The rare earths are known for their diverse magnetic properties which inspires essential techniques
deeply embedded in our green deal and climate ambition. Researchers are inspired to exploring
magnetic separation techniques for their separation and recycling which are more efficient at the
mean time not energy and chemical intensive. Dry separators and high-gradient magnetic separators
are routinely used to separate micrometer-sized paramagnetic particles or even nanometer-sized
ferro/ferrimagnetic particles. However, for the separation of RE(III) mixtures – a process currently
carried out by solvent extraction – the interaction of RE(III) extraction kinetics with magnetically
driven transport processes would need to be considered.

To address the challenges, we are developing an integrated sensor platform to monitor the key
parameters in real time. This modular sensing framework is embedded in both portable and inline
demonstrators, enabling noninvasive, dynamic characterization of RE mixtures during various
processing stages. The system comprises a multi-sensor array capable of tracking essential physical
and chemical parameters that have traditionally been difficult to observe simultaneously in an
operational environment, e.g. magnetic, and interfacial properties. Furthermore, these sensor readings
will be processed through an AI-enhanced decision layer. This enables the system to predict
separation outcomes and respond to variability in material streams, facilitating a transition from
static, chemically intensive methods to intelligent, adaptive separation strategies grounded in real
world process data.