« first day (556 days earlier)      last day (835 days later) » 

8:13 AM
0
Q: Geometrical Optimization not converging for a double perovskite quantum dot, using Gaussian

Parmeet Singh EP 066I am trying to optimize the geometry of a Cs4CuSb2Cl12 Stoichiometric Quantum dot and I have checked the input multiple times, but the opt+freq calculation is not converging. The quantum dot has about 80 atoms and is charge neutral furthermore I don't see any dangling bonds as such. Here is my in...

 
 
14 hours later…
9:43 PM
I'm not familiar with Gaussian at all, but it doesn't look like you've given any lattice vectors or specified periodic boundary conditions. Is this correct? — Phil Hasnip 12 hours ago
+1 but showing the output file that shows the lack of convergence would help. Also providing the coordinates will help people to do testing on their own in order to figure out how to solve your problem. — Nike Dattani 8 hours ago
Even if the output file is too long, it would help just to clarify what specifically isn't converging. Does the SCF fail or does an opt step or does are the forces just not getting smaller and it times out? — Tyberius ♦ 7 hours ago
I'm not familiar with Gaussian at all, but it doesn't look like you've given any lattice vectors or specified periodic boundary conditions. Is this correct? — Phil Hasnip 12 hours ago
So its the SCF that is failing and it looks like it is after quite some time. Usually, this would mean the optimization wound up at a bad structure for which the SCF was too difficult to converge. Was the energy throughout that SCF cycle significantly different from the energies in the previous optimization steps (search SCF Done throughout the file)? How close was the geometry to convergence (search Item, under which there should be a list of convergence criteria and how close they were). — Tyberius ♦ 7 hours ago
It seems like scf did come really close at two places one of which was (SCF Done: E(RCAM-B3LYP) = -23237.5922015 A.U. after 19 cycles Convg = 0.3136D-08 -V/T = 2.025) but under the Item column only RMS converged, and that too only once. — Parmeet Singh EP 066 6 hours ago
It probably would be easiest to include the whole output file. I'm guessing its too big to put in the post itself, but you can add it this Github repo or include a pastebin link. — Tyberius ♦ 6 hours ago
@Tyberius I uploded the log file which has all the information using google drive — Parmeet Singh EP 066 4 hours ago
 
9:56 PM
@ParmeetSinghEP066 Your input geometry seems to be the unit cell for the bulk crystal. I could be mistaken on this, but I would think the nanodot would potentially have a very different structure and that you would maybe have a tough time converging the geom/SCF for a small slab.
 

« first day (556 days earlier)      last day (835 days later) »