Yambo Community Forum

Dear all,
For a magnetic system, if its topmost valence band and the bottommost conduction band possess opposite spins , the exciton involved in these two bands must be spin-triplet. See https://journals.aps.org/prl/abstract/1 ... 124.166401 (Phys. Rev. Lett. 124, 166401). In this case, how do we accurate calculate the triplet exciton ? Do we need to take into account the SOC? Or set BSENGexx = 0 eV for spin polarization?

Thank You !

Dear lz Dong,
if you compute spin-flip transitions ("Delta Sz = 1"), you will get the spin flip channel of BSE.

Setting the exchange cutoff to zero in input you will get instead triplet transitions in the spin-conserving channel "Delta Sz=0"

"Delta Sz = 1" transitions are degenerate to "Delta Sz = 0" transitions for non magnetic systems.
However, they are significantly different for magnetic systems.
Indeed "Delta Sz = 1" transitions describe magnons. See for example here:
https://doi.org/10.1103/PhysRevB.103.155152

You can compute spin-flip transition by setting Please be aware that this is an experimental feature of the code, and that the convergence of magnons in the spin-flip channel is quite different from the convergence of excitons.
I advise to use the last yambo version 5.2

Best,
D.