Unstable Baseline using Headspace Injection

[Yorkie31]

Having a problem with an unstable baseline at a certain point in the run for a 12 solvent GC method development. The problem is when the last solvent elutesbefore the diluent peak. Sometimes the chromatoraphy is perfect for this last solvent, and at other times the peak is overshadowed by a rise in baseline before this last solvent peak elutes. The 12 solvents are at a concentration of 1ppm diluted in Dimethylsulphoxide.
Column used is a ZB-1 30M x 0.53 mm x 5µm which is giving good seperation of each solvent.
Headspace oven temp is 160°C, Loop 170°C, transfer line 180°C, GC Inlet 210°C. Injection volume is 1mL of headspace with a split ratio of 20/1.

[Rafael Chust]

Can you provide more details of your temperature program?

I would say it is too low and you have late elution of stuff stuck inseide your column...

[KennyH]

Cheers for the reply Rafael

oven program is 40°C for 4 mins, increased to 110°C at 8°C/minute, to elute all solvents of interest, then increased to 300°C at 50°C/minute and held for 2 minutes.
Inlet temp is 210°C and detector temp is 250°C.
Have steam cleaned headspace system but with no improvement to chromatography.
Have ran the oven program without performing an injection and the profile is clean so the contamination would appear to be coming from the headspace.

[Rafael Chust]

Does your headspace have a purge/cleaning option?

Which kind of septas are you using in your system?

Have you tried "bake" your headspace at high temp, and them repeat the analysis?

[KennyH]

The maintenance of the system only recommends steam cleaning. Done know of a purge option.

THe septa used are agilent low bleed optimised septa, so should be good for the job, its the ones ive always used in the past.

I left the headspace loop and transfer line at 220°C over last night, but hasnt made any difference.

[Rafael Chust]

Stupid question: this baking was done with gas flowing through?

[KennyH]

yes, gas was on.

[Rafael Chust]

Are you working with standards or the sample itself?

Did you try to inject the solvents in separate to see the behaviour?

[chromatographer1]

I am trying to determine exactly what is happening from your post. If I misunderstand please forgive.

One question I have is WHEN this occurs. First injection of many ?

or after several injections have occurred?

The rise in baseline is probably the residual DMSO left in the inlet (placement of the needle in the injection port) or on the column (the 'tail' of the previous diluent peak injection).

One solution is to increase the time you 'cook off' the column at the end of the run. I would suggest you do not use 300°C for 2 minutes but 280°C for 10 min. and see if that helps to better elute the DMSO 'tail' from the column.

I would elute DMSO diluent from my HS columns at a much lower temperature but for a longer time when I did my HS work 10 years ago. I also directly hooked up the transfer line to my column instead of using a split injection.

Perhaps a reading of the article might be helpful.

Rapid Analysis of USP Organic Volatile Impurities and Thirteen Other Common Residual Solvents by Static Headspace Analysis
Journal of Analytical Chemistry 1997 June

best wishes,

Rodney George
Senior Research and Development Scientist
Gas Separations Research
Supelco
595 North Harrison Road
Bellefonte, PA 16823
rgeorge@sial.com