Artificial intelligence driven design of catalysts and materials for ring opening polymerization using a domain-specific language.

Advances in machine learning (ML) and automated experimentation are poised to vastly accelerate research in polymer science. Data representation is a critical aspect for enabling ML integration in research workflows, yet many data models impose significant rigidity making it difficult to accommodate a broad array of experiment and data types found in polymer science. This inflexibility presents a significant barrier for researchers to leverage their historical data in ML development. Here we show that a domain specific language, termed Chemical Markdown Language (CMDL), provides flexible, extensible, and consistent representation of disparate experiment types and polymer structures. CMDL enables seamless use of historical experimental data to fine-tune regression transformer (RT) models for generative molecular design tasks. We demonstrate the utility of this approach through the generation and the experimental validation of catalysts and polymers in the context of ring-opening polymerization-although we provide examples of how CMDL can be more broadly applied to other polymer classes. Critically, we show how the CMDL tuned model preserves key functional groups within the polymer structure, allowing for experimental validation. These results reveal the versatility of CMDL and how it facilitates translation of historical data into meaningful predictive and generative models to produce experimentally actionable output.

Simple and Efficient Synthesis of Functionalized Cyclic Carbonate Monomers Using Carbon Dioxide.

Aliphatic polycarbonates represent an important class of materials with diverse applications ranging from battery electrolytes, polyurethane intermediates, and materials for biomedical applications. These materials can be produced via the ring-opening polymerization (ROP) of six- to eight-membered cyclic carbonates derived from precursor 1,3- and 1,5-diols. These diols can contain a range of functional groups depending on the desired thermal, mechanical, and solution properties. Generally, the ring closure to form the cyclic carbonate requires the use of undesirable and hazardous reagents. Advances in synthetic methodologies and catalysis have enabled the use of carbon dioxide (CO2) to perform these transformations with a high conversion of diol to cyclic carbonate, yet modest isolated yields due to oligomerization side reactions. In this Letter, we evaluate a series of bases in the presence of p-toluenesulfonyl chloride and the appropriate diol to better understand their effect on the yield and presence of oligomer byproducts during cyclic carbonate formation from CO2. From this study, N,N-tetramethylethylenediamine (TMEDA) was identified as an optimal base, facilitating the preparation of a diverse array of both six- and eight-membered carbonates from CO2 within 10 to 15 min. The robust conditions for both, the preparation of the diol precursor, and the TMEDA-mediated carbonate synthesis enabled readily telescoping the two-step reaction sequence, greatly simplifying the process of monomer preparation.

Mechanistic Insight Leads to a Ligand Which Facilitates the Palladium-Catalyzed Formation of 2-(Hetero)Arylaminooxazoles and 4-(Hetero)Arylaminothiazoles.

By using mechanistic insight, a new ligand (EPhos) for the palladium-catalyzed C-N cross-coupling between primary amines and aryl halides has been developed. Employing an isopropoxy group at the C3-position favors the C-bound isomer of the ligand-supported palladium(II) complexes and leads to significantly improved reactivity. The use of a catalyst system based on EPhos with NaOPh as a mild homogeneous base proved to be very effective in the formation of 4-arylaminothiazoles and highly functionalized 2-arylaminooxazoles. Previously, these were not readily accessible using palladium catalysis.

Control of pore hydrophilicity in ordered nanoporous polystyrene using an AB/AC block copolymer blending strategy.

Ordered nanoporous plastics with hydrophilic pore surfaces were prepared by the degradative removal of polylactide from a self-organised, multi-component composite containing two block copolymers: polystyrene-polylactide and polystyrene-polyethylene oxide. The solid-state characterization of blends containing up to 12 wt.% polyethylene oxide was consistent with nanoscopic cylinders of mixed polyethylene oxide and polylactide hexagonally packed in a polystyrene matrix. Orientation of these materials through simple channel die processing resulted in good cylinder alignment. Subsequent methanolysis/hydrolysis of the polylactide component gave nanoporous polystyrene with polyethylene oxide coated pores. The resulting nanoporous materials were able to imbibe water, in contrast to nanoporous polystyrene with no polyethylene oxide component.