Application of response surface methodology (RSM) to the optimization of a post-column luminol chemiluminescence analysis of silyl peroxides

The possibility of the utilization of chemiluminescence post-column luminol oxidation (CL) in a HPLC system for silyl peroxides analysis has been investigated. The conditions of HPLC separation for 12 silyl peroxides, representing bissilyl and alkyl-silyl peroxides, as well as their potential impurities, were established. Optimal chemiluminescent post-column reaction conditions were found using central composite design (CCD) and response surface methodology (RSM). The interaction effects of four of the most important operating variables - the concentrations of luminol, hemin, sodium hydroxide and the post-column solution flow rate - on the light intensity were evaluated. The optimized conditions for analysis were the same for bissilyl and alkyl-silyl peroxides for the base concentration (0.03 M), the luminol concentration (0.4 g L(-1)) and the hemin concentration (0.3 g L(-1)). The only differences occurred in a reagent flow rate (for bissilyl peroxide -0.3 mL min(-1) and for alkyl-silyl peroxides -0.9 mL min(-1)). Under optimal conditions, the detection limits were in the 0.07-0.16 nM range for bissilyl, and 0.53-1.01 for alkyl-silyl peroxides. The calibration curves were linear in the 0.25-3 nM range for bissilyl and the 2.5-25 range for alkyl-silyl peroxides. Intra-day and inter-day precision was lower than 5.5% for each tested concentration level. A mechanism of luminol oxidation by silyl peroxides involving a hydrolysis step with the formation of hydrogen peroxide or hydroperoxide was proposed.