Author Topic: Basic AIMD setup  (Read 6068 times)

mpjohans

  • Full Member
  • ***
  • Posts: 26
  • Karma: +0/-0
    • .
Basic AIMD setup
« on: September 17, 2008, 03:31:42 PM »
Hello All!

I'm about to do my first MD runs with Turbomole, and have a few basic questions, which I couldn't exactly figure out from the short description in the manual.

In short, what I want to do is to study the available conformation space of a molecule at finite (room) temperature. For this, I guess I should set up the following things:

1) Heat up time, getting the molecule from 0 to 298 K smoothly
2) Keeping the molecule at ~298 K for a long period, printing out the coords at some time intervals

Especially 1) seems problematic. For 2) I've inserted the following into mdmaster:

$md_action
  set temperature at t=41342.8146188192 to x=298.0 K
  fix temperature from t=41342.8146188192

(That is 1000 timesteps รก 1 fs from start). Is there a way of controlling how strictly this temperature is enforced? Is it possible to print out the temperature at each step?

Any help and general hints on how to properly do this kind of stuff with TM would be highly appreciated.

Have a nice day,
    Mikael J.
    http://www.iki.fi/~mpjohans/

selliott

  • Newbie
  • *
  • Posts: 1
  • Karma: +0/-0
Re: Basic AIMD setup
« Reply #1 on: October 31, 2008, 03:43:11 PM »
Hi Mikael,
I don't know of an automatic way to heat up the molecule as the code is currently written.  You could increase the temperature in a series of steps:
$md_action
  set temperature at t=10000.0 to x=50.0 K
  set temperature at t=20000.0 to x=100.0 K
  set temperature at t=30000.0 to x=150.0 K

etc.
The velocities should be scaled each time and then the molecule will redistribute that energy.
Best wishes,
Simon.