Method for predicting organic compound and hydroxyl radical reaction rate constant in water phase

Abstract

The invention discloses a method for predicting organic compound and hydroxyl radical reaction rate constant in a water phase based on a quantitative structure-activity relationship. On the base of obtaining a compound molecular structure, a quantitative structure-activity relationship model is built by calculating a molecular structure descriptor through a multi-linear-regression method, and the water phase kOH value of organic compounds can be predicted quickly and efficiently. According to the method, modeling is conducted according to the building and using guide rules of a QSAR model of an economic cooperation and development organization, the model application field is clear, 526 kinds of different organic compounds are included, and particularly compounds containing most compounds of atoms such as nitrogen, phosphorus, and sulphur which are not contained in previous models are covered. The model has good fitting capacity, robustness and predicting capacity, the water phase kOH value of the organic compounds in an application field can be effectively predicted, and an important data support is provided for environmental durability evaluation and ecological risk assessment of organic chemical products.

Description

technical field [0001] The invention relates to a method for predicting the reaction rate constant of an organic compound and a hydroxyl radical in an aqueous phase by establishing a quantitative structure-activity relationship model (QSAR), and belongs to the field of ecological risk assessment test strategies. Background technique [0002] Hydroxyl radicals (·OH) are a common active species in the water environment, which can be generated through photochemical transformation processes of dissolved substances in water (such as dissolved organic matter) or artificial advanced oxidation processes (such as Fenton system). The standard oxidation-reduction potential of OH is 1.9V, which is an important strong oxidizing agent, which can oxidize and degrade organic pollutants. The second-order reaction rate constant (k OH ) is a parameter that characterizes the intensity and ability of organic pollutants to react with OH, and is also an important indicator for evaluating the pers...

AI Technical Summary

Problems solved by technology

Second, some models use experimentally determined parameters as descriptors, which limits the application of the model, such as the literature "Physical Chemistry Chemical Physics.2004,6(16):4118-4126", "International Journal of Chemical Kinetics.2008, 40(4), 174-188." Using the experimentally determined bond dissociation energy (BDE) to predict the aqueous phase k of the compound OH , its popularization and application is largely limited by the lack of BDE experimental values

Third, the model is not transparent enough, such as the water phase k based on molecular structure fragments and artificial neural network (ANN) in the literature "Atmospheric Environment.2003,37(2):269-276" OH prediction model, but because the neural network model is a "black box model", it cannot use k OH The relationship with the model parameters is presented with a clear mathematical expression, and it is difficult to explain the mechanism of the model, so it is not suitable for acceptance and promotion

Fourth, some models, such as the model constructed in the literature "Atmospheric Environment.2005,39(40):7667-7688", have good fitting performance, but the model verification and characterization are not comprehensive, such as the lack of internal and external verification , model application domain characterization, etc., do not meet the requirements of OECD for the construction and use of QSAR models

In the classification model, the prediction model of some compounds is generally better, but there are also categories with unsatisfactory prediction results (such as low fitting R value), and the model form is complex, which is not easy to apply

Images