Hi,
I am using lammps to study Fe-C-Si system. I am using meam potential comprising Fe Mn Si and C available from the site
https://www.ctcms.nist.gov/potentials/system/Fe/#C-Fe-Mn-Si.
It is expected that the Fe-C-Si system will be more denser than Fe-C system after energy minimization at same NPT condition. I am not getting that trend at (I merely replaced few C atoms with Si atoms keeping other conditions same and modified the pair_coeff command).
My section of specifying potential looks as follows
For Fe-C system :
pair_style meam/c
pair_coeff * * ./ibrary.meam Fe Mn Si C ./parameter.meam Fe C
For Fe-C-Si system
pair_style meam/c
pair_coeff * * ./ibrary.meam Fe Mn Si C ./parameter.meam Fe Si C
- As per the manual, The arguments (between the file names) are elements for which info will be extracted from the library file. If my system consist only Fe Si and C then I should extract info of only Fe Si and C thereby writing :
pair_coeff * * ./ibrary.meam Fe Si C ./parameter.meam Fe Si C .
But lammps gives error : "MEAM library error 3 (…/pair_meamc.cpp:594) " and forces me to write “Fe Mn Si C” between the file names. Does This means, all elements’ info from library.meam file should be extracted irrespective of elements under study ?
- The order of elements after the second file is : atom 1 atom 2 atom 3 of the input structure format of lammps where in my case is Fe Si C. Is that correct.?
Any guidance will be much helpful.
PS : I successfully obtained the properties of Fe-C system using another meam potential from https://www.ctcms.nist.gov/potentials/system/Fe/#C-Fe . Hence, other section of input files has been previously verified.
Thanks
Sharad
Hi,
I am using lammps to study Fe-C-Si system. I am using meam potential comprising Fe Mn Si and C available from the site
https://www.ctcms.nist.gov/potentials/system/Fe/#C-Fe-Mn-Si.
[…]
- As per the manual, The arguments (between the file names) are elements for which info will be extracted from the library file. If my system consist only Fe Si and C then I should extract info of only Fe Si and C thereby writing :
pair_coeff * * ./ibrary.meam Fe Si C ./parameter.meam Fe Si C .
But lammps gives error : "MEAM library error 3 (…/pair_meamc.cpp:594) " and forces me to write “Fe Mn Si C” between the file names. Does This means, all elements’ info from library.meam file should be extracted irrespective of elements under study ?
The number and order(!) of elements extracted from the MEAM library file is entirely determined by the MEAM parameter file. This file is indexed by numbers, so it cannot reorder them and it cannot detect if you mixed them up. This second file contains additional settings for specific tuples of elements that override the generic settings from the library file. Your URL suggests that your parameter file requires: pair_coeff * * library.meam C Fe Mn Si parameter.meam …
this is irrespective of the order of atom types in your LAMMPS input.
you should consider this whole “library.meam parameter.meam” input block like one tersoff or eam parameter file.
- The order of elements after the second file is : atom 1 atom 2 atom 3 of the input structure format of lammps where in my case is Fe Si C. Is that correct.?
the number and order of Elements following the parameter file is the mapping of the extracted elements to the atom types. the atom types may be in any order and you may assign multiple atom types to the same element or skip elements extracted from the library.
please note that this is all explained in the documentation.
axel
Thank You for the explanation. Now it’s more clear to me.
I have a minor clarification.
As you said, the url https://www.ctcms.nist.gov/potentials/system/Fe/#C-Fe-Mn-Si suggest my parameter file requires :
pair_coeff * * library.meam C Fe Mn Si parameter.meam …
However, I cross checked the parameters in parameters.meam file with tabulated values in https://doi.org/10.1016/j.mtla.2019.100473 (Appendix B) with the parameter file.
Eg : Ec(Fe,Mn)= 3.002=Ec(1,2) , Ec(Fe,Si)=4.22=Ec(1,3), Ec(Fe,C)=5.285=Ec(1,4) which suggest that the order of elements are 1-Fe, 2-Mn, 3-Si and 4-C.
Hence shouldn’t it be : pair_coeff * * library.meam Fe Mn Si C parameter.meam … ?
Thank You.
Sharad
i said, that the URL suggests, but I don’t have the time to verify it.
Use whatever matches the publication or anything said on the webpage directly.
If needed, you should get the confirmation from the NIST database admin or the original uploader of the files.
Axel.
If you’re using Fe-C-Si system you only write:
pair_coeff * * library.meam Fe Si C parameter.meam
Because Mn is not in your structure file.
that is not correct!
the elements after library.meam must be those elements that have additional parameters in parameter.meam in the order in which they are referenced in marameter.meam. those are referenced with numerical indices and those numbers are inferred from the order of the elements after library.meam.
this is explained in detail in the documentation of pair style meam/c.
axel.