| Re: thermal stability with openmx ( No.1 )|
- Date: 2022/07/13 18:07
- Name: Yung-Ting Lee
- (1) The Nose Hoover method is recommended, i.e. NVT_NH.
Or, one can check Journal papers about method of MD simulations to be references.
(2) As for
The first number '4' gives the number of the following lines to control the temperature, i.e. next 4 command lines for controlling temperatures.
In the consecutive lines, the first and second columns give MD steps and a given temperature for nuclear motion.
1 1000 -> 1st MD step at a given temperature 1000 K.
100 1000 -> 100th MD step at a given temperature 1000 K.
400 700 -> 400th MD step at a given temperature 700 K.
700 600 -> 700th MD step at a given temperature 600 K.
This example is to show you how to change temperature at n-th MD step in a system if changing a given temperature is necessarily.
Reference : http://www.openmx-square.org/openmx_man3.9/node61.html
(3) In a system at 300 K constantly, one can just set a given temperature 300 K.
| Re: thermal stability with openmx ( No.2 )|
- Date: 2022/07/25 07:01
- Name: Zeinab_mrd
- Dr Yung-Ting Lee
Thanks in advance for your help.
In the third comment, do you mean that I should use the following md control to calculate the temperature stability in 700 K using the hover nose?
Or this one is true:
Because when I use the first one, I see a lot of temperature fluctuations.
| Re: thermal stability with openmx ( No.3 )|
- Date: 2022/07/26 15:00
- Name: Yung-Ting Lee
- As you mentioned, temperature fluctuates during a MD simulation with Nose-Hoover thermostat. It is a normal behavior at Nose-Hoover scheme.
And, average temperature should be equal to a given temperature when total time of this md simulation is long engough.
One can test temperature variation in a classical MD simulation with Lennard Jones potential by using velocity scaling method or Nose-Hoover method for checking temperature fluctuation.
Velcoities of atoms are related to temperature and coupled to heat bath with a bath mass. After coupling, temperature will be rescaled and subsequently velcoities of atoms will be redistributed for the next MD step.
On the other hand, in NVE scheme, energy also fluctuates and average energy will be close to a given energy.