A couple of graphics and links for those interested.from Jun...

A couple of graphics and links for those interested.

from Jun 2022: Materials | Free Full-Text | Recovery and Separation of Dysprosium from Waste Neodymium Magnets through Cyphos IL 104 Extraction | HTML (mdpi.com)
This is the determined composition of real-world EV motor magnets sent for recycling and obtained by the researchers:

Note that Nd/Dy = 20.73/3.55 = 5.8. A far cry from those who cry "30:1", but then these were EV motor magnets not overall usage.
-----------------------------------------
Second article from July 2022 (paywall for most, I suspect)
Significant enhancement in the coercivity and thermal stability of bulk hot-deformed Nd-Fe-B magnets by intergranular addition of Nd-Dy/Tb-Cu-Ga alloys - ScienceDirect
This work looked at using a low-melting gallium alloy (Ga melts at body temp as I recall) to enhance the subsequent Dy diffusion.

What you see here is an excerpted figure that shows current work (stars) as well as a literature search of recent Chinese efforts to lower Dy usage by more effective "GBD" by multiple methods; as you can see, some compare well with current grades of sintered and HD magnets (dotted lines) which span 3-8% and 2-6%, respectively. But in general, the experimental grades below 2% Dy have significantly lower coercivity H, one of the two most important properties of magnets that determines BH(max).


NOTE: GBD is an imprecise term. It was initially used to describe a vapor phase method where the DyTb was applied after the magnet was sintered. It is only sometimes used for solid state two-powder methods.