Thesis presented December 15, 2023

Abstract:
Dy-containing sintered NdFeB permanent magnets are key elements in the transition to low-carbon transport and the mass deployment of electric vehicles. However, Dy has been identified as a critical element by the European Union, due to the Chinese monopoly, which poses a supply risk. To address this supply vulnerability, a Grain Boundary Diffusion Process (GBDP) has been developed. This method seeks to minimize the amount of Dy used in these magnets by limiting its presence to the grain boundaries and their peripheries. This results in a core-shell microstructure, i.e., a grain whose core is low in Dy and whose periphery is high in Dy. This heterogeneous distribution of Dy is interesting because it maximizes the gain in magnetic performance provided by the heavy rare earth, while limiting the use of this critical material.
The aim of this thesis is to characterize the microstructure of magnets at different scales, in order to better understand the mechanism behind the formation of this microstructure, and thus be able to propose levers for optimizing the grain boundary diffusion process.
As part of this thesis, NdFeB magnets were manufactured by powder metallurgy on the LMCM laboratory pilot line at CEA/LITEN. Some of these samples were coated with a binary Dy-Co alloy and then annealed to allow Dy diffusion into the magnet (GBDP). Several electron microscopy techniques (SEM, EDX, EBSD and TEM) were used to carry out a detailed study of the microstructure of these magnets before and after this diffusion treatment. In particular, quantitative EDS/WDS measurements were used to determine the composition and thickness of Dy-rich shells within the grains as a function of distance from the surface. The magnetic performances of the magnets (remanence and coercivity) are also determined before and after GBDP from experimental demagnetization curves. The various mechanisms proposed in the literature to explain the formation of the core-shell microstructure involve the formation of a liquid phase at the grain boundaries, from which the shells can develop either by volume diffusion, solidification, or liquid film migration. We review the input data on which these mechanisms are based, comparing them with our results. We show that these hypotheses are not completely satisfactory for our observations. We propose improvements to the diffusion model using finite element simulations. We conclude this work by proposing new elements to be considered in order to better understand the mechanisms behind the formation of the core-shell microstructure.

Keywords:
NdFeB, permanent magnet, Grain boundary diffusion, Microstructure, Electron microscopy, Coercivity​