Layered Double Hydroxide (LDH)-Derived Mixed Oxides for Enhanced Light Hydrocarbon Production from CO₂ Hydrogenation

Abstract

Layered double hydroxide (LDH)-derived mixed oxides offer a promising approach for CO₂ hydrogenation to light hydrocarbons. Herein, we explore the impact of various transition metals (X = Mn, Co, Cu, and Zn) incorporated into the M-Al or M-(Al+Fe) LDH structures, with the aim of exploring possible synergistic effects. Structural and compositional analyses reveal that an abundance of Fe over Al (Fe/Al ratio ~4) leads to the formation of mixed oxide crystalline phases attributed to CoFe₂O₄, CuFe₂O₄, and ZnFe₂O₄. Catalytic evaluation results demonstrate that the X-Al LDH-derived oxides exhibit high CO₂ conversion yet are selective to CH₄ or CO. In contrast, Fe incorporation shifts selectivity toward higher hydrocarbons. Specifically, the yield to higher hydrocarbons (C₂⁺) follows the order Ζn-Al-Fe > Cu-Al-Fe > Mn-Al-Fe > Co-Al-Fe >> Mn-Al, Co-Al, Zn-Al, Cu-Al, highlighting the pivotal role of Fe. Moreover, Zn-Al-Fe and Mn-Al-Fe catalysts have been shown to be the most selective towards light olefins. Zn-based systems also exhibit high thermal and structural stability with minimal coke formation, whereas Co-, Cu-, and Mn-based catalysts, when modified with Fe, experience increased carbon deposition or structural changes that may impact long-term stability. This work provides insights into the combined role of Fe and a second transition metal in LDHs for modulating catalytic activity, phase transformations, and stability, underscoring the need for further optimization to balance selectivity and catalyst durability in CO₂ hydrogenation applications.

1. Introduction

The rise in energy consumption and commodity goods has led to a proportional increase in CO₂ emissions, driving the need for effective reduction strategies. To that end, Carbon Capture and Utilization (CCU) for hydrocarbon production emerges as a promising solution, addressing greenhouse gas emissions and recycling carbon resources. More specifically, using green hydrogen from renewable sources in CO₂ hydrogenation enhances the production of carbon-neutral chemicals and fuels, such as light olefins (C₂–C₄⁼), which are major components of the chemical industry. Light olefins, namely ethylene, propylene, and butene, are essential raw materials for the production of polymers, resins, pharmaceuticals, and rubbers, with current production numbers exceeding 250 Mt per year [1]. However, today, C₂–C₄⁼ are produced from energy-intensive routes, such as steam cracking of fossil fuels, emitting around 400 Mt CO₂ each year [2]. Consequently, converting CO₂ into value-added products offers a viable approach to reducing fossil fuel dependence, addressing significant environmental and economic challenges.

Currently, higher hydrocarbons can be synthesized directly from CO₂ hydrogenation following the modified Fischer–Tropsch (mFTS) catalytic route. This process follows two consecutive steps, namely the production of CO from CO₂ via the Reverse Water–Gas Shift (RWGS) reaction and its subsequent transformation to hydrocarbons via Fischer–Tropsch Synthesis (FTS). However, this intricate process faces several challenges due to the chemical inertness of CO₂ and its high C-C coupling energy barrier, which requires a compromise between the different temperature ranges that favor CO₂ activation and FTS, respectively [3]. Additionally, the distribution of produced hydrocarbons is limited by the Anderson–Schulz–Flory (ASF) distribution, which evaluates C₂–C₄ production (both olefins and paraffins) around 58% and CH₄ around 29% [4]. For that purpose, many attempts have been made to break the limit of ASF distribution, such as the selection and the development of tandem catalysts with multifunctional active sites both for RWGS and FTS reactions and their further tuning.

Iron-based catalysts have been recognized as the most selective catalysts towards higher hydrocarbon synthesis in FTS due to their ability to form co-existing phases of Fe in situ, such as Fe₃O₄, which is considered the active site for the RWGS reaction and several Fe carbides, with Fe₅C₂ being the most active phase in FTS [5,6,7]. Additionally, it has been demonstrated that doping Fe-based catalysts with adequate amounts of transition metals, such as Mn, Co, Cu, and Zn, can enhance the catalytic activity due to the synergy of the bimetallic active centers [8,9,10,11] and act as structural promoters that hinder catalyst deactivation [7]. In recent studies, Co has been widely explored as a secondary metallic promoter, demonstrating its ability to enhance higher hydrocarbon formation and promote CO₂ adsorption, thus facilitating its conversion [10,12]. Chaipraditgul et al. demonstrated that Co addition enabled the further hydrogenation of intermediate species, favoring paraffin production with a closer affinity to CH₄ formation [13]. In addition, Cu has proven to facilitate Fe reducibility, enhance its carburization rate, and thus promote Hägg Carbide formation [14,15]. Furthermore, the synergy between Fe and Cu increased the carbon chain propagation, resulting in higher C₅⁺ selectivity [16,17]. Zinc, on the other side, hinders the carburization ability of the catalytic system and enables the formation of ZnFe₂O₄, which enhances Fe dispersion and suppresses carbon deposition, which leads to sintering [11,18]. Consequently, the intricacy of Fe-based catalysts presents significant challenges, and the impact of transition-metal modifiers is still controversial.

Overall, it is important to develop catalysts with enhanced CO₂ adsorption ability, with facile and stable formation of active species. This design principle guides the focus of this work in Layered Double Hydroxides (LDHs) or hydrotalcites, a class of two-dimensional clays consisting of brucite-like layers containing different divalent and trivalent metal cations. Currently, Ni, Co-Al LDH-derived catalysts have been extensively explored in FTS applications [19,20,21]. Additionally, many studies have demonstrated the benefits of LDHs as catalyst precursors since the derived mixed oxides have a homogeneous distribution of metals (ions) in the formed mixed oxide crystalline phases, along with relatively high specific surface area and improved catalytic performance compared with similar materials synthesized by simple co-precipitation methods [22]. In addition to the LDH properties, the enhanced dispersion of metal sites at an atomic level with a tunable M₁²⁺/M₂³⁺ ratio holds great opportunity for fine-tuning the structural characteristics of the catalytic system [23]. Moreover, it has been demonstrated that upon thermal activation at 300 to 550 °C, the hydrotalcites’ layered structure transforms into a house-of-cards mesostructure consisting of the respective mixed oxides or spinels [24]. Nevertheless, the oxides of the divalent metal always prevail in spinels due to the high atomic ratio of M₁²⁺/M₂³⁺ (~2–5) in the lattice of hydrotalcite layers [25,26]. The latter was also discussed by Xu et al., who developed K-containing spinel-type ZnCo_xFe_2−xO₄ nanoparticles from LDH calcination with an FTY⁼ value of 29.1 μmol_CO2·gFe^−1.s⁻¹ for CO₂ conversion to C₂–C₄⁼ [10].

Herein, different transition metals (X = Mn, Co, Cu, Zn) were used to prepare X-Al and X-(Al+Fe) LDH-derived catalysts and their impact on light hydrocarbon production was examined. The role of Fe in facilitating higher hydrocarbon formation was assessed via comparison with the performance of the X-Al mixed-oxide series. Furthermore, preliminary investigations utilizing a combination of characterization techniques on fresh and spent samples provide insights into how the choice of transition metal influences CO₂ hydrogenation activity, selectivity, and stability. These findings highlight the potential of Fe-modified LDH-derived catalysts in CO₂ hydrogenation, a relatively underexplored topic, and pave the way for further optimization to tune selectivity and long-term stability.

2. Results

2.1. Characterization of Catalysts

2.1.1. Characterization of As-Prepared Materials

The elemental composition of the as-prepared materials was determined via ICP-AES analysis to confirm the amounts and molar ratios of metals, and the results are shown in

Table 1. Composition of as-made materials, as determined via ICP-AES.

Materials	Theoretical Molar Ratios	Experimental Molar Ratios	Co (wt.%)	Cu (wt.%)	Zn (wt.%)	Mn (wt.%)	Al (wt.%)	Fe (wt.%)
Co-Al	Co/Al = 2	Co/Al = 2.4	61.1	-	-	-	11.6	-
Co-Al-Fe	Co/(Al+Fe) = 2, Fe/Al = 4	Co/(Al+Fe) = 2.3, Fe/Al = 4.0	44.7	-	-	-	1.8	14.9
Cu-Al	Cu/Al = 2	Cu/Al = 2.3	-	39.9	-	-	7.3	-
Cu-Al-Fe	Cu/(Al+Fe) = 2, Fe/Al = 4	Cu/(Al+Fe) = 2.3, Fe/Al = 4.1	-	50.5	-	-	1.8	15.3
Zn-Al	Zn/Al = 2	Zn/Al = 2.6	-	-	69.5	-	11.1	-
Zn-Al-Fe	Zn/(Al+Fe) = 2.2, Fe/Al = 4	Zn/(Al+Fe) = 2.2, Fe/Al = 4.3	-	-	48.0	-	1.7	15.1
Mn-Al	Mn/Al = 2	Mn/Al = 3.1	-	-	-	53.7	8.5	-
Mn-Al-Fe	Mn/(Al+Fe) = 2, Fe/Al = 4	Mn/(Al+Fe) = 2.8, Fe/Al = 4.6	-	-	-	50.1	1.6	15.1

. The results indicate that the experimentally determined molar ratio M³⁺/(M²⁺/M³⁺) between the divalent (M²⁺) and trivalent metals (M³⁺) ranges between 0.27 and 0.39, being within the ratio 0.2–0.4 of cations required to achieve a stable hydrotalcite structure [27]. Furthermore, the ratio M²⁺/M³⁺ is close to 2.2–2.6 for Co, Cu, and Zn materials and is slightly increased to 2.8–3.1 for Mn. The addition of iron led to lower M²⁺/M³⁺ ratios compared to the LDHs without iron. Regarding the ratio of the trivalent metals, the experimentally determined ratio of Fe/Al is slightly higher than the theoretical ratio (Fe/Al = 4) and ranges between 4.0 and 4.6. A higher value (Fe/Al = 4.6) was obtained for the manganese material.

The crystal structure of the as-prepared LDHs was determined by X-ray Powder Diffraction (XRD). As can be observed in Figure 1, Co-Al LDH exhibits all the characteristic reflections of the hydrotalcite structure of the R-3m space group. More specifically, the three main peaks at 2θ = 11.6°, 23.6°, and 34.7° correspond to the reflections (003), (006), and (012) of the hydrotalcite structure and reveal the high crystallinity of the material. The partial substitution of aluminum by iron maintained the structure of the hydrotalcite, while the three main peaks slightly shifted to 2θ = 11.7°, 23.5° and 34.2°. Similarly, the Zn-Al LDH exhibits the hydrotalcite structure, and the partial substitution of aluminum by iron maintains the hydrotalcite structure, but the material is less crystalline. The lower crystallinity of Zn-Al LDH compared to the Co-Al LDH can be attributed to the pH of precipitation. All the materials were synthesized at pH = 9–10, which is close to the pH value required for the precipitation of cobalt hydroxides (pH = 9), while the precipitation of zinc hydroxides is favored in a slightly lower pH of ~8 [27]. On the other hand, Cu-Al does not exhibit the hydrotalcite structure, but it exhibits all the characteristic peaks of malachite (Cu₂(OH)₂CO₃) indexed in the P 1 21/a 1 space group and monoclinic crystal system. The formation of the malachite phase is mainly attributed to the high copper content and the Jahn–Teller effect, which predominates under the current conditions of synthesis (metal precursors and pH of precipitation) and does not favor the formation of hydrotalcite, as also referred in the literature [28,29]. The malachite structure is maintained after the partial substitution of aluminum by iron. In a similar trend, Mn-Al and Mn-Al-Fe materials exhibit the main reflections of manganese carbonate (rhodochrosite) of the R-3m space group in the trigonal crystal system. The Mn-Al sample also exhibits some of the characteristic peaks of Al(OH)₃ of the bayerite structure. Furthermore, the partial substitution of aluminum by iron led only to the MnCO₃ phase, as can be seen in the XRD patterns of Figure 1.

The thermal decomposition of the layered double hydroxides was examined via the Thermogravimetric Analysis (TGA). The degradation of the materials undergoes several steps, which are dependent on the different metals used in the synthesis, as can be observed in Figure 2. The degradation of Co-Al is carried out in two distinct steps. The first weight loss (15.0%) starts from room temperature to 220 °C (endodermic peak at DTG_max = 198 °C) and is attributed to the removal of physically absorbed water and the metal precursor used during the synthesis. The second weight loss (13.4%) starts at 220 °C, finishes at 320 °C (endodermic peak at DTG_max = 252 °C), and is due to the dehydroxylation and decarbonation, which leads to the removal of hydroxyl and carbonate ions from the brucite structure according to the literature. The partial substitution of aluminum with iron led to the one-step weight loss, which shifted to lower temperatures (DTG_max = 195 °C), accounting for 26.9%.

Regarding the thermal behavior of zinc-based LDHs, the distinction in the two above-described decomposition steps is not so clear for Zn-Al, which exhibits a broader single-step degradation, starting from room temperature up to 480 °C (DTG_max = 163 °C) with the weight being equal to 29%. The partial substitution of aluminum by iron led to two more distinct steps of weight loss. The first step is in the temperature range of 28–200 °C (endodermic peak at DTG_max = 167 °C), corresponding to 8.9% weight loss, and the second step ranges between 200 and 400 °C (endodermic peak at DTG_max = 259 °C). Similar to the cobalt-based materials, the first weight loss step can be attributed to the removal of water, while the second weight loss can be attributed to the removal of hydroxyl, nitrate, and carbonate ions, leading to the collapse of the LDH structure, in accordance with results from the literature [30].

In the case of copper LDHs, the main weight loss of both Cu-Al (19.2%) and Cu-Al-Fe (15.2%) has occurred in the temperature range of 250–400 °C (endodermic peak at DTG_max = 339 °C and 350 °C, respectively) and is attributed to the thermal decomposition of malachite (Cu₂(OH)₂CO₃) via dehydroxylation and decarbonation, as also confirmed by other studies in the literature [28]. Apart from this peak, both materials exhibit an extra weight loss (12.1% for Cu-Al and 11.5% for Cu-Al-Fe) at temperatures below 250 °C, which is attributed to the removal of physically absorbed water and the impurities of the metal precursor used for the synthesis of the materials.

Finally, Mn-Al exhibits two distinct steps of weight loss. The first step of weight loss (9.6%) occurred in the temperature range of 97–308 °C (endodermic peak at DTG_max = 255 °C) and is due to the thermal decomposition of MnCO₃ towards MnO₂ and CO₂, which is slightly higher compared to similar results from the literature [31]. The second step of weight loss (18.9%) is observed in the temperature range of 300–630 °C (endodermic peak at DTG_max= 562 °C) and is due to the further transformation of MnO₂ towards Mn₂O₃ [31]. The material with partial substitution of Al by Fe exhibits a first step of weight loss (9.4%) at DTG_max = 98 °C due to the removal of physically absorbed water molecules. At higher temperatures, multiple steps of weight loss can be observed and can be attributed to the degradation of MnCO₃, as well as to the subsequent oxidation of manganese and iron oxides. The weight loss within the temperature range of 260–700 °C is 12.8%.

Considering the thermal decomposition of the as-made materials, the temperature of 500 °C proved to be a sufficient calcination temperature to obtain a pure phase of mixed oxides.

2.1.2. Characterization of Calcined Mixed Oxides

The calcination of as-prepared materials led to the collapse of the layered structure towards the formation of the respective mixed oxides. The XRD analysis of the calcined Co-Al material reveals that all the reflections correspond to a Co₃O₄ spinel structure of the cubic Fd-3m space group (Figure 3). The existence of the CoAl₂O₄ was excluded due to the absence of reflection (331) at 2θ = 49°, which is attributed only to CoAl₂O₄, as well as based on the ratio of I₂₂₀/I₃₁₁. The ratio I₂₂₀/I₃₁₁ of the Co-Al material is equal to 0.38, which is closer to the standard Co₃O₄ with I₂₂₀/I₃₁₁ = 0.31 than the CoAl₂O₄ with I₂₂₀/I₃₁₁ = 0.7 [32,33]. The crystallite size was calculated via the Scherrer equation based on the peak with the highest intensity, which was equal to 6.7 nm (

Table 2. Physicochemical characterization of mixed oxides.

Materials	Theoretical Molar Ratio	Dcryst a (nm)	SBET (m2/g) b	Vtotal (cm3/g) c	dparticle (nm) d
Co-Al	Co/Al = 2	6.7	139	1.014	200
Co-Al-Fe	Co/(Al+Fe) = 2, Fe/Al = 4	7.5	65	0.888	260
Cu-Al	Cu/Al = 2	9.8	55	1.608	480
Cu-Al-Fe	Cu/(Al+Fe) = 2, Fe/Al = 4	9.2	53	0.367	330
Zn-Al	Zn/Al = 2	4.9	95	1.102	390
Zn-Al-Fe	Zn/(Al+Fe) = 2.2, Fe/Al = 4	12.7	61	0.556	430
Mn-Al	Mn/Al = 2	22.3	134	0.557	710
Mn-Al-Fe	Mn/(Al+Fe) = 2, Fe/Al = 4	-	157	0.863	330

^a Calculated via XRD analysis and the Scherrer equation. ^b Calculated via multipoint BET equation from N₂ sorption at −196 °C. ^c Total pore volume at P/Po = 0.99. ^d Determined via DLS measurements.

). The partial substitution of aluminum by iron led to the formation of Co₃O₄ and Fe₃O₄ spinels and slightly increased the crystallite size to 7.5 nm. The calcination of Zn-Al LDH led to the formation of ZnO in the hexagonal crystal system with a space group of P63mc. The peaks at 2θ = 31.8, 34.5, 36.4, 47.8, 56.7, 63.1, and 68.2° correspond to the diffractions by the (100), (002), (101), (012), (110), (013), and (112) planes. The calculation of the crystallite size was based on the peak (101) at 2θ = 36.4°, which was found to be equal to 4.9 nm. The partial (but significant) substitution of aluminum by iron led to the formation of a ZnFe₂O₄ spinel crystalline phase in the cubic crystal system and the Fd-3m space group, in addition to the ZnO phase like the one formed in the Zn-Al LDH-derived mixed oxide, as can be observed in the XRD pattern of Figure 3. However, the crystallite size of ZnO was calculated to be 12.7 nm (reflection (101) at 36.3°), substantially bigger than the size of ZnO in the Zn-Al LDH-derived material. The very small crystal size of ZnO (derived from Zn-Al LDH) can be attributed to the incorporation of Al³⁺ in the ZnO lattice. Upon calcination of the Zn-Al LDH, Al³⁺ can replace Zn²⁺ in the ZnO crystal structure due to its small ionic radius, which also leads to defect formations. Both of the above factors lead to lower crystallinity and smaller size ZnO phase crystals. On the other hand, when the LDH is deficient in Al (as in the Zn-AlFe LDH), the formed ZnO phase contains much less Al³⁺; thus, crystallization of the ZnO phase is not disturbed, leading to higher crystallinity and larger ZnO crystals.

Regarding the material obtained via the calcination of malachite (Cu-Al) at 500 °C, the formation of monoclinic tenorite (CuO) is observed, and the average crystallite size based on the (111) reflection at 2θ = 35.4° is found to be 9.8 nm. The formation of the tenorite phase via the calcination of malachite is consistent with literature results for the synthesis of Cu-Al LDH-derived mixed oxides [29]. The partial substitution of aluminum by iron did not change the CuO phase, and the extra peaks are assumed to be either CuFe₂O₄ or Fe₃O₄ cubic spinel structure. The average crystallite size calculated based on the (111) reflection at 2θ = 35.5° was found to be 9.2 nm, slightly smaller than the Cu-Al crystallite size. Finally, the calcination of rhodochrosite-enriched Mn-Al material led to the formation of Mn₂O₃ (bixbyite) in the cubic crystal system and I a-3 space group, as the main phase, which is in accordance with TGA and XRD data of the as-made material. The rest of the peaks correspond to Mn₃O₄ of the hausmannite structure (tetragonal, I 41/a m d), as well as MnAl₂O₄ of the cubic spinel structure. The crystallite size was calculated based on the reflection (222) of the Mn₂O₃ structure (2θ = 32.95°) and is equal to 22.3 nm, significantly larger among the mixed oxides synthesized within this work. The substitution of aluminum for iron led to an almost amorphous material with lower crystallinity compared to the Mn-Al. Despite the lower crystallinity, the main phases identified via the XRD analysis are the Mn₂O₃ (bixbyite) phase, the Mn₃O₄ hausmannite structure, and the Fe₃O₄ magnetite phase.

The different crystallinity of the mixed oxide catalyst samples can be attributed to the different composition and structure of the as-synthesized materials. Based on the XRD results, it is observed that the calcination of the well-formed crystalline hydrotalcite phase of both Co-Al and Co-Al/Fe LDHs led to the formation of Co₃O₄ and Fe₃O₄ spinel structures with similar distinct crystallinities. On the other hand, the calcination of Zn-Al and Zn-Al/Fe LDHs induced the formation of ZnO and ZnFe₂O₄. The latter spinel structure is highly crystalline, but the ZnO phase shows somewhat lower crystallinity and smaller crystal size. Thus, it can be suggested that Zn interacts in a different manner with Al and Fe towards the formation of composite oxide/spinel phases than Co does, which induces single metal spinel phases. In the case of Cu-Al/Fe and Mn-Al/Fe systems, the formation of the LDH (hydrotalcite) structure was not successful under the applied synthesis conditions (i.e., mainly pH), leading to the (co)precipitation of (Cu₂(OH)₂CO₃), Al(OH)₃, and MnCO₃, which were subsequently converted upon calcination to the CuO/CuFe₂O₄ and Mn₂O₃/Mn₃O₄/MnAl₂O₄ phases of varying crystallinity.

The porous properties of the calcined materials were studied via nitrogen porosimetry. The nitrogen adsorption–desorption isotherms are shown in Figure 4. All the mixed oxides exhibit isotherms of Type II, which is characteristic of non-porous or macroporous materials without intra-crystal porosity, according to the IUPAC classification [34]. The existence of macropores can be associated with the unrestricted adsorption of nitrogen at high relative pressures. An exception to this is the Mn-Al-Fe material, which exhibits an isotherm of Type IV, corresponding to mesoporous materials. Those results are representative of materials derived via the calcination of LDHs in accordance with other studies [35,36,37]. The same conclusions could also be drawn for the pore size distributions determined via the Barrett–Joyner–Halenda (BJH) method using the adsorption branch. As can be observed in Figure 5, all the materials exhibit relatively broad distributions and a pore diameter > 50 Å. Although the materials are highly macroporous, some mesopores are formed in Co-Al (45 Å) and Zn-Al (45 Å) mixed oxides. Regarding the BET-specific surface areas, they range between 53 and 157 m²/g, as shown in Table 2. The higher values of surface areas correspond to Mn-Al and Mn-Al-Fe materials, followed by Co-Al, while the rest of the materials exhibit significantly lower surface areas. The highest surface area of Mn-Al-Fe materials can be attributed to the lower crystallinity and amorphous mixed oxides phases, as determined by XRD analysis. On the other hand, the pore volumes exhibit a different trend. The materials Co-Al, Zn-Al, and Cu-Al exhibit high pore volumes (V_p = 1.014–1.608 cm³/g) compared to the Mn-Al, which has almost half the pore volume (V_p = 0.557 cm³/g). Furthermore, the partial substitution of aluminum by iron led to a considerable decrease in pore volume. An exception to this trend is the material Mn-Al-Fe, whose pore volume increased upon the substitution of aluminum.

The particle size distribution was determined via Dynamic Laser Scattering (DLS) after dispersion of the materials in an aqueous solution (Figure 6). The materials exhibit a relatively broad distribution with one or more modal diameters, apart from the Cu-Al-Fe and Mn-Al mixed oxides, which exhibit a very narrow distribution, indicative of the high homogeneity in terms of particle size. The particle diameters range from 0.08–0.71 μm, as can be observed in Table 2. The cobalt-based materials exhibit almost the same particle diameters, 0.20 and 0.26 μm, for Co-Al and Co-Al-Fe, respectively. Larger particle diameters are observed for Zn-based materials, with 0.39 and 0.43 μm for Zn-Al and Zn-Al-Fe, respectively. The above results indicate that the partial substitution of aluminum by iron in the LDH-derived mixed oxides led to broader particle size distributions without significant changes in the particle diameters. The Cu-Al mixed oxide exhibits a bimodal and broad particle size distribution. The majority of the particles exhibit diameters of 0.48 μm while the second maximum corresponds to 0.11 μm. The partial substitution of aluminum by iron led to a very narrow distribution and a shift in particle diameter towards lower values at 0.33 μm. Regarding the manganese-based mixed oxides, an opposite trend is observed compared to the copper mixed oxides. The Mn-Al material exhibited a narrow particle size distribution, and the modal particle diameter was close to 0.71 μm, while the substitution of aluminum by iron led to a broad and bimodal distribution with two maxima, at 0.33 μm and 0.08 μm.

2.2. CO₂ Hydrogenation

The m-FTS reaction performance of the LDH-derived mixed oxides was evaluated over a continuous fixed-bed reactor after 6 h on stream, where a pseudo-steady-state was achieved. The CO₂ conversion and the product selectivity in terms of CO, CH₄, and light (C₂–C₄) and heavy (C₅⁺) hydrocarbons are depicted in

Table 3. CO₂ conversion and product selectivity over the tested catalytic systems.

Materials	XCO2 (%)	Product Selectivity (%)
		CO	CH4	C2−C4	C5+
Co-Al	80.3	0.0	98.7	0.2	1.0
Co-Al-Fe	71.7	0.3	70.3	23.0	6.4
Cu-Al	23.0	89.7	7.5	2.8	0.0
Cu-Al-Fe	59.2	2.3	54.3	19.8	23.5
Zn-Al	8.4	86.5	2.6	4.8	6.1
Zn-Al-Fe	40.6	7.8	18.9	35.9	37.3
Mn-Al	19.6	69.4	19.8	8.4	2.5
Mn-Al-Fe	22.3	21.1	30.4	27.8	20.7

Reaction conditions: 340 °C, H₂/CO₂ = 3:1, WHSV = 5.5 L·g_cat^−1.h⁻¹, P = 20 bar, and TOS = 6 h.

, while the hydrocarbon distribution is further illustrated in Figure 7 for all tested samples. The results demonstrate that Co-Al exhibits the highest catalytic performance in terms of CO₂ conversion (80.3), being almost exclusively selective towards CH₄ (98.7%). The incorporation of Fe in Co-Al-Fe slightly reduces the conversion (XCO₂ = 71.7%) but shifts selectivity toward higher hydrocarbons. In contrast, Cu-Al displays significantly lower CO₂ conversion (23.0%) and is predominantly selective for CO (89.7%), indicating minimal hydrogenation activity. The incorporation of Fe enhances the conversion (XCO₂ = 59.2%) and redirects selectivity toward higher hydrocarbon formation, suggesting that Fe promotes chain growth while suppressing CO or CH₄ production. A comparable effect is also observed in Zn-Al, which exhibits poor CO₂ hydrogenation performance, with low conversion (8.4%) and high CO selectivity (86.5%). However, Fe incorporation substantially improves conversion (XCO₂ = 40.6%) and favors the production of longer-chain hydrocarbons, with C₂–C₄ and C₅⁺ selectivity increasing to 35.9% and 37.3%, respectively. Similarly, Mn-Al, which initially demonstrates low CO₂ conversion (19.6%) and moderate CH₄ selectivity (19.8%), experiences a notable shift upon Fe addition, leading to an increase in C₂–C₄ (27.8%) and C₅⁺ (20.7%) fractions. These trends consistently highlight the crucial role of Fe in facilitating hydrocarbon chain growth and enhancing the selectivity of heavier products.

Figure 7 further illustrates the hydrocarbon distribution across the examined samples. Among the detected C₂–C₄ products, paraffins dominate, while light olefin formation remains limited. Notably, Co-Al-Fe and Cu-Al-Fe exhibit minimal C₂–C₄⁼ production, indicating an inhibition of olefin formation. In contrast, Zn-Al-Fe and Mn-Al-Fe demonstrate a more favorable selectivity towards light olefins, yielding 11.4% and 8.3% C₂–C₄=, respectively. Notably, it is demonstrated that Fe incorporation systematically shifts selectivity away from C₁ products, with varying efficiency depending on the type of X metal in the X-(Al+Fe) initial LDH structure.

The present findings unambiguously reveal the combining effect of Fe with a second transition metal toward determining the conversion and selectivity performance of the LDH-derived mixed oxide catalysts. The basic, Fe-free, Co-Al sample is highly active for methane production, whereas Zn-, Mn-, and Cu-Al LDHs are selective towards CO via the rWGS reaction. This trend is notably changed when the same transition metals are combined with Fe, leading to the formation of higher hydrocarbons at the expense of C₁ products, with C₂⁺ yield following the order Zn-Al-Fe > Cu-Al-Fe > Mn-Al-Fe > Co-Al-Fe >> Mn-Al, Co-Al, Zn-Al, Cu-Al. Interestingly, the Zn-Al-Fe composites provide a yield to higher hydrocarbons of 29.7%, with an almost equal distribution between C₂–C₄ (14.6%) and C₅⁺ (15.1%), as illustrated in Figure 8. Similarly, Cu-Al-Fe exhibits a notably higher hydrocarbon yield (25.6%), with a slight preference for C₅⁺ production (13.9%). In contrast, Mn-Al-Fe, despite its lower overall product yield, demonstrates a stronger selectivity toward C₂–C₄ hydrocarbons.

Motivated by these encouraging results, work is in progress towards improving the yield to higher hydrocarbons, as well as their distribution, by appropriately adjusting the amount and ratio of Zn and Fe in the LDH structure.

2.3. Characterization of Spent Catalysts

The spent catalysts recovered after the CO₂ hydrogenation were analyzed via elemental analysis to determine the amount of carbon deposits, as well as by Thermogravimetric Analysis (TGA) in air and X-Ray Diffraction (XRD) to examine any changes in the crystal structures. The structure of spent catalysts seems to be quite complex and is strongly dependent on the metal combinations. Co-Al material exhibits high stability, and the cubic spinel structure of Co₃O₄ is maintained after the CO₂ hydrogenation, as can be observed in Figure 9. Peaks of lower intensity are attributed to metallic cobalt (Co⁰) in the hexagonal crystal system. Furthermore, the average crystallite size of the catalyst used is 7.6 nm, which is slightly increased from 6.7 nm in the fresh catalyst. On the other hand, the Co-Al-Fe structure seems to undergo significant changes during the hydrogenation of carbon dioxide. The peaks are assigned to cubic cobalt oxide (CoO), cubic spinel structure (CoFe₂O₄), and the metallic phase of iron (Fe⁰) or cobalt (Co⁰), which, due to the same crystal system, cannot be efficiently distinguished via XRD analysis. It can be assumed that during the hydrogenation reaction, the reductive conditions induce a reduction of Co₃O₄ towards CoO formation and most probably to metallic Co. Similarly, the iron oxide is reduced towards metallic iron. The average crystallite size was calculated based on the (200) reflection of CoO and found to be 10.3 nm.

Zn-Al is not affected by CO₂ hydrogenation, and all the reflections correspond to the hexagonal zincite (ZnO) structure of the P 63 mc space group. Only the crystallite size increased from 4.9 nm (fresh catalyst) to 7.6 nm due to thermal sintering during the reaction. Furthermore, the Zn-Al-Fe material maintains the structure of cubic spinels Fe₃O₄, ZnFe₂O₄, and the hexagonal zincite (ZnO) structure. The average crystallite size that was calculated based on the most abundant phase (ZnO, 2θ = 34.4°) is equal to 17.6 nm, larger than the Zn-Al spent catalyst, as well as the fresh Zn-Al-Fe catalyst, which exhibits a crystallite size equal to 12.7 nm.

The reductive environment during the CO₂ hydrogenation or during the pretreatment under the CO atmosphere induced a complete reduction of the CuO phase in Cu-Al mixed oxide towards the formation of metallic copper (Cu⁰), as can be observed in Figure 9. All the reflections correspond to (111), (200), and (202) planes at 2θ = 43.3, 50.5, and 74.2°, respectively. The average crystallite size of metallic copper is 17.1 nm. The partial reduction of the CuO phase to metallic Co is also observed in the Cu-Al-Fe material. In that material, the peaks can be assigned to monoclinic tenorite CuO, the cubic structure of metallic copper, as well as to the cubic spinels Fe₃O₄ and CuFe₂O₄. The crystallite size was calculated for the (111) reflection of Cu⁰ at 2θ = 43.3° and found to be equal to 24.4 nm, bigger than the Cu-Al spent catalyst.

Manganese-based mixed oxides are the most vulnerable materials according to the XRD analysis. The analysis revealed that the main crystal structure is the rhodochrosite (MnCO₃) in the trigonal system and R-3c space group, which formed due to the adsorption and interactions of manganese oxides with CO_x species during the CO₂ hydrogenation [38,39]. In the Mn-Al pattern, peaks corresponding to cubic MnO were also identified. The crystallite size calculation is based on the reflection (104) of the MnCO₃ phase and found to be 30.2 nm, slightly increased compared to the fresh catalyst. In the Mn-Al-Fe material with the partial substitution of aluminum by iron, the main structures identified are MnCO₃ and Fe₃O₄. The latter structure proved to be more stable than the manganese oxides and is less affected by carbon species. The crystallite size was calculated based on the (104) reflection of the MnCO₃ phase and found to be 30 nm, similar to the Mn-Al material.

Further physico-chemical characteristics of the spent catalysts were derived by thermogravimetric analysis (Figure 9). The spent Co-Al material does not exhibit any weight loss in the TGA, which is indicative of negligible carbon deposition, as also confirmed by elemental analysis (

Table 4. Physicochemical characterization of spent catalysts.

Materials	Dcryst (nm) a	C (wt.%) b	Weight Gain (wt.%) c	DTGmax (°C) c	Weight Loss (wt.%) c	DTGmax (°C) c
Co-Al	7.6	0.0	3.3	264	-	-
Co-Al-Fe	10.3	32.8	1.0	250	27.0	342–383
Cu-Al	17.1	0.2	9.8	314	-	-
Cu-Al-Fe	24.4	8.3	2.7	254	12.0	367
Zn-Al	7.6	1.2	-	-	4.7	136
Zn-Al-Fe	17.0	1.7	-	-	-	-
Mn-Al	30.2	4.4	-	-	14.0	525
Mn-Al-Fe	30.0	8.4	-	-	23.8	477

^a Determined via XRD analysis and Scherrer equation ^b Determined via elemental analysis. ^c Determined via TGA.

). The small weight gain (3.3%) in the temperature range of 170–375 °C (DTG_max = 264 °C) is attributed to the adsorption of oxygen and the subsequent oxidation of metallic cobalt to cobalt oxide. This phenomenon is more obvious in the analysis of spent Co-Al-Fe materials. A first weight gain is observed in the temperature range of 158–330 °C, which is attributed to the oxidation of metals (Co or Fe) to the respective metal oxides. The results are in accordance with the literature results, where the oxidation of metallic iron occurs in the temperature range of 200–250 °C, in TGA run under an air atmosphere [40,41,42]. Afterwards, a weight loss (28%) coupled with an exothermic peak is observed in the temperature range of 330–500 °C, which is attributed to the coke burning. Indeed, the carbon content measured via elemental analysis is 32.8 wt.%, which is very close to the weight loss in TGA.

Regarding the copper-based materials, in the Cu-Al spent catalyst, the weight changes are performed in two distinct steps, at DTG_max = 163 °C and DTG_max = 314 °C with a total weight gain of 9.8%. The first increase in weight can be assigned to the oxidation of metallic Cu to Cu₂O, and the second increase at higher temperatures to the subsequent oxidation of Cu₂O to CuO, in which reactions take place at 150–250 °C and 250–350 °C, according to the literature [43,44]. The partial substitution of aluminum by iron led to an initial weight gain of 2.7% within the temperature range of 220–317 °C (DTG_max = 254 °C), followed by a weight loss (12.0%) at 320–530 °C. The weight gain is assigned to the oxidation of metallic copper to copper oxide. The phenomenon is shifted to lower temperatures compared to the Cu-Al spent catalysts due to the lower amount of metallic copper, as confirmed by XRD analysis. Furthermore, the weight decrease is associated with the burning of coke deposits, which are significantly higher than Cu-Al, as can be observed via the elemental analysis (Table 4). The carbon content determined for the Cu-Al-Fe spent catalyst is 8.3 wt.%, higher than the 0.2 wt.% of Cu-Al.

The Zn-Al and Zn-Al-Fe materials exhibit almost no weight changes during the thermogravimetric analysis. As can be observed in Figure 9, the Zn-Al exhibits a weight loss (4.7%) at very low temperatures of 24–360 °C (DTG_max = 136 °C), which is attributed to the absorbed water or C₅⁺ formed during the hydrogenation of CO₂. Similarly, the TGA of Zn-Al-Fe did not show any changes in weight. Combining the results obtained via the thermogravimetric, elemental, and XRD analyses, it can be concluded that the Zn-Al and Zn-Al-Fe are the most stable materials with almost no changes in the crystal structure and very low carbon formation (<2 wt.%) compared to the rest materials of this work. Zn-Al-Fe LDHs are highly active in the formation of higher hydrocarbons (Figure 10), rendering them promising materials for CO₂ hydrogenation to hydrocarbons. To the best of our knowledge, this is the first work revealing the adequate catalytic performance in terms of activity and stability of Zn-Al-Fe LDHS for CO₂ hydrogenation. A different behavior is exhibited by the manganese-based catalysts. The spent catalysts recovered after the CO₂ hydrogenation exhibit a single-step weight loss in the temperature range of 300–620 °C, which can be attributed to the decomposition of MnCO₃, the main structure identified via XRD analysis. More specifically, in Mn-Al, a weight loss of 14% is centered at DTG_max = 525 °C, while in the Mn-Al-Fe, the weight is slightly higher at 23.8%, and the maximum rate of weight loss is at DTG_max = 477 °C. The shift at lower temperatures is attributed to the presence of coke deposits on the catalysts, as also confirmed by the elemental analysis, where the Mn-Al-Fe spent catalyst exhibits a higher carbon amount (8.4 wt.%) compared to the Mn-Al, which exhibits 4.4 wt.%.

Based on the above results for the characterization of spent catalysts, it can be concluded that cobalt and copper-based catalysts exhibited a significant reduction during the CO₂ hydrogenation process, as confirmed by the XRD analysis (identification of metallic phases) and the thermogravimetric analysis (increase of weight due to oxidation of metals to metal oxides). On the other hand, in manganese-based oxides, manganese carbonates are formed due to the strong adsorption of carbon dioxide or carbon monoxide. The results from the XRD analysis are consistent with the TGA results and the identification of MnCO₃ as the predominant phase in spent catalysts. Finally, zinc-based materials are the most stable materials with almost negligible changes in structure, as confirmed via TGA and XRD analysis, and they have very low deposited carbon. In general, the addition of iron enhanced the activity of the materials but also induced coke formation, as can be observed via the elemental analysis.

3. Discussion

In the present study, the CO₂ hydrogenation performance of LDH-derived transition metal mixed oxides (Co, Cu, Zn, Mn) was evaluated using a continuous fixed-bed reactor under identical conditions. The aim was to investigate the influence of these metals, as well as their interaction with Fe, on CO₂ conversion, product selectivity, and hydrocarbon chain growth. The obtained results were complemented with detailed characterization of the spent catalysts, including XRD, Thermogravimetric Analysis (TGA), and elemental analysis, to gain insight into the structural, redox, and material stability changes arising from the reactions.

XRD analysis provided crucial insights into the structural evolution of the catalysts, revealing significant differences between the calcined and spent samples. The Co-Al catalyst was initially composed of cubic spinel Co₃O₄, which remained largely intact after the reaction, with only minor crystallite growth due to sintering. In contrast, the incorporation of Fe in Co-Al-Fe led to substantial phase transformations, with the formation of CoO, CoFe₂O₄, and metallic Co and Fe phases. The presence of CoFe₂O₄ spinels highlights Fe’s ability to interact with cobalt, potentially modifying the electronic environment of active sites [45]. Furthermore, the addition of iron in cobalt oxide and the subsequent formation of the above-mentioned structure during CO pretreatment or in situ during CO₂ hydrogenation has already been proven to shift the selectivity of the reaction from methane to other hydrocarbons due to the easier formation of iron carbides [46,47,48].

A similar reduction behavior was observed in Cu-based catalysts. While calcined Cu-Al initially contained monoclinic CuO, the spent catalyst exhibited complete reduction to metallic Cu (Cu⁰) under reaction conditions, leading to significant sintering. This behavior is linked to a dominant selectivity for CO due to copper’s limited ability to further hydrogenate CO intermediates [49]. Unlike in the Co system, where Fe induced strong phase transformations, the Cu-Al-Fe system retained CuFe₂O₄ and Fe₃O₄ phases, along with partially reduced Cu, indicating that Fe incorporation stabilized copper oxides to some extent, preventing full reduction. This suggests that Fe modifies the redox dynamics of Cu, affecting the balance between CO and hydrocarbon formation. As a result, the Cu-Al-Fe catalyst exhibited lower CH₄ selectivity compared to Co-Al-Fe systems, where phase transformations were more pronounced.

Zn-based catalysts displayed a high degree of structural stability. The Zn-Al catalyst, composed primarily of hexagonal ZnO, remained virtually unchanged after the reaction, with only a slight increase in crystallite size due to sintering. Similarly, Zn-Al-Fe, which retains both ZnO and ZnFe₂O₄ phases after reaction, shows a pronounced shift in product selectivity toward higher hydrocarbons, particularly light olefins. The presence of Fe₃O₄ and ZnFe₂O₄ appears to create active sites that favor hydrocarbon chain propagation over CO desorption [50,51,52]. Unlike Co and Cu systems, Zn-based catalysts exhibited minimal carbon deposition, further supporting their potential as highly active and stable hydrogenation catalysts.

Manganese-based catalysts exhibited significant structural changes, with the Mn-Al catalyst initially containing MnO and amorphous Mn oxides, transforming into MnCO₃ upon reaction. This indicates strong interactions with CO_x species and demonstrates Mn’s tendency to form carbonates, potentially hindering catalytic performance. The Mn-Al-Fe comprised of Fe₃O₄ and MnCO₃ showed higher carbon deposition, suggesting that Fe promotes hydrocarbon chain growth at the cost of increased coke formation. In this system, the presence of Fe₃O₄ correlated with a shift in product distribution toward higher hydrocarbons but also increased catalyst deactivation due to coking [53,54].

The thermogravimetric analysis of the calcined and spent samples further supports these structural transformations observed in XRD analysis. Zn-based catalysts (Zn-Al and Zn-Al-Fe) exhibit the highest thermal stability, with minimal weight changes and negligible carbon deposition, maintaining their crystalline structure after reaction. The limited tendency for coke formation of the Zn-Al-Fe catalyst is probably attributed to the different basicity of those catalysts and the highly active and stable ZnFe₂O₄ phase. The formation of the spinel structure prevents the excessive Fe carburization and the formation of Fe⁰, which is known to promote the Boudouard reaction and carbon deposition [11,55,56,57]. The presence of Fe₃O₄ rather than Fe⁰ suggests a controlled oxidation–reduction balance that limits iron carbide overgrowth and reduces coking [50,58]. Additionally, ZnFe₂O₄ enhances surface basicity, promoting CO₂ adsorption and reducing carbon accumulation [58,59,60]. Beyond structural stabilization, Zn addition directly modifies catalyst selectivity and deactivation behavior. Zinc promotes RWGS activity, increasing CO formation while reducing direct carbon-forming pathways, which suppresses coking [10,58]. Compared to Mn, Cu, and Co, which promote deeper iron carburization and Fe-rich metallic sites that accelerate carbon buildup, Zn stabilizes Fe₃O₄, prevents excessive iron carbide formation, and reduces Fe sintering, thereby maintaining catalyst dispersion and limiting carbon deposition [10,56]. These effects contribute to the improved long-term stability of Zn-Fe-Al catalysts, as ZnFe₂O₄ remains structurally intact even after extended reaction periods, further reinforcing its role in reducing coke deposition and preventing deactivation [59].

In contrast, Fe-containing Co, Cu, and Mn catalysts experience significant coke formation, as indicated by notable weight loss in TGA and elevated carbon content from elemental analysis. This aligns with XRD findings, which reveal phase transformations and the formation of metallic phases in these catalysts. Additionally, Co and Cu-based systems show oxidation in air, evidenced by weight gain in TGA, reinforcing the XRD observations of redox cycling between oxide and metallic states. These findings demonstrate that while Fe enhances catalytic performance by promoting hydrocarbon chain growth, it also accelerates carbon deposition, which can reduce long-term stability. Consequently, the reasons and mechanisms of coke formation are quite complex and strongly dependent on catalyst basicity and phase composition, which can influence the CO_x sorption and carburization degree [12,61].

The catalytic data confirm that Fe incorporation universally promotes hydrocarbon chain growth, but its impact on CO₂ conversion and selectivity varies by system. While Co-Fe catalysts retain a strong tendency toward CH₄ formation, Fe addition enables higher hydrocarbon yields, aligning with findings from comparable Co-Fe systems [10,62,63]. Despite a slight decrease in conversion, the Fe-Co-Al catalyst maintained a remarkably high CO₂ conversion relative to reported samples, highlighting the effectiveness of the preparation method. Similarly, Fe in Cu-based catalysts shifted selectivity from CO toward hydrocarbons while stabilizing Cu oxides, which is in agreement with established Fe-Cu behavior [17,64,65]. Zn-based catalysts primarily favored CO formation, but Fe addition significantly enhanced light olefin production, a trend consistent with previous reports on Fe-Zn interactions [66,67]. The Mn-based catalyst displayed the most distinct transformation, with Fe shifting selectivity from CO toward hydrocarbons but at the cost of increased carbon deposition.

Overall, the incorporation of Fe into the Fe-based catalysts consistently promotes the formation of mixed oxide phases—such as CoFe₂O₄, CuFe₂O₄, ZnFe₂O₄, and Fe₃O₄—which correlates with a shift in product selectivity toward higher hydrocarbons and light olefins, as evidenced in similar catalytic systems shown in

Table 5. CO₂ hydrogenation performance and product selectivity for similar catalytic systems reported in the literature.

Catalyst	Reaction Conditions T (°C), P (bar), WHSV (mL·gcat−1.h−1)	XCO2 (%)	SCH4 (%)	SCO (%)	SC2–C4 (%)	SC2+ (%)	Ref.
ZnFe2O4 a	320, 25, 4800	42.1	10.1	10.7	28.8	79.0	[10]
Fe6Zn1Al1	330, 15, 15,000	39.1	12.4	22.5	21.1 c	62.6	[66]
10Fe3Zn1K/Al2O3	400, 30, 3600	38.6	35.8	33.8	24.1	30.9	[17]
FeZn	300, 10, 5000	40.6	27.3	10.7	-	46.9	[67]
Fe-Zn-Al	320, 10, 13,000	31.6	11.9	20.7	36.2	68.08	[68]
ZnCo0.5Fe1.5O4 a	320, 25, 4800	49.6	17.8	5.8	39.8	76.2	[10]
FeCo-1:2-LDH a	320, 30, 7200	48.2	43.9	4.2	36.7	51.9	[62]
Co-Fe a	240, 30, 5500	19.6	68.1	2.9	2.2	28.8	[63]
ZnCo0.5Fe1.5O4 a	320, 25, 4800	49.6	17.8	5.8	39.8	76.2	[10]
Fe-Co/Al2O3	300, 11, 3600	25.0	44.0	13.0	-	43.0	[69]
Co4Fe1	320, 20, 8000	43.7	34.1	7.6	31.7	53.5	[9]
Na-AlFeCu a	320, 50, 35 b	44.5	13.6	9.9	23.8	76.3	[65]
Fe–Cu(0.75)/Al2O3	300, 11, 3600	22.8	18.0	45.0	-	37.0	[8]
10Fe3Cu1K/Al2O3	400, 30, 3600	41.7	27.8	26.5	31.9	45.7	[17]
0.3-CuFe/ZrO2	320, 20, 4800	35.4	12.3	11.9	38.0	77.1	[64]
10Fe3Mn1K/Al2O3	400, 30, 3600	42.0	36.1	23.0	29.8	40.9	[17]
3MnNaFe	290, 15, 20,000	30.1	30.9	24.2	33.3	43.9	[70]
10Mn-Na/Fe	320, 30, 2040	37.7	14.0	12.9	34.1	73.8	[71]
Fe3O4	320, 30, 2000	29.3	50.2	16.6	30.4	33.02	[72]
Fe-Al-O Nanobelts	300, 10, 1 d	48.0	10.0	16.0	57.0 e	69.0	[73]
FeAl350	330, 15, 9000	48.2	17.3	10.1	38.1	72.6	[74]

^a LDH-derived samples, ^b gcat^.mol^−1.h⁻¹, ^c C₂–C₃ products, ^d h⁻¹, ^e C₂–C₅ products.

. These findings highlight the complex interplay among phase stability, reducibility, and catalytic function in Fe-based systems. While Fe enhances hydrocarbon chain growth, it also alters the reduction behavior of Co, Cu, Zn, and Mn oxides, leading to distinct structural and electronic environments in the spent catalysts. The TGA results confirm these changes, particularly showing increased carbon deposition in Fe-modified Co, Cu, and Mn catalysts. To further elucidate these phase transformations in real time, future studies should employ operando and spectroscopic techniques, providing deeper insight into the nature of active sites. In this direction, work is in progress towards optimizing Zn-Al-Fe composites by adjusting the metal loading at divalent and trivalent positions of LDHs while unveiling the underlying mechanism of high activity and stability originated by Zn-Fe synergistic effects.

Building on these findings, ongoing efforts are focused on refining Zn-Al-Fe composites by adjusting metal loading at the divalent and trivalent positions of LDHs. This optimization aims to unravel the underlying mechanisms governing their high activity and stability, potentially opening new paths for improved catalytic performance in CO₂ hydrogenation.

4. Materials and Methods

4.1. Materials

The catalysts were synthesized using Cu(NO₃)₃·3H₂O (99%, Merck, Darmstadt, Germany), Co(NO₃)₃·6H₂O (99%, Strem, Newburyport, MA, USA), Mn(NO₃)₃·4H₂O (>97%, Fluka, Seelze, Germany), Zn(NO₃)₃·9H₂O (99%, Merck, Darmstadt, Germany), Al(NO₃)₃·9H₂O (>98%, Sigma Aldrich, St. Louis, MO, USA), Fe(NO₃)₃·9H₂O (>98%, Fluka), as metal precursors while NaOH (99%, ChemLab, Kenilworth, NJ, USA) and Na₂CO₃ (99.8%, Panreac, Barcelona, Spain) were used for the precipitation.

4.2. Synthesis of Catalysts

The mixed oxides were derived via the calcination of layered double hydroxides (LDHs). The LDHs were synthesized via the co-precipitation method followed by hydrothermal treatment. In a typical synthesis, appropriate amounts of metal precursors were dissolved in deionized water under stirring. After complete dilution, they were mixed with a molar ratio M²⁺/M³⁺ = 2.0–2.2 and added dropwise to an aqueous solution of Na₂CO₃ (moles CO₃²⁻/moles M³⁺ = 2.5) under vigorous stirring at room temperature. The pH was maintained in the range 9–10 by adding dropwise a 50% (w/w) NaOH aqueous solution. Afterwards, the dispersions were left under stirring for 18 h in sealed polypropylene bottles at 60 °C for aging. The precipitate was filtered, washed several times with deionized water, and dried at room temperature. The final mixed oxides were derived via the calcination of LDHs at 500 °C for 2 h and a heating rate of 2 °C/min.

4.3. Characterization of Catalysts

The total metal content of the samples was determined by Atomic Emission Spectroscopy (ICP-AES, Plasma 40, Perkin Elmer instrument, Waltham, MA, USA) after the appropriate dissolution of the solid samples. The structural characterization of LDHs and mixed oxides was performed via X-ray Powder Diffraction (XRD) using a Rigaku Ultima+ 2cycles X-ray Diffractometer (Akishima, Japan) with CuKa X-ray radiation. The materials were scanned over 2θ = 5–85° range with 0.02°/step and 2 s/step. The porous properties of mixed oxides were determined via N₂ adsorption–desorption at −196 °C using an Automatic Volumetric Sorption Analyzer (Autosorb-1 MP, Quantachrome, Boynton Beac, FL, USA). Prior to the measurements, the samples were outgassed at 150 °C for 19 h under a 5 × 10⁻⁹ Torr vacuum. The total surface areas were determined via the multipoint BET method, the total pore volume was determined at P/Po = 0.99, while the pore size distribution was obtained via the Barrett–Joyner–Halenda (BJH) method. Particle size distribution was obtained using a Dynamic Light Scattering (DLS, Anton Parr, Litesizer 500, Graz, Austria) instrument. Before the measurement, 10 mg of mixed oxides were dispersed in 10 mL of deionized water and ultrasonicated for 10 min.

Spent catalysts were characterized by XRD, elemental, and Thermogravimetric Analysis (TGA). Elemental analysis was performed using the Eurovector EA3100 Series CHNS-O elemental analyzer (Pavia, Italy). The oxygen content was calculated by difference. Thermogravimetric analysis was carried out using a Netzsch STA449F5 instrument (Selb, Germany) under airflow (50 mL/min) and a constant heating rate of 10 °C/min in the temperature range of 25–950 °C.

4.4. Catalytic Activity Test

Catalytic performance tests were carried out in a fixed bed reactor with a 13 mm inner diameter arranged in a down-flow orientation. In the center of the catalyst bed, a K-type thermocouple was placed to monitor the catalyst’s temperature. Typically, 1 g of catalyst was loaded into the reactor, located between plugs of quartz wool, and activated under 20% CO/He in situ at 350 °C and atmospheric pressure for 5 h. Consequently, the catalyst was cooled to room temperature under pure N₂ flow and pressurized to 20 bar after exposure to a stoichiometric reactant mixture of H₂:CO₂ = 3 and WHSV = 5500 mL·g_cat⁻¹·h⁻¹. A cold trap was used to collect the liquid products before gaseous product analysis. The latter were quantified by an online gas chromatograph (Shimadzu GC-2014, Kyoto, Japan) equipped with a Molecular Sieve 13X column (Restek, Bellefonte, PA, USA) connected to a Thermal Conductivity Detector (TCD, Shimadzu, Kyoto, Japan), while the hydrocarbons were quantified in a Porapak Q packed column (Restek, Bellefonte, PA, USA) connected to a Flame Ionization Detector (FID, Shimadzu, Kyoto, Japan). The reaction was maintained for at least 6.5 h on stream, and the CO₂ Conversion (${\mathrm{X}}_{{\mathrm{C}\mathrm{O}}_2}$), product selectivities (${\mathrm{S}}_{\mathrm{C}\mathrm{O}}$, ${\mathrm{S}}_{\mathrm{C}i}$, $S_{{C5}^{+}}$) and yields (${{\rm Y}}_{Ci}$,${{\rm Y}}_{{C2}^{+}}$) were determined through the following Equations (1)–(6):

$${\mathrm{X}}_{{\mathrm{C}\mathrm{O}}_2}\left(\%\right)={\displaystyle \frac{{CO}_{2_{in}}-{CO}_{2_{out}}}{{CO}_{2_{in}}}}·100$$

(1)

$${\mathrm{S}}_{\mathrm{C}\mathrm{O}}(\%)={\displaystyle \frac{{CO}_{out}}{{CO}_{2_{in}}-{CO}_{2_{out}}}}·100$$

(2)

$${\mathrm{S}}_{\mathrm{C}i}(\%)={\displaystyle \frac{moleof{C}_{i_{out}}\times i}{{CO}_{2_{in}}-{CO}_{2_{out}}}}·100$$

(3)

$$S_{{C5}^{+}}\left(\%\right)=100-{\mathrm{S}}_{\mathrm{C}O}\left(\%\right)-{\mathrm{S}}_{C1-C4}\left(\%\right)$$

(4)

$${{\rm Y}}_{Ci}\left(\%\right)={\displaystyle \frac{{\mathrm{X}}_{{\mathrm{C}\mathrm{O}}_2}\times {\mathrm{S}}_{\mathrm{C}i}}{100}}$$

(5)

$${{\rm Y}}_{{C2}^{+}}\left(\%\right)={\displaystyle \frac{{\mathrm{X}}_{{\mathrm{C}\mathrm{O}}_2}\times {\mathrm{S}}_{\mathrm{C}2-\mathrm{C}4}\times S_{{C5}^{+}}}{100}}$$

(6)

where ${CO}_{2_{in}}$ represents the molar flow of CO₂ in the inlet of the reactor, and ${CO}_{2_{out}}$, CO_out, and $C_{i_{out}}$ represent the molar flow of CO₂, CO, and C₁–C₄ hydrocarbons in the outlet of the reactor.

5. Conclusions

In this study, X-Al and X-(Al+Fe) LDH-derived catalysts (X = Co, Cu, Zn, Mn) were evaluated for CO₂ hydrogenation. It was shown that the catalytic performance (activity, selectivity, coking) depends greatly on the composition, i.e., type and interaction between the different metals. Co-Al containing mixed oxide/spinels favor CH₄ formation, while Cu-Al and Zn-Al-based catalysts produce mainly CO. Partial replacement of Al by Fe (Fe/Al molar ratio 4:1) in the LDH structure enhanced the selectivity of the derived mixed oxide/spinel catalysts towards higher hydrocarbons (C₂–C₄). Among the studied catalysts, Zn-Al-Fe emerged as the most stable and promising material, combining improved CO₂ conversion (40.6%) with an enhanced C₂⁺ yield (29.7%) and minimal carbon deposition. In contrast, Fe-modified Co, Cu, and Mn catalysts experienced significant coke formation. These preliminary results highlight the remarkable synergistic effect of Fe addition in promoting C–C coupling and underscore the potential of Zn-Al-Fe catalysts for further modification to optimize product distribution toward added-value chemicals.