Human-AI collaborative autonomous synthesis with pulsed laser deposition for remote epitaxy

2025-11-14

AI Summary

This study addresses the dual challenges of growth optimization and graphene preservation in pulsed laser deposition (PLD) of BaTiO₃/graphene remote epitaxial heterostructures. Methodologically, we establish a tightly coupled human–AI collaborative autonomous experimental workflow: a large language model (LLM) drives scientific hypothesis generation, experimental planning, and in situ Raman analysis, while a dynamically updated collaboration protocol enables real-time integration of human expert knowledge and AI reasoning across experimental batches—overcoming the latency limitations of conventional “human-in-the-loop” systems. Key contributions include: (i) the first closed-loop intelligent exploration of PLD process parameter space; (ii) efficient construction of a graphene-damage versus film-growth phase diagram, revealing the trade-off between graphene integrity and BaTiO₃ crystallinity within low-oxygen, low-pressure, and low-temperature regimes; and (iii) the proposal and experimental validation of a two-step Ar/O₂ deposition strategy enabling successful exfoliation of ferroelectric BaTiO₃ films while fully preserving the monolayer graphene interlayer.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Autonomous laboratories typically rely on data-driven decision-making, occasionally with human-in-the-loop oversight to inject domain expertise. Fully leveraging AI agents, however, requires tightly coupled, collaborative workflows spanning hypothesis generation, experimental planning, execution, and interpretation. To address this, we develop and deploy a human-AI collaborative (HAIC) workflow that integrates large language models for hypothesis generation and analysis, with collaborative policy updates driving autonomous pulsed laser deposition (PLD) experiments for remote epitaxy of BaTiO$_3$/graphene. HAIC accelerated the hypothesis formation and experimental design and efficiently mapped the growth space to graphene-damage. In situ Raman spectroscopy reveals that chemistry drives degradation while the highest energy plume components seed defects, identifying a low-O$_2$ pressure low-temperature synthesis window that preserves graphene but is incompatible with optimal BaTiO$_3$ growth. Thus, we show a two-step Ar/O$_2$ deposition is required to exfoliate ferroelectric BaTiO$_3$ while maintaining a monolayer graphene interlayer. HAIC stages human insight with AI reasoning between autonomous batches to drive rapid scientific progress, providing an evolution to many existing human-in-the-loop autonomous workflows.

Problem

• Developing human-AI collaboration for autonomous pulsed laser deposition experiments
• Mapping growth conditions to prevent graphene damage during remote epitaxy
• Identifying synthesis windows balancing BaTiO3 growth and graphene preservation

Innovation

• Human-AI collaborative workflow with large language models
• Autonomous pulsed laser deposition for remote epitaxy
• Two-step Ar/O2 deposition preserving graphene interlayer