Indirect sampling

Nansen, C., Phillips, T.W., Parajulee, M.N., and Franqui, R.A. 2004. Comparison of direct and indirect sampling procedures for Plodia interpunctella in a maize storage facility. J. Stored Prod. Res. 40, 151-168. 

Some powder blends may present an unacceptable safety risk when directly sampled. The safety risk, once described, may justify an alternate procedure. In such cases, process knowledge and data from indirect sampling combined with additional in-process dosage unit data may be adequate to demonstrate the adequacy of the powder mix. Data analysis used to justify using these alternate procedures should be described in a summary report that is maintained at the manufacturing facility. 

The amount of sample required to obtain a spectrum is of the order of 0.01 mg (direct sample introduction system) to 1 mg (indirect sample introduction system). Purity of the sample is desirable, but in many cases not essential. Usually up to 10% of impurity (much more even of low molecular weight compounds) does not seriously interfere with the interpretation. In some instances, information about the structures of the different compounds may be obtained even from mixtures. 

Figure 1441. Sampling from a fermenter for on-line analysis (after [366]). 1. Direct removal of fermentation broth (analyte A) 2. indirect sampling by ultrafiltration, dialysis, electrodialysis, per-vaporation, providing an analyte A of proportional concentration, normally diluted 3. indirect sampling by extraction of fermentation broth by external buffer 4. in situ measurement by means of an enzyme electrode or using a sterile housing with inserted electrode. F = fermenter, W = waste.

Sampling limitations have more severe consequences in the case of indirectly sampled dimensions, where acquisition of each sampling point takes up to a few seconds. Even in 3D NMR experiments of proteins, featuring relatively fast transverse relaxation, it is almost impossible to reach the natural (determined by relaxation) line width in a reasonable experimental time. Limited experiment duration causes signal tmncation and results in broadened spectral peaks, according to the Fourier Uncertainty Principle [11]. 

Indirect sampling TLC-MS setups are the least state-of-the-art techniques used. They rely on extraction of the analyte from TLC spots using appropriate solvents, purification of solutions obtained, and the obtainment of mass spectra (sometimes after concentration of the analyte solution or even isolation of the pure analyte). The extraction is usually preceded by scraping the gel containing the analyte from the TLC surface. These techniques are technically undemanding and effective, but overall are time-consuming and tedious due to the steps that must be carried out to obtain a mass spectrum. Because the end analyte is in solution or pure form, however, the choice of mass spectrometer is straightforward and is determined solely by the availability and the type of the analyte. 

Direct sampling TLC-MS setups are usually more sophisticated than indirect sampling setups, but allow for more rapid analysis they are sometimes partially or fully automated. Generally, direct sampling techniques rely on desorption of the analyte from the TLC surface followed by its ionization, or the use of a special sampling probe that extracts the analyte and transfers it directly to the ion source. Some of these setups are now available commercially and can be attached to different types of mass spectrometers. 

OSP thickness is normally determined by indirect sampling. No method exists for determining the OSP thickness dnring prodnction on the actual PCB. Instead, a sample coupon is processed along with the prodnction parts. The conpon is immersed in an acid/solvent solution that dissolves the coating. The solntion is transferred to a UV-VIS cell, where the absorbance of light allows the estimation of the thickness of OSP on the coupon. That measurement is used as a quality control for the prodnction boards. 

Considerably smaller samples are necessary for the much narrower capillary columns. Since small fractions of a microliter can be neither reproducibly measured nor easily introduced into the capillary GC system, indirect sampling techniques are employed. In a commonly used sampling method, a sample volume of approximately l ul, or slightly less, is injected into a heated T-piece, where an uneven separation of the vaporized sample stream occurs. While the major part of the sample is allowed to escape from the system, a small fraction (typically, less than 1%) enters the first section of a capillary column. Sampling devices based on this principle are called splitting injectors or splitters. They are generally adequate in situations where samples with high concentrations of the analyzed substances are encountered. 

Other ways of indirect sampling onto a capillary column involve the injections of (relatively nonvolatile) samples diluted in a sufficiently large (measurable) volume of a volatile solvent (which serves as a sample vehicle ). With the column inlet kept at a sufficiently low temperature, the nonvolatile sample trace is trapped at the inlet and focused into a narrow zone, while the volatile solvent is allowed to pass through the column and widely separate from the sample. A subsequent increase of temperature permits the sample zone to desorb from its inlet position and enter the usual separation process. 

---