Convolutional neural networks for approximating electrical and thermal conductivities of Cu-CNT composites.

This article explores the deep learning approach towards approximating the effective electrical and thermal conductivities of copper (Cu)-carbon nanotube (CNT) composites with CNTs aligned to the field direction. Convolutional neural networks (CNN) are trained to map the two-dimensional images of stochastic Cu-CNT networks to corresponding conductivities. The CNN model learns to estimate the Cu-CNT composite conductivities for various CNT volume fractions, interfacial electrical resistances, Rc = 20 Ω-20 kΩ, and interfacial thermal resistances, R″t,c = 10-10-10-7 m2K/W. For training the CNNs, the hyperparameters such as learning rate, minibatch size, and hidden layer neurons are optimized. Without iteratively solving the physical governing equations, the trained CNN model approximates the electrical and thermal conductivities within a second with the coefficient of determination (R2) greater than 98%, which may take longer than 100 min for a convectional numerical simulation. This work demonstrates the potential of the deep learning surrogate model for the complex transport processes in composite materials.

Cu2Se-based thermoelectric cellular architectures for efficient and durable power generation.

Thermoelectric power generation offers a promising way to recover waste heat. The geometrical design of thermoelectric legs in modules is important to ensure sustainable power generation but cannot be easily achieved by traditional fabrication processes. Herein, we propose the design of cellular thermoelectric architectures for efficient and durable power generation, realized by the extrusion-based 3D printing process of Cu2Se thermoelectric materials. We design the optimum aspect ratio of a cuboid thermoelectric leg to maximize the power output and extend this design to the mechanically stiff cellular architectures of hollow hexagonal column- and honeycomb-based thermoelectric legs. Moreover, we develop organic binder-free Cu2Se-based 3D-printing inks with desirable viscoelasticity, tailored with an additive of inorganic Se82- polyanion, fabricating the designed topologies. The computational simulation and experimental measurement demonstrate the superior power output and mechanical stiffness of the proposed cellular thermoelectric architectures to other designs, unveiling the importance of topological designs of thermoelectric legs toward higher power and longer durability.