The demand of the lithium ion battery (LIB) is increasing exponentially, as it is the state-of-the-art solu tion to achieve electrification of the mobility sector and to ensure the stability of the electricity grid. The anode and cathode of these batteries consist of graphite and a lithium transition metal oxide (LTMO), respectively, which have both been designated as critical raw materials and face serious supply risks. At present, only a small amount of cathode material can be recovered through chemical separation processes, which are expensive and energy-intensive and produce a lot of waste. Direct physical re cycling of these materials is a far more efficient approach, but no scalable direct recycling routes are currently available. Sink-float separation using dense media had been successfully performed, indi cating that density is a suitable differentiating property. However, dense media are associated with serious drawbacks due to their high viscosity, toxicity and cost, making them unsuitable for large-scale application. Therefore, the objective of this research is to evaluate whether magnetic density separa tion (MDS), which applies a magnetic liquid subjected to a magnetic field to create an artificial density gradient within the liquid, can be applied in practice for the separation between anode and cathode ma terials from spent LIBs. This is achieved through magnetic field simulations, developing and executing a method for measuring the magnetic susceptibility, and Magnetic Levitation (MagLev) experiments. Following from the magnetic field simulations, an array of permanent disc magnets, stacked into one larger cylindrical magnet, was selected as the optimal magnetic configuration. Further, samples of para magnetic MnCl2 solutions were prepared at different concentrations, up to the saturation concentration. Their densities and magnetic susceptibilities had to be determined in order to be able to estimate the achievable apparent densities. From the literature it became clear that the magnetic susceptibility is a property that cannot be easily measured and a very wide range of values was used by different sources. Because of this, a new method was developed to measure the magnetic susceptibility of a liquid: the magnetic pendant drop method. The deformation of the drop was studied as it was brought into prox imity of the magnets. The results from this experiment were of the same order of magnitude as the reported magnetic susceptibilities found from the literature, which were quite dispersed. Finally, MagLev experiments were performed with anode and cathode materials. The cathode mate rial sunk to the bottom for each concentration. On the other hand, MagLev of anode material was achieved in samples of high MnCl2 concentration. From their levitation heights, the magnetic suscep tibility was once again calculated, and was even closer to the value reported in one of the sources. The equilibrium height at which the anode material levitated was at approximately 6 mm above the magnet’s surface. This means that careful design considerations will need to be taken in the design of a continuous process, and possibly a different option of magnets and paramagnetic medium should be selected. Nonetheless, the fact that levitation of graphite can be achieved, offers a positive prospect for separation by MDS on lab-scale.

The demand of the lithium ion battery (LIB) is increasing exponentially, as it is the state-of-the-art solution to achieve electrification of the mobility sector and to ensure the stability of the electricity grid. The anode and cathode of these batteries consist of graphite and a lithium transition metal oxide (LTMO), respectively, which have both been designated as critical raw materials and face serious supply risks. At present, only a small amount of cathode material can be recovered through chemical separation processes, which are expensive and energy-intensive and produce a lot of waste. Direct physical recycling of these materials is a far more efficient approach, but no scalable direct recycling routes are currently available. Sink-float separation using dense media had been successfully performed, indicating that density is a suitable differentiating property. However, dense media are associated with serious drawbacks due to their high viscosity, toxicity and cost, making them unsuitable for large-scale application. Therefore, the objective of this research is to evaluate whether magnetic density separation (MDS), which applies a magnetic liquid subjected to a magnetic field to create an artificial density gradient within the liquid, can be applied in practice for the separation between anode and cathode materials from spent LIBs. This is achieved through magnetic field simulations, developing and executing a method for measuring the magnetic susceptibility, and Magnetic Levitation (MagLev) experiments.

Following from the magnetic field simulations, an array of permanent disc magnets, stacked into one larger cylindrical magnet, was selected as the optimal magnetic configuration. Further, samples of paramagnetic MnCl2 solutions were prepared at different concentrations, up to the saturation concentration. Their densities and magnetic susceptibilities had to be determined in order to be able to estimate the achievable apparent densities. From the literature it became clear that the magnetic susceptibility is a property that cannot be easily measured and a very wide range of values was used by different sources. Because of this, a new method was developed to measure the magnetic susceptibility of a liquid: the magnetic pendant drop method. The deformation of the drop was studied as it was brought into proximity of the magnets. The results from this experiment were of the same order of magnitude as the reported magnetic susceptibilities found from the literature, which were quite dispersed.

Finally, MagLev experiments were performed with anode and cathode materials. The cathode material sunk to the bottom for each concentration. On the other hand, MagLev of anode material was achieved in samples of high MnCl2 concentration. From their levitation heights, the magnetic susceptibility was once again calculated, and was even closer to the value reported in one of the sources. The equilibrium height at which the anode material levitated was at approximately 6 mm above the magnet’s surface. This means that careful design considerations will need to be taken in the design of a continuous process, and possibly a different option of magnets and paramagnetic medium should be selected. Nonetheless, the fact that levitation of graphite can be achieved, offers a positive prospect for separation by MDS on lab-scale.