Application of multivariate analysis to the selection of test solutes for characterizing stationary phase selectivity in gas chromatography

Separation characteristics of wall-coated open-tubular columns for gas chromatography

The application of the solvation parameter model for the classification of wall-coated open-tubular columns for gas chromatography is reviewed. A system constants database for 50 wall-coated open-tubular columns at five equally spaced temperatures between 60 and 140 degrees C is constructed and statistical and chemometric techniques used to identify stationary phases with equivalent selectivity, the effect of monomer chemistry on selectivity, and the selection of stationary phases for method development. The system constants database contains examples of virtually all commercially available common stationary phases.

Application of the solvation parameter model to poly(methylcyanopropylsiloxane) stationary phases

The solvation parameter model has been applied to the specific retention volumes of 65 solutes of varied polarity on glass capillary columns coated with commercial and synthesized poly(methylcyanopropyl)siloxanes (CNPXX) with eight different percentages of cyanopropyl group (CNP). Their system constants were determined at 75, 90, 105 and 120 degrees C. The polymers examined do not either show any acidity (b = 0) or interact with solute pi/n electrons (e = 0); the prominent constants, dipolarity/polarizability and hydrogen-bond basicity, are of the same order (s approximately a), and the cavity formation/dispersive forces have normal values. Constants s, l and a decrease linearly with temperature for each cyanopropyl percentage. At each temperature, the constants s and a increase with polarity of polymer according to a curve, while the constant l decreases slightly. Cluster analysis shows that six CNPXX with medium to high cyanopropyl substitution integrate into a group with other high-polarity cyano-containing stationary phases taken from the literature, while the other three CNPXX with low CNP percentage form a group with other low-polarity stationary phases of different chemical nature. These clusters are supported by the dendrogram of 52 stationary phases made with the nine polymers presented here and other 43 taken from the literature.

System constants of synthesized poly(methyl-3,3,3-trifluoropropyl) siloxanes

The method of solvation model has been applied to five poly (methyl-trifluoropropyl) siloxanes (TFPSXX) prepared in our laboratories, at five trifluoropropyl (TFP) group contents, XX = 0, 11.5, 26.3, 35.5 and 50.0%, at 80, 100, 120 and 140 degrees C. Previously, specific retention volumes of 60-odd solutes of varied polarities were measured upon each of these stationary phases within the above temperature range. Constant s prevails over all other constants, TFPSXX stationary phases showing strong dipole/induced dipole forces with the solutes, moderate acidity and no basicity at all. Constant e is zero in the stationary phase without TFP groups, but has negative low-medium values for the other fluorine contents, XX from 11.5 to 50.0%, hinting at repulsive forces, as expected. Normal values for constant l, decreasing from the less cohesive TFPS00 to the more cohesive TFPS50, were found. For each TFP content constants s, a and l show a negative temperature dependence, while constant e increases as temperature increases. Constant c also decreases with increasing temperature. At each temperature, constants s and a increase with increasing %TFP (or increasing stationary phase polarity), whereas constants e and l show the opposite trend, diminishing with increasing polarity of the stationary phase. Principal component analysis shows that the five stationary phases presented in this work conform a group with other earlier synthesized trifluoropropyl siloxanes and other fluorinated stationary phases taken from literature: VB-210, QF-1, DB-200, DB-210 and PFS6, showing the same selectivity which only the fluorine atom confers. A dendrogram of 38 stationary phases supports these results.

Application of self-organizing maps for the classification of chromatographic systems and prediction of values of chromatographic quantities

The applicability of self-organizing maps (SOM) for the classification of chromatographic systems or components of chromatographic systems based on data taken from literature is shown. The SOM approach is compared to dendrogram and principal components analysis (PCA) approaches. It has been shown that the distance between classified objects could reveal linear correspondence with quantity to be optimized, e.g. resolution, so it can be applied in the chromatographic method development. SOMs can also be applied for prediction of chromatographic quantities. It is shown that SOM-based response surface modeling is comparable to triangular presentation of mobile phase composition response surfaces.

Development of comprehensive descriptors for multiple linear regression and artificial neural network modeling of retention behaviors of a variety of compounds on different stationary phases

A new series of six comprehensive descriptors that represent different features of the gas-liquid partition coefficient, K(L), for commonly used stationary phases is developed. These descriptors can be considered as counterparts of the parameters in the Abraham solvatochromic model of solution. A separate multiple linear regression (MLR) model was developed by using the six descriptors for each stationary phase of poly(ethylene glycol adipate) (EGAD), N,N,N',N'-tetrakis(2-hydroxypropyl) ethylenediamine (THPED), poly(ethylene glycol) (Ucon 50 HB 660) (U50HB), di(2-ethylhexyl)phosphoric acid (DEHPA) and tetra-n-butylammonium N,N-(bis-2-hydroxylethyl)-2-aminoethanesulfonate (QBES). The results obtained using these models are in good agreement with the experiment and with the results of the empirical model based on the solvatochromic theory. A 6-6-5 neural network was developed using the descriptors appearing in the MLR models as inputs. Comparison of the mean square errors (MSEs) shows the superiority of the artificial neural network (ANN) over that of the MLR. This indicates that the retention behavior of the molecules on different columns show some nonlinear characteristics. The experimental solvatochromic parameters proposed by Abraham can be replaced by the calculated descriptors in this work.