AI-assisted Discovery of New Materials for Energy Storage

About Drug Repurposing

Electrostatic capacitors are critical energy storage components in advanced electrical systems in defense, aerospace, energy and transportation. Although electrostatic capacitors provide extremely high power densities, existing dielectric materials, which degrade rapidly at elevated temperatures, limit the energy density of the capacitors. Most industrial-grade polymer dielectric materials are flexible polyolefins or rigid aromatic compounds that offer either high energy density or high thermal stability, but not both.

The research literature we cite here describes an AI-driven method called polyVERSE for generating and screening polymeric materials with excellent high-temperature dielectric properties.

Highlights of the Research

It is well known that the functionality of polymers is largely governed by their chemical composition, and AI's ability to quickly process vast amounts of data beyond human imagination can help discover new materials with high performance.

The research team's polyVERSE approach consists of four steps: chemical structure generation, property prediction, screening of the best candidates, and synthesis and characterization of the selected candidates.

First, the researchers used an expert system to generate polymers from commercially available monomers. Then, graph neural networks (GNNs) were used to evaluate the properties of these materials. These property predictions were used to screen out promising polymers from a large number of candidate materials.

Through this approach, the researchers identified a number of dielectrics that exhibit high thermal stability and high energy density over a wide temperature range. One of them, a new polynorbornene dielectric material called PONB-2Me5Cl, has a high energy density over a wide temperature range. At 200 ℃, the polymer achieves an energy density of 8.3 J/cc, which is more than 10 times that of any commercial alternative, making it one of the best polymer dielectric materials reported to date at this temperature.

Multitask Graph Neural Networks.Fig. 1. Structure-property models are trained using multitask GNNs. (Gurnani R.; et al. 2024)

Material Characterization

PONB-2Me5Cl's success is attributed to its excellent combination of glass transition temperature, band gap and dielectric constant properties. The polymer's high glass transition temperature ensures mechanical stability at elevated temperatures, while its high band gap acts as a significant barrier to electronic conductivity, leading to unprecedented breakdown field strengths. In addition, PONB-2Me5Cl has a modest, thermally stable dielectric constant, which leads to high energy densities from room temperature up to 200 ℃.

The researchers also evaluated the environmental impact of the proposed materials, emphasizing the importance of lightweighting and chemical synthesis. The high energy density and thermal stability of PONB-2Me5Cl eliminate the need for a cooling system for the capacitor, resulting in a reduction in weight and size. In addition, the researchers explored the possibility of using green solvents for synthesis and proposed several candidate materials for polyimides with potentially high performance.

These important findings by the researchers extend the potential applications of electrostatic capacitors in the 85-200 ℃ temperature range and highlight the power of AI in producing advanced polymer dielectrics with excellent energy storage capabilities.

This article demonstrates the great potential of AI in materials science, particularly in accelerating the discovery of high-temperature dielectric materials. Through the polyVERSE approach, researchers significantly increased the number of known high-temperature dielectric materials and provided new avenues for further research and applications of the materials.

Our company offers AI-assisted new material discovery services designed to identify and develop materials with specific properties, with specific solutions including:

  • Advanced catalyst screening
  • High performance polymer discovery
  • Nanomaterial development

If you are interested in our services or have a question, please feel free to contact us for more details.

Original Article:

Gurnani R.; et al. (2024). AI-assisted discovery of high-temperature dielectrics for energy storage. Nature communications. 2024, 15(1): 6107.

Services Related in the Article:

Discovering New Materials

Inquiry

logo

We harness the power of artificial intelligence to transform the landscape of drug discovery. Our mission is to accelerate the development of life-saving medicines by leveraging cutting-edge AI technologies.

CONTACT US
  • Tel:
  • E-mail:
  • Address:
Top