Multiwfn forum

Ahmed Wahba

Member

Registered: 2023-04-01

Posts: 4

How were temperature and pressure controlled in MD calculations?

I made an MD in his manuscript and a comment came to me saying as follows
How were temperature and pressure controlled in MD calculations?

I have made the following comment:

In MD, physical properties are derived under isothermal, isobaric, or both conditions (NVT and NPT). When NVT is used, the thermostat adapts the kinetic energy of the particles to the target temperature by controlling their momenta. In the case of NVT, kinetic energy is determined by evaluating temperature, while in the case of NPT, it is determined by evaluating temperature and pressure. Dynamic simulations were traditionally conducted at a constant temperature as NVT, but production runs will continue to be conducted at constant energy (NVE).  There is a possibility that some thermostats will affect diffusion coefficients since they can interfere with system dynamics. The frame output interval was reduced and the number of steps increased to collect more data for analysis. So I only did 5,000 steps. Two charts are updated during the calculation. In one plot, power is plotted over time over different components, and in the other, the temperature is plotted over time. Since this is an NVE group arithmetic, the total power should be constant. The exchange of kinetic and potential energy should occur, but there should be no net exchange as long as the equilibrium lasts for a long time. It should also fluctuate around 300 K on average without a systematic change, and the pressure should average 1 atm or 0.0001GPa to obtain a balanced system.

His response to me was as follows:

Their reply to my nineth comment is wrong. However, they did not apply even this wrong reply in the revised manuscript. temperature can be controlled by applying thermostat or heat bath and velocity rescaling whereas the pressure is fixed by applying barostate or volume rescaling.
C9. How were temperature and pressure controlled in MD calculations?

Please, Doctor, can you explain to me the answer to this comment?

sobereva

Tian Lu (Multiwfn developer)

From: Beijing

Registered: 2017-09-11

Posts: 1,631

Website

Re: How were temperature and pressure controlled in MD calculations?

If "How were temperature and pressure controlled in MD calculations?" is the comment raised by a reviewer, and you are trying to reply his comment, you should explicitly clarify whether or not thermostat and barostat have been applied in your MD simulation. From your statement, it seems that you conducted NVE simulation in production stage, therefore you should tell reviewer that neither thermostat and barostat were employed, and meantime explain why. If your main purpose of the simulation is calculating diffusion coefficients, it is indeed somewhat reasonable to avoid using barostat since volume scaling will introduce artificial atomic displacements and thus affect diffusion coefficient, however using thermostat in fact is generally not harmful for this purpose, see J. Chem. Theory Comput., 9, 2887 (2013).

In addition, you should mention that the volume used in NVE simulation is the well equilibrated one obtained in previous simulation, and the temperature maintained during the NVE simulation is found to be very close to the expected value. If the two points cannot be guaranteed, then your NVE simulation cannot produce a meaningful diffusion coefficient.

Ahmed Wahba

Member

Registered: 2023-04-01

Posts: 4

Re: How were temperature and pressure controlled in MD calculations?

I learn a lot from you
Thank you very much, Dr. Tian Lu