A multiscale modelling study on the sense and nonsense of thermal conductivity enhancement of liquid chromatography packings and other potential solutions for viscous heating effects

Detailed computational fluid dynamics study of the parameters contributing to the viscous heating band broadening in liquid chromatography at pressures up to 2500 bar in 2.1 mm columns

Over the past years viscous heating band broadening occurring in high pressure liquid chromatography has been studied extensively. In the present numerical study, we investigate the fine details of this band broadening contribution under extreme high-pressure conditions (2500 bar). To analyze the results, we first show that viscous heating leads to two clearly distinguishable band broadening effects, one originating from the radial differences in the species migration velocity and the other from the axial variations. It was found that the radial contribution is independent of the intrinsic band broadening of the bed (i.e. band broadening in absence of viscous heating) while it strongly depends on the radial dispersion coefficient and the retention enthalpy of the analytes. On the other hand, the axial contribution is strongly dependent on the bed intrinsic band broadening and it is found to be 4 to 5 times lower than the radial contribution. We also show the strong effect of the endfittings on the temperature gradients inside the column thus on the resulting viscous heating band broadening.

Computational fluid dynamics study of potential solutions to alleviate viscous heating band broadening in 2.1 millimeter liquid chromatography columns

We report on a numerical simulation study of a number of potential column technology solutions to minimize the plate height contribution (H_vh) originating from the use of ultra-high pressures and their concomitant viscous heating effect. Looking as far as possible into the future of UHPLC, all main results are obtained for the case of a 2500 bar pressure gradient. However, to generalize the result, a correlation is given that can be used to interpolate the results to lower pressures with some 10% accuracy. For the considered case of a 2.1mm column, a liquid flow rate of 0.45 ml/min, an analyte with retention factor k(25°C)=3 and a retention enthalpy chosen such that ΔH_R/R= -1000 K, it is found that, in order to keep the global plate height as measured at the column outlet (H_vh,glob,out) below 1 μm, the bed conductivity would need to be raised to λ_bed=2.4 W/m•K, i.e., 4 times higher than a typical packed bed of fully-porous or core-shell silica particles. An equivalent effect on the band broadening could be obtained if it would be possible to replace the steel column wall with a low conductivity material. In this case, a wall conductivity of 0.25 W/m•K, i.e., 64 times smaller than the conductivity of steel, would be needed to keep H_vh,glob,out below 1 μm. Results are also interpreted based on contour plots of the axial and radial velocity variation of a retained analyte.

Methods to determine the kinetic performance limit of contemporary chromatographic techniques

By combining separation efficiency data as a function of flow rate with the column permeability, the kinetic plot method allows to determine the limits of separation power (time vs. efficiency) of different chromatographic techniques and methods. The technique can be applied for all different types of chromatography (liquid, gas, or supercritical fluid), for different types of column morphologies (packed beds, monoliths, open tubular, micromachined columns), for pressure and electro-driven separations and in both isocratic and gradient elution mode. The present contribution gives an overview of the methods and calculations required to correctly determine these kinetic performance limits and their underlying limitations.