Stability and Migration of Silicalite-1 Zeolite-Encapsulated Pt Nanoclusters Using Artificial Neural Network Potential

The stability and migration behavior of metal nanoclusters are critical factors in subnanometer-scale heterogeneous catalysis, directly impacting catalytic performance and selectivity. Encapsulation of metal nanoclusters within porous frameworks, such as zeolites, presents a promising strategy to enhance the stability and control migration dynamics. A comprehensive understanding of the interplay among nanocluster size, zeolite framework properties, and their influence on stability and mobility is essential for the rational design of efficient zeolite-encapsulated catalysts. This study introduces a global neural network (G-NN) potential specifically trained for platinum (Pt) nanoclusters encapsulated in silicalite-1 zeolite, validated through density functional theory (DFT) calculations. The validation encompassed structural optimization, adsorption energy analysis for Ptn (n = 1–6) nanoclusters, and migration energy barrier assessments for Pt1 and Pt2 clusters. Using the G-NN potential, adsorption sites within silicalite-1 channels were identified, and migration pathways between straight and sinusoidal channels were explored. The results revealed size-dependent interaction strengths between Ptn nanoclusters and the silicalite-1 framework, with larger clusters exhibiting stronger confinement. Migration barrier calculations and molecular dynamics simulations further demonstrated that larger Ptn nanoclusters (n ≥ 6) are effectively immobilized within the zeolite framework over nanosecond time scales. These findings provide critical insights into the design of advanced zeolite-encapsulated metal catalysts, paving the way for improved catalytic performance in various applications.