Mapping the Evolution of Surface Strain in PtCo Core-Shell Catalysts By 4D-STEM

Abstract

With the recent shift in attention from light-duty to heavy-duty vehicles for fuel cell research, the efficiency and durability requirements for catalyst materials have significantly increased [1]. As a result, it is more important than ever to understand the effects of degradation on catalyst nanoparticle surface structure and properties. Surface strain, for example, plays an important role in dictating the activity of catalyst nanoparticles [2], and understanding its evolution over the lifetime of a cell is therefore important. State-of-the-art PtCo core-shell catalyst nanoparticles contain complex strain gradients since, in addition to the surface strain that can occur for even single-metal nanoparticles [3], the lattice constant of the ordered PtCo core is smaller than that of the Pt shell [4]. These strain gradients may evolve over the lifetime of the cell due to cycling-induced metal dissolution, leading to compositional and shell thickness changes. Since these particles are generally <10 nm in diameter and have Pt shells only a few monolayers thick, a technique that can directly reveal their surface strain with high spatial resolution is needed to enable a better understanding of the relationship between surface strain and the performance and durability of heavy-duty fuel cells.High-resolution strain mapping can be performed by conventional high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) techniques, for example through geometric phase analysis [5] or direct tracking of atomic column positions [6,7]. These methods require the material to be tilted to a low-order zone axis for data acquisition, a very time-consuming process for randomly oriented nanoparticles that significantly limits the number of particles, and hence the statistically relevant information, that can be included in a given study [6-8]. Four-dimensional (4D)-STEM provides an alternative approach for mapping strain at high resolution by tracking diffraction disk shifts in patterns recorded in a two-dimensional STEM scan [9]. Recently, 4D-STEM was shown to offer advantages over more conventional methods for nanoparticle strain mapping, such as reduced noise and/or artifacts [10]. While initial demonstrations of off-axis 4D-STEM strain measurements have been shown [11], strain analysis based on 4D-STEM typically still requires particles to be oriented on a zone axis, which again limits the number of particles measurable for a given study. A strain mapping technique that incorporates the strengths of 4D-STEM while allowing off-axis measurements to be made would therefore significantly increase the throughput of the technique, enabling statistically relevant information about particle surface strain as a function of size, composition, or cycling history to be obtained.In this talk, we will demonstrate advancements in 4D-STEM strain mapping that allow strain to be mapped in off-axis particles, therefore enabling hundreds of catalyst nanoparticles to be measured with unit-cell spatial resolution in a single instrument session. This allows mean strain profiles to be generated for the particles as a function of distance from the surface and correlate this with particle size, composition, and cycling history. By significantly increasing the throughput of high-resolution strain mapping, this technique paves the way to a better understanding of the role strain plays in catalyst performance and degradation, accelerating development of advanced heavy-duty fuel cell catalyst materials [12].