Repeatability and Standard Additions in USP 467

[4meljones]

Does anyone use USP 467, procedure C to quantitate their residual solvents (in water soluble articles)? It is a standard additions method using headspace GC and FID detection. I am having trouble making it work. My issues are:

1) For one of my analytes, chloroform, the peak area detected by FID of the "spiked" sample is very close to that of my "unspiked sample" and this analyte is not very soluble in the extraction solvent (water);
2) My variability is very high (% RSD of replicate injections of standard or sample > 5%);


Generally, what kind of %RSD should I be expecting to make a standard additions method work? There seems to be overlap between the response in my spiked and unspiked samples. For headspace methods where water is the extraction solvent, is there any reason why I should not have headspace sample loop and transfer line temperatures > 100°C to avoid variability caused by possible condensation of water, where my vial equilibration temperature is 80°C?

Melissa

[chromatographer1]

Sample prep errors usually are the cause for this problem. CHCl3 has a very poor solubility in water and if the sample is not sealed immediately after spiking OR if the spiking solution is not homogeneous in its CHCl3 content losses of CHCl3 become apparent when the samples are run.

I validated my own method for water and non-aqueous solutions because of the problem so many had with the USP method.

Good luck.

Rod