Peak Capacity and Resolution

Usually, peak capacity is defined as the number of peaks per unit time, see Eq. (3.5). The peak capacity as a separation criterion proves to be important when very similar components have to be separated, such as components of a homologous series, for example, ohgomers in this case, one expects equidistant spacing of the peaks. With similar components, we can hardly expect different interactions and thus a good selectivity. The situation is similar in the case of complex mixtures and/or a difficult matrix. Again, realistically here we do not come any further via selectivity. A separation in this case will be possible, when the many (similar ) components can be eluted distributed as narrow peaks throughout the chromatogram. 

When UHPLC is available, it is additionally recommended to use a long column filled with sub 2 pm particles - possibly think also of core shell or monolith columns, see Chapter 5 - at a high flow, run a steep gradient and additionally increase the temperature. A trick that, with many similar components, often leads to good resolution and narrow peaks start with high %B and run a relatively flat gradient. For further information regarding peak capacity, see Chapter 3. 

Resolution Now let us turn to the objective of good resolution. This is by far the most difficult case. Make the following clear - good resolution means. 

Furthermore, in the case of slow kinetics it must be reckoned with another -sometimes substantial - contribution to peak broadening. Due to, for example, additional ionic interactions - but also complex formation or displacement of equilibria due to an inadvertent pH gradient during the run — wide/tailing peaks can also be obtained with gradient runs. Regarding peak broadening due to slow kinetics, see below. 

Separation Capability Selectivity Term In practice, we have to deal with a-values between about 1.02 and 1.1, in the case of substantially differing components perhaps of 1.2. Selectivity is by far the most sensitive function for the resolution, even a minimal improvement in the a-value leads to a dramatic increase in resolution, for details and numerical examples see Chapter 3. There, as also in Chapter 5 (RP columns), is reported in detail how the selectivity can be influenced. Let us go 

---

The resolution of these columns for protein mixtures, however, was comparably poor. The peak capacity for human serum albumin was near 3 during 20 min gradient elution. Improvement has been reached by covalent binding of PEI (M = 400-600) onto a 330 A silica of 5 pm particle size [38], The peak capacities of ovalbumin and 2a -arid glycoprotein were 30-40 (tgradienl = 20 min). Enhanced peak capacity and resolution probably were due to the more diffuse structure of PEI coupled to silane moieties than that of strictly adsorbed on silica and cross-linked (see Sect, 2.2). Other applications of covalently adsorbed PEI are discussed in Sect. 4.1. 

Multidimensional chromatography is a very attractive technique for the analysis of complex mixtures where a mono-dimensional separation cannot be sufficient to resolve all the components of interest. Obvious advantages are the higher peak capacity and resolution offered by these systems. Typically, one part of the chromatogram from the first column is transferred to another column via a suitable interface. 

A practical method for enhancing the peak capacity, and thus the resolution of analytes in multicomponent complex mixtures, can be achieved by changing the mode of the separation during the chromatographic analysis, employing a column switching system in order to optimize a separation. 

The choice of columns used for 2DLC is based upon experience with the sample and resolution required. The HPLC column descriptors of selectivity, resolution, peak capacity, sample capacity, degree of sample recovery, and speed of separation have been discussed previously (Unger et al., 2000). Columns with higher peak capacity and sample capacity (IEC, HIC, NPLC, and RPLC) are preferred in the first dimension, and higher speed columns (SEC and RPLC) are better in the second dimension. This is discussed in detail in Chapters 2 and 6. 

A commercial HPLC system and columns capable of performing ultra high-pressure LC were recendy introduced at PITTCON 2004 (ACQUITY Ultra Performance LC System by Waters). This HPLC system was designed to take full advantage of the potential of novel, sub-2-micron particles to give scientists chromatographic run times that are up to 9 times shorter than current fast HPLC systems, up to 2 times better peak capacity or resolution, and 3 times better routine sensitivity. 

The approaches described above are designed to increase peak capacity and thus reduce peak overlap. A radically different approach involves accepting component overlap as inevitable and directing attention at numerical rather than physical peak resolution. Numerical resolution allows the recovery of analytical information but not the recovery of purified components. 

This chapter provides an overview of basic terminology and essential concepts in HPLC including retention, selectivity, efficiency, resolution, and peak symmetry as well as their relationships with key column and mobile phase parameters. The resolution and van Deemter equations are discussed. The concepts of peak capacity and method orthogonality as well as key gradient parameters such as gradient time and flow rate are described. An abbreviated glossary of HPLC terms is listed. 

For speed and resolution comparison, the nearest GC separahon on carbon nanotubes known to the authors is the one by Saridara and Mitra at the N.J. Inst. Tech. [13], which took 60x longer, or 4 min, to achieve a separahon of C -C alkenes and a comparable peak capacity of 16. They used a 300-cm capillary column with 500- tm ID, temperature ramping of 50 °C min , and a carrier gas velocity of -85 cm s . We will return later to the importance of peak capacity and its influence on FAR in the section on FAR (Sechon 9.3.5.4). 

We can approach an undostanding of macromolecular HPLC in various ways, based on the present model (//, 12,28,40). On the one hand, we can use the model to carry out exact predi ons (computer simulation of HPLC runs), and see how separation varies with experimental conditions. Altema tively, we can simplify our model to provide >proximate but explicit equations for resolution, peak capacity, and peak height as functions of variables such as flow rate, column length, gradient time. 

As we have described, TGF has successfully demonstrated the simultaneous concentration and separation of a wide range of species in a variety of implementations. Key figures of merit for TGF as a separation modality are the concentration factor and peak capacity. Static TGF, where the externally applied pressure remains constant, has demonstrated concentration factors in excess of 20,000 but is limited to peak capacities <10 [56], As a result, there has been recent work to develop a dynamic form of TGF called scanning TGF [67], where the externally applied pressure is varied with time. This technique allows higher peak capacity and tunable resolution and concentration by adjusting the rate at which the procedure scans through the applied pressures. 

---