Effect of interfacial bridging atoms on the strength of Al/CNT composites: machine-learning-based prediction and experimental validation

Abstract

Weak interfacial adhesion is one of key obstacles to develop aluminum matrix composites containing carbon nanotubes (CNTs). This study suggests the concept of bridging atoms to enhance the interfacial wetting between aluminum and CNTs. Machine learning and sensitivity analyses were employed to determine the most favorable element as a bridging atom. Copper was identified as the most effective bridging atom, and its bridging efficiency (enhancement of strengthening efficiency of CNTs) was experimentally validated by comparison with those in the monolithic Al and Al–Si matrix. As a result, the strengthening efficiencies of the CNTs were measured to be ∼43, 27, and 73 MPa/vol% for the Al, Al–Si, and Al–Cu matrices, respectively, which is comparable with the prediction by the machine learning model.

Keywords

• Metal matrix composites
• Mechanical properties
• Mechanical alloying
• Carbon nanotubes
• Machine learning
• Interface