Model discovery for studies of surface morphological modifications based on Kuramoto-Sivashinsky dynamics

Area of Science:

• Materials Science
• Computational Physics
• Surface Science

Background:

• Continuum models, specifically Kuramoto-Sivashinsky types, successfully explain ion beam surface interactions.
• Current applications often rely on phenomenological determination of unknown model parameters.
• Atomistic models offer theoretical insights but practical applications require robust parameter extraction.

Purpose of the Study:

• To develop and investigate a numerical tool for determining model coefficients and structures from experimental data.
• To address the limitations of phenomenological approaches in analyzing ion beam surface modification.
• To provide a data-driven method for understanding complex surface dynamics.

Main Methods:

• A numerical tool inspired by machine learning and data-driven reconstruction techniques was developed.
• The tool was tested on a scaled model system using numerical simulations.
• The algorithm was applied to a system with lognormal distributed ion bombardment to assess its performance with noise.

Main Results:

• The reconstruction technique demonstrated high accuracy in recovering input parameters for noise-free simulations.
• The algorithm successfully applied to a system with lognormal distributed ion bombardment.
• The impact of beam fluctuations on the model was analyzed.

Conclusions:

• The developed numerical tool offers a powerful approach for analyzing experimental data in ion beam surface studies.
• This data-driven method enhances the accuracy of continuum model parameterization.
• The algorithm shows promise for exploring unknown systems and understanding beam fluctuation effects.