Reaction coordinates and rate constants for liquid droplet nucleation: Quantifying the interplay between driving...

Area of Science:

• Physical Chemistry
• Chemical Physics
• Materials Science

Background:

• Homogeneous nucleation of liquid droplets from supersaturated vapor is a fundamental process.
• Understanding nucleation dynamics is key to controlling phase transitions and material properties.

Purpose of the Study:

• To develop and validate a more accurate reaction coordinate (RC) for nucleation.
• To investigate the role of local density fluctuations in nucleation at varying supersaturations.
• To improve the efficiency of molecular dynamics simulations for nucleation studies.

Main Methods:

• Utilized biased and unbiased molecular dynamics (MD) simulations.
• Employed the spectral gap optimization approach (SGOOP) to calculate the RC.
• Constructed the RC as a linear combination of liquid-like atom count and two local density fluctuation parameters.
• Applied the optimized RC within the infrequent metadynamics framework.

Main Results:

• The reaction coordinate for nucleation transitions from solely counting liquid-like atoms to including local density fluctuations as supersaturation decreases.
• Density fluctuations become non-Markovian at lower supersaturations, indicating longer memory effects.
• The optimized RC significantly improves nucleation rate accuracy and accelerates simulations by four orders of magnitude compared to unbiased MD.

Conclusions:

• Local density fluctuations are essential components of the nucleation reaction coordinate, especially at lower supersaturations.
• The SGOOP-derived RC and infrequent metadynamics offer a powerful and efficient method for studying nucleation.
• This work provides a more nuanced understanding of droplet nucleation dynamics and simulation methodologies.