Competitive Hydrogen Migration in Silicon Nitride Nanoclusters: Reaction Kinetics Generalized from Supervised Machine...

Area of Science:

• Computational Chemistry
• Materials Science
• Chemical Kinetics

Background:

• Hydrogen shift reactions are crucial in silicon nitride chemistry.
• Understanding reaction pathways and kinetics is essential for material design and synthesis.

Purpose of the Study:

• To calculate rate coefficients for 52 hydrogen shift reactions in silicon nitrides.
• To investigate the influence of reaction type and molecular structure on kinetic parameters.
• To develop machine learning models for predicting reaction energy barriers.

Main Methods:

• G3//B3LYP composite method for electronic structure calculations.
• Statistical thermodynamics for rate coefficient determination.
• Supervised machine learning for predictive modeling.

Main Results:

• Identified three main types of hydrogen shift reactions (N-Si, Si-Si endothermic, Si-Si exothermic).
• Exothermic Si-Si shifts involve a stable intermediate and a two-step pathway.
• Kinetic parameters (pre-exponential factors and activation energies) were computed for various silicon nitrides.
• Machine learning models successfully predicted energy barriers based on molecular descriptors.

Conclusions:

• The type of hydrogen shift reaction significantly impacts kinetic parameters.
• Computational and machine learning approaches provide valuable insights into silicon nitride reaction mechanisms.
• Findings can guide the development of new silicon nitride materials and processes.