[ase-users] LJ parameters for atom types

Wed Nov 6 18:14:46 CET 2019

Dear Daria,

Your MM LJ input shouldn't be floats, as water has 3 atoms per molecule. I know that the eps and sig for the water Hs are 0, but you still have to tell that to LJInteractionsGeneral.

So I.e., if the water part of your atoms object is ordered like 'OHHOHHOHH...' then your MM Lj arrays should look like: eps_mm = np.array([epsilonOtip3p, 0, 0]) and likewise for sig.

Hope that helps.

Best,

Asmus
________________________________
From: Daria Tombolelli <tombolelli.daria at gmail.com>
Sent: Wednesday, November 6, 2019 3:06 PM
To: Asmus Ougaard Dohn <asod at hi.is>; ase-users at listserv.fysik.dtu.dk <ase-users at listserv.fysik.dtu.dk>
Subject: Re: [ase-users] LJ parameters for atom types

Dear Asmus Dohn,

I tried to use the class you suggested me LJInteractionsGeneral but I encountered a problem.

For now my system is a CO in water where CO is the qm part and waters are the mm part.

I wanted then to insert different LJ parameters for O of CO and O of water.

This is my input statement for LJ parameters:

interaction = LJInteractionsGeneral(sigmaqm=(sigmaOM,sigmaCM),epsilonqm=(epsilonOM,epsilonCM),epsilonmm = epsilonOtip3p,sigmamm=sigmaOtip3p,qm_molecule_size=2)

where OM is for Oxygen of CO, CM is for Carbon of CO.

The problem raise when the function combine_lj_lorenz_berthelot is called at line 332
sigma_c = np.zeros((len(sigmaqm), len(sigmamm[cc])))

since my sigmamm[cc] is a float (and it has no len()).

I solved changing this line to:
sigma_c = np.zeros((len(sigmaqm), 1))

but I am not sure this is correct.

It would be really nice if you would be able to help me.

Best regards,
Daria Tombolelli

On Wed, 23 Oct 2019 at 11:22, Asmus Ougaard Dohn <asod at hi.is<mailto:asod at hi.is>> wrote:
Hello Daria,

Not silly questions!

If you need to insert different LJ parameters for different atom types of the same element, you need to use the LJInteractionsGeneral class instead. Apologies for the lack of documentation of that one, we need to get that fixed. This calculator simply takes in epsilon/sigma arrays of length qm_molecule_size for the qm subsystem and of length mm_molecule_size of the mm subsystem and combines them using the Lorenz-Berthelot rules. So with that class you have full control of which atom(type)s get which parameters.

Second question: Yes.

Addendum: I've made some tools of automatically assigning LJ parameters that could maybe come in handy for larger systems, you can find them here: https://gitlab.com/asod/aseambertools  . It uses AmberTools for atom typing.

Best,

Asmus

________________________________
From: ase-users-bounces at listserv.fysik.dtu.dk<mailto:ase-users-bounces at listserv.fysik.dtu.dk> <ase-users-bounces at listserv.fysik.dtu.dk<mailto:ase-users-bounces at listserv.fysik.dtu.dk>> on behalf of Daria Tombolelli via ase-users <ase-users at listserv.fysik.dtu.dk<mailto:ase-users at listserv.fysik.dtu.dk>>
Sent: Wednesday, October 23, 2019 8:52 AM
To: ase-users at listserv.fysik.dtu.dk<mailto:ase-users at listserv.fysik.dtu.dk> <ase-users at listserv.fysik.dtu.dk<mailto:ase-users at listserv.fysik.dtu.dk>>
Subject: [ase-users] LJ parameters for atom types

Hi everyone,

I am trying to use the QM/MM calculator (EIQMMM) for simple calculations of i.e. CO in water box, but I don't find anywhere the information about how to insert different LJ epsilon and sigma values for different atom types, in this case the Oxygen of CO and the Oxygen of water molecules. Is this is not possible without using interfaces to other (external) MM calculators?

The second question is about the molecular_size parameter in the Embedding class: i.e I did a simple QM/MM calculation of water dimer in a water box and I left this value to 3. Is it correct?

Sorry for the maybe silly questions and thank you for your help.

Daria

--
Dr. Daria Tombolelli
Technische Universität Berlin
Fakultät II
Institut für Chemie
Sekretariat PC 14
Straße des 17. Juni 135
10623 Berlin

E-Mail: daria.tombolelli(at)tu-berlin.de<http://tu-berlin.de/>

[https://static.licdn.com/scds/common/u/img/webpromo/btn_profile_bluetxt_80x15.png]<http://it.linkedin.com/in/dariatombolelli1987>
Researchgate<https://www.researchgate.net/profile/Daria_Tombolelli>

Raum: SE-RH 108

wiss. Mitarbeiterin in der Arbeitsgruppe Prof. Dr. M. Mroginski
http://www.modeling.tu-berlin.de<http://www.modeling.tu-berlin.de/menue/modellierung_biomolekularer_systeme/>

--
Dr. Daria Tombolelli
Technische Universität Berlin
Fakultät II
Institut für Chemie
Sekretariat PC 14
Straße des 17. Juni 135
10623 Berlin

E-Mail: daria.tombolelli(at)tu-berlin.de<http://tu-berlin.de/>

[https://static.licdn.com/scds/common/u/img/webpromo/btn_profile_bluetxt_80x15.png]<http://it.linkedin.com/in/dariatombolelli1987>
Researchgate<https://www.researchgate.net/profile/Daria_Tombolelli>

Raum: SE-RH 108

wiss. Mitarbeiterin in der Arbeitsgruppe Prof. Dr. M. Mroginski
http://www.modeling.tu-berlin.de<http://www.modeling.tu-berlin.de/menue/modellierung_biomolekularer_systeme/>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/ase-users/attachments/20191106/9b086eeb/attachment-0001.html>