Unlock the electrochemical performance of g-SiC/Nb2CO2 heterostructure as lithium/sodium ion storage anode with first-principles and machine learning

Abstract

Heterojunction has achieved some breakthroughs in the performance aspects such as specific capacity, rate performance and service life through interface structure, charge coupling and interaction regulation, and has become a key regulatory means for high-performance electrode materials. Herein, a heterojunction combined g-SiC with Nb₂CO₂ is established to systematically predict the electrochemical properties through first-principles calculations and machine learning. Based on the results of AIMD simulation at 300 K, it is found that the heterojunction has neither structural deformation nor bond breakage, revealing good thermodynamic stability. Moreover, the lowest diffusion barriers of Li and Na migration are 0.58 eV and 0.27 eV, with the open circuit voltage lowered 0.97 V and 0.18 V, and theoretical specific capacities of 910.54 mAh/g and 505.86 mAh/g, respectively. These findings indicate that g-SiC/Nb₂CO₂ is a promising anode material for Li/Na batteries. Meanwhile, combining the characteristics and physical properties of atoms as descriptors for machine learning, their key influence on adsorption performance is explored, which provides valuable insights for the design of anode materials toward high-performance metal ion batteries in the future.