This review is an update of a previous review in 2009 and covers publications from 2009 to 2019. The review focuses on experimental design, referred to as the design of experiments (DoE), used in developing bioanalytical solid-phase microextraction (SPME) methods. Characteristics of different SPME approaches are illustrated and critically discussed. The literature selection evidences that two-level full factorial designs, with a limited number of factors (<5), are most frequently used for preliminary factors screening. When applying the response surface methodology for the quantitative assessment of factorial effects, few quadratic models were used. The most popular were the rotatable central composite and Box-Benkhen designs. Models including more than four factors, such as fractional factorial designs (including the Plackett-Burman and Taguchi designs), were rarely used. Definitive screening and D-Optimal designs were not reported anywhere in the literature selection. When examining the diagnostic criteria used to evaluate different model's quality and validity, it was apparent the researchers relied heavily on commercial software for experimental design, analysis, and reporting of the results.
Experimental designs for solid-phase microextraction method development in bioanalysis: A review
Dugheri S.;
2020-01-01
Abstract
This review is an update of a previous review in 2009 and covers publications from 2009 to 2019. The review focuses on experimental design, referred to as the design of experiments (DoE), used in developing bioanalytical solid-phase microextraction (SPME) methods. Characteristics of different SPME approaches are illustrated and critically discussed. The literature selection evidences that two-level full factorial designs, with a limited number of factors (<5), are most frequently used for preliminary factors screening. When applying the response surface methodology for the quantitative assessment of factorial effects, few quadratic models were used. The most popular were the rotatable central composite and Box-Benkhen designs. Models including more than four factors, such as fractional factorial designs (including the Plackett-Burman and Taguchi designs), were rarely used. Definitive screening and D-Optimal designs were not reported anywhere in the literature selection. When examining the diagnostic criteria used to evaluate different model's quality and validity, it was apparent the researchers relied heavily on commercial software for experimental design, analysis, and reporting of the results.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
