Statistically Equivalent Representative Volume Element Modeling of Carbon Nanotube–Copper Composite Fibers Based on Cross-Sectional Microstructures

Carbon nanotube (CNT) fibers coated with metals represent a promising class of multifunctional materials that combine enhanced mechanical strength with high electrical conductivity. In this study, we develop a computational framework for two-dimensional porous CNT fibers and metal-coated CNT fibers by integrating statistically equivalent representative volume element (SERVE) models with finite element methods. The SERVE framework generates synthetic microstructures that reproduce key statistical descriptors extracted from microscopy images, enabling efficient and representative simulations without replicating specific physical samples. Using the SERVE approach, the tensile response of pristine CNT fibers and Cu/CNT fibers is numerically evaluated as a function of packing density and copper volume fraction, assuming a porosity-induced necking failure mechanism. For pristine CNT fibers, tensile strength increases with packing density, whereas for Cu/CNT fibers, tensile strength is strongly dependent on microstructural morphology. Among the configurations studied, Cu/CNT fibers featuring a high-packing-density CNT core encapsulated by a copper shell exhibit the highest performance index, indicating the most favorable balance between tensile strength and electrical conductivity. Although the SERVE approach is demonstrated using in-plane microstructures derived from cross-sectional images without accounting for out-of-plane structural variations, it establishes a scalable framework for modeling stochastically porous Cu/CNT fiber microstructures that can be extended to fully three-dimensional analyses in future work.