Collective Solvation and Transport at Tetrahydrofuran-Silica Interfaces for Separation of Aromatic Compounds: Insight from Molecular Dynamics Simulations

We have performed umbrella sampling molecular dynamics simulations to study the separation mechanism of aromatic compounds at the tetrahydrofuran (THF)-methanol-silica interface by liquid chromatography. Solute molecules with different polarities (naphthol and naphthalene) are selected as representative aromatic compounds. For the polar solute (naphthol), the free energy profile shows a deep minimum near the THF-silica interface, suggesting strong interactions with the polar surface. When methanol is added to the interface, there is a sharp increase in naphthol's free energy minimum, and the corresponding diffusion dynamics also undergoes a dramatic change. These findings explain the fast separation mechanism in recent experiments of separating fused ring compounds in asphaltenes with liquid chromatography. Further solvation structure and orientation analysis suggest that apolar and polar solutes may find their own comfort zones several angstroms away from the interface, and their phenyl ring's orientations would undergo a parallel-to-perpendicular transition as the solute molecule moves away from the surface. Extending our simulation studies to systems with different solute concentrations reveals that there is a decrease in the adsorption free energy accompanied by enhanced surface diffusion as the solute concentration increases, which is related to the crowding in the interfacial layers. Our simulation analysis gives a detailed microscopic description of solute solvation and transport at the THF-silica chromatography interface and will be helpful for improving separation protocols in future applications.

The joint effect of surface polarity and concentration on the structure and dynamics of acetonitrile solution: a molecular dynamics simulation study

The interfacial properties of the acetonitrile (ACN)-water-silica interface have great implications in both liquid chromatography and heterogeneous catalysis. We have performed molecular dynamics (MD) simulations of ACN and water binary solutions to give a comprehensive study of the collective effect of silica surface polarity and ACN concentration on interfacial structures and dynamics by tuning both surface charges and ACN concentration. MD simulation results indicate that many properties in the liquid-solid interface region undergo a monotonic change as the silica surface is tuned from polar to apolar due to the weakening of hydrogen bonding, while their dependence on ACN concentration is presumably governed by the preferential adsorption of water at the silica surface over ACN. However, at apolar surfaces, the interfacial structures of both water and ACN behave like the liquid-vapor interface, and this resemblance leads to an enrichment of ACN at the interface as well as accelerated dynamics, which is very different from that in the bulk solution. The organization of ACN molecules at both polar and apolar surfaces can be attributed to the amphiphilic nature of ACN, by which the micro-heterogeneity domain formed can persist both in the bulk and at the liquid-solid interface. Moreover, extending diffusion analysis to the second layer of the interface shows that the interfacial transport pathways at polar surfaces are likely very different from that of apolar surfaces. These simulation results give a full spectrum description of the ACN/water liquid-solid interface at the microscopic level and will be helpful for explaining related spectroscopic experiments and understanding the microscopic mechanisms of separation protocols in current chromatography applications.

The effect of surface polarity on the structure and collective dynamics of liquid ethanol

Typical configurations of ethanol during polarity modulation.

Effects of pH and the presence of micelles on the resolution of diuretics by reversed-phase liquid chromatography

A comparative study on the performance of two RPLC modes on the separation of 18 diuretics with diverse acid-base behaviour (acetazolamide, althiazide, amiloride, bendroflumethiazide, benzthiazide, bumetanide, canrenoic acid, chlorothiazide, chlorthalidone, ethacrynic acid, furosemide, hydrochlorothiazide, piretanide, probenecid, spironolactone, triamterene, trichloromethiazide and xipamide) was carried out. A conventional octadecylsilane column and acidic acetonitrile-water mobile phases, in the absence and presence of micelles of the anionic surfactant sodium dodecyl sulphate (SDS), were used. The effects of pH and the modifiers acetonitrile and SDS on peak asymmetry, efficiency, selectivity, resolution and analysis time, were examined. The comparison of both RPLC modes (aqueous- and micellar-organics) was done using the same processing tools, applying several polynomial and mechanistic equations to describe the retention. The best separations were obtained by maximising the product of peak purities, considering a wide range of experimental conditions. The study illustrates that, despite the theoretical and practical complexity of the problem, the predicted optimal chromatograms can be reproduced experimentally with great accuracy. None of the examined RPLC modes was able to yield baseline separation of the 18 diuretics. However, their selectivity was complementary, being appropriate for different combinations of a smaller number of the assayed diuretics.

New equations describing the combined effect of pH and organic modifier concentration on the retention in reversed-phase liquid chromatography

Six equations that express the combined effect of mobile phase pH and organic modifier content on sample retention in reversed-phase liquid chromatography (RPLC) are developed based on either the adsorption or the partition model for retention. The equations are tested against five retention data sets taken from literature. In the tests two pH scales are used, w(w)pH and s(s)pH. It is shown that a new seven-parameter equation works more satisfactorily, because it exhibits good numerical behavior, gives low values of the sum of squares of residuals and represents the experimental retention surfaces successfully. In addition, the danger of overfitting, which leads to the prediction of physically meaningless retention surfaces, is minimized by using the proposed new seven-parameter equation. Finally, the possibility of obtaining reliable pK values of weak acids or bases chromatographically by means of the derived equations is also considered and discussed.

Effect of the organic modifier concentration on the retention in reversed-phase liquid chromatography I. General semi-thermodynamic treatment for adsorption and partition mechanisms

A semi-thermodynamic treatment is adopted to account for adsorption or partition of solute molecules from aqueous mobile phases on/in reversed-phase liquid chromatography stationary phases. The theoretical expressions of ln k' versus organic modifier content are tested against 10 data sets covering a variety of solute molecules. It is shown that the mean field approximation, adopted widely in ptevious studies, is marginally valid in aqueous mobile phases, especially in the presence of solute molecules, and the lattice model approximation, which is also used in relevant studies, is a poor approximation. Clear conclusions about the validity of either the adsorption or the partition model for the retention mechanism could not be drawn. The equations of the adsorption model describe all data sets absolutely satisfactorily and yield a physically reasonable picture about the behavior of modifier and solvent at the adsorbed layer. However, the high applicability of the adsorption model may not safely entail the validity of the adsorption mechanism at a molecular level, especially in the case of solutes with small and non-polar molecules, where our analysis gives strong indications about the validity of the partition mechanism. The next steps needed for the final elucidation of the retention mechanism in reversed-phase chromatographic columns are indicated.

On the use of genetic algorithms for response surface modeling in high-performance liquid chromatography and their combination with the Microsoft Solver

Four genetic algorithms--the classical, Haupt's, Brunetti's and a modification of the classical algorithm suggested in the present paper--are examined when they are used for the modeling of response surfaces in high-performance liquid chromatography (HPLC). We found that the best results are obtained from our modification and the worst by Haupt's algorithm. The classical genetic algorithm gives satisfactory results, better than those of Brunetti's algorithm. We also ascertained that all genetic algorithms may get stuck in a local minimum other than the global one, except for our modification, which can be considered to approach a global method. Finally, the time needed for the optimization of a genetic algorithm and the combination of a genetic algorithm with a non-linear least-squares routine are considered and discussed.