Use of experimental design in the optimisation of a solid phase preconcentration system for Cobalt determination by GFAAS

Abstract

In this work is proposed a solid phase preconcentration system of Co 2+ ions and its posterior
determination by GFAAS in which fractional factorial design and response surface methodology (RSM)
were used for optimization of the variables associated with preconcentration system performance. The
method is based on cobalt extraction as a complex Co 2+ -PAN (1:2) in a mini-column of polyurethane foam
(PUF) impregnated with 1-(2-pyridylazo)-naphthol (PAN) followed by elution with HCl solution and its
determination by GFAAS. The chemical and flow variables studied were pH, buffer concentration, eluent
concentration and preconcentration and elution flow rates. Results obtained from fractional factorial
design 2 5-1 showed that only the variables pH, buffer concentration and interaction (pH X buffer
concentration) based on analysis of variance (ANOVA) were statistically significant at 95% confidence
level. Under optimised conditions, the method provided an enrichment factor of 11.6 fold with limit of
detection and quantification of 38 and 130 ng L -1 , respectively, and linear range varying from 0.13 to 10
μg L -1 . The precision (n = 9) assessed by relative standard deviation (RSD) was respectively 5.18 and
2.87% for 0.3 and 3.0 μg L -1 cobalt concentrations.