Real-time Multi-instrument Autonomous Discovery of Novel Phase-change Memory Materials

2026-05-18

AI Summary

This work addresses the limitations of traditional materials discovery, which relies on post-experimental analysis and struggles to integrate heterogeneous, asynchronous data from multiple instruments for real-time decision-making. To overcome these challenges, the authors propose a Multi-instrument Autonomous Discovery (MAD) framework that, for the first time, enables concurrent optimization of both structural and functional objectives within a closed-loop experimental paradigm. By sharing uncertainty information across tasks, MAD facilitates parallel decision-making and breaks away from conventional sequential experimentation. The approach integrates X-ray diffraction and resistivity measurements through multi-output Gaussian processes with a co-kriging kernel, non-negative matrix factorization, and Bayesian optimization. Demonstrated in the discovery of high-performance phase-change memory materials, the method identifies optimal candidates within 25 iterations—revealing synthesis–structure–property relationships and achieving a sevenfold improvement in efficiency over traditional approaches.

Abstract

Autonomous labs enable the integration of automated experiment execution, data analysis and decision making. The main challenge remains the integration of diverse data streams from multiple instruments, where the data is often heterogeneous and unsynchronized. The standard learning process of undetermined synthesis-process-structure-property relationships (SPSPR) usually relies on post-experiment analysis after data is fully collected, not during live experiments, and decision making is carried out independently across characterization equipment. Here, we demonstrate the Multi-instrument Autonomous Discovery (MAD) framework -- combining structural property mapping and functional property optimization simultaneously in a closed-loop manner. As an example, we applied MAD to phase change memory (PCM) materials, and, in particular on the Mn-Sb-Te ternary, a previously unexplored materials system for PCM. A multi-output model is employed to merge data from x-ray diffraction (XRD) and electrical resistance measurements simultaneously through a co-regionalization kernel that models the relationship between them. The output probabilistic posterior and uncertainty quantification facilitate decision making with shared knowledge, while the goals are different across tasks. We aimed to maximize the knowledge of crystal structure distribution using non-negative matrix factorization (NMF), while in parallel, we find the composition with the maximum resistance value, an important figure of merit for PCM. Leveraging MAD, we found promising electrical PCMs and identified the SPSPR within 25 closed-loop iterations, corresponding to a seven-fold speed-up. The framework opens a new path of study in large-scale autonomous facilities, where future experiments can be run in parallel together, not independently.

Problem

• autonomous discovery
• multi-instrument integration
• phase-change memory
• heterogeneous data
• real-time decision making

Innovation

• Multi-instrument Autonomous Discovery
• co-regionalization kernel
• phase-change memory
• closed-loop experimentation
• synthesis-process-structure-property relationships