Multiple residual solvents analysis

[Pegry]

Hi everybody,

We have to develope and validate a residual solvents method for an API. The solvents we want to analyze are: methanol, ethanol, acetone, ethyl acetate, isopropanol, dichloromethane, 1,4-dioxane, toluene and xylene. I'm going to use a GCMS and a RTX-624 column for the separation. We have an headspace autosampler (with the heated syringe) and our products is soluble in DMSO.
My question are: Is better to use an unique internal standard for the quantification (es. n-propanol) for all the solvents or to use external calibration?
Which is the best way to analize all the solvents with only 1 method?
I've found in literature numerous methods for residual solvents analysis that don't use an internal standard (but they use a different autosampler with the transfer line and not the syringe):is this system more reproducible than the heated syringe system?


Thanks again for the help,

Have a good day,

Davide

[Peter Apps]

Hi Davide

Because it is so difficult (in practise impossible) to find internal standards whose partition ratio between sample and headspace is the same as the analytes, you have to run a calibration series to calibrate the ratio of analyte peak area to internal standard area against analyte concentration.

Check the thread started by zimanli on this subject, and the thread started by Anthony Doyle on headspace ethanol.

All other things being equal, loop and syringe injection give about the same repeatability - the critical step in headspace analysis is control of the sample equilibration temperature, which in some instruments (no names no pack drill !) is not as good as it should be.

Peter