The temperature coefficient of resistance (TCR) determines the electrical performance of materials in electronics. For a carbon nanotube (CNT) nanocomposite, change of resistivity with temperature depends on changes in CNT intrinsic conductivity, tunnelling thresholds and distances, matrix’ coefficient of thermal expansion, and other factors. In our study, we add one more influencing factor–the degree of cure. Complexities of the curing process cause difficulties to predict, or even measure, the curing state of the polymer matrix while uncertainty in the degree of cure influences TCR measurements leading to biased values. Here we study the influence of the cure state on the TCR of a single-walled CNT/epoxy polymer nanocomposite. For the given degree of cure, TCR measurements are conducted in the temperature range 25–100 °C, followed by the next 24 h of post-curing and a new cycle of measurements, 8 cycles in total. We find that contrary to industry practice to expect a high degree of cure after 3 h at 130 °C, the curing process is far from reaching the steady state of the material and continues at least for the next 72 h at 120 °C, as we observe by changes in the material electrical resistivity. If TCR measurements are conducted in this period, we find them significantly influenced by the post-curing process continuing in parallel, leading in particular to non-monotonic temperature dependence and the appearance of negative values. The unbiased TCR values we observe only when the material reaches the steady state are no longer influenced by the heat input. The dependence becomes steady, monotonically increasing from near zero value at room temperature to 0.001 1/°C at 100 °C.

This research explores the influence of interfacial debonding between a broken fiber and matrix on stress redistribution surrounding a fiber break within a unidirectional (UD) impregnated fiber bundle, accounting for misalignment of fibers and fiber diameter distribution in randomly packed fiber configurations. Finite-element modelling is conducted on carbon-reinforced epoxy UD bundles with one fiber broken for different combinations of the bundle parameters: aligned/misaligned fibers and constant/randomly distributed fiber diameters. Two definitions of stress concentration factor (SCF) are examined, based on average and maximum stress over the fiber cross-section. The study reveals a statistically significant difference for average SCF for misaligned bundles with both constant and variable diameters of fibers compared to the case of aligned bundles of fibers with constant diameter. When the calculated SCFs are incorporated in a bundle strength model, the failure strain of unidirectional composites can be more realistically predicted.