PolyID: Artificial Intelligence for Discovering Performance-Advantaged and Sustainable Polymers.

A necessary transformation for a sustainable economy is the transition from fossil-derived plastics to polymers derived from biomass and waste resources. While renewable feedstocks can enhance material performance through unique chemical moieties, probing the vast material design space by experiment alone is not practically feasible. Here, we develop a machine-learning-based tool, PolyID, to reduce the design space of renewable feedstocks to enable efficient discovery of performance-advantaged, biobased polymers. PolyID is a multioutput, graph neural network specifically designed to increase accuracy and to enable quantitative structure-property relationship (QSPR) analysis for polymers. It includes a novel domain-of-validity method that was developed and applied to demonstrate how gaps in training data can be filled to improve accuracy. The model was benchmarked with both a 20% held-out subset of the original training data and 22 experimentally synthesized polymers. A mean absolute error for the glass transition temperatures of 19.8 and 26.4 °C was achieved for the test and experimental data sets, respectively. Predictions were made on polymers composed of monomers from four databases that contain biologically accessible small molecules: MetaCyc, MINEs, KEGG, and BiGG. From 1.4 × 106 accessible biobased polymers, we identified five poly(ethylene terephthalate) (PET) analogues with predicted improvements to thermal and transport performance. Experimental validation for one of the PET analogues demonstrated a glass transition temperature between 85 and 112 °C, which is higher than PET and within the predicted range of the PolyID tool. In addition to accurate predictions, we show how the model's predictions are explainable through analysis of individual bond importance for a biobased nylon. Overall, PolyID can aid the biobased polymer practitioner to navigate the vast number of renewable polymers to discover sustainable materials with enhanced performance.

Ni- and Ni/Pd-Catalyzed Reductive Coupling of Lignin-Derived Aromatics to Access Biobased Plasticizers.

Lignin-derived aromatic chemicals offer a compelling alternative to petrochemical feedstocks, and new applications are the focus of extensive interest. 4-Hydroxybenzoic acid (H), vanillic acid (G), and syringic acid (S) are readily obtained via oxidative depolymerization of hardwood lignin substrates. Here, we explore the use of these compounds to access biaryl dicarboxylate esters that represent biobased, less toxic alternatives to phthalate plasticizers. Chemical and electrochemical methods are developed for catalytic reductive coupling of sulfonate derivatives of H, G, and S to access all possible homo- and cross-coupling products. A conventional NiCl2/bipyridine catalyst is able to access the H-H and G-G products, but new catalysts are identified to afford the more challenging coupling products, including a NiCl2/bisphosphine catalyst for S-S and a NiCl2/phenanthroline/PdCl2/phosphine cocatalyst system for H-G, H-S, and G-S. High-throughput experimentation methods with a chemical reductant (Zn powder) are shown to provide an efficient screening platform for identification of new catalysts, while electrochemical methods can access improved yields and/or facilitate implementation on larger scale. Plasticizer tests are performed with poly(vinyl chloride), using esters of the 4,4'-biaryl dicarboxylate products. The H-G and G-G derivatives, in particular, exhibit performance advantages relative to an established petroleum-based phthalate ester plasticizer.

Metabolism of syringyl lignin-derived compounds in Pseudomonas putida enables convergent production of 2-pyrone-4,6-dicarboxylic acid.

Valorization of lignin, an abundant component of plant cell walls, is critical to enabling the lignocellulosic bioeconomy. Biological funneling using microbial biocatalysts has emerged as an attractive approach to convert complex mixtures of lignin depolymerization products to value-added compounds. Ideally, biocatalysts would convert aromatic compounds derived from the three canonical types of lignin: syringyl (S), guaiacyl (G), and p-hydroxyphenyl (H). Pseudomonas putida KT2440 (hereafter KT2440) has been developed as a biocatalyst owing in part to its native catabolic capabilities but is not known to catabolize S-type lignin-derived compounds. Here, we demonstrate that syringate, a common S-type lignin-derived compound, is utilized by KT2440 only in the presence of another energy source or when vanAB was overexpressed, as syringate was found to be O-demethylated to gallate by VanAB, a two-component monooxygenase, and further catabolized via extradiol cleavage. Unexpectedly, the specificity (kcat/KM) of VanAB for syringate was within 25% that for vanillate and O-demethylation of both substrates was well-coupled to O2 consumption. However, the native KT2440 gallate-cleaving dioxygenase, GalA, was potently inactivated by 3-O-methylgallate. To engineer a biocatalyst to simultaneously convert S-, G-, and H-type monomers, we therefore employed VanAB from Pseudomonas sp. HR199, which has lower activity for 3MGA, and LigAB, an extradiol dioxygenase able to cleave protocatechuate and 3-O-methylgallate. This strain converted 93% of a mixture of lignin monomers to 2-pyrone-4,6-dicarboxylate, a promising bio-based chemical. Overall, this study elucidates a native pathway in KT2440 for catabolizing S-type lignin-derived compounds and demonstrates the potential of this robust chassis for lignin valorization.

Reversible Photoswitching in Poly(2-oxazoline) Nanoreactors.

This contribution reports light responsive catalytic nanoreactors based on poly(2-oxazoline) diblock copolymers. The hydrophobic block of the copolymer is a random copolymer consisting of a spiropyran functionalized 2-oxazoline (SPOx) and 2-(but-3-yn-1-yl)-4,5-dihydrooxazole (ButynOx), while the hydrophilic block is based on 2-methyl-2-oxazoline (MeOx). The block copolymer is terminated with tris(2-aminoethyl) amine (TREN) that serves as catalyst in a Knoevenagel condensation. Four block copolymers with different ButynOx/SPOx and hydrophilic/hydrophobic ratios are synthesized and self-assembled through solvent exchange. Micelles and vesicles of various sizes are observed by TEM, which undergo morphological and size changes in response to irradiation with UV light. We hypothesize that these transformations in the nanostructures are caused by increases in the hydrophilicity of the hydrophobic block when spiropyran (SP) isomerizes to merocyanine (MC) in the presence of UV light. The reversible transition from micellar to vesicular nanoreactors resulted in increased reaction kinetics through improved substrate accessibility to the catalytic site, or termination of the catalytic reaction due to polymer precipitation. These nanoreactors present a promising platform towards photoregulating reaction outcomes based on changes in nanostructure morphology.

An Immobilized-Dirhodium Hollow-Fiber Flow Reactor for Scalable and Sustainable C-H Functionalization in Continuous Flow.

A scalable flow reactor is demonstrated for enantioselective and regioselective rhodium carbene reactions (cyclopropanation and C-H functionalization) by developing cascade reaction methods employing a microfluidic flow reactor system containing immobilized dirhodium catalysts in conjunction with the flow synthesis of diazo compounds. This allows the utilization of the energetic diazo compounds in a safe manner and the recycling of the dirhodium catalysts multiple times. This approach is amenable to application in a bulk-scale synthesis employing asymmetric C-H functionalization by stacking multiple fibers in one reactor module. The products from this sequential flow-flow reactor are compared with a conventional batch reactor or flow-batch reactor in terms of yield, regioselectivity, and enantioselectivity.

The Conversion of Starch and Sugars into Branched C10 and C11 Hydrocarbons.

Oligosaccharides, such as starch, cellulose, and hemicelluloses, are abundant and easily obtainable bio-derived materials that can potentially be used as precursors for fuels and chemical feedstocks. To access the pertinent molecular building blocks (i.e., 5- or 6-carbon containing sugar units) located within these biopolymers and transform them into useful fuel precursors, oligosaccharide depolymerization followed by chain extension is required. This chain extension can readily be performed via a Garcia-Gonzalez-like approach using ß-diketones under mild conditions to provide fuel precursors containing an increased carbon atom content that meets fuel requirements. In a subsequent step, ring opening and hydrodeoxygenation chemistry of these species allows for the preparation of branched alkanes under relatively mild conditions. This approach can be applied to monomeric sugars (glucose and xylose), oligosaccharides (starch), and potentially to hydrolyzed dedicated energy crops to allow the conversion of real biomass into fuel type molecules.

Probing the Role of Zr Addition versus Textural Properties in Enhancement of CO2 Adsorption Performance in Silica/PEI Composite Sorbents.

Polymeric amines such as poly(ethylenimine) (PEI) supported on mesoporous oxides are promising candidate adsorbents for CO2 capture processes. An important aspect to the design and optimization of these materials is a fundamental understanding of how the properties of the oxide support such as pore structure, particle morphology, and surface properties affect the efficiency of the guest polymer in its interactions with CO2. Previously, the efficiency of impregnated PEI to adsorb CO2 was shown to increase upon the addition of Zr as a surface modifier in SBA-15. However, the efficacy of this method to tune the adsorption performance has not been explored in materials of differing textural and morphological nature. Here, these issues are directly addressed via the preparation of an array of SBA-15 support materials with varying textural and morphological properties, as well as varying content of zirconium doped into the material. Zirconium is incorporated into the SBA-15 either during the synthesis of the SBA-15, or postsynthetically via deposition of Zr species onto pure-silica SBA-15. The method of Zr incorporation alters the textural and morphological properties of the parent SBA-15 in different ways. Importantly, the CO2 capacity of SBA-15 impregnated with PEI increases by a maximum of ∼60% with the quantity of doped Zr for a "standard" SBA-15 containing significant microporosity, while no increase in the CO2 capacity is observed upon Zr incorporation for an SBA-15 with reduced microporosity and a larger pore size, pore volume, and particle size. Finally, adsorbents supported on SBA-15 with controlled particle morphology show only modest increases in CO2 capacity upon inclusion of Zr to the silica framework. The data demonstrate that the textural and morphological properties of the support have a more significant impact on the ability of PEI to capture CO2 than the support surface composition.