It looks like you somehow got too high , highly overestimated radiation power. E.g. magnetic polarization of electron in storage rings takes minutes (although it slowed many time sby relativistic factor) e.g. https://inis.iaea.org/collection/NCLCollectionStore/_Public/08/343/8343801.pdf

Thank you, the detailed calculation of radiation power here might be quite complex, but what orders of magnitude would you expect?

E.g. in NMR they also have this alignment, https://www.cis.rit.edu/htbooks/nmr/chap-3/chap-3.htm : "When a group of spins is placed in a magnetic field, each spin aligns in one of the two possible orientations"

Atom deexcitation can similarly be seen as EM radiation from electron dynamics in EM fields, and measured times are attosecond-scale, e.g. 22as delay of photoemission: Article Delay in Photoemission

Crisis and emergency alert http://youtu.be/Ng1-KJueYiU Time for the people to stand together to bypass, help us build the bypass. We have the foundation's know

I think the answer can be made using formula here https://en.wikipedia.org/wiki/Sokolov%E2%80%93Ternov_effect but considering static electrons so the formula simplifies using E=mc^2

Thanks, but I see it starts with this parallel-antiparallel alignment, again only suggesting it is extremely fast: "An electron in a magnetic field can have its spin oriented in the same ("spin up") or in the opposite ("spin down") direction with respect to the direction of the magnetic field (which is assumed to be oriented "up")"

To estimate its speed there is needed calculation of power of EM radiated energy from such precessing/rotating dipole.

Jarek Duda In quantum spin mechanics you just have superposition of parallel and anit parallel wavefunctions and the polarization time constant is decay tau tehn means just a decay tau from higher energy state to lower.

Yes I know this QM way of thinking, but electron is still a magnetic dipole, precess in magnetic field ... and rotating dipole radiates energy as EM waves.

This way already classical magnets should finally align in magnetic field, bringing a question of the smallest objects it applies to ... and it agrees e.g. experimentally with atoms in Stern-Gerlach.

I was suggested very nice article: "Phenomenological theory of the Stern-Gerlach experimen" by Sergey A. Rashkovskiy with very detailed calculations - getting ∼10−10s times for such alignment of atoms in Stern-Gerlach: https://www.preprints.org/manuscript/202210.0478/v1

Calculations are straightforward from equation (3) there for magnetic dipole in external magnetic field.

My very approximated evaluation from radiation of abundant energy suggested a few orders of magnitude fasted alignment - bringing very interesting question if they are equivalent, how does energy balance looks above (?)

Anyway, this is another confirmation that classical magnetic dipoles in external magnetic field have tendency to align in parallel or anti-parallel way.

This "classical measurement" is deterministic and time-reversible: if recreating reversed EM, in theory one could reverse the process ...

What is nonintuive here is that such EM radiation carrying energy difference here seems different than in "optical photon", might be delocalized (?).

The big question is the minimal size to be able to apply this "classical measurement" - minimal size of such magnet: a million atoms? Thousand atoms? Single atoms? Electron?

Experimentally in Stern-Gerlach they observe the same conclusion, such alignment is also well known for electrons (e.g. https://en.wikipedia.org/wiki/Sokolov%E2%80%93Ternov_effect ), for which they observe both Larmor precession, but also much more complex acrobatics in EM field: spin echo ( https://en.wikipedia.org/wiki/Electron_paramagnetic_resonance#Pulsed_electron_paramagnetic_resonance )

So where is the classcial-quantum boundary here?

I don't see it, nor need for "unitary evolution - hocus pocus/measurement/abracadabra - unitary evolution" ...