<div style="direction: ltr;"><span class="q"><br>
<br>
<br>
>>>>>>>> <br>
</span></div>
<div style="direction: ltr;">would you mind reading the answer that you get before<br>asking new questions, or more exactly, the same question?</div>
<br><br>
Dear Paolo,<br>
I read your every answer very carefully.<br>
My question was not the same.<br>
I agree that Methfessel-Paxton or Gaussian broadening might change the absolute <br>
value of Lambda little bit.<br>
However, <br>
last time I wanted to say that electron-phonon matrix elements are not<br>
converged using 32x32x32 k-point grid. I know achieving convergency is very <br>
slow and painful. But I was very surprised to see the value of averaged 'lambda' in the<br>
example 's out put directrory. How come it's so close to the experimental value<br>
even for very low el-ph broadening (0.01 Ryd).<br>
<br>
Please check it once.<br>
<br>
Electron-phonon coupling constant, lambda<br>
<br>
Broadening 0.0100 lambda 0.3845 dos_el 1.8818<br>
Broadening 0.0200 lambda 0.3744 dos_el 2.2498<br>
Broadening 0.0300 lambda 0.3447 dos_el 2.3960<br>
Broadening 0.0400 lambda 0.3446 dos_el 2.5079<br>
Broadening 0.0500 lambda 0.3515 dos_el 2.5896<br>
Broadening 0.0600 lambda 0.3555 dos_el 2.6381<br>
Broadening 0.0700 lambda 0.3551 dos_el 2.6616<br>
Broadening 0.0800 lambda 0.3516 dos_el 2.6709<br>
Broadening 0.0900 lambda 0.3463 dos_el 2.6737<br>
Broadening 0.1000 lambda 0.3405 dos_el 2.6743<br>
<br>
<br>
When individual matrix elements are not well converged (Please see my last mail)<br>
then how come lambda value is so good even for small broadening and <br>
small nq value (4 4 4)???? Many things confuse me lot ----- <br>
<br>
Which dos_el or Fermi energy or double delta integral are acceptable???<br>
<br>
Sometime with increasing k-point grid we did not get better convergence ---<br>
it seems the value changes very slowly with increasing K-point.<br>
We can't take arbitrary large k-point grid (millions of K-points) because it demands large <br>
disk space and very long computational time.<br>
<br>
<br>
Do you think the following numbers are converged???????<br>
<br>
. Gaussian Broadening: 0.010 Ry, ngauss= 0<br>
DOS = 1.881758 states/spin/Ry/Unit Cell at Ef= 8.327154 eV<br>
lambda( 1)= 0.0253 gamma= 0.92 GHz<br>
lambda( 2)= 0.0291 gamma= 1.05 GHz<br>
lambda( 3)= 0.0403 gamma= 6.35 GHz<br>
Gaussian Broadening: 0.020 Ry, ngauss= 0<br>
DOS = 2.249756 states/spin/Ry/Unit Cell at Ef= 8.324326 eV<br>
lambda( 1)= 0.0699 gamma= 3.02 GHz<br>
lambda( 2)= 0.0781 gamma= 3.37 GHz<br>
lambda( 3)= 0.1272 gamma= 24.01 GHz<br>
Gaussian Broadening: 0.030 Ry, ngauss= 0<br>
DOS = 2.396042 states/spin/Ry/Unit Cell at Ef= 8.311302 eV<br>
lambda( 1)= 0.0799 gamma= 3.67 GHz<br>
lambda( 2)= 0.0856 gamma= 3.93 GHz<br>
lambda( 3)= 0.1515 gamma= 30.44 GHz<br>
Gaussian Broadening: 0.040 Ry, ngauss= 0<br>
DOS = 2.507879 states/spin/Ry/Unit Cell at Ef= 8.299961 eV<br>
lambda( 1)= 0.0851 gamma= 4.10 GHz<br>
lambda( 2)= 0.0885 gamma= 4.26 GHz<br>
lambda( 3)= 0.1599 gamma= 33.63 GHz<br>
Gaussian Broadening: 0.050 Ry, ngauss= 0<br>
DOS = 2.589584 states/spin/Ry/Unit Cell at Ef= 8.291558 eV<br>
lambda( 1)= 0.0881 gamma= 4.38 GHz<br>
lambda( 2)= 0.0901 gamma= 4.48 GHz<br>
lambda( 3)= 0.1645 gamma= 35.73 GHz<br>
Gaussian Broadening: 0.060 Ry, ngauss= 0<br>
DOS = 2.638140 states/spin/Ry/Unit Cell at Ef= 8.285378 eV<br>
lambda( 1)= 0.0887 gamma= 4.49 GHz<br>
lambda( 2)= 0.0900 gamma= 4.56 GHz<br>
lambda( 3)= 0.1673 gamma= 37.02 GHz<br>
Gaussian Broadening: 0.070 Ry, ngauss= 0<br>
DOS = 2.661607 states/spin/Ry/Unit Cell at Ef= 8.280404 eV<br>
lambda( 1)= 0.0876 gamma= 4.47 GHz<br>
lambda( 2)= 0.0883 gamma= 4.51 GHz<br>
lambda( 3)= 0.1695 gamma= 37.82 GHz<br>
Gaussian Broadening: 0.080 Ry, ngauss= 0<br>
DOS = 2.670887 states/spin/Ry/Unit Cell at Ef= 8.275903 eV<br>
lambda( 1)= 0.0856 gamma= 4.39 GHz<br>
lambda( 2)= 0.0859 gamma= 4.40 GHz<br>
lambda( 3)= 0.1717 gamma= 38.47 GHz<br>
Gaussian Broadening: 0.090 Ry, ngauss= 0<br>
DOS = 2.673746 states/spin/Ry/Unit Cell at Ef= 8.271433 eV<br>
lambda( 1)= 0.0834 gamma= 4.28 GHz<br>
lambda( 2)= 0.0834 gamma= 4.28 GHz<br>
lambda( 3)= 0.1744 gamma= 39.10 GHz<br>
Gaussian Broadening: 0.100 Ry, ngauss= 0<br>
DOS = 2.674314 states/spin/Ry/Unit Cell at Ef= 8.266772 eV<br>
lambda( 1)= 0.0813 gamma= 4.17 GHz<br>
lambda( 2)= 0.0811 gamma= 4.16 GHz<br>
lambda( 3)= 0.1773 gamma= 39.76 GHz<br>
<br>
and so on ..........................<br>
<br>
It keeps on incresaing forever even with a very large K-point grid.<br>
<br>
Then how come averaged 'lamda' value is so closed to the experimental value <br>
even with small Gaussian broadennig and small K-point grids (like 16 16 16) ???????<br>
<br>
Is it accidental?????? <br>
<br>
Should we take large value of nq like nq1=32, nq2=32, nq3=32 <br>
like large value of nk for better results?????????<br>
<br>
Sometime even in total energy calculatiion we may get accidental convergence.<br>
In MIT lecture notes, it's written that <br>
<br>
<font face="Times" size="3"><span style="font-size: 16px; font-family: Times;">You do need to be careful though. It is possible to get "false" or "accidental"<br>
</span></font><font face="Times" size="3"><span style="font-size: 16px; font-family: Times;">convergence as well. That is, your energy at a 2x2x2 k-grid may be the same as<br>
</span></font><font face="Times" size="3"><span style="font-size: 16px; font-family: Times;">the energy at a 8x8x8 k-grid, but the energy at a 4x4x4 might be very different<br>
</span></font><font face="Times" size="3"><span style="font-size: 16px; font-family: Times;">from both of these. In this case, you aren't really converged at a 2x2x2 k-grid.</span></font><font face="Times" size="3"><span style="font-size: 16px; font-family: Times;">
<br>
</span></font><br>
<br>
Is it possible to calculate EPC for arbitrary q -point like <br>
0.13579 0.3474 0.83765 ???????????<br>
<br>
Looking forward to your valuable suggestions.<br>
<br>
<br>
With best regards,<br>
Amit <br>
<br>
P.S. Dear Nicola, Thank you very much for your useful reference. <br>