Non-reciprocal electromagnetic devices are fairly common but not static magnetic devices; reciprocity tells us that the coupling between two coils must be the same from one to the other as from the other to the one.
This paper claims to have found a way to break that symmetry:
It is modestly titled:
"Circumventing Magnetostatic Reciprocity: a Diode for Magnetic Fields"
I've skimmed through it and while some of the math is beyond my rusty competence but I didn't see any obvious "divide by zero" or other unphysical boo-boos.
(There's a kind of "trick" involving a moving conductor near two otherwise non-moving coils but I think that can be similarly done with a time-varying field in a static conductor.)
Anybody? Does this break some conservation law? Possible applications?
Mark L. Fergerson