Agilent 5975C, water, and sensitivity

[randy]

I have come upon a problem that has developed with our 5975C and cryogenic preconcentrator analyzing 6L SUMMA canisters. I say developed because the MS seemed to chug along just fine for its first 3 months of use. Now when I inject samples that contain water, the response of the MS decreases significantly. If I go back to injecting internal standard without anything else, I can watch the peak areas of the ISTDs go back up. The preconcentrator does have a form of water control, but I don't think I can eliminate the water completely from these samples.

If the 5975 is really this sensitive to water, how does anyone successfully analyze purge and trap or canister samples with this model? I know our 5971 isn't this sensitive to water. Our 5973 is more sensitive to water than the 5971, but I don't think it's as sensitive to water as the 5975.

Is there anything I can tweak with the MS to lessen the effects of water? I don't know how much is being put onto the column, but it does raise the baseline for about three minutes.

[gpronger]

One of thee things that can happen over time when running purge and trap and related techniques where there is a lot of water entering the mas-spec, is that your rough pump oil becomes saturated with water, which then due to the partial pressure of water, decreases the rough pumps ability to move it out of your mass-spec. The upshot is that every else seems fine, but if the system is hit with some water vapor, you can get reduced response (some type of ionization suppression).

Degassing the rough pump or simply changing the oil may make a significant difference.

Greg

[randy]

That sounds reasonable, and I'll certainly try it, but at high vacuum doesn't the turbo pump primarily remove the water?

[danicrd]

In our lab we currently analyze samples with purge e trap and 5975. The most often symptom we have seen is decreasing area of IS during a batch.
We have bypassed the problem with a clean source.
Perhaps we use Autocan concentrator for analysis of 6l summa can and 5973. Everytime the area of ISTD was a problem is water related: the solution we have found is a longer drypurge time in order to remove more humidity from sample.

good luck

[randy]

That sounds reasonable, and I'll certainly try it, but at high vacuum doesn't the turbo pump primarily remove the water?

I realized the folly of my question once I saw that the roughing pump is connected to the turbomolecular pump.

I have some Inland 45 on order.