# How to make input script file for nanoparticle using LAMMPS?

Hope you're all doing well. I am a student of Doctorate in Beijing Institute of Technology. We are working on a project "DNA Regulated Nanozyme". and we intend to build an interaction between DNA and Nanozyme and for this we need to do the MD simulation of our nanoparticle "Iron Manganese Silicone Nanoparticle" (IMSN). It is actually a silicon nanoparticle. And I am using LAMMPS for its simulation. But, I am new to LAMMPS so I can't understand how to build its input script file? Like what are the basic things I need to have for LAMMPS input script? Like I tried Avogadro to build its structure but couldn't able to make its structure. is there any way that how can I build my nanoparticle structure in Avogadro and then convert it into LAMMPS format for input script file? I shall be very thankful to you for this help Regards and Thanks Misbah Naseem

I don't think you can build a nanoparticle using Avogadro, I personally use ASE which is a python package that can output a lot of different formats (xyz, cif, vasp, ...)

You can directly output LAMMPS format using ASE or you can use Atomsk to convert whatever format to LAMMPS.

By doing that you will obtain a "lammps-data" file, which needs to be read by a LAMMPS input using the read_data command.

If you want to use LAMMPS for your project, it is a really powerful program but there is no way around it: you need to read the documentation.

If you are a new computational chemist I would advise you to use ASE, it is not only useful for nanoparticles, I'm using it nearly every day.

• thank you so much for your response but as I am a beginner so I don't know how to make input file for LAMMPS because I don't know my nanoparticle structure only know its formula which isn't enough I think. Can you please suggest me some tutorial regarding nanoparticle simulation by LAMMPS? Also How to use this ASE? Can you please give me website of this one? Thank you Sep 15 at 13:01
• Not knowing the structure is another problem, you should read the literature about it. The structure should depend on the chemical nature and some energetical considerations (Wulff construction), but no one can guess it without research. Concerning ASE there is a link in my answer but if you are not familiar with Python it might be complicated. Good luck. Sep 15 at 14:55
• Thank you so much Sep 16 at 18:09

You can either use python or MATLAB to generate the coordinates, or even use the built in tools in LAMMPS. To do this in LAMMPS, you only need to follow three steps.

1. Use the lattice command to define the lattice type (diamond for Silicon).

2. Use the region command to define a region with the desired dimensions. (Note that you may use boolean operations to obtain the exact shape).

3. Populate that region using atoms which can be achieved using the Create_atoms command.

However, an energy minimization step followed by relaxation in the NPT and/or NVE ensembles should be recommended to ensure that the structure is stable at the particular conditions.

• thank you so much I will surely try this and let you know if I find any difficulty Sep 23 at 11:56

# Atomsk

You can use this software to play with the structure and then, generate de coordinate file for you program.

Atomsk is a free, Open Source command-line program dedicated to the creation, manipulation, and conversion of data files for atomic-scale simulations. It supports the file formats used by many programs, including:

• Visualization and analysis softwares: Atomeye, ddplot, OVITO, VESTA, XCrySDen;
• Ab initio calculation softwares: ABINIT, Quantum Espresso, VASP, SIESTA;
• Classical force-field simulation softwares: DL_POLY, GULP, IMD, LAMMPS, XMD;
• TEM image simulation softwares: Dr Probe, JEMS, QSTEM.

The reference paper is: Atomsk: A tool for manipulating and converting atomic data files" Pierre Hirel, Comput. Phys. Comm. 197 (2015) 212-219 doi:10.1016/j.cpc.2015.07.012.

The program documentation is here.

There are some tutorials than can help/guide you here.

• thank you so much I will surely try this Sep 21 at 21:57