Best column selection

It should be noted that there is no best column selectivity, because separation depends on both the column and the sample a column which is best for one sample will not be best for all other samples. 

Another important parameter for column selection is the proper choice of sorbent porosity. The pore size of the sorbent determines the fractionation range of the column. The best way of doing this is by looking at the calibration curves of the columns, which are normally documented by the column vendor (cf. Fig. 9.3 for PSS SDV column calibration curves and PSS SDV fractionation ranges) (7). 

Figure 9.4 illustrates a simple way of selecting the best column for SEC work based on the calibration curves of two sorbents with different pore sizes (4). 

Scheme 4.5 illustrates HPLCphase selection. Column manufacturers may have an applications database from which they can recommend a column and a method. Specific methods have been established for quite a large number of analytes, such as additives (e.g. antioxidants). Column selection and column technology have been reviewed [549]. Contrary to GC, and with the exception of SEC, selectivity in HPLC is determined not by the column alone but also by the mobile phase. There is therefore no one-for-one assignment between an analytical problem and the best column for this problem. 

Rohrschneider went on to show that one can use the liquid phase constants in reverse to get values that would represent the polarities of other solutes. The values in Table 14 were compiled by him by running a new solute on five liquid phases whose Rohrschneider constants had been previously determined. The five A/ values thus obtained are substituted into five equations [like Eq. (6)], which are solved simultaneously for the five unknowns a, b, c, d, and e. As expected, the a value for toluene is close to 100 (108), the b value for propanol is close to 100 (105 for n-propanol), and the c value for acetone is nearly 100 (95) because these three solutes have the same functional groups as the original probes. To answer the question How polar is chloroform, one can look at the five constants and tell what functional groups (or interactions) it is like and what groups it is not like. In principle, one can determine the solute constants for the components of any sample, multiply them by the Rohrschneider/McReynolds constants for a variety of common columns, sum the values, and thus find the best column for the separation. This approach is not the one analysts have taken to select the best liquid phase for a given separation. Before we take a look at the procedures that are used, let us note some related uses that can be made of the Rohrschneider/ McReynolds constants. 

Note that multiple pH values and columns can be screened in gradient mode at the same time as well. This will increase the efficiency/probabifity of obtaining the best column/conditions and the best demonstrated chromatographic selectivity. Note that the aqueous phase pH values that would be chosen for these pH/column screening studies should be based on knowledge of the physicochemical properties of the molecule, taking into consideration the mobile-phase pH and analyte pKa shifts in the hydro-organic media. 

The column efficiency, backpressure and lifetime should be taken into account and compromised when selecting the best column particle size. Most often, porous particles with diameters of 5 pm are used in conventional analytical columns and 3 pm (exceptionally 2 pm) porous particles are usually used in short high-speed columns for... 

Perhaps 90% of all rHPLC separations use a Cig stationary phase. In SFC, there is a much wider choice. The vast majority of SFC separations have been performed on totally porous silica, usually with a bonded organic phase. A full range of achiral stationary phases is available for use in SFC. The most common are silica, cyano, amino, and diol. Several relatively new phases, such as an ethyl pyridine phase, promise to decrease the need for additives in the mobile phase. Almost everything elutes from cyano columns. Diol and bare silica often give the best selectivity for acids and alcohols. Amino is often more retentive but also often yields the best selectivity for amines. Although any of these columns is likely to give a reasonable separation, the use of an automated column selection valve, along with a solvent selection valve, makes it easy to find the one with the best selectivity for a specific application. 

A chromatograph can employ a column selection valve (49) and a solvent selection valve with automation to empirically find the best column, and to a lesser extent, mobile phase, for a particular application. Samples are usually run overnight and the chromatograms evaluated in the morning. After the column giving the best separation is found, it usually only takes a few runs to further develop the method to the level needed for the separation goals. 

After the solvent is chosen, the slurry concentration needs to be selected. As shown in Reference 11, there is an optimal value for the slurry concentration that depends on the other parameters of the packing procedure. Therefore, the slurry concentration that results in the best column needs to be determined empirically. The concentration that you start with might be dictated by the packing equipment, but if you are free to choose, a concentration of 10%v/v is a good starting point. 

Many modern packed column SFCs are equipped with a column selection valve and a solvent selection valve. Software Wizards are available that allow simple programming where a large number of different stationary and mobile phases can be automatically tested to find the best conditions without operator intervention. 

Of the two important decisions in setting up a gas chromatographic analysis, choosing the best column (usually the best stationary phase) is the more important. The other, selecting the column temperature, is less critical because the temperature can be easily programmed through a range of values to find the optimum value. (See Chapter 9.)... 

Stationary phase that has a selectivity, a, for this pair, which is not so close to 1. Of course, we must hope that resolution for the other analyte pairs does not decrease too much with the new column. Such are the trade-offs in finding the best column and conditions for a mixture separation. One can easily imagine that the probability of problems and conflicts increases with the number of components we would like to measure. 

Band spreading is also related to an obstructive factor that is not a constant in a column bed. Both Kubin [6] and Pfannkoch et al. [11] have shown that plate height varies as a function of K. It appears that as its molecular size approaches the pore dimensions, a solute experiences diffusion limitations, decreasing its effective diffusion coefficient. This influence of restricted molecular movement on plate height (//) can be readily observed in SEC profiles. The first peak to elute after the void volume marker is frequently the broadest peak in the chromatogram. If one were able to obtain columns with different pore diameters but similar plate counts, pore volumes, and calibration curve slopes, it would be best to select a column on which would be 0.2 or greater for the solutes to minimize the effect of the obstructive factor. 

Choosing the best column is the most important decision in HPLC. Reviews have discussed this in detail [45-48]. Marchylo [45] noted that column selection can be bewildering due to the many columns available but that many columns offer satisfactory performance, with some differences in resolution and selectivity. Verevacek and Huyghebaert [47] also reviewed commercial columns for lE-HPLC, SE-HPLC, and RP-HPLC of food proteins, along with column characteristics and manufacturers. Here I will briefly review the status and recent developments in columns for RP-, IE-, and SE-HPLC. 

At the end of the working day, a further six columns should be inserted into the column selecting valve, preferably columns with phases that are similar to the best column of the previous night s experiment. This is the column that yielded... 

After selecting the best column material, other column parameters have to be taken into account to meet the analysis goals. Here, the column dimensions, the particle size, and the porosity are of importance. 

---