Structure search for molecular clusters. Towards a rational understanding of their size-dependent properties

Inorganica Chimica Acta, Год журнала: 2024, Номер unknown, С. 122376 - 122376

Опубликована: Сен. 1, 2024

Язык: Английский

---

Directed Electrostatics Strategy Integrated as a Graph Neural Network Approach for Accelerated Cluster Structure Prediction

Journal of Chemical Theory and Computation, Год журнала: 2025, Номер unknown

Опубликована: Янв. 15, 2025

We present a directed electrostatics strategy integrated as graph neural network (DESIGNN) approach for predicting stable nanocluster structures on their potential energy surfaces (PESs). The DESIGNN is (GNN)-based model building of large atomic clusters with specific sizes and point-group symmetry. This assists in the structure metal by molecular electrostatic (MESP) topography minima structural evolution paths. benchmarked prototype Mgn n < 150. predicted MESP clusters, 70, fairly agrees whole-molecule calculations. Moreover, ground-state (n = 4-32) generated through corroborate well global minimum reported literature. Furthermore, this could generate novel symmetric isomers medium to size regime, 150, constraining symmetry parent clusters. growth (GP) cluster gives measure its accommodate more atoms characterize DESIGNN-guided path. GP can also be interpreted cooperative interaction relative cluster. Along highest paths, further employed nanoclusters 228, 236, 257, 260. Therefore, holds great promise accelerating search prediction guided topography.

Язык: Английский

---

5-step algorithm to accelerate the prediction of [Au₂₅(SR)₁₉]^z clusters (z = 1−, 0, 1+)

RSC Advances, Год журнала: 2024, Номер 14(27), С. 19483 - 19489

Опубликована: Янв. 1, 2024

Prediction of the structure thiolated gold clusters is time demanding, and new strategies are needed to expedite this process.

Язык: Английский

---