Research

Paper

AI LLM March 10, 2026

AI-driven Inverse Design of Complex Oxide Thin Films for Semiconductor Devices

Authors

Bonwook Gu, Trinh Ngoc Le, Wonjoong Kim, Zunair Masroor, Han-Bo-Ram Lee

Abstract

Bridging generative foundation models with non-equilibrium thin-film synthesis remains a central challenge, limiting the practical impact of AI-driven materials discovery on semiconductor dielectrics. Here, we introduce IDEAL (Inverse Design for Experimental Atomic Layers), an inverse-design platform that links generative diffusion models, machine learning interatomic potentials, and graph neural network property predictors with atomic layer deposition (ALD). We demonstrate IDEAL using the Hf-Zr-O system as a stringent benchmark for semiconductor-relevant complex oxides. The platform statistically enumerates thermodynamically plausible structures and constructs a composition-structure-property map. Crucially, it identifies a narrow composition window where low-energy tetragonal and orthorhombic phases cluster, revealing trade-offs between band gap and dielectric response. Experimental validation using atomic layer modulation (ALM) corroborates these predictions, demonstrating predictive guidance under realistic, non-equilibrium thin-film growth. By experimentally closing the loop, IDEAL provides a transferable and generalizable route to the precision synthesis of next-generation semiconductor dielectrics.

Metadata

arXiv ID: 2603.09744
Provider: ARXIV
Primary Category: cond-mat.mtrl-sci
Published: 2026-03-10
Fetched: 2026-03-11 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.09744v1</id>\n    <title>AI-driven Inverse Design of Complex Oxide Thin Films for Semiconductor Devices</title>\n    <updated>2026-03-10T14:48:35Z</updated>\n    <link href='https://arxiv.org/abs/2603.09744v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.09744v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Bridging generative foundation models with non-equilibrium thin-film synthesis remains a central challenge, limiting the practical impact of AI-driven materials discovery on semiconductor dielectrics. Here, we introduce IDEAL (Inverse Design for Experimental Atomic Layers), an inverse-design platform that links generative diffusion models, machine learning interatomic potentials, and graph neural network property predictors with atomic layer deposition (ALD). We demonstrate IDEAL using the Hf-Zr-O system as a stringent benchmark for semiconductor-relevant complex oxides. The platform statistically enumerates thermodynamically plausible structures and constructs a composition-structure-property map. Crucially, it identifies a narrow composition window where low-energy tetragonal and orthorhombic phases cluster, revealing trade-offs between band gap and dielectric response. Experimental validation using atomic layer modulation (ALM) corroborates these predictions, demonstrating predictive guidance under realistic, non-equilibrium thin-film growth. By experimentally closing the loop, IDEAL provides a transferable and generalizable route to the precision synthesis of next-generation semiconductor dielectrics.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cond-mat.mtrl-sci'/>\n    <published>2026-03-10T14:48:35Z</published>\n    <arxiv:comment>25 pages, 7 figures</arxiv:comment>\n    <arxiv:primary_category term='cond-mat.mtrl-sci'/>\n    <author>\n      <name>Bonwook Gu</name>\n    </author>\n    <author>\n      <name>Trinh Ngoc Le</name>\n    </author>\n    <author>\n      <name>Wonjoong Kim</name>\n    </author>\n    <author>\n      <name>Zunair Masroor</name>\n    </author>\n    <author>\n      <name>Han-Bo-Ram Lee</name>\n    </author>\n  </entry>"
}