Non-cryogenic separation of Helium_3 from Helium_4?

IIRC, He_3 is usually separated from much less rare He_4 by cryogenic cooling of gas mix to 'liquid', at which point the mix divides to two phases, one with each isotope...

IIRC, Hydrogen and Deuterium, as gas mix, may be progressively separated at near-ambient conditions by differential membrane diffusion.
Is Helium simply 'too slippery' to practicably resolve thus ?

A quick google scholar search suggests that it might be possible to do it using diffusion.
That said, I would be surprised if it would be better than doing using cryogenically. The latter method is very easy and cheap (you just need a 4K system of some type, it can even be a liquid helium dewar) and has the added advantage that other impurities are frozen out (you usually want your He3 to be very, very clean).
It might be worth it if done industrially, but even then using a modern high throughput He liquefier to liquefy the He4 and siphon off the He3 is probably more efficient. That said, the latter is just a guess.

I know a guy working on this via diffusion. "Engineering nightmare" doesn't begin to cover it. Last I talked to him, his target was to enrich the helium so you'd still cryoseparate it but have a better yield at that phase.