Predicting polymorphism in molecular crystals using orientational entropy.

Area of Science:

• Computational chemistry
• Materials science
• Crystallography

Background:

• Discovering molecular crystal polymorphs is crucial for materials science.
• Simulating crystallization processes computationally is challenging due to long timescales.

Purpose of the Study:

• To develop an efficient computational method for discovering polymorphs at finite temperatures.
• To enable the identification and classification of different crystal structures within accessible simulation times.

Main Methods:

• Utilized well-tempered metadynamics to enhance fluctuations of a collective variable.
• Employed an entropy surrogate based on an extended pair correlation function.
• Developed a similarity metric using generalized Kullback-Leibler divergence for configuration classification.
• Applied hierarchical clustering to automatically group discovered configurations into polymorphs.

Main Results:

• Successfully applied the method to urea and naphthalene, identifying multiple polymorphs for each.
• Demonstrated the stabilization of at least one polymorph at finite temperature due to entropic effects.
• Validated the computational approach for polymorph discovery and classification.

Conclusions:

• The developed computational method effectively discovers and classifies molecular crystal polymorphs at finite temperatures.
• Entropic effects play a significant role in stabilizing certain polymorphs.
• This approach accelerates the exploration of the crystal structure landscape.