Can someone help me understand how to find the cohesive energy of a compound from its constituents?
Please see the screenshot attached below.
I am unable to understand how to get the equation for Ec for Mg-Al from the constituents Mg and Al. I understand that the coeff 0.25 and 0.75 are according to the proportion of each element in the compound but how to determine the +0.005 factor?
The literature says that these can be obtained from material properties like enthalpy of formation but the Energy of Formation for MgAl3 (as stated on materials project website) is 0.016ev/atom which is 0.064ev for 4atoms of MgAl3.
You are not making much sense here. E_c is an input parameter. Please have a closer look at the documentation for pair style meam. It has parameters for individual elements (the library file) and then for pairs or triples of elements (the parameter file). Since this appears to be a 2NN MEAM potential, it will require additional settings in the parameter files.
Dear @akohlmey
I am aware of the other parameters used in the 2NN MEAM potential file. However I am trying to create a new potential of my own and needed to understand how do i fix the Ec for my potential file.
Well, what does the paper say how they got it? That should be the major source. Or look for some other papers that determine 2NN MEAM parameters, if this one is not detailed enough.
The paper says they got it from Enthalpy of formation of GaN (as can be seen in the screenshot). However, the enthalpy of formation of GaN as reported in another research paper is 117kJ/mol which is arounf 1.21 eV. And the cohesive energy of GaN is reported to be 8-10 eV in the literature.
These values do not match the Ec value given in the MEAM file. I’ve looked through a couple of MEAM papers and everywhere they just give a formula for Ec based on the Ec of its constituent elements and an adjustment factor. I am unable to understand how do they determine this adjustment factor.
The theory for these papers say that Ec has been taken from material properties directly so doing a trial-and-error method for fitting Ec doesn’t make sense.
This is something you need to discuss with the authors of that paper.
This is becoming completely off-topic for a fourm about LAMMPS.
