Decoding α-MoC_{1- x} Nanoparticle Formation in Continuous Flow via Machine Learning.

Area of Science:

• Materials Science
• Nanotechnology
• Chemical Engineering

Background:

• Molybdenum carbide nanoparticles (α-MoC_{1- x} NPs) present a cost-effective alternative to noble metal catalysts.
• Developing efficient and scalable synthesis methods for these nanoparticles is crucial for their widespread application.

Purpose of the Study:

• To report a mild continuous-flow synthesis of α-MoC_{1- x} NPs.
• To integrate in-line spectroscopic monitoring and machine learning (ML) for real-time quantification of synthesis.
• To elucidate the reaction pathway and kinetics governing nanoparticle formation.

Main Methods:

• Continuous-flow synthesis utilizing molybdenum hexacarbonyl (Mo(CO)₆) as a precursor.
• In-line spectroscopic monitoring coupled with a multilayer perceptron machine learning model.
• Ex situ characterization using small angle X-ray scattering (SAXS) and X-ray diffraction (XRD).

Main Results:

• Accurate real-time quantification of precursor conversion and α-MoC_{1- x} NP formation.
• Identification of a two-step reaction pathway: precursor conversion to an amorphous intermediate, followed by intraparticle crystallization.
• Confirmation that precursor conversion is the rate-limiting step.

Conclusions:

• Machine learning provides powerful insights into nanoparticle nucleation and growth dynamics.
• The integrated approach enables precise control over nanoparticle synthesis.
• This work paves the way for self-driving, flow-based platforms for nanoparticle synthesis.