Abstract: High-potential molecules derived from biomass sources may suitably replace or supple-ment traditional nonrenewable hydrocarbon fuels to reduce pollution and fuel processing cost. Ex-perimental property testing of these bioproducts is usually conducted years after initial bench-scale experiments, due to high experimental costs and/or high volume requirements. However, neglecting to conduct property testing early in the pathway development cycle can lead to investments spent on scaling-up production of bioproducts and biofuels that do not perform as expected. Instead, machine-learning techniques can be used to develop quantitative structure–property relationships for molecules using a relatively large training set of molecular descriptor data. For this study, we compiled measured properties, IR spectra, and molecular descriptors of bio-based molecules from databases and published studies for training models of bioproduct properties. We trained regres-sion models with molecular descriptors and will compare results of different estimators. This study describes the first steps towards a performance prediction tool for bio-based alternative fuels. Keywords: Machine learning, biofuels, jet fuels, fuel properties

Machine learning has proven to be a powerful tool for accelerating biofuel development. Although numerous models are available to predict a range of properties using chemical descriptors, there is a trade-off between interpretability and performance. Neural networks provide predictive models with high accuracy at the expense of some interpretability, while simpler models such as linear regression often lack in accuracy. In addition to model architecture, feature selection is also critical for developing interpretable and accurate predictive models. We present a method for systematically selecting molecular descriptor features and developing interpretable machine learning models without sacrificing accuracy. Our method simplifies the process of selecting features by reducing feature multicollinearity and enables discoveries of new relationships between global properties and molecular descriptors. To demonstrate our approach, we developed models for predicting melting point, boiling point, flash point, yield sooting index, and net heat of combustion with the help of the Tree-based Pipeline Optimization Tool (TPOT). For training, we used publicly available experimental data for up to 8351 molecules. Our models accurately predict various molecular properties for organic molecules (mean absolute percent error (MAPE) ranges from 3.3% to 10.5%) and provide a set of features that are well-correlated to the property. This method enables researchers to explore sets of features that significantly contribute to the prediction of the property, offering new scientific insights. To help accelerate early stage biofuel research and development, we also integrated the data and models into a open-source, interactive web tool.

The combustion of petroleum-based fuels contributes to increases in atmospheric CO2, contributing to climate change. As other sectors electrify, current battery technology makes this impractical for the aviation industry. The fastest pathway to reducing the carbon contributions of aviation is to create low or no net carbon-emitting drop-in petroleum alternatives. This research explores 3 potential sorghum derived jet fuel molecules from 4 different production routes, hydro-processed esters and fatty acids (HEFA), and Jet A, to identify how blends including these fuels could offer improved performance (MJ/kg, MJ/L) and improved emissions (gCO2/MJ) while minimizing costs. Different applications will value each target metric differently. This research utilizes the Jet Fuel Blend Optimizer (JudO) to create a 4-dimensional Pareto front across potential solutions. The molecules considered will not offer reduced cost compared to conventional jet fuel (Jet A), but they can be utilized to create blends with higher specific energy, greater energy density, and reduced greenhouse gas emissions. Considerations including carbon credits, the sale of byproducts, and the valuation of improved performance will make the proposed molecules more commercially viable. At the lowest region of GHG solutions, coupled with equivalent LCFS approximations, JudO determined blends that could achieve as high as 69% overall carbon reductions at a premium of $0.34/L.