Sampling splitting techniques

USP 429) stress the importance of sampling and advise that a representative sample be prepared using a sample splitting technique. The golden rules of sampling suggested by Allen 1990 are ... 

These sample sizes are often obtained by sample splitting techniques. 

An alternative form of split injection is the timed split technique, Figure 6.11 [130,131,133]. In this case the column is connected directly to the valve and the valve actuator is controlled electronically to turn the valve to the inject position and back very rapidly with only a portion of the sample in the loop displaced to the column. Timed split allows variable volumes to be injected by changing the valve actuator tine and provides more reproducible splitting than the dynamic split technique. However, it suffers from many of the same problems as dynamic split, namely, poor accuracy, split ratios that depend on pressure, and high detection limits. 

General texts on GC are numerous [118,119] narrow-bore GC was addressed by van Es [120]. Sample introduction techniques and GC inlet systems have been reviewed [25,90] and split/splitless [121] and on-column injection [122] were considered specifically. Stationary phases [123], multiple detection [103], derivatisation [124,125], and quantitative analysis in GC [109] have been described. High-speed GC has recently been reviewed [126]. For a compendium of GC terms and techniques, see Hinshaw [127]. 

Ideally, a sample is introduced into a chromatograph as a perfect plug. In practice, this is not the case, and diffusion occurs because of the injector. For narrow-bore and microbore applications, injectors capable of introducing the required sample volumes are commercially available and optimized to reduce dispersion. This is not the case for capillary LC, and homemade injection systems include the sample tube technique, in-column injection, stopped-flow injection, pressure pulse-driven stopped-flow injection (PSI), groove injection, split injection, heart-cut injection, and the moving injection technique (MIT). Of the injection techniques, only the split injector, MIT and PSI approaches can introduce subnanoliter sample volumes accu-... 

For biomarker identification, it is also possible to separate out substances of interest from a complex biofluid sample using techniques such as solid phase extraction or HPLC. For metabolite identification, directly coupled chromatography-NMR spectroscopy methods can be used. The most powerful of these hyphenated approaches is HPLC-NMR-MS [24] in which the eluting HPLC peak is split with parallel analysis by directly coupled NMR and MS techniques. This can be operated in on-flow, stopped-flow, and loop-storage modes and thus can provide the full array of NMR and MS-based molecular identification tools. These include MS-MS for identification of fragment ions and FT-MS or TOF-MS for accurate mass measurement and hence derivation of molecular empirical formulae. 

The use of conventional columns in the first dimension and a carrier gas-flow rate similar to that used in ID GC enables all injection techniques to be used (e.g., split, splitless, large volume, PTV and SPME) [7]. That further makes any method based on a conventional GC separation amenable to a GCxGC approach without reconsideration or reoptimization of the existing sample introduction technique. 

It must be emphasized, however, that many of today s automated electronic size analysis instruments reqnire very small qnantities of sample for optimum performance. This places considerable importance on the pretreatment, sample splitting, and preparation techniques (see also discussions in Matthews, 1991a McManns, 1988). 

It is even possible to couple LC and GC here, LC plays the role of a sample preparation technique that eliminates compounds that would affect the gas chromatographic separation. Because GC cannot tolerate high volumes of liquid, it is necessary to use narrow-bore LC columns, to split the eluate, or to use a special interface that eliminates most of the liquid. 

Hankemeier et al., studied large-volume injection combined with GC/DD-FTIR. A loop-type injection interface was chosen because of its rather simple optimization. Large-volume injection by means of a loop-type interface can be carried out successfully in conjunction with GC/DD-FTIR. The hyphenation permits enhanced detectability of analytes by about two orders of magnitude when compared with conventional split/splitless ones. As demonstrated, the determination and identification of PAHs in river water is possible down to a level of 0.5 p-g/L, even when using simple micro hquid-liquid extraction as a sample preparation technique. The present system may, therefore, be considered a viable approach to trace-level environmental analysis. 

Because of the variety of columns and samples that can be analyzed by GC, several injection techniques have been developed. The packed inlet system is designed mainly for packed and wide-bore columns. However, an adapter can be used to enable capillary columns to be used. When injection is carried out in the on-column mode, glass wool can be used for packing the injector. For capillary GC, spht technique is most common, which is used for high concentration samples. This technique allows injection of samples virtually independent of the selection of solvent, at any column temperature, with httle risk of band broadening or disturbing solvent effects. The sphtless technique, on the other hand, is used for trace level analysis. The so-called cold injection techniques (on-column, temperature programed vaporization, cooled needle split) have also been recently developed. " " ... 

Fig. 3-3 Efficiency characteristic VH vs. sample size for adsorption capillary column with aluminum oxide layer on the inner walls [26], splitless (a) and split-type (b) sample injection technique.

Fig. 3-3 shows the H-W dependence according to equation (3-21) for the same adsorption capillary column but with the use of different sample injection techniques, viz., split-type (b) and splitless (a). It is evident that the dependence of the HETP value on sample size in these coordinates is linear, which is corroborated by equation (3-21). When a splitter is used, the efficiency shows a market increase (the H value decreases). 

Interestingly, the Hq value is virtually the same irrespective of sample injection technique (split or splitless). This characteristic is an additional test of equation (3-21). 

---