Separation of very hydrophobic analytes by MEKC. II. Carbon number equivalents as analyte descriptors – influence of the composition of the separation electrolyte

Abstract

The use of the parameter carbon number equivalent, N(c)(*), as analyte descriptor for neutral hydrophobic solutes in EKC is investigated. The carbon number equivalent of an analyte is defined analogously to the retention index, RI, however, it is directly calculated from the regression line plotting for the members of a homologous series the logarithm of the retention factor against the number of methylene units N(c) in the alkyl chain. This plot can be obtained by using simultaneously an iteration procedure for the calculation of t(MC). For separation electrolytes containing SDS as micellar phase employing ACN, urea, and CaCl(2) as additives within a wide span of concentrations the parameter N(c)(*) is quasi-independent of the composition of the separation electrolyte, being of importance concerning identification of analytes and improving repeatability and reproducibility. With known octanol-water partition coefficients for the members of the homologous series taken as reference it is possible to transform the conventional time scale of an electropherogram into a scale of N(c)(*) or a scale of log P(OW). Due to the quasi-independence of N(c)(*) from the composition of the separation electrolyte these transformed electropherograms can be used as normalized electropherograms for identification purposes. It is also shown that the linear dependence of log P(OW) on N(c) can be the basis of the determination of octanol-water partition coefficients of solutes within a very large range (-4 < or = log P(OW) < or = 9).

Separation of very hydrophobic analytes by micellar electrokinetic chromatography. III. Characterization and optimization of the composition of the separation electrolyte using carbon number equivalents

“…Taking alkyl phenones as reference solutes, N Ã C for a number of sassafras oil constituents were calculated for different compositions of the separation electrolyte (see Section 3) [29]. The resulting mean values and RSDs are given in Table 1.…”

“…However, in order to simplify further calculations we did not transfer retention data into retention indices but used carbon number equivalents N Ã C . In a previous publication [29], it was shown that for the analytes selected taking the homologous series of alkyl phenones as reference N Ã C is quasi-independent of the composition of the separation electrolyte regarding type and concentration of the modifier. Other publications [36,37] showed the independence of retention indices from the concentration of the PSP and the temperature.…”

“…This paper also addresses the optimization of the composition of the sample injection solution. In the second part of this series [29] we investigated the use of carbon number equivalents N Ã C as analyte descriptors. It was shown that for the analytes selected taking the homologous series of alkyl phenones as reference carbon number equivalents are quasi-independent of the composition of the separation electrolyte with RSD 4% for a wide parameter range varying the concentration of urea, the volume fraction of ACN, and the concentration of CaCl 2 .…”

“…In ref. [29], c = 1.229 and d = 0.502 were obtained via linear regression of log P OW plotted against N c for alkyl phenones, for log P OW = c 1 d6N c , with log P OW taken from (www.vcclab.org/lab/alogps, 23.10.2006). With these parameters, carbon number equivalents can be obtained directly from octanol-water partitioning coefficients.…”

Marker compounds for the determination of retention factors in EKC

“…The slope b of the regression plot shows the methylene selectivities that represent an average for the entire homologous series 106, 91. Huhn et al 111, 112, 119 proposed to use regression parameters together with determined values of t 0 and t ps as a set of parameters to characterize the properties of the pseudostationary phase. Consequently, the influence of additives on the properties of the BGE solution can be characterized using these four parameters.…”