Dear ELK Developers,
I am trying to obtain core level XPS spectra of simple elements using a method mentioned by Manabu Takahash et al. in PHYSICAL REVIEW B 78, 155108 (2008). To begin with I want to reproduce the results using ELK code for iron (as the authors have reported). This requires one to create a supercell and obtain the excited state which in proposed paper has been computed by removing one electron from core state (say 3s). In case of ELK it may be achieved reducing occupancy of core state in species.in file (if I am not wrong). To maintain charge neutrality we can add the same charge (one electron) in form of excess charge (exschg) in unit cell and thus obtain a ground state for such system. I have following questions related to this approach using ELK code:
1. To use equation (1) in the mentioned paper, we need to compute <f|s|i> which is given by deperminant of equation (4) which requires evaluation of integral given by equation (5) requiring Bloch wavefunctions for ground and sxcited state. How can we otain data for Bloch wavefunctions so that it can be used to perform mentioned integration (say numerically).
2. The authors mention about core hole with spin up or spin down state. How can we decide the spin of hole created in core state in species.in file.
3. ELK older versions (elk-1.2.15) had options to calculate valance band VB-XPS spectra. This feature does not seem availble in later versions. Was there some specific reason for not keeping it or it is still available and I am not aware of the same?
Thanks in advance
Jagdish
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Dear ELK Developers,
I am trying to obtain core level XPS spectra of simple elements using a method mentioned by Manabu Takahash et al. in PHYSICAL REVIEW B 78, 155108 (2008). To begin with I want to reproduce the results using ELK code for iron (as the authors have reported). This requires one to create a supercell and obtain the excited state which in proposed paper has been computed by removing one electron from core state (say 3s). In case of ELK it may be achieved reducing occupancy of core state in species.in file (if I am not wrong). To maintain charge neutrality we can add the same charge (one electron) in form of excess charge (exschg) in unit cell and thus obtain a ground state for such system. I have following questions related to this approach using ELK code:
1. To use equation (1) in the mentioned paper, we need to compute <f|s|i> which is given by deperminant of equation (4) which requires evaluation of integral given by equation (5) requiring Bloch wavefunctions for ground and sxcited state. How can we otain data for Bloch wavefunctions so that it can be used to perform mentioned integration (say numerically).
2. The authors mention about core hole with spin up or spin down state. How can we decide the spin of hole created in core state in species.in file.
3. ELK older versions (elk-1.2.15) had options to calculate valance band VB-XPS spectra. This feature does not seem availble in later versions. Was there some specific reason for not keeping it or it is still available and I am not aware of the same?
Thanks in advance
Jagdish