Composing Nanomaterials – with AI and Chemistry
Researchers developed a tool that combines automated chemical synthesis, high-throughput characterization, and data-driven modeling
Advertisement
LMU researcher Professor Alexander Urban and his team have developed a tool that could revolutionize the design of new materials. Synthesizer is a platform that combines automated chemical synthesis, high-throughput characterization, and data-driven modeling. The goal is to control the growth of nanocrystals with unprecedented precision, thereby creating materials with tailor-made optical properties. The results of their work, funded by the e-conversion Excellence Cluster, have now been published by the LMU team in Advanced Materials.
Unlike previous data-driven approaches, Synthesizer is the first platform to connect the entire chain from automated synthesis and optical high-throughput characterization to the AI-supported derivation of concrete design rules within an open and modular system. “Today, we can compose material properties almost like a melody, note by note, parameter by parameter,” says Alexander Urban. That is exactly what Synthesizer enables. Using the platform, variants of halide perovskites can be produced and characterized automatically, while an AI model learns which chemical combinations lead to specific colors, brightness levels or stabilities.
The optical properties of halide perovskites such as color, brightness or emission width determine their use in LEDs, solar cells or sensors. “Even the smallest differences in nanocrystal size, shape and structure can shift light emission,” explains Nina Henke, first author and doctoral researcher in Urban’s team. “Fine-tuning is therefore essential in order to develop materials that are precisely tailored to specific applications.”
A turbo boost for the development of halide perovskites
What makes Synthesizer special is that the platform is open, flexible, and expandable. It was originally developed for halide perovskites but is, in principle, suitable for other material classes as well. In the future, researchers will be able to automate syntheses, systematically vary parameters and generate valuable data sets in a very short time. The AI model then translates this data into concrete design rules. In the journal article, the researchers not only present the concept but also release Synthesizer as a freely available and modularly adaptable platform.
“Our goal is to accelerate materials research and enable precise predictions,” says Alexander Urban. “This makes it possible to create crystals with specifically tuned optical and physical properties and to further advance optoelectronics and photonics.” The synthesizer platform is compatible with existing systems for automated synthesis. The LMU team is currently working on integrating its development into the laboratory routine.
Original publication
Nina A. Henke, Leo Luber, Ioannis Kouroudis, Jonathan Paul, Alexander Schuhbeck, Lukas M. Rescher, Tizian Lorenzen, Veronika Mayer, Knut Müller‐Caspary, Bert Nickel, Alessio Gagliardi, Alexander S. Urban; "Synthesizer: Chemistry‐Aware Machine Learning for Precision Control of Nanocrystal Growth"; Advanced Materials, 2025-11-5
Other news from the department science
Most read news
More news from our other portals
See the theme worlds for related content
Topic world Synthesis
Chemical synthesis is at the heart of modern chemistry and enables the targeted production of molecules with specific properties. By combining starting materials in defined reaction conditions, chemists can create a wide range of compounds, from simple molecules to complex active ingredients.
Topic world Synthesis
Chemical synthesis is at the heart of modern chemistry and enables the targeted production of molecules with specific properties. By combining starting materials in defined reaction conditions, chemists can create a wide range of compounds, from simple molecules to complex active ingredients.