In a groundbreaking development poised to revolutionize materials science, researchers at North Carolina State University have unveiled a pioneering autonomous laboratory system capable of navigating an immense universe of material synthesis recipes with unprecedented speed and precision. Named PoLARIS—short for perovskite laboratory for autonomous reaction inference and synthesis—this self-driving microfluidic lab excels at rapidly synthesizing and optimizing lead-free, light-emitting nanomaterials known as double perovskite nanoplatelets. Remarkably, PoLARIS completed a targeted discovery campaign in just 12 hours, a process that conventionally might take years, demonstrating a transformative leap in material discovery and optimization.
Double perovskite nanoplatelets belong to an intriguing class of two-dimensional nanocrystals characterized by their extraordinary thinness—mere billionths of a meter thick—and their tunable optical properties. The materials under study are lead-free, a critical advancement given growing concerns over the toxicity of lead in conventional perovskites. By precisely adjusting atomic compositions within this family of materials, scientists can modulate light absorption and emission to design safer, brighter, and more efficient optoelectronic devices, ranging from photodetectors to solar fuel production systems.
Milad Abolhasani, the Alcoa Professor and University Faculty Scholar who spearheaded the project, emphasized the staggering complexity inherent to developing such materials. “The materials universe is virtually infinite,” he stated, highlighting the daunting combination of factors including chemical components, stoichiometric ratios, reaction temperatures, and environmental conditions that define the synthesis process. Conventionally, mastering these multidimensional parameters has relied on laborious trial-and-error methods prone to missing subtle, yet crucial, interactions between variables.
What sets PoLARIS apart is its integration of artificial intelligence with a sophisticated microfluidic platform, enabling high-throughput, automated experiments within minuscule flowing droplets that serve as microscale reaction vessels. Each experiment tunes reaction parameters, such as precursor concentrations, temperature settings, and reaction durations, to iteratively refine nanoplatelet synthesis. The system autonomously analyzes the resulting photoluminescence—a direct measure of brightness—and feeds the data back into machine learning algorithms that strategize subsequent experimental recipes. This self-correcting loop makes it possible to rapidly converge on optimal material formulations while economizing on reagents and time.
During a single operating session spanning just half a day, PoLARIS conducted 120 discrete experiments, systematically elevating the photoluminescence intensity of the double perovskites. This rapid-fire experimental approach uncovered best-in-class formulations that meet the crucial “safer” criterion of being free from lead and other heavy metals, without compromising performance. Such acceleration in discovery not only propels safer nanomaterials to market readiness but also opens avenues for environmentally benign applications in next-generation photonic and energy devices.
Beyond its impressive throughput and optimization prowess, PoLARIS shines in its interpretative capabilities. Unlike many AI systems that deliver optimized outcomes as opaque “black box” solutions, PoLARIS constructs mechanistic maps elucidating the interplay between chemical reactivity, compositional changes, and thermal conditions that govern nanoplatelet formation. This scientific insight deepens understanding of the fundamental processes, fashioning a new paradigm where AI-aided discovery transcends mere trial and error to become a tool for exploratory science and knowledge generation.
The modular design of PoLARIS enhances its versatility and scalability. It functions not only as an autonomous materials discovery engine but also seamlessly transitions into a miniature continuous-flow manufacturing setup, capable of producing optimized materials at scale without human intervention. Such dual functionality bodes well for expediting commercialization pathways of emergent nanomaterials, shortening the timeline from lab discovery to industrial application.
PoLARIS serves as a proof-of-concept for the broader vision of autonomous laboratories that couple real-time experimental feedback with intelligent experiment planning, thereby navigating high-dimensional, multi-parameter chemical landscapes at speeds impossible for traditional discovery approaches. As materials science increasingly confronts the complexities of multi-element, compositionally intricate substances—like high-entropy colloidal nanocrystals—autonomous platforms like PoLARIS will be crucial for efficient exploration and innovation.
This work marks a significant milestone in the integration of microfluidics, machine learning, and materials chemistry. The continuous-flow heat-up synthesis employed within PoLARIS permits precise thermal management and reaction scaling, while dynamic flow experimentation facilitates probing of precursor interactions and reaction pathways. Combining these elements in a closed-loop autonomous framework paves the way for a new era of mechanistically informed, accelerated chemical synthesis.
The research team, including a cadre of Ph.D. candidates from NC State and collaborators from Brown University, plans to further generalize this approach to encompass other nanocrystal systems beyond the double perovskite family. They envision a future where AI-guided, self-driving labs become indispensable tools in the rapid generation of functional materials designed for energy, electronics, and sustainability applications.
Published in the prestigious journal Nature Communications in May 2026, the study represents a collaboration across disciplines and institutions, funded by the National Science Foundation. It signals a paradigm shift in how complex optical nanomaterials are discovered, optimized, and manufactured—a fusion of human creativity and machine intelligence poised to accelerate technological breakthroughs in the years ahead.
As Milad Abolhasani aptly summarized, PoLARIS is both a GPS and a factory in one unified platform: it charts unexplored chemical territories, deciphers the underlying reasons why specific synthesis paths succeed, and then produces the material continuously. This convergence of autonomous technology with the scientific method embodies the future of materials discovery, promising transformative impacts on science and industry alike.
Subject of Research:
Not applicable
Article Title:
“Autonomous microfluidic experimentation for exploring reaction interference and synthesizing double perovskite nanoplatelets”
News Publication Date:
4-May-2026
Web References:
DOI: 10.1038/s41467-026-72765-2
References:
Li, J., Delgado-Licona, F., Perry, H., Xu, J., Mukhin, N., Sadeghi, S., Abolhasani, M., Liu, Z., & Chen, O. (2026). Autonomous microfluidic experimentation for exploring reaction interference and synthesizing double perovskite nanoplatelets. Nature Communications, DOI:10.1038/s41467-026-72765-2
Image Credits:
Not provided
Keywords
Autonomous laboratory, PoLARIS, double perovskites, nanoplatelets, microfluidics, machine learning, materials discovery, lead-free nanomaterials, photoluminescence, continuous-flow synthesis, reaction optimization, nanochemistry
