Re: Problem with SPC/E water simulation

From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Fri Mar 04 2011 - 15:50:13 CST

On Fri, Mar 4, 2011 at 4:07 PM, Davide Vanzo/FS/VCU <dvanzo_at_vcu.edu> wrote:
> Dear all,

dear davide,

> this is the first time I use Namd to run a MD simulation and I need to
> equilibrate a simple SPC/E water sample at 300K and 1atm with SHAKE. I
> started from a lattice of water molecules with SPC/E geometry at very low
> density in order to avoid superimpositions. The PBC cell is correctly
> centered and sized and the simulation ran without evidence of
> superimpositions. As you can see from the input script below I ran an NPT

it would have been much simple to take an already equilibrated water
restart, e.g. the one that ships with VMD for the solvate plugin. even
though it was built for a different water potential, this would have sped
up the process of equilibration significantly.

> simulation with temperature rescaling and langevin barostat. However the
> calculated pressure never reached the desired value, with strong
> oscillations (hundreds atm) around zero, and the compression stopped at
> densities higher than that expected for water.
> What is wrong?

probably nothing. you didn't say, how many water molecules you system
contains in total. the instantaneous pressure fluctuates a lot in dense
and not very compressible systems. you have to look at the long time
average. also, you didn't say what density you were comparing to what.
did you compute the average density of the last part of your MD? did
you compare to experimental water or SPC/E water (which is not exactly
the same!).

cheers,
     axel.

>
>
>
> numsteps                        100000
> timestep                        1
> firsttimestep                0
> stepspercycle                20
>
> nonbondedFreq                1
> fullElectFrequency        1
>
> cutoff                        10
> switching                        off
> switchdist                        8
> pairlistdist                12
>
> exclude                        scaled1-4
> 1-4scaling                        1
>
> rigidBonds                        all
> rigidTolerance                0.00000001
> rigidIterations                100
> rigidDieOnError                on
>
> vdwGeometricSigma                yes
>
> PME                        on
> PMETolerance        0.00001
> PMEInterpOrder        4
> PMEGridSpacing        1
>
> dielectric                 1
>
> ZeroMomentum            yes
>
> cellBasisVector1   68   0    0
> cellBasisVector2    0  68    0
> cellBasisVector3    0   0   68
> cellOrigin                    34  34   34
>
> wrapAll         off
>
> temperature                300
> COMmotion               no
>
> rescaleTemp                300
> rescalefreq             5
>
> langevinPiston                  on
> langevinPistonTarget            1.01325
> langevinPistonTemp              300
> langevinPistonDecay             200
> langevinPistonPeriod            400
>
> useFlexibleCell         no
> useGroupPressure        yes
>
> coordinates                water_5K.pdb
> structure                water_5K.psf
> paratypeCharmm        on
> parameters                spce_water.par
>
> outputEnergies        50
>
> binaryoutput        no
> outputname                water_r1
>
> binaryrestart        yes
> restartname                water_r1_rst
> restartfreq                10000
>
> DCDfile                water_r1.dcd
> DCDfreq                 1000
>
> XSTfile                water_r1.xst
> XSTfreq                 1000
>
>
>
> This is my parameter file for SPC/E water
>
>
>
> BONDS
> OSP  HSP   450.000     1.0000
> HSP  HSP     0.000     1.6330
>
> ANGLES
> HSP  OSP  HSP    55.000   109.4700
>
> NONBOND
> HSP    0.000000  -0.000001     0.000000
> OSP    0.000000  -0.155350     1.582760
>
> END
>
>
>
> whether the topology file is
>
>
> 18  1
>
> MASS     1 HSP    1.00797  H
> MASS     2 OSP    15.99940 O
>
> DEFAULT FIRST NONE LAST NONE
>
> RESI SPE         0.0000
> GROUP
> ATOM O  OSP     -0.8476
> ATOM HA   HSP     0.4238
> ATOM HB   HSP     0.4238
> BOND O HA O HB HA HB
> ANGLE HA O HB
> DONOR HA O
> DONOR HB O
> ACCEPTOR O
> PATCHING FIRST NONE LAST NONE
> END
>
>
>
> Thank you to anyone could help me.
>
> Davide

-- 
Dr. Axel Kohlmeyer
akohlmey_at_gmail.com  http://goo.gl/1wk0
Institute for Computational Molecular Science
Temple University, Philadelphia PA, USA.

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:19:53 CST