Thermal/flush desorption

If the thermal desorption unit is able to provide inverted (back-flush) gas flow during trap desorption (Figure 1.2), the same technique described for tubes can also be used for the cold trap. A multi-bed trap can considerably extend the analytical window of the instruments. Commercial cold-traps packed with quartz beads, quartz wool and Tenax TA are reported to cover a substance range from C6 to C40. 

These semi-preparative methods are useful where identification is required but for quantitative and comparative analytical purposes much more rapid sampling techniques, such as automated headspace and solid phase microextraction (SPME), may be preferred. Both of these techniques give similar results for most volatiles. In the former, the vapour above a heated sample is removed by a syringe or gas flushing and injected onto a GC column, either directly or after trapping on a suitable absorbent and thermal desorption. In SPME, the vapour is absorbed on to a suitable bonded medium on a special needle and then injected into the gas chromatogram. 

If the target analytes do not readily evaporate from the matrix they are in, then another recently improved technique - direct thermal desorption - is more effective. This is often used for dry, non-volatile matrices like wood, soil, spices or resins. The sample is placed directly in the liner or desorption chamber, which is then flushed with inert carrier gas and heated rapidly to transfer the volatiles to the analytical column. 

Acidity of the samples were determined by temperature programmed desorption of ammonia. The calcined samples were heated at 450°C for 3 h in helium flow followed by cooling to 100°C. To minimise the amount of physisoibed ammonia adsorption was carried out at 100°C and any physisorbed ammonia was then flushed out in helium flow (50ml/min) for 3 hours. Finally, TPD spectra were recorded by heating from 100 to 600"C at a heating rate of 10 C/min. The amount of ammonia desorbed was monitored using a calibrated thermal conductivity detector. 

This problem appears during the up-side of the desorption trace and on the downside. During the up-side the output concentration is too low to account for all the material desorbed due to the accumulation of material in the void space while during the down-side it is too high due to the flushing out of accumulated adsorbate from the void volume. The overall effect is that the output concentration peaks are broadened, their maxima are shifted to longer clock times and their amplitude is reduced. In gas phase systems the peak is further flattened by the thermal expansion of the sweeping gas. 

Figure 15.5 Schematic of CHNS analyser. 1. Flushing gas (He), 2. Combustion gas (O ), 3. Gas flow meter, 4. Moisture trap (P2O5), 5. Carousel, 6. Ash finger, 7. Combustion tube, 8. Reduction tube, 9. Furnace, 10. SO adsorption / desorption tube, 11. H O adsorption / desorption tube, 12. CO adsorption / desorption tube, 13. Reference gas (He), 14 Thermal conductivity detector.

The sample chamber is set to a temperature of 40 C for pre-heating and extraction. This temperature was chosen for a good representation of the released VOCs of the cheeses as a mock-up of the typical mouth feeling. In this dynamic headspace sampling step the volatile headspace is flushed from the cheese samples by using the inert gas nitrogen for collection onto the sorbent packed tubes. After collection the tubes had been transferred to a thermal desorption unit. 

The total acidity of the solids was evaluated by Programmed Temperature Desorption (TPD) of NH3. For those experiments 0.10 g of samples seived (2OO<0<315 p.) were treated for two hours under helium flow up to 400 or 500°C. After adsorption of NH3 for 15 mn at 100°C, the solids were flushed with helium flow for 1 h at 100°C in order to eliminate physisorbed ammonia. The heating rate during TPD experiments was 10°C min i. The NH3 desorbed was detected by a thermal conductivity detector. 

---