Asensi-Bernardi L, Escuder-Gilabert L, Martín-Biosca Y, Medina-Hernández MJ, Sagrado S. Modeling the chiral resolution ability of highly sulfated β-cyclodextrin for basic compounds in electrokinetic chromatography.
J Chromatogr A 2013;
1308:152-60. [PMID:
23958689 DOI:
10.1016/j.chroma.2013.08.003]
[Citation(s) in RCA: 19] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/18/2013] [Revised: 07/31/2013] [Accepted: 08/01/2013] [Indexed: 11/30/2022]
Abstract
Despite the fact that extensive research in the field of enantioseparations by capillary electrophoresis has been carried out, it is difficult to predict whether a concrete chiral selector would be useful for the separation of a racemic compound. Hence, several experimental effort is necessary to test the abilities of individual chiral selectors, usually by trial and error procedures. Thus, the enantioseparation of a new racemate becomes a time- and money-consuming task. In this work, the ability of highly sulfated β-cyclodextrin (HS-β-CD) as chiral selector in electrokinetic chromatography (EKC) is modeled for the first time, using exclusively directly-available structural data of forty compounds (structurally unrelated basic drugs and pesticides). A discriminant partial least squares (PLS)-based quantitative structure-property relationship (QSPR) approach is simplified, resulting in a consistent, predictive and descriptive model. It is converted into an explicit equation able to predict the enantioresolution level (Rs) of new compounds, from four structure properties available in an on-line open database: logarithm of octanol-water partition coefficient estimated at pH 7.4 (lgD), polar surface area (PSA), number of hydrogen bond donors (HBD) and acceptors (HBA). For the cases in which the model predicts good Rs only in concrete experimental conditions, a Box-Behnken experimental design is proposed for the fast PLS-based optimization of the most influential experimental variables: cyclodextrin concentration, temperature and pH.
Collapse