Abstract:Materials with high-dielectric constant easily polarize under external electric fields, allowing them to perform essential functions in many modern electronic devices. Their practical utility is determined by two conflicting properties: high dielectric constants tend to occur in materials with narrow band gaps, limiting the operating voltage before dielectric breakdown. We present a high-throughput workflow that combines element substitution, ML pre-screening, ab initio simulation and human expert intuition to efficiently explore the vast space of unknown materials for potential dielectrics, leading to the synthesis and characterization of two novel dielectric materials, CsTaTeO6 and Bi2Zr2O7. Our key idea is to deploy ML in a multi-objective optimization setting with concave Pareto front. While usually considered more challenging than single-objective optimization, we argue and show preliminary evidence that the $1/x$-correlation between band gap and permittivity in fact makes the task more amenable to ML methods by allowing separate models for band gap and permittivity to each operate in regions of good training support while still predicting materials of exceptional merit. To our knowledge, this is the first instance of successful ML-guided multi-objective materials optimization achieving experimental synthesis and characterization. CsTaTeO6 is a structure generated via element substitution not present in our reference data sources, thus exemplifying successful de-novo materials design. Meanwhile, we report the first high-purity synthesis and dielectric characterization of Bi2Zr2O7 with a band gap of 2.27 eV and a permittivity of 20.5, meeting all target metrics of our multi-objective search.

What problem does this paper attempt to address?

This paper focuses on the discovery of high dielectric constant materials widely used in electronic devices. The practical value of these materials lies in their ability to easily polarize under an external electric field, but high dielectric constants often coexist with narrow band gaps (which limit the operating voltage) leading to dielectric breakdown. The research team proposed a high-throughput workflow combining element substitution, machine learning (ML) pre-screening, initial value simulation, and human expert intuition to effectively explore the unknown material space and identify potential dielectric materials. They specifically used a multi-objective optimization approach, utilizing ML to predict materials with excellent performance even when balancing multiple conflicting properties such as band gap and dielectric constant. The research team synthesized two new dielectric materials, CsTaTeO6 and Bi2Zr2O7, using this approach. CsTaTeO6 was generated through element substitution to create a new structure, while Bi2Zr2O7 achieved high purity synthesis and dielectric characterization for the first time, with a band gap of 2.27 eV and a dielectric constant of 20.5. Experimental validation showed that these two materials partially or fully met the target indicators of the multi-objective search. The paper indicates that although ML may not be as reliable as traditional methods in single-objective optimization tasks, its application in multi-objective optimization can reduce the need for extrapolation and operate within well-supported regions, thus predicting materials with excellent performance. Furthermore, the challenges of experimental synthesis and characterization remain a key bottleneck in material design, but ML-guided screening can improve the hit rate of successful materials and reduce the risk of trying new synthesis procedures. Overall, this paper demonstrates the potential of ML in the discovery of dielectric materials, enabling efficient navigation through vast material spaces and balancing multiple material properties to achieve optimal device performance.

Pushing the Pareto front of band gap and permittivity: ML-guided search for dielectric materials

High Dielectric Ternary Oxides from Crystal Structure Prediction and High-throughput Screening

Identification of High-Dielectric Constant Compounds from Statistical Design

Assessing data-driven predictions of band gap and electrical conductivity for transparent conducting materials

Giant permittivity in Nb-doped SrTiO3 single crystal: Compositional gradient and local structure

Band engineering and rational design of high-performance thermoelectric materials by first-principles

Advances in dielectric performance of atomically engineered Sr1.8Bi0.2Nb3O10 perovskite nanosheet thin films

Accelerated Search for BaTiO3‐Based Ceramics with Large Energy Storage at Low Fields Using Machine Learning and Experimental Design

Evolutionary search for new high-k dielectric materials: methodology and applications to hafnia-based oxides

Ternary Wide Band Gap Oxides for High-Power Electronics Identified Computationally

High-refractive-index materials screening from machine learning and ab initio methods

First-principles design and subsequent synthesis of a material to search for the permanent electric dipole moment of the electron

Compositional engineering of perovskites with machine learning

Discovering ABO3-type perovskite with different dielectric constants via intelligent optimization algorithm

Candidate ferroelectrics via ab initio high-throughput screening of polar materials

Enhanced tunable dielectric properties of Ba0.6Sr0.4TiO3/PVDF composites through dual-gradient structural engineering

Polymorphic Localized Heterostructure Design for High-Performance Amorphous/Nanocrystalline Composite Film

Polarons, free charge localisation and effective dielectric permittivity in oxides

Computational framework for polymer synthesis to study dielectric properties using polarizable reactive molecular dynamics

Defect Engineering in Two Common Types of Dielectric Materials for Electromagnetic Absorption Applications

Microwave Dielectric Materials with Defect-Dipole Clusters Induced Colossal Permittivity and Ultra-low Loss