GC question about seperating DMSO

[cbowen]

Hi all,

I have been running a split residual solvents method on GC to separate ethanol (10%), acetonitrile (410ppm) and DMSO (5000 ppm). However my boss wants us to switch to splitless because all our other lab sites do and we weren't getting great seperation between EtOH and ACN. So I started working on that and came up with a decent separation of all. However, when I started to run system suitability and some accuracy/precision runs DMSO failed with %RSD numbers of greater than 10% and also did not pass a linearity exp. I repreped all my samples and still the same happened so I do not think it's my sample prep.

My question is how to get DMSO to pass? Is it even possible? Here are my conditions for my new method. I'm still quite new to GC (only about 6 months experience) so I have a feeling I am overlooking something really simple here! Thanks in advance for the help!

GC: Hewlett Packard 5890 series 2 with FID and an autoinjector
Column: Restek Stabilwax, 0.53mm x 30m
Carrier Gas is Helium at 6.0 ml/min
Initial temp is 68C for 2 min
Temp ramp is 50C/min to a final temp of 210 for 1.5 min
The purge valve opens at 0.2 min
My sample is diluted 50 fold as well.

Let me know if you need anymore information! I also realized today we are still using the same split linear with a small bit of steel wool inside...I'm not sure if that could effect this. I also used to notice in the old method a lot of carryover of DMSO that would sometimes require me to make more injections or do blank water washes till it finally passed system suitability.

[Peter Apps]

What is the injector temperature ?

Surely you do not really have steel wool in the inlet isn't it glass wool ?

The concentrations that you give are unbelievably high for residual solvents.

Why the very rapid temperature programme ?

What solvent do you use for dilutions ?

Peter

[cbowen]

Yes I mean glass wool! Sorry!

My injector temp is 225...I didn't set this, just carried it over from the old method. What should I consider when deciding on an injection temp?

The concentrations are our limits in our release specs. 5000ppm for DMSO and 410 for ACN, 10% EtOH shows up because it is in our diluent. I make a solution with these concentrations then dilute 50 fold for the actual 1uL injection.

The temp program doesn't necessarily have to be quite so aggressive but I am doing QC for a radiopharmaceutical with a half life of about 5-6 hours so our old method was about 10 minutes long that allowed for reinjections and other problems that would arrive before our patient was injected, etc etc etc.

The solvent is water and ethanol (9:1) because that is what our drug diluent is.

I called Restek and the rep said he thought it was blackflash contamination from DMSO so we are switching to a proper inlet (with a larger volume) and I am following agilents cleaning procedures so hopefully that will help solve the problem.

[Peter Apps]

If you want the analysis to run faster, and as long as you do not have peaks that are barely resolved, then cut the column in half.

Injecting 90% water will give you pressure pulse problems for sure, and will also eat up the column. You would do better diluting just in ethanol - if I understand correctly this is not actually a residual solvent.

Your existing inlet temp is OK.

Peter

[cbowen]

Thanks Peter I will try your suggestions!