A New Methodology for Evaluating the Structural Similarity between Different Phases Using a Dimensionality Reduction...

Area of Science:

• Materials Science
• Computational Chemistry
• Crystallography

Background:

• Evaluating structural similarity between crystalline and amorphous phases is crucial in materials science.
• Traditional methods can be subjective or computationally intensive.
• A quantitative and impartial approach is needed for complex systems.

Purpose of the Study:

• To propose a novel methodology for evaluating structural similarity between different phases.
• To apply this method to analyze amorphous calcium carbonate (CaCO3) and its crystal polymorphs.
• To investigate the influence of additives (Mg2+, Sr2+, water) on structural similarity.

Main Methods:

• Utilizing a dimensionality reduction (DR) technique from machine learning and statistics.
• Analyzing the geometrical similarity of pair and angular distribution functions.
• Employing molecular dynamics simulations to obtain atomic-scale structures of CaCO3 phases.

Main Results:

• Without additives, amorphous CaCO3 shows higher structural similarity to vaterite than calcite or aragonite.
• Mg2+ and Sr2+ ions decrease the structural similarity between amorphous CaCO3 and vaterite.
• High water concentration leads to amorphous CaCO3 separating into particles, increasing similarity to vaterite.

Conclusions:

• The proposed DR-based methodology provides an impartial evaluation of structural similarity.
• Additives significantly alter the structural relationship between amorphous and crystalline CaCO3.
• The method is applicable to complex multicomponent systems for phase similarity assessment.