GC MHE problems/residual solvents

[Sprucebar]

Hi,

We are analyzing residual solvents by GC head space. We inject 4 MHE from a standard vial with e.g. EtOH and then we inject our sample 4 times. We have acceptance criterias for the slope of the logarithm of the 4 areas; R2>= 0.95.

BUT my question is: Is not the slope more of interest, of course should the regression look fine but if the slope of EtOH in sample is totaly different from the slope in standard?? Can we really compare areas between sample and standard if the slope differ? In this case the slope differ by a factor of 10...

Another problem: The first injection often give rise to a smaller peak of EtOH in samples then the second injection (out of 4). The areas should decrease over number of injections. This problem can we only see for the solvent EtOH and none of our other 10 in the same method. Some kind of equlibration problem?

(Thermostating time 20min at 65°C)

BR
Sprucebar

[Sprucebar]

Regarding the slope question:

If we spike our API with a solvent, e.g. EtOH, we get the same slope over the intervall 0.02-0.4 w/w%, and this slope is similar/identical to our standard. When we analyse API we get another slope, but our total area gives rise to a content >0.02%...

[krickos]

Hi

Somewhat rusty regarding MHE try to avoid it when analysing APIs.

Regarding slope. Even if you log the peak area do not expect identical slopes as the matrix is not the same in sample as in standard.
Kolb and Ettre (Static HS GC) has an example where difference in slope between standard and coffe powder with water present differs by a factor 5 but have also seen smaller differences (application note below on polymers).

Regarding first sample injection somewhat lower than expected. Can happen due to that in first extraction the equilibrum is between liquid and air, in extraction 2- 4 the equilibrum is between liquid and gas.

A couple of questions:

Is sample dissolved or not? (guess not as you went for MHE)

General precations (at least for Agilent HS)
Equilibration time you have stated but:
You do pressurize the vial at each extraction?
What is the vent time? The pressure has to be relased properly between each extraction, something like 0,5min or so.

There is a decent application note, might be newer ones too, from HP nr 228-356 from 1996 that perhaps can be found att Agilents site.

[Sprucebar]

Tack för svaret!

I will try to find the agilent document. Perhaps we should calculate the amount from the three last injections!? I found an appl. note that also described the different slopes of std and samples.

The sample is only wetted by water, not dissolved.

We are using a perkin elmer headspace and an Agilent GC, we pressurize tha vial for 3 min each sampling, using a withdrawal time of 0.5min.

//