13th International Colloquium Fuels - September 2021 93 Improve by Ind 4.0 the Heterogeneous Catalysts for CO 2 Hydrogenation to Liquid Fuels Dr. rer. nat. Paul Olaru AGIR-SETEC Head Lab. IMNR-INCDMNR Bucharest, Romania Summery The process for methanol synthesis using captured CO2 was shown by a system engineering analysis based on planetary boundaries. Reverse water-gas shift reaction (RWGS: CO2 + H2 → CO + H2O) is mostly determined by the heterolytic splitting of H2 on the oxygen vacancies acting as catalytic center Its thermocatalytic production (CO2 + 3H2 → CH3OH + H2O) from captured CO2 and renewable H2 or CO2-rich feeds obtained via technology gasification is an attractive approach to effectively combat global warming. Today, the large-scale commercial production of methanol is mainly from syngas (a CO and H2 mixture), which are generated by fossil resources (mainly coal and natural gas) via processes including the gasification of coal as well as the steam reforming of natural gas. Small amounts of CO2 (about 2-8%) are typically added to the CO/ H2 stream to balance the H/ C ratio to the desired stoichiometry and to accelerate the reaction rate. Methanol formation: (1) CO + 2H2 = CH3OH, ΔH298K = ̶ 90.6 kJ mol¯1; (2) CO2 + 3H2 = CH3OH + H2O, ΔH298K = ̶ 49.5 kJ mol¯1; Reverse water-gas-shift reaction (RWGS): (3) CO2 + H2 = CO + H2O, ΔH298K = ̶ 41.2 kJ mol¯1. 1. Introduction Industry 4.0 from talk to practice. Digital trust and data analyses are the foundation of Ind 4.0 Data fuels Industry 4.0 and successful data analytics is the prerequisite for successful implementation of digital enterprise applications. It’s time to move from a phase of discovery and understanding what data is available and what it is worth to one of insights and action. ‘First movers’ are already making the shift and using data analytics to help drive decision-making. As digital ecosystems expand, so does the importance of establishing strong levels of digital trust, backed up by transparency and non-repudiation that provides proof of the integrity and origin of one’s own and third party data. More than half of respondents expect ( fig.1) their Industry 4.0 investments to yield a return within two years or less, given investment of around 5% p.a. of annualrevenue. Fig. 1 2. Process In comparing the commercial production of methanol via syngas ( see eq. (1)), methanol formation from CO 2 hydrogenation requires an extra amount of hydrogen since extra H 2 is necessary to remove one oxygen atom from CO 2 through water formation as a by-product (see eq. (2)). Moreover, the thermodynamics for methanol production from CO 2 are not as favorable as that from CO, thus the one-pass methanol yield of the CO 2 -based process is lower than that of the syngas-based process. 94 13th International Colloquium Fuels - September 2021 Improve by Ind 4.0 the Heterogeneous Catalysts for CO2 Hydrogenation to Liquid Fuels The equilibrium yield of methanol from CO at 200ºC is greater than 80%, while that from CO 2 at 200ºC is less than 40%. In addition, several side reactions (e.g. the formation ethers, ketones, higher alcohols or hydrocarbons). Methanol synthesis from CO 2 and H 2 is an exothermal reaction with a decrease in reaction molecule number, according to Le Chatelier’s principle, the optimization of the reaction conditions such as high pressure and low temperature thermodynamically favors CO 2 conversion to methanol. However, taking the chemically inert nature of CO 2 and the reaction rate into consideration, a reaction temperature higher than 242°C that favors the CO 2 activation and methanol formation is usually adopted. In addition, under methanol synthesis conditions, the RWGS reaction becomes highly facile in thermodynamics. 2.1 Hydrogenation CO 2 to Methanol Recently, significant progress has been made indeveloping more efficient catalysts for CO2 hydrogenation to CH3OH, including metal-supported catalysts,bimetallic systems, and reducible metal oxides. During CO2 hydrogenation, the formation of the undesired CO through RWGS is a competitive reaction to methanol synthesis. However, the low temperature and/ or high space velocity usually result in low single pass CO2 conversion. Thus, it remains a great challenge to simultaneously obtain high CO2 conversion and high methanol selectivity. Industrial methanol production from CO2-containing syngas uses the well-known Cu-ZnO-Al2O3 catalysts. Currently, CH3OH synthesis from catalytic CO2 hydrogenation has been implemented at the pilot-plant level. In Europe, power-to-liquid (PTL) fuels obtained from surplus renewable electricity via water electrolysis-derived H2 and CO2 have received much attention, and a number of PTL-related studies were recently reported. Liquid hydrocarbon (C5+) fuels including gasoline (C5-11), jet fuel (C8-16), and diesel (C10-20) play an instrumental role in the global energy supply chain and are wildly used as transportation fuels around the world. Besides, alcohols such as methanol, ethanol, and other higher alcohols (C2+OH) are clean and multipurpose fuels.In both academic researchand industrial practices, (fig.2), great progress has been made in the synthesis of C1 molecules such as methanol (CH3OH) from CO2 hydrogenation. Currently, the largest plant, Carbon Recycling International (CRI)’s CO2-to-renewable-methanol plant in Iceland, is capable of producing 4000 t/ year of methanol by converting about 5500 t/ year of CO2.However, direct reduction of CO2 to C2+ products is still a grand challenge due to the lack of efficient catalysts with high stability, as the activity of C-C coupling is low and the formation of byproduct water can easily deactivatethevarious catalysts for CO2 conversion. Syngas (CO/ H2) and CH3OH are the most important C1 platform molecules, and their conversions to value-added products via the Fischer-Tropsch synthesis (FTS) and methanol to hydrocarbons (MTH) processes, respectively, were extensively applied in industry.Therefore, combining the reverse water-gas shift (RWGS) with FTS and combining high-temperature methanol synthesis with MTH over bifunctional/ multifunctional catalysts are two efficient strategies for direct CO2 hydrogenation to C2+ hydrocarbons including liquid hydrocarbons. Moreover, the synthesis of C2+ alcohols is even more challenging than C2+ hydrocarbons either in CO2 or CO hydrogenation, which requires more precise control of the C-C coupling. Fig. 2 The supported copper materials have attracted much attention and have been extensively investigated for CO2 hydrogenation to CH3OH. The methanol synthesis reaction is known to show strong support effects. Using a suitable support material can enhance CH3OH selectivity, attributed to structural, electronic, and chemical promotional effects. For example, CH3OH selectivity is usually high (>70%) at 200-260 °C when ZrO2 is used as the support.[30,32]. Recently, several works have clarified the origin of the promotional effect of the ZrO2 support that is unknown before. Larmier et al. confirmed that the Cu/ ZrO2 interface can promote the conversion of the formate intermediates to methanol by combined experimental and computational investigations.[30]. Tada et al. found that the interfacial sites on Cu/ a-ZrO2 (a-: amorphous) are favored for methanol production compared to those on tetragonal and monoclinic ZrO2 supported Cu.[33]. Additionally, the oxygen vacancies on tetragonal ZrO2 were proposed to play a crucial role in enhancing the activity of methanol formation by stabilizing the Cu+ active sites adjacent to them.[32]. Lam et al. suggested that the surface Lewis acid Zr(IV) sites with Cu particles in the vicinity are responsible for improving CH3OH activity and selectivity on the Cu/ ZrO2-based catalysts.[31]. Apart from conventional metal oxides, researchers recently also explored other supports such as TiO2 nanotubes [34] and Mg-Al layered double hydroxide (LDH) [35] to promote methanol formation. To simultaneously improve the intrinsic activity and catalyst stability, researchers directly used Cu-Zn-based LDH as the precursor to synthesize an efficient methanol synthesis catalyst with a confined structure, in which the ac- 13th International Colloquium Fuels - September 2021 95 Improve by Ind 4.0 the Heterogeneous Catalysts for CO2 Hydrogenation to Liquid Fuels tive metallic Cu phase is highly dispersed and partially embedded in the remaining oxide matrix.[36,29] Li et al. prepared ultrathin Cu-Zn-Ga LDH nanosheets by the aqueous miscible organic solvent treatment method and further increased Cu surface areas and dispersion of the resulting Cu-based catalysts (fig. 1A).[37].The CH3OH space time yield (STY) is 0.59 g MeOH g cat -1 h -1 with selectivity of ~49% at CO2 conversion of ~20%. On the other hand, scientists also tried to use silica and metal-organic frameworks to confine small Cu nanoparticles and developed highly efficient Cu-based catalysts for CO2 hydrogenation.[40-44].Example, Liu et al. synthesized three-dimensional porous Cu@ZrO2 framework catalysts using the Cu@UiO-66 precursor, which displays CH3OH STY of 0.796 g MeOH g cat -1 h -1 and a long-term stability for 105 h with a selectivity of 78.8% and conversion of 13.1% at 258°C, 4.5 MPa, 21 600 mL g -1 h -1 and H2/ CO2 = 3 [43].Recently, several groups have developed novel cobalt (Co)based catalytic systems that display a promising performance for methanol synthesis from CO2 hydrogenation.[45-48]. Co catalysts are widely used and extensively studied for FTS and can easily catalyze the CO2 methanation reaction.[2,17,46,49-52]. The high performance for CH3OH synthesis over Cobased catalysts was attributed to the formation of a new active phase, rather than the conventional metallic Co phase. Our works [7],were conducted to predict theeffect of operating conditions on syngas conversionand temperature profiles of the cobalt-based FTSreaction in the FT reactor. The kinetic modelproposed in the present study was shown to beeffective for application to a commercial softwarepackage (Multiphysics). Li et al. Reported MnOx nanoparticles supported on a mesoporous Co3O4 for methanol production at low pressure (0.1-0.6 MPa). [45].They showed the importance of the CoO phase and revealed the strong interaction between MnOx nanoparticles and Co@CoO core-shell grains, which enhances the performance of CO2 hydrogenation to CH3OH. Lian et al. ascribed the increased CH3OH selectivity to the oxygen defects on the surface of Co@Co3O4 core-shell active species supported on the nitrogen-doped carbon material, which is derived from zeolitic imidazolate framework precursors.[47].Wang et al. Also inferred that the cobalt oxide phase on silica supported Co catalyst with Co-O-SiOn linkages via Co phyllosilicates could suppress the CO and CH4 formation and promote methanol production (fig.3B) [48]. This novel Co catalyst shows CH3OH STY of 0.096 g MeOH g cat -1 h -1 (3.0 mmol g cat -1 h -1 ) with a selectivity of 70.5% at 8.6% conversion at 320 °C. Because the H2 splitting ability of cobalt oxide is lower than the metallic Co phase, the activities of these Co-based methanol synthesis catalysts are to be further enhanced. Moreover, the mechanism of this new active site inhibiting the CO2 methanation and RWGS reactions needs to be further clarified. Various bimetallic materials including: Pd-Cu,[3,5], Pd- Ga,[6,17], Pd-In,[6,8], Pd- Zn,[9], Ni-Ga,[2.6], In-Rh,[17], In-Cu,[8,9], In-Co,[7], In Ni,[11,12] and Ni-Cu [15-17], have also been examined for methanol production from CO2 hydrogenation. Among these catalysts, some non-noble metal-based bimetallics were designed for efficient CO2 hydrogenation to CH3OH at low pressure. In recent years, reducible oxides have received considerable attention due to their excellent performance with high CH3OH selectivity in a wide range of temperatures (210-310 °C). Nearly 100% selectivity can be attained over cubic In2O3 nanomaterial and In2O3 supported on monoclinic ZrO2 at CO2 conversions of less 5.5% under 300 °C, 5.0 MPa, 20 000 mL g -1 h -1 and H2/ CO2 = 4,[4]. Researchers have clearly demonstrated the structure sensitivity of the In2O3 catalyst in terms of both the phase and the exposed facet by combined computational and experimental studies.[7,8], Danget al. reported a successful work of computer-aided rational design of more efficient In2O3 catalysts for CO2 hydrogenation to CH3OH.[8]. Fig. 3: (A) TEM images of Cu-Zn-Ga catalysts derived from conventional hydroxyl-carbonate phases and ultrathin LDH (left) as well as their performance (right). [37]. Reaction conditions: 270 °C, 4.5 MPa, 18 000 mL g -1 h -1 , and H2/ CO2 = 3. (B) Synthesis and catalysis strategies (left) as well as CO2 hydrogenation performance (right) of Co@Six catalysts.[48]. Reaction con- 96 13th International Colloquium Fuels - September 2021 Improve by Ind 4.0 the Heterogeneous Catalysts for CO2 Hydrogenation to Liquid Fuels ditions: 260-380 °C, 2.0 MPa, 6000 mL g -1 h -1 and H2/ CO2 = 3. (C) HAADF-STEM images and corresponding EDX maps (top) of indium (blue) and palladium (green) for the spent coprecipitated (CP, left) and dry impregnated (DI, right) Pd-In2O3 catalysts with 0.75 wt % Pd as well as illustration for the distinct role of Pd in equilibrated S1 and S2 catalysts with pure In2O3 as a reference.[24].The thickness of the arrows qualitatively suggests the methanol and CO formation rates. Reprinted with permission from refs [37,48] Copyright 2020 American Chemical Society The In2O3/ ZrO2 catalyst system, the ZrO2 support remarkably boosted the activity of the In2O3 catalyst and prevented its sintering.[24.29]. It was suggested that synergic effects between In2O3 and ZrO2 carriers favorably tuned CH3OH selectivity by changing thereaction pathway.[8], Frei et al. also investigated the electronic, geometric, and interfacial phenomena related to the peculiar promotional effects of the monoclinic ZrO2 carrier on In2O3.[31] Less-pronounced lattice mismatching between In2O3 and ZrO2 favors the formation of more surface oxygen vacancies on In2O3, which is beneficial for methanol synthesis. The lower H2 splitting ability of In2O3 compared with metal catalyst, palladium (Pd) was introduced to enhance H2 activation and facilitate oxygen vacancy formation and thereby substantially promote the activity of In2O3.[12,13], CO2 conversion over In2O3 supported highly dispersed Pd nanocatalyst reached above 20% with CH3OH selectivity of ~70%and STY up to 0.89 g MeOH g cat -1 h -1 ,[13] which is about 2-5 times higher than pure In2O3 under similar reaction conditions. Moreover, the incorporation of Pd atoms in the In2O3 matrix forming low-nuclearity Pd clusters can simultaneously increase the activity, selectivity and long-term stability (fig.3C),[14]. Different from dry impregnated Pd-In2O3, this nanostructure can effectively avoid Pd clustering and minimize In2O3 sintering; thus, CH3OH STY dropped slightly from 1.01 to 0.96 g MeOH g cat -1 h -1 after time-onstream of 500 h at 280 °C, 5.0 MPa, 48 000 mL g -1 h -1 , and H2/ CO2 = 4. The addition of Pt, Au, Cu, Co, or Ni can also benefit the generation of active hydrogen species.[19,25-29]. 3. Ind 4.0 -possible new solutions Palladium atoms replacing indium atoms in the active In3O5 ensemble attract additional palladium atoms deposited onto the surface forming low-nuclearity clusters, which foster H2 activation and remain unaltered, enabling record productivities for 500 h. In2O3, recently discovered for CO2 hydrogenation tomethanol [17,18]. Indeed, although this reducible oxide is very selective and durable, especially when supported on ZrO2, owing to the presence of frustrated Lewis pairs [19,20], and single ensembles [21,22], its H2-splitting ability is limited [23] Palladium was shown to generate bulk intermetallic compounds with indium [24,25], which were superior to In2O3 in liquid-phase CO2 hydrogenation [26], but were exclusively active for the competitive reverse water-gas shift (RWGS) reaction when applied in the gas phase as nanoparticles supported on silica [27]. A synthetic method devised to deposit palladium nanoparticles on In2O3 minimizing alloy formation [28], led to a system displaying enhanced methanol formation, which was ascribed to the higher availability of activated hydrogen, fostering the desired hydrogenation reaction and vacancy formation on the oxide surface. These contrasting findings highlight that the true relevant speciation and action mechanism of palladium are poorly understood. More importantly, to maximize the promoting effect and thus process-leveladvantages, the H2splitting ability of palladium should be exploited while minimizing its high efficacy for the RWGS reaction as a stand-alone metallic phase [29,31] As single palladium atoms supported on carbon nitride are active in alkyne semihydrogenation and photocatalytic CO2 reduction [32-34], and possess distinct electronic properties compared to agglomeratedpalladium, calculations indicate that the (R)WGS reaction on Pd(111) requires at least three palladium atoms [35] , and the selectivity of platinum on MoS2 in CO2 hydrogenation was dictated by the cluster mono-/ binuclearity [36], we conceived atomic dispersion as a strategy to curtail detrimental effects while preserving beneficial attributes. This approach would simultaneously grant improved metal utilization and robustness, if palladium atoms are well anchored to the In2O3 structure. Aiming at deriving synthesis-structure-performance relations, we emphasized experimental methods as well as theoretical means, which comprise a prominent tool to uncover the impact of structure on activity and selectivity [37-39] but have rarely been applied to link preparation to structure 3.1 Solution of the palladium application The samples co-precipitated catalyst (CP)=S1 and dry impregnation (DI)=S2 were studied through a battery of characterization techniques to elucidate the palladium speciation. X-ray diffraction (XRD)confirmed the bixbyite (i.e., a defective fluorite-type) structure of In2O3 in both systems ( Fig. 4), the lowest-energy surface of which presents a protruding In3O5(O) substructure, which generates the active site in acetylene and CO2 hydrogenation after removal of the (O) atom [21], and an adjacent sunken region. The average particle size of In2O3 was calculated at 9 nmfor the fresh materials and increased to 15 and 25 nm for used S1and S2 catalysts, respectively. Moderate sintering was also observed earlier for unpromoted In2O3 [17,22]. A weak reflection at 39.5° 2θ specific to Pd0 was detected only for the used S2 material. High angle annular dark-field scanning transmissionelectron microscopy (H-STEM) images and indium and palladium maps acquired by energy-disper- 13th International Colloquium Fuels - September 2021 97 Improve by Ind 4.0 the Heterogeneous Catalysts for CO2 Hydrogenation to Liquid Fuels sive X-ray spectroscopy (EDX) indicate that palladium is extremely well dispersed in the fresh samples (fig.4) ). The S1 material used in the reaction for 93 h appeared virtually unaltered. In contrast, In2O3 particles increased their size to ca. 30 nm and palladium nanoparticles formed using the S2 catalyst for 74 h (fig.4) the latter having an average size of 2.8 nm based on high resolution transmission electron microscopy (HRTEM, (figs.4). Fig. 4: Microscopy analysis of palladium-promoted In2O3 catalysts. H-STEM images and corresponding EDX maps of indium (blue) and palladium (green) for the catalysts produced by a-S1 and b-S2 with 0.75 wt.% Pd in fresh form and after the tests depicted in Fig. 4b. The scale bar in all panels is 50 nm 4. Experimental nuclearity correlation. Intrigued by the selectivity-nuclearity correlation uncovered by the theoretical studies for S1 catalysts, we conducted a complementary experiment to link palladium agglomeration and its impact on methanol selectivity. When the S2 catalyst was tested keeping the CO2 conversion at 2-3%, the methanol selectivity remained at ca. 83% for 6 h, followed by a linear decline owing to a progressively greater impact of the RWGS reaction till a value of 58% after 9 h on stream. Characterization of the specimen retrieved at the end of the test confirmed severe sintering of palladium and In2O3 (particle sizes of 12 and 28 nm, respectively). A sample generated by a repeated experiment ending before the onset of the selectivity change still showed an extremely high-palladium dispersion based on STEM-EDX, as expected. This supports that agglomerates of only few Pd atoms are required to start favoring the RWGS reaction and highlights the importance of controlling the promoter nanostructure upon synthesis and preserving it upon .reaction. (figs.5). Figure 5 presents the most significant S1 and S2 catalysts’ models and their features, highlighting the remarkable matching of information derived from experiments and from theory. Gibbs energy profiles for methanol and CO formation from CO2 were calculated based on the elementary steps found optimal for In2O3. The activation of the S1 catalyst starts with H2 dissociation onto two O atoms located nearby the lattice palladium species, followed by transfer of one H atom to form water, which desorbs creating a vacancy. It was introduced as a novel technique to elucidate vacancy formation and short-range crystallinity in (palladium-promoted) In2O3 catalysts, and theoretical modeling provided sound understanding of the structure and functioning of the active sites in the two systems. In the S1 catalyst, embedding one Pd atom in the In2O3 matrix enables a controlled growth of extralattice atoms, leading to the stabilization of low-nuclearity palladium clusters. The number of vacancies does not increase but their electronic properties are strongly modified, permitting improved H2 dissociation. The small size of the palladium cluster is crucial to curtail the RWGS reaction on palladium sites, which is relevant already for clusters of 3 extra-lattice promoter atoms. In the S2 catalyst, palladium steadily agglomerates with time on stream forming particles. These aid methanol production by supplying activated hydrogen that fosters the formation of surface vacancies on In2O3 and the hydrogenation of the CO2 therein adsorbed. Still, they consume a significant part of the H2 split by converting CO2 into CO on their own surface (fig. 6). Fig. 5: Experimental and theoretical description of palladium sites. Most relevant surface models representing the S1 and S2 catalyst in fresh, activated, and used forms and their main features derived from experiments and theory. 98 13th International Colloquium Fuels - September 2021 Improve by Ind 4.0 the Heterogeneous Catalysts for CO2 Hydrogenation to Liquid Fuels Fig. 6: Schematic representation of the distinct role of palladium in equilibrated S1 and S2 systems with In2O3 and Pd-TiO2 as references. The thickness of the arrows qualitatively indicates the products formation rates. Reaction conditions: P = 5 MPa, H2: CO2 = 4, and WHSV = 48,000 cm3 STP h -1 g cat -1 5. Results and Discussion Our fixed bed reactor model for FTS using a cobalt catalyst was developed. This kinetic model has twoseries of multi-tubular FT reactors over a cobalt catalyst, and the Multiphysics was applied to simulate and predict the profiles of syngas conversion, feed concentration, oil selectivity, and temperature in the reactor under a variety of conditions including: feed temperature, syngas space velocity, and coolant temperature. Results show overall CO conversion of 50% and controlled temperature applied in operatingtemperatures of 215 °C to 235 °C in the first step ofthe multi tubular reactor. Moreover, the effect of coolant temperature was evaluated to determine optimum operating conditions to achieve more conversion and higher heat flux to keep the FT tubes near isothermal conditions. Non-isothermal reactor was investigated in this study for different feed temperature to get higher selectivity of jet fuel (this sentence does not make sense. Further simulation shows higher productivity and selectivity to control the catalytic Fischer-Tropsch (FT) packed bed rector near isothermal conditions, which started at a temperature of 235 °C. One of these problems is that this type of kinetic model cannot estimate the exact amount of heat released by exothermic reactions through the FT reactor. It is well known that the heat released from producing one mole of decane is different from the heat released by producing ten moles of methane. 154 kJ and 207 kJ of heat are released per CO moleconsumed in each case. The syngas molar ratio of H2/ CO=4 was fed through the top of the first stage with a temperature of 235 °C, was kept at the operating conditions of 300 psig, and was finally passed through the packed bed. The FT product from the first stage (unconverted syngas) was removed from the end of the reactor tube and fed to the top of the second stage of the reactor. The synthesis temperature was maintained by circulation of pressurized water in the shell. To control the inside temperature of the reactor according to theexothermic reactions therein, saturated water wasused to adjust the profile temperature inside the reactor to prevent coking of the catalysts. Water was heated to create saturated liquid water and was then pumped into the reactor. The average particle size of In2O3 increased by only 2 nm compared with S2-92 h (14 and 18 nm, respectively). Owing to the addition of TiO2 as a diluent in this test, which could not be separated from the used catalyst, changes in porous properties and vacancy density could not be investigated. The methanol STY was 0.92 g MeOH h -1 g cat -1 , which is, to our knowledge, the highest sustained productivity reported for CO2-based methanol synthesis over a heterogeneous catalyst. Specifically, our catalyst reaches a 16%-higher methanol STY than the best performer reported and at a cca. 62% shorter residence time. The latter enables a reduction in reactor size by 62%, which is a strong gain for a prospective industrial process. Moreover, S1 is a synthesis method that has been successfully implemented for catalyst manufacture at the large scale for numerous applications, whereas the polymer-assisted route used for the benchmark appears of limited industrial amenability. Controlled growth of extra-lattice atoms, leading to the stabilization of low-nuclearity palladium clusters. The number of vacancies does not increase but their electronic properties are strongly modified, permitting improved H2 dissociation. The small size of the palladium cluster is crucial to curtail the RWGS reaction on palladium sites, which is relevant already for clusters of 3 extra-lattice promoter atoms. In the S2 catalyst, palladium steadily agglomerates with time on stream forming particles. These aid methanol production by supplying activated hydrogen that fosters the formation of surface vacancies on In2O3 and the hydrogenation of the CO2 therein adsorbed. Still, they consume a significant part of the H2 split by converting CO2 into CO on their own surface (Fig. 6). These contrasting palladium speciations also explain the profoundly diverse time-on-stream behaviors of the two catalysts. Palladium embedded into indium oxide attracts the scarce palladium in pockets, avoiding its clustering, and facilitates water desorption minimizing In2O3 sintering, whereas palladium deposited onto In2O3 easily agglomerates and leads to catalyst sintering by spilling too much activated H2 to the oxide, causing over-reduction, and by excessively retaining water, provoking crystals coalescence. Hence, anchoring of palladium clusters to the In2O3 lattice is an unprecedented approach to effectively enhance activity and selectivity, and more importantly, to maintain the promotional effect in the long term, an aspect fully disregarded in previous studies on palladium-promoted In2O3 systems. In addition, low-nuclearity palladium clusters grant the possibility to operate the catalyst at reduced temperature, with hydrogen-lean feeds, and in the presence of greater amounts of water, altogether implying strong economic and ecologic benefits at a process level. So, propose our strategy to engineer promotion at the atomic scale marks a step ahead toward green methanol production and holds great general potential for tailoring new or existing promoted systems in current and emerging applications of heterogeneous catalysis. Most companies believe they will see a return on investment (ROI) within two years or less (fig. 7) for their Industry 4.0 projects. Just over a third of companies anticipate a longer timescale of three to five years, for Industry 4.0 investments to pay for themselves. 13th International Colloquium Fuels - September 2021 99 Improve by Ind 4.0 the Heterogeneous Catalysts for CO2 Hydrogenation to Liquid Fuels References [1] Rui, N.; Zhang, F.; Sun, K.; Liu, Z.; Xu, W.; Stavitski, E.; Senanayake, S. D.; Rodriguez, J. A.; Liu, C.-J. Hydrogenation of CO2 to Methanol on a Auδ+-In2O3-x Catalyst. ACS Catal. 2020, [2] Li, M. M.; Zou, H.; Zheng, J.; Wu, T. S.; Chan, T. S.; Soo, Y. L.; Wu, X. P.; Gong, X. Q.; Chen, T.; Roy, K.; Held, G.; Tsang, S. C. E. Methanol Synthesis at a Wide Range of H2/ CO2 Ratios over a Rh-In Bimetallic Catalyst. Angew. Chem., [3] Guil-Lopez, R.; Mota, N.; Llorente, J.; Millan, E.; Pawelec, B.; Fierro, J. L. G.; Navarro, R. M. Methanol Synthesis from CO2: A Review of the Latest Developments in Heterogeneous Catalysis. Materials 2019, 12 (23), 3902. [4] Din, I. U.; Shaharun, M. S.; Alotaibi, M. A.; Alharthi, A. I.; Naeem, A. Recent developments on heterogeneous catalytic CO2 reduction to methanol. J. CO2 Util. 2019, 34, 20-33. [5] S. Mitchell, T.F. Segawa, R. Hauert, C. Drouilly, D. Curulla Ferr´e, J. P´erez-Ramírez, Indium oxide as a superior catalyst for methanol synthesis by CO2 hydrogenation, Angew. Chem. Int. Ed. 55 (2016)6261-6265. [6] B. Qin, Y. Yang, H. Wang, J. Cai, Y. Han, S. Li, P. Gao, Y. Sun, Rationally designed indium oxide catalysts for CO2 hydrogenation to methanol with high activity and selectivity, Sci. Adv. 6 (2020), [7] Olaru, P.et el. New Carbon Capture, Converted in Electricity (CCCE) technology by (UCnF) Reinforced Shape Memory Polymers-(SMPs) Nanostrip Multilayers, International Colloquium Fuels -TAE 27 to 29 June 2017 in Stuttgart / Ostfidern, Germany [8] Behr, A. Methanol: the basic chemical and energy feedstock of the future. Angew. Chem. Int. Ed. 53, 12674-12674 (2014). [9] Álvarez, A. et al. Challenges in the greener production of formates/ formic acid, methanol, and DME by heterogeneously catalyzed CO2 hydrogenation processes. Chem. Rev. 117, 9804-9838 (2017). [10]. Wang, J. et al. A highly selective and stable ZnO- ZrO2 solid solution catalyst for CO2 hydrogenation to methanol. Sci. Adv. 3, e1701290 (2017). [11 Vogt, C. et al. Unravelling structure sensitivity in CO2 hydrogenation over nickel. Nat. Catal. 1, 127- 134 (2018). [12] Wang, L. et al. Incorporating nitrogen atoms into cobalt nanosheets as a strategy to boost catalytic activity toward CO2 hydrogenation. Nat. Energy 2,869-876 (2017). [13] Behrens, M. Promoting the synthesis of methanol: understanding the requirements for an industrial catalyst for the conversion of CO2. Angew. Chem. Int. Ed. 55, 14906-14908 (2016). [14] Xu, J. et al. Methanol synthesis from CO2 and H2 over Pd/ ZnO/ Al2O3: catalyst structure dependence of methanol selectivity. Appl. Catal. A 514, 51-59 (2016). [15] Yang, H. B. et al. Atomically dispersed Ni(I) as the active site for electrochemical CO2 reduction. Nat. Energy 3, 140-147 (2018). [16] Loiland, J. A., Wulfers, M. J., Marinkovic, N. S. & Lobo, R. F. Fe/ γ-Al2O3 and Fe-K/ γ-Al2O3 as reverse water-gas shift catalysts. Catal. Sci. Technol. 6,5267-5279 (2016). [17] Martin, O. et al. Indium oxide as a superior catalyst for methanol synthesis by CO2 hydrogenation. Angew. Chem. Int. Ed. 55, 6261-6265 (2016). [18] Gao, P. et al. Direct conversion of CO2 into liquid fuels with high selectivity over a bifunctional catalyst. Nat. Chem. 9, 1019-1024 (2017). [19] Zhang, S. et al. Solid frustrated-Lewis-pair catalysts constructed by regulations on surface defects of porous nanorods of CeO2. Nat. Commun. 8, 15266(2017). [20] Ghuman, K. K. et al. Photoexcited surface frustrated Lewis pairs for heterogeneous photocatalytic CO2 reduction. J. Am. Chem. Soc. 138, 1206-1214 (2016). [21] Albani, D. et al. Semihydrogenation of acetylene on indium oxide: proposed single-ensemble catalysis. Angew. Chem. Int. Ed. 56, 10755-10760 (2017). 100 13th International Colloquium Fuels - September 2021 Improve by Ind 4.0 the Heterogeneous Catalysts for CO2 Hydrogenation to Liquid Fuels [22] Frei, M. S. et al. Mechanism and microkinetics of methanol synthesis via CO2 hydrogenation on indium oxide. J. Catal. 361, 313-321 (2018). [23] Ye, J., Liu, C.-j, Mei, D. & Ge, Q. Methanol synthesis from CO2 hydrogenation over a Pd4/ In2O3 model catalyst: a combined DFT and kinetic study. J. Catal. 317, 44-53 (2014). [24] Lorenz, H. et al. Pd-In2O3 interaction due to reduction in hydrogen: consequences for methanol steam reforming. Appl. Catal. A 374, 180-188 (2010). [25] Rameshan, C. et al. Impregnated and co-precipitated Pd-Ga2O3, Pd-In2O3 and Pd-Ga2O3-In2O3 catalysts: influence of the microstructure on the CO2 selectivity in methanol steam reforming. Catal. Lett. 148, 3062-3071 (2018). [26] García-Trenco, A. et al. PdIn intermetallic nanoparticles for the hydrogenation of CO2 to methanol. Appl. Catal. B 220, 9-18 (2018). [27] Ye, J., Ge, Q. & Liu, C.-J. Effect of PdIn bimetallic particle formation on CO2reduction over the PdIn/ SiO2 catalyst. Chem. Eng. Sci. 135, 193-201 (2015). [28] Rui, N. et al. CO2 hydrogenation to methanol over Pd/ In2O3: effects of Pd and oxygen vacancy. Appl. Catal. B 218, 488-497 (2017). [29] Arena, F. et al. Solid-state interactions, adsorption sites and functionality of Cu-ZnO/ ZrO2 catalysts in the CO2 hydrogenation to CH3OH. Appl. Catal. A 350, 16-23 (2008). [30] Erdöhelyi, A., Pásztor, M. & Solymosi, F. Catalytic hydrogenation of CO2 over supported palladium. J. Catal. 98, 166-177 (1986). [31] Pettigrew, D. J., Trimm, D. L. & Cant, N. W. The effects of rare earth oxides on the reverse water-gas shift reaction on palladium/ alumina. Catal. Lett. 28,313-319 (1994). [32] Vilé, G. et al. A stable single-site palladium catalyst for hydrogenations. Angew. Chem. Int. Ed. 54, 11265-11269 (2015). [33] Mitchell, S., Vorobyeva, E. & Pérez-Ramírez, J. The multifaceted reactivity of single-atom heterogeneous catalysts. Angew. Chem. Int. Ed. 130, 15538-15552 (2018). [34] Gao, G., Jiao, Y., Waclawik, E. R. & Du, A. Single atom (Pd/ Pt) supported on graphitic carbon nitride as an efficient photocatalyst for visible-light reduction of carbon dioxide. J. Am. Chem. Soc. 138, 6292-6297 (2016). [35] Li, Q., García-Muelas, R. & López, N. Microkinetics of alcohol reforming for H2 production from a fair density functional theory database. Nat. Commun. 9, 526 (2018). [36] Li, H. et al. Synergetic interaction between neighbouring platinum monomers in CO2 hydrogenation. Nat. Nanotechnol. 13, 411-417 (2018). [37] Nørskov, J. K., Bligaard, T., Rossmeisl, J. & Christensen, C. H. Towards the computational design of solid catalysts. Nat. Chem. 1, 37-46 (2009). [38] Medford, A. J. et al. From the Sabatier principle to a predictive theory of transition-metal heterogeneous catalysis. J. Catal. 328, 36-42 (2015). [39] Jimenez-Izal, E. & Alexandrova, A. N. Computational design of clusters for catalysis. Annu. Rev. Phys. Chem. 69, 377-400 (2018). [40] Tauster, S. Strong metal-support interactions. Acc. Chem. Res. 20, 389-394 (1987). [41] Abbet, S. et al. Identification of defect sites on MgO(100) thin films by decoration with Pd atoms and studying CO adsorption properties. J. Am. Chem. Soc. 123, 6172-6178 (2001). [42] Bergeret, G. & Gallezot, P. in Handbook of Heterogeneous Catalysis Vol. 2. (eds Ertl, G., Knözinger, H., Schüth, F. & Weitkamp, J.) 738-765 (Wiley- VCH, Weinheim, 2008). [43] Capdevila-Cortada, M. & López, N. Entropic contributions enhance polarity compensation for CeO2(100) surfaces. Nat. Mater. 16, 328-334 (2017). [44] Snider, J. L. et al. Revealing the synergy between oxide and alloy phases on the performance of bimetallic In-Pd catalysts for CO2 hydrogenation tomethanol. ACS Catal. 9, 3399-3412 (2019). [45]. Garcıa-Muelas, R., Li, Q. & López, N. Density functional theory comparison of ́ methanol decomposition and reverse reactions on metal surfaces. ACS Catal. 5, 1027-1036 (2015). [46] Chorkendorff, I. & Niemantsverdriet, J. W. in Concepts of Modern Catalysis and Kinetics Vol. 1 (eds Chorkendorff, I. & Niemantsverdriet) 79-128 (WileyVCH, Weinheim, 2003) [47] Kresse, G. & Furthmüller, J. Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Comput. Mater. Sci. 6, 15-50 (1996). [48] Kresse, G. & Hafner, J. Ab initio molecular dynamics for liquid metals. Phys.Rev. B47, 558-561 (1993). [49] Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple, Phys. Rev. Lett. 77, 3865-3868 (1996). [50] Blöchl, P. E. Projector augmented-wave method, Phys. Rev. B50, 17953-17979 (1994). [51] Frisch, M. J. et al. Gaussian 09, Revision A.02. (Gaussian, Inc., Wallingford CT,2016). [52] Köhler, L. & Kresse, G. Density functional study of CO on Rh(111). Phys. Rev.B 70, 165405 (2004)