coarse k-mesh needs to be strictly positive

[mirshowkat07]

Hi
I want to calculate the carrier mobility of my system. The scf, phonon and nscf calculations finished normally. Now, when, I am running the epw.x code,
I am getting the error as given below. Kindly, suggest, how can I fix this issue with epw.x

G-vector sticks info
--------------------
sticks: dense smooth PW G-vecs: dense smooth PW
Sum 9787 4903 1405 3514205 1242867 191063

negative rho (up, down): 1.604E-02 0.000E+00

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Error in routine epw_setup (1):
coarse k-mesh needs to be strictly positive in 1st BZ
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
stopping ...


with best regards
Showkat
Department of Chemical Engineering
IIT Kanpur, India

[roxana]

Hi,

You need to run a non self-consistent calculation on a homogeneous k grid. You cannot use the K_POINTS automatic feature of pw.x, but you need to generate a positive definite grid between 0 and 1 and specify K_POINTS crystal. You can use, for example, the kmesh.pl script provided in the wannier90 distribution to
generate the homogeneous k grid (in wannier90/utility).

Best,
Roxana

[mirshowkat07]

Dear Roxana
Thank you very much. I can generate the k-point mesh using kmesh.pl script, but what about the weight.
The script does not generate the weight on the 4th column. Will the EPW code work without specifying the weight of k-points in nscf calculation.
Any help in this regard will be appreciated.

best regards
Showkat

[roxana]

Hi,

The script should provide the k-point weight if used as "kmesh.pl n1 n2 n3". Otherwise, the weight is just 1/(n1*n2*n3)

Best,
Roxana

[mirshowkat07]

Thank you for your quick reply.


best regards
Showkat