|
The separation and detection of explosives by chromatographic techniques is fraught with danger, especially for the more unstable compounds like triacetone triperoxide (TATP) the detonator of choice for the infamous shoe bomber. In gas chromatography, the risks are exacerbated by the need to heat the column during separation, so rapid carrier gas flow rates and thin stationary phases films have been proposed to minimise residence times in the column. The stationary phase normally recommended for GC columns running explosives is 5% phenyl polydimethylsiloxane (PDMS) but two scientists in the USA have noted that column selection is not often discussed in the literature. Scott Harvey from the Pacific Northwest National Laboratory in Richland, WA, and Thomas Wenzel from Bates College, ME, have had some success isolating explosives by SPME, using a polymeric metal beta-diketonate as the fibre coating. They used the same material in packed columns for the selective retention of explosives. They decided to see if this polymeric material would be as effective as the stationary phase in capillary GC columns. They examined the suitability of four lanthanide beta-diketonate polymers based on lanthanum, terbium and europium and the ligand p-bis(4,4,5,5,6,6,6-heptafluoro-1,3-hexanedionyl)benzene. However, it was not a straightforward exercise because the polymers have limited solubility in methanol and are insoluble in most other solvents. In the first instance, the duo adopted a dynamic coating process, in which they attempted to coat an existing crosslinked PDMS thin film by passing a plug of a methanolic solution of the polymer through the column under pressure at 1 cm/s. The retention index for TNT on the resulting column was more or less the same as that on the PDMS-coated column alone. The coating process had not worked. They tried a static coating process next in order to co-deposit the polymer and SE-30 polymers in a fused silica capillary with non-polar surface deactivation, using concentrations based on published work. The lanthanide polymers dissolved only in methanol. SE-30 required a hydrophobic solvent and the solvent mixture should have as low a boiling point as possible for removal after coating. They found that an azeotropic mixture of methanol and dichloromethane (92.7:7.3, w/w) dissolved the required concentration of SE-30 (0.4%) and provided sufficient solubility for the lanthanide polymers. Its boiling point (37.8°C) was lower than those of the individual solvents. The columns were filled with the coating solution, plugged at one end, and immersed in a water bath held at 30°C while a vacuum was applied to the other end. After coating, they were purged with helium for several hours then thermally conditioned in the GC oven before use. The researchers believe that this is the first time that an azeotropic solvent has been used for GC column coating, so they evaluated the feasibility of the process by preparing a control column from the same solution in the absence of the metal polymer. The separation efficiency was 2700 plates/m, indicating a reasonably efficient column. The column efficiencies, examined using a mixture of n-alkanes, were 2200 and 2600 plates/m for Tb and Eu polymeric columns, respectively. For one La column (La1), the efficiency was 290 plates/ms but for a second (La2), containing 50% of the La polymer in the first, the value rose to 1000 plates/m. Column selectivity was evaluated with an eight-component mixture of nitro-based explosives and cyclohexanone, a volatile component associated with RDX and of forensic interest. The columns with the best efficiency displayed the poorest retention behaviour. So, La1 gave the strongest overall retention with Eu the weakest and Tb and La2 intermediate. The thermally sensitive explosives TATP and nitroglycerine were chromatographed using a reduced injection temperature to minimise decomposition. Nevertheless, TATP decomposed on the LA1 column and was not selectively retained on any of them. Nitroglycerine was selectively retained to a moderate extent on the La1 column but not on the Tb and Eu columns. The test data have illustrated the potential of these novel columns. The separation mechanism is very different to those on most GC stationary phases, being based on Lewis base interactions with the Lewis acidic stationary phase. This led Harvey and Wenzel to conclude that they would be of value in multidimensional GC methods, especially as the first dimension. The novel protocol, employing the first published use of an azeotrope for the static coating of a GC capillary column, has produced new stationary phases for the GC separation of explosives and related compounds. Related Links:
The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd. |
 |
|
|
|
|
