[gmx-users] Density of pure tetrolic acid does not agree with experimental data

[Alex]

Nathan,

The experimental data ideally corresponds to an infinitely long,
well-converged MD simulation at the corresponding set of temperature and pressure, not after EM
and most certainly not after just filling a box, where spacing is
according to a preset cut-off and not due to interatomic interactions.

Alex

NKH> Hello Gromacs users,

NKH> I tried to create a 6.5 nm cubic box of tetrolic acid (otherwise
NKH> known as 2-butynoic acid), but the density is too low. According
NKH> to this, http://www.chemspider.com/Chemical-Structure.61810.html,
NKH> the density should be about 0.964
NKH> g/mL. Given the molecular weight of tetrolic acid, this
NKH> corresponds to right around 1900 molecules in the 6.5 nm box I
NKH> prepared. Unfortunately, when I attempted to fill the box,
NKH> Gromacs only found room for 1676 molecules, resulting in
NKH> a density of 0.852 g/mL. I know that models are not 100%
NKH> accurate, but I'm worried about a difference that big. Should I
NKH> expect the system to condense when I minimize, and just shrink
NKH> the box afterwards? Or do you think there is
NKH> another problem?

NKH> Thanks for your help,
NKH> Nathan