A sucrose octaacetate (SOA) gradient HPLC evaporative light scattering detection (ELSD) and low-wavelength UV-diode array detection (UV-DAD)-specific stability-indicating method development and validation comparison is reported. A central composite response surface design and multicriteria optimization was used to maximize molten SOA area-under-the-curve response and signal-to-noise ratio. The ELSD data were also analyzed using multivariate principal component analysis, analysis of variance, and standard least squares effects modeling. The method suitability and validation parameters of both methods were compared. To the authors' knowledge, this is the first report that validates an ELSD method using a molten analyte. SOA exhibited a low molar absorptivity of 439 absorption units/cm/M in water at 210 nm requiring low-wavelength UV-DAD detection. The low-wavelength UV-DAD method provided substantially better intraday and interday precision, intraday and interday goodness-of-fit, detection limit, and quantitation limit than ELSD. ELSD exhibited a 60-fold greater area-under-the-curve response, better resolution, and 58% more theoretical plates. On balance, the UV-DAD method was chosen for SOA chemical kinetic studies. This study illustrates that ELSD may not always be the best alternative to gradient HPLC low-wavelength UV detection.
Keywords: design of experiments; evaporative light scattering detection; modeling; molten analyte; multicriteria optimization; multivariate analysis; pharmaceutical analysis; principal component analysis; specific stability-indicating method; sucrose octaacetate; ultraviolet diode array detection.
Copyright © 2016. Published by Elsevier Inc.