I'm using GULP to test an interatomic potential for a ceramic material. I'm interested in describing how the volume of the unit cell changes with temperature. I've used GULP to minimize de Free-energy of the system at different temperatures and I can find a reasonable change in the volume. However, when using a NPT molecular dynamics using stochastic integration of the equations of motion the system reaches equilibrium at a smaller volume. Is this a common phenomenon in MD?
Equilibrating the system with MD should result in the minimization of free energy. So, in principle, you should get the same results. But you should be careful about the definition and the choice of the free energy itself. If you have set up your MD correctly, the NPT simulation would minimize the Gibbs free energy (G = μN). Are you also applying the same conditions in your simulation using GULP?
Thank you for your response, Mr. Sadeghi. Yes, when running a MD in GULP, I'm using a NPT ensemble. When using lattice dynamics, I'm also using constant temperature and pressure (conp). I get a difference in volume of approximately 4%. Could this be due to the classical treatment of the system in low temperatures in the molecular dynamics?
- Badges
- Science topic
- Similar topics
- Chemistry
- Theoretical Chemistry
- Computational Chemistry
- Molecular Dynamics