Fig. 1: Workflow for constructing and validating the deep learning potential for CF3SO2F decomposition. | Communications Chemistry

Fig. 1: Workflow for constructing and validating the deep learning potential for CF3SO2F decomposition.

From: Probing the thermal decomposition mechanism of CF3SO2F by deep learning molecular dynamics

Fig. 1

The workflow comprises four key steps: (1) Configuration sampling from AIMD trajectories, (2) Training of the DLP, (3) Validation of the DLP against DFT benchmarks, and (4) Large-scale molecular dynamics simulations of CF3SO2F decomposition across a range of temperatures (T1 to Tn), with supercell size adjusted to model different system pressures.

Back to article page