Top Page > Browsing

Unit using in exchange coupling parameter Date: 2012/08/07 17:05 Name: Pui-Wai Ma   <leo.ma@ccfe.ac.uk> Dear Prof. Ozaki, I am trying to use OpenMX to calculate the exchange coupling parameter between iron atoms. I looked at your program, and also your paper with Myung Joon Han on 2004 in PRB. I am a bit confused with the unit of J_ij, which is cm^-1. The Heisenberg Hamiltonian that was stated in the paper is H = -Sum_{i,j} J_ij Si . Sj. May I know the unit of each symbol, and especially the meaning of Si? If I am using a Hamiltonian, say, H = -Sum_{i,j} J'_ij Mi . Mj , where Mi is the magnetic moment vector, in the unit of Bohr magneton, muB, the unit for J'_ij could be eV/(muB^2). How could I convert the J_ij into J'_ij? Thank you very much for your help. Best Regards, Pui-Wai (Leo) Ma Senior Scientist Culham Centre for Fusion Energy (CCFE) Abingdon, Oxfordshire OX14 3DB, UK tel. +44 (0)1235 466681 fax. +44 (0)1235 466435 email: leo.ma@ccfe.ac.uk

Page: [1]

Re: Unit using in exchange coupling parameter ( No.1 ) Date: 2012/08/29 15:14 Name: T. Ozaki Hi, > The Heisenberg Hamiltonian that was stated in the paper is > H = -Sum_{i,j} J_ij Si . Sj. May I know the unit of each symbol, > and especially the meaning of Si? The calculation of J_ij is based on a perturbation theory where it is assumed that the magnitude of spin moment does not vary under small rotation of spin angle. Then, S_i and S_j are just unit vectors representing the spin direction. They are neither local spin moment nor magnetic moment vector in the unit of Bohr magneton. If you want to get a normalized J_ij, you only have to divide J_ij by the absolute value of |S_i||S_j| for the collinear treatment. Regards, TO

Page: [1]