From: Aron Broom (broomsday_at_gmail.com)
Date: Mon Nov 11 2013 - 19:12:33 CST
I think that normally if one is using temperature to melt things or for say
replica exchange, you would do constant volume, i.e. NVT, rather than
constant pressure, i.e. NPT. I may be wrong. Of course, the TIP3P water
model isn't meant for phase transitions, but one would still expect
something of that sort at some temperature once the kinetic energy
overcomes the attractive potentials. So I would think what you saw at
~550K was TIP3P "boiling".
Probably a good bet would be to check out what people do with replica
exchange (maybe NVT as above), since they would need to solve similar
On Mon, Nov 11, 2013 at 6:55 PM, Martin, Erik W <Erik.Martin_at_stjude.org>wrote:
> So, I have set up simulations where every 2 nanoseconds I increase the
> temperature by ten degrees. The goal of this simulation was to attempt to
> "melt" structures that had previously formed at constant temperature. It
> is possible that I went about this in a rather naive way, and would
> appreciate any insight.
> I accomplished this by simply rescaling velocities and resetting Langevin
> Temp and Piston Temp to the current temperature in a short tcl script. I
> did not anticipate that the volume would be static in these simulations,
> but was a bit surprised by what I saw.
> The simulation seemed to progress fine with actually very little change in
> Volume until it got to 550 degrees. At this point I saw a sudden explosion
> in volume accompanied by a drop to near zero pressure. These jumps appear
> to also include a sudden increase in total energy. I've never done this
> sort of thing before and am not sure what I am actually seeing happen.
> If anyone can interpret this for me, I'd greatly appreciate it.
> PS. This is a simulation in TIP3 water of ~90k atoms.
> Email Disclaimer: www.stjude.org/emaildisclaimer
> Consultation Disclaimer: www.stjude.org/consultationdisclaimer
-- Aron Broom M.Sc PhD Student Department of Chemistry University of Waterloo
This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:23:58 CST