Kinetic modeling of CO2 + CO hydrogenation to DME over a CuO-ZnO-ZrO2@SAPO-11 core-shell catalyst

Kinetic modeling of CO2 + CO hydrogenation to DME over a CuOZnO-ZrO2@SAPO-11 core-shell catalyst Ainara Ateka*, Miguel Sánchez-Contador, Ander Portillo, Javier Bilbao, Andres T. Aguayo Department of Chemical Engineering, University of the Basque Country UPV/EHU, P.O. Box 644, 48080 Bilbao, Spain *Corresponding author. Tel.: 34-94-6015361. E-mail address: [email protected] ABSTRACT A kinetic model for the CO2 + CO hydrogenation to dimethyl ether (DME) in a single step over an original core-shell structured CuO-ZnO-ZrO2@SAPO-11 bifunctional catalyst (metallic in the core and acid in shell) has been established. The catalytic runs have been carried out in an isothermal fixed bed reactor under the following conditions: 250-320 ºC; 10-50 bar; space time, 1.25-15 gcat·h·molC-1; H2/COx molar fraction in the feed, 2.5-4, and CO2/COx, 0-1. The catalyst has a high activity and stability as a result of the separation of reactions in the two functions.The model describes the effect of the operating conditions (temperature, pressure and feed composition) over the evolution of product distribution with time on stream. For this, the individual reactions (CO2 and CO hydrogenation to methanol, its dehydration to DME, the WGS reaction and the side reaction of hydrocarbons formation) are considered together with catalyst deactivation. Using the model, simulation studies allow for establishing suitable operating conditions (305 ºC,70 bar, CO2/COx of 0.75 and H2/COx of 3) to attain a good compromise between DME yield and CO2 conversion, reaching a value of 23 % for both objectives. This is the Accepted Manuscript version of a Published Work that appeared in final form in Fuel Processing Technology 206 : (2020) , 106434. To access the final edited and published work see https://doi.org/10.1016/j.fuproc.2020.106434 © 2020. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/ KEYWORDS: Model; deactivation; core-shell; CO2; valorization; dimethyl ether GRAPHICAL ABSTRACT Yield (%) HC MeOH H2O DME CO Time on stream (h) CO2 CuO-ZnO-ZrO2 SAPO-11 CO H2 CO2 1. INTRODUCTION The halt of climate change requires reducing the consumption of fossil sources and the implementation of new sustainable processes for the valorization of CO2 and for the alternative production of fuels and energy vectors [1,2]. Among the catalytic processes under study for the conversion of CO2, the direct synthesis of dimethyl ether (DME) receives great attention and has good prospects for its large-scale industrial implementation, due to the interest of DME economy and the capacity of the process to valorize CO2 co-fed with synthesis gas. DME is a good domestic and diesel engine fuel because its properties are similar to those of the Liquefied Petroleum Gases (LPG) and it has a high cetane number, which facilitates its storage and distribution [3,4]. Its utilization for power production in turbines is also interesting [5]. In addition, the viability of the selective conversion of DME into light olefins [6,7], aromatics or gasoline [8,9] is well established. Besides, DME has other applications, as for example, its use as refrigerant or green solvent [10,11], or in the enhanced oil recovery [12]. The synthesis of DME comprises the following reactions: Methanol synthesis from CO: CO + 2H2 ↔ CH3OH (1) Methanol dehydration to DME: 2CH3OH ↔ CH3OCH3 + H2O (2) Reverse water gas shift (rWGS): CO2 + H2 ↔ CO + H2O (3) Besides, the direct hydrogenation of CO2 to methanol (MeOH) also take place, and is described according to the following reaction: CO2 + 3H2 ↔ CH3OH+ H2O (4) Moreover, the undesired side reaction of paraffins formation may also takes place, giving way mainly to methane: OnHHCH)1n2(OCn 22n2n2    (n=1-3) (5) The direct synthesis of DME is carried out with bifunctional catalysts under pressure and temperature conditions intermediate to those corresponding to the individual reactions of methanol synthesis (Eqs. 1 and 4) and its dehydration to DME (Eq. 2). Conceptually, the integration of the two reactions in a single reactor has lower equipment costs and also facilitates the displacement of the thermodynamic equilibrium of methanol synthesis reactions, since it is in situ dehydrated to DME. The thermodynamic advantages of direct synthesis of DME with respect to two-stage synthesis and to the synthesis of methanol have been compared in the literature [13,14]. Among the practical consequences of these advantages, the following are to be mentioned: i) the greater conversion of CO2 when it is co-fed together with syngas, and; ii) the lower H2/CO ratio required, which facilitates the valorization of the syngas derived from biomass and from different sources (coal, natural gas, biomass, plastics, tires). These advantages and the availability of natural gas and the important development of gasification and reforming technologies justify the attention received in the literature by the direct synthesis of DME [15,16]. This attention has focused mainly on the development of new catalysts [17] and new reactors [18]. The most studied reactors are fixed-bed reactors. Moradi et al. carried out a three dimensional dynamic CFD simulation for the direct DME production from CO and CO2 hydrogenation in a fixed bed reactor. [19,20]. Slurry reactors have also been used for DME synthesis from CO hydrogenation. Papari et al. developed an axial dispersion mathematical model to simulate a slurry bubble column reactor for this reaction [21,22]. This model has been extended to other reactor types [23]. The isothermicity of the fluidized bed reactor is interesting to control the temperature in different catalytic processes, in which the gas flow is considered with a two-phase model [24]. In this regard, Abashar et al. described a model to simulate a two-phase fluidized bed reactor for DME synthesis. [25]. Traditionally, in the bifunctional catalysts used in the direct synthesis of DME the metallic (for methanol synthesis) and acid (for its dehydration to DME) functions are integrated into the same particle by pelletization, in order to achieve the required mechanical strength for its use in the reactor and also to favor the synergy of the catalytic activity of the two functions. As acid function, HZSM-5 zeolite (less hydrophilic) has replaced the -Al2O3 initially used together with the CuO-ZnO metallic function (with different promoters). Despite the moderate-medium acidity of HZSM-5 zeolite, in order to limit the formation of hydrocarbons (coke precursors) in methanol dehydration, the incorporation of metals is used to passivate the strong acid sites [26]. This strategy and the partial dealumination are effective for minimizing side reactions activity and stabilizing zeolites [27]. However, the close contact of the metallic and acid sites also favors the synergy of the coke formation mechanisms in each type of sites and the migration of components, which causes the irreversible deactivation of these sites [28-30]. The use of catalyst particles with core-shell structure is an attractive initiative to preserve the properties of the metallic catalysts and attenuate their sintering [31-33], poisoning [34] or the formation of coke through side reactions [35]. In addition, the separation of the individual reactions in different regions of the catalyst particle improves the selectivity in complex reactions such as Fischer-Tropsch [36]. Thus, the direct synthesis of DME by CO and CO2 hydrogenation has been studied in the literature, with core-shell catalysts of different composition such as Cu-ZnO-Al2O3@HZSM-5 [37-39], CuO-ZnO@HZSM-5 [40], Cr-ZnO@HZSM-5 [41], CuO-ZnO-Al2O3@SiO2-Al2O3 [42] or Al2O3@Cu [43]. In previous works, the preparation and the advantages of a bifunctional CuO-ZnO-ZrO2@SAPO-11 (CZZr@S-11) core-shell catalyst for the direct synthesis of DME have been studied, and its performance has been compared with that of a catalyst with conventional configuration (prepared by pelletization of the metallic and acid functions) [44]. Among these advantages, the greater activity and stability (lower deactivation) of the CZZr@S-11 catalyst are to be highlighted. The separation of methanol synthesis and dehydration reactions in two regions facilitates the separation of water from the first, favoring the activity of the catalyst for methanol synthesis, which explains the higher activity of the catalyst. The separation of the two reactions also prevents the deactivation phenomena previously stated [28-30]. Consequently, the CZZr@S-11 coreshell catalyst is very stable below 325 ºC, and only suffers a slow deactivation by coke [44,45]. The implementation of the CZZr@S-11 catalyst on a larger scale requires having a kinetic model suitable for the design of the reactor, which allows assessing the effects of the process conditions on products yields and distribution. Given the industrial relevance of the main reactions involved, the mechanisms and kinetic models for methanol synthesis [46-50]; its dehydration [51-54] and WGS reaction [55-59] are well established in the literature. However, these kinetic equations have been obtained under the suitable conditions (pressure, temperature) for each of these reactions and with a composition of the reaction medium that is also different from that in the direct synthesis of DME. It is also noteworthy that catalyst deactivation is not quantified. As to the direct synthesis of DME regards, kinetic models have been previously reported for different conventional bifunctional catalysts (hybrid) such as CuO-ZnO-Al2O3/γ-Al2O3 [60] or CuO-ZnO-MnO/SAPO-18 [61], but no kinetic equation has been established for a catalyst with core-shell configuration. In the present work, a kinetic model has been established for the direct synthesis of DME with the CZZr@S-11 core-shell catalyst, in a wide range of reaction conditions (temperature, pressure, space time, CO2/(CO+CO2) and H2/(CO+CO2) molar ratios in the feed). 2. EXPERIMENTAL 2.1. Catalyst preparation and characterization The CZZr@S-11core-shell-like catalyst, has been prepared by physically coating the CZZr metallic function with the S-11 acid function, in a mass ratio of 1/2 as described in detail in previous works [44,45]. The good performance of the CuO-ZnO-ZrO2 (CZZr) function and its adequate composition for the synthesis of methanol were studied in a previous work [62]. SAPO-11 (S-11) has a structure made up of elliptical one-dimensional channels of 0.4x0.6 nm. In addition, it has a high total acidity, but with sites of weak acid strength. These properties lead to high activity for the stage of methanol dehydration to DME, but a low activity for the side reactions of hydrocarbons and coke formation. This good behavior has been ascertained in a previous work [63]. As to the core-shell preparation methodology respects, over the CZZr cores (90-120 μm) the adhesion of the S-11 has been conducted by using a silica solution (Ludox TMA-34, Aldrich) as adhesive, following procedures described in the literature [64-66]. The resulting particles have been dried and calcined at 400 ºC for 2 h, and the strengthened core-shell particles sieved to 125-800 μm. For this purpose, the CuO-ZnZrO2 metallic function was previously prepared following a conventional method of coprecipitation of the metallic nitrates in the desired proportions (Cu:Zn:Zr = 2:1:1) with Na2CO3 and calcined at 300 ºC for 10 h [62]; and the SAPO-11 crystallized (at 195 ºC for 24 h) in a Berghof Highpreactor BR-300 teflon coated autoclave from H3PO4 (Merk), Ludox AS-40 (Aldrich) and Disperal (Sasol) and di-propylamine (Aldrich) as organic template, and calcined at 575 ºC for 8 h [63]. The textural properties of the catalyst were characterized by N2 adsorption–desorption at -196 ºC, using a Micromeritics ASAP 2010. Prior to the measurements, the sample was degassed at 150 ºC for 8 h as for removing possible impurities. Using the Brunauer-Emmett-Teller equation, the specific surface area was determined from the isotherm; and using the BJH method in the adsorption branch of the isotherm, the total pore volume and the micropore volume were determined. The metallic content (Cu:Zn:Zr) has been analyzed by ICP-OES (Inductively Coupled Plasma Optical Emission Spectrometry) in a Perkin Elmer Optima 8300 apparatus; whereas the metallic properties (Cu surface area and dispersion) were determined by selective N2O chemisorption in a Micromeritics Autochem 2920 Apparatus coupled on-line to a Mass Spectrometer (Pfeiffer-Vacuum Omnistar). The acidity and acid strength have been measured by combining thermogravimetry and calorimetry of NH3 adsorption at 150 ºC and subsequent temperature programmed desorption (at 5 ºC·min-1 rate up to 550 ºC) using a Setaram TG-DSC 111 equipment coupled to a Balzers Instruments Thermostar Mass Spectrometer. Table 1 summarizes the most relevant properties of the catalyst. However, further analyzes such as, Scanning Electron Spectroscopy to assess the internal structure of the catalyst; Energy Dispersive X-ray spectroscopy to analyze each section of the catalyst; XRD to study the structural properties, and; Temperature Programmed Reduction (TPR) to ensure the complete reduction of CuO to Cu0 [62] have been carried out. For the characterization of the coke content deposited on the used catalysts, the CO2 signal resulting from Temperature Programmed Oxidation analyzes (conducted in a TA Instruments TGA Q5000 apparatus) has been registered in a Mass Spectrometer (Balzers Instruments). For the quantitative measurement of CO2 in the combustion gases, CaCO3 has been added to each sample as internal standard (decomposes at higher temperature than the combustion compounds) [67]. Table 1. Textural, metallic and acid properties of the CZZr@S-11 catalyst. Textural properties SBET (m2·g-1) Vm (cm3·g-1) Vp (cm3·g-1) 123 0.031 0.300 Metallic properties SCu (m2·gCu-1) SCu (m2·gcat-1) Dispersion (%) 33.3 3.9 5.1 Acid properties Acid strength (kJ·molNH3-1) Total acidity (mmolNH3·gcat-1) 85 0.186 2.2. Reaction equipment and catalytic runs The reaction runs have been carried out in an automated reaction equipment (Microactivity reference, PID Eng. Tech. Micromeritics) provided with a high pressure packed bed stainless steel 316 reactor, of 9 mm of internal diameter and 100 mm of effective length. The used equipment is capable for operating up to 100 atm and 700 ºC, and has been described in detail in previous works [45,68]. In order to ascertain that the kinetics is not affected by the limitations of the stages of diffusion inside and outside the particles, theoretical and experimental criteria of the literature have been adopted [69]. Thus, the absence of diffusion limitations for catalyst particles in the 0.1-0.5 mm range and feeding a total gas flowrate of 60 cm3·min-1 to the reactor has been determined. Consequently, those have been selected as run conditions. and progressive coke deposition on the acid sites, which requires an induction period, and has been related to the routes of the hydrocarbon pool mechanism, side products also containing methoxy ions (produced from the methanol and DME adsorbed on the sites of the acid function) as intermediates. These mechanisms of coke formation are well established for the MTO (methanol to olefins) and DTO (DME to olefins) processes [76]. However, the high H2 pressure in the studied conditions and the limited acid strength of the sites of the SAPO-11 in the catalyst are key features to attenuate the formation of coke. Giving the complexity of considering both routes, due to their different kinetics and bearing in mind that the first one only occurs during the first hour of reaction, special attention has been paid to the second route as this cause of deactivation progresses slowly with time on stream. Consequently, a deactivation kinetic equation has been established dependent on the concentration of the oxygenates (methanol and DME) in the reaction medium, due to its role in the generation of methoxy ions and subsequent formation of hydrocarbons precursors of coke [77,63]:   a··ffk dt da dDMEMeOHd  (28) where kd is the kinetic constant for deactivation. In Eq. (28) a term θd has been considered for quantifying the attenuating effect of H2O and CO2 concentrations on the deposition of coke due to the limitation of the methoxy ions formation [77] and to the competition of these components with coke precursor hydrocarbons for their adsorption on the active sites (both on metallic and on acid sites). As mentioned, term θd gathers the attenuating effect of H2O and CO2 adsorption; anyhow, different expressions have been used in the kinetic models previously described. Thus, in Models 1-3 this effect has not been considered, and therefore, θd = 1 has been established in Eq. (28). In Models 4 and 5 term θd has been considered according to the expressions described in Eqs. (29) and (30), respectively. dOH,adsOH d 22 K·f1 1   (model 4) (29) dCO,adsCO dOH,adsOH d 2222 K·fK·f1 1   (model 5) (30) where dOH,ads 2 K and dCO,ads 2 K correspond to the adsorption constants of H2O and CO2 on the active sites involved in the deactivation by coke. The equations considered in the different models have been summarized in Table 2. Table 2. Studied kinetic models. Model Equations rMeOH rDME rWGS rHC 1 10 11 12 13 2 14 11 12 13 3A 14 11 12 15 3B 14 11 12 16 3C 14 11 12 17 4A 18 11 12 13 4B 14 19 12 13 4C 14 11 20 13 4D 14 11 12 21 5A 23 11 12 13 5B 18 11 24 13 5C 18 11 12 25 3.3. Discrimination of the kinetic models Table 3 shows the main statistic parameters (sum of squares of the errors, SSE; degrees of freedom, ν; variances for the lack of fit, s2) for each model and for the experimental results, whereas Table 4 gathers the values (Fischer distribution, Fa-b; and the critical value of Fisher distribution, F1-α) used for model discrimination (following the methodology described in the Supporting Information). Table 3. Statistic parameters for each model, and for the experimental error. Model SSE ν s2 1 3.36E-01 11251 2.69·10-1 2 3.36E-01 11246 2.61·10-4 3A 3.26E-01 1248 2.61·10-4 3B 3.26E-01 1247 2.61·10-4 3C 3.26E-01 1246 2.61·10-4 4A 3.16E-01 1242 2.54·10-4 4B 3.26E-01 1242 2.63·10-4 4C 3.26E-01 1242 2.62·10-4 4D 3.26E-01 1242 2.62·10-4 5A 2.52E-01 1238 2.04·10-4 5B 2.46E-01 1238 1.99·10-4 5C 2.59E-01 1238 2.09·10-4 Experimental 1.35E-01 581 2.32·10-4 Table 4. Statistic comparison for model discrimination. Fa-b Fa-b F1-α Improvement Selected F1-2 18.67 3.001 Yes 2 F2-3A 0.08 3.85 No 2 F2-3B 0.67 3.00 No 2 F2-3C 0.59 2.61 No 2 F2-4A 5.70 2.02 Yes 4A F2-4B 0.08 2.02 No 2 F2-4C 0.17 2.02 No 2 F2-4D 0.20 2.02 No 2 F4A-5A 78.08 2.38 Yes 5A F4A-5B 87.89 2.38 Yes 5B F4A-5C 69.09 2.38 Yes 5C From the results in Tables 3 and 4, it is evident that considering successively methanol formation from CO2 (Model 2), the attenuating effect of H2O adsorption on the reaction and deactivation rates (Model 4) and the attenuating effect of CO2 adsorption (Model 5) lead to relevant improvements on the fitting. On the other hand, given the low paraffin amount reported, using more kinetic parameters to analyze the origin of their formation is not worth it (Model 3). For comparing models 5A-5C, as they have equal degrees of freedom, a variance analysis has been carried out. Thus, that of lower variance has been selected (5B) since any of them implies an improvement over the other on fitting the experimental data. Finally it has been ascertained that the selected Model 5B satisfies the significance test in Eq. (S5) (Fs= 0.86), which means that the error associated to the lack of fit is lower than the experimental error, and so, that the model represents satisfactorily the experimental results. The kinetic parameters of best fit (kinetic and adsorption constants at reference temperature, k* and K*, respectively, and activation energies and reaction heats, E and H, respectively) for the selected model (5B) have been listed in Table 5. It is noteworthy that the activation energy of methanol synthesis from CO (12.8 kJ·mol-1) is notably lower than that corresponding to its synthesis from CO2 (84.5 kJ·mol-1). Furthermore, the kinetic constant at the reference temperature is greater for the synthesis from CO (1.14 10-5 molMeOH·g-1·h-1·bar-3) than from CO2 (9.47 10-7 molMeOH·g-1·h-1·bar-4). These results, obtained by fitting the results to an empirical kinetic model, are consistent with the molecular simulation results of the DME synthesis by Qin et al. [78]. These authors determined by density functional theory (DFT) that methanol synthesis mechanism takes place through formate ions, with a lower energy barrier for the synthesis from CO than from CO2. Consequently, they consider in their intrinsic reaction model that r-WGS (Eq. (3)) is key for the synthesis of methanol. In Table 5, the value of the kinetic constant at the reference temperature of methanol dehydration is very high (25.6 molDME·g-1·h-1·bar-2), which is also in accordance with the consideration of Qin et al. that the stage of methanol synthesis is slower than that of methanol dehydration and conditions the hydrogenation of CO2 to DME [78]. On the other hand, the reaction heats corresponding to the constants related to de adsorption of H2O and CO2 on the metallic and acid sites (Kads,H2O and Kads,CO2) are small, as correspond to physical adsorption. An interpretation of the values of the adsorption heats of H2O and CO2 cannot be made for the constants that quantify deactivation due to their empirical meaning. Table 5. Kinetic parameters for Model 5B considering deactivation. Parameter units k* or K* (at 275 ºC) E or ∆H (kJ·mol-1) k1 (molMeOH·g-1·h-1·bar-3) 1.14·10-5 1.28·101 k2 (molDME·g-1·h-1·bar-2) 2.56·101 2.07·102 k3 (mol·g-1·h-1·bar-2) 4.63·101 9.33·101 k4 (molMeOH·g-1·h-1·bar-4) 9.47·10-7 8.45·101 k5 (molHC·g-1·h-1) 1.30·10-3 - Kads,H2O (bar-1) 3.17·100 8.70·10-2 Kads, CO2 (bar-1) 1.16·10-1 1.56·10-1 kd (h-1·bar-1) 1.31·10-1 5.73·100 Kads,H2O,d (bar-1) 1.37·10-2 9.12·10-1 Kads,CO2, d (bar-1) 1.26·10-2 9.71·10-1 In order to show visually the fitting of the tested models to the experimental data, further information of the fitting obtained with Models 1, 2, 4A and 5B can be found in the Supporting Information (Fig. S1) and in Fig. 1, where the fitting to all the components in the reaction medium is depicted. For this and subsequent figures product yield has been defined as: 100· F F·n Y0 COx ii i (31) where ni is the number of carbon atoms in a molecule of component i; Fi the molar flowrate of component i at the reactor outlet, and 0 COx F the molar flowrate of carbon in the reactor inlet stream fed as CO and/or CO2. 0 2 4 6 8 10 12 14 16 0 10 20 30 40 50 60 0 2 4 6 8 10 12 14 16 0 10 20 30 40 50 60 0 2 4 6 8 10 12 14 16 0 10 20 30 40 50 60 0 2 4 6 8 10 12 14 16 0 10 20 30 40 50 60 Yield (%) Space time (gcat·h·molC -1) a) CO CO2 MeOH DME HC Model Model 1 Experimental Yield (%) Space time (gcat·h·molC -1) Model 2 b) Yield (%) Space time (gcat·h·molC -1) Model 4A c) Yield (%) Space time (gcat·h·molC -1) Model 5B d) Figure 1. Fitting of models 1 (a), 2 (b), 4A (c) and 5B (d) to the experimental values of CO, CO2, MeOH, DME and HC yields. Reaction conditions: 300 ºC, 30 bar, CO2/COx= 0.5, H2/COx= 3. 3.4. Fitting of the model to the experimental values As an example, the fitting obtained with Model 5B at different operating conditions is depicted in Figure 2. The reaction conditions unless other indicated have been: 300 ºC; 30 bar; 5 gcat·h·molC-1; H2/COx molar ratio in the feed of 3, CO2/COx molar ratio in the feed of 0.5, and 5 h TOS. The study has been extended in the Supplementary Information Section, for other operating conditions, Figures S2-S6. 0 1 2 3 4 5 0 4 8 12 16 40 50 60 0 1 2 3 4 5 0 4 8 12 40 50 60 0 1 2 3 4 5 0 20 40 60 80 Yield (%) Time on stream (h) a) CO CO2 MeOH DME HC Model Experimental Yield (%) Time on stream (h) b) Yield (%) Time on stream (h) c) 0 1 2 3 4 5 0 4 8 12 16 40 50 60 0 1 2 3 4 5 0 4 8 12 40 50 60 0 1 2 3 4 5 0 20 40 60 80 Yield (%) Time on stream (h) a) CO CO2 MeOH DME HC Model Experimental Yield (%) Time on stream (h) b) Yield (%) Time on stream (h) c) 0 1 2 3 4 5 0 4 8 12 16 40 50 60 0 1 2 3 4 5 0 4 8 12 40 50 60 0 1 2 3 4 5 0 20 40 60 80 Yield (%) Time on stream (h) a) CO CO2 MeOH DME HC Model Experimental Yield (%) Time on stream (h) b) Yield (%) Time on stream (h) b) 0 1 2 3 4 5 0 10 20 60 80 100 0 1 2 3 4 5 0 10 20 30 60 80 100 0 1 2 3 4 5 0 4 8 20 40 60 80 100 Yield (%) Time on stream (h) c) CO CO2 MeOH DME HC Model Experimental Yield (%) Time on stream (h) b)b)b)b) Yield (%) Time on stream (h) c)c)c) c) 0 1 2 3 4 5 0 4 8 40 60 80 0 1 2 3 4 5 0 4 8 40 50 60 0 1 2 3 4 5 0 4 8 40 60 80 0 1 2 3 4 5 0 1 2 3 4 40 60 Yield (%) Time on stream (h) a) CO CO2 MeOH DME HC Model Experimental Yield (%) Time on stream (h) b)b)b)b) Yield (%) Time on stream (h) d) Yield (%) Time on stream (h) d) 0 1 2 3 4 5 0 10 20 30 80 100 0 1 2 3 4 5 0 4 8 20 40 60 80 100 0 1 2 3 4 5 0 5 10 15 20 60 80 100 Yield (%) Time on stream (h) a) CO CO2 MeOH DME HC Model Experimental Yield (%) Time on stream (h) b)b)b)b) Yield (%) Time on stream (h) c)c)c) e) 0 1 2 3 4 5 0 2 4 6 8 40 50 60 0 1 2 3 4 5 0 2 4 6 8 40 50 60 0 1 2 3 4 5 0 4 8 12 16 30 40 50 60 0 1 2 3 4 5 0 4 8 12 16 20 30 40 50 60 Yield (%) Time on stream (h) f) CO CO2 MeOH DME HC Model Experimental Yield (%) Time on stream (h) b)b)b)b) Yield (%) Time on stream (h) c) Yield (%) Time on stream (h) d) Figure 2. Fitting of the Model 5B to the experimental values of CO, CO2, MeOH, DME and HC yields evolution with time on stream. Reaction conditions: a) 300 ºC; 30 bar; H2/COx, 2.5; CO2/COx, 0.5; space time, 5 gcat·h·molC-1; b) idem except H2/COx, 4; c) idem except H2/COx, 3 and CO2/COx, 0 (syngas); d) idem except 325 ºC and CO2/COx, 0.5; e) idem except 300 ºC and 40 bar; f) idem except 30 bar and space time 1.25 gcat·h·molC-1. 3.5. Reactor simulation Once proved in the previous Section 3.4 that the proposed kinetic model is capable of describing the evolution of product distribution with TOS within the studied range of operating conditions, it has been used for simulating the operation in a fixed-bed isothermal reactor. Figure 3, shows the operating maps of DME yield for two different feeds; syngas (CO+H2) and CO2+H2, as a function of reaction temperature and pressure. It can be observed that the CO2 content in the feed has a remarkable influence on the yield of DME (YDME), decreasing from around 50 % for CO+H2 feeds, to almost 10 % for CO2+H2 feeds at the most suitable conditions. For both feed compositions, YDME increases noteworthy upon increasing reaction pressure, and the optimum is located within the 280-300 ºC range, the lower limit corresponding to the maximum at higher pressure. Moradi et al. have studied by simulation of a fixed bed reactor the importance of pressure and temperature in the conversion of CO and selectivity of DME, obtaining as optimum a pressure of 50 bar [19,20]. The optimal temperature for these authors is 270 ºC in an adiabatic regime and 260 ºC in an isothermal regime. The differences in the results of these authors with those shown in Figure 3 for CO hydrogenation are moderate and are a consequence of the differences in the kinetic model (different catalyst). Furthermore, the results in Figure 3 correspond to an H2/COx ratio of 3, and this ratio (suitable for CO and CO2 hydrogenation) is of great relevance in the results [45], as also verified by Moradi et al. in the hydrogenation of CO [20]. Figure 3. Evolution of DME yield with reaction temperature and pressure for CO2/COx ratios in the feed of 0 (syngas, CO+H2) and 1 (CO2+H2). Reaction conditions: H2+CO+CO2 feed; H2/COx, 3; space time 5 gcat·h·molC-1; time on stream, 1 h. As a favorable feature, the effect of feeding CO2 on the attenuation of the deactivation is remarkable. Thus, after 1 h time on stream this deactivation is very slow when feeding CO2, as it can be observed in Figures 2a, 2b, 2d, 2e and 1f, and in Figures S2-S6, when CO2 is fed with a CO2/COx ratio of 0.5 or above. This result is of great interest for the industrial viability of the process with this catalyst and is consistent with the deactivation kinetic equation, Eq. (30), which considers the competence of the adsorption of H2O and CO2 with coke precursors. Presumably, these precursors are hydrocarbons formed in the metallic sites by Fischer-Tropsch synthesis from CO, and in the acidic sites from the oxygenates (methanol and DME), by the generation of methoxy ions in this case, which are also active to generate hydrocarbons through the well established hydrocarbon pool mechanism [79]. In addition, the relationship between the doi:10.1016/j.micromeso.2019.02.005. [8] J. Li, D. Jan, T. He, G. Liu, Z. Zi, Z. Wang, J. Wu, J. Wu, Nanocrystal H[Fe, Al]ZSM-5 zeolites with different silica-alumina composition for conversion of dimethyl ether to gasoline, Fuel. Process. Technol. 191 (2019) 104–110. doi: 10.1016/j.fuproc.2019.03.029. [9] X. Su, K. Zhang, Y. Snatenkova, Z. Matieva, X. Bai, N. Kolesnichenko, W. Wu, High-efficiency nano [Zn,Al]ZSM-5 bifunctional catalysts for dimethyl ether conversion to isoparaffin-rich gasoline, Fuel Process. Technol. 198 (2020) 106242–106253. doi:10.1016/j.fuproc.2019.106242. [10] M.C. Bauer, A. Kruse, The use of dimethyl ether as an organic extraction solvent for biomass applications in future biorefineries: A user-oriented review, Fuel. 254 (2019) 115703–115716. doi:10.1016/j.fuel.2019.115703. [11] A. Subratti, L.J. Lalgee, N.K. Jalsa, Liquified dimethyl ether (DME): A green solvent for the extraction of hemp (Cannabis sativa L.) seed oil, Sustain. Chem. Pharm. 12 (2019) 100144–100149. doi:10.1016/j.scp.2019.100144. [12] H. Javanmard, M. Seyyedi, S.A. Jones, S.M. Nielsen, Dimethyl Ether enhanced oil recovery in fractured reservoirs and aspects of phase behavior, Energy & Fuels. 33 (2019) 10718–10727. doi:10.1021/acs.energyfuels.9b02600. [13] M. De Falco, M. Capocelli, G. Centi, Dimethyl ether production from CO2 rich feedstocks in a one-step process: Thermodynamic evaluation and reactor simulation, Chem. Eng. J. 294 (2016) 400–409. doi:http://dx.doi.org/10.1016/j.cej.2016.03.009. [14] A. Ateka, P. Pérez-Uriarte, M. Gamero, J. Ereña, A.T. Aguayo, J. Bilbao, A comparative thermodynamic study on the CO2 conversion in the synthesis of methanol and of DME, Energy. 120 (2017) 796–804. doi:10.1016/j.energy.2016.11.129. [15] Z. Azizi, M. Rezaeimanesh, T. Tohidian, M.R. Rahimpour, Dimethyl ether: A review of technologies and production challenges, Chem. Eng. Process. Process Intensif. 82 (2014) 150–172. doi:http://dx.doi.org/10.1016/j.cep.2014.06.007. [16] G. Leonzio, State of art and perspectives about the production of methanol, dimethyl ether and syngas by carbon dioxide hydrogenation, J. CO2 Util. 27 (2018) 326–354. doi:10.1016/J.JCOU.2018.08.005. [17] U. Mondal, G.D. Yadav, Perspective of dimethyl ether as fuel: Part I. Catalysis, J. CO2 Util. 32 (2019) 299–320. doi:10.1016/j.jcou.2019.02.003. [18] U. Mondal, G.D. Yadav, Perspective of dimethyl ether as fuel: Part IIanalysis of reactor systems and industrial processes, 32 (2019) 321–338. doi:10.1016/j.jcou.2019.02.006. [19] F. Moradi, M. Kazemeini, L. Vafajoo, M. Fattahi, A three dimensonal dynamic CFD simulation for the direct DME production in a fixed bed reactor, Comp. Aided Chem. Eng. 32 (2013) 247–252. [20] F. Moradi, M. Kazemeini, M. Fattahi, A three dimensonal dynamic CFD simulation and optimization of direct DME synthesis in a fixed bed reactor, Pet. Sci. 11 (2014) 323–330. [21] S. Papari, M. Kazemeini, M. Fattahi, Mathematical modeling of a slurry reactor for DME direct synthesis from syngas, J. Nat. Gas Chem. 21 (2012) 148-157. [22] S. Papari, M. Kazemeini, M. Fattahi, M. Fatahi, DME direct synthesis from syngas in a large-scale three phase slurry bubble column reactor: Transient modeling, Chem. Eng. Comm. 201 (2014) 612-634. [23] B. Khadem-Hamedani, S. Yaghmaei, M. Fattahi, S-Mashayekhan, S.M. HosseniArdali, Mathematical modeling of a slurry bubble column reactor for hydrodesulfurization of diesel fuel: Singleand two-bubble configurations, Chem. Eng. Res. Des. 100 (2015) 362-376. [24] A. Torabi, M. Kazemeini, M. Fattahi, Developing a mathematical model for the oxidative dehydrogenation of propane in a fluidized bed reactor, Asia-Pacific J. Chem. Eng. 11 (2016) 448-459 [25] M.E.E. Abashar, Dimethyl ether synthesis in a multi-stage fluidized bed reactor, Chem. Eng. Process. Process Intensif, 122 (2017) 172-180. [26] M.Z. Pedram, M. Kazemeini, M. Fattahi, A. Amjadian, A physicochemical evaluation of modified HZSM-5 catalyst utilized for production of dimethyl ether from methanol, Pet. Sci. Technol. 32 (2014) 904-911. [27] F.B. Shareh, M. Kazemeini, M. Asadi, M. Fattahi, Metal promoted mordenite catalyst for methanol conversion into light olefins, Pet. Sci. Technol. 32 (2014) 1349-1356. [28] A. García-Trenco, A. Vidal-Moya, A. Martínez, Study of the interaction between components in hybrid CuZnAl/HZSM-5 catalysts and its impact in the syngas-toDME reaction, Catal. Today. 179 (2012) 43–51. doi:http://dx.doi.org/10.1016/j.cattod.2011.06.034. [29] A. García-Trenco, A. Martínez, Direct synthesis of DME from syngas on hybrid CuZnAl/ZSM-5 catalysts: New insights into the role of zeolite acidity, Appl. Catal. A Gen. 411–412 (2012) 170–179. doi:http://dx.doi.org/10.1016/j.apcata.2011.10.036. [30] G. Bonura, C. Cannilla, L. Frusteri, E. Catizzone, S. Todaro, M. Migliori, G. Giordano, F. Frusteri, Interaction effects between CuO-ZnO-ZrO2 methanol phase and zeolite surface affecting stability of hybrid systems during one-step CO2 hydrogenation to DME, Catal. Today. (2019). In press. doi:10.1016/j.cattod.2019.08.014. [31] E. Baktash, P. Littlewood, R. Schomäcker, A. Thomas, P.C. Stair, Alumina coated nickel nanoparticles as a highly active catalyst for dry reforming of methane, Appl. Catal. B Environ. 179 (2015) 122–127. doi:10.1016/j.apcatb.2015.05.018. [32] P. Lakshmanan, M.S. Kim, E.D. Park, A highly loaded Ni@SiO2 core-shell catalyst for CO methanation, Appl. Catal. A Gen. 513 (2016) 98–105. doi:10.1016/j.apcata.2015.12.038. [33] M.A. Lucchini, A. Testino, A. Kambolis, C. Proff, C. Ludwig, Sintering and coking resistant core-shell microporous silica-nickel nanoparticles for CO methanation: Towards advanced catalysts production, Appl. Catal. B Environ. 182 (2016) 94–101. doi:10.1016/j.apcatb.2015.09.012. [34] B. Liu, H. Xu, Z. Zhang, Platinum based core-shell catalysts for sour water-gas shift reaction, Catal. Commun. 26 (2012) 159–163. doi:10.1016/j.catcom.2012.05.013. [35] A. Abdalla, P. Arudra, S.S. Al-Khattaf, Catalytic cracking of 1-butene to propylene using modified H-ZSM-5 catalyst: A comparative study of surface modification and core-shell synthesis, Appl. Catal. A Gen. 533 (2017) 109–120. doi:10.1016/j.apcata.2017.01.003. [36] B. Zeng, B. Hou, L. Jia, D. Li, Y. Sun, Fischer-Tropsch synthesis over different structured catalysts: The effect of silica coating onto nanoparticles, J. Mol. Catal. A Chem. 379 (2013) 263–268. doi:10.1016/j.molcata.2013.08.008. [37] G. Yang, N. Tsubaki, J. Shamoto, Y. Yoneyama, Y. Zhang, Confinement effect and synergistic function of H-ZSM-5/Cu-ZnO-Al2O3 capsule catalyst for onestep controlled synthesis, J. Am. Chem. Soc. 132 (2010) 8129–8136. doi:10.1021/ja101882a. [38] R. Liu, H. Tian, A. Yang, F. Zha, J. Ding, Y. Chang, Preparation of HZSM-5 membrane packed CuO–ZnO–Al2O3 nanoparticles for catalysing carbon dioxide hydrogenation to dimethyl ether, Appl. Surf. Sci. 345 (2015) 1–9. doi:https://doi.org/10.1016/j.apsusc.2015.03.125. [39] Y. Sun, X. Han, Z. Zhao, Direct coating copper-zinc-aluminum oxalate with HZSM-5 to fabricate a highly efficient capsule-structured bifunctional catalyst for dimethyl ether production from syngas, Catal. Sci. Technol. 9 (2019) 3763–3770. doi:10.1039/c9cy00980a. [40] R. Nie, H. Lei, S. Pan, L. Wang, J. Fei, Z. Hou, Core–shell structured CuO– ZnO@H-ZSM-5 catalysts for CO hydrogenation to dimethyl ether, Fuel. 96 (2012) 419–425. doi:http://dx.doi.org/10.1016/j.fuel.2011.12.048. [41] G. Yang, M. Thongkam, T. Vitidsant, Y. Yoneyama, Y. Tan, N. Tsubaki, A double-shell capsule catalyst with core–shell-like structure for one-step exactly controlled synthesis of dimethyl ether from CO2 containing syngas, Catal. Today. 171 (2011) 229–235. doi:http://dx.doi.org/10.1016/j.cattod.2011.02.021. [42] Y. Wang, W. Wang, Y. Chen, J. Ma, R. Li, Synthesis of dimethyl ether from syngas over core–shell structure catalyst CuO–ZnO–Al2O3@SiO2–Al2O3, Chem. Eng. J. 250 (2014) 248–256. doi:https://doi.org/10.1016/j.cej.2014.04.018. [43] L. Tan, P. Zhang, Y. Suzuki, H. Li, L. Guo, Y. Yoneyama, J. Chen, X. Peng, N. Tsubaki, Bifunctional capsule catalyst of Al2O3@Cu with strengthened dehydration reaction field for direct synthesis of dimethyl ether from syngas, Ind. Eng. Chem. Res. 58 (2019) 22905–22911. doi:10.1021/acs.iecr.9b04864. [44] M. Sánchez-Contador, A. Ateka, A.T. Aguayo, J. Bilbao, Direct synthesis of dimethyl ether from CO and CO2 over a core-shell structured CuO-ZnO- ZrO2@SAPO-11 catalyst, Fuel Process. Technol. 179 (2018) 258–268. doi:https://doi.org/10.1016/j.fuproc.2018.07.009. [45] M. Sánchez-Contador, A. Ateka, M. Ibáñez, J. Bilbao, A.T. Aguayo, Influence of the operating conditions on the behavior and deactivation of a CuO-ZnOZrO2@SAPO-11 core-shell-like catalyst in the direct synthesis of DME, Renew. Energy. 138 (2019) 585–597. doi:10.1016/J.RENENE.2019.01.093. [46] G. Natta, Catalysis, Reinhold, New York, 1955. [47] G.H. Graaf, E.J. Stamhuis, A.A.C.M. Beenackers, Kinetics of low-pressure methanol synthesis, Chem. Eng. Sci. 43 (1988) 3185–3195. doi:https://doi.org/10.1016/0009-2509(88)85127-3. [48] G.H. Graaf, H. Scholtens, E.J. Stamhuis, A.A.C.M. Beenackers, Intra-particle diffusion limitations in low-pressure methanol synthesis, Chem. Eng. Sci. 45 (1990) 773–783. doi:10.1016/0009-2509(90)85001-T. [49] R.G. Herman, K. Klier, G.W. Simmons, B.P. Finn, J.B. Bulko, T.P. Kobylinski, Catalytic synthesis of methanol from CO H2. I. Phase composition, electronic properties, and activities of the Cu/ZnO/M2O3 catalysts, J. Catal. 56 (1979) 407– 429. doi:10.1016/0021-9517(79)90132-5. [50] L.C. Grabow, M. Mavrikakis, Mechanism of methanol synthesis on cu through CO2 and CO hydrogenation, ACS Catal. 1 (2011) 365–384. doi:10.1021/cs200055d. [51] B.C. Gates, L.N. Johanson, Langmuir-Hinshelwood kinetics of the dehydration of methanol catalyzed by cation exchange resin, AIChE J. 17 (1971) 981–983. doi:10.1002/aic.690170435. [52] C. Ortega, M. Rezaei, V. Hessel, G. Kolb, Methanol to dimethyl ether conversion over a ZSM-5 catalyst: Intrinsic kinetic study on an external recycle reactor, Chem. Eng. J. 347 (2018) 741–753. doi:10.1016/j.cej.2018.04.160. [53] P.O. Ibeh, F.J. García-Mateos, R. Ruiz-Rosas, J.M. Rosas, J. Rodríguez-Mirasol, T. Cordero, Acid mesoporous carbon monoliths from lignocellulosic biomass waste for methanol dehydration, Materials. 12 (2019) 2394. doi:10.3390/ma12152394. [54] J. Palomo, M.A. Rodríguez-Cano, J. Rodríguez-Mirasol, T. Cordero, On the kinetics of methanol dehydration to dimethyl ether on Zr-loaded P-containing mesoporous activated carbon catalyst, Chem. Eng. J. 378 (2019) 122198– 122207. doi:10.1016/j.cej.2019.122198. [55] C. V. Ovesen, P. Stoltze, J.K. Nørskov, C.T. Campbell, A kinetic model of the water gas shift reaction, J. Catal. 134 (1992) 445–468. doi:10.1016/00219517(92)90334-E. [56] C. Rhodes, G.J. Hutchings, A.M. Ward, Water-gas shift reaction: finding the mechanistic boundary, Catal. Today. 23 (1995) 43–58. doi:https://doi.org/10.1016/0920-5861(94)00135-O. [57] D. Ma, C.R.F. Lund, Assessing high-temperature water-gas shift membrane reactors, Ind. Eng. Chem. Res. 42 (2003) 711–717. https://www.scopus.com/inward/record.uri?eid=2-s2.00037453781&partnerID=40&md5=043d971824a854e31239b2d6d3bcc692. [58] A.R. De La Osa, A. De Lucas, A. Romero, J.L. Valverde, P. Sánchez, Kinetic models discrimination for the high pressure WGS reaction over a commercial CoMo catalyst, Int. J. Hydrogen Energy. 36 (2011) 9673–9684. doi:10.1016/j.ijhydene.2011.05.043. [59] R. Burch, A. Goguet, F.C. Meunier, A critical analysis of the experimental evidence for and against a formate mechanism for high activity water-gas shift catalysts, Appl. Catal. A Gen. 409–410 (2011) 3–12. doi:10.1016/j.apcata.2011.09.034. [60] J. Ereña, I. Sierra, A.T. Aguayo, A. Ateka, M. Olazar, J. Bilbao, Kinetic modelling of dimethyl ether synthesis from (H2+CO2) by considering catalyst deactivation, Chem. Eng. J. 174 (2011) 660–667. doi:10.1016/j.cej.2011.09.067. [61] A. Ateka, J. Ereña, J. Bilbao, A.T. Aguayo, Kinetic modeling of the direct synthesis of dimethyl ether over a CuO‑ZnO‑MnO/SAPO‑18 catalyst and assessment of the CO2 conversion, Fuel Process. Technol. 181 (2018) 233–243. doi:10.1016/j.fuproc.2018.09.024. [62] M. Sánchez-Contador, A. Ateka, P. Rodriguez-Vega, J. Bilbao, A.T. Aguayo, Optimization of the Zr content in the CuO-ZnO-ZrO2/SAPO-11 catalyst for the selective hydrogenation of CO+CO2 mixtures in the direct synthesis of dimethyl ether, Ind. Eng. Chem. Res. 57 (2018) 1169–1178. doi:10.1021/acs.iecr.7b04345. [63] M. Sánchez-Contador, A. Ateka, A.T. Aguayo, J. Bilbao, Behavior of SAPO-11 as acid function in the direct synthesis of dimethyl ether from syngas and CO2, J. Ind. Eng. Chem. 63 (2018) 245–254. doi:10.1016/j.jiec.2018.02.022. [64] G. Yang, C. Xing, W. Hirohama, Y. Jin, C. Zeng, Y. Suehiro, T. Wang, Y. Yoneyama, N. Tsubaki, Tandem catalytic synthesis of light isoparaffin from syngas via Fischer–Tropsch synthesis by newly developed core–shell-like zeolite capsule catalysts, Catal. Today. 215 (2013) 29–35. doi:https://doi.org/10.1016/j.cattod.2013.01.010. [65] K. Pinkaew, G. Yang, T. Vitidsant, Y. Jin, C. Zeng, Y. Yoneyama, N. Tsubaki, A new core–shell-like capsule catalyst with SAPO-46 zeolite shell encapsulated Cr/ZnO for the controlled tandem synthesis of dimethyl ether from syngas, Fuel. 111 (2013) 727–732. doi:http://dx.doi.org/10.1016/j.fuel.2013.03.027. [66] R. Phienluphon, K. Pinkaew, G. Yang, J. Li, Q. Wei, Y. Yoneyama, T. Vitidsant, N. Tsubaki, Designing core (Cu/ZnO/Al2O3)–shell (SAPO-11) zeolite capsule catalyst with a facile physical way for dimethyl ether direct synthesis from syngas, Chem. Eng. J. 270 (2015) 605–611. doi:https://doi.org/10.1016/j.cej.2015.02.071. [67] A.G. Gayubo, J. Vicente, J. Ereña, L. Oar-Arteta, M.J. Azkoiti, M. Olazar, J. Bilbao, Causes of deactivation of bifunctional catalysts made up of CuO-ZnOAl2O3 and desilicated HZSM-5 zeolite in DME steam reforming, Appl. Catal. A Gen. 483 (2014) 76–84. doi:10.1016/j.apcata.2014.06.031. [68] A. Ateka, J. Ereña, M. Sánchez-Contador, P. Perez-Uriarte, J. Bilbao, A.T. Aguayo, Capability of the direct dimethyl ether synthesis process for the conversion of carbon dioxide, Appl. Sci. 8 (2018) 677–690. doi:10.3390/app8050677. [69] M. Fattahi, M. Kazemeini, F. Khorasheh, A. Rashidi, Kinetic modeling of oxidative dehydrogenation of propane (ODHP) over a vanadium-graphene catalyst: Application of the DOE and ANN methodologies, J. Ind. Eng. Chem. 20 (2014) 2236–2247. [70] K. Toch, J.W. Thybaut, G.B. Marin, A systematic methodology for kinetic modeling of chemical reactions applied to n-hexane hydroisomerization, AIChE J. 61 (2015) 880–892. doi:10.1002/aic.14680. [71] T. Cordero-Lanzac, A.T. Aguayo, A.G. Gayubo, P. Castaño, J. Bilbao, Simultaneous modeling of the kinetics for n-pentane cracking and the deactivation of a HZSM-5 based catalyst, Chem. Eng. J. 331 (2018) 818–830. doi:10.1016/j.cej.2017.08.106. [72] A.G. Gayubo, A.T. Aguayo, A.L. Morán, M. Olazar, J. Bilbao, Role of water in the kinetic modeling of catalyst deactivation in the MTG process, AIChE J. 48 (2002) 1561–1571. doi:10.1002/aic.690480718. [73] A.G. Gayubo, A.T. Aguayo, A.E. Sánchez Del Campo, A.M. Tarrío, J. Bilbao, Kinetic modeling of methanol transformation into olefins on a SAPO-34 catalyst, Ind. Eng. Chem. Res. 39 (2000) 292–300. doi:10.1021/ie990188z. [74] G. Bonura, M. Migliori, L. Frusteri, C. Cannilla, E. Catizzone, G. Giordano, F. Frusteri, Acidity control of zeolite functionality on activity and stability of hybrid catalysts during DME production via CO2 hydrogenation, J. CO2 Util. 24 (2018) 398–406. doi:10.1016/j.jcou.2018.01.028. [75] S. Ren, W.R. Shoemaker, X. Wang, Z. Shang, N. Klinghoffer, S. Li, M. Yu, X. He, T.A. White, X. Liang, Highly active and selective Cu-ZnO based catalyst for methanol and dimethyl ether synthesis via CO2 hydrogenation, Fuel. 239 (2019) 1125–1133. doi:10.1016/J.FUEL.2018.11.105. [76] T. Cordero-Lanzac, A. Ateka, P. Pérez-Uriarte, P. Castaño, A.T. Aguayo, J. Bilbao, Insight into the deactivation and regeneration of HZSM-5 zeolite catalysts in the conversion of dimethyl ether to olefins, Ind. Eng. Chem. Res. 57 (2018) 13689–13702. doi:10.1021/acs.iecr.8b03308. [77] I. Sierra, J. Ereña, A.T. Aguayo, M. Olazar, J. Bilbao, Deactivation kinetics for direct dimethyl ether synthesis on a CuO-ZnO-Al2O3/γ-Al2O3 Catalyst, Ind. Eng. Chem. Res. 49 (2010) 481–489. doi:10.1021/ie900978a. [78] Z-Z. Qin, T.M. Su, H-B. Ji, Y-X, Jiang, R-W. Liu, J-H. Chen, Experimental and theoretical study of the intrinsic kinetics for dimethyl ether synthesis from CO2 over Cu–Fe–Zr/HZSM-5, AIChE J. 61 (2015) 1613-1627. doi 10.1002/aic.14743 [79] M. Bjørgen, S. Svelle, F. Joensen, J. Nerlov, S. Kolboe, F. Bonino, L. Palumbo, S. Bordiga, U. Olsbye, Conversion of methanol to hydrocarbons over zeolite H- The following Figures S2-S6 show the fitting quality of the selected model (5B) to the experimental results at different operating conditions. 0 1 2 3 4 5 0 4 8 12 16 40 50 60 0 1 2 3 4 5 0 4 8 12 40 50 60 0 1 2 3 4 5 0 20 40 60 80 Yield (%) Time on stream (h) a) CO CO2 MeOH DME HC Model Experimental Yield (%) Time on stream (h) b) Yield (%) Time on stream (h) c) Figure S2. Fitting of the model to the experimental values of CO, CO2, MeOH, DME and HC yields evolution with time on stream for different H2/COx ratios in the feed: 2.5 (a), 3 (b) and 4 (c). Reaction conditions: 300 ºC, 30 bar, CO2/COx= 0.5, space time 5 gcat·h·molC-1. 0 1 2 3 4 5 0 10 20 60 80 100 0 1 2 3 4 5 0 10 20 30 60 80 100 0 1 2 3 4 5 0 4 8 20 40 60 80 100 Yield (%) Time on stream (h) a) CO CO2 MeOH DME HC Model Experimental Yield (%) Time on stream (h) b)b)b)b) Yield (%) Time on stream (h) c)c)c) c) Figure S3. Fitting of the model to the experimental values of CO, CO2, MeOH, DME and HC yields evolution with time on stream for different CO2/COx ratios in the feed: 0 (a), 0.75 (b) and 1 (c). Reaction conditions: 300 ºC, 30 bar, H2/COx= 3, space time 5 gcat·h·molC-1. 0 1 2 3 4 5 0 4 8 40 60 80 0 1 2 3 4 5 0 4 8 40 50 60 0 1 2 3 4 5 0 4 8 40 60 80 Yield (%) Time on stream (h) a) CO CO2 MeOH DME HC Model Experimental Yield (%) Time on stream (h) b)b)b)b) Yield (%) Time on stream (h) c) Figure S4. Fitting of the model to the experimental values of CO, CO2, MeOH, DME and HC yields evolution with time on stream for different reaction temperatures: 250 ºC (a), 275 ºC (b) and 325 ºC (c). Reaction conditions: 30 bar, H2/COx= 3, CO2/COx= 0.5, space time 5 gcat·h·molC-1. 0 1 2 3 4 5 0 10 20 30 80 100 0 1 2 3 4 5 0 4 8 20 40 60 80 100 0 1 2 3 4 5 0 5 10 15 20 60 80 100 Yield (%) Time on stream (h) a) CO CO2 MeOH DME HC Model Experimental Yield (%) Time on stream (h) b)b)b)b) Yield (%) Time on stream (h) c)c)c) c) Figure S5. Fitting of the model to the experimental values of CO, CO2, MeOH, DME and HC yields evolution with time on stream for different reaction pressures: 20 bar (a), 30 bar (b) and 40 bar (c). Reaction conditions: 300 ºC, H2/COx= 3, CO2/COx= 1, space time 5 gcat·h·molC-1. 0 1 2 3 4 5 0 2 4 6 8 40 50 60 0 1 2 3 4 5 0 2 4 6 8 40 50 60 0 1 2 3 4 5 0 4 8 12 16 30 40 50 60 0 1 2 3 4 5 0 4 8 12 16 20 30 40 50 60 Yield (%) Time on stream (h) a) CO CO2 MeOH DME HC Model Experimental Yield (%) Time on stream (h) b)b)b)b) Yield (%) Time on stream (h) c) Yield (%) Time on stream (h) d) Figure S6. Fitting of the model to the experimental values of CO, CO2, MeOH, DME and HC yields evolution with time on stream for different space time values: 1.25 (a), 2.5 (b), 10 (c) and 15 (d). Reaction conditions: 300 ºC, 30 bar, H2/COx= 3, CO2/COx= 0.5. References [1] A.K. Agarwal, M.L. Brisk, Sequential Experimental Design for Precise Parameter Estimation. 1. Use of Reparameterization, Ind. Eng. Chem. Process Des. Dev. 24 (1985) 203–207. doi:10.1021/i200028a034. [2] A.G. Gayubo, A.T. Aguayo, A.E. Sánchez Del Campo, A.M. Tarrío, J. Bilbao, Kinetic modeling of methanol transformation into olefins on a SAPO-34 catalyst, Ind. Eng. Chem. Res. 39 (2000) 292–300. doi:10.1021/ie990188z. [3] P. Pérez-Uriarte, A. Ateka, A.T. Aguayo, A.G. Gayubo, J. Bilbao, Kinetic model for the reaction of DME to olefins over a HZSM-5 zeolite catalyst, Chem. Eng. J. 302 (2016). doi:10.1016/j.cej.2016.05.096. [4] A. Ateka, J. Ereña, J. Bilbao, A.T. Aguayo, Kinetic modeling of the direct synthesis of dimethyl ether over a CuO-ZnO-MnO/SAPO-18 catalyst and assessment of the CO2 conversion, Fuel Process. Technol. 181 (2018) 233–243. doi:10.1016/j.fuproc.2018.09.024.