Special Issue on “Chemical Process Design, Simulation and Optimization”

Since humanity has been able to transform materials, such as raw minerals, and produce food or beverages, a central question was the type of operation and how and where it should be performed [...]

Machine learning with physicochemical relationships: solubility prediction in organic solvents and water

Solubility prediction remains a critical challenge in drug development, synthetic route and chemical process design, extraction and crystallisation. Here we report a successful approach to solubility prediction in organic solvents and water using a combination of machine learning (ANN, SVM, RF, ExtraTrees, Bagging and GP) and computational chemistry. Rational interpretation of dissolution process into a numerical problem led to a small set of selected descriptors and subsequent predictions which are independent of the applied machine learning method. These models gave significantly more accurate predictions compared to benchmarked open-access and commercial tools, achieving accuracy close to the expected level of noise in training data (LogS ± 0.7). Finally, they reproduced physicochemical relationship between solubility and molecular properties in different solvents, which led to rational approaches to improve the accuracy of each models.

Computer-Aided Environmental Assessment Applied for Estimation of Ecological Impacts Derived from Topological Pathways Based on Lignocellulosic Biomass Transformation

The growing awareness to include sustainability goals in the chemical process design has been making palpable since many governments and research institutions have made many efforts precisely to progress new ways to transform available resources into valuable chemicals. In this sense, this work is presenting a computer-aided evaluation based on environmental impact assessment and comparison of technical parameters for estimating the potential effects of two biorefinery designs. The first process involved a multiproduct production of acetone, butanol, and ethanol from cassava waste, while the second biorefinery comprised of succinic acid and bioethanol production from a mixture of cassava waste and banana rachis. These residues are highly available in the North Colombia region due to the agroindustrial activities of that zone. The developed environmental analysis employed the waste reduction algorithm (WAR) for estimating impact generation and output rates considering atmospheric and toxicological categories. Otherwise, process simulation of biorefineries showed production of 546.3 kg/h of acetone, 280.0 kg/h of ethanol, and 1305 kg/h of butanol for topology 1, while topology 2 delivered a synthesis of 13,865.7 kg/h of acetic acid and 2277.9 kg/h of ethanol. Data generated from process simulation allowed performing a technical comparison between evaluated biorefineries, showing a higher performance of evaluated indicators for topology 2. These evaluated variables included resource energy efficiency, and production yield, among others. The environmental analysis provided relevant information, indicating that topology 2 is a better alternative from an ecological viewpoint since this design would emit substances with lower effects than topology 1.