Tunable and scalable production of nanostar particle platforms for diverse applications using an AI-integrated automated synthesis system.

The tunable optical properties and exceptional electromagnetic field enhancement of nanostar-based plasmonic nanoparticles have garnered much attention for use in a wide array of biomedical applications. However, a great challenge for their widespread use is the time-sensitive nature of nanostar synthesis, which could lead to unprecise control of their homogeneity and high batch-to-batch variability. We have developed an automated synthesis system with AI capability to reproducibly synthesize large quantities of nanostar particles. Using this platform, synthesis parameters such as reagent volume and reagent addition timing can be varied to observe how these factors determine the optical properties and SERS enhancement of gold nanostars and bimetallic nanostars, which are used as two model systems. An artificial intelligence (AI) system based on two tree-based machine learning models was developed and trained using nanoparticle characterization data to predict absorbance features and SERS enhancement from synthesis parameters. A grid matrix was fed into the final trained models to create a lookup table to synthesize gold nanostars with an absorbance maximum at specific wavelengths, culminating in the reproducible synthesis of desired nanostar platforms with a peak absorbance wavelength of less than 1.2% difference compared to the target peak absorbance. This machine learning-integrated automated synthesis platform has the potential to enable the next phase of investigation for nanostar-based technologies and expand the scope of their current applications.