Dear all,

Im currently using siesta to calculate the momentum (not crystal momentum) of the each state in CNT(carbon nanotube). 

My calculation was gamma point only calculation due to some constraints of my research.

After ran some of my calculations, I have two question about siesta.

1. How can I decompose the band structure energy in output file.

In the siesta document about the siesta algorithm (JSoler-Internals-Of-Siesta.pdf (, E^KS could be decomposed in to like below.

E^KS = Ekin + Enl + Ena + Eions + DEna + DUscf + Exc

But there is no memtion about band structure energy (Ebs).

I thought that Ebs could be a compostion of the terms in the right side of E^KS equation but I couldn't find it.

How can I get Ebs from the parameter written above(Ekin, Enl ... ). Or does it need other parameter that was provided in the output file?

(also how does siesta calculate the Ekin internally. Does it calculate from the system's hamiltonian or using different operator(matrix) from hamiltonian?)

2. Why not wavefunctions in gamma point are complex? 

After calculation, I ran denchar and get real and imaginary components of the wavefunction for each states.

But whole value of imaginary part of wavefunctions were zero.

I thought that it was occur because of CNT but single atom (with 2p orbital, with large vacuum) also return zero imaginary.

I think due to 2p orbital, it need to be non zero imaginary.

Do wavefunctions in gamma point need to be real valued? If so, Why?

Thanks in advance,

Best regards,

Hyunwook Ha

SIESTA is supported by the Spanish Research Agency (AEI) and by the European 
H2020 MaX Centre of Excellence (

Responder a