MC and MD inconsistency

[skvara.jiri]

Dear RASPA users,
I tried to compare MD and MC runs in NPT ensemble for 2-methylbutane molecules which forcefield is attached to Raspa as an example. I did not modify the force field but I am not able to reach the same density in MC and MD (in MD the density is lower by 100 kg/m^3). I tried simulations at different state point and I was not able to match MC and MD Raspa results. I also tried simulations of different molecules and my results match only for water all other molecules differed in density.

I had found that MD results are changes when I am using the option "ComputeMolecularPressure yes". I can't see why there should be any change in density with this option if the barostat is working properly. If this option is necessary for barostat, then the NPT run should not be performed without this command.


Attached you can find my MD simulation script (system did not allow me to attach the file itself). I will be glad for your advice. I believed that problem is in my simulation settings.

SimulationType                   MolecularDynamics
NumberOfCycles         50000000
NumberOfInitializationCycles   1000
NumberOfEquilibrationCycles   10000
PrintEvery                       100
RestartFile                      no

InitEnsemble    NVT
RunEnsemble   NPT
TimeStep   0.001

Forcefield                       local
CutOff                           12
ChargeMethod               Ewald
CutOffVDW         12
CutOffCoulomb         12

Box 0
BoxLengths 37 37 37
ExternalTemperature 293.15
ComputeMolecularPressure yes
ExternalPressure 101325

Movies yes
WriteMoviesEvery 1000


Component 0 MoleculeName             2-methylbutane
            MoleculeDefinition       local
       ReinsertionProbability   1.0
       TranslationProbability   1.0
            CreateNumberOfMolecules  256
      

[David Dubbeldam]

A common inconstancy between MC and MD is the VDW potential. Make sure you use a shifted VDW potential for MC.