Sometimes ago, I found a paper in Journal of Magnetism and Magnetic Materials 299:64~69. They used Wien2k to obtain the susceptibility of alpha-iron. The procedure they use is very simple. First they make simulations under different magnetic fields at 0 K. This gives them energy and magnetic moment (m). Then, they put the energy differences , say E (i.e., energy with magnetic field minus energy without the magnetic field) into the formula:
M=m*tanh(E/KT)(K is the Boltzmann constant and T the temperature)
to obtain M at different temperature. Thus they can get susceptibility at different temperatures by dividing magnetisation by field strength at each temperature.
I would like to repeat the work using EXCITING with the parameters "bfinite" and "bfieldc". What I found is that I need a much higher field strength to cause a similar change in energy. For example, the magnetic moment is ~13 Ampere.meter^2/mole. To make it reach 1.6 A.m^2/mole at 1043 K, the paper I mentioned above use ~1 Tesla, and I need to use ~100 Tesla at the same temperature. As a consequence, suscpetibility is ~ 100 times smaller.
I remember you once cautioned the use of "bfinite". So my question is that maybe at the present moment EXCITING cannot be used to study material properties under a magnetic field?
With best regards!
zhangyg
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The DFT ground state of alpha-Fe is ferromagnetic, so applying a magnetic field won't change the magnetisation very much, since the system is saturated. Therefore the magnetic field energy is just the moment times the magnetic field. Why do you have to perform calculations at different field strengths?
Best wishes,
Kay.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I did some reference search these days and try to understand the problem better. I think you are right. What the paper that I mentioned previously did was, from my current understanding, just a Fixed Spin Moment study. For the moment, I still do not undertsand how they can relate such a calculation to the behavior of the system at high temperatures.
With best regards!
zhangyg
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I did some calculations of hcp iron and bcc iron using EXCITING. I exerted external magnetic field "bfieldc" (e.g., 0 0 1.0e-3) with bfinite =.true. One thing I found is that the resulted magnetic moment is negative for both iron structures. This puzzles me a little. Does this mean that bcc and hcp iron are diamagnetic? Or I misunderstood something here?
Looking forward to your kind answer.
with best regards!
zhangyg
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
EXCITING does not follow the usual convention of using a minus sign in the coupling of B(r) and m(r). This is why the moment is opposite to the applied field.
Best wishes,
Kay.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Dear Kay:
Sometimes ago, I found a paper in Journal of Magnetism and Magnetic Materials 299:64~69. They used Wien2k to obtain the susceptibility of alpha-iron. The procedure they use is very simple. First they make simulations under different magnetic fields at 0 K. This gives them energy and magnetic moment (m). Then, they put the energy differences , say E (i.e., energy with magnetic field minus energy without the magnetic field) into the formula:
M=m*tanh(E/KT)(K is the Boltzmann constant and T the temperature)
to obtain M at different temperature. Thus they can get susceptibility at different temperatures by dividing magnetisation by field strength at each temperature.
I would like to repeat the work using EXCITING with the parameters "bfinite" and "bfieldc". What I found is that I need a much higher field strength to cause a similar change in energy. For example, the magnetic moment is ~13 Ampere.meter^2/mole. To make it reach 1.6 A.m^2/mole at 1043 K, the paper I mentioned above use ~1 Tesla, and I need to use ~100 Tesla at the same temperature. As a consequence, suscpetibility is ~ 100 times smaller.
I remember you once cautioned the use of "bfinite". So my question is that maybe at the present moment EXCITING cannot be used to study material properties under a magnetic field?
With best regards!
zhangyg
Dear zhangyg,
Could you explain a bit more:
The DFT ground state of alpha-Fe is ferromagnetic, so applying a magnetic field won't change the magnetisation very much, since the system is saturated. Therefore the magnetic field energy is just the moment times the magnetic field. Why do you have to perform calculations at different field strengths?
Best wishes,
Kay.
Dear Kay:
I did some reference search these days and try to understand the problem better. I think you are right. What the paper that I mentioned previously did was, from my current understanding, just a Fixed Spin Moment study. For the moment, I still do not undertsand how they can relate such a calculation to the behavior of the system at high temperatures.
With best regards!
zhangyg
Dear Kay:
I did some calculations of hcp iron and bcc iron using EXCITING. I exerted external magnetic field "bfieldc" (e.g., 0 0 1.0e-3) with bfinite =.true. One thing I found is that the resulted magnetic moment is negative for both iron structures. This puzzles me a little. Does this mean that bcc and hcp iron are diamagnetic? Or I misunderstood something here?
Looking forward to your kind answer.
with best regards!
zhangyg
Dear zhangyg,
EXCITING does not follow the usual convention of using a minus sign in the coupling of B(r) and m(r). This is why the moment is opposite to the applied field.
Best wishes,
Kay.