The bulk modulus of diamond calculated by LAMMPS and CP2K(MM) is very different

[Yanyan Zhang]

Dear All,

Firstly, I calculated bulk modulus of diamond by cp2k with QS METHOD, the lattice constant varied from 3.537 to 3.607, then the bulk modulus was calculated to 434.85 GPa by fitting the correlation between volume and energy (Birch–Murnaghan equation of state). And the result matches the real experiment. 

Then, I calculated bulk modulus of diamond by cp2k with MM METHOD. I choose a SiC.tersoff as the potential type whose bulk modulus was calculated to 425.68 GPa by LAMMPS(https://openkim.org/id/Tersoff_LAMMPS_Tersoff_1989_SiC__MO_171585019474_003). Similarly, the lattice constant varied from 3.537 to 3.607, but then the bulk modulus was calculated to 54482 GPa by fitting the correlation between volume and energy (Birch–Murnaghan equation of state). The CP2K's result does not match the LAMMPS's result.

Question1 : Why do their results have a big difference?(CP2K and LAMMPS) And is my input file wrong? 

Question2 : I want to find a classical force field of diamond for QMMM, do you have a better advice?

The input files and results are attached.