【标题】Quantitativestructure-property relationships on dissolvability of PCDD/Fs usingquantum chemical descriptors and partial least squares.

【期刊】Journalof Theoretical and Computational Chemistry

【第一作者】卢桂宁

【摘要】Theenvironmental fate of polychlorinated dibenzo-p-dioxins andpolychlorinated dibenzofurans (PCDD/Fs) has become a major issue inrecent decades. Quantitative structure-property relationship (QSPR)modeling is a powerful approach for predicting the properties ofenvironmental organic pollutants from their structure descriptors. Inthis study, QSPR models were established for estimating watersolubility (- log SW) and n-octanol/water partition coefficient (logK-OW) of PCDD/Fs. Quantum chemical descriptors computed with densityfunctional theory at the B3LYP/6-31G(d) level and partial leastsquares (PLS) analysis with an optimizing procedure were used togenerate QSPR models for - log SW and log K-OW of PCDD/Fs. Optimizedmodels with high correlation coefficients (R-2 > 0.983) wereobtained for estimating - log S-W and log K-OW of PCDD/Fs. Both theinternal cross validation test (Q(cum)(2) > 0.978) and externalvalidation test (R-2 > 0.965) results showed that the obtainedmodels had high-precision and good prediction capability. The - logS-W and log K-OW values predicted by the obtained models are veryclose to those observed. The PLS analysis indicated that PCDD/Fs withlarger electronic spatial extent (R-e), lower molecular total energy(E-T), and smaller energy gap between the lowest unoccupied and thehighest occupied molecular orbitals (E-LUMO - E-HOMO) tend to be lesssoluble in water but more lipophilic.

【文章链接】http://apps.webofknowledge.com/full_record.do?product=UA&search_mode=GeneralSearch&qid=4&SID=6FvFjW868Mgf3segmqk&page=1&doc=1