Residual Solvents by GC Headspace - DMF response

[eboyles90]

Hello,
I am developing a Residual Solvent method by GC headspace the solvents along with the reporting limits I need are listed below. The method is going to be a limits method, so a standard at the reporting limit for each solvent and samples are greater than or less than the limit. I have two primary problems.

Problem #1 - DMF is high boiling point and anything I do to improve the response causes the low boiling point solvents to lose resolution, tail and carryover.

Problem #2 - In order to separate Acetone and IPA I have a 4 min hold at 40C (35C works too). I am having tailing issues with MeOH,I have done some research and its possible the hold may be causing MeOH to tail. The tailing is worse at 35C than at 40C which I think helps support that theory. Granted Problem #1 above isn't helping either.

MeOH ~3000ppm
Acetone ~5000ppm
IPA ~5000ppm
MeCl ~600ppm
THF ~1000ppm
Toluene ~1000ppm
DMF ~900ppm


Aside form the obvious of changing the limits or having two methods does anyone have any advice or experience with a method like this than can offer any suggestions. I've tried different starting temps, ramps, flow rates, pressures autosampler parameters to no avail. Any advice or comments would be much appreciated.

Thanks,
Eric

[rb6banjo]

A description of your sampling parameters and your chromatographic method would be helpful to us as we try to help you with this problem. Please include as much detail about how you are sampling and the analysis conditions (temperatures, flows, column, etc.).

[HPLC chemist]

Why are you not using the USP <467> HSGC/FID method for residual solvents (OVI's)? Why are you reinventing the wheel?

[AaronAIT]

DMF residual solvent by HS-GC can be a nightmare at the best of times, by USP <467> or otherwise. I have used standard liquid injection methods for residual DMF solvent in the past and these methods generally do not give too many problems.

Your best bet might be to run two separate methods for such a situation - HS on a 624 column for everything apart from DMF, and then a liquid injection method for quantification of DMF. You should not be facing long run times in either case, if your temp. programs are well optimised.

As HPLC Chemist says, use <467> as a starting point for the HS method in terms of column phase, flow rate, inlet temperatures etc and if you need to modify your temperature ramp or HS equilibration time to reduce the run time, you can do so.

What is your limit for symmetry factor that Methanol is failing? On a brand new DB-624 column at 40DegC and ~35 cm/sec carrier flow, Methanol peak shape won't be fantastic, but it shouldn't be a game changer.