I am trying to achive a high-spin state in a calculation without the control filed.
I assume the field might introduce artifacts and thus I am checking out other options.
For the sake of comparison, I do that on a geometry which has been optimized with the field (with high spin), and in which the magnetization has a distinct localization spot.
The number of unpaired electrons is 1.
So far, I've got an oscillating moment situation - as the calculation reaches convergence, the moment goes up, and then goes down. Sometimes it goes down, changes sign, goes up, repeat.
To make it go up again, I start the calculation from the last STATE.OUT and increase rgkmax a bit.
The moment yet again goes up, but then, at RMS in potential of about 0.1e-4, it goes down.
I assume that if I increase rgkmax every few SCF loops, and maybe even by a variable increment, the calculation might end up in the state I want it to end up.
The question: where in the code do I have to insert the lines to increase rgkmax?
Also, I would need to reallocate arrays that will change size due to the increase. Is there a subroutine that does it and can be called?
Is it simpler to just automate it outside elk?
Thank you.
Andrew.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I've achieved my goal of the field-free magnetization.
The answer was to reduce swidth.
I have the following questions now:
1. Can I interpret my odd-electron non-magnetized system (majority spin density ~= minority spin density) as a paramagnetic state? After all, it's not just one cell with just one electron - the cell represents an infinite continuum of cells. A half of the "unpaired" electrons can be up, and another half can be down. Am I wrong?
2. Lets say I make a series of calculations with the different values of swidth, and then interpret swidth as temperature. Do I get a Curie temperature this way?
3. Without the field, in the unmagnetized system, the up-DOS and down-DOS are basically the same. With the field, or with the self-magnetization (small swidth), the occupied eigenvalue (band) in the up-DOS shifts down in energies (slightly), while the corresponding virtual eigenvalue (empty band) in the down-DOS shifts up in energies (noticeably). With the field, the shift is reasonable: electrons with the spins pointing along the field should have energies different from the electrons with the spins pointing against the field. Why does this happen in the self-induced magnetization case? The system's own field does the shift?
Thank you.
Andrew
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Dear Users,
I am trying to achive a high-spin state in a calculation without the control filed.
I assume the field might introduce artifacts and thus I am checking out other options.
For the sake of comparison, I do that on a geometry which has been optimized with the field (with high spin), and in which the magnetization has a distinct localization spot.
The number of unpaired electrons is 1.
So far, I've got an oscillating moment situation - as the calculation reaches convergence, the moment goes up, and then goes down. Sometimes it goes down, changes sign, goes up, repeat.
To make it go up again, I start the calculation from the last STATE.OUT and increase rgkmax a bit.
The moment yet again goes up, but then, at RMS in potential of about 0.1e-4, it goes down.
I assume that if I increase rgkmax every few SCF loops, and maybe even by a variable increment, the calculation might end up in the state I want it to end up.
The question: where in the code do I have to insert the lines to increase rgkmax?
Also, I would need to reallocate arrays that will change size due to the increase. Is there a subroutine that does it and can be called?
Is it simpler to just automate it outside elk?
Thank you.
Andrew.
UPD. The described method works, although not perfectly. I will stick to manual restarts for not.
Dear Users,
I've achieved my goal of the field-free magnetization.
The answer was to reduce swidth.
I have the following questions now:
1. Can I interpret my odd-electron non-magnetized system (majority spin density ~= minority spin density) as a paramagnetic state? After all, it's not just one cell with just one electron - the cell represents an infinite continuum of cells. A half of the "unpaired" electrons can be up, and another half can be down. Am I wrong?
2. Lets say I make a series of calculations with the different values of swidth, and then interpret swidth as temperature. Do I get a Curie temperature this way?
3. Without the field, in the unmagnetized system, the up-DOS and down-DOS are basically the same. With the field, or with the self-magnetization (small swidth), the occupied eigenvalue (band) in the up-DOS shifts down in energies (slightly), while the corresponding virtual eigenvalue (empty band) in the down-DOS shifts up in energies (noticeably). With the field, the shift is reasonable: electrons with the spins pointing along the field should have energies different from the electrons with the spins pointing against the field. Why does this happen in the self-induced magnetization case? The system's own field does the shift?
Thank you.
Andrew