In-Line Molar Mass Determination in Semidilute Polymer Solutions

Integrating data science into materials discovery demands automated polymer synthesis and rapid, on-the-fly characterization to generate structure–property datasets for machine learning. A key bottleneck is molecular characterization at a pace commensurate with automated synthesis, limiting fully closed-loop “self-driving labs.” We develop a real-time method for determining weight-average molar mass (M_w) at concentrations typical of polymerization and processing. The approach exploits a theory linking specific viscosity of semidilute solutions to Mw via a correlation blob description, where polymers are treated as chains of correlation blobs. Measuring  over a range of concentrations and for polymers with systematically varied M_w generates a robust calibration curve. This approach enables M_w determination for unknown samples, independent of polymerization mechanism, sample dispersity, and over a wide concentration range. We demonstrate this method in a tubular reactor geometry by measuring viscosity of semidilute solutions of polystyrene in toluene and THF from a pressure drop across a capillary enabling a rapid, in-line determination of sample M_w. This capability advances high-throughput, closed-loop polymer discovery compatible with “big-data” workflows.