Metal oxide overlayer- and bulk metal oxide-supported Fe3Ni-based catalysts for CO2 activation via the reverse water-gas shift reaction

Verfaille, Thierry Philippe

Metal oxide overlayer- and bulk metal oxide-supported Fe3Ni-based catalysts for CO2 activation via the reverse water-gas shift reaction

Thesis / Dissertation

2026

Publisher

University of Cape Town

Department

Department of Chemical Engineering

Faculty

Faculty of Engineering and the Built Environment

Abstract

In order to combat the growing effects of global warming, the research and development of carbon capture and utilisation (CCU) and carbon capture and sequestration (CCS) technologies must be treated as a priority. For such technologies to be capable of making a meaningful impact, large-scale and cost-effective processes must be developed. The Power-to-Liquids (PtL) pathway, whereby readily available renewable energy is harnessed to synthesise carbon-neutral liquid hydrocarbon fuels from green H2 and captured CO2 via the reverse water-gas shift (RWGS) and Fischer-Tropsch synthesis (FTS), is among the viable proposed solutions. Whilst the FTS is well-understood, significant catalyst development efforts are still necessary in the RWGS in order for the PtL pathway to be considered viable. The challenges associated with the RWGS are due to it being a mildly endothermic and thermodynamic equilibrium-limited reaction. As a result, the RWGS requires a catalyst which can kinetically suppress the competing CO methanation (MCO) and CO2 methanation (MCO2) reactions which are thermodynamically favoured under industrially relevant conditions, involving high gas hourly space velocities (GHSV), H2:CO2 feed ratios and operating pressures, at medium temperatures. Studies conducted thus far have identified bimetallic FeNi-based catalysts as potential candidates to facilitate the RWGS at close to equilibrium CO2 conversion while attaining near-complete CO selectivity. While previous studies have gone so far as to identify the likely optimal Fe:Ni ratio of 3:1 for RWGS operation at mild operating conditions, significant variations in performance have been observed over different support materials. Correlations have been observed between the extent of CO2 activation and surface acidity and reducibility of a given support material, but further investigations over a larger library of supports are required to fully understand the interactions taking place between the metal and support under RWGS conditions. In the present work, a series of Fe3Ni-based catalysts are synthesised by employing a non-aqueous surfactant-free hydrothermal synthesis route to form (Ni0.75Fe0.25)Fe2O4 inverse spinel nanoparticles. These nanoparticles are reduced to form Fe3Ni nanoalloys, after being supported on the chosen support materials by ultrasonication. These catalysts are then tested in three phases under increasingly harsh operating conditions, including those considered to be of industrial relevance for the viability of the PtL pathway. The first phase of testing involves the use of Fe3Ni nanoalloys supported on metal oxide-overlayers of CrOx, GaOx, SmOx, TiOx, VOx and ZrOx coated on a common γ-Al2O3 carrier. Unlike using bulk metal oxides, the use of such Fe3Ni/MOx@Al2O3 catalysts allows for the influence of the supports to be studied with focus on the various surface chemical properties and with decoupled textural properties. Analysis of the CO2 activation behaviours of the different Fe3Ni/MOx@Al2O3 catalysts in the absence of H2 is used to provide insight into the dominant RWGS reaction mechanism. Additional analyses performed on both the fresh, reduced and spent catalysts are used to develop an understanding of the RWGS active sites, in the context of the differing surface chemistry of the support materials. The second phase of the study involves the testing of bulk metal oxide-supported Fe3Ni-based catalysts, where the difference in performance between these and their associated metal oxide overlayer-supported catalysts is investigated. The results of the RWGS testing in this phase are used to gain insights into the accuracy of using metal oxide overlayered supports to forecast the strength of the interactions between the metal and the support that take place on bulk metal oxide-supported Fe3Ni-based catalysts. Furthermore, this testing phase is used to identify which of the bulk metal oxide-supported Fe3Ni-based catalysts, if any, show promise for further testing at conditions of relevance to the PtL pathway. The third phase of RWGS testing takes those catalysts which are identified in the second phase as having the potential to facilitate the RWGS approaching equilibrium CO2 conversion and investigates their catalytic performance in the RWGS at increasingly harsh operating conditions. Initially, the effect of increasing GHSV on CO2 conversion and CO selectivity is investigated, which is followed by repeating the same experiment, but in a stepwise fashion and at an increased H2:CO2 feed ratio. Finally, the most promising candidate catalyst is tested at elevated operating pressure. Temperature-programmed CO2 activation (CO2-TPA) experiments indicate that not all Fe3Ni/MOx@Al2O3 catalysts are capable of directly activating CO2. Furthermore, the extent to which this occurs per catalyst correlates with the surface acidity of the support, suggesting that there is a surface acidity threshold below which direct CO2 activation is no longer possible. CO2 activation is also observed to be significantly more facile on the two catalysts supported on reducible metal oxide overlayers, Fe3Ni/CrOx@Al2O3 and Fe3Ni/VOx@Al2O3, indicating that support reducibility also plays a role in enhancing this behaviour. RWGS experimental results in Phase 1 correlate with those of the CO2-TPA experiments, with a strong synergistic effect between the alloy and supports boosting CO2 conversion on the most acidic and reducible supports. A moderate synergistic effect is observed on moderately acidic but nonreducible supports, and negligible synergistic effect observed on weakly acidic and nonreducible supports. Together, these results could suggest that the dominant reaction mechanism taking place on these catalysts is the MvK-type mechanism, given that those catalysts demonstrating a synergistic effect between the nanoalloy and support in the RWGS are the ones able to directly activate CO2 in the absence of H2. STEM-EDX analyses of the freshly reduced and RWGS spent catalysts reveal that, under reaction conditions, Fe species exsolve out of the Fe3Ni nanoalloys and become dispersed over the support surface, with the extent of dispersion directly correlating to the extent of the synergistic interaction between the metal and support observed in the RWGS testing. These Fe species, believed to be amorphous FeOx species, are thought to enhance catalytic activity through the formation of specific FeOx-MOx interfacial sites, which boost the activation of CO2. Finally, it is shown that the exsolution and dispersion of these FeOx species from the Fe3Ni nanoalloy is the likely cause for the observed BCC to FCC phase transformation seen on these catalysts under reaction conditions. Phase 2 results show that there is no correlation between the RWGS performance of a given Fe3Ni/MOx@Al2O3 and its related bulk metal oxide-supported Fe3Ni-based catalyst. No correlation between the surface acidity of the support and the strength of the interaction between the metal and support can be drawn from these results. Both the Fe3Ni/V2O5 and Fe3Ni/ZrO2 catalysts approach equilibrium CO2 conversion under the reaction conditions, and are chosen for Phase 3 testing. The Phase 3 experiments reveal a large gap in performance between the two selected catalysts. It is observed that increasing GHSV does reduce CO2 conversion on both catalysts, as well as catalyst stability on Fe3Ni/V2O5, but does not have a strong effect on CO selectivity. Increasing the H2:CO2 feed ratio results in an increase in CO2 conversion in line with the increase in the thermodynamic equilibrium conversion, but reduces CO selectivity. Finally, the Fe3Ni/ZrO2 catalyst is tested at 20 bar in combination with the maximum GHSV and H2:CO2 feed ratio tested prior. This increase in operating pressure does not have a significant impact on the CO2 conversion or stability of the Fe3Ni/ZrO2 catalyst, but significantly decreases CO selectivity. This suggests the need for further catalyst modifications before Fe3Ni/ZrO2 can be considered for use in large scale operations.