Yes.
--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
cp2k+uns...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/cp2k/6b67935a-b8b5-4e7d-b2b7-75b03cfbd73e%40googlegroups.com.
Yes.
From: cp...@googlegroups.com <cp...@googlegroups.com> On Behalf Of Dev Rana
Sent: Mittwoch, 3. Juni 2020 18:19
To: cp2k <cp...@googlegroups.com>
Subject: [CP2K:13443] Is it possible to run a Andressui or Fattebert Solvation Model with Molecular Dynamics?
As the question in the subject suggests, is it possible to run a solvation model with MD?
--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp...@googlegroups.com.
Dev,
your setup of a system of six separated carbon atoms does not have a band gap and the addition of an implicit solvent won’t change that most likely. The problem should show up already for a single isolated carbon atom which you can converge either as a triplet within a spin polarized run (LSD and MULTIP 3) or with smearing, i.e. with a fractional occupation of 2/3 for the three degenerated C 2p orbitals. The latter approach works only with diagonalization (try broyden_mixing).
Matthias
To unsubscribe from this group and stop receiving emails from it, send an email to
cp2k+uns...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/cp2k/206d8b04-4fb7-4076-8ad6-d7f609f50bf6%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/206d8b04-4fb7-4076-8ad6-d7f609f50bf6%40googlegroups.com.
Hi Dev
a stress tensor is available with SCCS only in the current development version of CP2K and not in any earlier version including the release version 7.1. Note, that the stress tensor is only needed for MD ensembles with a barostat like NpT, i.e. NVT or NVE should work also with earlier versions.
HTH
Matthias
From: cp...@googlegroups.com <cp...@googlegroups.com>
On Behalf Of Dev Rana
Sent: Donnerstag, 4. Juni 2020 18:26
To: cp2k <cp...@googlegroups.com>
Subject: Re: [CP2K:13461] Is it possible to run a Andressui or Fattebert Solvation Model with Molecular Dynamics?
Hi Matthias,
Thank you for your input. I understand that as the distance from C1 (carbon #1) to C2 (carbon #2) increases, the band gap decreases until only the valence electrons remains. When the distance between atoms becomes short, the electrons converge into the conduction band which allows a bond to form with overlapping electrons. Therefore, since my electrons were so far apart, they had no band gap to overcome. I understand your rationale for using a spin polarized run or smearing, and I have tried this using a 6 membered carbon ring, instead. i.e. all 6 atoms are close. Using SCCS & MD, it tells me that I need a stress tensor enabled. So I turn it on, then it says that SCCS is not configured for using the stress tensor. Therefore, I can't really run SCCS with MD, unless I use a SCF OT method.
I guess I just don't understand how to run SCCS and MD with my system in which I'd like to see a reaction occurring between the 6 carbon atoms as they react into a ring or chain.
I apologize for the many questions, I'm a complete newbie with this stuff, with very little guidance as I am self-teaching these topics.
To unsubscribe from this group and stop receiving emails from it, send an email to
cp2k+uns...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/cp2k/461f1bb0-1c1e-4841-94c4-99731983785d%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/461f1bb0-1c1e-4841-94c4-99731983785d%40googlegroups.com.