Assuming your electronic structure calculations are done correctly, still your question doesn't specify whether you're calculating Seebeck coefficient for chemical potentials above the conduction band minimum (n-doping) or below the valence band maximum (p-doping). Within each doping range, also, Seebeck can show sign reveal both in terms of energy and temperature. It is because Mg2Si has a multi-valley band structure, with Fermi pockets, some of which show electron-like behavior and some others hole-like.

Mohammad Saeed Bahramy Thank you very much for your answer sir, but I still have a question.. so, in which parameter or how can I specify more details about the doping range, so it should show the n-doping.

Actually, I do not specify it for calculation I've been doing.

In file prefix.intrans, I only edit the (efcut) energy range from 0.15 (before) to 0.35 (after I change it) and the temperature (from 800 to 900l). So which parameter I should change so it shows the n-type doping?

Thank you

Dear Monalisa Nur Ramadhan

The first and most important parameter to set in your input file is the value of Fermi Level. If you want to consider n-doping set this value to the energy corresponding to the bottom of the conduction band. On the other hand if you want to study the effect of p-doping fix it at the top of valence band. The parameters, deltae and ecut, then allow you to vary the chemical potential around the Fermi level at regular steps (delte) within your desired range (ecut). Alternatively, if you use the latest version of Boltztrap you can directly specify the doping concentrations in the input as well (still you need to specify Fermi level). For example to compare the Seebeck resulting from doping 1x10^20 cm^-3 electrons with that of hope doping of 2x10^20 cm^-3 you can add the following commands to the input:

2

1E20 -2E20

Above, The first line is the number of fixed dopings and the second line defines the corresponding values. For more information, please refer to the manual of Boltztrap (attached here)