Chromatography Forum

Hello all,

I have recently changed jobs and inherited some methods which were carved in stone and carried down the mountain long ago.

We have a method for volatile acids (C2 -C5) which involves aqueous injection of a 0.2%HCl extract of the acids. Concentrations are around 100-400ppm.

The biggest problem is carryover - we run more blanks than samples.

NOW FOR SOME DIAGNOSTICS:

--->If I run the 400ppm standard, followed by blank injections of 0.2%HCl in water, it takes about 5 blanks to clear out the carryover.

--->If I run a temperature profile with no injection - no ghost peaks appear.

--->This seems to suggest that the carryover is not residing on the column, but within the injector plumbing.

--->We are making a 1ul aqueous injection at 150C and 6.3psi. If my HP Flow calculator program serves me correctly I believe we are exceeding the internal volume of the 4mm single gooseneck liner we are using. Therefore the excess aqueous, acidic crud has to go somewhere....probably someplace bad.

--->If I give the system a front end overhaul, new liner, new seal, chop off a foot of column, the system performs marginally better for about 10 injections.

SO:

The first obvious thing to do would be to reduce injection size and stop overloading the injector.

The next step would be to move away from acidified aquous injections..but that's another post.

Any thoughts?

I would like to get some second opinions. I want to make some improvements, and I am up against a certain amount of inertia...
Thanks!

Mike

Mike,

There are some things you can try.

First inject the sample SLOWLY, as you delay the vaporization, you make the flash cloud smaller in volume you have less backflash into the cooler portions of the injector pneumatics. Take 6 seconds to inject the 1 microliter. Some autosamplers can be programmed to very slow injection rates.

Inject a smaller amount. Cut the size in half. Repeat if necessary.

Increase the flow rate just before and after injection. This also helps to minimize the flash cloud.

Get a larger injector volume by changing injection liners.

Temperature program the injection port. Inject cold, heat ballistically to the proper temperature.

Is cold on column possible in your lab?

Good luck,

Rod

Thanks.

It's a 6890 with a 4mm liner so I don't think I can increase the size of the physical injector.

I never thought of doing a very slow injection, I may try that.

The first obvious thing I can do is to try a 0.5ul injection.

Any thoughts on exchanging the acids from the aqueous extract into solvent - hexane or MeCl2 or something a little more GC friendly?

Thanks,

Mike

If you can change solvents without loss of the acids then that would be advisable.

Easier said than done however.

To reduce the number of blanks, use the slower injection rate with a smaller sample and have a plug of a suitable solvent behind your sample to flush any residue of acids out of the injection liner. Again, the goal is to form a small vapor cloud and not to flash sample back into your pneumatics.

Remember, if we knew what we were doing we would not call it research.

best wishes,

Rod

Hi Mike.

I recently put together a method thats almost the exact same as yours and I came up against the same problems and reached pretty much the same conclusions.

My eventual solution was to remove the inlet from the equation entirely and switch to on-column with a single wash of 4% formic acid solution between each sample.

Good luck

Rich

experiencing the same problem here, at first used a new liner, new seal & new needle.

Was working fine for a while, usually 1-2blanks to get a 0 reading.
Using methanol for rinsing the needle.

But now after a while i notice more carry over.

The plunger seems to be much looser in the barrel and stuff, and the inside seems to be alot rougher then when i started out with this needle.

Sounds to me like your needle/syringe needs more washing. Have you tried to increase the amount of washing steps?