Visualizing Interfacial Processes and Surface Instabilities in Nanoparticle Systems Through In Situ TEM and Artificial Intelligence

Interfacial processes control not only crystal growth but also crystal dissolution. For example, during classic nucleation and growth, nuclei are forming and dissolving in solution and only when a seed crystal grows above the critical size, can a larger crystal evolve. The structural evolution of a crystal surface may be strongly influenced by the ambient environment. In small nanoparticles (&lt;5 nm) in the presence of a surface stimuli, such as gas or liquid ambient, the thermodynamic driving force to minimize surface energy is continuously perturbed due to the adsorption and desorption associated with molecular exchange processes. This leads to an evolving interfacial energy landscape with the system constantly “chasing” an elusive energy minimum. The nanoparticle surface is far from equilibrium, but local energy minima give rise to a variety of short-lived metastable states. There is considerable interest in exploring the system evolution as it transitions between stable and less stable surface configurations. Not only is this of fundamental interest to materials science, but it may impact applications in fields where surface reactivity is important such as catalysis and corrosion. The recent development of direct electron detectors with high sensitivity and fast readout rates together with advances in AI based denoisers [1] now makes it possible to perform atomic resolution exploration of nanoparticle surface processes with time resolutions ultimately limited by detector readout speeds (0.0001 – 0.001 s is now common). We are investigating the surface dynamics of Pt nanoparticles with temporal resolutions of ~0.01 s under different ambient conditions. Typical kinetic pathways are observed to follow elementary surface transformation steps including atom column activity (e.g., adatom migration, adatom fluxionality, surface layer growth, surface layer fluxionality, surface layer removal), shearing (e.g., surface shear, subsurface shear creating or annihilating stacking faults). Since there is a tight connection between the surface and subsurface structure in small particles, instabilities in the surface structure can trigger instabilities in the entire nanoparticle crystal. The relationship between these different metastable states and kinetic pathway will be discussed in the presentation.<br/><br/><i><b>References</b></i><br/>[1] Sheth, D.Y., et al. <i>Unsupervised deep video denoising</i>. in <i>Proceedings of the IEEE/CVF International Conference on Computer Vision</i>. 2021, pages: 1759-1768<br/>[2] We gratefully acknowledge the support of the following NSF grants to ASU (OAC 1940263, 2104105, CBET 1604971, and DMR 184084 and 1920335) and NYU (HDR-1940097 and OAC-2103936). We also acknowledge the support from DOE grant BES DE-SC0004954. The authors acknowledge HPC resources available through ASU, and NYU as well as the John M. Cowley Center for High Resolution Electron Microscopy at Arizona State University.