Prediction of Alternative Gasoline Sorption in a Semicrystalline Poly(ethylene)

Abstract

In this work, we first report the acquisition of new experimental data and then the development of quantitative structure-property relationships on the basis of sorption values for neat compounds and up to quinary mixtures of some hydrocarbons, alcohols, and ethers, in a semicrystalline poly(ethylene). Two machine learning methods (i.e., genetic function approximation and support vector machines) and two families of descriptors (i.e., functional group counts and substructural molecular fragments) were used to derive predictive models. Models were then used to predict sorption variations when increasing the number of carbon atoms in a series of hydrocarbons and for n-alkan-1-ols. In addition to the performed internal/external validations, the model was further tested for surrogate gasolines containing ca. 300 compounds, and predicted sorption values were in excellent agreement with experimental data (R(2) = 0.940).

Fuel sorption into polymers: Experimental and machine learning studies

“…The term sorption is commonly used to describe the dissolution of a penetrant into a polymer matrix. Measurements of liquid sorption into polymers were performed using a gravimetric method, as detailed in our previous works [44]. Experiments consists in recording the mass variation (weight gain or loss) of a polymer sample with time when immersed into a large excess of the studied liquid.…”

“…From conclusions drawn in previous studies [2,44], we chose to solely consider functional group count descriptors (FGCD). Such a simple representation of compounds has been shown to provide relevant descriptors us-…”

“…able in QSPR procedure [3,44]. This family of molecular descriptors gathers some counts of groups identified as relevant under chemical aspects.…”

“…Li et al proposed models for predicting polymer/brine partition coefficients for chemicals, with polymers such as polyethylene, polypropylene and polystyrene [43]. Our group has previously proposed QSPR models to predict sorption values for neat compounds and up to quinary mixtures of hydrocarbons, alcohols, and ethers, and demonstrated their applicability to predict sorption values for some alternative fuels into a poly(ethylene) [44].…”

Chemoinformatics at IFP Energies Nouvelles: Applications in the Fields of Energy, Transport, and Environment

“…Currently, such compatibility tests are performed with experiments of thousands hour duration. Consequently, we envisaged the use of chemoinformatics tools to predict possible changes in physical and chemical properties of polymers after contact with biofuels . We conducted a series of measurements to establish a database for the sorption of gasoline components – neat compounds and up to quinary mixtures of hydrocarbons, alcohols, and ethers – in a semicrystalline poly(ethylene).…”