LAMMPS WWW Site - LAMMPS Documentation - LAMMPS Mailing List Archives
Re: [lammps-users] Can I mix coarse grained atoms with full atoms in LAMMPS?
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lammps-users] Can I mix coarse grained atoms with full atoms in LAMMPS?

From: Axel Kohlmeyer <akohlmey@...24...>
Date: Wed, 16 May 2018 11:10:52 -0400

On Tue, May 15, 2018 at 11:00 PM, #KIEU TRUNG HIEU# <KIEU0001@...36.....2775...> wrote:

Dear LAMMPS users,


I'm going to build a simulation of water behavior with a Graphene oxide-Zif8 composite. The problem is the domain of the system is too large (20x20x15 nm3) and thus I'm thinking about doing coarse-graining of water molecules. However, I want to maintain the atom structure of the GO/MOF composite.

​200x200x150 A^3 isn't really that large for a code like LAMMPS. a quick back of the envelope guesstimate would make this a system with less than 1 million atoms. LAMMPS is capable of running simulations with *billions* of atoms (and people have done it). you simply may have to look for a larger parallel machine to run your simulation.​


I would like to ask whether it is possible to have coarse-grained atom and full-atom in a LAMMPS simulations?

​yes, technically it is possible. LAMMPS doesn't care whether the simulated (point) particles represent ​a single or multiple atoms. it just applies to model you provide, computes the forces and moves the particles.
the tricky thing is to determine a suitable parameterization and figure out whether this is representing what you want to model well enough. specifically water is a tricky issue. in many cases, the interactions of individual water molecules with your substrate are crucial to be represented explicitly, e.g. to be able to have hydrogen bonds, and thus a coarse grain water representation (as good as it may be to represent bulk water), may not be suitable for your setup.

Have anyone done that before in the literature?

​IIRC, people in the group of mario ors​i have done mixed coarse-grain / all atom simulations as i recall integrating some code into the r-RESPA integrator in LAMMPS to make it more efficient.
nobody here is going to do a literature search for you, so you have to get your hands dirty and search for it yourself.




Thanks a lot.



Check out the vibrant tech community on one of the world's most
engaging tech sites,!
lammps-users mailing list

Dr. Axel Kohlmeyer  akohlmey@...12...24...
College of Science & Technology, Temple University, Philadelphia PA, USA
International Centre for Theoretical Physics, Trieste. Italy.