QSAR/QSPR Study on Some Pesticides and Fluorobenzene Derivatives
|School||Jiangxi University of Technology|
|Keywords||Quantitative Structure Property/Activity Relationships (QSAR/QSPR) pesticides fluorobenzene derivatives water solubility (–lgSw) toxicity (–lgEC50)|
Pesticides and fluorobenzene derivatives are often characterized by low water solubility, high lipid solubility, semi-volatile and so on. Once they enter the environment, they are permitted to be adsorbed in soil, either to be vaporized or to be adsorbed on atmospheric particles. They distribute among the soil, water and gas. They have long-term and serious impact on environment because of their corresponding degradation, adsorption, desorption, volatilization and leaching, such as biochemical and physico-chemical changes. Because of pesticides and fluorobenzene derivatives with bioaccumulation, carcinogenicity, environmental sustainability and the potential harm to human health, they become global concern of organic pollutants. The study of these compounds’activity and property has great environmental and practical significance.Quantitative structure activity/property relationships (QSAR/QSPR) which meet the needs of our study in organic pollutants show that there are quantitative functional relationships between the molecular structure descriptor and various physical and chemical activity/property. It is a cross subject covering chemistry, biology, mathematics and computer science, etc. It has become an important means for evaluating ecological risk of the transference, transformation and distribution behavior in environment for organic pollutants. With the development of the computer, QSAR/QSPR will be more comprehensive, theoretical and procedural.This thesis mainly studied QSAR/QSPR of pesticides and fluorobenzene derivatives. First of all, the experimental values of toxicity (–lgEC50) to Vibrio qinghaiensis sp. Nov. (Q67) and water solubility (–lgSw) of pesticides were determined. And the experimental values of toxicity (–lgEC50) of fluorobenzene derivatives to photobacterium phosphoreum T3 and Vibrio qinghaiensis sp. Nov. (Q67),–lgSw and n-octanol/water partition coefficient (lgKow) were also determined. Then quantum chemistry parameters of pesticides and fluorobenzene derivatives were computed using density functional theory (DFT) method. Based on the modified theoretical linear solvation energy relationship model, the obtained structural and thermodynamic parameters were taken as theoretical descriptors to establish the QSAR/QSPR models by using GQSARF 2.0 and SPSS 12.0 for Windows programs. The obtained models are as follows:(1)Optimized calculation of 30 pesticides’molecules were carried out at B3LYP/6-31G* level with DFT method. The obtained parameters were taken as theoretical descriptors to establish the–lgEC50 model, in which the conventional correlation coefficient (R2) is 0.970, the LOO cross-validation correlation coefficient (q2) is 0.952, and the standard error (SE) is 0.096.(2)The–lgSw of pesticides were determined by shake-flask method. The obtained parameters were taken as theoretical descriptors to establish the–lgSw model, in which the R2 is 0.963, the q2 is 0.939, and the SE is 0.153. (3)Optimized calculation of 28 fluorobenzene derivatives’molecules were carried out at B3LYP/6-311G** level with DFT method. The–lgEC50 to photobacterium phosphoreum T3 and Vibrio qinghaiensis sp. Nov. (Q67), the obtained parameters were taken as theoretical descriptors to establish the–lgEC50 models, in which the R2 are 0.981 and 0.953, the q2 are 0.972 and 0.931, and SE are 0.060 and 0.066, respectively.(4)The obtained parameters were taken as theoretical descriptors to establish the lgKow and–lgSw models, in which the R2 are 0.966 and 0.906, the q2 are 0.931 and 0.867, and the SE are 0.083 and 0.244, respectively.Statistical test and evaluation for the obtained models were performed with LOO cross-validation, t-test, F-test and variation inflation factors (VIF). The results indicated that all of these models have good correlation and stability and can be used to predict activity and property of similar compounds.