Directed Electrostatics Strategy Integrated as a Graph Neural Network Approach for Accelerated Cluster Structure
Sridatri Nandy1, K V Jovan Jose1
1Advanced Artificial Intelligence Theoretical and Computational Chemistry Laboratory, School of Chemistry, University of Hyderabad, Hyderabad, Telangana 500046, India.
Journal of Chemical Theory and Computation
|January 15, 2025
Summary
We developed a graph neural network (DESIGNN) to predict stable nanocluster structures using directed electrostatics. This method accurately identifies magnesium cluster structures and generates novel isomers, accelerating materials discovery.
Area of Science:
- Computational chemistry and materials science
- Application of artificial intelligence in predicting atomic structures
Background:
- Predicting stable nanocluster structures is computationally challenging.
- Understanding the potential energy surfaces (PESs) is crucial for identifying stable atomic cluster configurations.
Purpose of the Study:
- To introduce a novel graph neural network (GNN)-based approach, DESIGNN, for predicting stable nanocluster structures.
- To utilize directed electrostatics and molecular electrostatic potential (MESP) topography for guiding structure prediction.
Main Methods:
- Developed the DESIGNN approach, integrating graph neural networks with directed electrostatics.
- Benchmarked the model on magnesium (Mg) clusters (n < 150).
- Predicted MESP topography minima and analyzed parent growth potential (GP) for cluster evolution.
Main Results:
- DESIGNN accurately predicts MESP topography minima for Mg clusters (n < 70), aligning with full calculations.
- Generated ground-state structures for Mg clusters (n = 4-32) that match literature findings.
- Successfully generated novel symmetric isomers and large stable Mg nanoclusters (n up to 260).
Conclusions:
- The DESIGNN approach effectively accelerates the search and prediction of stable nanocluster structures.
- This method shows significant promise for exploring large metal clusters guided by MESP topography.
- DESIGNN facilitates the discovery of new materials with specific symmetries and properties.
Related Concept Videos
02:27Predicting Molecular Geometry
34.0K
VSEPR Theory for Determination of Electron Pair Geometries
34.0K
02:52VSEPR Theory and the Basic Shapes
67.4K
Overview of VSEPR Theory
67.4K
01:25Electric Field Lines
7.4K
The three-dimensional representation of the electric field of a positive point charge requires tracing the electric field vectors, whose lengths decrease as the square of their distance from the charge and which point away from the charge at each point. This vector field is no doubt challenging to visualize. The visualization of electric fields becomes quickly intractable as the number of charges increases.
The solution to this problem is to use electric field lines, which are not vectors but...
The solution to this problem is to use electric field lines, which are not vectors but...
7.4K
02:37VSEPR Theory
9.0K
Valence shell electron-pair repulsion theory (VSEPR theory) enables us to predict the molecular structure around a central atom from an examination of the number of bonds and lone electron pairs in its Lewis structure. The VSEPR model assumes that electron pairs in the valence shell of a central atom will adopt an arrangement that minimizes repulsions between these electron pairs by maximizing the distance between them. The electrons in the valence shell of a central atom form either bonding...
9.0K