DDT Breakdown

[kdamme]

I'm running a method for pesticides for little over a year now on Agilent dual-channel 7890 GC's.

In the past we didn't have any problems with DDT-breakdown, but since a few weeks we see that during a series DDT in the first standards is OK, but 15 or 20 injections later no peak is found anymore.

We use the same type of liners and analytical column since we started. The guard column was replaced by another type, but the new type is also used by another pesticide analysis we do without any problems (that analysis uses cold on-colum injection) and our suplier says that this type should be at least equal to the type we used before.
The frequency of replacing liners, guard columns and analytical columns hasn't changed (or even increased since we have this problem).

We do not use press-fits, but use a self-made system in which the analytical column is put into the guard column for about 1 cm. This connection is used for years and didn't give any problems so far and is still working on other applications in which pesticides are measured.

The problem occurs on both GC's on both channels.

Any ideas on the things I could check now?

Kevin

[chemwipe]

Any ideas on the things I could check now?

I've had problems before with my Endrin/DDT solution not passing the method criteria.

You can try the following:

Remake your degradation solution.
Trim the ends of the analytical and guard column where they're connected (you might also want to cut a foot or two on the injection side - both the guard column and analytical column).
I've had a bad lot of liners before, maybe see if you have a different lot available.

There's a method online for cleaning the injection port with a gun brush, methanol (can't seem to find it right now) and acetone. I've been told that it's not that effective, but I was able to resolve my problem with my suggestions above in addition to cleaning out the port.


John

[kdamme]

Great idea for the different liner type: I'll order them tomorrow from a different supplier.

I'll also cut some cm from our columns, although I think that's not the solution I'm looking for: the column is recently replaced completely and that didn't help. Nevertheless: an extra cut is little work and might help.

Both systems are serviced about 2 months ago by Agilent and as they know the type of application we do, I suspect they also cleaned the injection port. I'll check that on the service report. If thet didn't do that [i]and[/i] you can give me the web-address I'm going to clean it myself.

Kevin

[Ron]

Have the gold seals at the base of the injection ports been changed? If they are a year old I would suspect that as a potential problem area for breakdown.

[chemwipe]

Have the gold seals at the base of the injection ports been changed? If they are a year old I would suspect that as a potential problem area for breakdown.

Just remembered another thing: I always use siltek treated liners and siltek treated baseplates. They seem to be more stable than the gold ones.

And I would think a year is a long time to go without replacing them!

John

[Balderquell]

Is that 15-20 injections later referring to standards or samples?

If standards, are you getting septa shavings in your liner?

If samples, I'd blame it on the samples. Running dilutions or performing cleanup procedures? One dark colored and smelly extract can be enough to cause degradation and murder a CCV for DDT and Endrin.

If you can get your degradation to pass at all, then you're doing something right. I'd expect 20 or so samples to be enough to cause problems, however.

[kdamme]

Ron, Chemwipe: I contacted Agilent, and they say that the gold seals should have been replaced in June when the systems were serviced. I will order a set of gold seals anyway to make sure they are not the problem (or maybe they are..)

Balderquell: There are 15-20 injections of samples. But I try to avoid dark colored extracts being injected into the system: not only will they have effects on the liner, but there also might be heavy compounds giving problems with my columns. I suppose -but have not tested yet- that injecting 20 standards will show a decrease in DDT (and Endrin) hights.

[chemwipe]

give me the web-address I'm going to clean it myself.

Kevin

Here you go:

http://www.chem.agilent.com/Library/Sup ... a16022.pdf

John