation of by-products.

6.2. DMO hydrogenation The traditional industrial approach to produce EG is hydration of ethylene oxide, which is produced by oxidation of ethylene. Due to constantly increasing demand for EG, alternative routes for synthesis of EG from syngas have become more significant. Hydrogenation of DMO is one of the crucial steps in the indirect synthesis of EG from syngas via DMO. Dimethyl oxalate has been the most widely examined substrate featured in literature reports of ester hydrogenation, and the reaction includes two steps – formation of methyl glycolate (MGL) and hydrogenation of MGL to EG [162]. Both homogeneous and heterogeneous catalytic systems have been reported based on base- and noble-metals such as Ru [162–167], Ag [14,168,169], Cu [156,170–173], Al [174,175] and even Au. Currently, all commercial fatty ester hydrogenation plants employ heterogeneous catalysts, which are required for the efficient conversion of an ester to the corresponding alcohol. A suitable homogeneous alternative has yet to be identified [163]. Ruthenium-based compounds were primarily explored for homogeneous systems and, in general, hydrogenation of DMO to EG has two steps: hydrogenation of DMO to MGL and further hydrogenation of MGL to EG. Usually, MGL to EG requires more severe conditions [P (H2 ) 200 bar, 180 ◦ C] [17] while DMO is relatively easily reduced to MGL. Thus, severe conditions are required

170

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

Fig. 13. TEM images of the catalyst made with (C1 , a) and without (C2 , c) Cu2+ and the corresponding size distributions of Pd NPs (insets), and HRTEM images of C1 (b) and C2 (d), respectively [16].

for the efficient conversion of an ester to the corresponding alcohol, unless the ester is activated by electron-withdrawing substituents [163,165,166,176], or an increased electron density can be introduced on the ruthenium complex center, which enhances the nucleophilicity of the intermediate. For example, Matteoli’s group has reported a series of ruthenium catalysts such as H4 Ru4 (CO)8 (PBu3 )4 , Ru(CO)2 (CH3 COO)2 (PBu3)2 ,

Fig. 14. The mechanism of CO coupling reaction to DMO [23].

and Ru(CO)2 (Ac)2 (PBu)3 , which can catalyze the hydrogenation of DMO with a maximum EG yield of 82%, with H2 pressure as high as 200 bar [177–179]. However, Elsevier and co-worker have developed Ru-based homogeneous catalysts which enabled complete conversion of DMO under relatively mild conditions (80–100 bar H2 , 120 ◦ C for 16 h) with an EG yield of 95%. Those catalysts were obtained from ruthenium(III) acetylacetonate (Ru(acac)3) with the facially coordinating tridentate phosphine ligand CH3 C(CH2 PPh2 )3 for the hydrogenation of DMO [163]. Boardman et al. reported the utilization of sulfur-based ligands [MeC(CH2 SBu)3 ] coupled with ruthenium complex due to sulfur’s electron-rich property [162]. This catalyst first used a simple thioether ligand in combination with ruthenium as a hydrogenation catalyst. However, this catalyst was only highly selective for MGL and could not convert MGL to EG. Table 3 shows that the ruthenium catalysts with phosphine ligands perform with higher activity than the nitrogen or arsenic compounds, which indicates that a ruthenium complex with electron density coordinating ligand is essential for high catalytic activity [163]. Inexpensive heterogeneous catalysts other than ruthenium for hydrogenation of oxalates have attracted extensive attention during the past years. Copper-based catalysts have been reported on numerous occasions, due to high activity for hydrogenation of DMO to EG, especially Cu/SiO2 catalysts for gas-phase hydrogenation of DMO. Copper-containing hexagonal mesoporous silica (HMS) catalysts have been developed, which exhibited superior catalytic performance in the selective hydrogenation of dimethyl oxalate to

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

171

Table 3 Influence of the ligand in the ruthenium catalyzed hydrogenation of DMO to EG. Reaction conditions: T = 100 ◦ C, P (H2 ) = 70 bar. T = 16 h. MeOH (12 ml) [163]. Entry DMO/ mmol

Ligand

Ru:DMO (%)

Ligand:Ru (%)

Zn:DMO (%)

Conversion DMO (%)

Yield MGL (%)

Selectivity MGL (%)

Yield EGL (%)

Selectivity EGL (%)

Turnover number

Turnover freq./h−1

1 2 3 4 5 6 7 8 9 10 11 12

Nonea PPh3 AsPh3 1,10-Phenanthroline 2,2 :6 ,2 -Terpyridine Pyrazolyl ligandb P(C6 H11 )3 Ph2 PC2 H4 PPh2 PhP(C2 H4 PPh2 )2 [CH2 P(Ph)C2 H4 PPh2 ]2 MeC(CH2 PPh2 )3 Ru(CO)2 (OAc)2 (PBu3 )2 c

1.64 1.98 1.19 1.96 1.74 0.21 2.18 1.82 1.75 2.38 1.19 1.34

0.00 5.88 8.91 6.39 1.79 2.32 4.63 2.94 1.68 1.02 1.37 4.00

0.27 1.32 0.57 0.33 0.38 0.06 0.25 0.42 0.36 0.33 0.26 0.00

18 73 1 20 11 14 7 18 76 91 100 100

2 36 0 0 0 1 1 11 67 85 1 18

10 49 0 0 0 4 18 60 89 93 1 18

0 0 0 0 0 0 0 0 0 0 95 82

0 0 0 0 0 0 0 0 0 0 95 82

1 18.1 0 0 0 2.9 0.5 5.9 38.2 35.7 160 136

0 0.9 0 0 0 0.2 0 0.4 2.5 2.2 10 0.9

a b c

0.96 0.99 1.41 1.04 0.89 9.25 0.89 0.88 1.14 0.96 1.77 11.00

Reaction time 41 h. The ligand is tris(3,5-dimethylpyrazol-1-yl)borohydride. T = 180 ◦ C, P (H2 ) = 200 bar with reaction time 144 h.

ethylene glycol, as compared to those with the traditional SiO2 support (Fig. 15) [180]. The dimethyl oxalate conversion can reach 100% and the ethylene glycol selectivity can reach 92% over the 5 wt% Cu-HMS catalyst. The enhanced catalytic performance of copperHMS catalysts may be attributed to the homogeneous dispersion and uniformity of the active copper species and to the larger copper surface areas attained on the HMS supports with large pore diameters and surface areas. Ma and co-workers reported a Cu/SiO2 -based monolithic catalyst for hydrogenation of DMO to EG [181], which was prepared by dip-coating cordierite with highly dispersed Cu/SiO2 slurry via an ammonia evaporation method. This structure guarantees high dispersion of copper species within the mesopores of silica matrix in the form of copper phyllosilicate (Fig. 16). The catalyst is low cost, stable, and exhibits high activity for the DMO hydrogenation reaction, achieving a 100% conversion of DMO and more than 95% selectivity to EG. This catalyst could be an excellent candidate for industrial applications. Cupreous sites account for selective hydrogenation of the carbon–oxygen bond and are relatively inactive for hydrogenolysis of the carbon–carbon bond. We note that strong acidic sites induce the intermolecular dehydration of EG to ethanol, whereas strong basic sites catalyze the Guerbet reaction leading to the formation of 1,2-butanediol (1,2-BDO) [170]. Both products diminish the selectivity of EG. Therefore, catalysts for this reaction are generally supported on neutral or weak acid/alkaline carriers. Representative

copper-based catalytic systems for hydrogenation of DMO to EG are listed in Table 4 [182]. The active sites for the hydrogenation of DMO to EG have also been extensively researched [183–185]. Especially, the valence state of active copper sites is still unclear. Mokhtar et al. [186] suggested that only Cu0 acted as the active site in the ester hydrogenation. However, in methyl acetate hydrogenation, it was reported that while Cu0 dissociatively adsorbs H2 , Cu+ stabilizes the methoxy and the acyl species which are also important intermediates in DMO hydrogenation [187]. Chen et al. [173] reported that Cu/SiO2 catalysts prepared at different temperatures of ammonia evaporation with different ratios of Cu+ /Cu0 and copper surface areas. They proposed that the optimal catalytic activity on the catalyst lies in the cooperative effect between Cu0 and Cu+ to dissociate hydrogen and to activate DMO, respectively. Moreover, Cu+ may function as electrophilic or Lewis acid sites to polarize the C O bond via the electron lone pair on oxygen and thus improve the reactivity of the ester group in DMO. However, the precise role of Cu0 or Cu+ and the synergetic effect of Cu0 and Cu+ are still needs further exploration. Several other base- and noble-metals such as Ru, Ag, Al and even Au have also been reported active for the synthesis of EG from DMO. Ag/SiO2 catalyst with 15 wt% silver loadings was demonstrated as a facile and efficient one pot route for the fabrication of methyl glycolate and EG with the selectivity of 92% and 84%, respectively [169]. Tripodal phosphine ruthenium ligands were developed as

Fig. 15. TEM images of HMS materials (a) and traditional SiO2 (b) and DMO conversion and EG selectivity of Cu-HMS compared with Cu-SiO2 (c). Remodeled from Ref. [180].

172

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

Fig. 16. Illustration and morphology of CuSi/monolithic catalyst. (a) Scheme of monolith catalyst, (b) SEM images of a monolith channel, and (c) cross-sectional cordierite monolith wash-coated with Cu/SiO2 catalyst, and (d) TEM image of calcined wash-coat layer [181]. Table 4 Copper-based catalytic systems for hydrogenation of DMO to EG [182]. Catalyst

Preparation methoda

LHSV/h−1

20%Cu/SiO2 -363 1B-Cu/SiO2 b 5%Cu/HMS 50%Cu/SBA-15 25%Cu/Al-HMS 20%Cu/HMS-333 20%Cu/SiO2 20%Cu/SiO2 15.6%Cu/SiO2 CuB/HMS(2/1)c Cu/H1S1d 20%Cu/SiO2 e

AE UAG IWI ADP DP IE IM DP UH EVI MIM AE

0.5 0.7 0.2 0.6 1.8 0.7 0.8 0.8 0.8 2.5 1.5 2.7

H2 /DMO (mol mol−1 )

50 80 120 100 50 100 60 60 200 120 100 80

T/K

473 463 473 473 473 473 516 516 473 473 473 473

Conversion (%) DMO

79 99.7 100 100 100 100 100 100 100 100 100 95.4

Selectivity (%)

Ref.

EG

MG

EtOH

57 93.0 92 99 98 98 62.5 94.1 90 98 98 92.8

43 3.3 n/a n/a n/a n/a 9.8 1.2 n/a n/a n/a 5.1

1.8 1.8 n/a n/a n/a n/a 25.6 2.0 n/a n/a n/a 1.1

[5] [9] [91] [92] [93] [95] [97] [97] [98] [99] [100] [102]

n/a: not available. a AE: ammonia evaporation, IWI: incipient wet impregnation, ADP: ammonia-driving deposition-precipitation, DP: deposition-precipitation, IE: ion-exchange, UAG: urea-assisted gelation, UH: urea hydrolysis, EVI: ammonia evaporation-induced, MIM: modified impregnation. b 1B-Cu: 1% B and 27% Cu. c Cu/B = 2/1. d H1S1:HMS/silica sol mass ratio of 1/1. e LHSV = 2.75 h−1 .

catalysts of hydrogenation of DMO [166]. A Cu–Zn–Al layered double hydroxide catalyst has been tried for the hydrogenation of DMO to EG [175]. High EG yield of 94.7% was achieved by this novel catalyst and the molar ratios of these three metals might change the selectivity of the product. 7. Conclusion ICL technology provides a sustainable way to use our limited energy sources. In China, this technology is approaching maturity and is poised to play important roles in the growing markets. For

example, large-scale plants to produce EG and diesel via ICL technology have been setup in recent years. In consideration of the continuous increasing price of crude oil and ever-dwindling sources of petroleum, ICL is a promising alternative source of liquid fuels, especially for countries with abundant coal reserves. However, several issues still need to be optimized before this technology can be fully industrialized and eventually surpass traditional petrochemical technology. First, more engineering studies need to be conducted to make this process a commercial success on large scale. Upfront costs are still too high for the rapid industrialization of ICL. Second, effective capture and reuse of CO2 in this process should

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

be further optimized, especially when lignite or brown coal is used. Environmental issues associated with coal liquefaction must be overcome. Lastly, co-productions of fuels, chemicals, heat, and electricity. from the process need to be fully considered and balanced to achieve maximum cost reduction and energy efficiency. Acknowledgements The research was supported by the U.S. Department of Energy and the Wyoming Clean Coal Program. References [1] B. Dutcher, M. Fan, B. Leonard, M.D. Dyar, J.K. Tang, E.A. Speicher, P. Liu, Y.L. Zhang, J. Phys. Chem. C 115 (2011) 15532–15544. [2] R.A. Monterrozo, M. Fan, M.D. Argyle, J. Environ. Eng. – ASCE 138 (2012) 386–391. [3] H. Hu, Q. Yang, X. Lu, W.C. Wang, S.S. Wang, M. Fan, Crit. Rev. Environ. Sci. Technol. 40 (2010) 452–518. [4] L.L. He, M. Fan, B. Dutcher, S. Cui, X.D. Shen, Y. Kong, A.G. Russell, P. McCurdy, Chem. Eng. J. 189 (2012) 13–23. [5] S. Cui, W.W. Cheng, X.D. Shen, M. Fan, A. Russell, Z.W. Wu, X.B. Yi, Energy Environ. Sci. 4 (2011) 2070–2074. [6] B.T. Zhang, M. Fan, A.E. Bland, Energy Fuels 25 (2011) 1919–1925. [7] A. Tuwatit, M. Fan, A.G. Russell, J.J. Wang, H.F.M. Dacosta, Energy Fuels 27 (2013) 7628–7636. [8] B. Voillequin, F. Luck, Actual. Chim. (2011) 16–25. [9] N.O. Elbashir, D.B. Bukur, E. Durham, C.B. Roberts, AIChE J. 56 (2010) 997–1015. [10] F.G. Botes, Catal. Rev. 50 (2008) 471–491. [11] R.A. van Santen, M.M. Ghouri, S. Shetty, E.M.H. Hensen, Catal. Sci. Technol. 1 (2011) 891–911. [12] V.R. Calderone, N.R. Shiju, D.C. Ferre, G. Rothenberg, Green Chem. 13 (2011) 1950–1959. [13] S.H. Zhu, X.Q. Gao, Y.L. Zhu, Y.F. Zhu, H.Y. Zheng, Y.W. Li, J. Catal. 303 (2013) 70–79. [14] J.W. Zheng, H.Q. Lin, Y.N. Wang, X.L. Zheng, X.P. Duan, Y.Z. Yuan, J. Catal. 297 (2013) 110–118. [15] H.Y. Song, R.H. Jin, M.R. Kang, J. Chen, Chin. J. Catal. 34 (2013) 1035–1050. [16] S.Y. Peng, Z.N. Xu, Q.S. Chen, Y.M. Chen, J. Sun, Z.Q. Wang, M.S. Wang, G.C. Guo, Chem. Commun. 49 (2013) 5718–5720. [17] H.R. Yue, Y.J. Zhao, X.B. Ma, J.L. Gong, Chem. Soc. Rev. 41 (2012) 4218–4244. [18] T.H. Fleisch, A. Basu, R.A. Sills, J. Nat. Gas Sci. Eng. 9 (2012) 94–107. [19] A. Kumar, D.D. Jones, M.A. Hanna, Energies 2 (2009) 556–581. [20] Z.N. Xu, J. Sun, C.S. Lin, X.M. Jiang, Q.S. Chen, S.Y. Peng, M.S. Wang, G.C. Guo, ACS Catal. 3 (2013) 118–122. [21] Y.N. Wang, X.P. Duan, J.W. Zheng, H.Q. Lin, Y.Z. Yuan, H. Ariga, S. Takakusagi, K. Asakura, Catal. Sci. Technol. 2 (2012) 1637–1639. [22] X.C. Gao, Y.J. Zhao, S.P. Wang, Y.L. Yin, B.W. Wang, X.B. Ma, Chem. Eng. Sci. 66 (2011) 3513–3522. [23] Y. Ji, G. Liu, W. Li, W.D. Xiao, J. Mol. Catal. A: Chem. 314 (2009) 63–70. [24] T.J. Zhao, D. Chen, Y.C. Dai, W.K. Yuan, A. Holmen, Ind. Eng. Chem. Res. 43 (2004) 4595–4601. [25] J.G. Fang, B.W. Wang, Z.H. Li, G.H. Xu, React. Kinet. Catal. Lett. 80 (2003) 293–301. [26] J.W. Bae, S.H. Kang, Y.J. Lee, K.W. Jun, J. Ind. Eng. Chem. 15 (2009) 566–572. [27] A. Indarto, J.W. Choi, H. Lee, H.K. Song, J. Rare Earth 26 (2008) 1–6. [28] G.X. Qi, J.H. Fei, X.M. Zheng, Z.Y. Hou, React. Kinet. Catal. Lett. 73 (2001) 245–256. [29] P.W.N.M. van Leeuwen, N.D. Clement, M.J.L. Tschan, Coord. Chem. Rev. 255 (2011) 1499–1517. [30] W.Y. Li, Z. Li, K.C. Xie, Energy Sources Part A 31 (2009) 1673–1679. [31] Z. Jiang, T. Xiao, V.L. Kuznetsov, P.P. Edwards, Philos. Trans. R. Soc. A: Math. Phys. Eng. Sci. 368 (2010) 3343–3364. [32] Q.L. Chen, W.M. Yang, J.W. Teng, Chin. J. Catal. 34 (2013) 217–224. [33] U. Olsbye, S. Svelle, M. Bjorgen, P. Beato, T.V.W. Janssens, F. Joensen, S. Bordiga, K.P. Lillerud, Angew. Chem. Int. Ed. 51 (2012) 5810–5831. [34] J.L. White, Catal. Sci. Technol. 1 (2011) 1630–1635. [35] X.B. Yu, Y. Liu, Y.R. Yang, J.D. Wang, Prog. Chem. 21 (2009) 1757–1762. [36] W. Wang, Y.J. Jiang, M. Hunger, Catal. Today 113 (2006) 102–114. [37] M. Stocker, Microporous Mesoporous Mater. 29 (1999) 3–48. [38] F.J. Keil, Microporous Mesoporous Mater. 29 (1999) 49–66. [39] R.A. Dictor, A.T. Bell, Appl. Catal. 20 (1986) 145–162. [40] D. Leckel, Fuel Process. Technol. 92 (2011) 959–969. [41] S.N. Naik, V.V. Goud, P.K. Rout, A.K. Dalai, Renew. Sustain. Energy Rev. 14 (2010) 578–597. [42] M.E. Dry, Catal. Today 71 (2002) 227–241. [43] Y.N. Wang, Y.Y. Xu, Y.W. Li, Y.L. Zhao, B.J. Zhang, Chem. Eng. Sci. 58 (2003) 867–875. [44] A. Irankhah, A. Haghtalab, Chem. Eng. Technol. 31 (2008) 525–536. [45] Y.H. Bi, A.K. Dalai, Can. J. Chem. Eng. 81 (2003) 230–242. [46] C. Knottenbelt, Catal. Today 71 (2002) 437–445.

173

[47] Erdol & Kohle Erdgas Petrochemie 33 (1980) 545. [48] P. Molander, E. Ommundsen, T. Greibrokk, Chromatographia 51 (2000) 349–352. [49] R.H. Crabtree, Stud. Surf. Sci. Catal. 81 (1994) 85–92. [50] L.M. Tau, H.A. Dabbagh, B.H. Davis, Energy Fuels 5 (1991) 174–179. [51] V.A. Kirillov, V.M. Khanayev, V.D. Mescheryakov, S.I. Fadeev, R.G. Lukyanova, Stud. Surf. Sci. Catal. 119 (1998) 149–154. [52] J. Anfray, M. Bremaud, P. Fongarland, A. Khodakov, S. Jallais, D. Schweich, Chem. Eng. Sci. 62 (2007) 5353–5356. [53] E. de Smit, I. Swart, J.F. Creemer, G.H. Hoveling, M.K. Gilles, T. Tyliszczak, P.J. Kooyman, H.W. Zandbergen, C. Morin, B.M. Weckhuysen, F.M.F. de Groot, Nature 456 (2008) 222–225. [54] V.A. Nivin, Dokl. Earth Sci. 429 (2009) 1580–1582. [55] L. Neveux, D. Chiche, D. Bazer-Bachi, L. Favergeon, M. Pijolat, Chem. Eng. J. 181 (2012) 508–515. [56] M.K. Zhuo, K.F. Tan, A. Borgna, M. Saeys, J. Phys. Chem. C 113 (2009) 8357–8365. [57] R.B. Anderson, B. Seligman, J.F. Shultz, R. Kelly, M.A. Elliott, Ind. Eng. Chem. 44 (1952) 391–397. [58] M.E. Dry, T. Shingles, L.J. Boshoff, G.J. Oosthuiz, J. Catal. 15 (1969) 190. [59] H.J. Zhang, H.F. Ma, H.T. Zhang, W.Y. Ying, D.Y. Fang, Catal. Lett. 142 (2012) 131–137. [60] J. Gaube, H.F. Klein, Appl. Catal. A 350 (2008) 126–132. [61] J.Y. Park, Y.J. Lee, P.R. Karandikar, K.W. Jun, J.W. Bae, K.S. Ha, J. Mol. Catal. A: Chem. 344 (2011) 153–160. [62] A.R. de la Osa, A. De Lucas, A. Romero, J.L. Valverde, P. Sanchez, Fuel 90 (2011) 1935–1945. [63] J.M. Xiong, Y.J. Ding, T. Wang, Y. Lu, H.J. Zhu, H.Y. Luo, L.W. Lin, Chin. J. Catal. 26 (2005) 874–878. [64] M.V. Cagnoli, S.G. Marchetti, N.G. Gallegos, A.M. Alvarez, A.A. Yeramian, R.C. Mercader, Mater. Chem. Phys. 27 (1991) 403–418. [65] J. Yang, Y.C. Sun, Y. Tang, Y. Liu, H.L. Wang, L. Tian, H. Wang, Z.X. Zhang, H.W. Xiang, Y.W. Li, J. Mol. Catal. A: Chem. 245 (2006) 26–36. [66] J.F. Li, C.H. Zhang, X.F. Cheng, M. Qing, J. Xu, B.S. Wu, Y. Yang, Y.W. Li, Appl. Catal. A 464 (2013) 10–19. [67] W. Ngantsoue-Hoc, Y.Q. Zhang, R.J. O’Brien, M.S. Luo, B.H. Davis, Appl. Catal. A 236 (2002) 77–89. [68] G. Connell, J.A. Dumesic, J. Catal. 92 (1985) 17–24. [69] C.H. Zhang, Y. Yang, B.T. Teng, T.Z. Li, H.Y. Zheng, H.W. Xiang, Y.W. Li, J. Catal. 237 (2006) 405–415. [70] C. Wang, Q.X. Wang, X.D. Sun, L.Y. Xu, Catal. Lett. 105 (2005) 93–101. [71] V.R. Surisetty, E. Epelde, M. Trepanier, J. Kozinski, A.K. Dalai, Int. J. Chem. React. Eng. 10 (2012). [72] G.L. Bezemer, J.H. Bitter, H.P.C.E. Kuipers, H. Oosterbeek, J.E. Holewijn, X.D. Xu, F. Kapteijn, A.J. van Dillen, K.P. de Jong, J. Am. Chem. Soc. 128 (2006) 3956–3964. [73] J. Chen, J.L. Li, Y.X. Zhao, Y.H. Zhang, J.P. Hong, J. Nat. Gas Chem. 21 (2012) 673–679. [74] A. Martinez, C. Lopez, F. Marquez, I. Diaz, J. Catal. 220 (2003) 486–499. [75] S. Storsaeter, O. Borg, E.A. Blekkan, A. Holmen, J. Catal. 231 (2005) 405–419. [76] M.M. Hauman, A. Saib, D.J. Moodley, E. du Plessis, M. Claeys, E. van Steen, ChemCatChem 4 (2012) 1411–1419. [77] C.J. Weststrate, A.C. Kizilkaya, E.T.R. Rossen, M.W.G.M. Verhoeven, I.M. Ciobica, A.M. Saib, J.W. Niemantsverdriet, J. Phys. Chem. C 116 (2012) 11575–11583. [78] D.J. Moodley, A.M. Saib, J. van de Loosdrecht, C.A. Welker-Nieuwoudt, B.H. Sigwebela, J.W. Niemantsverdriet, Catal. Today 171 (2011) 192–200. [79] A.M. Saib, D.J. Moodley, I.M. Ciobica, M.M. Hauman, B.H. Sigwebela, C.J. Weststrate, J.W. Niemantsverdriet, J. van de Loosdrecht, Catal. Today 154 (2010) 271–282. [80] D.J. Moodley, J. van de Loosdrecht, A.M. Saib, M.J. Overett, A.K. Datye, J.W. Niemantsverdriet, Appl. Catal. A 354 (2009) 102–110. [81] E. van Steen, M. Claeys, M.E. Dry, J. van de Loosdrecht, E.L. Viljoen, J.L. Visagie, J. Phys. Chem. B 109 (2005) 3575–3577. [82] N.M. Ostrovskii, Chem. Eng. J. 120 (2006) 73–82. [83] C.H. Bartholomew, Appl. Catal. A 212 (2001) 17–60. [84] P. Oconnor, A.C. Pouwels, H.N. Wijngaards, Abstr. Pap. Am. Chem. Soc. 210 (1995) (46-Petr.). [85] N.M. Ostrovskii, Izv. Vyssh. Uchebn. Zaved. Khim. I Khim. Tekhnol. 36 (1993) 47–59. [86] J.Y. Park, Y.J. Lee, K.W. Jun, Abstr. Pap. Am. Chem. Soc. 243 (2012). [87] Y.Q. Zhang, N. Sirimanothan, R.J. O’Brien, H.H. Hamdeh, B.H. Davis, Stud. Surf. Sci. Catal. 139 (2001) 125–132. [88] M.E. Dry, Abstr. Pap. Am. Chem. Soc. 219 (2000) U254. [89] B. Jager, R. Espinoza, Catal. Today 23 (1995) 17–28. [90] S. All, B. Chen, J.G. Goodwin, J. Catal. 157 (1995) 35–41. [91] X.H. Mo, J. Gao, N. Umnajkaseam, J.G. Goodwin, J. Catal. 267 (2009) 167–176. [92] B. Viswanathan, R. Gopalakrishnan, J. Catal. 99 (1986) 342–348. [93] Z.M. Liu, J. Liang, Curr. Opin. Solid State Mater. Sci. 4 (1999) 80–84. [94] H.G. Jang, H.K. Min, J.K. Lee, S.B. Hong, G. Seo, Appl. Catal. A 437 (2012) 120–130. [95] W.Z. Wu, W.Y. Guo, W.D. Xiao, M. Luo, Fuel Process. Technol. 108 (2013) 19–24. [96] E. Iglesia, T. Wang, S.Y. Yu, Stud. Surf. Sci. Catal. 119 (1998) 527–532. [97] B.M. Lok, C.A. Messina, R.L. Patton, R.T. Gajek, T.R. Cannan, E.M. Flanigen, J. Am. Chem. Soc. 106 (1984) 6092–6093.

174

E. Jin et al. / Applied Catalysis A: General 476 (2014) 158–174

[98] A. Izadbakhsh, F. Farhadi, F. Khorasheh, S. Sahebdelfar, M. Asadi, Z.F. Yan, Microporous Mesoporous Mater. 126 (2009) 1–7. [99] G.J. Yang, Y.X. Wei, S.T. Xu, J.R. Chen, J.Z. Li, Z.M. Li, J.H. Yu, R.R. Xu, J. Phys. Chem. C 117 (2013) 8214–8222. [100] W.L. Dai, G.J. Wu, L.D. Li, N.J. Guan, M. Hunger, ACS Catal. 3 (2013) 588–596. [101] L. Wu, Z.Y. Liu, L. Xia, M.H. Qiu, X. Liu, H.J. Zhu, Y.H. Sun, Chin. J. Catal. 34 (2013) 1348–1356. [102] Y. Cui, Q. Zhang, J. He, Y. Wang, F. Wei, Particuology 11 (2013) 468–474. [103] Q. Wang, L. Wang, H. Wang, Z.X. Li, H. Wu, G.M. Li, X.P. Zhang, S.J. Zhang, Asia Pac. J. Chem. Eng. 6 (2011) 596–605. [104] A.H. Zhang, S.L. Sun, Z.J.A. Komon, N. Osterwalder, S. Gadewar, P. Stoimenov, D.J. Auerbach, G.D. Stucky, E.W. McFarland, Phys. Chem. Chem. Phys. 13 (2011) 2550–2555. [105] M. Salmasi, S. Fatemi, A.T. Najafabadi, J. Ind. Eng. Chem. 17 (2011) 755–761. [106] J.F. Haw, W.G. Song, D.M. Marcus, J.B. Nicholas, Acc. Chem. Res. 36 (2003) 317–326. [107] A. Sassi, M.A. Wildman, H.J. Ahn, P. Prasad, J.B. Nicholas, J.F. Haw, J. Phys. Chem. B 106 (2002) 2294–2303. [108] D.K. Murray, J.W. Chang, J.F. Haw, J. Am. Chem. Soc. 115 (1993) 4732–4741. [109] E.J. Munson, N.D. Lazo, M.E. Moellenhoff, J.F. Haw, J. Am. Chem. Soc. 113 (1991) 2783–2784. [110] E.J. Munson, A.A. Kheir, J.F. Haw, J. Phys. Chem. 97 (1993) 7321–7327. [111] D.A. Wood, C. Nwaoha, B.F. Towler, J. Nat. Gas Sci. Eng. 9 (2012) 196–208. [112] G.A. Olah, A. Goeppert, G.K.S. Prakash, J. Org. Chem. 74 (2009) 487–498. [113] S. Ledakowicz, Przem. Chem. 85 (2006) 551–555. [114] J. Shen, G. Stiegel, A.C. Bose, Fuel Sci. Technol. Int. 14 (1996) 559–576. [115] A. Boretti, Int. J. Hydrogen Energy 38 (2013) 1806–1812. [116] A.M. Ribeiro, J.C. Santos, A.E. Rodrigues, S. Rifflart, Sep. Sci. Technol. 47 (2012) 850–866. [117] K.C. Waugh, Catal. Lett. 142 (2012) 1153–1166. [118] Y.C. Ma, Q.J. Ge, W.Z. Li, H.Y. Xu, Catal. Commun. 10 (2008) 6–10. [119] J.W. Bae, S.H. Kang, Y.J. Lee, K.W. Jun, Appl. Catal. B 90 (2009) 426–435. [120] R. Khoshbin, M. Haghighi, J. Nanosci. Nanotechnol. 13 (2013) 4996–5003. [121] V. Chatikavanij, G.W. Simmons, K. Klier, Abstr. Pap. Am. Chem. Soc. 182 (1981) 20–30. [122] K. Klier, Adv. Catal. 31 (1982) 243–313. [123] J.M. Dominquez, G.W. Simmons, K. Klier, J. Mol. Catal. 20 (1983) 369–385. [124] R.G. Herman, K. Klier, G.W. Simmons, B.P. Finn, J.B. Bulko, T.P. Kobylinski, J. Catal. 56 (1979) 407–429. [125] S. Mehta, G.W. Simmons, K. Klier, R.G. Herman, J. Catal. 57 (1979) 339–360. [126] J.M. Bermudez, B. Fidalgo, A. Arenillas, J.A. Menendez, Fuel 94 (2012) 197–203. [127] N. Wang, M. Sasaki, T. Yoshida, T. Kotanigawa, Energy Fuels 12 (1998) 531–535. [128] A. Sardesai, S. Lee, Energy Sources 27 (2005) 489–500. [129] S.G. Lee, A. Sardesai, Top. Catal. 32 (2005) 197–207. [130] R. Quinn, T.A. Dahl, B.A. Toseland, Appl. Catal. A 272 (2004) 61–68. [131] M.V. Twigg, M.S. Spencer, Appl. Catal. A 212 (2001) 161–174. [132] R. Quinn, T. Mebrahtu, T.A. Dahl, F.A. Lucrezi, B.A. Toseland, Appl. Catal. A: Gen. 264 (2004) 103–109. [133] G. Dwivedi, S. Jain, M.P. Sharma, Renew. Sustain. Energy Rev. 15 (2011) 4633–4641. [134] J.H. Fei, Z.Y. Hou, B. Zhu, H. Lou, X.M. Zheng, Appl. Catal. A 304 (2006) 49–54. [135] A. Garcia-Trenco, A. Martinez, Appl. Catal. A 411 (2012) 170–179. [136] J.L. Hu, Y. Wang, C.S. Cao, D.C. Elliott, D.J. Stevens, J.F. White, Ind. Eng. Chem. Res. 44 (2005) 1722–1727. [137] M.H. Khademi, M. Farsi, M.R. Rahimpour, A. Jahanmiri, Chem. Eng. Process. 50 (2011) 113–123. [138] J.H. Kim, M.J. Park, S.J. Kim, O.S. Joo, K.D. Jung, Appl. Catal. A 264 (2004) 37–41. [139] J. Topp-Jørgensen, Stud. Surf. Sci. Catal. 36 (1988) 293. [140] J.K. Jeon, S.K. Ihm, Catal. Lett. 31 (1995) 341–350. [141] K.P. Sun, W.W. Lu, M. Wang, X.L. Xu, Catal. Commun. 5 (2004) 367–370. [142] J.Y. Wang, C.Y. Zeng, C.Z. Wu, Chin. J. Catal. 27 (2006) 927–931. [143] Q. Wang, S.R. Wang, X.B. Li, L. Guo, Bioresources 8 (2013) 2897–2909. [144] R. Khoshbin, M. Haghighi, Chem. Eng. Res. Des. 91 (2013) 1111–1122. [145] G.H. Yang, N. Tsubaki, J. Shamoto, Y. Yoneyama, Y. Zhang, J. Am. Chem. Soc. 132 (2010) 8129–8136.

[146] R.F. Nie, H. Lei, S.Y. Pan, L.N. Wang, J.H. Fei, Z.Y. Hou, Fuel 96 (2012) 419–425. [147] Z.H. Li, Z.J. Zuo, W. Huang, K.C. Xie, Appl. Surf. Sci. 257 (2011) 2180–2183. [148] Z.H. Gao, W. Huang, J.F. Li, L.H. Yin, K.C. Xie, Chem. J. Chin. Univ. 30 (2009) 534–538. [149] Z.H. Gao, W. Huang, L.H. Yin, L.F. Hao, K.C. Xie, Catal. Lett. 127 (2009) 354–359. [150] X.D. Peng, B.A. Toseland, R.P. Underwood, Stud. Surf. Sci. Catal. 111 (1997) 175–182. [151] S. Uchiumi, K. Ataka, T. Matsuzaki, J. Organomet. Chem. 576 (1999) 279–289. [152] P. Stepnicka, I. Cisarova, Collect. Czech. Chem. Commun. 74 (2009) 799–809. [153] Y.Y. Zhu, S.R. Wang, L.J. Zhu, X.L. Ge, X.B. Li, Z.Y. Luo, Catal. Lett. 135 (2010) 275–281. [154] Q. Lin, X.G. Zhao, W. Bi, W.D. Xiao, Chin. J. Catal. 27 (2006) 911–915. [155] Q. Lin, Y. Ji, Z.D. Jiang, W.D. Xiao, Ind. Eng. Chem. Res. 46 (2007) 7950–7954. [156] X.B. Ma, H.W. Chi, H.R. Yue, Y.J. Zhao, Y. Xu, J. Lv, S.P. Wang, J.L. Gong, AIChE J. 59 (2013) 2530–2539. [157] Q. Lin, Y. Ji, J. Tan, W. Xiao, Chin. J. Catal. 29 (2008) 325–329. [158] X.Z. Liang, W.J. Zhao, R.H. Yang, M.H. Feng, Z.W. Wang, X.B. Ma, G.H. Xu, Chem. J. Chin. Univ. 18 (1997) 777–781. [159] X.G. Zhao, Q. Lin, W.D. Xiao, Appl. Catal. A 284 (2005) 253–257. [160] F.J. Waller, J. Mol. Catal. 31 (1985) 123–136. [161] Z.H. Gao, Z.C. Liu, F. He, G.H. Xu, J. Mol. Catal. A: Chem. 235 (2005) 143–149. [162] B. Boardman, M.J. Hanton, H. van Rensburg, R.P. Tooze, Chem. Commun. (2006) 2289–2291. [163] H.T. Teunissen, C.J. Elsevier, Chem. Commun. (1997) 667–668. [164] P. Pattanayak, D. Patra, J.L. Pratihar, A. Burrows, M.F. Mahon, S. Chattopadhyay, Inorg. Chim. Acta 363 (2010) 2865–2873. [165] U. Matteoli, M. Bianchi, G. Menchi, P. Frediani, F. Piacenti, J. Mol. Catal. 29 (1985) 269–270. [166] M.J. Hanton, S. Tin, B. Boardman, P. Miller, J. Mol. Catal. A: Chem. 346 (2011) 70–78. [167] M.L. Clarke, M.B. Diaz-Valenzuela, A.M.Z. Slawin, Organometallics 26 (2007) 16–19. [168] J.W. Zheng, H.Q. Lin, X.L. Zheng, X.P. Duan, Y.Z. Yuan, Catal. Commun. 40 (2013) 129–133. [169] A.Y. Yin, X.Y. Guo, W.L. Dai, K.N. Fan, Chem. Commun. 46 (2010) 4348–4350. [170] B. Zhang, Y.L. Zhu, G.Q. Ding, H.Y. Zheng, Y.W. Li, Appl. Catal. A 443 (2012) 191–201. [171] X.Y. Guo, A.Y. Yin, W.L. Dai, K.N. Fan, Catal. Lett. 132 (2009) 22–27. [172] C. Wen, Y.Y. Cui, W.L. Dai, S.H. Xie, K.N. Fan, Chem. Commun. 49 (2013) 5195–5197. [173] L.F. Chen, P.J. Guo, M.H. Qiao, S.R. Yan, H.X. Li, W. Shen, H.L. Xu, K.N. Fan, J. Catal. 257 (2008) 172–180. [174] A.Y. Yin, X.Y. Guo, W.L. Dai, K.N. Fan, J. Phys. Chem. C 114 (2010) 8523–8532. [175] S.Y. Zhang, Q.Y. Liu, G.L. Fan, F. Li, Catal. Lett. 142 (2012) 1121–1127. [176] P. Frediani, C. Giannelli, A. Salvini, S. Ianelli, J. Organomet. Chem. 667 (2003) 197–208. [177] U. Matteoli, M. Bianchi, G. Menchi, P. Frediani, F. Piacenti, J. Mol. Catal. 22 (1984) 353–362. [178] U. Matteoli, G. Menchi, M. Bianchi, F. Piacenti, J. Organomet. Chem. 299 (1986) 233–238. [179] U. Matteoli, G. Menchi, M. Bianchi, F. Piacenti, J. Mol. Catal. 64 (1991) 257–267. [180] A.Y. Yin, X.Y. Guo, W.L. Dai, H.X. Li, K.N. Fan, Appl. Catal. A 349 (2008) 91–99. [181] H.R. Yue, Y.J. Zhao, L. Zhao, J. Lv, S.P. Wang, J.L. Gong, X.B. Ma, AIChE J. 58 (2012) 2798–2809. [182] B. Zhang, S.G. Hui, S.H. Zhang, Y. Ji, W. Li, D.Y. Fang, J. Nat. Gas Chem. 21 (2012) 563–570. [183] L.F. Chen, Y. Zhu, X.Y. Liu, Y. Pei, M.H. Qiao, W. Shen, H.L. Xu, K.N.A. Fan, Acta Chim. Sin. 67 (2009) 2739–2744. [184] D.S. Brands, E.K. Poels, A. Bliek, Appl. Catal. A 184 (1999) 279–289. [185] U. Matteoli, G. Menchi, M. Bianchi, F. Piacenti, J. Mol. Catal. 44 (1988) 347–355. [186] M. Mokhtar, C. Ohlinger, J.H. Schlander, T. Turek, Chem. Eng. Technol. 24 (2001) 423–426. [187] E.K. Poels, D.S. Brands, Appl. Catal. A 191 (2000) 83–96.