Nucleation rates from small scale atomistic simulations and transition state theory

Area of Science:

• Physical Chemistry
• Computational Materials Science
• Chemical Physics

Background:

• Evaluating nucleation rates via molecular dynamics is challenging due to slow timescales and finite system size effects.
• Classical nucleation theory approximations can limit accuracy in simulations.
• Accurate nucleation rate determination is crucial for understanding phase transitions.

Purpose of the Study:

• To develop a general and accurate method for calculating nucleation rates from molecular dynamics simulations.
• To overcome the limitations of slow nucleation timescales and finite size effects in computational studies.
• To provide a robust alternative to classical nucleation theory approximations.

Main Methods:

• Utilized enhanced sampling techniques, specifically metadynamics, to address slow nucleation timescales.
• Reconstructed the free energy surface from an appropriate ensemble to circumvent finite size effects.
• Applied transition state theory with a dynamical correction for diffusive recrossing.

Main Results:

• Achieved accurate macroscopic droplet nucleation rates for argon vapor.
• Calculations spanned 16 orders of magnitude, showing excellent agreement with literature data.
• Demonstrated the method's efficacy using small system sizes (512 atoms).

Conclusions:

• The developed method accurately determines nucleation rates without relying on classical nucleation theory approximations.
• Enhanced sampling and free energy surface reconstruction effectively overcome common simulation limitations.
• This approach offers a powerful tool for studying nucleation phenomena across various systems.