*2.2. Phase Transformations*

Phase transformations are processes of deactivation involving the conversion of an active crystalline phase of the catalyst (or one of its components) into a different inactive one. These transformations can involve both metal-supported and metal-oxide catalysts. In the former type of catalyst, atoms from the catalyst's support will diffuse into the catalyst's surface. A reaction at the surface can then result in an inactive phase, deactivating the catalyst.

Riedel et al. was able to demonstrate that the steady states of the synthesis of hydrocarbons using iron oxides could be separated into five episodes of distinct kinetic regimes. In episode I, the adsorption of the reactants takes place on the catalyst surface and carbonization occurs. During episodes II and III, products from the RWGS reaction dominate during ongoing carbon deposition. In episode IV, the rate of FT activity increases up to the steady

state, and the maintenance of the steady state occurs in episode V. Prior to the reaction, the iron phases of the reduced catalyst are mainly α-Fe and Fe3O4, along with a small amount of Fe2O3. As the process proceeds, the Fe3O4 and Fe2O3 phases are consumed and a new oxidic iron amorphous phase is formed, which appears to be active for the RWGS reaction. Through a reaction of iron with carbon from the CO dissociation, FTS activity commences with the formation of iron carbide (Fe5C2). Upon the formation of the stable but inactive carbide (Fe3C), which is the result of Fe5C2 carburization, the catalyst begins deactivating [57–59]. Lee et al. studied the causes of the deactivation of Fe–K/γ-Al2O3 for CO2 hydrogenation to hydrocarbons, and found the causes for deactivation varied based on positioning inside the reactor. Over time, the Fe2O3 was reduced to active phase χ-Fe5C3, and then the χ-Fe5C3 was transformed to θ-FeC3, a form which is not active for CO2 hydrogenation. The primary reason for deactivation was the phase transformation at the top of the reactor. Conversely, at the bottom of the reactor, deactivation was largely the result of deposited coke generated by secondary reactions [57].

Zhang et al. reported the structure evolution of the iron catalyst during its full catalytic life cycle of CO2 to olefins (CTO), including the catalyst activation, reaction/deactivation (120 h) and regeneration. The phase transition during the CO activation was observed to follow the sequence of Fe2O3 → Fe3O4 → Fe → Fe5C2. The primary deactivation mechanism during CTO was identified as the irreversible transition of iron phases under reaction conditions. Two possible pathways of the phase transition of the iron catalyst under CTO conditions have been identified, i.e., Fe5C2 → Fe3O4 and Fe5C2 → Fe3C → Fe3O4. Moreover, carbon deposition and the agglomeration of the catalyst particle proves to have relatively minor impacts on the catalytic activity compared with phase transition during the 120 h of reaction [60].

It appears that transformation to iron oxides will destroy catalyst activity. There is some question as to whether the cementite phase itself is problematic to activity, as higher-surface-area cementite phases have been reported to perform CO2 hydrogenation quite effectively [61].
