<div dir="auto"><div dir="auto">Robert,</div><div>Lorenzo is correct in that it's reasonably expensive to do it with fd_ef.x. </div><div dir="auto"><br></div><div dir="auto">I'm attaching two files:</div><div dir="auto"><br></div><div dir="auto">1.) fd_ef.x input </div><div dir="auto">2.) An scf file for one of the polarization directions needed. </div><div dir="auto"><br></div><div dir="auto">The seven directions and their order are mentioned in the fd_ef.x input</div><div dir="auto"><br></div><div dir="auto">Depending on the symmetry of your system you can reduce the number of polarization directions you need to actually calculate.</div><div dir="auto"><br></div><div dir="auto">You will have to run the scf calculations for each polarization direction and copy from each output the converged values (the last one) for the electronic dipole: </div><div dir="auto"><br></div><div dir="auto"><font color="#2196f3">Electronic Dipole on Cartesian axes</font></div><div dir="auto"><font color="#2196f3"> 1 -3.224342350515147E-003</font></div><div dir="auto"><font color="#2196f3"> 2 3.224063842650654E-003</font></div><div dir="auto"><font color="#2196f3"> 3 0.889627066087568</font></div><div dir="auto"><br></div><div dir="auto">Into the fd_ef.x input.</div><div dir="auto"><br></div><div dir="auto">Keep in mind that the example scf input is just an example and you should test the scf input parameters (particularly the kp grid) to verify they're adequate for your system. </div><div dir="auto"><br></div><div dir="auto">Good luck! </div><div dir="auto"><br></div><div dir="auto">Andrew Supka</div><div dir="auto">PhD Student</div><div dir="auto">Central Michigan University</div><div dir="auto"><div class="gmail_extra" dir="auto"><br><div class="gmail_quote">On Feb 25, 2018 5:33 PM, "Lorenzo Paulatto" <<a href="mailto:paulatz@gmail.com" target="_blank">paulatz@gmail.com</a>> wrote:<br type="attribution"><blockquote class="m_1642896117054207054m_7115283680841619047quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="m_1642896117054207054m_7115283680841619047quoted-text"><br>
<br>
On 02/25/2018 07:58 PM, Robert Wexler wrote:<br>
> force constants. Since ph.x, to the best of my knowledge, cannot handle<br>
> vdW corrections,<br>
</div> > ... QE version 5.1.<br>
<br>
The current version ph.x can handle vdW corrections with no problems,<br>
but you'll have to upgrade, because v5.1 is very old.<br>
<br>
To get epsilon doing finite derivatives from modern theory of<br>
polarization, via fd_ef, or not by hand, the only difficulty are the<br>
units of measure, is not very efficient it would probably take a billion<br>
k-points. If you insist, and manage to do it, please send us a working<br>
example.<br>
<br>
hth<br>
<font color="#888888"><br>
<br>
--<br>
Lorenzo Paulatto - Paris<br>
______________________________<wbr>_________________<br>
Pw_forum mailing list<br>
<a href="mailto:Pw_forum@pwscf.org" target="_blank">Pw_forum@pwscf.org</a><br>
<a href="http://pwscf.org/mailman/listinfo/pw_forum" rel="noreferrer" target="_blank">http://pwscf.org/mailman/listi<wbr>nfo/pw_forum</a><br>
</font></blockquote></div><br></div></div></div>