Heating run of metal oxide with Langevin thermostat (MDALGO=3)

[kai_oshiro]

Dear Experts,

I have been testing heating runs of MD calculation before training force fields at high temperatures.
Therefore, I tested several conditions and confirmed that heating from 100K to 1000K works appropriately using the Nose-Hoover thermostat.
However, I couldn't raise the temperature of the MD calculations with the Langevin thermostat (NpT or NVT).

I read tutorials and forums but could not solve it.
Could anyone help me with this issue?

Attached files include input and some output.
All calculations were done in vasp.6.3.2.

Best,
Kai

[ferenc_karsai]

You have a problem with your with your LANGEVIN_GAMMA settings, because you set them to zero.
You must set these values to a non-zero value.

Values between 5.0-10.0 should work fine for your calculation.

[kai_oshiro]

Dear ferenc_karsai,

Thank you for your quick reply, and sorry for causing you trouble with my small mistake.
I have confirmed that temperature increases correctly with LANGEVIN_GAMMA = 5 5.

About my wrong MD simulation, is it correct to understand that I calculated atoms in a solid system (small velocity and kinetic energy) without random force (γ=0), so the temperature remained much lower?

Best,
Kai

[ferenc_karsai]

Yes. So in principle you calculated without thermostat so that's why the temperature didn't change.

[kai_oshiro]

Understood! Thanks for helping me out. I really appreciate that!