Yambo Community Forum

Dear all，

when I calculated GW-BSE-SHG, I found several places that talk about setting the direction of the electric field. In GW step, LongDrXp; in reduced symmetries, Efield1; in collision, LongDrXs; and in nloptics, Field1_Dir.

According to my understanding, LongDrXp and LongDrXs should be consistent, representing the external field perturbation direction. For 2D materials with not very strong anisotropy, 100, 010 and 110 can all be used, and it is best to use 110. For 3D materials, it is best to use 111. Efield1 and Filed1_Dir should be consistent, indicating the input directions of the different components of the SHG to be calculated. For example, if you want to calculate SHG in the yxx direction, set them to 100. If I have misunderstood something, please do help me correct my mistake. Thank you all.

Furthermore, I would like to ask if LongDrXs and LongDrXp refer to the same parameter? When I calculate BSE and generate the input file according to the instructions, some instructions result in parameters with the suffix "Xs", while others result in "Xp". Can we just choose one of these and it won't affect the calculation result?

Best,
sunxl

Dear Sunxl,

LongDrXp and LongDrXs have the same meaning, perturbation direction when calculating the screening matrix. The difference is that one (p) is used in plasmon-pole calculation and the other (s) for static screening. The two are exclusive, you should not have them in the same input as it is useless. In BSE you can caluclate the static screening LongDrXs, or use the screening previously calculated in a GW calculation LongDrXp.

About SHG E-field: I'm not an expert but I would say that Efield1 and Field1_Dir have to be consistent.

Best,
Daniele