(88b) Solution Coating of Polymer Blend Libraries for High-Throughput Experimentation Via Passive Mixing

As many research methodologies traditionally demand a considerable investment in time and resources, materials informatics approaches have emerged as a key development focus in enabling accelerated materials discovery. Such approaches, which often leverage machine learning and advanced data analytics techniques, require sufficiently large experimental datasets. In some materials systems, high-throughput experimentation methods can potentially bridge this small data gap by accelerating the curation of laboratory data. For high-throughput experimentation of polymer materials, one protocol is to enable the deposition of thin-film libraries that can be rapidly characterized within the parameter space of interest. However, the need to overcome unfavorable solubility constraints at ambient conditions, high viscosities, and other processing challenges can impede the ability to generate such polymer libraries. To overcome these obstacles, this work demonstrates an automated flow system capable of operation under aggressive solvent conditions up to 120°C for the solution casting of polymer thin-films. The system includes a custom-designed passive mixer suitable for effective mixing of polymer solutions under solvent use or elevated temperatures. Residence time distribution modeling was leveraged to successfully inform flow conditions and generate composition gradient films based on poly(3-hexylthiophene), poly(styrene), and poly(propylene). These composition libraries were interrogated using high-throughput property measurements and mapped to spatially resolved composition measurements. Results display the feasibility of our gradient film approach as a key component within informatics-driven workflows. Integration of data science techniques on such curated data can facilitate the discovery and development of high-performance polymer coatings, plastics, and clean energy materials, all of which often display the processing challenges that are traditionally difficult for high throughput experimentation.