Toward self-driving laboratory 2.0 for chemistry and materials discovery

Area of Science:

• Chemistry and Materials Science
• Laboratory Automation
• Artificial Intelligence

Background:

• Early self-driving laboratories (SDL 1.0) showed potential but faced limitations in scope and interoperability.
• The convergence of automation, AI, and data science drives the development of autonomous research platforms.
• Minimal human input is required for experiment design, execution, and analysis in these systems.

Purpose of the Study:

• To outline the vision and defining characteristics of the next generation of self-driving laboratories (SDL 2.0).
• To discuss recent advancements enabling more flexible, scalable, and collaborative scientific discovery engines.
• To propose SDL 2.0 as a transformative platform for reproducible experimentation and democratized research access.

Main Methods:

• Review of recent advances in modular hardware, AI decision-making (Bayesian optimization, computer vision, LLMs), and orchestration software.
• Integration of scheduling, data management, and safety protocols within autonomous laboratory systems.
• Proposal of six key characteristics for SDL 2.0: interoperable, collaborative, generalizable, orchestrated, safe, and creative.

Main Results:

• SDL 2.0 is envisioned as a new generation of discovery engines for chemistry and materials science.
• Key advancements include modular hardware, sophisticated AI decision-making, and integrated orchestration software.
• The proposed characteristics define SDL 2.0 as networked, reproducible, and accessible research infrastructure.

Conclusions:

• SDL 2.0 represents a significant leap forward from SDL 1.0, overcoming previous limitations.
• These advanced platforms will accelerate scientific innovation through reproducible experimentation and collaborative research.
• SDL 2.0 aims to democratize access to cutting-edge research infrastructure, transforming scientific discovery.