I’m trying to do HIP modeling, and in order to keep the boundary structure of the model unchanged, I need to set the boundary conditions in the X and Y directions to a fixed boundary, but NPT can’t control the pressure in the X and Y directions of the fixed boundary, so how should I control the temperature in this case?
Original Literature：A new simulation box of dimensions 2031 × 909 × 57.2088 Å with fixed boundary along x and y axes and periodic boundary along the z-axis is created by removing the loose powders from the fused nano-powder bed. The new simulation box contains approximately 5,900,000 atoms. All the HIP calculations are performed with a time step of 0.0005 ps. The initial configuration is relaxed by performing energy minimization using the conjugate gradient method. Then, the system of atoms are equilibrated for 20 ps using time integration on Nose-Hoover style non-Hamiltonian equations of motion in an NPT (constant Number of atoms, Pressure, and Temperature) ensemble at finite temperature of 550 K and zero pressure along z-direction. The pressure along the x, y, and z directions is gradually amplified to 10 GPa in 50 ps and then held steady at 10 GPa for another 65 ps to facilitate the densification of the porous fusednano-powder. After the voids in the as-built microstructure have vanished, the temperature and pressure are brought down to 293 K and 0 GPa, respectively, in the next 100 ps.
What do you mean precisely when you say this?
If your model’s “boundary structure” is reasonably stable at your desired lateral pressures, you should simply be able to barostat the X and Y dimensions and proceed.
If your model’s “boundary structure” is not stable at your desired lateral pressures, then artificially stabilising it will not give you scientifically useful results.
If your model’s “boundary structure” is supposed to be stable at your desired lateral pressures in reality, but not stable in an MD simulation, then the MD simulation is not giving you realistic behaviour (!!), and using that MD simulation will not give you scientifically useful results.
Thanks!What I mean by this is that the model I’m building needs to have the boundaries set to a fixed boundary condition