Panels A and B are simulated chromatograms for 15 min gradient separations of a 20-component mixture, where each separation has a peak capacity of 200. The bottom line is that as chromatograms become increasing crowded, the chance of encountering an area of refractory peak overlap increases dramatically.įigure 2 captures these ideas in the form of chromatograms. (16) In spite of their sophistication, such model driven optimization approaches are also limited by low methodological peak capacities.
#STELLA ARCHITECT TUTORIAL ADVANTAGES DISADVANTAGES TRIAL#
Indeed, this issue is the driving force behind hundreds of papers devoted to replacing Edisonian trial and error with sophisticated methods to model retention and selectivity as a function of the method variables leading to the algorithmically directed optimization of the separation variables. (15) Ultimately though, we do not find ourselves much further ahead after changing the phase chemistry. The benefits of adjusting stationary phase differences can be enhanced somewhat through column coupling and selectivity tuning. Changes in the stationary phase type can often effect greater changes in the elution pattern, but improvements in one critical pair frequently result in the lessening in the separation of another pair. When only a limited time is available to resolve an apparently “simple” mixture, we often find that the analysis time is controlled by the resolution of one or a few stubborn pairs of compounds which resist improvement despite changes to the easily manipulated separation variables such as eluent composition, stationary phase type, or column temperature. This issue is familiar to chromatographers in many application areas. This surely is a valuable increase, but the difference becomes more profound as the sample complexity increases as shown in Figure 1. We see that for a 50-component mixture, only about 50% of the compounds will be observed as distinct (i.e., resolved) peaks, whereas a peak capacity of 3000 increases the fraction of compounds observable as singlet peaks to about 95%. Figure 1 shows the calculated results for sample complexities in the range of 5 to 1000 and peak capacities in the range of 100 (easily achieved by 1D-LC) to 3000 (only achievable in a reasonable time (<2 h) by 2D-LC).
(12, 13) The Statistical Theory of peak Overlap (STO) of Davis and Giddings (14) allows estimation of the fraction of constituents in a sample that would be (on average) observed as chromatographically distinct (singlet) peaks for different combinations of sample complexity and peak capacities. These numbers roughly double when discussing separations of peptides, in part because the extent of gradient peak compression is greater in this case, and their inherently greater sensitivity to a change in composition produces narrower peaks even if there were no effect of on-column compression of the tailing side of the peak. (11) Recent estimates for the maximum peak capacity that is achievable in 1D-LC of small molecules range from about 100 in analyses of ∼5–10 min to a few hundred in analyses of a few hours (assuming solvent gradient elution and reversed-phase liquid chromatography (RPLC).
Peak capacity is a theoretical construct that estimates the maximum number of equal-height peaks that can be fit, side-by-side at equal resolution (typically 1.0), within a given separation space. Peak capacity is the metric most often used to define the limitations of 1D-LC in tackling type-A problems. For more detailed explanations of some terms readers are advised to consult the most recent listing of terms, definitions, and nomenclature. Finally, most of the nomenclature and terminology used here that are specific to 2D-LC are nominally consistent with those prescribed by Marriott and Schoenmakers. Where appropriate, we have drawn attention throughout this Feature to particular reviews that are especially relevant to specific topics. Those looking for more depth and breadth are encouraged to consult a number of recent reviews (1-7) and books (8, 9) on 2D-LC. This tutorial is most certainly not a comprehensive review rather, it is an introduction for interested but unfamiliar readers. Finally, we show by way of selected examples the ways in which 2D-LC is being used in practice to efficiently and effectively solve challenging analytical problems in a variety of fields. We then review some of the most fundamental principles of 2D-LC. In this tutorial, we aim to capture the state of the art of two-dimensional liquid chromatography (2D-LC) by highlighting key background that will help readers understand where 2D-LC methods fit in the analytical chemist’s toolbox.