Optimum peak shape

The temperature of the ineoming mobile phase will also alter the environment within tire eoluirrtr.. Some laboratories have advoeated that equilibration of the ineoming mobile phase with the eolumn temperature is essential for good peak shapes and optimum effieieney. However, a number of studies have now demonstrated that if the mobile phase is eooler than the eolumn the effieieney is improved, in some eases quite markedly. 

The furnace is heated by low voltage (usually 10 V) and high current (up to 500 A) from a well stabilized step-down transformer. For optimum precision, the voltage should be well stabilized, often by a feedback loop which may be temperature feed-back based (see Section 3.6.1). A rapid rise-time of the temperature is also preferable, because of theoretical considerations of peak shapes. This has implications for power supply design and furnace design, as will be shown below. Currently, furnaces are available that reach temperatures of up to 3000°C, and temperatures of 2500°C should be reached in less than 2 s in a well designed furnace. 

This CE method provides an efficient approach for rapid and effective separation of serum conjugated BAs with an analysis time of 8 min. It is important to mention that each micellar solution plays an important role in the analysis use of 20 mM SDS is to modify the electro-osmotic flow, whereas solubility of glycine-conjugated BAs and the peak shape of all BAs are maintained with 20% acetonitrile and the neutral pH of the phosphate buffer. The optimum condition for baseline separation is achieved with the addition of 8 mM CD. 

A similar argument holds for the influence of the peak shape on the separation criterion. In the non-linear part of the distribution isotherm, the shape of the peak will be a function of the injected quantity. Hence, once again, the location of the optimum may be affected by the composition of the sample. Also, the effect of column dimensions on the peak shape may be hard to predict, and the peak shape may to a large extent be determined by the characteristics of the instrument, rather than of the column. Therefore, if the composition (or the concentration) of the sample can be expected to vary considerably, and if it is desirable that the result of an optimization process can be extrapolated to different columns (of the same type) and to different instruments, then it is advisable to use criteria that are not affected by the relative peak areas, nor by the shape of the peaks. 

Engelhardt and Mliller reported on the differences in the physical properties, such as specific surface area, specific pore volume and average pore diameter - and on the different amounts of stationary and mobile phase per unit column volume for various commercially available silica gels. If the retention for various solutes were normalized for these factors, distinct selectivities were still noticed. This could be explained by differences in the surface pH of the silicas. Irregular ones were usually neutral or weakly acidic, whereas the spherical ones were either acidic (pH ca.4) or basic (pH ca.9) (see Table 1.4).To obtain the required and optimum selectivity, the pH of silica gel can easily be adjusted. For basic compounds more symmetrical peak shapes were obtained on silica with a basic character. 

Optimum pH for the separation was 6, and the addition of methanol improved the peak shape. The method had to be modified for the analysis of colchiceinamide and its demethylated meta-... 

Figure 24.1 Peak shapes as obtained with different column diameters and separation performances (as a function of particle diameter at a given column length). Columns 1 and 3 are packed with a coarse stationary phase, columns 2 and 4 with a fine one. The packing quality, defined as reduced plate height h, is the same in all four cases. Separation performance is independent of column inner diameter therefore peaks 1 and 3 as well as 2 and 4, respectively, are ofthe same width. The peak height in the eluate can be calculated from equation (15) it is higher when the column is thinner and the separation performance is better. Peak areas cannot be compared although the same amount is injected in any case if a concentration-sensitive detector is used optimum flow rate depends on particle size and hence also the residence time in the detector.

Another interesting question is Is it possible to determine optimal peak integration intervals on the basis of known or even unknown peak shapes and known noise characteristics And if an optimum integration interval can be estimated, what is the error variance ... 

Note that smoothing increases the width of peaks and makes resolution of overlapping peaks more difficult. The optimum filter described above broadens the peak by a factor of y/2. In general, the optimum filter will depend on the background shape as well as the peak shape and will be affected by the peak-to-background ratio. 

SFC uses the same stationary phases as FIPLC. Selectivity is similar but not identical. One of the greatest differences from n-FIPLC is in speed. SFC optimum flow is inherently three to five times faster, while peak shapes are often significantly better. Unlike n-HPLC, SFC reequilibrates after passage of only a few column volumes. Overall, SFC is often much more than 10 times faster than FIPLC. A fast chiral SFC separation is shown in Figure 4. Note that the column is not high speed but a standard 4.6x250 mm column with 10 pm particles. Some industrial pharmaceutical companies have dropped HPLC for chiral analysis and use SFC for less polar solutes, and high-performance capillary electrophoresis (HPCE) for water-soluble solutes. Unlike HPCE, SEC is scalable. 

There is no single optimum sample size. Some general guidelines are available, however. Table 2.1 lists typical sample sizes for three types of columns. For the best peak shape and maximum resolution, the smallest possible sample size should always be used. 

---