[Wannier] How to use wannier-kslice-curv to calculate chern number
杨晟祺
12307110406 at fudan.edu.cn
Mon Mar 14 13:18:32 CET 2016
I am computing the berry curvature and chern number of a 2D material.
Below is part of my wannier.win:
#berry=true
#berry_task=ahc
#berry_kmesh=25 25 1
fermi_energy=-2.423600000000
berry_curv_unit=ang2
kpath=false
#kpath_task=bands+curv
#kpath_bands_colour=spin
#kpath_num_points=1000
kslice=true
kslice_task=curv+fermi_lines
kslice_corner=0.0 0.0 0.0
kslice_b1=1.0 0.0 0.0
kslice_b2=0.0 1.0 0.0
kslice_2dkmesh=200 200
Below is part of the wannier.wpout:
------
SYSTEM
------
Lattice Vectors (Ang)
a_1 5.767388 0.000000 0.000000
a_2 -2.883694 4.994704 0.000000
a_3 0.000000 0.000000 14.349700
Unit Cell Volume: 413.36313 (Ang^3)
Reciprocal-Space Vectors (Ang^-1)
b_1 1.089434 0.628985 0.000000
b_2 0.000000 1.257969 0.000000
b_3 0.000000 0.000000 0.437862
After running postw90, I get wannier-kslice-curv.dat. I believe the three rows in this file is berry curvature on x,y,z direction respectively. According to the definition of Chern number, C=(SumCurv*Sbz)/(kmesh1*kmesh2*2*pi),where SunCurv is the sum of all berry curvature in certain row. Sbz is the area of the first BZ, kmesh1*kmesh2 is the multiplication of the two number in kslice_2dkmesh. However, I cannot get an integer. This material is an insulator so its chern number must be an integer.
I sumed all data in each row of wannier-kslice-curv.dat and below is the result.
berry =
1.0e+003 *
0.0004 0.0010 -3.9866
Chern number of this system should be -2. Does anyone know how to compute chern number with wannier90?
Thank you very much for your help!
--
Department of Physics, Fudan University
Shengqi Yang
Tel: (+86)18817873184
More information about the Wannier
mailing list