Researchers from North Carolina State University and Brown University have developed a microfluidic self-driving laboratory, termed PoLARIS (perovskite laboratory for autonomous reaction inference and synthesis), capable of rapidly optimizing and analyzing the synthesis of complex, multi-element perovskite nanocrystals. The platform addresses a key challenge in materials discovery: efficiently navigating the vast, high-dimensional parameter space associated with compositionally complex systems.
Self-driving laboratories combine automated experimentation with machine-learning-guided decision-making, but their application to materials with multiple coupled reaction pathways has remained limited. In this work, the researchers demonstrate that PoLARIS can autonomously synthesize and optimize metal halide double perovskite nanoplatelets containing up to six distinct elements, using a continuous-flow heat-up reaction.
The PoLARIS platform integrates a modular microfluidic reactor with closed-loop experiment selection. In practice, the system varies synthesis parameters - including precursor composition, ratios, temperature, and reaction time - across a sequence of experiments. Each experiment is conducted within a flowing microdroplet that acts as an individual reaction vessel, enabling highly time- and material-efficient screening.
The scale of the accessible chemical space is substantial, with billions of possible synthesis “recipes.” Traditional trial-and-error approaches are therefore prohibitively slow and often fail to capture complex parameter interactions. By contrast, PoLARIS executes iterative optimization cycles in which experimental results are analyzed in real time and used to guide subsequent experiments.
In a single 12-hour autonomous campaign, the system performed 120 experiments and successfully identified brighter, lead-free double perovskite nanoplatelets. These materials are of particular interest due to their tunable optical properties and potential use in applications such as photodetectors and solar fuel generation, while avoiding toxic heavy metals.
Beyond optimization, a key innovation of PoLARIS lies in its ability to extract mechanistic insight. Through dynamic flow experimentation, the platform probes precursor reactivity and reaction pathways governing nanoplatelet formation. This enables the construction of a data-driven map linking composition, reaction conditions, and optical performance.
As noted by North Carolina State University's Milad Abolhasani, “One of the big challenges in developing safer optical nanomaterials is the sheer size of the material universe.” He adds, “There are a vast number of possible combinations of ingredients, ratios, temperatures, and reaction environments that need to be explored to synthesize light-emitting nanoplatelets with the desired optical properties.”
Importantly, PoLARIS does not simply identify optimal conditions but also rationalizes them. “What is exciting about PoLARIS is that it does more than speed up trial and error,” Abolhasani says. “It learns from every experiment and builds a map of how chemistry, composition and temperature control material performance. That means we can discover promising materials faster, use fewer materials and understand why the best recipes work.”
The system is also inherently scalable. Once optimal synthesis conditions are identified, PoLARIS can transition from exploration to continuous production, effectively functioning as both a discovery engine and a micro-scale manufacturing platform. As Abolhasani describes it, “The beauty of PoLARIS is that it is both a GPS for materials discovery and a miniature materials factory.”
This work establishes microfluidic self-driving laboratories as a generalizable framework for unifying autonomous synthesis, optimization, and mechanistic understanding in complex colloidal systems. The PoLARIS approach may be extended to other multi-element and high-entropy nanomaterials, offering a scalable pathway toward accelerated discovery in next-generation energy and optoelectronic materials.
