Probing the kinetic performance limits for ion chromatography. II. Gradient conditions for small ions

https://doi.org/10.1016/j.chroma.2010.05.002Get rights and content

Abstract

A gradient kinetic plot method is used for theoretical characterisation of the performance of polymeric particulate anion exchange columns for gradient separations of small inorganic anions. The method employed requires only information obtained from a series of isocratic column performance measurements and in silico predictions of retention time and peak width under gradient conditions. Results obtained under practically constrained conditions provide parameters for the generation of high peak capacities and rapid peak production for fast analysis to be determined. Using this prediction method, a maximum theoretical peak capacity of 84 could be used to achieve separation of 26 components using a 120 min gradient (Rs > 1). This approach provides a highly convenient tool for development of both mono- and multidimensional ion chromatography (IC) methodologies as it yields comprehensive understanding of the influence of gradient slope, analysis time, column length and temperature upon kinetically optimised gradient performance.

Introduction

One mode of liquid chromatography (LC) with a long history of in silico optimisation is ion chromatography (IC) [1], [2], [3], [4], [5]. Numerous hard and soft modelling procedures have been described for the analysis of small ionic species allowing development of analyte retention models for computer-based simulations to select optimum conditions for separation. This approach has been developed substantially beyond isocratic modelling to incorporate both single [4] and multistep gradient profiles [6], [7], which are commonly employed for the analysis of small ionic species [4] and large ionogenic pharmaceutical compounds [8]. The primary aim of these simulations is to provide a rapid in silico approach to method development allowing variation in the elution profile to predict optimum separation conditions.

However, this method only allows for the peak capacity to be varied through alteration of the gradient profile. Changes in the morphology of the stationary phase, flow-rate, available pressure drop and column length can also be utilised to improve separations by virtue of increased efficiency and/or speed. There is good reason to understand the influence of these factors for IC given recent approaches for fast separations including the use of short (3 cm) particulate columns [9], [10] and highly permeable silica monolithic columns [11], [12], [13]. Conversely, the development of high peak capacities for IC has received much less attention in spite of the substantial improvements in column technology [5] and the recent implementation of IC × IC methodologies [14]. The validity of any approach for improving peak production and/or total peak capacity should be assessed using a method which accounts for different support formats and analysis conditions.

Methods for prediction of efficiency under kinetically optimised conditions have been demonstrated to provide practical solutions for method development [15]. The most recent innovation in this area has been the proliferation of graphical kinetic plot (KP) optimisation approaches for quantifying the influences of maximum inlet pressure, particle size, temperature and column length [16]. Most examples of this approach thus far have been confined to the reversed-phase separations of small molecules and pharmaceuticals [17], [18], but some more recent examples have employed this approach for the analysis of peptides [19], hydrophilic interaction liquid chromatography (HILIC) [20] and isocratic IC separations [21]. Of these examples, the majority have been tailored for the optimisation of efficiency under isocratic conditions, but the KP method can also be readily extended for understanding the more useful metric of peak capacity under normalised gradient conditions [22].

As IC is particularly dependent upon the implementation of gradient elution profiles to yield satisfactory peak capacities [5], it would seem that separation optimisation procedures should consider the influence of a wider range of analytical conditions upon total peak capacity and peak production. Coupled with a numerical model for prediction of retention time and peak width [6], it should be possible to predict the expected maximum peak capacity for any single step gradient analysis considering variation in column length, gradient time, ramp rate and pressure drop.

To this end, experimental H, u0 data for a single analyte across a wide range of flow-rates were collected and these data translated to kinetically optimised isocratic performance data using the method of Desmet et al. [16]. Retention time and peak width models [6] were then used to calculate the peak capacity for various normalised gradient conditions. Finally, some experimental constraints were applied to consider the separations which were achievable in practice using currently available columns and instrumentation. From the resultant response surfaces, it is relatively straightforward to select kinetically favourable conditions for achieving a given peak capacity.

Section snippets

Reagents

All chemicals used were of analytical reagent grade and were used as supplied by Sigma–Aldrich (Sydney, Australia) unless stated otherwise. The eluent was prepared using deionised 18.2 MΩ water from a Millipore Milli-Q water purification system (Bedford, MA, USA). Working standards were prepared using deionised water from 1000 mg/L stock solutions of sodium salts (fluoride, acetate, formate, pyruvate, nitrite, sulfate, fumarate, methacrylate, tungstate, phosphate, azide, molybdate, bromide and iso

Method for calculation of peak capacity

It is well established that calculating peak capacity is a very useful parameter for measuring the quality of LC separations under gradient conditions [25], [26]. There are a number of methods for calculating this performance metric for gradient separations which may yield a maximum theoretical peak capacity or a sample peak capacity representing the maximum number of components that can be resolved between the first and last peaks of the actual chromatogram [27]. In either case, the

Conclusions

A new gradient kinetic plot method has been introduced for a greater range of experimental variables and has been applied to separations of small anions by IC. The approach provides a powerful tool for developing methodologies to satisfy peak capacity targets and is based on calculations performed using isocratic retention data. Utilising extended column lengths at the maximum allowable flow-rate was found to yield high peak capacity values even with practical restrictions on the maximum

Acknowledgements

This work was supported by the Australian Research Council's Discovery funding scheme (project numbers DP0663781 and DP0987318) The support of Dionex Corporation is gratefully acknowledged, in particular the loan of the columns used in this work.

References (38)

  • J.E. Madden et al.

    J. Chromatogr. A

    (2001)
  • P.R. Haddad et al.

    J. Chromatogr. A

    (2008)
  • V. Drgan et al.

    J. Chromatogr. A

    (2009)
  • P. Zakaria et al.

    J. Chromatogr. A

    (2009)
  • E. Tyrrell et al.

    J. Chromatogr. A

    (2008)
  • E. Tyrrell et al.

    J. Chromatogr. A

    (2009)
  • H. Poppe

    J. Chromatogr. A

    (1997)
  • A. de Villiers et al.

    J. Chromatogr. A

    (2009)
  • D. Cabooter et al.

    J. Chromatogr. A

    (2008)
  • T.J. Causon et al.

    J. Chromatogr. A

    (2010)
  • S. Louw et al.

    J. Chromatogr. A

    (2010)
  • K. Broeckhoven et al.

    J. Chromatogr. A

    (2010)
  • P. Jandera et al.

    J. Chromatogr.

    (1974)
  • U.D. Neue

    J. Chromatogr. A

    (2008)
  • U.D. Neue

    J. Chromatogr. A

    (2005)
  • J.W. Dolan et al.

    J. Chromatogr. A

    (1999)
  • C.A. Pohl et al.

    J. Chromatogr. A

    (1997)
  • X.L. Wang et al.

    J. Chromatogr. A

    (2006)
  • G. Guiochon et al.

    J. Chromatogr. A

    (2008)
  • Cited by (0)

    View full text