I am currently doing a Ph.D. in the field of simulation of hydrogen storage materials. I have to simulate the process where hydrogen get absorbed into a mixture and form a metal hydride through a chemical reaction. At the molecular level, we use Lammps to understand the diffusion process. I have to use this result at the macro level. Could anyone suggest some open source softwares that are popular in the market for this kind of work?
I think that a possible solution almost free (depending on your university partnerships) is Simulink. You can perform multiscale modelling by simply using your LAMMPS results as an external function for example, and then modelling your system either with standard s. libraries or with more specific libraries (e.g. Simscape).
When you say multiscale models, you have to specify what kind of modelling. If you start with molecular level, the you may use atomistic MD, that follows Newtons laws. In the mesoscale, you still use MD, but with coarse-grained particles. On a macro level, you may use some continuum models, where particles are not present. Choice of software depends on what method you use.
Thank you for your answers.
Luigi Candido you mentioned using my Lammps result as an external function and modeling with standard libraries. It is a very good solution but at the macro level, should I use libraries from python or some other language? Unfortunately, matlab is not free for me. On the other hand, should I use the lammps result as a file and write a script to include it in software like elmer or moose? These are some free software for fem simulation. Which is a better solution in your opinion?
Kolattukudy P. Santo I am trying to do a phase-field simulation of the diffusion phenomenon. More details are given in the description part of the question. If you still need any specific information please write another comment here. At the atomistic level I am using lammps and at macro level I could use something like elmer. My question however lies in the scale bridging part of the simulation. What is the common practice? Do people write a script to take the molecular data, employ the scale bridging strategy (e.g. homogenization) within the script and then add the result as an input boundary condition in the macro simulation? I would be happy to have your opinion on this.
Dear Arnab,
Simulink support also personalized functions, concerning the FEM analysis you can add your results by user-defined S-functions.
Luigi Candido Thank you for your answer. I have never used Simulink. I will have a look.
Arnab Majumdar : I am not familiar with phase-field model. However , if you can explain a little more about your approach, we may discuss.
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India