(431e) Exploration of the Retention Mechanism in Supercritical Fluid Chromatography by Molecular Simulations

This presentation discusses molecular insights on the retention mechanism for prototypical analytes in supercritical fluid chromatography (SFC) and a comparison to the retention mechanism in reversed-phase liquid chromatography (RPLC). Gibbs ensemble Monte Carlo simulations are used to investigate the structure of the retentive phase in contact with different mobile phase compositions, temperatures, and pressures. The retentive phase model is a planar slit pore (width of 9 nm) with dimethyl octadecyl silane (ODS) ligands grafted to a cristobalite surface at a grafting density of 2.9 μmol/m² without end-capping that matches the one used in prior studies of RPLC retention [Rafferty et al., J. Chromatogr. A, 1218 (2011) 2203]. The mobile phases investigated are CO₂/methanol mixtures (0_, 5, and 10 mol% methanol in CO₂). The analyte molecules consist of C₁-C₆ linear alkanes and alcohols, and the conditions examined are at 293 (a subcritical temperature), 313, and 333 K with pressures of 15 and 30 MPa at each temperature. The molecular models are validated by comparison to experimental vaporliquid coexistence data for bulk CO₂/methanol mixtures.

The simulations indicate that the SFC retention of a methylene group (separating the homologous series of n-alkanes) is significantly weaker than in RPLC because the CO₂/methanol mixtures are much stronger solvents for the CH₂ group than neat water or an aqueous mobile phase with up to 50% v/v methanol or acetonitrile. In addition, the solubility of CO₂ in the ODS region is much larger than for water and methanol and makes the stationary phase less retentive for CH₂ groups. The presence of a hydroxyl group (separating a primary alcohol from the linear alkane with the same number of carbon atoms) reduces retention in RPLC because the mobile phase offers a more favorable environment than the stationary phase. The reverse is true in SFC with a neat CO₂ mobile phase where the availability of hydrogen-bonding sites due to residual silanols in the non-endcapped stationary phase makes it a more favorable environment. Addition of methanol allows for hydrogen-bonding in the SFC mobile phase and reduces the retention increment of the OH group due to stronger solvophilic interactions. Besides the stronger interactions with solvents, the methanol molecules “de-activate” the residual silanol groups, causing the loss of the retention of the OH group.

In addition to thermodynamic information, the simulation trajectories are analyzed to yield detailed information on the ODS conformations, the spatial distribution of CO₂ and methanol molecules in the stationary phase, the spatial and orientational preferences of analyte molecules, and the propensities for silanol-methanol, silanol-analyte, and methanol-analyte hydrogen bond formation. The chain conformation in the SFC system does not show substantial changes over the range of temperatures and pressures investigated here, whereas there are significant differences to the chain conformation in the RPLC systems.

This research is supported by the Chemical Measurement and Imaging program in the National Science Foundation Division of Chemistry under Grant No. CHE-2003246 (with partial co-funding from the Division of Chemical, Bioengineering, Environmental, and Transport Systems).