Four-level orthogonal array design as a chemometric approach to the optimization of polarographic reaction system for phosphorus determination

Abstract

It is the purpose of the present work to provide information on the four-level orthogonal array design and data analysis for the optimization of analytical procedures. In the theoretical part, the construction and characteristics of the OA16(45) matrix is described in detail, followed by the data analysis strategy, in which the significance of the different factors is quantitatively evaluated by an analysis of variance (ANOVA) method including per cent contribution, and the difference among four levels for each factor is determined by Duncan's multiple F test. Furthermore, a third-order polynomial model representing response surface is established to estimate the effects for the factors with significant influences. In the application part, the proposed four-level orthogonal array design and data analysis method were applied to optimize polarographic reaction system for phosphorus determination. By conducting 16 preplanned experiments that span the maximum working range of the system, the best experimental conditions for achieving the largest response can be obtained. The expected value for each experimental trial calculated by the third-order regression equation established is in good agreement with the corresponding experimental value. To confirm the validity of the optimization procedure, additional experiments using the recommended conditions were performed. The results demonstrate that satisfactory results can be acquired. Therefore, the proposed four-level orthogonal array design as a chemometric approach to optimize the polarographic reaction system for phosphorus determination is rather efficient and effective. © 1994.