something strange happened when using pair potential lj/class2

Hello Axel and Steve :

Thank you for your time!

When I use lj/class2 pair potential to simulate the wetting behavior of argon liquid droplet ontop of platium, some of the argon atoms diffuse into the Pt substrate. It happens no matter how to vary the pair potential parameters. Though the simulation still is ok without any warning (usually there should be some warnings such as “lost of atoms”).

While lj/cut potential works well for such system, which means Ar liquid droplets wet the substrate surface other than diffuse into it.

BTW, the fix NVT is just applied just to Ar atoms to fix the Pt atom position to avoid strain effect in the wetting course.

Looking forward to your suggestions !

Best regards

Hello Axel and Steve :

Thank you for your time!

When I use lj/class2 pair potential to simulate the wetting behavior of
argon liquid droplet ontop of platium, some of the argon atoms diffuse into
the Pt substrate. It happens no matter how to vary the pair potential
parameters. Though the simulation still is ok without any warning (usually
there should be some warnings such as "lost of atoms").

While lj/cut potential works well for such system, which means Ar liquid
droplets wet the substrate surface other than diffuse into it.

BTW, the fix NVT is just applied just to Ar atoms to fix the Pt atom
position to avoid strain effect in the wetting course.

i don't see anything that indicates that lj/class2 is *not* working.
what you describe doesn't sound strange to me. different potentials
show different behavior and if they don't behave like you expect it,
it doesn't mean that there is an error in the implementation.

axel.