Re-visitation of Two Models for Predicting Mechanically-Induced Disordering after Cryogenic Impact Milling

Area of Science:

• Materials Science
• Solid State Physics

Background:

• Cryogenic milling is used to induce disorder in materials.
• Predicting the extent of disordering is crucial for material applications.
• Existing models for disordering potential include critical dislocation density and empirical approaches.

Purpose of the Study:

• To compare the predictive accuracy of two models for material disordering potential after cryogenic milling.
• To evaluate model performance with expanded datasets.

Main Methods:

• Simulated elastic shear moduli (μs) in silico.
• Predicted complete disordering potential using critical dislocation density (ρcrit) and bivariate empirical models.
• Characterized mechanical disordering via PXRD and DSC after cryomilling.

Main Results:

• The ρcrit model showed 55% accuracy (13/29 misclassifications).
• The bivariate empirical model achieved 97% accuracy (31/32 correct classifications).
• Recalibration of the empirical model maintained 94% accuracy with only 2 misclassifications.

Conclusions:

• The critical dislocation density model's accuracy decreased with dataset expansion.
• The bivariate empirical model demonstrated superior and robust prediction accuracy.
• Empirical model suggests glass transition temperature (Tg) and molar volume (Mv) are key factors in amorphous material formation via cryomilling.