Innovative Materials Science via Machine Learning

Abstract

Nowadays, the research on materials science is rapidly entering a phase of data-driven age. Machine learning, one of the most powerful data-driven methods, have been being applied to materials discovery and performances prediction with undoubtedly tremendous application foreground. Herein, the challenges and current progress of machine learning are summarized in materials science, the design strategies are classified and highlighted, and possible perspectives are proposed for the future development. It is hoped this review can provide important scientific guidance for innovating materials science and technology via machine learning in the future.

Temperature-Dependent Growth of 36 Inner Nanotubes inside Nickelocene, Cobaltocene and Ferrocene-Filled Single-Walled Carbon Nanotubes

“…The filling of SWCNTs with metallocene molecules opens the way of n-doping of the nanotubes, and the further annealing of the filled SWCNTs allows modifying the electronic properties of SWCNTs in a tailored manner. Combining the controlled growth kinetics of inner SWCNTs [21,29,30] and tailored electronic properties of these nanostructures [42,43] allows applying these systems in various fields, such as nanoelectronics, thermoelectric power generation, catalysis, sensors, electrochemical energy storage, spintronics, magnetic recording and biomedicine [44][45][46][47].…”

Studying Complex Evolution of Hyperelastic Materials under External Field Stimuli using Artificial Neural Networks with Spatiotemporal Features in a Small‐Scale Dataset

“…Therefore, building a high-quality DL prediction model from small-scale experimental data is a universally beneficial strategy to promote the development of DL in materials science. [18,19] In this work, we take the nonlinear stress and strain behavior of hyperelastic materials as the focus of our research to achieve high-precision DL prediction models based on a small sample space. This example is of general interest in material mechanical experiments because it can be regarded as a universal case in the mechanical response process of materials whose spatial structure features continuously evolve under sustained external stimuli.…”

Pd Doped Co3O4 Loaded on Carbon Nanofibers as Highly Efficient Free-Standing Electrocatalyst for Oxygen Reduction and Oxygen Evolution Reactions

“…Besides, the intrinsic electrocatalyst activity is also significant to manufacture high performance electrodes ( Li et al, 2020 ; Xiao et al, 2021 ). Cobalt oxide has been studied extensively, because of their sustainability against corrosion, outstanding redox capability, distinct 3 days electron orbitals, and superior theoretical activity ( Xiang et al, 2018 ; Yu et al, 2018 ; Huang et al, 2020b ; Hou et al, 2020 ; Gao et al, 2021 ; Liu et al, 2021 ). Although there are many studies on cobalt oxide nanofibers ( Huang et al, 2019a ; Li et al, 2019b ), it is still facing great challenges to directly use it as a self-supporting electrode because of its lack of mechanical strength.…”