Hi,
I am experimenting with using amber to simulate a non-biological polymer (Polyethylene terephthalate in this case). The system consists of 10 strands of polymer, each of MW ~ 4,000. I am using a periodic boundary (but no water molecules) as I was wanting to see what the box size came down to at equilibrium and I was hoping that I might be able to use AMBER as a *much* faster substitute for something like Forcite in Materials Studio.
I have minimised then heated the system to 400 K just fine but during equilibration I get strange results as follows:
------------------------------------------------------------------------------
NSTEP = 110000 TIME(PS) = 360.000 TEMP(K) = 393.95 PRESS = 2.0
Etot = 12309.5181 EKtot = 5166.8032 EPtot = 7142.7148
BOND = 1485.0625 ANGLE = 2578.6577 DIHED = 2946.0975
1-4 NB = 2092.4314 1-4 EEL = 5210.4485 VDWAALS = -2527.1920
EELEC = -4642.7908 EHBOND = 0.0000 RESTRAINT = 0.0000
EKCMT = 13.1973 VIRIAL = -30.8884 VOLUME = 1013509.8700
Density = 0.0694
------------------------------------------------------------------------------
wrapping first mol.: -1866860.98880 -795847.12907 -1555380.40974
wrapping first mol.: -1866860.98880 -795847.12907 -1555380.40974
NSTEP = 115000 TIME(PS) = 365.000 TEMP(K) =********* PRESS = -6357.0
Etot = ************** EKtot = ************** EPtot = **************
BOND = -0.0000 ANGLE = 554318.5662 DIHED = 23858.3639
1-4 NB = 0.0000 1-4 EEL = 0.0622 VDWAALS = **************
EELEC = 26.1532 EHBOND = 0.0000 RESTRAINT = 0.0000
EKCMT = 12582912.0000 VIRIAL = 13190994.3106 VOLUME = 4430319.6150
Density = 0.0159
------------------------------------------------------------------------------
wrapping first mol.:-35683624.36994 -3720160.40395 -5584955.18246
wrapping first mol.:-35683624.36994 -3720160.40395 -5584955.18246
NSTEP = 120000 TIME(PS) = 370.000 TEMP(K) =********* PRESS =********
Etot = ************** EKtot = ************** EPtot = 106105996.2891
BOND = -0.0000 ANGLE = 625557.8344 DIHED = 22922.0560
1-4 NB = 0.0000 1-4 EEL = 0.0532 VDWAALS = 105458199.0325
EELEC = -682.6869 EHBOND = 0.0000 RESTRAINT = 0.0000
EKCMT = 12582912.0000 VIRIAL = 32401261.8614 VOLUME = 5158252.5762
Density = 0.0136
------------------------------------------------------------------------------
Once those blank values (**************) occur the .mdcrd stops being written, the system density drops with a corresponding jump in volume (as you can see) but the simulation continues. Looking at the trajectory to that point gives no indication that anything has gone awry - the polymer bunches up nicely in the middle of the box. I even put a restraint on a central atom, and then a polymer chain, in case there was something wandering that I could not see but this phenomenon keeps occurring. I also tried running in shorter trajectories (200,000 steps) and using an updated .rst as the -ref. I also tried iwrap = 0.
Equil
&cntrl
imin = 0,
ig = -1,
iwrap = 1,
irest = 0, ntx = 1,
ntc = 2, ntf = 2,
ntp = 1, pres0 = 1.0, taup = 5.0,
cut = 10.0,
tempi = 400.0, temp0 = 400.0,
ntt = 3, gamma_ln = 2.0,
nstlim = 200000,
ntwx = 5000, ntwe = 5000, ntpr = 5000, ntwr = 5000,
dt = 0.001,
ntr = 1,
restraint_wt=0.01,
restraintmask=':8.C',
/
Hoping the collective AMBER wisdom can help.
Thanks in advance.
Andrew
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jun 24 2014 - 22:30:02 PDT
