SFE Trace Enrichment and Windowing

To get the most out of the HPLC as a time machine, we have to analyze the separation, then put to work the extraction techniques we have discussed to speed the separation. As an example, we will take a metabolite study carried out by one of my customers. His problem was to study the dispersion of a compound, XX, in the environment. He had to follow its fate and metabolites in samples of water, sludge, soil, and fish tissue. 

In examining environmental run-off water he ran into a problem. His samples were very dilute, so the first step would be some form of concentration. Studies with pure material show that it is sufficiently nonpolar to stick easily to a Ci8 SFE cartridge from an aqueous solution. 

He prepared the SPE by first wetting it with 2 mL of methanol and then with 2 mL of water. Next, the SPE was attached below a particulate filter in a line leading to a 1-L suction flask (Fig. 12.1a). The inlet of the SFE was fitted with a tube dipped into a 1-L flask of the runoff water containing our compound, XX, at 5 parts per trillion. (He was unable to detect this level of compound by direct injection into an HPLC system.) 

To check recovery of compound from the medium, he ran a labeling study. Radiolabeled sample was added to runoff water at 5 ppt, it was sonicated, extracted with a wetted SPE, eluted as before, and counted. He found a 97% recovery of radiolabel from the SFE. 

The problem with the separation at that point was that he had increased concentration but not improved his run time. Even though he was interested in only a single component, he had to run a gradient to separate it from the rest of the mixture. He still needed to prevent early running, more polar compounds from running into the sample and washing out late runners before the next injection. His bottom limit on run time was about 45 min, with a 15-min reequilibration to achieve reproducible results. 

---