Water/N2 In GC/MS

[S. Chasse]

I have an AGILENT GC 7890A (3 years old). Recntly I began to see Water in our Autotune. From that time the level of water have raise from a low level to a level high enough to stop working with that GC/MS.

I have speak with AGILENT and they sent me on some track, but the water is always the same level now, but still high enough to give bad result.

We know that it is not the MS, we had isolate our MS and no Water was showed.

The problem is I have more Water than N2, like 2X.

I Know it's not coming from our Helium, we have same line for multiple GC and they are all good.
I have changed all ferrule, septum, o-ring, liners, gold seal, look for leaks with a leak detector found nothing.
I have changed o-ring from EPC too.
I have done a GC Leak check and it shows a variation of 0.2PSI in 10 mins.

I am now trying to see if someone had an idea of what happened to my GC.

Thank you,

S.Chasse

[dr_bahram1977]

Do you have any Trap on He line for this Gc? check your traps, ( all traps, maybe hidden).

[Peter Apps]

Can you exclude the possibility that there is water in your samples ?

Peter

[S. Chasse]

Hi,

My Helium line is shared with many GC's and I don't see water for the other GC's.
The trap is shared at the start of the line.

I have not running sample on this GC for weeks and I let oven at 200 Celsius for hours, and I still have water.
I see water in autotune and when I scan. The water is always there, but if I isolated my MS there is no water.

Thank you,

S. Chasse

[Peter Apps]

When you were not running samples for weeks was the MS vented ? And did you have carrier gas flowing through the column ?

Peter

[S. Chasse]

Hi,
MS was on (Diffusion pump Hot, and Vacuum pump on).
I have a flow of 1.3mL of Helium in the column.

Thank you,

S. Chasse

[Peter Apps]

Hi,
MS was on (Diffusion pump Hot, and Vacuum pump on).
I have a flow of 1.3mL of Helium in the column.

Thank you,

S. Chasse

Sorry to be pedantic, but it is important. Did you have carrier gas flowing through the column when the GC-MS was not being used ? The flow you have now is less important.

Peter

[S. Chasse]

Hi,
Yes there was Helium in the column.
And When I installed the new column, I did purge the column with Helium.

[Peter Apps]

Hi,
Yes there was Helium in the column.
And When I installed the new column, I did purge the column with Helium.

This is the first time that you mention a new column. When was it installed ?, before or after the Ms being on standby ? If after, did you check water levels before you changed the column.

Your first post says that the water levels increased progressively - were the water levels normal immediately after you brought the MS back into use ?

Peter

[S. Chasse]

Hi,
I want to let you know that I am pleased that you are trying to help me.

I have remake 2-3 times my first post, and forgot to let the new column in my remake.

The problem occured one day without changing something to the GC, at first the water level was low and increased until I can't run anymore sample.

I tried to find where that water is coming, without succes.
Then I changed the column, for the same result.
The only thing that made water level low enough to run back sample, is when I changed all o-ring and ferrule.
But there are still water that should not be there.

The MS was shutdown once the problem occured, and the level before after was the same. But like I said when isolating the MS, there is no water.

Thank you,

S. Chasse

[Peter Apps]

If changing the o-ring and ferrules made a difference then there must have been a leak. If there is still water then there must still be a leak.

Does the level of water change if you change the split ratio, or if you change from split to splitless (without injecting anything) ?

Peter

[S. Chasse]

Hi,
I changed from split to splitless and here is my result :
Water % stayed all the way around 8%.
N2 level when split around 18%
N2 level when splitless around 13%
N2 level when put back to split 19%

Now there's something I don't understand :

Again I put the inlet to splitless and got a N2 level around 30%
and when I switched back to split still got a N2 level around 30%.

Between each test I waited 5min, and I calculated that it should take 3.5min for Helium to pass from inlet to MS (calculated from velocity and column lenght).

Thank you,

S. Chassse

[Peter Apps]

This confirms a leak somewhere - if the helium was contaminated the water and nitrogen would stay the same.

What split ratio did you use, and what is the septum purge flow ?

Peter

[S. Chasse]

Hi,
Split ratio 25 : 1, 32.5mL/min.
Septum purge flow : 3mL/min

Thank you,

S. Chasse

[Peter Apps]

You need to check again for leaks with the leak seeker - including all the plumbing upstream and downstream of the inlet.

Peter