*3.4. Structure Effect*

The structure of the catalysts plays an important role in converting CO2 to light olefins. In this section, we will report the recent progress on the ways in which morphology changes in both the Fe-based and methanol zeolite composite catalysts can improve the catalytic performance. Some representative catalysts on the structure effect for CO2 hydrogenation to light olefins with improved catalyst stability are presented in Table 4.

**Table 4.** Some representative catalysts on the structure effect for CO2 hydrogenation to light olefins.


<sup>a</sup> Refers to hydrocarbon distribution %; <sup>b</sup> refers to C2-C5+ hydrocarbons.

Wang et al. developed a layered metal oxides (LMO) structure, K-Fe-Ti, that displayed high catalytic activity, olefin selectivity and decent stability toward CO2–FTS. The light olefin selectivity achieved approximately 60% with an olefin/paraffin ratio of 7.3 over the catalyst 0.8K-2.4Fe-1.3Ti (Figure 23). The LMO structure exfoliated through the acid treatment was found to weaken the interaction between Fe and Ti, which made it easier for the reduction and activation of iron oxides to form active iron carbide species that favored a shift from the RWGS to the FTS reaction. Meantime, C2H4 adsorption was hindered due to the low surface area of the LMO structure, contributing to higher olefin selectivity by inhibiting the secondary hydrogenation of primary olefins. The acid treatment played a key role in the formation of a slice structure that favored CO2 conversion to light olefins with lower CO selectivity [123]. Fujiwara et al. found the composite catalysts obtained from the simple mixing of Cu–Zn–Al oxide together with HB zeolite, which was modified with 1,4-bis(hydroxydimethylsilyl) benzene, to be very effective for CO2 hydrogenation to C2+ hydrocarbons. The modification of zeolite with the disilane compound made the catalysts' surface hydrophobic, a characteristic which was effective in preventing catalyst deactivation by the formation of water during CO2 hydrogenation. The highest yield of C2+ hydrocarbons over the modified composite catalysts reached about 12.6 C-mol% at 573 K under a pressure of 0.98 Mpa. The diminishing of the deactivation of the strong acid sites of HB zeolite with the hydrophobic surface is the source of the enhanced catalytic activity [129].

Liu et al. synthesized a unique structure with ZnO and nitrogen-doped carbon (NC)-overcoated Fe-based catalysts (Fe@NC) (Figure 24), and found that the reaction rate increased by ~25%, while the O/P ratio increased from 0.07 to 1.68 when compared with the benchmark Fe3O4 catalyst. The inactive θ-Fe3C phase disappeared, and the active phases (Fe3O4 and Fe5C2) formed for CO2 hydrogenation. The introduction of NC to the surface of the Fe catalysts significantly boosted the catalyst activity, the selectivity toward light olefins, and the stability due to the enhanced metal–support-reactant interaction and interfacial charge transfer [36].

**Figure 23.** (**a**) Catalytic performance over different catalysts. (**b**) The catalytic stability of 0.8K-2.4Fe-1.3Ti at TOS (testing conditions: H2/CO2 molar ratio = 3/1, T = 593 K, P = 2.0 MPa and GHSV = 10,000 mL gcat<sup>−</sup>1h−1). Adapted with permission from ref. [123]. Copyright 2019 Elsevier.

**Figure 24.** Schematic illustration of the formation of Fe@NC catalysts and the reaction for CO2 hydrogenation. Adapted with permission from ref. [36]. Copyright 2019 American Chemical Society.

Numpilai et al. studied the hydrogenation of CO2 to light olefins over Fe-Co/K-Al2O3 catalysts, and discovered that the pore sizes of the Al2O3 support had profound effects on the Fe2O3 crystallite size, the reducibility, the adsorption–desorption of CO2 and H2, and the catalytic performances. The highest olefins to paraffins ratio of 6.82 was obtained from the largest pore catalyst (CL-Al2O3) due to the suppression of the hydrogenation of olefins to paraffins by increasing the pore sizes of Al2O3 to eliminate diffusion limitation. The maximum light olefin yield of 14.38% was obtained over the catalyst with an appropriated Al2O3 pore size (49.7 nm) owing to the suppression of the olefins' hydrogenation and chain growth reaction [37].

The electrospun ceramic K/Fe-Al-O nanobelt catalysts synthesized by Elishav et al. showed a much higher CO2 conversion of 48%, a C2-C5 olefin selectivity of 52%, and a high olefin/paraffin ratio of 10.4, while the K/Fe-Al-O spinel powder catalyst produced mainly

C6+ hydrocarbons. The enhanced olefin selectivity of the electrospun materials is related to a high degree of reduction of the surface Fe atoms due to the more efficient interaction with the K promoter [124].

A defect-rich MgH2/Cu*x*O hydrogen storage composite might inspire the catalysts' design for the hydrogenation of CO2 to lower olefins. Chen et al. presented a defect-rich MgH2/Cu*x*O composite catalyst that achieved a C2 =–C4 <sup>=</sup> selectivity of 54.8% and a CO2 conversion of 20.7% at 623 K under a low H2/CO2 ratio of 1:5. It is the defective structure of MgH2/CuxO that promotes CO2 molecule adsorption and activation, while the electronic structure of MgH2 was more conducive to the provision of lattice H− for the hydrogenation of the CO2 molecule. The lattice H− could combine with the C site of the CO2 molecule to promote the formation of Mg formate, which was further hydrogenated to lower olefins under a low H− concentration [125]. The same group reported carbon-confined MgH2 nano-lamellae which stored solid hydrogen for the hydrogenation of CO2 to lower olefins and demonstrated a high selectivity under low H2/CO2 ratios. The high selectivity of lower olefins was attributed to the low concentration of solid hydrogen under low H2/CO2 ratios that suppressed the further hydrogenation of light olefins from Mg formate [128].

SAPO-34 molecular sieves were considered to be the best catalysts due to their excellent structure selectivity, suitable acidity, favorable thermal stability, and hydrothermal stability, as well as their high selectivity for light olefins. Tian et al. used Palygorskite as a silicon and partial aluminum source, and DEA, TEA, MOR and TEAOH as template agents to prepare SAPO-34 molecular sieves with higher purity. Composite catalysts of CuO-ZnO-Al2O3/SAPO-34 were prepared by mechanically mixing SAPO-34 molecular sieves with CuO-ZnO-Al2O3 (CZA), and a superb CO2 conversion of 53.5%, a light olefin selectivity of 62.1% and a yield of 33.2% were obtained over the CZA/SAPO-34(TEAOH)HCl composite catalyst [126]. CO2 conversion and product distribution are strongly dependent on the oxide composition and structure. Li et al. developed a bifunctional catalyst composed of ZnO-Y2O3 oxide and SAPO-34 zeolite that offered a CO2 conversion of 27.6% and a light olefin selectivity of 83.6% [40].

Some Fe-containing catalysts can also be improved by creating unique architectures. Wei et al. created Fe-based catalysts with honeycomb-structured graphene (HSG) as the catalyst support and K as the promoter, and achieved the 59% selectivity of light olefins over a FeK1.5/HSG catalyst. No obvious deactivation was observed within 120 h on stream (Figure 25). The excellent catalytic performance was ascribed to the confinement effect of HSG and the K promotion effect on the activation of inert CO2 and the formation of iron carbide. The complex three-dimensional (3D) architecture of the porous HSG effectively impeded the sintering of the active sites' iron carbide nanoparticles (NPs). Meanwhile, CO2 and H2 could more easily permeate the mesoporous–macroporous framework of HSG and access the catalysts' active sites. Similarly, the generated light olefins could more easily emerge from the catalyst so as to avoid further unwanted hydrogenation [127].

Consequently, multiple reports indicate that the modification of the morphology of zeolite–methanol synthesis composites by creating core–shell configurations can have a beneficial effect [16,120,130]. For example, dual-function composite catalysts containing CuZnZr (CZZ) and SAPO-34 were synthesized by Chen et al. for the tandem reactions of CO2 to methanol and methanol to olefins. The assembled core–shell CZZ@SAPO-34 catalyst, as shown in Figure 26, exhibited an enhanced light olefin selectivity of 72% and inhibited CH4 formation due to reduced contact interface between CZZ and SAPO-34 and weakened hydrogenation ability at the metal sites. Furthermore, the addition of Zn reduced the acidity of SAPO-34; as a result, the secondary reactions of the primary olefins were significantly diminished (Figure 26) [120].

In summary of this section, the structure and the properties associated with the structure of the catalysts are pivotal for CO2 hydrogenation to light olefins. The low surface area of the LMO structure could hinder the C2H4 secondary reaction, contributing to higher olefin selectivity. The surface modification of zeolite from hydrophilic to hydrophobic could prevent the catalyst deactivation caused by the formation of water. The unique structure of

Fe@NC enables phase transformation from the inactive (θ-Fe3C) phase to active species (Fe3O4 and Fe5C2). Increasing the pore sizes of Al2O3 could eliminate the diffusion limitation for CO2 and H2. The electrospun ceramic K/Fe-Al-O nanobelt catalysts led to a high degree of the reduction of surface iron atoms. The defective structure of MgH2/CuxO and carbon-confined MgH2/C nano-lamellae could promote CO2 adsorption and activation, with the electronic structure of MgH2 offering lattice H− for CO2 hydrogenation. The 3D architecture of the porous HSG could impede the sintering of the active sites' iron carbide NPs. The confinement of core–shell CZZ@SAPO-34 structure could increase the access frequency of the methanol intermediate to the active zeolite sites, consequently improving the light olefine selectivity.

**Figure 25.** (**a**) N2 physisorption isotherms, (**b**) SEM image, (**c**) HAADF−STEM image, and (**d**) TEM image and particle distribution of the FeK1.5/HSG catalyst. (**e**) CO2 hydrogenation over the catalyst FeK1.5/HSG during a TOS of 120 h (testing conditions: mass of catalyst = 0.15 g, T = 613 K, P = 20 bar, H2/CO2 molar ratio = 3, and GHSV = 26 L h−1g−1). Adapted with permission from ref. [127]. Copyright 2018 American Chemical Society.

**Figure 26.** (**a**) Schematic illustration of the interface between CZZ and SAPO-34. (**b**) Core–shell interface of CZZ and SAPO-34. (**c**) Stability of the composite catalyst CZZ@Zn-SAPO-34 at TOS (testing conditions: H2/CO2 molar ratio = 3, T = 673 K). Adapted with permission from ref. [120]. Copyright 2019 Elsevier.
