The chromatographic simulation of reversed-phase separations has become widely used for the optimization of gradient conditions. Such simulations use the linear solvent strength (LSS) model developed by Lloyd Snyder and John Dolan. However, elution modes such as hydrophilic interaction chromatography (HILIC) and supercritical fluid chromatography (SFC) are becoming more popular for the analysis of ionizable and/or polar compounds, and for the analysis of a wide range of analytes. Here, the conventional LSS-model can no longer be used to predict retention under various gradient conditions. In these elution modes, retention is a result of a combination of several mechanisms, including partitioning and adsorption, which results in non-linear retention relationships. Clearly, more elaborate retention models are needed to describe the curvature.
I understand that this suggestion might seem frightening for the “hands-on” chromatographer, but I sincerely believe that existing modeling software, such as Drylab or Chromsword, dedicated to reversed-phase approach, could – and should – be extended to “new” and emerging elution modes, using non-linear retention models. I am aware that the parameter estimation using such non-linear models becomes more difficult, especially when performing a limited number of initial gradient scouting runs. And the parameter fitting becomes really problematic for the non-linear models for which no analytical expression for the retention factor in gradient elution exists. However, these non-linear models can provide a significant improvement in the accuracy of (gradient) retention time predictions, compared to the conventional LSS-model – and the extra effort would therefore pay off during the optimization of gradient conditions.
Several research groups have initiated isocratic modeling for these non-linear elution modes for a wide range of compounds; for example, the groups of Pavel Jandera, Thomas Letzel and Xinmiao Liang are working on HILIC, and the groups of Caroline West and Eric Lesellier on SFC. Recently, I started investigating the possibilities of gradient retention modeling in these two elution modes, together with Davy Guillarme, using a mixed model (combining both an adsorption and a partitioning term, proposed by Liang) and an empirical (reversed-phase) model, proposed by Uwe Neue. The use of these non-linear models for the separation optimization in HILIC and SFC should in the near future be validated with real-life mixtures, including compounds with a wide range of chemical properties.
I am aware that, before computer-aided method development can be used routinely for the optimization of HILIC and SFC separations, an evaluation of the accuracy and the robustness is needed. However, it would be great to see the promising retention models proposed by several research groups being implemented by the vendors of the existing reversed-phase optimization software packages... Clearly, we have to tackle the problems that might arise and answer the concerns of the chromatography community, but I believe there is a strong future for computer-aided method development that goes beyond traditional reversed-phase separations. Furthermore, don’t we all want a greener future? The solvents saved by reducing the number of required experiments during the chromatographic method development has got to be a good thing.
Newsletters
Receive the latest analytical science news, personalities, education, and career development – weekly to your inbox.
