(474d) High-Throughput Experimentation and Compositional Screening of Polymer-Based Semiconductor/Insulator Blends for Organic Device Applications

Conjugated polymer semiconductors are promising materials for their imminent potential in enabling stretchable, flexible, and deformable device applications, such as in transistors, photovoltaics, and others. Recently, polymer semiconductor/insulator blends have gained popularity as a formulation method to afford mechanical properties without sacrificing device performance. For example, when insulating polymers are blended with the active semiconducting layer of organic field-effect transistors (OFETs), the charge-carrier mobility (the main device metric) can match or even exceed devices made with only the single semiconducting component, even at very low compositions of the semiconductor. Crucial to understanding this behavior is interrogating the interplay between solution processing parameters, the thin-film morphology, structural measurements, and device metrics within a rich process-structure-property space that can be time-consuming to measure. Unfortunately, capturing the exact location of behavior onsets, phase boundaries, and other transitions associated with improved performance can be challenging as they take place at very sensitive windows. The number of experiments required to fully understand the process-structure-property relationships quickly multiplies when subsequently exploring this composition-performance trend with respect to different polymers/solvents, or incorporating solution processing steps (e.g., UV-irradiation, aging, etc.). In this work, to aid in traversing the rich compositional landscape of polymer semiconductor/insulator blends, we incorporate a gradient film methodology through an automated flow system as a means of high-throughput experimentation. The screening of the P3HT/PS and DPP-DTT/PS composition libraries presented herein represents how the gradient film approach can facilitate preliminary exploration of process-structure-property relationships in polymer thin-film systems. The large amount of data generated from the gradient film libraries can then be integrated with data science techniques (e.g. machine learning) to model the relationship between polymer composition and OFET device performance. By leveraging the techniques reported herein, we envision the high throughput experimental screening approach to be broadly applicable to other solution-based thin-film systems, including organic device applications and beyond.