Effect of water and methanol concentration in the feed on the deactivation of In2O3-ZrO2/SAPO-34 catalyst in the conversion of CO2/CO to olefins by hydrogenation

Fuel 346 (2023) 128298 Available online 13 April 2023 0016-2361/© 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/bync/4.0/). Full Length Article Effect of water and methanol concentration in the feed on the deactivation of In 2 O 3 -ZrO 2 /SAPO-34 catalyst in the conversion of CO 2 /CO to olefins by hydrogenation A. Portillo, O. Parra, J. Ere˜ na, A.T. Aguayo, J. Bilbao, A. Ateka * Department of Chemical Engineering, University of the Basque Country UPV/EHU, P.O. Box 644, 48080 Bilbao, Spain ARTICLE INFO Keywords: CO 2 valorization Olefins SAPO-34 Tandem catalysts Deactivation ABSTRACT On account of the superior performance of In 2 O 3 -ZrO 2 /SAPO-34 tandem catalyst in the direct synthesis of olefins from CO, CO 2 and CO/CO 2 mixture by hydrogenation, it is interesting to establish the conditions to avoid its deactivation due to the rapid coke deposition on SAPO-34. The co-feeding of H 2 O and/or methanol together with H 2 +CO 2 /CO was studied in a packed bed reactor at: 400 ◦C, 30 bar; CO 2 /CO x in the feed, 0–1; H 2 /CO x in the feed, 1–3; and space time of 5 g cat h mol C −1 , quantifying the evolution with time on stream (up to 16 h) of CO 2 and CO x conversions and olefin, paraffin and CH 4 yields. The effects of the co-feeding on coke content and its nature were determined by temperature programmed oxidation (TPO) analyses of the spent catalyst. The results highlighted the complex effect of the concentration of H 2 O and oxygenates (methanol/dimethyl ether (DME)) on the deactivation of SAPO-34 and on the products yields in the pseudo-steady state of the catalyst. Co-feeding H 2 O lessens coke deactivation, however, high H 2 O concentration leads to attenuate the acidity of SAPO-34, limiting the performance of the tandem catalyst (mainly in the CO 2 conversion). Oxygenates co-feeding concentration limit value lies on its favoring effect for coke formation. In addition to this effect, the favorable attenuation of coke deactivation by the high H 2 concentration (studied in runs with H 2 /CO x ratio in the feed in the 1–3 range) plays a key role in the viability of the process, leading to a pseudo-steady catalyst state in which the activity is constant. The proven effect of H 2 O and methanol concentrations will be useful for establishing new catalysts and reaction conditions at which their presence in the reactor will attenuate deactivation. 1. Introduction In order to mitigate climate change, the carbon capture and utilization (CCU) technologies have received a great deal of attention, encouraging the valorization of CO 2 mainly through its transformation into chemicals and fuels [1–3]. Moreover, the joint valorization of CO 2 together with syngas is especially interesting when syngas (with CO and H 2 as main components) is obtained from sustainable routes such as biomass gasification or reforming of its derivatives [4–7]. In addition, with biosyngas co-feeding part of the needed H 2 is supplied. Among the catalytic processes for CO and CO 2 hydrogenation, the direct synthesis of hydrocarbons offers the advantage of integrating in the same reactor the stages for CO/CO 2 hydrogenation and the conversion of the intermediates into the targeted hydrocarbon products such as fuels and chemicals. The extent of the conversion of the intermediates favors the displacement of the equilibrium of thermodynamically limited reactions, boosting the conversion of CO 2 . This process can be carried out via the modified Fischer Tropsch synthesis or with methanol/dimethyl ether (DME) as intermediates. Tandem catalysts are used, combining metal oxides and acidic zeotypes for each reaction stage [8–10]. Not being limited by the Anderson-Shulz-Flory (ASF) distribution, the route with methanol/DME as intermediates is suitable for obtaining high olefin selectivity [11]. Guo et al. [12] studied the thermodynamics of the methanol synthesis stage, emphasizing the complex role of CO. Thus, in the CO 2 hydrogenation at equilibrium conditions, CO formation by the reverse WGS (rWGS) had a negative effect on the conversion of CO 2 and on methanol yield. Whereas co-feeding CO with CO 2 increased these equilibrium values. As these authors stated, the results obtained under favorable thermodynamic conditions will be conditioned by the activity and selectivity of the catalysts used, highlighting that the results in the methanol to olefins conversion stage are a direct consequence of * Corresponding author. E-mail address: [email protected] (A. Ateka). Contents lists available at ScienceDirect Fuel journal homepage: www.elsevier.com/locate/fuel https://doi.org/10.1016/j.fuel.2023.128298 Received 30 January 2023; Received in revised form 27 March 2023; Accepted 31 March 2023 Fuel 346 (2023) 128298 2 the selectivity of the zeotype used in the tandem catalyst composed of a metallic oxide and acid zeotype (OX/ZEO tandem catalysts). The most studied catalysts are prepared mainly using oxides of In or Zn with ZrO 2 as promoter for methanol synthesis: ZnO-ZrO 2 [13–15], In 2 O 3 [16–18], In 2 O 3 -ZrO 2 [19–21] or In 2 O 3 -ZnO-ZrO 2 [22]. These catalysts are more stable at the required operating temperature (≥300 ◦C for the oxygenate conversion into hydrocarbons) than the conventionally employed Cubased catalysts [23]. The zeotype catalysts for the selective production of olefins are primarily based on the silicoaluminophosphate SAPO-34 [13,15,16,18–21,23], mainly thanks to its severe shape selectivity (with CHA structure). Among the attempts to improve the performance of SAPO-34 by modulating its acidity, the addition of HF during the synthesis was ascertained to provide more acid strength to the zeolite and boost light olefins selectivity using a GamCrOx/H-SAPO (F) tandem catalyst, while reducing CO selectivity [24]. Alternatively, for the production of higher hydrocarbons catalysts based on HZSM-5 zeolite (MFI structure) are applied, with lower severity of shape selectivity [14,17]. Even if the severe shape selectivity of the SAPO-34 catalyst is suitable for the selective production of light olefins, it is well established for the MTO/DTO processes that as a drawback it promotes the confinement of the coke precursor intermediates [25,26] and their evolution towards polyaromatic structures, which block the acid sites and the porous structure [27]. It is also remarkable the foreseeable autocatalytic nature of coke formation, whose rate in the MTO/DTO processes is dependent on the methanol/DME concentration [28]. Among the strategies to attenuate coke deactivation, the decrease of the crystal particle size and the hierarchical organization of the pore structure of the catalyst particle stand out [29]. The concentration of water has a favorable incidence on the attenuation of the deactivation [30,31], which motivates its cofeeding [32,33], although competition for the adsorption of water and methanol/DME and reaction intermediates in the acid sites also diminishes their activity for the production of olefins [34–36]. Moreover, the favorable contribution of the high H 2 partial pressure for attenuating coke deposition on methanol/DME conversion into hydrocarbons is noteworthy [37–39] due to the hydrogenation of the coke precursors [40]. Presumably, these effects will exist in the direct synthesis of olefins from CO 2 /CO hydrogenation, although the conditions and the composition of the reaction medium are different. Hence, the presence of H 2 in the direct synthesis of olefins from CO 2 /CO hydrogenation results in a lower deactivation of the SAPO-34 catalyst [41]. In a previous work [42], the optimal composition of the In 2 O 3 -ZrO 2 catalyst was determined (with an In/Zr ratio of 1/2), in order to favor the conversion of CO 2 and CO x (CO 2 /CO) mixtures. This way, the sintering of In by over-reduction was avoided, and the secondary methanation reaction was suppresed. In addition, the optimal conditions (moderate pressure of 30 bar, 400 ◦C, H 2 /CO x ratio in the feed of 3), suitable for obtaining with the In 2 O 3 -ZrO 2 /SAPO-34 tandem catalyst a high per pass yield of olefins (greater than 4%) with a selectivity greater than 70 % being propylene the major olefin, were determined [41]. It should be pointed out that the catalyst undergoes a rapid initial deactivation, and subsequently reaches a pseudo-steady state of constant activity, in which coke formation rate is equal to its elimination rate by hydrogenation. This work addresses the drawback of the deactivation of the In 2 O 3 - ZrO 2 /SAPO-34 tandem catalyst, which conditions its feasibility for the direct synthesis of olefins from CO 2 /CO feeds. The study emphasizes the effect of reaction conditions (H 2 partial pressure, water and methanol concentration in the reaction medium) that will have an impact on the coke deposition in SAPO-34 (origin of the deactivation). The results will be of interest to adapt the properties of the catalyst and operating conditions with the objective of lessening the deactivation. 2. Experimental 2.1. Catalyst synthesis and characterization In 2 O 3 -ZrO 2 catalyst was synthesized following a co-precipitation method described in detail in a previous work [42]. Briefly, metallic nitrates (In(NO 3 ) 3 and Zr(NO 3 ) 4 , Panreac) were dissolved in water (1 M) in a 2/1 In/Zr ratio and coprecipitated with (NH 4 ) 2 CO 3 at 70 ◦C and maintaining the pH close to 7. Subsequently, the sample was dried, calcined and pelletized. The best configuration was ascertained to be the mixture of individually pelletized In 2 O 3 -ZrO 2 (125–250 µm) and SAPO34 (300–400 µm) catalyst particles in a 2/1 mass ratio diluted in SiC [43]. The different particle size eases the separation of the individual spent catalysts by sieving in order to analyze them. In 2 O 3 -ZrO 2 and SAPO-34 catalysts were characterized by N 2 adsorption–desorption analyses (Micromeritics ASAP 2010), NH 3 -temperature programmed desorption (NH 3 -TPD) analyses and H 2 and CO temperature programmed reduction (H 2 -TPR and CO-TPR). Characterization results are shown in Table 1 (physical properties and acidity) and in Supporting Information (Fig. S1) (TPR analyses). The high value of the BET surface area of the SAPO-34 catalyst (651.8 m 2 g −1 ), consistet with its microporous structure, as well as its high acidity (777.6 µmol NH3 g cat −1 ) should be noted. The remarkable surface area of the In 2 O 3 -ZrO 2 catalyst is consequence of the mesoporous structure of both oxides. Thus, S BET accounts for 53.4 m 2 g −1 for In 2 O 3 and 95.4 m 2 g −1 for ZrO 2 , and the pore volume 0.25 and 0.16 cm 3 g −1 , respectively [42]. The acidity (122.6 µmol NH3 g cat −1 ) is attributable to ZrO 2 . The results in Fig. S1 of In 2 O 3 -ZrO 2 catalyst TPR are in line with previous results [42], where the higher reducing capability of CO compared to H 2 is confirmed, in line with the literature [42,44,45]. 2.2. Catalytic activity test CO 2 /CO hydrogenation reactions were carried out in a reaction equipment (PID End & Tech Microactivity Reference) described in previous works [42,46] equipped with an isothermal stainless steel packed bed reactor, with an effective length of 10 cm and an inner diameter of 9 mm and it can operate at conditions up to 800 ◦C and 100 bar. It is provided with an internal ceramic coating to avoid side reactions with the reactor walls. For each run the reactor was loaded with a mixture of the desired catalyst amount and SiC (up to 5 g of solid mixture) to avoid preferential pathways, ensure isothermal conditions and sufficient bed-height when using small space time values. In order to analyze the products on-line, a representative fraction of the reactor outlet stream was diluted in He and sent to an on-line connected gas chromatograph (Varian CP-4900). The microGC is equipped with MS-5, Porapak Q and CPSiL columns, allowing the quantification of compounds comprising H 2 , H 2 O, oxygenates (methanol and DME) and hydrocarbons up to C 9 . The coke deposited over the spent catalysts was studied by temperature programmed oxidation (TPO) analyses (TA Instruments TGA Q5000 thermobalance), consisting of: i) sweeping of the sample with He at 120 ◦C for 30 min to remove the humidity, ii) stabilizing of the sample in air at 120 ◦C (10 cm 3 min −1 ) and iii) heating of the sample up to 680 ◦C with a heating rate of 7 ◦C min −1 . The weight of the sample was measured during the whole experiment to ascertain that all humidity Table 1 Physical and acid properties of the metallic and acid catalysts. Catalyst S BET (m 2 g −1 ) V micropore (cm 3 g −1 ) V pore (cm 3 g −1 ) d p (nm) Acidity (µmol NH3 g cat −1 ) In 2 O 3 - ZrO 2 86 0.003 0.23 9.0 122.6 SAPO-34 652 0.219 0.23 1.5 777.6 A. Portillo et al. Fuel 346 (2023) 128298 3 was desorbed during the first step. The reaction runs were carried out under the following conditions: 350–425 ◦C; 20–50 bar; space time, 5 g cat h mol C −1 ; H 2 /CO x , 1–3; H 2 / CO x , 0–1. The effect of H 2 O and methanol content in the reaction medium was studied co-feeding both in a 5 % molar fraction. The results were quantified from the molar flows at the inlet and outlet of the reactor in contained C units, as follows. Conversion of CO 2 (X CO2 ): XCO2=F0 CO2−FCO2 F0 CO2 100 (1) where F 0 CO2 and F CO2 are the CO 2 molar flowrates at the inlet and outlet of the reactor, respectively. The conversion of the CO and CO 2 mixture (the carbon fraction converted into hydrocarbons or oxygenates) (X COx ) was defined as: XCOx=F0 COx−FCOx F0 COx 100 (2) where F 0 COx is inlet molar flowrate in C atoms, and F COx its analogous at the outlet of the reactor. Carbonaceous product yields (Y i ) and selectivities (S i ) were defined by grouping the products in lumps: CH 4 , C 2 -C 4 olefins, C 2 -C 4 paraffins and oxygenates (methanol and DME) according to the following expressions: Yi=niFi F0 COx 100 (3) Si=niFi ∑niFi 100 (4) being n i the number of C atoms in a molecule of i component (or lump) and F i the corresponding molar flowrate at the reactor outlet stream. 3. Results 3.1. Coke deposition over SAPO-34 Fig. 1 shows the TPO profiles of the two individual catalysts after a certain reaction. It is evident that most of the coke was formed over SAPO-34 (where coke amount accounted up to 10.9 wt% while in In 2 O 3 - ZrO 2 did not reach 0.6 wt%). The high coke content for SAPO-34 is typical of the MTO/DTO processes [26,47–49] and it is explained by the confinement capacity of the intermediate arenes formed in the dual cycle mechanism in the cages of SAPO-34 catalyst. The wide temperature range (250–650 ◦C) of the TPO profile corresponds to a coke formed by heterogeneous carbonaceous species, whose condensation towards polyaromatic structures during the reaction and its combustion in the TPO analysis will be limited by the diffusional restrictions in the cages of the SAPO-34. On the contrary, The hydrogenation capability of the In 2 O 3 -ZrO 2 catalyst enables the hydrogenation of coke precursors during the reaction and its mesoporous structure eases its diffusion outwards the catalyst particle, explaining its low content in 16 h on stream. Furthermore, in Fig. 1, the maximum combustion peak is observed at 450 ◦C for SAPO-34 and 270 ◦C for In 2 O 3 -ZrO 2 , as the combustion of the coke deposited on the metallic oxides is catalyzed (which is interesting to facilitate its regeneration). In all the studied reaction conditions similar results to these plotted in Fig. 1 were obtained, pointing to coke formation over SAPO-34 as the main responsible of the deactivation of In 2 O 3 -ZrO 2 /SAPO-34 tandem catalyst. Consequently, the same species that deactivate SAPO-34 during the MTO/DTO reactions will presumably be the reason for the deactivation of the tandem catalyst in the direct CO/CO 2 to olefins conversion too, as will be discussed below. 3.2. Effect of the feed composition 3.2.1. Water and methanol co-feeding with H 2 /CO and H 2 /CO 2 feeds Various runs, where H 2 O, methanol, and an equimolar mixture of H 2 O and methanol were added to the H 2 +CO/CO 2 feed, were conducted to assess the role of H 2 O and methanol intermediates in the direct hydrogenation of CO/CO 2 to olefins over the tandem catalysts. The H 2 O flow rate fed was calculated to simulate the same concentration as that generated stoichiometrically by the rWGS reaction with H 2 /CO 2 feeds, which corresponds to approximately 5 mol%. Consequently, the effect of CO and CO 2 can be compared as in all cases H 2 O concentration was similar. For the co-feeding of methanol, the CO and CO 2 flow rates were adjusted to keep the same space time value. H 2 /CO x ratio in the feed was 3 in all runs. In Fig. 2 product yields evolution with time on stream with H 2 /CO feed is depicted together with the results obtained co-feeding H 2 O and methanol. With the H 2 /CO feed (Fig. 2a), the initial deactivation of SAPO-34 was very fast, which caused a decrease in olefin and paraffin (formed by hydrogenation of the former) yields, and led to an increase of the concentration of oxygenates (non converted) and CH 4 (formed by oxygenates cracking). These yields reached a constant value after 4 h on stream corresponding to the catalyst remaining activity. H 2 O co-feeding (Fig. 2b) attenuated the deactivation rate, delaying the time for acquiring a pseudo-steady state of the catalyst (up to 8 h) in which olefin and paraffin yields were similar to those obtained without H 2 O in the feed. The higher yield of oxygenates and lower CH 4 yield demonstrate the effect of H 2 O to attenuate the activity of SAPO-34 acid sites for the dual cycle mechanism and also for methanol/DME cracking [50,51]. Consequently, besides the lower deactivation, the smaller CH 4 yield is another interesting result that facilitates the recirculation of the products stream and helps to obtain higher conversion. On the other hand, methanol co-feeding (Fig. 2c), did not lead to a higher olefin yield. Because the deactivation rate is directly dependent to oxygenates concentration. Thus, the yield of olefins and paraffins was stable after 2 h on stream and the values were significantly lower than at previous conditions. In this reached catalyst state, the oxygenate stream was mainly formed by DME, because methanol dehydration activity was maintained, which only required a low density of acid sites of reduced acid strength [35,52]. The high CH 4 yield obtained when co-feeding methanol suggests that CH 4 was being formed due to the high oxygenates (methanol and DME) partial pressure, which underwent thermal cracking at 400 ◦C (mainly DME as it is less stable than methanol) [53]. Thus, it is observed that an excessive formation of oxygenates has the drawback of strongly favoring the fast deactivation of SAPO-34 catalyst by coke. Based on the results, the incidence of H 2 O co-feeding on the deactivation of the SAPO-34 catalyst was similar to the well-established 100200300400500600700 0.000 0.002 0.004 0.006 Cat. In2O3-ZrO2 SAPO-34 DTG ( g c min -1 g -1 cat ) Temperature ( C) Fig. 1. Comparison of TPO profiles for the individual In 2 O 3 -ZrO 2 and SAPO-34 spent catalysts. Reaction conditions: 400 ◦C; 30 bar; CO 2 /CO x , 0.5; space time, 5 g cat h mol C −1 ; H 2 /CO x , 3, and; time on stream, 16 h. A. Portillo et al. Fuel 346 (2023) 128298 4 effect on the conversion of methanol/DME to olefins (processes at atmospheric pressure and without H 2 presence). H 2 O decreases the activity and attenuates the deactivation of the catalyst, as it competes for the adsorption in the acid sites with reactants and coke precursors [51,52]. It is also well-established for the methanol/DME conversion to olefins over different acid catalysts that increasing the concentration of reactant oxygenates results in a higher coking rate [33,53]. In terms of scaling up, it is interesting to compare the yields in the pseudo-steady state of the catalyst for the different feeds (Fig. 2d). When co-feeding H 2 O, olefin yield in the pseudo-steady state of the catalyst was equal (and it was even higher before reaching this state, compared to pure H 2 /CO feeds) and higher oxygenate yield was achieved. The latter occurs due to the shift of the equilibrium of the methanol to DME dehydration step (hampered by H 2 O), which results in higher methanol yield in the medium and lower CH 4 , enlightening that it is formed preferentially from DME thermal cracking. These effects (derived from the presence of H 2 O) would take place without H 2 O co-feeding when modifying many other process conditions, considering that H 2 O is the product of numerous reactions of the direct CO 2 /CO to olefins process: i) rWGS reaction, ii) synthesis of methanol from CO 2 , iii) methanol dehydration to DME and iv) conversion of oxygenates to olefins. The effect of methanol co-feeding in Fig. 2d shows a high oxygenate yield and CH 4 . It is remarkable that DME is the major compound within the oxygenates (96 %). This occurs due to the fast deactivation of SAPO34 catalyst, only capable of dehydrating methanol. Moreover, the low remaining activity of SAPO-34 for the conversion of methanol/DME via the dual cycle mechanism favors the CH 4 formation by DME thermal cracking. Consequently, this effect must be taken into account, operating under conditions in which the excess of oxygenates in the reaction medium is avoided. Fig. 2d also gathers the results obtained with the catalyst in the pseudo-steady state with the joint co-feeding of H 2 O and methanol. In this case, the excess of both components in the reaction medium combined their negative effects, resulting in a decrease in olefin yield and in an increase of DME yield. The CH 4 yield attained was similar to that obtained without co-feeding H 2 O or methanol, suggesting that the effects of co-feeding methanol (favoring CH 4 formation) and H 2 O (attenuating CH 4 formation) were compensated. The effect of the co-feeding of H 2 O and methanol was also evaluated for H 2 /CO 2 feed (Fig. 3), comparing the results with the aforementioned for H 2 /CO feed. As observed previously in Fig. 2b with H 2 /CO feed, H 2 O co-feeding led to a very even evolution of products distribution with time on stream (Fig. 3b) compared to the run without H 2 O co-feeding (Fig. 3a). However, the conversion into olefins was strongly penalized by the thermodynamic effects (attenuation of the extent of rWGS, methanol formation, DME formation and MTO/DTO reactions). The results obtained by methanol addition (Fig. 3c) resulted in an excess of oxygenates that turned into a high CH 4 formation by DME (main oxygenate in this case) thermal cracking. Comparing the product yields in the pseudo-steady state of the catalyst (Fig. 3d), it is observed that H 2 O co-feeding also entailed a negative effect, in contrast to that observed for H 2 /CO feed (Fig. 2d). This occurs because for H 2 /CO 2 feed the rWGS reaction is favored and, thus, the H 2 O generated by rWGS is added to the fed H 2 O, resulting in an 0 4 8 12 16 0 1 2 3 4 5 6 0 4 8 12 16 0 1 2 3 4 5 6 0 4 8 12 16 0 1 2 3 4 5 6 H 2 / CO H 2 / CO + H 2 O H 2 / CO + H 2 O + CH 3 OH H 2 / CO + CH 3 OH 0 1 2 3 4 5 6 7 8 9 Yield (%) Olefins Paraffins Oxygenates Methane d) Olefins Paraffins Oxygenates Methane Yield (%) Time on stream (h) a) Yield (%) Time on stream (h) b) Yield (%) Time on stream (h) c) Fig. 2. Evolution of product yields with time on stream for H 2 /CO feed without any co-feeding (a), co-feeding H 2 O (b), cofeeding methanol (c), and their values at the pseudo-steady state of the catalyst (16 h on stream) (d). Reaction conditions: 400 ◦C; 30 bar; CO 2 /CO x , 0; space time, 5 g cat h mol C −1 , and; H 2 /CO x , 3. A. Portillo et al. Fuel 346 (2023) 128298 5 excess of H 2 O in the medium. Therefore, the high H 2 O concentration excessively attenuated the extent of the dual cycle mechanism for olefin formation. On the other hand, analogously to what was observed for the H 2 /CO feed, co-feeding methanol occasioned negative effects. CH 4 and DME were formed and olefin yield was lower. Furthermore, the joint cofeeding of H 2 O and methanol integrated the negative effects. These results ratify the importance of H 2 O and methanol/DME concentrations on the mechanisms of the direct synthesis of olefins from CO 2 and on deactivation. It is also noteworthy that the presence of H 2 O in a small concentration had the positive effect of attenuating the deactivation with the H 2 /CO feed. Nonetheless, an excessive H 2 O concentration with H 2 /CO 2 feed involved the negative effect of limiting the activity of the acid sites of SAPO-34 to be predominant. All these effects over the catalyst deactivation are related to the differences in coke formation on the SAPO-34 catalyst. These trends are clearly observed in the TPOs gathered in Fig. 4 corresponding to the runs with H 2 /CO 2 feed, without and with co-feeding H 2 O, methanol and a mixture of both. On the one hand, co-feeding H 2 O lessened coke content from 10.3 wt% to 4.3 wt%. Consequently, the noticeable limited activity of SAPO-34 observed in Fig. 3b and 3d is explained by the competence of H 2 O with the oxygenates for its adsorption in the acid sites [51,52]. On the other hand, coke content rose up to 14.4 wt% when methanol was co-fed. Moreover, in this case, TPO profile was displaced to higher temperature (the corresponding to the highest combustion peak rate is at 500 ◦C), which is in line with a more condensed coke [30,40]. This takes place due to the higher extent of the condensation reactions of the arenes retained in the cages of SAPO-34, that will evolve into polyaromatic condensed structures (coke with a lower H/C ratio). The joint co-feeding of H 2 O and methanol resulted in an intermediate coke profile (compared to the two individual feeds), both in area (coke content 10.1 wt%) and position (combustion rate peak slightly lower than 500 ◦C), even if the negative effect of increasing methanol concentration was 0 4 8 12 16 0 1 2 3 4 5 6 0 4 8 12 16 0 1 2 3 4 5 6 0 4 8 12 16 0 1 2 3 4 5 6 Olefins Paraffins Oxygenates Methane Yield (%) Time on stream (h) a) Yield (%) Time on stream (h) b) Yield (%) Time on stream (h) c) H 2 / CO 2 H 2 / CO 2 + H 2 O H 2 / CO 2 + H 2 O + CH 3 OH H 2 / CO 2 + CH 3 OH 0 1 2 3 4 5 6 7 8 Yield (%) Olefins Paraffins Oxygenates Methane d) Fig. 3. Evolution of products yield with time on stream for H 2 /CO 2 feed (a), with water co-feeding (b), with methanol co-feeding (c) and their values at pseudosteady state of the catalyst (16 h on stream) (d). Reaction conditions: 400 ◦C; 30 bar; CO 2 /CO x , 1; H 2 /CO x , 3, and; space time, 5 g cat h mol C −1 . 200300400500600700 0.000 0.002 0.004 0.006 H2/CO2/CH3OH H2/CO2/H2O/CH3OH H2/CO2/H2O H2/CO2 ( GTD g c nim -1 g -1 cat ) Temperature ( C) 14.4% 10.1% 10.3% 4.3% Fig. 4. Effect of the different co-feedings on the SAPO-34 TPO profiles and their corresponding coke content (wt%). Reaction conditions: 400 ◦C; 30 bar; CO 2 / CO x , 1; space time, 5 g cat h mol C −1 ; H 2 /CO x , 3; and time on stream, 16 h. A. Portillo et al. Fuel 346 (2023) 128298 6 predominant, favoring coke deposition and condensation. On the other hand, the symmetry of the TPO profiles hampers the identification of different coke types by deconvolution, which is indicative of the coke being deposited in a porous structure with uniform spaces (CHA topology with 10 ×6.7 cavities Å connected by 3.8 ×3.8 Å 8-ring cages) [54]. The long reaction time (16 h) contributes to the uniformity of the coke, since for all the conditions studied, the catalyst had already reached a pseudo-steady state of constant activity, and presumably the coke also reached a pseudo-steady state of condensation. 3.2.2. H 2 /CO x feeds Considering the aforementioned difference in catalyst deactivation for H 2 feed with CO and CO 2 (especially due to the higher H 2 O generated with CO 2 via rWGS) is interesting to assess the differences among H 2 / CO, H 2 /CO/CO 2 and H 2 /CO 2 feeds, although the lower reactivity of CO 2 in the methanol synthesis should be taken into account in the comparison. Fig. 5 compares the evolution of CO x conversion with time on stream for H 2 /CO, H 2 /CO x (equimolar mixture of CO 2 and CO) and H 2 /CO 2 feeds. Here, the difference in the initial deactivation rate was evident, in the order H 2 /CO >H 2 /CO x >H 2 /CO 2 and can be explained by the increase of the H 2 O concentration in the medium with the presence of CO 2 in the feed (as it favors the greater extent of the rWGS) and its attenuating effect on coke formation. Moreover, this effect was also observed for different operating conditions (different In 2 O 3 -ZrO 2 /SAPO-34 ratios, operating temperatures, pressures and space time values) (results not shown). The yields obtained after 16 h on stream, i.e., at pseudo-steady state, are depicted in Fig. 6. The results were obtained with runs with CO 2 /CO x ratios in the feed of: 0, 0.25, 0.5, 0.75 and 1. The maximum olefin yield value was achieved with a CO 2 /CO x ratio of 0.5, although the difference with the feeds with lower CO 2 amount was not remarkable. It is observed that, when increasing the CO 2 amount in the feed up to that value, the increase in H 2 O concentration (as product of the rWGS reaction) enhanced the process by attenuating the deactivation. Indeed, with a CO 2 /CO x ratio of 0.5, in pseudo-steady state, SAPO-34 maintained a sufficient remaining activity for the complete conversion of oxygenates into olefins, avoiding their cracking into CH 4 . The slight decrease of olefin yield for feeds with higher CO 2 content suggests that the effect of the lower reactivity of CO 2 with respect to CO [55] predominated over the expected higher attenuation of deactivation. The effect of co-feeding H 2 O and methanol together with H 2 /CO x on catalyst deactivation, for a CO 2 /CO x ratio of 0.5, was also studied. The results of evolution of the products with the time on stream revealed an intermediate trend to those observed previously for the H 2 /CO and H 2 / CO 2 feeds. Thus, the effect of co-feeding H 2 O on the yields at the pseudosteady state of the catalyst (Fig. 7) was in line with the aforementioned results. However, the unfavorable effects observed for the H 2 /CO 2 feed predominated, due to the high H 2 O concentration (generated by the rWGS reaction) that diminished the activity of the acid sites. According to this, H 2 O co-feeding also inhibited olefin formation. On the other hand, the effect of increasing methanol concentration, especially without the compensation of co-feeding H 2 O, was decisive to intensify the deactivation rate by coking, resulting in the inhibition of methanol conversion into olefins. 3.3. H 2 /CO x ratio in the feed The effect of this variable was analyzed separately considering that, in addition to affecting the deactivation of the In 2 O 3 -ZrO 2 /SAPO-34 tandem catalyst (leading to pseudo-steady states of different interest), it also affects directly the kinetics of the methanol/DME synthesis reactions. This direct effect will mask the results explained in the Section 3.2 about the concentration of H 2 O and oxygenates. 0 4 8 12 16 0 2 4 6 8 10 12 X COX (%) Time on stream (h) H 2 /CO H 2 /CO/CO 2 H 2 /CO 2 Fig. 5. Effect of CO 2 /CO x ratio in the feed (0, 0.5 and 1) on the evolution of CO x conversion with time on stream. Reaction conditions: 400 ◦C; 30 bar; H 2 / CO x , 3; and space time, 5 g cat h mol C −1 . 0 25 50 75 100 0 2 4 6 CO 2 /CO X (%) Olefins Paraffins Oxygenates Methane )%( dleiY Fig. 6. Effect of CO 2 /CO x ratio in the feed on products yield. Reaction conditions: 400 ◦C; 30 bar; H 2 /CO x , 3; space time, 5 g cat h mol C −1 ; and time on stream, 16 h. H 2 CO/CO 2 H 2 CO/CO 2 + H 2 O H 2 CO/CO 2 + H 2 O + CH 3 OH H 2 CO/CO 2 + CH 3 OH 0 1 2 3 4 5 6 7 8 )%( dleiY Olefins Paraffins Oxygenates Methane Fig. 7. Effect of different co-feedings on products yield. Reaction conditions: 400 ◦C; 30 bar; CO 2 /CO x , 0.5; H 2 /CO x , 3, 5 g cat h mol C −1 ; space time, 5 g cat h mol C −1 ; and time on stream, 16 h. A. Portillo et al. Fuel 346 (2023) 128298 7 The results of the evolution of products yields with time on stream shown in Fig. 8 correspond to feeds with H 2 /CO x molar ratio of 1, 2 and 3 (Fig. 8a, 8b and 8c, respectively). It is evidenced that the higher H 2 partial pressure augmented the initial olefin yield, but also favored their hydrogenation into paraffins. Moreover, the catalyst deactivation was remarkably attenuated when increasing H 2 partial pressure due to the hydrogenation of coke precursors [31,38]. As a consequence, by increasing the H 2 /CO x ratio, the catalyst maintained a higher remaining activity in the pseudo-steady state and the CO 2 and CO x conversions values were higher (Fig. 8d). The attenuation of deactivation with increasing H 2 concentration may be attributed to the effect on the content and nature of coke in the SAPO-34 catalyst. In the TPO profiles gathered in Fig. 9, it is observed that at higher H 2 /CO x ratio, the coke content decreased. Furthermore, although the symmetry of the profiles evidenced the aforementioned uniformity of coke, the different extent of the evolution towards condensed structures is noteworthy. The effect of H 2 /CO x ratio in Fig. 9 is moderate because in all three conditions the partial pressure of H 2 was high enough to hydrogenate coke precursors [40], and consequently to difficult its condensation into polyaromatic structures. These hydrogenation reactions presumably are activated by the In 2 O 3 -ZrO 2 catalyst sites in contact with the SAPO-34 catalyst. It is noteworthy that the temperature corresponding to the maximum combustion rate in the TPO profiles was low (in the range of 450–500 ◦C), indicating relatively little condensed coke. Moreover, it should be noted that it was proved in a previous work that coke is partially eliminated by means of an inert gas stream [46]. These results are in good agreement with the literature of the MTO and DTO processes, in which the coke is relatively little condensed under usual conditions (atmospheric pressure) [30,35,56]. The increase in the H 2 /CO x ratio in the feed on coke content and nature led to the lower deactivation rate of the catalysts and its greater remaining activity in the pseudo-steady state in Fig. 8, due to the greater ease of the reactants to access the cages of the SAPO-34. Based on these results, an H 2 /CO x ratio above 3 could be considered adequate to further mitigate deactivation. However, H 2 availability must be considered to establish an optimal H 2 /CO x value. Accordingly, a value of 3 at 30 bar was set as appropriate, achieving a good compromise between olefin yield-selectivity and catalyst stability [41], bearing in mind that under 0 4 8 12 16 0 1 2 3 4 5 6 0 4 8 12 16 0 1 2 3 4 5 6 0 4 8 12 16 0 1 2 3 4 5 6 Paraffins Olefins Methane Oxygenates Yield (%) Time on stream (h) a) Yield (%) Time on stream (h) b) Yield (%) Time on stream (h) c) 123 0 5 10 15 20 25 30 Conversion (%) H 2 / CO X CO 2 CO X d) Fig. 8. Evolution of products yields with time on stream for H 2 /CO x values in the feed of 1 (a), 2 (b) and 3 (c), and CO 2 and COx conversion values at the pseudosteady state of the catalyst (16 h on stream) (d). Reaction conditions: 400 ◦C; 30 bar; CO 2 /CO x 0.5, and; space time 5 g cat h mol C −1 . 100200300400500600700 0.000 0.002 0.004 0.006 ( GTD gcnim 1g1cat) Temperature ( C) H 2 /CO x 1/1 2/1 3/1 14.2% 12.5% 11.5% Fig. 9. Effect of H 2 /CO x ratio in the feed on the TPO profiles of spent SAPO-34 and the corresponding coke content (wt%). Reaction conditions: 400 ◦C; 30 bar; CO 2 /CO x , 0.5; space time, 5 g cat h mol C −1 , and; time on stream, 16 h. A. Portillo et al. Fuel 346 (2023) 128298 8 these conditions H 2 produced with PEM electrolyzers could be used [57]. 4. Conclusions The rapid deactivation by coke of the SAPO-34 catalyst conditions the performance of the In 2 O 3 -ZrO 2 /SAPO-34 tandem catalyst in the direct synthesis of olefins from CO 2 /CO and the yield of olefins and byproducts (paraffins and CH 4 ) in the subsequent pseudo-steady state. The increase in H 2 concentration resulted in lower coke deposition and of less condensed nature, which contributes to lessen de deactivation rate and to increase the remaining activity of the catalyst in the pseudo-steady state. However, the role of H 2 O and oxygenates (methanol/DME) concentration on the deactivation of the SAPO-34 was complex, due to the implication of these compounds (reaction intermediates) in the mechanism of the main and deactivation reactions involved in the conversion of the oxygenates. Increasing H 2 O concentration lessened the deactivation of the SAPO-34 catalyst for H 2 /CO feed, boosting olefins yield. However, this did not occur for the H 2 /CO 2 feed, because the H 2 O concentration was too high at this condition, and its negative effect of attenuating the activity of the acid sites of the SAPO-34 predominated. In addition, the excessive concentration of methanol/DME, desired as reactants for olefin production, had the counterpart of favoring deactivation by coke, and consequently the presence of DME and CH 4 as by-products. These effects of H 2 O and methanol concentration affect the performance of the tandem In 2 O 3 -ZrO 2 /SAPO-34 catalyst in the conversion to olefins of CO 2 /CO mixtures, in which the H 2 O concentration in the feed would also influence the effect of other operating conditions (temperature, pressure, space time, In 2 O 3 -ZrO 2 /SAPO-34 ratio) on the deactivation. Consequently, when studying the effect of the process conditions for the interpretation of the results, their effect on the H 2 O and oxygenates concentrations must be taken into account. Moreover, the incidence of H 2 O and oxygenates concentrations should be taken into account in the development of kinetic models for this process, and in particular, in the kinetics of deactivation by coke. These effects are also of interest to progress in the design of new catalysts (more stable in the presence of H 2 O and with lower coke deposition) and reactors. Among the initiatives to modify the acid zeotype, the modulation of the acid strength and the generation of hierarchical porous structures (with mesopores inside the crystals or agglomerating the zeotype in a matrix) may contribute to attenuate coke deposition. To minimize the attenuation of the activity by the concentration of H 2 O, it would be reasonable to use less hydrophilic zeotypes and their incorporation into composites with H 2 O adsorbing materials. From the perspective of improving the reactor, hydrophilic membrane reactors may be a good solution to avoid the unfavorable effect of excess H 2 O with CO 2 /CO feeds. CRediT authorship contribution statement A. Portillo: Conceptualization, Investigation, Writing – original draft, Writing – review & editing. O. Parra: Validation, Visualization, Writing – original draft, Writing – review & editing. J. Ere˜ na: Project administration, Funding acquisition. A.T. Aguayo: Methodology, Resources, Supervision, Project administration, Funding acquisition. J. Bilbao: Conceptualization, Writing – original draft, Writing – review & editing, Project administration, Funding acquisition. A. Ateka: Conceptualization, Writing – original draft, Writing – review & editing. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Data availability The authors are unable or have chosen not to specify which data has been used. Acknowledgements This work has been carried out with the financial support of the Ministry of Science, Innovation and Universities of the Spanish Government (PID2019-108448RB-100); the Basque Government (Project IT1645-22), the European Regional Development Funds (ERDF) and the European Commission (HORIZON H2020-MSCA RISE-2018. Contract No. 823745). A. Portillo is grateful for the Ph.D. grant from the Ministry of Science, Innovation and Universities of the Spanish Government (BES2017-081135) and O. Parra is grateful for the PhD grant from the Basque Government (PRE_2021_1_0014). The authors thank for technical and human support provided by SGIker (UPV/EHU). Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.fuel.2023.128298. References [1] Zhou W, Cheng K, Kang J, Zhou C, Subramanian V, Zhang Q, et al. New horizon in C1 chemistry: breaking the selectivity limitation in transformation of syngas and hydrogenation of CO 2 into hydrocarbon chemicals and fuels. Chem Soc Rev 2019; 48:3193–228. https://doi.org/10.1039/c8cs00502h. [2] Ramirez A, Sarathy SM, Gascon J. CO 2 derived E-fuels: research trends, Misconceptions, and Future Directions. Trends Chem 2020;2:785–95. https://doi. org/10.1016/j.trechm.2020.07.005. [3] Cui X, Zhuang Y, Dong H, Du J. Multi-criteria screening of carbon dioxide utilization products combined with process optimization and evaluation. Fuel 2022;328:125319. https://doi.org/10.1016/j.fuel.2022.125319. [4] Cortazar M, Santamaria L, Lopez G, Alvarez J, Amutio M, Bilbao J, et al. Fe/olivine as primary catalyst in the biomass steam gasification in a fountain confined spouted bed reactor. J Ind Eng Chem 2021;99:364–79. https://doi.org/10.1016/j. jiec.2021.04.046. [5] Landa L, Remiro A, de la Torre R, Aguado R, Bilbao J, Gayubo AG. Global vision from the thermodynamics of the effect of the bio-oil composition and the reforming strategies in the H 2 production and the energy requirement. Energy Convers Manag 2021;239:114181. https://doi.org/10.1016/j.enconman.2021.114181. [6] Moreira R, Bimbela F, Gil-Lalaguna N, S´ anchez JL, Portugal A. Clean syngas production by gasification of lignocellulosic char: State of the art and future prospects. J Ind Eng Chem 2021;101:1–20. https://doi.org/10.1016/j. jiec.2021.05.040. [7] Rosha P, Rosha AK, Ibrahim H, Kumar S. Recent advances in biogas upgrading to value added products: A review. Int J Hydrogen Energy 2021;46:21318–37. https://doi.org/10.1016/j.ijhydene.2021.03.246. [8] De S, Dokania A, Ramirez A, Gascon J. Advances in the design of heterogeneous catalysts and thermocatalytic processes for CO 2 utilization. ACS Catal 2020;10: 14147–85. https://doi.org/10.1021/acscatal.0c04273. [9] Zhou Z, Gao P. Direct carbon dioxide hydrogenation to produce bulk chemicals and liquid fuels via heterogeneous catalysis, Chinese. J Catal 2022;43:2045–56. https://doi.org/10.1016/s1872-2067(22)64107-x. [10] Saeidi S, Najari S, Hessel V, Wilson K, Keil FJ, Concepci´ on P, et al. Recent advances in CO 2 hydrogenation to value-added products — Current challenges and future directions. Prog Energy Combust Sci 2021;85:100905. https://doi.org/10.1016/j. pecs.2021.100905. [11] Sharma P, Sebastian J, Ghosh S, Creaser D, Olsson L. Recent advances in hydrogenation of CO 2 into hydrocarbons via methanol intermediate over heterogeneous catalysts, Catal. Sci Technol 2021;11:1665–97. https://doi.org/ 10.1039/d0cy01913e. [12] Guo S, Wang H, Qin Z, Li Z, Wang G, Dong M, et al. Feasibility, limit, and suitable reaction conditions for the production of alcohols and hydrocarbons from CO and CO 2 through hydrogenation, a thermodynamic consideration. Ind Eng Chem Res 2022;61:17027–38. https://doi.org/10.1021/acs.iecr.2c02898. [13] Wang G, Zeng L, Cao J, Liu F, Lin Q, Yi Y, et al. Highly selective conversion of CO 2 to hydrocarbons over composite catalysts of ZnO-ZrO 2 and SAPO-34. Microporous Mesoporous Mater 2019;284:133–40. [14] Ticali P, Salusso D, Ahmad R, Ahoba-Sam C, Ramirez A, Shterk G, et al. CO 2 hydrogenation to methanol and hydrocarbons over bifunctional Zn-doped ZrO 2 / zeolite catalysts, Catal. Sci Technol 2021;11:1249–68. https://doi.org/10.1039/ d0cy01550d. [15] Liu Z, Ni Y, Hu Z, Fu Y, Fang X, Jiang Q, et al. Insights into effects of ZrO 2 crystal phase on syngas-to-olefin conversion over ZnO/ZrO 2 and SAPO-34 composite A. Portillo et al. Fuel 346 (2023) 128298 9 catalysts, Chinese. J Catal 2022;43:877–84. https://doi.org/10.1016/s1872-2067 (21)63908-6. [16] Numpilai T, Wattanakit C, Chareonpanich M, Limtrakul J, Witoon T. Optimization of synthesis condition for CO 2 hydrogenation to light olefins over In 2 O 3 admixed with SAPO-34. Energy Convers Manag 2019;180:511–23. https://doi.org/ 10.1016/j.enconman.2018.11.011. [17] Ghosh S, Olsson L, Creaser D. Methanol mediated direct CO 2 hydrogenation to hydrocarbons: Experimental and kinetic modeling study. Chem Eng J 2022;435: 135090. https://doi.org/10.1016/j.cej.2022.135090. [18] Wang S, Wang P, Qin Z, Yan W, Dong M, Li J, et al. Enhancement of light olefin production in CO 2 hydrogenation over In 2 O 3 -based oxide and SAPO-34 composite. J Catal 2020;391:459–70. https://doi.org/10.1016/j.jcat.2020.09.010. [19] Dang S, Gao P, Liu Z, Chen X, Yang C, Wang H, et al. Role of zirconium in direct CO 2 hydrogenation to lower olefins on oxide/zeolite bifunctional catalysts. J Catal 2018;364:382–93. https://doi.org/10.1016/j.jcat.2018.06.010. [20] Gao P, Dang S, Li S, Bu X, Liu Z, Qiu M, et al. Direct Production of lower olefins from CO 2 conversion via bifunctional catalysis. ACS Catal 2018;8:571–8. https:// doi.org/10.1021/acscatal.7b02649. [21] Tan L, Zhang P, Cui Y, Suzuki Y, Li H, Guo L, et al. Direct CO 2 hydrogenation to light olefins by suppressing CO by-product formation. Fuel Process Technol 2019; 196:106174. https://doi.org/10.1016/j.fuproc.2019.106174. [22] Dang S, Li S, Yang C, Chen X, Li X, Zhong L, et al. Selective transformation of CO 2 and H 2 into lower olefins over In 2 O 3 -ZnZrO x /SAPO-34 bifunctional catalysts. ChemSusChem 2019;12:3582–91. https://doi.org/10.1002/cssc.201900958. [23] Chen J, Wang X, Wu D, Zhang J, Ma Q, Gao X, et al. Hydrogenation of CO 2 to light olefins on CuZnZr@(Zn-)SAPO-34 catalysts: Strategy for product distribution. Fuel 2019;239:44–52. https://doi.org/10.1016/j.fuel.2018.10.148. [24] Zhang W, Wang S, Guo S, Qin Z, Dong M, Fan W, et al. GamCrOx/H-SAPO-34(F), a highly efficient bifunctional catalyst for the direct conversion of CO 2 into ethene and propene. Fuel 2022;329:125475. https://doi.org/10.1016/j. fuel.2022.125475. [25] P´ erez-Uriarte P, Ateka A, Aguayo AT, Bilbao J. Comparison of HZSM-5 zeolite and SAPO (-18 and-34) based catalysts for the production of light olefins from DME. Catal Lett 2016;146:1892–902. https://doi.org/10.1007/s10562-016-1829-z. [26] Luo M, Hu B, Mao G, Wang B. Trace compounds confined in SAPO-34 and a probable evolution route of coke in the MTO process. ACS Omega 2022;7:3277–83. https://doi.org/10.1021/acsomega.1c05336. [27] Ib´ a˜ nez M, P´ erez-Uriarte P, S´ anchez-Contador M, Cordero-Lanzac T, Aguayo AT, Bilbao J, et al. Nature and location of carbonaceous species in a composite HZSM-5 zeolite catalyst during the conversion of dimethyl ether into light olefins. Catalysts 2017;7:254. https://doi.org/10.3390/catal7090254. [28] Müller S, Liu Y, Vishnuvarthan M, Sun X, Van Veen AC, Haller GL, et al. Coke formation and deactivation pathways on H-ZSM-5 in the conversion of methanol to olefins. J Catal 2015;325:48–59. https://doi.org/10.1016/j.jcat.2015.02.013. [29] P´ erez-Uriarte P, Gamero M, Ateka A, Díaz M, Aguayo AT, Bilbao J. Effect of the acidity of HZSM-5 zeolite and the binder in the DME transformation to olefins. Ind Eng Chem Res 2016;55:1513–21. https://doi.org/10.1021/acs.iecr.5b04477. [30] Cordero-Lanzac T, Ateka A, P´ erez-Uriarte P, Casta˜ no P, Aguayo AT, Bilbao J. Insight into the deactivation and regeneration of HZSM-5 zeolite catalysts in the conversion of dimethyl ether to olefins. Ind Eng Chem Res 2018;57:13689–702. https://doi.org/10.1021/acs.iecr.8b03308. [31] Zhao X, Li J, Tian P, Wang L, Li X, Lin S, et al. Achieving a superlong lifetime in the zeolite-catalyzed MTO reaction under high pressure: synergistic effect of hydrogen and water. ACS Catal 2019;9:3017–25. https://doi.org/10.1021/ acscatal.8b04402. [32] Tian P, Wei Y, Ye M, Liu Z. Methanol to olefins (MTO): From fundamentals to commercialization. ACS Catal 2015;5:1922–38. https://doi.org/10.1021/ acscatal.5b00007. [33] Cordero-Lanzac T, Aguayo AT, Gayubo AG, Bilbao J. Consideration of the activity distribution using the population balance theory for designing a dual fluidized bed reactor-regenerator system. Application to the MTO process. Chem Eng J 2021; 405:126448. https://doi.org/10.1016/j.cej.2020.126448. [34] P´ erez-Uriarte P, Ateka A, Aguayo AT, Gayubo AG, Bilbao J. Kinetic model for the reaction of DME to olefins over a HZSM-5 zeolite catalyst. Chem Eng J 2016;302: 801–10. https://doi.org/10.1016/j.cej.2016.05.096. [35] P´ erez-Uriarte P, Ateka A, Gayubo AG, Cordero-Lanzac T, Aguayo AT, Bilbao J. Deactivation kinetics for the conversion of dimethyl ether to olefins over a HZSM-5 zeolite catalyst. Chem Eng J 2017;311:367–77. https://doi.org/10.1016/j. cej.2016.11.104. [36] Cordero-Lanzac T, Martínez C, Aguayo AT, Casta˜ no P, Bilbao J, Corma A. Activation of n-pentane while prolonging HZSM-5 catalyst lifetime during its combined reaction with methanol or dimethyl ether. Catal Today 2022;383:320–9. https://doi.org/10.1016/j.cattod.2020.09.015. [37] Arora SS, Nieskens DLS, Malek A, Bhan A. Lifetime improvement in methanol-toolefins catalysis over chabazite materials by high-pressure H 2 co-feeds. Nat Catal 2018;1:666–72. https://doi.org/10.1038/s41929-018-0125-2. [38] Xie J, Firth DS, Cordero-Lanzac T, Airi A, Negri C, Øien-Ødegaard S, et al. MAPO18 catalysts for the methanol to olefins process: influence of catalyst acidity in a high-pressure syngas (CO and H 2 ) environment. ACS Catal 2022;12:1520–31. https://doi.org/10.1021/acscatal.1c04694. [39] Arora SS, Shi Z, Bhan A. Mechanistic basis for effects of high-pressure H 2 cofeeds on methanol-to-hydrocarbons catalysis over zeolites. ACS Catal 2019;9:6407–14. https://doi.org/10.1021/acscatal.9b00969. [40] Nieskens DLS, Lunn JD, Malek A. Understanding the enhanced lifetime of SAPO-34 in a direct syngas-to-hydrocarbons process. ACS Catal 2019;9:691–700. https:// doi.org/10.1021/acscatal.8b03465. [41] Portillo A, Ateka A, Ere˜ na J, Aguayo AT, Bilbao J. Conditions for the joint conversion of CO 2 and syngas in the direct synthesis of light olefins using In 2 O 3 –ZrO 2 /SAPO-34 catalyst. Ind Eng Chem Res 2022;2021:10365–76. https:// doi.org/10.1021/acs.iecr.1c03556. [42] Portillo A, Ateka A, Ere˜ na J, Bilbao J, Aguayo AT. Role of Zr loading into In 2 O 3 catalysts for the direct conversion of CO 2 /CO mixtures into light olefins. J Environ Manage 2022;316:115329. https://doi.org/10.1016/j.jenvman.2022.115329. [43] Gao J, Jia C, Liu B. Direct and selective hydrogenation of CO 2 to ethylene and propene by bifunctional catalysts, Catal. Sci Technol 2017;7:5602–7. https://doi. org/10.1039/c7cy01549f. [44] Chen TY, Cao C, Chen TB, Ding X, Huang H, Shen L, et al. Unraveling highly tunable selectivity in CO 2 hydrogenation over bimetallic In-Zr Oxide catalysts. ACS Catal 2019;9:8785–97. https://doi.org/10.1021/acscatal.9b01869. [45] Frei MS, Mondelli C, Cesarini A, Krumeich F, Hauert R, Stewart JA, et al. Role of zirconia in indium oxide-catalyzed CO 2 hydrogenation to methanol. ACS Catal 2020;10:1133–45. https://doi.org/10.1021/acscatal.9b03305. [46] Portillo A, Ateka A, Ere˜ na J, Bilbao J, Aguayo A. Alternative acid catalysts for the stable and selective direct conversion of CO2/CO mixtures into light olefins. Fuel Process. Technol. 2022;238:107513. https://doi.org/10.1016/j. fuproc.2022.107513. [47] Aguayo AT, del Campo AES, Gayubo AG, Tarrío A, Bilbao J. Deactivation by coke of a catalyst based on a SAPO-34 in the transformation of methanol into olefins. J Chem Technol Biotechnol 1999;74:315–21. https://doi.org/10.1002/(sici)10974660(199904)74:4<315::aid-jctb34>3.0.co;2-g. [48] Aguayo AT, Gayubo AG, Vivanco R, Olazar M, Bilbao J. Role of acidity and microporous structure in alternative catalysts for the transformation of methanol into olefins. Appl Catal A Gen 2005;283:197–207. https://doi.org/10.1016/j. apcata.2005.01.006. [49] Chen J, Li J, Wei Y, Yuan C, Li B, Xu S, et al. Spatial confinement effects of cagetype SAPO molecular sieves on product distribution and coke formation in methanol-to-olefin reaction. Catal Commun 2014;46:36–40. https://doi.org/ 10.1016/j.catcom.2013.11.016. [50] Gayubo AG, Aguayo AT, S´ anchez Del Campo AE, Tarrío AM, Bilbao J. Kinetic modeling of methanol transformation into olefins on a SAPO-34 catalyst. Ind Eng Chem Res 2000;39:292–300. https://doi.org/10.1021/ie990188z. [51] Zapater D, Lasobras J, Soler J, Herguido J, Men´ endez M. Comparison of conventional and two-zone fluidized bed reactors for methanol to olefins. Effect of reaction conditions and the presence of water in the feed. Ind Eng Chem Res 2022; 61:5757–65. [52] Palomo J, Rodríguez-Cano MA, Berruezo-García J, Rodríguez-Mirasol J, Cordero T. Efficient methanol dehydration to DME and light hydrocarbons by submicrometric ZrO2-ZSM-5 fibrillar catalysts with a shell-like structure. Fuel 2022;315:123283. https://doi.org/10.1016/j.fuel.2022.123283. [53] P´ erez-Uriarte P, Ateka A, Gamero M, Aguayo AT, Bilbao J. Effect of the operating conditions in the transformation of DME to olefins over a HZSM-5 zeolite catalyst. Ind Eng Chem Res 2016;55:6569–78. https://doi.org/10.1021/acs.iecr.6b00627. [54] Hemelsoet K, Van Der Mynsbrugge J, De Wispelaere K, Waroquier M, Van Speybroeck V. Unraveling the reaction mechanisms governing methanol-to-olefins catalysis by theory and experiment. ChemPhysChem 2013;14:1526–45. https:// doi.org/10.1002/cphc.201201023. [55] Nielsen ND, Jensen AD, Christensen JM. The roles of CO and CO 2 in high pressure methanol synthesis over Cu-based catalysts. J Catal 2021;393:324–34. https://doi. org/10.1016/j.jcat.2020.11.035. [56] Gayubo AG, Aguayo AT, Alonso A, Bilbao J. Kinetic modeling of the methanol-toolefins process on a silicoaluminophosphate (SAPO-18) catalyst by considering deactivation and the formation of individual olefins. Ind Eng Chem Res 2007;46: 1981–9. https://doi.org/10.1021/ie061278o. [57] Shaner MR, Atwater HA, Lewis NS, McFarland EW. A comparative technoeconomic analysis of renewable hydrogen production using solar energy. Energy Environ Sci 2016;9:2354–71. A. Portillo et al.