Hello All,
I would like to know if it is possible in LAMMPS to delete entire molecules as they move across a fixed boundary. Currently I am simulating the explosive boiling of water on a gold substrate. I constrain the water using fix RATTLE with bond and angle constraints which results in the creation of 528 frozen angles (which makes sense because I have 528 water molecules). I have periodic boundary conditions in the x and y direction while I have a fixed boundary condition in the z-direction. As the water molecules increase in temperature they move in the positive z-direction for a time until they cross the positive z-boundary of my simulation box. At that point I get the error: Shake atoms 235 567 987 missing on proc 0 at step 35699. I believe I get this error because parts of molecules are being deleted resulting in molecule fragments. Fix evaporate provides a work around to this because of its molecule flag but I am curious if there is a simpler way to get rid entire water molecules as they cross a fixed boundary (in my case the positive z-boundary). I am using the 5-Sep-17, 64-bit Windows version of LAMMPS. Any insight will be greatly appreciated.
Sincerely,
Malcolm Porterfield
Hello All,
I would like to know if it is possible in LAMMPS to delete entire
molecules as they move across a fixed boundary. Currently I am simulating
the explosive boiling of water on a gold substrate. I constrain the water
using fix RATTLE with bond and angle constraints which results in the
creation of 528 frozen angles (which makes sense because I have 528 water
molecules). I have periodic boundary conditions in the x and y direction
while I have a fixed boundary condition in the z-direction. As the water
molecules increase in temperature they move in the positive z-direction for
a time until they cross the positive z-boundary of my simulation box. At
that point I get the error: Shake atoms 235 567 987 missing on proc 0 at
step 35699. I believe I get this error because parts of molecules are being
deleted resulting in molecule fragments. Fix evaporate provides a work
around to this because of its molecule flag but I am curious if there is a
simpler way to get rid entire water molecules as they cross a fixed
boundary (in my case the positive z-boundary). I am using the 5-Sep-17,
64-bit Windows version of LAMMPS. Any insight will be greatly appreciated.
fix evaporate is the tool to use.
when you tell LAMMPS to just ignore lost atoms, it cannot remove entire
molecules and their corresponding topology data, because that information
is already lost by the time the check is done.
axel.
Thank you for the clarification.
Sincerely,
Malcolm Porterfield