Quantitative structure—retention relationship analysis of nanoparticle compounds

Abstract

Genetic algorithm and partial least square (GA-PLS), the kernel PLS (KPLS) and Levenberg-<br />Marquardt artificial neural network (L-M ANN) techniques were used to investigate the correlation<br />between retention time (RT) and descriptors for 15 nanoparticle compounds which obtained by the<br />comprehensive two dimensional gas chromatography system (GC x GC). Application of the<br />dodecanethiol monolayer-protected gold nanoparticle (MPN) column was for a high-speed separation<br />as the second column of GC x GC. The L-M ANN model with the final optimum network<br />architecture of [9-4-1] gave a significantly better performance than the other models. This is the first<br />research on the quantitative structure—retention relationship (QSRR) of the nanoparticle compounds<br />using the GA-PLS, GA-KPLS and L-M ANN.