Orthogonal array design as a chemometric method for the optimization of analytical procedures. Part 3. Five-level design and its application in a polarographic reaction system for selenium determination

Abstract

The theory and methodology of a five-level orthogonal array design for the optimization of analytical procedures were developed. In the theoretical section, the construction and characteristics of the OA₂₅(5⁶) matrix are described in detail while orthogonality is proved by means of a fourth-order polynomial model. Next, the assignment of experiments in the OA₂₅(5⁶) matrix is illustrated, followed by the data analysis strategy, in which significant/non-significant influence for each factor is quantitatively evaluated by the analysis of variance (ANOVA) technique including the percentage contribution, and the differences among five levels for each factor that has a significant influence are determined by Duncan's multiple F-test. Finally, the response surface methodology for the OA₂₅(5⁶) matrix is developed by using the equations obtained from the proof on orthogonality. In the application section, the reaction system for selenium determination by differential-pulse polarography, as a practical example, is employed to demonstrate the application of the OA₂₅(5⁶) matrix in the area of analytical chemistry.