Hi, Fan
I'm Soonsung So and Thank you for your nice GPUMD package.
I have some questions today.
When I saw your paper 'PHYSICAL REVIEW B 99, 064308 (2019)
(Homogeneous nonequilibrium molecular dynamics method for heat transport and
spectral decomposition with many-body potentials)'
1. I found that in figure6(a) and (c), there were nodes around 20THz. If they
have minus value, is it the computational errors?
2. When it comes to MFP(mean free path) in Figure6(c), it consists of two
terms, 'in' and 'out'.
I wonder when it is decomposed to 'in' and 'out' terms, I think, it should
contain cross terms such as
k(w)=MFP(w)*G(w)=(MFP(w)in*G(w)in + MFP(w)in*G(w)out + MFP(w)out*G(w)in +
MFP(w)_out*G(w)_out)
because G(w) and k(w) have each of 'in' and 'out' terms
How can I understand it?
