Peak distortion

Figure 9. Peak Distortion by a Langmuir Type Isotherm...

Figure 11. Peak Distortion Caused by a Freundlich Type Isotherm...

As a consequence, the peak is distorted with sloping front and a sharp back as shown in Figure 11. There are a number of other effects that can cause peak distortion and... 

This extreme condition rarely happens but serious peak distortion and loss of resolution can still result. This is particularly so if the sensor volume is of the same order of magnitude as the peak volume. The problem can be particularly severe when open tubular columns and columns of small diameter are being used. Scott and Kucera measured the effective sensor cell volume on peak shape and their results are shown in Figure 13. 

When columns of the same polarity are used, the elution order of components in GC are not changed and there is no need for trapping. However, when columns of different polarities are used trapping or heart-cutting must be employed. Trapping can be used in trace analysis for enrichment of samples by repetitive preseparation before the main separation is initiated and the total amount or part of a mixture can then be effectively and quantitatively transferred to a second column. The main considerations for a trap are that it should attain either very high or very low temperatures over a short period of time and be chemically inactive. The enrichment is usually carried out with a cold trap, plus an open vent after this, where the trace components are held within the trap and the excess carrier gas is vented. Then, in the re-injection mode the vent behind the trap is closed, the trap is heated and the trapped compounds can be rapidly flushed from the trap and introduced into the second column. Peak broadening and peak distortion, which could occur in the preseparation, are suppressed or eliminated by this re-injection procedure (18). 

A definitive identification of the proteins in each peak is not possible, however, the elution times of the peaks at 13-14 min. and 15 min. are close to the times which would be expected for gamma-globulins and albumins, two of the principal classes of serum proteins. These data also indicate the loading capacity of this column with serum. More than 14 mg. of undiluted serum was injected before evidence of overloading in the form of band broadening and peak distortion was observed. 

Figure 3.6 Different peak distortion problems due to band broadening in time and band broadening in space observed during hot splitless injection. Band broadening in space is characterized by a broadening which grows proportionally with retention time and may result in peak splitting that is poorly reproducible. Band broadening in time is characterized by a constant broadening of all peaks. Partial solvent trapping results in characteristic chair and stool shaped peaks. (Adapted with permission from ref.

Samay, G. and Kubin, M., Peak distortion in gel permeation chromatography at high sample concentration, /. Appl. Polym. Sci., 23, 1879, 1979. 

Trying to determine which column is ideal for a specific analysis can be difficult with over 1000 different columns on the market [74]. A proper choice implies a definition of parameters such as column material, stationary phase (polarity), i.d., film thickness and column length. Guides to column selection are available [74,75]. The most important consideration is the stationary phase. When selecting an i.d., sample concentration and instrumentation must be considered. If the concentration of the sample exceeds the column s capacity, then loss of resolution, poor reproducibility and peak distortion will result. Film thickness has a direct effect on retention and the elution temperature for each sample compound. Longer columns provide more resolving probe, increase analysis times and cost. 

A commonly used procedure for the determination of phosphate in seawater and estuarine waters uses the formation of the molybdenum blue complex at 35-40 °C in an autoanalyser and spectrophotometric evaluation of the resulting colour. Unfortunately, when applied to seawater samples, depending on the chloride content of the sample, peak distortion or even negative peaks occur which make it impossible to obtain reliable phosphate values (Fig. 2.7). This effect can be overcome by the replacement of the distilled water-wash solution used in such methods by a solution of sodium chloride of an appropriate concentration related to the chloride concentration of the sample. The chloride content of the wash solution need not be exactly equal to that of the sample. For chloride contents in the sample up to 18 000 mg/1 (i.e., seawater),... 

The peak distortion in CZE electrophoregrams by electromigration dispersion has been modelled and the peak centre was related by... 

It has been found that the equations describe correctly the peak shape determined experimentally, and can be applied for the prediction of peak distortion [118]. 

The most serious causes of error are (i) wave and peak distortion caused by excessively fast scan rates, which are themselves caused by diffusion being an inefficient mass transport mechanism, (ii) current maxima caused by convective effects as the mercury drop forms and then grows, and (iii) IR drop, i.e. the resistance of the solution being non-zero. Other causes of error can be minimized by careful experimental design. 

The mobile phase composition at the starting conditions of the gradient should preferably also be used as sample solvent. Differences in solvent strengths are usually the cause of peak distortion (splitting) in the early part of the chromatogram, as discussed in Section IV. A. 

Clean-up and filter samples to prevent system and column damage. Dissolve or dilute the final sample solutions in mobile phase, if possible. If a final sample solvent is stronger than the mobile phase, inject smaller volumes to minimize peak distortion. 

Equation (67) clearly shows that when d is decreased either the detector cell volume has to be decreased proportionally or the column diameter should be increased in order to avoid significant peak distortion. With typical values such as 6 = 0.1 [Pg.26]

As with all window-based smoothers, the choice of the window size in polynomial smoothers is very imponani. Another decision to make for polynomial smoothers is the order of the polynomial to be fit (Barak, 1995). Typically, a second- or third-order polynomial is used. An example of applying a polynomial smoother is shown in Figure 3.7, wliere a second-order polynomial is fit with window sizes of 7, 13, and 25 points. As the window size increases, the noise is continually reduced. Howe -er, when the window is too large, sharp peaks may be removed and the remaining peaks distorted. This is demonstrated in Figure 3.8 where a spectrum is shown before (solid) and after (dashed) applying a 49-point second-order polynomial smoother. 

Figure 3.8. Illustration of peak distortion when using a smoother with a too-iarge window width (49 point). (Solid—raw data dashed—smoothed data.)...

A major problem with viscous fingering is that reproducible peak shapes, albeit distorted, may be observed this may be highly misleading in interpreting SEC results. To obtain reliable results, molecular weight distributions of samples should be obtained as a function of polymer concentrations to arrive at a value in which peak distortion is not present or peak shape does... 

The data handling or integration software can also cause peak distortion if the settings are altered throughout the lifetime of the method and it is often necessary to test these for ruggedness. 

A freezing of any one of these equilibria at low temperature would result in peak distortions... 

---