| 6:37:37 PM PST - Fri, Feb 26th 2016 |
|
Hi Edoapra,
I used your suggested script. The initial few iterations suggest that the DIIS will converge in a while.
I have a few follow up questions.
1. I did some comparison of pc-3 and def2-svp for a single atom Silicon and found the pc-3 value to be 98 mHa lower than the def2-svp value, which seems quite high for any reliable calculation.
2. The reason I had chosen pc-3 in the first place was that I wanted to check systematic convergence of ground state energy with respect basis set refinement. So I started with pc-1 and went up to pc-4 and fixed my basis-set whenever the difference in ground state energy between two consecutive basis set fell below 1mHa . I was particular about this systematic convergence aspect of basis set because I wanted a reference value to compare against my own DFT code. For smaller silicon clusters (18, 31 and 95 atoms), my code was in good agreement with the pc-3 values (within 1mHa). So I was hoping of getting the pc-3 value for the current 280 atoms silicon cluster and compare it against my code. Is there a way to make pc-3 work for the 280 atoms case by tinkering with certain parameters?
I would really appreciate your inputs on it.
Thanks again,
Bikash
|