Dear Giovanni Thanks for your patient reply. Under your hint, I found a paper which proofed the off diagonal elements had to be real under T-symmetry. So I think maybe some parameters I used need to be modified. I meant the Hamiltionian in real space. I read the following two papers http://journals.aps.org/prb/abstract/10.1103/PhysRevB.81.245108 http://www.pnas.org/content/108/46/18622 There they seemed to use off diagonal elements directly as hopping parameters. I just simply added hr_plot=true in examples in PP and Wannier90. Nonzero imaginary parts appeared in some examples like Si and diamond. Maybe they are not well convergened. As you suggest, I will post further questions on Wannier list if I have.
Best wishes Pang Rui On Sun, 12 Oct 2014 21:54:11 +0000, Giovanni Pizzi <[email protected]> wrote: > Dear Pang, > just a comment on my previous message: I was thinking to the Hamiltonian > in k space, but actually reading more carefully your message, I think you > meant the Hamiltonian in real space. > > In which system do you find imaginary components in the off-diagonal > matrix elements between WF? Did you check if your Wannier functions were > converged and real? Do you have spin-orbit in your Hamiltonian? > > In case, we can continue this discussion on the Wannier mailing-list, that > I think is more suited to this topic. > > Giovanni > > > > On 12 Oct 2014, at 21:28, Giovanni Pizzi wrote: > > While the Slater-Koster parameters are real, the Hamiltonian matrix > elements are complex. > You should study how to build a tight-binding Hamiltonian; one good point > to start is the original paper by Slater and Koster > http://journals.aps.org/pr/abstract/10.1103/PhysRev.94.1498 > or you can check many solid-state books or any other resource on > tight-binding. > > Best, > Giovanni > > -- > Giovanni Pizzi > Post-doctoral Research Scientist > EPFL STI IMX THEOS > MXC 340 (Bâtiment MXC) > Station 12 > CH-1015 Lausanne (Switzerland) > Phone: +41 21 69 31124 > > > > > > On 11 Oct 2014, at 14:35, Pang Rui wrote: > > Dear all > Wannier functions were used in many papers to construct a tight binding > model. The off diaganalization matrix elements were interpreted as the > hopping parameters. However, I found that in most of the times, these > elements were complex numbers. How can I related these complex numbers to > real numbers to parameter a TB model? > > Best wishes! > Pang Rui > -- > PostDoc > Department of Physics, South University of Science and Technology of China > Shenzhen, Guangdong, PRC, 518500 > _______________________________________________ > Pw_forum mailing list > [email protected]<mailto:[email protected]> > http://pwscf.org/mailman/listinfo/pw_forum > > _______________________________________________ > Pw_forum mailing list > [email protected]<mailto:[email protected]> > http://pwscf.org/mailman/listinfo/pw_forum -- PostDoc Department of Physics, South University of Science and Technology of China Shenzhen, Guangdong, PRC, 518500 _______________________________________________ Pw_forum mailing list [email protected] http://pwscf.org/mailman/listinfo/pw_forum
