<html><head></head><body><div class="yahoo-style-wrap" style="font-family:Helvetica Neue, Helvetica, Arial, sans-serif;font-size:16px;"><div dir="ltr" data-setdir="false"><div><div>Dear QE developers,</div><div><br></div><div>I am trying to understand the meaning of “orbital-resolved magnetic anisotropy”</div><div>as demonstrated in: PP/examples/ForceTheorem_example.</div><div><br></div><div>From the documentation and the projwfc.x output, it appears that QE provides</div><div>the contribution of each orbital character (e.g., dxy, dx2−y2, dz2, …) to the</div><div>band-energy difference (MAE) obtained via the force theorem.</div><div><br></div><div>Question:</div><div>Is my understanding correct that QE reports the *projected occupied-state</div><div>contribution* of a given orbital character (e.g., dxy = 0.9 meV), rather than</div><div>an explicit second-order SOC coupling term between occupied and unoccupied</div><div>orbitals (as typically reported in VASP, e.g., dxy|dz2 = 0.3 meV, dxy|dx2−y2 = 0.4 meV, …)?</div><div><br></div><div>In other words, if VASP reports:</div><div> dxy|dz2 = 0.3 meV</div><div> dxy|dx2−y2 = 0.4 meV</div><div> dxy|dyz = 0.2 meV</div><div><br></div><div>then the QE-projected value:</div><div> dxy = 0.9 meV</div><div><br></div><div>would correspond to the *sum of all pairwise contributions* involving dxy as the</div><div>occupied partner. Could you please confirm whether projwfc.x computes this</div><div>projection only from occupied states, and whether this interpretation is correct?</div><div><br></div><div>Thank you very much.</div></div><br></div></div></body></html>