Domain, Specialties : Neural networks
Keywords: Machin learning-Artificial intelligence-alloys-recycling
Research Unit : NIMBE / LICSEN
Summary
The production of waste electrical and electronic equipment (WEEE) is increasing and poses recycling challenges, particularly for printed circuit boards (PCBs), which contain precious metals and reusable components. The aim of the internship is to use machine learning methods to discover and characterize new environmentally-friendly soldering alloys, reducing the energy consumption and the environmental risks associated with current recycling processes.
Full description
The production of waste electrical and electronic equipment (WEEE) is rising sharply worldwide, with several million tonnes generated every year, only 17% of which being properly collected and recycled.
Printed Circuit Boards (PCBs) account for most of the value of this waste, thanks to the many metals they contain (gold, silver, tin, copper, etc.). The electronic components (capacitors, diodes, etc.) are also reusable, but are soldered to the surface of the PCBs.
In order to recover these electronic components and reuse/recycle them, thermal solder melting processes are widely used worldwide. These processes require considerable energy consumption to reach the melting temperature of the solder used. Currently, most solder is made from SAC (silver-tin-copper mix), with melting temperatures higher than 200°C. At these temperatures, there is a risk of damaging plastics and electronic components, as well as volatilizing or dispersing toxic substances into the environment (arsenic, lead, flame retardants, etc.).
The aim of this internship is to use machine learning methods to accelerate the discovery and characterization of new alloys more suitable for disassembly and recycling, as well as their properties. Indeed, it is essential to ensure that their properties are adequate to enable their deployment on an industrial scale.
The student will be required to build up a database by studying the literature, to deploy machine learning algorithms and eventually to experimentally verify the properties predicted by these algorithms.
The student will be working in an international environment and will be involved in several industrial collaborations. In this context, this project should potentially lead to significant societal benefits.
This internship is therefore an excellent opportunity for professional growth, both in terms of your knowledge and know-how.
The candidate is a student in his/her final year of Master’s degree. Strong skills in machine learning and data visualization, particularly in Python and associated libraries (scikit-learn, matplotlib…). Experience in the use of algorithms for the prediction of continuous data, such as random forests and support vector machines (SVM), is required. A background in chemistry or physics is desirable, with an appetite for experimental laboratory work. The candidate must be able to carry out bibliographical research. Autonomy, an analytical mind and an interest in environmental issues are important qualities.
Location
CEA-Saclay, (91) Essonne, France
Internship conditions
- Internship duration: 6 months
- Level of study: Bac+5
- Training: Master 2
- Continuation in PhD thesis: No
- Application deadline: 3 mars 2025
Experimental skills
Language : English
Useful methods and technics:
- Programmation (Python…),
- Machine learning (SVM, random forests…),
- Data visualisation (matplotlib…)
Computer languages and software: Python
Supervisor
Guillaume Zante
Phone: 0169089083
Email :