large amount of water in GC/MS

[allz13]

Hi!
I recently started to work with an old GC/MS. The GC is an HP 5890 and the MSD is a HP 5973.
The autotune doesn't show anything wrong apart of too many picks (around 350) and high percentages of water and N2
The water is around 25% and N2 around 11% even after several days from being pumped down.
We vented the system and pump it down again to be sure that the side plate of the MSD was perfectly close. And the software indicates that the turbo pump is working at 100% speed. We also verify that all nuts were tightened.
My question is that if this kind of water percentage could only be caused by a leak or if it could be something else.

Thanks

[bhuvfe]

25% after days of pumping down is not typical.

Few thoughts.

Was your system exposed to the moisture of the lab for a long time (I mean months)?

Check if the heating elements are OK. A defective ion source and quadrupole heaters (or the wrong T) will have an effect on the removal of water by the pump.

Is the transfer line heater working correctly and at the right T?

Is the column polar and has it been exposed to the lab (or it is new)?
Some columns retain some water but this generally disappear after few blank runs.

Last but not least. Do you have any reason to suspect the purity of the carrier gas or the efficiency of the traps?

Regarding the air you have first to know if it's coming from the inlet or not
(search the forum and you'll find several info about this).

Good luck.
bhuvfe

[thohry]

You can also isolate MS from GC to see if the problem comes from column and/or carrier gas.

[Don_Hilton]

A quick way to check to see if the water is coming from the inlet: With the showing a profile, turn a can of dust-off upside down and spray liquid across the column close to the transfer line. This will chill the column and cold trap moisture going thrugh the column. The m/z 18 signal will drop and then spike once the column warms up. If the m/z 18 sigaal does not drop, the water is entering from somewhere else.