peak in DMF blank injection

[emmcoo]

Hi
i am running a GC headspace residual solvents method. one of the residual solvent of interest is methanol. I am using DMF as the diluent and when i run a blank injection of the diluent i get a significant peak at the methanol retention time which would cause interference. The brand of DMF I am using is sigma aldrich GC headspace tested >= 99.9%. any suggestions on what one should do? would other grades of DMF provide less interference? use this DMF and when determining methanol in the samples account for the area obtained in the diluent?

thanks

[chromatographer1]

If water contaminates the DMF, a reaction by-product will be formed called methylamine. It will elute very close to methanol on most columns and appear to be methanol. It only takes a few parts per million to produce an interfering peak.

You have a choice of distilling the DMF to remove the methylamine or purchasing a fresh bottle or another brand.

I advise you to buy bottles in smaller sizes so the DMF is used more quickly from container to container and to purge the bottle after opening with an inert dry gas after each use.

Do not keep bulk amounts in storage but have a continuing purchase and delivery of fresh solvent if you wish to avoid this contamination in the future.

Of course, if the stock you purchase is old and was packaged some time ago this contamination may be unavoidable.

Good luck,

Rod

[sisay]

Hi all

I have dug all topics regarding interference in GC analysis as listed below:

viewtopic.php?f=2&t=17592&hilit=interference interference from DMF
viewtopic.php?f=2&t=8733&hilit=interference interference from DMSO
viewtopic.php?f=2&t=2224&hilit=interference interference from DMF/water

Currently we have the same problem regarding the interference from DMF/water (1:5) blank solution in the region where ethanol elutes when verifying USP <467> procedures B:

Chromatograph system:
Agilent 7890A with FID
Agilent G1888 headspace autosampler
DB-WAX 30m*032mm*0.25um USP G16
Temperature: 50C for 20 min, to 165C at 6C/min and hold for 20 min
Carrier gas: He at 35 ml/min
Injection temp: 140C
Detector temp:250C
Detector gas: H2 at 35ml/min, Air at 350ml/min, N2 at 30 ml/min
Split ratio: 1:1
DMF: Merck GC grade
Water: Merck LC grade
DMSO: sigma
(It is a USP <467> procedure B settings)

Injected samples:
1. Two peaks were observed in a 6 mL of DMF/water (1:5) solution. 2nd peak at 2.96 min interfered with ethanol at 3.07 min.
2. 2nd peak was reduced (about 50% reduced) in a 6 mL of DMF/water (5:1) solution (invert the ratio), while the area of the other peak was not changed.
3. 2nd peak disappeared in a 6mL of water and a 6mL of DMF respectively, while the area of the other peak was not changed. -->Proposed something produced after mixing these two solvents-->as discussed in the topic above, it may be “methylamine” or any other by-product or stabilizer?
4. 1st peak was observed in an empty headspace vial (i.e., neither solution nor sample is inside the vial)-->Proposed 1st peak belongs to environmental interference.
5. Two peaks were observed in a 6 mL of DMSO/water (1:5) solution. 2nd peak at 3.08 min exactly interfered with ethanol at 3.07 min. The peak area is 200% of sample No. 1. (USP <467> allows us to use DMSO to substitute DMF)
6. Follow USP <621> to adjust initial temp from 50C to 45C and 55C (+/- 10%), but two peaks still can’t be resolved.

We are planning to:
1. Use different brand name of DMF (including Fisher, ACROS, Sigma, etc.) to evaluate the interference from difference sources.
2. Use newly opened DB-WAX column to see if these two peaks can be resolved.

Please help to provide any idea to dismiss this interference. Thank you very much!

Happy thanksgiving,

[kubowicz.tomasz]

Hello

I'd try to run blank - empty capped vial, to see if it is not peak from headspace sampler. DMF is very "nasty" solvent and it has tendency to stick in sample loop. The only way to solve the problem is to replace sample loop in headspace sampler.

Regards

Tomasz Kubowicz