Using machine learning and quantum chemistry descriptors to predict the toxicity of ionic liquids

Cao, Lingdi; Zhu, Peng; Zhao, Yongsheng; Zhao, Jihong

doi:10.1016/j.jhazmat.2018.03.025

Cited by 84 publications

(36 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To determine the predictive capacity of the dataset ( N m =119), we split the molecules into training and test sets, making sure that the training set spanned the entire response variable space and that the test set included ≈15 % of the total dataset . To create a predictive QSAR model, we tested three different algorithms for dividing molecules: (i) random selection, (ii) cluster by rank, in which molecules were sorted by Δ G TS1 values into a specific number of groups and then randomly selected from each group for the test set, and (iii) modified cluster by rank, in which a rational division of molecules was executed to ensure a reasonable dispersion of conjugate acid p K a values within the training and test sets.…”

Section: Resultsmentioning

confidence: 99%

Screening for Improved Nerve Agent Simulants and Insights into Organophosphate Hydrolysis Reactions from DFT and QSAR Modeling

Mendonca

Snurr

2019

Chemistry A European J

View full text Add to dashboard Cite

The effective detoxification of chemical warfare agents, specifically nerve agents, is a pressing issue in the modern world. Due to the high toxicity of these molecules, simulants are often used in experiments as substitutes for the agents. However, there is little reason to believe that the current simulants used in the literature are optimal predictors of nerve agent reactivity. Density functional theory calculations were performed on the alkaline hydrolysis of over 100 organophosphate molecules to identify improved simulants for the G‐series nerve agents soman and sarin, based on low toxicity and similarity to nerve agent hydrolysis energetics and degradation mechanism. This screening highlighted 5 molecules that have nearly identical reaction barriers to the actual agents, while being far less toxic. Quantitative structure–activity relationship (QSAR) models were also derived to determine the most significant molecular descriptors for describing the hydrolysis free energy barriers of these reactions. The optimal QSAR model was subjected to a thorough statistical analysis and validation procedure to confirm its predictive capacity, showing excellent quantitative and ranking accuracy. It was further shown that the model trained on G‐series agents can reliably predict energetics for other organophosphate classes as well, including VX. Through these computational insights, experimentalists may be aided in accurately and safely studying these reactions with less toxic simulants.

show abstract

Section: Resultsmentioning

confidence: 99%

Screening for Improved Nerve Agent Simulants and Insights into Organophosphate Hydrolysis Reactions from DFT and QSAR Modeling

Mendonca

Snurr

2019

Chemistry A European J

View full text Add to dashboard Cite

show abstract

“…Nowadays, many experimental studies [21,33] have been conducted to evaluate the environmental impact of ILs. In addition, computational approaches [34][35][36] are also employed to assess their toxicity. In general, cations and anions do have an influence on the ILs toxicity particularly, cations have a greater impact than anions.…”

Section: Biodegradability and Toxicitymentioning

confidence: 99%

Ionic Liquids as High-Performance Lubricants and Lubricant Additives

Guo¹,

Iglesias²

2021

Ionic Liquids - Thermophysical Properties and Applications

View full text Add to dashboard Cite

Taking into account the environmental awareness and ever-growing restrictive regulations over contamination, the study of new lubricants or lubricant additives with high performance and low toxicity over the traditional lubes to reduce the negative impact on the environment is needed. In this chapter, the current literature on the use of ionic liquids, particularly protic ionic liquids, as high-performance lubricants and lubricant additives to different types of base lubricants are reviewed and described. The relation between ionic liquids structures and their physicochemical properties, such as viscosity, thermal stability, corrosion behavior, biodegradability, and toxicity, is elaborated. Friction reduction and wear protection mechanisms of the ionic liquids are discussed with relation to their molecular structures and physicochemical properties.

show abstract

“…Stolte et al [37] tested the toxicity of a number of imidazolium salts. Cao et al, Erfurt et al [38,39] studied cytotoxicity of selected ILs. However, the imidazolium and ammonium ILs presented in this work were not completely studied.…”

Section: Introductionmentioning

confidence: 99%

Antibacterial, Antifungal and Ecotoxic Effects of Ammonium and Imidazolium Ionic Liquids Synthesized in Microwaves

et al. 2020

View full text Add to dashboard Cite

Ionic liquids are increasingly used for their superior properties. Four water-immiscible ionic liquids (butyltriethylammonium bis(trifluoromethylsulfonyl)imide, octyltriethylammonium bis(trifluoromethylsulfonyl)imide, dodecyltriethylammonium bis(trifluoromethylsulfonyl)imide, butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide) and their water miscible precursors (bromides) were synthesized in a microwave reactor and by conventional heating. The best conditions for microwave-assisted synthesis concerning the yield and the purity of the product are proposed. The heating in the microwave reactor significantly shortened the reaction time. Biocide and ecotoxic effects of synthesized ionic liquids and their precursors were investigated. All tested compounds had at least a little effect on the growth or living of microorganisms (bacteria or mold). The precursor dodecyltriethylammonium bromide was found to be the strongest biocide, but posed a risk to the aquatic environment due to its relatively high EC50 value in the test with Vibrio fischeri. We assumed that apart from the alkyl chain length, the solubility in water, duration of action, or type of anion can influence the final biocide and ecotoxic effect.

show abstract

Using machine learning and quantum chemistry descriptors to predict the toxicity of ionic liquids

Cited by 84 publications

References 46 publications

Screening for Improved Nerve Agent Simulants and Insights into Organophosphate Hydrolysis Reactions from DFT and QSAR Modeling

Screening for Improved Nerve Agent Simulants and Insights into Organophosphate Hydrolysis Reactions from DFT and QSAR Modeling

Ionic Liquids as High-Performance Lubricants and Lubricant Additives

Antibacterial, Antifungal and Ecotoxic Effects of Ammonium and Imidazolium Ionic Liquids Synthesized in Microwaves

Contact Info

Product

Resources

About