(702b) Atomistic Simulation of the Aqueous-Phase Synthesis of Au Nanoparticles

The catalytic, electronic, and optical properties of metal nanoparticles strongly depend on their three-dimensional atomic structure. In particular, Au nanoparticles have demonstrated a wide range of properties, which can be tuned for different catalytic and biomedical applications. In this work, we have adopted a kinetic Monte Carlo (KMC) simulation approach for modeling the atomistic growth behavior of Au nanoparticles, in order to provide a direct link to the experimentally-observed growth behavior. Such a model can provide information about the impact of individual system events on the final atomic-scale structural features of the Au nanoparticles, and this can help guide the experimental synthesis process. The advantage of our approach is that, compared to traditional molecular dynamics simulations or electronic structure calculations, we can track the atomistic nanoparticle structural evolution on time scales that approach the actual experiments. This has allowed us to perform several different comparisons against experimental benchmarks, and it has helped transition our KMC simulations from a hypothetical toy model into a more experimentally-relevant test-bed. In addition, while previous analytical models have been used for predicting nanoparticle growth, the KMC approach can track atomistic-scale details and account for neighbor-neighbor interactions, structural defects, and structural heterogeneity. These aspects can be particularly influential during nanoparticle growth, leading to a wide variety of interesting nanoparticle geometries.

Here, we report the effects of synthesis temperature and initial precursor concentration on the Au nanoparticle growth behavior, and we find that our model can be trained (via automated feedback) to adequately reproduce the experimental growth curves at the same conditions. The fitting procedure results in reasonable parameter values, including activation energy barriers that are consistent with related experimental measurements. In addition, since our KMC simulations preserve the atomistic details of our growing Au nanoparticle, this allows for detailed structural analyses. Although our nanoparticles are roughly spherical, the maximum/minimum dimensions deviate from the average by approximately 12.5%, which is consistent with the corresponding experiments. Also, our surface texture analysis highlights the changes in the surface structure as a function of time. While the nanoparticles show similar surface structures throughout the growth process, there can be some significant differences during the initial growth at different synthesis conditions. To the best of our knowledge, this is one of the first attempts at modeling the atomistic growth behavior of realistic metal nanoparticles in solution. In the future, we intend to perform additional benchmarking of more atomistic growth features, including an extension of our model to capture anisotropic nanoparticle growth.