Specifying sigma and epsilon values for an lj potential in real units

Hi,

I’m trying to model a system using lj potentials using real units. However, unlike lj units which have a unit value of sigma, epsilon, mass, I see no way of specifying them explicitly in the real style. The best command I found using the online documentation is pair_style and pair_coeff. Pair_style lj/cut only requires the cutoff value in angstrom, and pair_coeff only requires the sigma and epsilon values as a fraction of the overall sigma and epsilon value of the simulation. If instead I specify it as the actual sigma and epsilon value in real units, it seems to give me an error. So how do I specify this global value that’s (obviously) not unity?

Regards,
Siddharth

Hi,

I’m trying to model a system using lj potentials using real units. However, unlike lj units which have a unit value of sigma, epsilon, mass, I see no way of specifying them explicitly in the real style.

​this is nonsense. in reduced units, you don’t necessarily have values of 1.0 and 1.0 for epsilon and sigma, respectively. this would only apply to single type systems, and even then, you can choose a different reference and then your values for epsilon and sigma would be != 1.0.​

​in LAMMPS you have to enter the values for epsilon and sigma in real units in the exact same way as in reduced units.

The best command I found using the online documentation is pair_style and pair_coeff. Pair_style lj/cut only requires the cutoff value in angstrom, and pair_coeff only requires the sigma and epsilon values as a fraction of the overall sigma and epsilon value of the simulation. If instead I specify it as the actual sigma and epsilon value in real units, it seems to give me an error. So how do I specify this global value that’s (obviously) not unity?

​this makes no sense. the choice of units is defined with the units keyword and that then defines with energy or length or other unit is relevant for parameters in the input. please note, that the documentation only mentions specific units when commands are only valid for a specific choice of units. otherwise it mentions when kind of unit is used and then you can look up in the documentation of the units command how this unit is represented in the specific choice.​

​this is all logical and straightforward. you can review this in the examples in the examples folder; some of which are in lj units, others in real or metal units.

axel.​