Predicting thermophysical properties of dialkylimidazolium ionic liquids from sigma profiles

Mapping the Flammability Space of Sustainable Refrigerant Mixtures through an Artificial Neural Network Based on Molecular Descriptors

As the EU's mandates to phase out high-GWP refrigerants come into effect, the refrigeration industry is facing a new, unexpected reality: the introduction of more flammable yet environmentally compliant alternatives. This paradigm shift amplifies the need for a rapid, reliable screening methodology to assess the propensity for flammability of emerging fourth generation blends, offering a pragmatic alternative to laborious and time-intensive traditional experimental assessments. In this study, an artificial neural network (ANN) is meticulously constructed, evaluated, and validated to address this emerging challenge by predicting the normalized flammability index (NFI) for an extensive array of pure, binary, and ternary mixtures, reflecting a substantial diversity of compounds like CO2, hydrofluorocarbons (HFCs), hydrofluoroolefins (HFOs), six saturated hydrocarbons (sHCs), hydroolefins (HOs), and others. The optimal configuration ([61 (I) × 14 (HL1) × 24 (HL2) × 1 (O)]) demonstrated a profound fit to the data, with metrics like R 2 of 0.999, root-mean-square error (RMSE) of 0.1735, average absolute relative deviation (AARD)% of 0.8091, and SDav of ±0.0434. Exhaustive assessments were conducted to ensure the most efficient architecture without compromising the accuracy. Additionally, the analysis of the standardized residuals (SDR) and applicability domain (AD) exhibited fine control and consistency over the data points. External validation using quaternary mixtures further attested to the model's adaptability and predictive capability. The exploration into the relative contribution of descriptors led to the identification of 23 significant sigma descriptors derived from conductor-like screening model (COSMO), responsible for 90.98% of the total contribution, revealing potential avenues for model simplification without a substantial loss in predictive power. Moreover, the model successfully predicted the behavior of prospective industry-relevant mixtures, reinforcing its reliability and opening the door to experimentation with untested blends. The results collectively manifest the developed ANN's efficiency, robustness, and adaptability in modeling flammability, catering to the demands of industry standards, environmental concerns, and safety requirements.

Unveiling CO2 capture in tailorable green neoteric solvents: An ensemble learning approach informed by quantum chemistry

In the relentless battle against the impending climate crisis, deep eutectic solvents (DESs) have emerged as beacons of hope in the realm of green chemistry, igniting a resurgence of scientific exploration. These versatile compounds hold the promise of revolutionizing carbon capture, effectively countering the rising tide of carbon dioxide (CO2) emissions responsible for global warming and climate instability. Their adaptability offers a tantalizing prospect, as they can be finely tailored for a multitude of applications, thereby encompassing the uncharted territory of potential DESs. Navigating this unexplored terrain underscores the vital need for predictive computational methods, which serve as our guiding compass in the expansive landscape of DESs. Thermodynamic modeling and solubility prognostications stand as our unwavering navigational aides on this treacherous odyssey. In this direction, the COSMO-RS model intertwined with the captivating Stochastic Gradient Boosting (SGB) algorithm. Together, they unveil the elusive truths pertaining to CO2 solubility in DESs, forging a path toward a sustainable future. Our quest is substantiated by two exhaustive datasets, a repository of knowledge encompassing 1973 and 2327 CO2 solubility data points spanning 132 and 150 distinct DESs respectively, encapsulating a spectrum of conditions. The SGB models, incorporating features derived from COSMO-RS, as well as accounting for pressure and temperature variables, furnishes predictions that harmonize seamlessly with experimental CO2 solubility values, boasting an impressive Average Absolute Relative Deviation (AARD) of a mere 0.85% and 2.30% respectively. When juxtaposed with literature-reported methodologies like different EoS, as well as Computational Solvation, and machine learning (ML) models, our SGB model emerges as the epitome of reliability, offering robust forecasts of CO2 solubility in DESs. It emerges as a potent tool for the design and selection of DESs for CO2 capture and utilization, heralding a sustainable and environmentally conscientious future in the battle against climate change.

Computational study on organochlorine insecticides extraction using ionic liquids

Insecticides pose hazardous environmental effects and can enter the food chain and contaminate water resources. Ionic liquids (ILs) have recently drawn much interest as environmentally friendly solvents and have been an efficient choice for extracting pesticides because of their outstanding thermophysical characteristics and tunable nature. In this study, ILs were screened using COSMO-RS (Conductor-like Screening Model for Real Solvents) to extract organochlorine insecticides from water at 289 K. A total of 165 ILs, a combination of 33 cations with five anions, were screened by COSMO-RS to predict the selectivity and capacity of the organochlorine insecticides at infinite dilution. The Organochlorine insecticide compounds, such as benzene hexachloride (BHC), Heptachlor, Aldrin, Gamma-Chlordane (γ-Chlordane), Endrin, and Methoxychlor are selected for this study. Charge density profiles show that Endrin and Methoxychlor compounds are strong H-bond acceptors and weak H-bond donors, while the rest of the compounds are H-bond donors with no H-bond acceptor potential. Moreover, it has been shown that ILs composed of halides and heteroatomic anions in conjunction with cations have enhanced selectivity and capacity for insecticides. Moreover, the hydrophobic phosphonium-based ILs have enhanced selectivity and capacity for insecticides. In BHC extraction, the selectivity of 1,3-dimethyl-imidazolium chloride was found to be the highest at 1074.06, whereas 2-hydroxyethyl trimethyl ammonium chloride exhibited the highest capacity being 84.0.1,3-dimethyl-imidazolium chloride exhibits the highest performance index, which is 57064.77. In addition, the ILs that have been chosen are well-recognized as environmentally friendly and very effective solvents to extract insecticides from water. As a result, this study evaluated that ILs could be promising solvents that may be further developed for the extraction of insecticides from contaminated water.

Boosting Graph Neural Networks with Molecular Mechanics: A Case Study of Sigma Profile Prediction

Sigma profiles are quantum-chemistry-derived molecular descriptors that encode the polarity of molecules. They have shown great performance when used as a feature in machine learning applications. To accelerate the development of these models and the construction of large sigma profile databases, this work proposes a graph convolutional network (GCN) architecture to predict sigma profiles from molecule structures. To do so, the usage of molecular mechanics (force field atom types) is explored as a computationally inexpensive node-level featurization technique to encode the local and global chemical environments of atoms in molecules. The GCN models developed in this work accurately predict the sigma profiles of assorted organic and inorganic compounds. The best GCN model here reported, obtained using Merck molecular force field (MMFF) atom types, displayed training and testing set coefficients of determination of 0.98 and 0.96, respectively, which are superior to previous methodologies reported in the literature. This performance boost is shown to be due to both the usage of a convolutional architecture and node-level features based on force field atom types. Finally, to demonstrate their practical applicability, we used GCN-predicted sigma profiles as the input to machine learning models previously developed in the literature that predict boiling temperatures and aqueous solubilities. Using the predicted sigma profiles as input, these models were able to compute both physicochemical properties using significantly less computational resources and displayed only a slight decrease in performance when compared with sigma profiles obtained from quantum chemistry methods.

Novel hybrid QSPR-GPR approach for modeling of carbon dioxide capture using deep eutectic solvents

In recent years, deep eutectic solvents (DESs) have garnered considerable attention for their potential in carbon capture and utilization processes. Predicting the carbon dioxide (CO2) solubility in DES is crucial for optimizing these solvent systems and advancing their application in sustainable technologies. In this study, we presented an evolving hybrid Quantitative Structure-Property Relationship and Gaussian Process Regression (QSPR-GPR) model that enables accurate predictions of CO2 solubility in various DESs. The QSPR-GPR model combined the strengths of both approaches, leveraging molecular descriptors and structural features of DES components to establish a robust and adaptable predictive framework. Through a systematic evolution process, we iteratively refined the model, enhancing its performance and generalization capacity. By incorporating experimental CO2 solubility data in varied DES compositions and temperatures, we trained the model to capture the intricate solubility behaviour precisely. The analytical capability of the evolving hybrid model was validated against an extensive dataset of experimental CO2 solubility values, demonstrating its superiority over individual QSPR and GPR models. The model achieves high accuracy, capturing the complex interactions between CO2 and DES components under varying thermodynamic conditions. The versatility of the evolving hybrid model was highlighted by its ability to accommodate new experimental data and adapt to different DES compositions and temperatures. The proposed QSPR-GPR model presented a powerful tool for predicting CO2 solubility in DES, providing valuable insights for designing and optimizing solvent systems in carbon capture technologies. The model's remarkable performance enhances our understanding of CO2 solubility mechanisms and contributes to sustainable solutions for mitigating greenhouse gas emissions. As research in DESs progresses, the evolving hybrid QSPR-GPR model offers a versatile and accurate means for predicting CO2 solubility, supporting advancements in carbon capture and utilization processes towards a greener and more sustainable future.

Sigma profiles in deep learning: towards a universal molecular descriptor

This work showcases the remarkable ability of sigma profiles to function as molecular descriptors in deep learning. The sigma profiles of 1432 compounds are used to train convolutional neural networks that accurately correlate and predict a wide range of physicochemical properties. The architectures developed are then exploited to include temperature as an additional feature.

A review on machine learning algorithms for the ionic liquid chemical space

There are thousands of papers published every year investigating the properties and possible applications of ionic liquids. Industrial use of these exceptional fluids requires adequate understanding of their physical properties, in order to create the ionic liquid that will optimally suit the application. Computational property prediction arose from the urgent need to minimise the time and cost that would be required to experimentally test different combinations of ions. This review discusses the use of machine learning algorithms as property prediction tools for ionic liquids (either as standalone methods or in conjunction with molecular dynamics simulations), presents common problems of training datasets and proposes ways that could lead to more accurate and efficient models.