I want to calculate the viscosity of CaO-SiO2-MgO-Al2O3 Quaternary slag system for which i have Buckingham potential files for each species. I have seen the example of how to run a code for single species viscosity calculation. Can I have an example of how to calculate viscosity in lammps for multi-component system with different potential files?
I don’t think that the method is different. Why would it be? But if you want help, please at least tell us which method you use for single species viscosity calculation.
Thank you for your reply. I just wanted to know that should i be using a hybrid on potentials for each interaction(I have potentials for O-O, Mg-O interactions, etc) to calculate the viscosity for the multi-component system or is there a need for me to get a single potential for the entire system ?
With the hybrid style, you can mix several potentials within the same simulation:
Yes, understood. Thank you so much for helping me out with this! Appreciate your time and insights to this problem.
Using pair_style hybrid is always the second best choice for this. It is best suited for cases where you have two components like a metal and a machining tool or a graphene sheet and some molecules.
If you combine multiple different pair styles the challenge is always to determine the mixed interactions.
I am aware of previous studies of viscosity for molten salt mixtures from the group of Ed Maginn. You should study those papers carefully before embarking on your study. The biggest challenge will be the slow relaxation dynamics of slag, which I imagine to be quite sluggish. That will directly impact the computational cost of your simulations i.e. how many days/weeks/months you will need to run for. I recommend starting with the lowest-viscosity pure component, and if you survive that experience, proceed to binaries, and then the full mixture.
Correct. I would proceed with a pure melt and subsequently move to binary interaction and full mixture. Thank you for your insights!
Thank you for mentioning the previous studies and I am able to realise the time complexity of the problem. I would proceed as per your suggestion…with a pure melt and subsequently to the mixture.