GW-PPA

[xjxiao]

Dear all,
I am writing to you with three questions, and I would be grateful for your response.

1. Regarding GW-PPA calculations, are these three commands equivalent: yambo -x -p p -g n -F gw_ppa.in, yambo -p p -g n -F gw_ppa.in, and yambo -p p -F gw_ppa.in? It seems they generate the same parameter list.

2. When performing convergence tests for the quasi-particle band gap in GW-PPA calculations with respect to different parameters, should the calculation order be as follows: first test the convergence of other parameters using a smaller k-point mesh, then fix these other parameters at their converged values, and subsequently investigate at what k-point mesh the band gap converges? Furthermore, what is the convergence criterion for the band gap?

3. Regarding GW-PPA calculations, in 2D materials, how should LongDrXp be set? How should it be set in 3D materials?

Thanks！
Yours，
Xiao

[Daniele Varsano]

Dear Xiao,

1) The first two command line are equivalent, while the third (yambo -p p -F gw_ppa.in) does not include the dyson equation solver runlevel so the qp equation will be not solved.

2) The strategy you indicate is correct, as the k point sampling convergence is independent of the other parameters, while the dimension of the screening and number of unoccupied states are intertwined. The convergence criterion depends on the precision you need, usually tenth of meV is a reasonable choice.

3) A good choice in 3D is an average of the Cartesian direction (1,1,1), while in 2D you can disregard the direction perpendicular to the 2D system, so it can be set to (1,1,0) if your system lies in the xy plane.

Best,

Daniele