Dear All,
I am working with an alloy with 5 elements (Fe, Al, Mn, Ni and C) and am struggling to define a potential. I found one paper with Co,Cr,Fe,Mn,Ni meam potential.
Could I use the above without Cr & Co and use Tersoff or Lennard jones in conjunction (for C & Mn)?
I am aware that meam is a many-body potential and I don’t know if I can combine LJ with this. I also know that this potential was optimized for that combination of elements in alloy itself so this would probably be an assumption (or a mistake?) Please advise.
If not meam+ lj, then is there any other way to define the potential for this alloy?
Thanks for your help.
Dear All,
I am working with an alloy with 5 elements (Fe, Al, Mn, Ni and C) and am struggling to define a potential. I found one paper with Co,Cr,Fe,Mn,Ni meam potential.
Could I use the above without Cr & Co and use Tersoff or Lennard jones in conjunction (for C & Mn)?
leaving out elements is usually not a problem. adding missing elements with a different style of pair potential istechnically possible, but in practice, for a real alloy (and not a system like a metal in contact with a diamond tool) i would expect poor results.
I am aware that meam is a many-body potential and I don’t know if I can combine LJ with this. I also know that this potential was optimized for that combination of elements in alloy itself so this would probably be an assumption (or a mistake?) Please advise.
If not meam+ lj, then is there any other way to define the potential for this alloy?
the clean solution would be to keep searching or learn how to parameterize such a potential yourself. since there are many degrees of freedom, it is going to be very tricky (unlike a 2 element alloy).
have you done the math and computed how many atoms your typical total system would have, and how many of them you would have for each element? if one of them is a rather small fraction, that it would probably not make a big different to leave that element out (after all, there would only be a few atoms of that kind in the total system, anyway, and your statistical errors for typical atom scale simulations are fairly large.
axel.