Pendant Group Functionalization of Cyclic Olefin for High Temperature and High-Density Energy Storage.

High-temperature flexible polymer dielectrics are critical for high density energy storage and conversion. The need to simultaneously possess a high bandgap, dielectric constant and glass transition temperature forms a substantial design challenge for novel dielectric polymers. Here, by varying halogen substituents of an aromatic pendant hanging off a bicyclic mainchain polymer, a class of high-temperature olefins with adjustable thermal stability are obtained, all with uncompromised large bandgaps. Halogens substitution of the pendant groups at para or ortho position of polyoxanorborneneimides (PONB) imparts it with tunable high glass transition from 220 to 245 °C, while with high breakdown strength of 625-800 MV/m. A high energy density of 7.1 J/cc at 200 °C is achieved with p-POClNB, representing the highest energy density reported among homo-polymers. Molecular dynamic simulations and ultrafast infrared spectroscopy are used to probe the free volume element distribution and chain relaxations pertinent to dielectric thermal properties. An increase in free volume element is observed with the change in the pendant group from fluorine to bromine at the para position; however, smaller free volume element is observed for the same pendant when at the ortho position due to steric hindrance. With the dielectric constant and bandgap remaining stable, properly designing the pendant groups of PONB boosts its thermal stability for high density electrification.

Polyphosphazene-Based Anion-Anchored Polymer Electrolytes For All-Solid-State Lithium Metal Batteries.

Safety concerns of traditional liquid electrolytes, especially when paired with lithium (Li) metal anodes, have stimulated research of solid polymer electrolytes (SPEs) to exploit the superior thermal and mechanical properties of polymers. Polyphosphazenes are primarily known for their use as flame retardant materials and have demonstrated high Li-ion conductivity owing to their highly flexible P = N backbone which promotes Li-ion conduction via inter- and intrachain hopping along the polymer backbone. While polyphosphazenes are largely unexplored as SPEs in the literature, a few existing examples showed promising ionic conductivity. By anchoring the anion to the polymer backbone, one may primarily allow the movement of Li ions, alleviating the detrimental effects of polarization that are common in conventional dual-ion conducting SPEs. Anion-anchored SPEs, known as single Li-ion conducting solid polymer electrolytes (SLiC-SPEs), exhibit high Li-ion transference numbers (tLi+), which limits Li dendrite growth, thus further increasing the safety of SPEs. However, previously reported SLiC-SPEs suffer from inadequate ionic conductivity, small electrochemical stability windows (ESWs), and limited cycling stability. Herein, we report three polyphosphazene-based SLiC-SPEs comprising lithiated polyphosphazenes. The SLiC polyphosphazenes were prepared through a facile synthesis route, opening the door for enhanced tunability of polymer properties via facile macromolecular nucleophilic substitution and subsequent lithiation. State-of-the-art characterization techniques, such as differential scanning calorimetry (DSC), electrochemical impedance spectroscopy (EIS), and solid-state nuclear magnetic resonance spectroscopy (ssNMR) were employed to probe the effect of the polymer structure on Li-ion dynamics and other electrochemical properties. Produced SPEs showed thermal stability up to ∼208 °C with ionic conductivities comparable to that of the best-reported SLiC-SPEs that definitively comprise no solvents or plasticizers. Among the three lithiated polyphosphazenes, the SPE containing dilithium poly[bis(trifluoroethylamino)phosphazene] (pTFAP2Li) exhibited the most promising electrochemical characteristics with tLi+ of 0.76 and compatibility with both Li metal anodes and LiFePO4 (LFP) cathodes; through 40 cycles at 100 °C, the PEO-pTFAP2Li blend showed 81.2% capacity utilization and 86.8% capacity retention. This work constitutes one of the first successful demonstrations of the cycling performance of a true all-solid-state Li-metal battery using SLiC polyphosphazene SPEs.

Machine-Guided Discovery of Acrylate Photopolymer Compositions.

Additive manufacturing (AM) can be advanced by the diverse characteristics offered by thermoplastic and thermoset polymers and the further benefits of copolymerization. However, the availability of suitable polymeric materials for AM is limited and may not always be ideal for specific applications. Additionally, the extensive number of potential monomers and their combinations make experimental determination of resin compositions extremely time-consuming and costly. To overcome these challenges, we develop an active learning (AL) approach to effectively choose compositions in a ternary monomer space ranging from rigid to elastomeric. Our AL algorithm dynamically suggests monomer composition ratios for the subsequent round of testing, allowing us to efficiently build a robust machine learning (ML) model capable of predicting polymer properties, including Young's modulus, peak stress, ultimate strain, and Shore A hardness based on composition while minimizing the number of experiments. As a demonstration of the effectiveness of our approach, we use the ML model to drive material selection for a specific property, namely, Young's modulus. The results indicate that the ML model can be used to select material compositions within at least 10% of a targeted value of Young's modulus. We then use the materials designed by the ML model to 3D print a multimaterial "hand" with soft "skin" and rigid "bones". This work presents a promising tool for enabling informed AM material selection tailored to user specifications and accelerating material discovery using a limited monomer space.

Informatics-Driven Design of Superhard B-C-O Compounds.

Materials containing B, C, and O, due to the advantages of forming strong covalent bonds, may lead to materials that are superhard, i.e., those with a Vicker's hardness larger than 40 GPa. However, the exploration of this vast chemical, compositional, and configurational space is nontrivial. Here, we leverage a combination of machine learning (ML) and first-principles calculations to enable and accelerate such a targeted search. The ML models first screen for potentially superhard B-C-O compositions from a large hypothetical B-C-O candidate space. Atomic-level structure search using density functional theory (DFT) within those identified compositions, followed by further detailed analyses, unravels on four potentially superhard B-C-O phases exhibiting thermodynamic, mechanical, and dynamic stability.

Chemical Circularity in 3D Printing with Biobased Δ-Valerolactone.

Digital Light Processing (DLP) is a vat photopolymerization-based 3D printing technology that fabricates parts typically made of chemically crosslinked polymers. The rapidly growing DLP market has an increasing demand for polymer raw materials, along with growing environmental concerns. Therefore, circular DLP printing with a closed-loop recyclable ink is of great importance for sustainability. The low-ceiling temperature alkyl-substituted δ-valerolactone (VL) is an industrially accessible biorenewable feedstock for developing recyclable polymers. In this work, acrylate-functionalized poly(δ-valerolactone) (PVLA), synthesized through the ring-opening transesterification polymerization of VL, is used as a platform photoprecursor to improve the chemical circularity in DLP printing. A small portion of photocurable reactive diluent (RD) turns the unprintable PVLA into DLP printable ink. Various photocurable monomers can serve as RDs to modulate the properties of printed structures for applications like sacrificial molds, soft actuators, sensors, etc. The intrinsic depolymerizability of PVLA is well preserved, regardless of whether the printed polymer is a thermoplastic or thermoset. The recovery yield of virgin quality VL monomer is 93% through direct bulk thermolysis of the printed structures. This work proposes the utilization of depolymerizable photoprecursors and highlights the feasibility of biorenewable VL as a versatile material platform toward circular DLP printing.

Probing Electronic Band Structures of Dielectric Polymers via Pre-Breakdown Conduction.

The electronic band structure, especially the defect states at the conduction band tail, dominates electron transport and electrical degradation of a dielectric material under an extremely high electric field. However, the electronic band structure in a dielectric is barely well studied due to experimental challenges in detecting the electrical conduction to an extremely high electric field, i.e., prebreakdown. In this work, the electronic band structure of polymer dielectric films is probed through an in situ prebreakdown conduction measurement method in conjunction with a space-charge-limited-current spectroscopic analysis. An exponential distribution of defect states at the conduction band tail with varying trap levels is observed in accordance with the specific morphological disorder in the polymer dielectric, and the experimental defect states show also a favorable agreement with the calculated density of states from the density functional theory. The methodology demonstrated in this work bridges the molecule-structure-determined electronic band structure and the macro electrical conduction behavior with a highly improved understanding of material properties that control the electrical breakdown, and paves a way for guiding the modification of existing material and the exploration of novel materials for high electric field applications.

Accelerated Scheme to Predict Ring-Opening Polymerization Enthalpy: Simulation-Experimental Data Fusion and Multitask Machine Learning.

Ring-opening enthalpy (ΔHROP) is a fundamental thermodynamic quantity controlling the polymerization and depolymerization of an important class of recyclable polymers, namely, those created from ring-opening polymerization (ROP). Highly accurate first-principles-based computational methods to compute ΔHROP are computationally too demanding to efficiently guide the design of depolymerizable polymers. In this work, we develop a generalizable machine-learning model that was trained on experimental measurements and reliably computed simulation results of ΔHROP (the latter provides a pathway to systematically increase the chemical diversity of the data). Predictions of ΔHROP using this machine-learning model require essentially no time while the prediction accuracy is about ∼8 kJ/mol, approaching the well-known chemical accuracy. We hope that this effort will contribute to the future development of new depolymerizable polymers.

Effect of Fluorine in Redesigning Energy-Storage Properties of High-Temperature Dielectric Polymers.

Exploration of novel polymer dielectrics exhibiting high electric-field stability and high energy density with high efficiency at elevated temperatures is urgently needed for ever-demanding energy-storage technologies. Conventional high-temperature polymers with conjugated backbone structures cannot fulfill this demand due to their deteriorated performance at elevated electric fields. Here, in search of new polymer structures, we have explored the effect of fluorine groups on the energy-storage properties of polyoxanorbornene imide polymers with simultaneous wide band gap and high glass transition temperature (Tg). The systematic synthesis of polymers with varying amounts of fluorine is carried out and characterized for the energy-storage properties. The incorporation of fluorine imparts flexibility to the polymer structure, and free-standing films can be obtained. An oxanorbornene copolymer with 25% fluorination exhibits a high breakdown strength of 700 MV/m and a discharged energy density of 6.3 J/cm3 with 90% efficiency. The incorporation of fluorine helps to increase the polymer band gap, as observed using UV-vis spectroscopy, but lowers the polymer Tg, as shown by differential scanning calorimetry. Both the displacement-electric field (D-E) hysteresis loop and high-field conduction measurements show increased conduction loss for polymers with higher fluorine content, despite their larger band gap. The presence of excess free volume may play a key role in increasing the conduction current and lowering the efficiency of polymers with high fluorine content. Such an improved understanding of the effect of fluorination on the polymer energy-storage properties, as revealed in this systematic molecular engineering study, broadens the basis of material-informatic proxies to enable a more targeted codesign of scalable and efficient polymer dielectrics.

Low-temperature 3D printing of transparent silica glass microstructures.

Transparent silica glass is one of the most essential materials used in society and industry, owing to its exceptional optical, thermal, and chemical properties. However, glass is extremely difficult to shape, especially into complex and miniaturized structures. Recent advances in three-dimensional (3D) printing have allowed for the creation of glass structures, but these methods involve time-consuming and high-temperature processes. Here, we report a photochemistry-based strategy for making glass structures of micrometer size under mild conditions. Our technique uses a photocurable polydimethylsiloxane resin that is 3D printed into complex structures and converted to silica glass via deep ultraviolet (DUV) irradiation in an ozone environment. The unique DUV-ozone conversion process for silica microstructures is low temperature (~220°C) and fast (<5 hours). The printed silica glass is highly transparent with smooth surface, comparable to commercial fused silica glass. This work enables the creation of arbitrary structures in silica glass through photochemistry and opens opportunities in unexplored territories for glass processing techniques.

Data-driven predictions of complex organic mixture permeation in polymer membranes.

Membrane-based organic solvent separations are rapidly emerging as a promising class of technologies for enhancing the energy efficiency of existing separation and purification systems. Polymeric membranes have shown promise in the fractionation or splitting of complex mixtures of organic molecules such as crude oil. Determining the separation performance of a polymer membrane when challenged with a complex mixture has thus far occurred in an ad hoc manner, and methods to predict the performance based on mixture composition and polymer chemistry are unavailable. Here, we combine physics-informed machine learning algorithms (ML) and mass transport simulations to create an integrated predictive model for the separation of complex mixtures containing up to 400 components via any arbitrary linear polymer membrane. We experimentally demonstrate the effectiveness of the model by predicting the separation of two crude oils within 6-7% of the measurements. Integration of ML predictors of diffusion and sorption properties of molecules with transport simulators enables for the rapid screening of polymer membranes prior to physical experimentation for the separation of complex liquid mixtures.

polyBERT: a chemical language model to enable fully machine-driven ultrafast polymer informatics.

Polymers are a vital part of everyday life. Their chemical universe is so large that it presents unprecedented opportunities as well as significant challenges to identify suitable application-specific candidates. We present a complete end-to-end machine-driven polymer informatics pipeline that can search this space for suitable candidates at unprecedented speed and accuracy. This pipeline includes a polymer chemical fingerprinting capability called polyBERT (inspired by Natural Language Processing concepts), and a multitask learning approach that maps the polyBERT fingerprints to a host of properties. polyBERT is a chemical linguist that treats the chemical structure of polymers as a chemical language. The present approach outstrips the best presently available concepts for polymer property prediction based on handcrafted fingerprint schemes in speed by two orders of magnitude while preserving accuracy, thus making it a strong candidate for deployment in scalable architectures including cloud infrastructures.

Chemically Recyclable Polymer System Based on Nucleophilic Aromatic Ring-Opening Polymerization.

The development of chemically recyclable polymers with desirable properties is a long-standing but challenging goal in polymer science. Central to this challenge is the need for reversible chemical reactions that can equilibrate at rapid rates and provide efficient polymerization and depolymerization cycles. Based on the dynamic chemistry of nucleophilic aromatic substitution (SNAr), we report a chemically recyclable polythioether system derived from readily accessible benzothiocane (BT) monomers. This system represents the first example of a well-defined monomer platform capable of chain-growth ring-opening polymerization through an SNAr manifold. The polymerizations reach completion in minutes, and the pendant functionalities are easily customized to tune material properties or render the polymers amenable to further functionalization. The resulting polythioether materials exhibit comparable performance to commercial thermoplastics and can be depolymerized to the original monomers in high yields.

A general-purpose material property data extraction pipeline from large polymer corpora using natural language processing.

The ever-increasing number of materials science articles makes it hard to infer chemistry-structure-property relations from literature. We used natural language processing methods to automatically extract material property data from the abstracts of polymer literature. As a component of our pipeline, we trained MaterialsBERT, a language model, using 2.4 million materials science abstracts, which outperforms other baseline models in three out of five named entity recognition datasets. Using this pipeline, we obtained ~300,000 material property records from ~130,000 abstracts in 60 hours. The extracted data was analyzed for a diverse range of applications such as fuel cells, supercapacitors, and polymer solar cells to recover non-trivial insights. The data extracted through our pipeline is made available at polymerscholar.org which can be used to locate material property data recorded in abstracts. This work demonstrates the feasibility of an automatic pipeline that starts from published literature and ends with extracted material property information.

One-Pot Synthesis of Depolymerizable δ-Lactone Based Vitrimers.

A depolymerizable vitrimer that allows both reprocessability and monomer recovery by a simple and scalable one-pot two-step synthesis of vitrimers from cyclic lactones is reported. Biobased δ-valerolactone with alkyl substituents (δ-lactone) has low ceiling temperature; thus, their ring-opening-polymerized aliphatic polyesters are capable of depolymerizing back to monomers. In this work, the amorphous poly(δ-lactone) is solidified into an elastomer (i.e., δ-lactone vitrimer) by a vinyl ether cross-linker with dynamic acetal linkages, giving the merits of reprocessing and healing. Thermolysis of the bulk δ-lactone vitrimer at 200 °C can recover 85-90 wt% of the material, allowing reuse without losing value and achieving a successful closed-loop life cycle. It further demonstrates that the new vitrimer has excellent properties, with the potential to serve as a biobased and sustainable replacement of conventional soft elastomers for various applications such as lenses, mold materials, soft robots, and microfluidic devices.

Polymer Informatics at Scale with Multitask Graph Neural Networks.

Artificial intelligence-based methods are becoming increasingly effective at screening libraries of polymers down to a selection that is manageable for experimental inquiry. The vast majority of presently adopted approaches for polymer screening rely on handcrafted chemostructural features extracted from polymer repeat units-a burdensome task as polymer libraries, which approximate the polymer chemical search space, progressively grow over time. Here, we demonstrate that directly "machine learning" important features from a polymer repeat unit is a cheap and viable alternative to extracting expensive features by hand. Our approach-based on graph neural networks, multitask learning, and other advanced deep learning techniques-speeds up feature extraction by 1-2 orders of magnitude relative to presently adopted handcrafted methods without compromising model accuracy for a variety of polymer property prediction tasks. We anticipate that our approach, which unlocks the screening of truly massive polymer libraries at scale, will enable more sophisticated and large scale screening technologies in the field of polymer informatics.

Solvent selection for polymers enabled by generalized chemical fingerprinting and machine learning.

We present machine learning models trained on experimental data to predict room-temperature solubility for any polymer-solvent pair. The new models are a significant advancement over past data-driven work, in terms of protocol, validity, and versatility. A generalizable fingerprinting method is used for the polymers and solvents, making it possible, in principle, to handle any polymer-solvent combination. Our data-driven approach achieves high accuracy when either both the polymer and solvent or just the polymer has been seen during the training phase. Model performance is modest though when a solvent (in a newly queried polymer-solvent pair) is not part of the training set. This is likely because the number of unique solvents in our data set is small (much smaller than the number of polymers). Nevertheless, as the data set increases in size, especially as the solvent set becomes more diverse, the overall predictive performance is expected to improve.

Vapor-Phase Infiltration of Polymer of Intrinsic Microporosity 1 (PIM-1) with Trimethylaluminum (TMA) and Water: A Combined Computational and Experimental Study.

Vapor-phase infiltration, a postpolymerization modification process, has demonstrated the ability to create organic-inorganic hybrid membranes with excellent stability in organic solvents while maintaining critical membrane properties of high permeability and selectivity. However, the chemical reaction pathways that occur during VPI and their implications on the hybrid membrane stability are poorly understood. This paper combines in situ quartz crystal microbalance gravimetry (QCM) and ex situ chemical characterization with first-principles simulations at the atomic scale to study each processing step in the infiltration of polymer of intrinsic microporosity 1 (PIM-1) with trimethylaluminum (TMA) and its co-reaction with water vapor. Building upon results from in situ QCM experiments and SEM/EDX, which find TMA remains within PIM-1 even under long desorption times, density functional theory (DFT) simulations identify that an energetically stable coordination forms between the metal-organic precursor and PIM-1's nitrile functional group during the precursor exposure step of VPI. In the subsequent water vapor exposure step, the system undergoes a series of exothermic reactions to form the final hybrid membrane. DFT simulations indicate that these reaction pathways result in aluminum oxyhydroxide species consistent with ex situ XPS and FTIR characterization. Both NMR and DFT simulations suggest that the final aluminum structure is primarily 6-fold coordinated and that the aluminum is at least dimerized, if not further "polymerized". According to the simulations, coordination of the aluminum with at least one nitrile group from the PIM-1 appears to weaken significantly as the final inorganic structure emerges but remains present to enable the formation of the 6-fold coordination species. Water molecules are proposed to complete the coordination complex without further increasing the aluminum's oxidation state. This study provides new insights into the infiltration process and the chemical structure of the final hybrid membrane including support for the possible mechanism of solvent stability.

Modulating Polymerization Thermodynamics of Thiolactones Through Substituent and Heteroatom Incorporation.

A central challenge in the development of next-generation sustainable materials is to design polymers that can easily revert back to their monomeric starting material through chemical recycling to monomer (CRM). An emerging monomer class that displays efficient CRM are thiolactones, which exhibit rapid rates of polymerization and depolymerization. This report details the polymerization thermodynamics for a series of thiolactone monomers through systematic changes to substitution patterns and sulfur heteroatom incorporation. Additionally, computational studies highlight the importance of conformation in modulating the enthalpy of polymerization, leading to monomers that display high conversions to polymer at near-ambient temperatures, while maintaining low ceiling temperatures (Tc). Specifically, the combination of a highly negative enthalpy (-19.3 kJ/mol) and entropy (-58.4 J/(mol·K)) of polymerization allows for a monomer whose equilibrium polymerization conversion is very sensitive to temperature.

Toward Recyclable Polymers: Ring-Opening Polymerization Enthalpy from First-Principles.

Ring-opening polymerization (ROP) enthalpy ΔHROP is an important thermodynamic property controlling the polymerization of cyclic monomers. While ΔHROP can be measured, computing ΔHROP for realistic polymer systems with an error of ≃5-10 kJ/mol is critical for designing new monomer systems for depolymerizable polymers. We have developed a first-principles computational scheme in which multiple challenges in computing ΔHROP are resolved definitively including extensive exploration of conformational states and adequately addressing finite size effects. This scheme is validated on a diverse benchmark set of 42 ROP polymers for which reliable experimental values of ΔHROP are available. For this set, the ΔHROP root-mean-square error is ≃7 kJ/mol, about 3-times smaller than conventional approaches. This development opens up new pathways to build up a high-quality database of ΔHROP for downstream predictive machine-learning models and ultimately to accelerate the design of depolymerizable polymers with desired properties.

High dielectric constant and high breakdown strength polyimide via tin complexation of the polyamide acid precursor.

Polymer dielectrics with ultra-high charge-discharge rates are significant for advanced electrical and electronic systems. Despite the fact that polymers possess high breakdown strength, the low dielectric constant (k) of polymers gives rise to low energy densities. Incorporating metal into polyimides (PI) at the polyamic acid (PAA) precursor stage of the synthetic process is a cheap and versatile way to improve the dielectric constant of the hybrid system while maintaining a high breakdown strength. Here, we explore inclusion of different percentages of Sn as a coordinated complex in a polyimide matrix to achieve metal homogeneity within the dielectric film to boost dielectric constant. Sn-O bonds with high atomic polarizability are intended to enhance the ionic polarization without sacrificing bandgap, a measurable property of the material to assess intrinsic breakdown strength. Enhancements of k from ca. 3.7 to 5.7 were achieved in going from the pure PI film to films containing 10 mol% tin.