Activity of the FID detection port: a big problem if underestimated.
31 Mar 2011
No, this is not an April fools joke although there will be readers that probably do not believe this.
Yesterday I received an article from Dr Matthew Klee about the FID detector and its operation. This detector is great but one has to be careful.
It immediately reminded me to some work that was done half a year ago on this detector, if it was not used in the correct way. The reason for doing some investigation was, that the contribution of detection-port activity is forgotten many times when discussing trace analysis. My question was: what happens if I do not install the column in the detector in its correct position? I looked at literature and could not find any data on experimentation.
During one of the many discussions I had with Dr. Jim Luong, a famous GC scientist at Dow Chemical, it was very clear that activity at the end of the GC system is often neglected.
We have seen columns and transfer lines get incredibly active, just in standby position in a normal FID, where the carrier pressure was stopped and the detector flows and temperatures were still on.
This is a big challenge for FID and reason for my question: What happens inside FID liner if you change column or when the pressure is stopped for some reason? Officially we always should cool down the injector and detector when doing maintenance, but is that always done in a routine laboratory? The FID usually is at high temperature, so if the column flow disappears, or the connection is opened, we can expect water/oxygen to enter this area.
After talking with colleagues they also told me that detector position is important, but again, there was no data. All shared it was very trivial to install the column correctly and if you do not do that, you are not a good chromatographer. My point is, to make people aware of the reason WHY we need to pay attention to correct column installation.
If we want to test a GC column for inertness, it is a valid concern to be aware of the discussion above. Columns are tested with nanograms of highly polar, base/acid analytes to verify its performance.
A simple experiment was done. A Rxi-5Sil MS column was tested in an Agilent 6890N GC, and a chromatogram was obtained as in figure 1;
Correct installation: Column end close to flame tip
Now the column was pushed 1 cm lower from the flame tip and the same analysis was done. Fig. 2 shows the result. The last 10mm of the FID caused a reduction in response for the polar analytes of almost a factor 2 and also note the tailing.. Figure 3 shows a detail of the first 4 polar components. No column would pass QA if it showed this behavior. The last 10mm of the flame tip in the 6890 showed huge adsorption and activity and caused loss of many components.
To maximize the impact of detector I reduced the detector temperature from 300 to 250C and took the column back another 20mm, just to see the effect. See figure 4.
This would not be very realistic, but it surely make a good point on what happens in the last 30mm of transport.
We tested this also on a second GC which gave similar results. Strange enough, testing in an old 5890 system did not make any difference in any position or temperature tested.
The activity /reactivity of the flame tip in FID always depends on the moment. One can replace the flame tip for special deactivated ones (there are Siltek deactivated tips available), but activity in detection systems can also often be “masked” by conditioning columns at upper temperature and as some bleed products are formed, they can deactivate the detector, so it will perform OK. This “conditioning-deactivation” is a known method for deactivating transfer lines and traps f.i. in SCLD and MS systems.
For good GC we need a full inert gas path: We can help with Inert liners(like Restek Premium) and Inert columns (Rxi-series), but for best results, one needs to be aware why columns are to be installed in certain positions in injection and detection systems.