Machine Learning for the Prediction of Aluminum-Based Quasicrystals

• Grant number: 25/26609-9
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: March 01, 2026
• End date: February 28, 2027
• Field of knowledge: Engineering - Materials and Metallurgical Engineering - Physical Metallurgy
• Principal Investigator: Witor Wolf
• Grantee: Daniel de Almeida Durán
• Host Institution: Escola de Engenharia de São Carlos (EESC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

Quasicrystalline forming aluminum alloys are of great interest to materials engineering as they present unique properties, especially related to surfaces. The main ones presented by those materials can be listed: low coefficient of friction, high hardness and Young modulus, making them promising materials for tribological protection. But thanks both to the metastability of the phases and the sensibility of the systems to changes in the chemical composition, only specific compositions of alloys lead to the formation of those phases. Through machine learning models, it is intended to observe correlations between physicochemical properties, e.g., mixing enthalpy and difference in atomic radius, of such materials to understand which phases are formed, specially the quasicrystalline ones. In this project, still, the validation will happen by synthesizing one alloy, chosen according to the final model. Its characterization will be done by diffraction techniques and electron microscopy to study the phases formed. (AU)