[Wannier] 答复: questions about quantum anomalous Hall conductivity

[Menghao Wu]

Dear Dr.Mostofi ,
Thank you very much for your kind help!
I have got wannier90.x and postw90.x in 2.0.0 version. But even with the same input files,
for version 1.2, *mmn, *eig files will be automatically generated after vasp run so I can continue the next step for wannier90.x;
but for version 2.0.0, after vasp run finishes, no files like *mmn, *eig are generated; so if I continue the wannier90.x step, .wout will show:
Error: Problem opening input file wannier90.mmn

I have no idea in which step can .mmn, .eig, .chk files  be generated in 2.0.0 version ?


 Best regards

Wu
MIT Department of NSE



________________________________
发件人: Wannier [wannier-bounces at quantum-espresso.org] 代表 Mostofi, Arash [a.mostofi at imperial.ac.uk]
发送时间: 2014年11月3日 16:17
收件人: wannier at quantum-espresso.org
主题: Re: [Wannier] questions about quantum anomalous Hall conductivity

Dear Wu,

I can answer some of your queries:

1. I believe the difference should be related to whether you are referring the conductance (units of S) or conductivity (units of S/m).

2. Try setting fermi_energy_step in the input file. The default value is 0.01 eV.

3. I’m not familiar with the VASP interface so I won’t speculate on why it might not be writing those files. But in case you had not realised, with wannier90 v2.0.0 there are now two executables that you can compile: one a serial executable (wannier90.x, obtained via “make wannier”) for first performing the spread minimisation, calculating the Wannier functions, and a few other things (i.e. pretty much everything that wannier90-1.2 does); the other a parallel executable (postw90.x, obtained via “make post”) that calculates several further properties (DoS, Berry curvature etc - see the user guide for full details.)

And finally, please don’t forget to sign off with your affiliation!

Best wishes,

Arash

―
Dr Arash Mostofi ― www.mostofigroup.org<http://www.mostofigroup.org>
Imperial College London
Director, Thomas Young Centre @Imperial
Assistant Director, CDT in Theory & Simulation of Materials
Warden, Wilkinson & Weeks Hall




On 3 Nov 2014, at 20:06, Menghao Wu <wmh1987 at mit.edu<mailto:wmh1987 at mit.edu>> wrote:

Dear Wannier developers & experts,
I am trying to calculate quantum anomalous Hall conductivity by using VASP+Wannier90.x+postw90.x, may I ask some questions:

1. Why the unit of ahc Anomalous Hall conductivity is S/cm?   von Klitzing constant RK = h/e2 = 25812.807557Ω<http://en.wikipedia.org/wiki/Ohm_(unit)>, I think the unit of  [\sigma = \frac{I_\text{channel}}{V_\text{Hall}} = \nu \; \frac{e^2}{h},]   should be S or Ω<http://en.wikipedia.org/wiki/Ohm_(unit)>-1 in all previous report I read. Why this conductivity is S over a length?

2.  As I set in .win file:

fermi_energy_min = -2.685
fermi_energy_max = -2.485
berry = true
berry_task = ahc
berry_kmesh = 25 25 1
In the oupt  .ahc. file I get:

   -2.685000    -0.009966    -0.000345   -69.735021
   -2.675000    -0.005905    -0.002186   -89.967445
   -2.665000    -0.002600    -0.001902  -266.247099
   -2.655000     0.002155    -0.003784  -266.146173
......

   -2.495000     0.011176     0.013206    51.615581
   -2.485000     0.014227     0.010831    32.024074

But the grid of fermi energy is constantly 0.01eV no matter how I tune the fermi_energy min and max. I cannot find any tag in manual on making it denser. But I need a much denser plot to show the plateau of AHC.  Is there any tag setup that can make it denser？


3. Seems only wannier90-2.0 has postw90.x dealing with berry curvature, but as I first run vasp, it finishes without generating *eig,*mmn files, and if I then run wannier90, .werr comes out asking for *eig, *mmn files so I cannot continue.
   But for wannier90-1.2, the first step by vasp will generate *eig, *mmn files and everything goes smoothly.
   So the problem is, will it be ok I first run vasp+ wannier90.x in --1.2 version, and the use postw90.x in --2.0 version?
  Or is there anyway to solve the problems in the first step that vaspwannier90--2.0.0 cannot generate *eig,*mmn files?

4. It was reported that Wannier90 can be used to calculate the edge Green’s function of the semi-infinite lattice using the recursive method:
   http://pubs.acs.org/doi/full/10.1021/nl401147u

   But I cannot find such function in the manual. Is there any tag setup that can calculate edge state of semi-infinite systems?

Thank you very much and your help is highly appreciated!
Best regards
Wu









_______________________________________________
Wannier mailing list
Wannier at quantum-espresso.org<mailto:Wannier at quantum-espresso.org>
http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20141103/af499a6b/attachment.html>