hello, everyone~~
the initial velocity of a group of atoms can be created by the “velocity group-id create temp seed” command. however i wonder if this command is a must in input files.
hello, everyone~~
the initial velocity of a group of atoms can be created by the “velocity group-id create temp seed” command. however i wonder if this command is a must in input files.
No, it isn’t. But in many cases it is incredibly convenient.
Axel.
Hi, as far as I know, usually we use this command before equilibration runs. And if you do NOT use it, the simulation can still goes fine.
But, this command can speed up the equilibration, because the velocities of the atoms get closer to the final state.
I am a beginner here too, so I hope somebody much more experienced at this field can give a more accurate answer.
2015-09-25
yeah, when this command is used, an ensemble of velocity can be created for a group of atoms and the temperature also reach the target temperature. thus the system can reach equilibration as soon as possible. however, will the results be the same for the cases with or without “velocity create” command?
That is a more tricky question. If the system is really in equilibrium and both setups reached the same equilibrium (i.e., the simulations are ergodic) then all thermodynamical properties of the system should be equal, at least up to statistical errors. However, I don’t think It is easy to determine if a system has reached equilibrium or if it is ergodic. I think the best you can do is monitor things like temperature, pressure, etc. and wait until they reach a steady state, and then hope that this steady state is also equilibrium.
Sorry, I do not know. But I think you can run two test input files for this purpose with or without velocity command.
Do you mean exact same results? Sorry, I do not understand what is meaning of your concern.
2015-09-25
It is not a concern of your specific case but more of a general philosophical thing: Unless you know beforehand what the equilibrium distributions should be, you can never be sure if a system is properly equilibrated. On the other hand, if you know what equilibrium distributions are, you do not need to do simulations.
and how do you confirm if a system reach equilibration when you do some simulations? the stable energy or something else?
Here is when you need to go pick a book about the topic and read it.
Carlos