Hi Axel,
There is a paper by Tersoff commenting on its applicability to amorphous carbon, titled "Empirical interatomic potential for carbon, with application to amorphous carbon":
-Tersoff, J. "Empirical interatomic potential for carbon, with applications to amorphous carbon." Physical Review Letters 61.25 (1988): 2879.
It is not perfect however, in its ability to represent the atom interactions within amorphous carbon, and one effect is its underestimation of the sp3 to sp2 ratio.
Nevertheless, there are several papers documenting the use of Tersoff potential for amorphous carbon:
-Lee, Seung-Hyeob, et al. "Meta-stable sites in amorphous carbon lattice generated by rapid quenching of liquid diamond." 2002 International Conference on Computational Nanoscience and Nanotechnology
-Sha, Z. D., et al. "A modified Tersoff potential for pure and hydrogenated diamond-like carbon." Computational Materials Science 67 (2013): 146-150.
-Li, Longqiu, et al. "The effect of empirical potential functions on modeling of amorphous carbon using molecular dynamics method." Applied Surface
Some of these papers show the RDF graph for amorphous carbon generated with Tersoff, but it looks more similar to the RDF graph I've generated with REBO than with Tersoff.
There are also past discussions (on this forum) on generating amorphous carbon with the Tersoff potential, specifically on the liquid-quench protocol to generate the sample.
Rgds,
Damien
Hi,
To those who have experience with using Tersoff to generate amorphous carbon, I would like to ask for some pointers in generating a sample with a valid RDF.
I have followed the guidelines to generate amorphous carbon:
- creating carbon atoms in a random configuration or an unstable lattice configuration such as simple cubic
- Heating to above 5000K (have tried 5000K, 8000K and 14000K)
- Quenching to 300K and equilibrating (both under NVT) until thermodynamic properties are stable under NVE
- density of about 3g/cm^3
I have varied the number of atoms from a thousand to 15,000 as well as used several versions of lammps including the latest (as well as using both windows and ubuntu versions), but could not get the two characteristic peaks in the RDF. Neither does the graph tends towards 1 at the end.
Using the same conditions but with the REBO potential however, I managed to get a valid RDF graph though.
do you have any reference saying, that tersoff *should* reproduce
amorphous carbon?
axel.
