Abstract
The rapid growth of electronic waste (e-waste) and the increasing demand for advanced materials have spurred innovative approaches to sustainable material production, particularly in the development of high-entropy alloys (HEAs). HEAs, composed of five or more principal elements in near-equal proportions, exhibit exceptional mechanical strength, corrosion resistance, and thermal stability, making them highly desirable for aerospace, automotive, and energy applications. This study explores the potential of utilising e-waste as a secondary resource for HEA production, addressing both environmental and material supply challenges. E-waste, rich in valuable metals such as gold, copper, and rare earth elements, presents a viable feedstock for HEA synthesis, offering a sustainable alternative to traditional mining.
The research systematically reviews current e-waste recycling techniques, including mechanical, hydrometallurgical, and biotechnological methods, and their integration into HEA production. Advanced recovery processes, such as selective leaching and bioleaching, are evaluated alongside alloying techniques like powder metallurgy and arc melting. The study further adopts the Machine Learning (ML) technique by employing an Artificial Neural Network to optimise the composition of HEA obtained from e-waste and ranking them based on CALPHAD-predicted properties such as density, hardness and modulus of elasticity. The key finding of the ML optimisation of the composition of HEA obtained from e-waste shows that ANN is suitable for the optimisation of the compositions of HEA sourced from e-waste and the HEA with composition Cu4.2Sn1.6Pb0.35Ni54.1Fe3.5W1.1Cr21.5Mo13.2 specified as design A had the highest optimisation score and ranked first amongst the other possible compositions.
Other findings of the study reveal that integrating e-waste into HEA production can significantly reduce the environmental footprint by minimising the need for primary mining
7
and lowering energy consumption. However, achieving scalability and overcoming technical limitations requires further research into advanced extraction technologies, life cycle analyses, and optimisation of recycling and alloying processes. The study also emphasises the importance of computational approaches, such as CALPHAD and machine learning, in predicting alloy properties and optimising compositions for industrial applications.
This research underscores the potential of e-waste as a sustainable feedstock for HEA production, contributing to a circular economy and reducing reliance on finite natural resources. By advancing recycling technologies and refining synthesis processes, the study provides a pathway for the industrial-scale production of high-performance HEAs from recycled materials, offering both environmental and economic benefits. Future work should focus on experimental validation, increased data for ML analysis, economic feasibility analysis, and the development of scalable, eco-friendly recycling methods to fully realize the potential of e-waste-derived HEAs.