Analytical methods
Static headspace gas chromatography (GC) often provides a convenient and easily auto mated mode of analysis for blood and other biological specimens that may be obtained without using special apparatus to collect the sample. Many analyses can be accomplished using flame ionisation detection (FID) and/or electron capture detection (ECD). Nitrous oxide and most halogenated compounds respond on the ECD, although the thermal conductivity detector (TCD) may be used as an alternative if nitrous oxide poisoning is suspected. Use of expired air collected into either a Tedlar bag or via a special device (Dyne et al. 1997; Fig. 1) with subsequent GC or GC-mass spectrometry (MS) analysis can facilitate the analysis of a number of compounds. Direct MS of expired air can also detect many compounds several days post exposure. However, the use of these techniques is limited by the need to take breath directly from the patient and the specialist equipment required (Ramsey 1984). Vapour phase infrared (IR) spectrophotometry may be useful in the analysis of abused products or ambient atmospheres. High-performance liquid chromatography (HPLC) is useful in the analysis of polar metabolites of certain solvents.
Figure 1 Device for capturing breath samples for solvent analysis.
Packed GC columns have been used extensively in conjunction with headspace sample preparation. Disadvantages include the poor resolution of some very volatile substances, a long total analysis time and variation in the peak shape given by alcohols between different batches of column packing. GC separation using capillary columns offers superior resolution to the use of packed columns. However, the most commonly used columns such as a 30 m X 0.2 mm i.d. column with a chemically bonded stationary phase of dimethylpolysiloxane (DMS) 0.1 μm film thick ness are unsuitable because volatile substances elute rapidly and resolution is poor. This difficulty can be overcome by the use of wide-bore capillary columns e.g. 60 m x 0.53 mm i.d. 5 μm film thickness DMS combined with large-volume injection. An alternative for the analysis of volatile substances is a porous layer open tubular (PLOT) column. These phases give good retention and thus resolution of compounds of similar relative formula mass, but peak shapes of polar compounds are poor and it is difficult to screen for compounds of widely different volatility in one analysis.
The use of a capillary column together with two different detectors (FID and ECD) confers a high degree of selectivity, particularly for low formula-mass compounds for which there are relatively few alternative structures. If more rigorous identification is required, GC combined with MS or Fourier transform IR spectrometry (FTIR) may be used. However, GC-MS can be difficult at high sensitivity when the fragments produced are less than m/z 40, particularly if the instrument is used for purposes other than solvent analyses. In particular, the available sensitivity and spectra of the low-molecular weight alkanes renders them very difficult to confirm by GC-MS. Inertial spray MS allows the introduction of biological fluids directly into the mass spectrometer without prior chromato graphic analysis and has been used in the analysis of halothane in blood during anaesthesia. A derivatisation method for toluene and ethylbenzene involving the use of chlorine gas prior to GC-MS has been described .
GC-FTIR may be more appropriate than GC MS in the analysis of volatiles, but sensitivity is poor, particularly compared with ECD. Moreover, the apparatus is expensive and not widely available. In addition, interference, particularly from water and carbon dioxide in the case of biological specimens, can be troublesome. ‘Purge and trap’ and multiple headspace extraction offer ways to increase sensitivity and, although not needed for most clinical and forensic applications, ‘purge and trap’ has been used in conjunction with GC-FTIR and FID in forensic casework. Pulse heating has also been employed in the analysis of volatiles in biological specimens. This method involves the use of a Curie point pyrolyser employing a ferromagnetic alloy that can accurately attain temperatures in the range 150 to 1040C very rapidly (4 s or so). Advantages of this technique include the use of a small sample volume (0.5–5 μL), short extraction time and lack of matrix effects. Headspace solid-phase microextraction (HS-SPME) has also been used in the analysis of volatile compounds in biological samples.
Chiral GCmethods are available and have been applied to the enantiomer separation of anaesthetics such as enflurane and isoflurane because they have different anaesthetic potencies and side-effects.
Analysis of Products Aerosols and fuel gases can be analysed after releasing a portion of the product into a head space vial, and then transferring a few microlitres of the vapour to another vial for analysis. Liquids can be analysed in the same way, except that it is often possible to withdraw a portion (5–50 μL) of the headspace directly from the container. In this latter case, however, the result may not be representative of the composition of the liquid as a whole. Adhesives and other liquid or semi-liquid products can be analysed by headspace GC.
Quantitative analysis Quantitative assays should involve analysis of standard solutions with addition of an appropriate internal standard.
For liquid and solid analytes, calibration solutions are prepared by adding a known volume of the liquid analyte to a volumetric flask that contains ‘volatile-free’ blood and ascertaining the exact amount added by weighing. Solid analytes are weighed in directly. After allowing time for equilibration, appropriate dilutions are performed, taking care to minimise loss of analyte by handling reagents and glassware at 4C and storing samples and standards at 4C with minimal headspace. Portions of the standards are transferred to headspace vials for analysis, as described above, Calibration mixtures for gaseous analytes are prepared directly into headspace vials. Details of sample preparation can be found in Moffat et al. (2004).
Calibration graphs of peak height or area ratio to the internal standard are usually used to measure analyte concentrations in a sample, although absolute calibration in terms of amount of analyte injected should be possible, especially if an automated headspace analyser is employed. Such apparatus not only permits unattended operation, but also gives much better reproducibility in quantitative work.
