Journal
PHYSICAL CHEMISTRY CHEMICAL PHYSICS
Volume 23, Issue 35, Pages 19457-19464Publisher
ROYAL SOC CHEMISTRY
DOI: 10.1039/d1cp01656c
Keywords
-
Funding
- National Natural Science Foundation of China (NSAF) [11504153]
- Introduction and Cultivation Plan of Youth Innovation Talents for Universities of Shandong Province
- Cultivation Plan of Youth of Science and Technology talents of the Education Department of Guizhou Province [KY[2018]398]
Ask authors/readers for more resources
In this work, an enhanced reactive force field model using message passing neural networks (MPNN) was proposed to compute bond order and energies, significantly improving the investigation of potential energy in atomic simulations. The new potential model, ReaxFF-MPNN, provided in an atomic simulation environment, showed promising results and active learning machine was found to be more effective than general optimizer for model training.
Reactive force field (ReaxFF) is a powerful computational tool for exploring material properties. In this work, we proposed an enhanced reactive force field model, which uses message passing neural networks (MPNN) to compute the bond order and bond energies. MPNN are a variation of graph neural networks (GNN), which are derived from graph theory. In MPNN or GNN, molecular structures are treated as a graph and atoms and chemical bonds are represented by nodes and edges. The edge states correspond to the bond order in ReaxFF and are updated by message functions according to the message passing algorithms. The results are very encouraging; the investigation of the potential, such as the potential energy surface, reaction energies and equation of state, are greatly improved by this simple improvement. The new potential model, called reactive force field with message passing neural networks (ReaxFF-MPNN), is provided as an interface in an atomic simulation environment (ASE) with which the original ReaxFF and ReaxFF-MPNN potential models can do MD simulations and geometry optimizations within the ASE. Furthermore, machine learning, based on an active learning algorithm and gradient optimizer, is designed to train the model. We found that the active learning machine not only saves the manual work to collect the training data but is also much more effective than the general optimizer.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available