Article
Chemistry, Physical
Ncomeka Mgxadeni, Oarabile Mmelesi, Bakusele Kabane, Indra Bahadur
Summary: Deep eutectic solvents (DESs) are cost-effective and easy-to-prepare solvents that play a crucial role in designing separation processes. This study utilized gas liquid chromatography technique (GLC) to determine the activity coefficients of 34 organic solutes in two DESs based on zinc chloride. The results showed that the DESs had high selectivity values for difficult separation systems, such as benzene/ethanol and cyclohexane/ethanol, making them suitable as extraction solvents.
JOURNAL OF MOLECULAR LIQUIDS
(2022)
Article
Energy & Fuels
Sergio M. Vilas-Boas, Monia A. . R. . Martins, Fabio R. . Tentor, Gabriel Teixeira, Juliana G. Sgorlon, Joao A. . P. Coutinho, Olga Ferreira, Simao P. Pinho
Summary: This study investigated the potential of imidazolium chloride ionic liquids (ILs) as separation agents in chemical separations. The activity coefficients were calculated to determine the selectivities, capacities, and solvent performance indices of different imidazolium chloride ILs. The results showed that the proportion of cations containing larger alkyl chains can be adjusted to meet specific purposes, and the COSMO-RS thermodynamic model was found to be a promising tool for qualitative screening of potential separation agents.
Article
Mechanics
Andrew Y. K. Hong, Michael A. Gallis
Summary: Recently, collision-specific parameters proposed for direct simulation Monte Carlo simulations were tested for binary mixtures of nitrogen, oxygen, and argon. The traditional collision-averaged parameters were found to be highly accurate under near ambient conditions, while the collision-specific parameters were not. However, simulating transport using the collision-averaged parameters for helium mixtures was inaccurate. Therefore, a novel method combining the Chapman-Enskog theory with empirical mixing rules and experimental data was proposed to determine molecular parameters. The optimized parameters greatly improved the accuracy of the simulated transport for binary mixtures of nitrogen, oxygen, and argon, as well as for helium mixtures.
Article
Thermodynamics
Jonathan M. Bonebrake, Viswanath R. Katta, Timothy M. Ombrello, David L. Blunck
Summary: Ignition of fuel/oxidizer mixtures is essential for advanced combustor operation. The presence of NO can decrease ignition delay times, but its influence on forced ignition is unclear. This study investigates ignition kernels generated by spark discharges in CH4/air flows diluted with NO and N2. It is found that the presence of NO increases the size of ignition kernels compared to mixtures diluted with N2. Numerical modeling reveals that low-temperature chemistry plays a role in this sensitivity. (c) 2023 The Combustion Institute. Published by Elsevier Inc. All rights reserved.
COMBUSTION AND FLAME
(2023)
Article
Engineering, Chemical
Thomas Brouwer, Sascha R. A. Kersten, Gerrald Bargeman, Boelo Schuur
Summary: The method utilizes the Margules equation to predict vapor-liquid equilibria in ternary systems, providing accurate pre-selection for non-hydrogen bond donating mixtures and highly dissimilar mixtures. By identifying molecular solvents and potential alternative solvents, it offers realistic pre-selection outcomes for industrial applications. The method also highlights the potential of morpholinium and ammonium structures in increasing relative volatilities, with lower toxicity compared to other cations.
SEPARATION AND PURIFICATION TECHNOLOGY
(2021)
Article
Engineering, Chemical
Thomas Specht, Hans Hasse, Fabian Jirasek
Summary: In chemical engineering processes, it is often difficult to determine the specific composition of mixtures. A method using NMR experiments has been developed to quantify the group composition of poorly specified mixtures. This method, combined with thermodynamic group-contribution methods, has shown excellent agreement in simulations of fluid separation processes.
INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH
(2023)
Article
Chemistry, Multidisciplinary
Hossein Jalaei Salmani, Hamed Karkhanechi, Mohammad Reza Moradi, Hideto Matsuyama
Summary: This study thermodynamically investigates the binary mixtures of ethylenediamine (EDA) with water, methanol, ethanol, and 2-propanol using the CPA equation of states, demonstrating its capability in accurately describing the thermodynamic behavior of the studied mixtures.
Article
Multidisciplinary Sciences
Aliaksandr Mialdun, Mounir Bou-Ali, Valentina Shevtsova
Summary: The Soret effect explains the transport of constituent species in multicomponent mixtures due to temperature gradient, and the introduction of the Soret vector presents a new concept for characterizing Soret-driven separation in ternary mixtures. The vector form offers advantages such as predicting Soret sign, controlling coefficient consistency, determining components causing separation, and identifying regions where Soret separation is inaccessible for certain techniques. This approach provides a systematic framework for future research on the challenging problem of thermodiffusion in multicomponent liquids.
SCIENTIFIC REPORTS
(2021)
Article
Engineering, Electrical & Electronic
Saurabh Parmar, Bishakha Ray, Suwarna Datar
Summary: Polymer modified Quartz tuning fork based sensors are a versatile replacement for conventional sensors in electronic nose applications. A sensor array was developed to detect mixtures of styrene, propylbenzene and acetone, achieving an overall accuracy of 98% after modification in the prediction process. This work demonstrates the potential of using QTF sensor array and suitable classifier for detecting binary and ternary mixtures in various electronic nose applications.
SENSORS AND ACTUATORS A-PHYSICAL
(2021)
Article
Automation & Control Systems
Dingchao Fan, Wenguang Zhu, Yusen Chen, Ke Xue, Tianxiong Liu, Peizhe Cui, Jianguang Qi, Zhaoyou Zhu, Yinglong Wang
Summary: In this study, deep learning models were used to predict the infinite dilution activity coefficient for ionic liquid-solute systems. The models were constructed using a large dataset and parameters were optimized using Bayesian optimization algorithm and cross validation. The results showed a high agreement between the predicted values and experimental values.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Environmental Sciences
Bertille Bonnaud, Cecile Miege, Amandine Daval, Vincent Fauvelle, Nicolas Mazzella
Summary: The diffusive gradient in thin film technique was applied to measure organic compounds in this study. Two methods for measuring the diffusional coefficient (D) were compared for 112 organic compounds. The D-stack method tends to be higher than the D-cell method, and is considered as a reference value for diffusion in gel.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2022)
Article
Mechanics
Jiandong Deng, Yanan Zhang, Guozhu Jia
Summary: Modeling predictions of surface tension for binary and ternary liquid mixtures is challenging, but this study proposes a machine learning model that accurately predicts the surface tension of such mixtures. The model combines machine learning algorithms, UNIFAC, ASP, and SHAP to analyze and characterize the effect of parameters. Among various algorithms, CatBoost performs the best, with MAE = 0.3338, RMSE = 0.7565, and R-2 = 0.9946. The SHAP results reveal that the surface tension decreases with an increase in the volume and surface area of the anion. This work not only provides accurate predictions for surface tension, but also offers insights into microscopic interactions and properties.
Article
Chemistry, Physical
Noora Hyttinen
Summary: Liquid-liquid phase separation (LLPS) affects the water uptake of aerosol particles in the atmosphere through Kelvin and Raoult effects. This study used a computational method, COSMO-RS, to investigate LLPS in ternary mixtures containing water and two organic compounds. The computations found LLPS in all of the studied mixtures, due to the limited solubility of the hydrophobic POA proxies in water. Experimental observations, however, did not observe the additional three-phase states predicted by the computations, likely due to the lower relative humidity used in the experiments.
PHYSICAL CHEMISTRY CHEMICAL PHYSICS
(2023)
Article
Engineering, Environmental
Sergio M. Vilas-Boas, Debora C. do Nascimento, Rafael M. Dias, Gabriel H. Rozo, Antonio M. Barbosa Neto, Olga Ferreira, Simao P. Pinho, Mariana C. da Costa
Summary: The flash point (FP) is a critical parameter for fire risk mitigation and prevention in the chemical and oil industry. This study provides experimental FP measurements and calculation models for binary and ternary mixtures of four different monoterpenes, highlighting the importance of considering nonideal effects in estimating the FP data of terpene mixtures. These findings offer essential information for accurate fire risk assessment in the essential oil industry.
PROCESS SAFETY AND ENVIRONMENTAL PROTECTION
(2023)
Article
Engineering, Chemical
Zhengtao Zhou, Mario Eden, Weifeng Shen
Summary: QSPR modeling is a widely used method for estimating molecular properties based on structural information, and it has been applied in exploring new solvents, pharmaceuticals, and materials with desired properties. SMILES is considered as a chemical language, and a deep pyramid convolutional neural network architecture is constructed to extract information from SMILES sentences. The effectiveness of this approach is proven through a case study of predicting the logarithm values of the octanol-water partition coefficient, showing better performance compared to a precedent reference model and providing insights for molecular information mining and exploration of chemical property space through natural language processing technologies.
INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH
(2023)