Machine Learning-Assisted Discovery of High-Voltage Organic Materials for Rechargeable Batteries

Abstract

Organic redox compounds are rich in elements and structural diversity, which are an ideal choice for lithium-ion batteries. However, most organic cathode materials show a trade-off between specific capacity and voltage, limiting energy density. By increasing the redox potential of cathode materials, the balance between redox potential and specific capacity can be broken to increase energy density. In this work, we use machine learning to train materials with different redox potentials to predict novel polymers with ideal potentials. In situ computer vision and infrared spectroscopy monitor the reaction in real time. We also theoretically studied the concentrationdependent yields by providing a depletion-force model. This work provides a new solution to material research flow, including training, prediction, synthesis, examination, and analysis, accelerating high-capacity organic cathode material discovery.

Machine Learning-Assisted Development of Organic Solar Cell Materials: Issues, Analyses, and Outlooks

“…Moreover, the synthesis and characterization of π-conjugated materials as well as their OPV device fabrication require considerable time and effort, which has precluded rapid exploration in the large molecular space. In contrast, ML can deal with the design of a complex OPV system at a remarkably fast speed, even though there may be a lack of established causality and universal theory. − In addition, automated combinatorial synthesis and characterization are compatible with ML to accelerate the experimental screening. , …”

“…30−36 In addition, automated combinatorial synthesis and characterization are compatible with ML to accelerate the experimental screening. 37,38 In this Perspective, we first introduce the present status of ML-assisted development of OPVs through a literature survey and statistics. Second, we discuss the general issues in applying ML for the development of OPVs, that is, the amount of data and explanatory variables needed.…”

Advanced Machine Learning Methods for Learning from Sparse Data in High-Dimensional Spaces: A Perspective on Uses in The Upstream of Development of Novel Energy Technologies

“…These examples from the fields catalysis and solar cells are barely illustrative. Machine learning (and the same methods as mentioned above) is also used to help design battery materials [23,24,67,68] and for other energy technologies as indicated above, but ML is in demand not just at the material or device level but also at the system level [69][70][71][72]. In an energy mix containing solar farms, wind farms, and other intermittent technologies alongside more established generation methods such as nuclear or natural gas powered stations, one needs to constantly balance supply and demand.…”

“…One area where ML is gaining more and more traction are novel energy conversion and storage technologies. These techniques are, in particular, intensely explored for application to the development of technologies typically associated with sustainable generation and use of energy such as advanced types (organic and inorganic materials based) of solar cells and LED (light-emitting diodes) [10][11][12][13][14][15][16][17][18][19][20][21][22], inorganic and organic metal ion batteries [23,24], fuel cells and generally heterogeneous catalysis including electro-and photocatalysis [25][26][27][28][29][30][31][32][33][34]. This is natural in the sense that the development of these technologies often passes through optimization and balancing of multiple factors acting simultaneously and to opposite ends; for example, in the case of organic solar cells, there is an optimum to be sought between the donor's bandgap, the band offset between the donor and the acceptor, the reorganization energies of both the donor and the acceptor, the charge transfer integral etc.…”

N‐Substituted Carbazole Derivate Salts as Stable Organic Electrodes for Anion Insertion

“…In a second paper, Xu and coworkers employed a machine learning algorithm to predict the electrochemical properties of different carbazole derivates. [23] From first principles, the authors linked the reversibility of the electron exchange process with the nature of the substituent of the nitrogen atom. An electronwithdrawing substituent can stabilize the structure with a consequent increase of redox potential of carbazole.…”