[lammps-users] heat capasity at constant pressure for water

My system contains 1000 tip4p water molecules (3000 atoms). For Cv calculation, initially I equilibrate the system in NPT ensemble for 10ps, and then follow by NPT for 100ps. Cv was calculated from RMS deviation of the total energy. For Cp calculation, the system was (straight away) equilibrated in NPT for 100ps. Cp was calculated from the RMS deviation of enthalpy(etotal+P*V). Please comment on this approach.

Thanks for your help.

Regards,
Christopher

Chris,

You should probably read the prior literature on these methods, as there are
lots of inherent pitfalls that you do not want to relearn. For example, I
suspect you are using the wrong definition of p. I would use the specified
external pressure, not the instantaneous internal pressure. Also, there is
an extensive literature demonstrating that NPT/NPH volume fluctuations do
not reliably predict bulk modulus. This will also be an issue for Cp.

A more robust method for obtaining Cv and Cp is to measure <KE+PE> and <T>
in NPH and NPT simulations, and then calculate the slope of <KE+PE> versus
<T>.

Aidan

Chris,

We have computed Cp’s for liquids using the method Aidan described. Take a look at Jayaraman, Thompson, von Lilienfeld, Maginn, Ind. Eng. Chem. Res., 2010, 49, 559-571

Sai.

Dear Dr. Saivenkataraman Jayaraman,

Thank you, I had read your article (Jayaraman, Thompson, von Lilienfeld, Maginn, Ind. Eng. Chem. Res., 2010, 49, 559-571). I have some questions and would like to ask you.

In the beginning, you use experimental data (lattice parameters,etc.) to initialize your system and then perform a NPT equilibration. I wonder why you did not initialize velocities for the particles in your system at 297K?
After that, you increase your system to a higher temperature and then equilibrate your system and follow by extracting quantities by averaging over time steps. I would like to ask you how did you increase your system initially at 297K to a higher temperature, say 400K. Did you use NVT thermostat (fix 1 all NVT temp 297 400 100) to do that? and then you use NPT barostat to equilibrate your system at 400K (fix 1 all npt temp 400 400 100 aniso 1.0 1.0 500)?

Thank you very much.

Regards,
Christopher