[Q-e-developers] About differiential calculation

[Paolo Giannozzi]

dV/dR is by definition the limit of Delta V/Delta R for Delta R=>0. If this
does not converge to a well defined function for dR->0, there is something
wrong in your calculation

On Mon, Jun 15, 2015 at 4:56 PM, Jiqiang Li <jqli14 at fudan.edu.cn> wrote:

> Dear developers,
>         I want to ask a trivial question about total local potential Vtot
> (V_Hartree + Vxc + V_ion), i.e., the variable *vrs *in electron.f90
> (correct me if I am wrong). When we move a specific atom along a specific
> direction, does the derivative quantity dVtot/dR remain unchanged ?
>        I try 0.01~0.1 Angstrom as the amplitude of the displacements, and
> find that dVtot/dR is not independent of dR. does the differece arise from
> the numerical error or theoretically should dVtot/dR depend on dR? If the
> former is right, then how long is the optimal value of dR for this
> calculation generally?
>
> Best regards!
>
> Jiqiang Li
>
>


-- 
Paolo Giannozzi, Dept. Chemistry&Physics&Environment,
Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
hone +39-0432-558216, fax +39-0432-558222
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/developers/attachments/20150615/5b7e0854/attachment.html>