Exsolution Modeling and Control to Improve the Catalytic Activity of Nanostructured Electrodes

Abstract

In situ exsolution for nanoscale electrode design has attracted considerable attention because of its promising activity and high stability. However, fundamental research on the mechanisms underlying particle growth remains insufficient. Herein, cation‐diffusion‐determined exsolution is presented using an analytical model based on classical nucleation and diffusion. In the designed perovskite system, the exsolution trend for particle growth is consistent with this diffusion model, which strongly depends on the initial cation concentration and reduction conditions. Based on the experimental and theoretical results, a highly Ni‐doped anode and an electrochemical switching technique are employed to promote exsolution and overcome growth limitations. The optimal cell exhibits an outstanding maximum power density of 1.7 W cm−2 at 900 °C and shows no evident degradation when operating at 800 °C for 240 h under wet H2. This study provides crucial insights into the developing and tuning of heterogeneous catalysts for energy‐conversion applications.

Chemical Stability of BaMg0.33Nb0.67-XFexO3-δ in High Temperature Methane Conversion Environments

“…The BMNF33 may have reached its saturation limit for iron, therefore unable to accommodate more Fe into B site lattice positions. As seen previously, there may be also tendency towards exsolution with perovskite materials that have slightly higher nonstoichiometric iron loading with otherwise stable structures (8)(9)(10)(11). This effect is clearly seen in Figure 1c, where there appears to be new peak formations for BMNF33, indicative of a second phase formation.…”

Local Structures of Ex-Solved Nanoparticles Identified by Machine-Learned Potentials

“…Consequently, dislocation model type I is energetically more favorable than the strain model. This result explains the lack of signs of strain during the growth of exo particles. ,− Notably, in Table , we find that the (111) interface energy is smaller than the (100) interface energy. This is why the nanoparticles were found to be more deeply embedded in the host material when they are grown on the (111) surface than on the (100) surface .…”

“…In previous studies, the growth kinetics of exo particles are determined as either strain-, reactant-, or diffusion-limited on the basis of fitting results from each respective classical growth model. However, the small differences between these fitting results have led to conflicting conclusions. ,,,− From our observations, which highlight the minimal strain surrounding exo particles, we infer that a strain-limited growth mechanism is unlikely in this scenario.…”

“…Therefore, we consider (100) and (111) surfaces and interfaces. For the interfaces, the lattice directions between metals and supports are set as identical, a fact well-established from previous experiments for Ni ex-solved nanoparticles. ,− Figure b shows the types of epitaxially grown interface structures examined in this work: one strain model and two dislocation interface models. For the strain model, we assume that the lattice parameters of the support remain consistent near the interface, while the lattice parameters of Ni adapt to match those of the support, as the previous study verfied that strain energy is predominantly stored in the Ni atoms close to the interface .…”

“…Furthermore, the intense strain at the interfaces is observed for endo particles, which are nanoparticles that remain embedded within the host material instead of being exposed on the surface (Figure a). , On the basis of these arguments, the interface structure between the metals and supports is smooth with minimal defects to ensure that the lattice mismatch is compensated by the distortion near the interface, as shown in the “strain model” in Figure b . In contrast, transmission electron microscopy (TEM) images of ex-solved nanoparticles shown in other studies reveal that the lattice parameters of ex-solved metals and support oxides deviate minimally from their original values, indicating insignificant lattice distortions. ,− Following these observations, the lattice mismatch would be compensated by the dislocations at the interfaces, instead of strain, as represented in the “dislocation models” in Figure b. In fact, Du et al observed moiré patterns at the interface of ex-solved nanoparticles, which is suggested to be an evidence supporting the dislocation interface model …”