incorrect interaction

Dear Steve, dear LAMMPS users,

I have run several MD simulations using the same settings, with no
problem at all. However, in one case I found a strange error: at the
very beginning of the simulation, two oxygen atoms (which should
be kept separated by a Buckingham potential plus Coulomb repulsion) get
in contact with one another, thus
making the calculation to fail.

I have checked the structure and it looks fine. If somebody wants to
have a look at my simulation, all
the relevant files are here:
http://dl.dropbox.com/u/4235054/s01_hyd2.tgz

Thank you for the help,

Otello

LA RICERCA C’È E SI VEDE:
5 per mille all'Università di Bologna - C.F.: 80007010376
http://www.unibo.it/Vademecum5permille.htm

Questa informativa è inserita in automatico dal sistema al fine esclusivo della realizzazione dei fini istituzionali dell’ente.

Otello:
Sounds like the infamous Exponential-6 catastrophe where the Buckingham class of potentials give negative infinity energies for small distances, thereby causing nuclear fusion...
There used to be a simple fix for this (inverting the curve at small distances), but I will have to dig up some old fortran code.

Mike

Dear Mike,

Sounds like the infamous Exponential-6 catastrophe where the Buckingham class of potentials give negative infinity energies for small distances, thereby causing nuclear fusion...

indeed this is what happens. However, the couple of particles that give
rise to this behavior (whose indexes in the LAMMPS input file are 6998
and 13335) are 1.37 Ang apart at the beginning of the simulation. At
this distance the Buckingham potential is well into the repulsive region.
It looks like the atom 13335 is behaving as a positive particle...

Thank you for the help,
Otello

There used to be a simple fix for this (inverting the curve at small distances), but I will have to dig up some old fortran code.

Mike

From: Otello Maria Roscioni<[email protected]...>
To: [email protected]
Cc:
Sent: Tuesday, June 21, 2011 5:22 PM
Subject: [lammps-users] incorrect interaction

Dear Steve, dear LAMMPS users,

I have run several MD simulations using the same settings, with no
problem at all. However, in one case I found a strange error: at the
very beginning of the simulation, two oxygen atoms (which should
be kept separated by a Buckingham potential plus Coulomb repulsion) get
in contact with one another, thus
making the calculation to fail.

I have checked the structure and it looks fine. If somebody wants to
have a look at my simulation, all
the relevant files are here:
http://dl.dropbox.com/u/4235054/s01_hyd2.tgz

Thank you for the help,

Otello

LA RICERCA C’È E SI VEDE:
5 per mille all'Università di Bologna - C.F.: 80007010376
http://www.unibo.it/Vademecum5permille.htm

Questa informativa è inserita in automatico dal sistema al fine esclusivo della realizzazione dei fini istituzionali dell’ente.

------------------------------------------------------------------------------
EditLive Enterprise is the world's most technically advanced content
authoring tool. Experience the power of Track Changes, Inline Image
Editing and ensure content is compliant with Accessibility Checking.
http://p.sf.net/sfu/ephox-dev2dev
_______________________________________________
lammps-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/lammps-users

LA RICERCA C’È E SI VEDE:
5 per mille all'Università di Bologna - C.F.: 80007010376
http://www.unibo.it/Vademecum5permille.htm

Questa informativa è inserita in automatico dal sistema al fine esclusivo della realizzazione dei fini istituzionali dell’ente.

Otello:
Even though it is supposed to be in the repulsive region, there is a problem with the Buckingham potential itself at small distances (about 0.3xr0 if I remember correctly).
Simply plot the function with Mathematica and you will see that rather than being repulsive at short distances, the potential becomes infinitely attractive, as
the dispersive 1/r6 term overwhelms the exponential (Pauli-repulsion) term.

Mike

Dear Mike,

I am aware of this problem. What I do not understand is why this problem
appears with one of the many structures I have generated, and not on all
of them. I have used the same procedure and scripts, so there should not
be random errors due to manual editing.
However, I am happy to find a solution for this very simulation.

Thanks!
Otello

I am writing to conclude this topic.
I have solved the problem in this way by carrying out firstly a short
annealing at 300K in the NVT ensemble with a modified Buckingham term
(C=0.0, so that the potential is purely repulsive). Then I have
restarted the simulation using the correct Buckingham potential, and
this time no nuclear fusion has been observed :wink:

Thank you again for the useful advice.
Best wishes,
Otello