The use of machine learning in computational molecular design has great potential to accelerate the discovery of innovative materials. However, its practical benefits still remain unproven in real-world applications, particularly in polymer science. We demonstrate the successful discovery of new polymers with high thermal conductivity, inspired by machine-learning-assisted polymer chemistry. This discovery was made by the interplay between machine intelligence trained on a substantially limited amount of polymeric properties data, expertise from laboratory synthesis and advanced technologies for thermophysical property measurements. Using a molecular design algorithm trained to recognize quantitative structure-property relationships with respect to thermal conductivity and other targeted polymeric properties, we identified thousands of promising hypothetical polymers. From these candidates, three were selected for monomer synthesis and polymerization because of their synthetic accessibility and their potential for ease of processing in further applications. The synthesized polymers reached thermal conductivities of 0.18-0.41 W/mK, which are comparable to those of state-of-the-art polymers in non-composite thermoplastics .

There is a growing demand for the use of machine learning (ML) to derive fast-to-evaluate surrogate models of materials properties. In recent years, a broad array of materials property databases have emerged as part of a digital transformation of materials science. However, recent technological advances in ML are not fully exploited because of the insufficient volume and diversity of materials data. An ML framework called “transfer learning” has considerable potential to overcome the problem of limited amounts of materials data. Transfer learning relies on the concept that various property types, such as physical, chemical, electronic, thermodynamic, and mechanical properties, are physically interrelated. For a given target property to be predicted from a limited supply of training data, models of related proxy properties are pretrained using sufficient data; these models capture common features relevant to the target task. Repurposing of such machine-acquired features on the target task yields outstanding prediction performance even with exceedingly small data sets, as if highly experienced human experts can make rational inferences even for considerably less experienced tasks. In this study, to facilitate widespread use of transfer learning, we develop a pretrained model library called XenonPy.MDL. In this first release, the library comprises more than 140 000 pretrained models for various properties of small molecules, polymers, and inorganic crystalline materials. Along with these pretrained models, we describe some outstanding successes of transfer learning in different scenarios such as building models with only dozens of materials data, increasing the ability of extrapolative prediction through a strategic model transfer, and so on. Remarkably, transfer learning has autonomously identified rather nontrivial transferability across different properties transcending the different disciplines of materials science; for example, our analysis has revealed underlying bridges between small molecules and polymers and between organic and inorganic chemistry.

With harmful ozone concentrations tied to meteorological conditions, EPA investigates the U.S. air quality implications of a changing climate. Consequently, the 03 NAAQS are most often exceeded during summertime hot spells in places with large natural or anthropogenic precursor emissions (e.g., cities and suburban areas). Table 2 The average maximum or minimum temperature and/or changes in their spatial distribution and duration, leading to a change in reaction rate coefficients and the solubility of gases in cloud water solution;The frequency and pattern of cloud cover, leading to a change in reaction rates and rates of conversion of S02to acid deposition;The frequency and intensity of stagnation episodes or a change in the mixing layer, leading to more or less mixing of polluted air with background air;Background boundary layer concentrations of water vapor, hydrocarbons, NOx, and 03, leading to more or less dilution of polluted air in the boundary layer and altering the chemical transformation rates;

The formation of secondary organic aerosols (SOA) is simulated for the Nashville/western Tennessee domain using three recent SOA modules incorporated into the three-dimensional air quality model, CMAQ. The Odum/Griffin et al. and CMU/STI modules represent SOA absorptive partitioning into a mixture of primary and secondary particulate organic compounds (OC), with some differences in the formulation of the absorption process and the selection of SOA species and their precursors. Empirical representations based on measured laboratory SOA yields are used for condensable organic products in both these modules. The AEC module simulates SOA absorption into organic and aqueous particulate phases, and a representation based on an explicit gas-phase mechanism is used in the AEC module. Predicted SOA concentrations can vary by a factor of 10 or more. In general, the gas-phase mechanistic approach predicts a higher yield of SOA than those based on laboratory yields. There exist some differences in the two empirical modules despite their similar basis on experimental data. All three modules predict a dominance of SOA of biogenic origin as compared to SOA of anthropogenic origin. The causes for differences among the three SOA modules include the representation of terpenes, the mechanistic versus empirical representation of SOA-forming reactions, the identities of SOA, and the parameters used in the gas/particle partitioning calculations. Two sensitivity studies show that formation of water-soluble SOA and temperature dependence may be areas of key uncertainties affecting current models.