[gmx-users] Is vacuum simulation NVT?

[Michael Shirts]

> In the absence of PBC, you simply have an infinite system.  In a loose
> sense, that may be NVT, but V is infinite, so whether or not you can
> consider that to be constant or not is theoretical math above what I know :)

A real molecule in vacuum is usually NVE -- it is not coupled to the
environment, and thus must have conserved energy.  You can certainly
add a thermostat, and then it will be NVT, though it won't be very
much like a real isolated gas molecule.  If you try to run NPT, the
simulation will likely crash because of numerical instabilities, and
there's not much of a point, since you are essentially either 1) in
the ideal gas limit if running with no periodic boundary conditions 2)
in some sort of weird superdilute crystal that really doesn't resemble
anything real if run with a periodic boundary conditions

If you are subtracting out the center of mass motion, then V is not
infinite -- you remove the center of mass degree of freedom, and thus
you have a very different ensemble than if you include the center of
mass motion.  You would need to multiply by V to get the partition
function for an actual gas.

Note that I would strongly suggest sd as the integrator/thermostat,
since there are real issues with ergodicity in systems with only a few
degrees of freedom