Loss of resolution after replacing inlet glass liner

[Don_Hilton]

On the question above regarding the seal with the vespel ring. I've used them and they worked fine for me. Like the gold seals, I only installed the seal once. On removal I made the assumption that you can never put it back in the same way and that the original cut would become a leak.

[DocHollywood]

The gold seal did not solve the problem. If anyone has any more advice i'd love to hear it.
Thanks
AD

[AICMM]

DocHollywood,

I would take out most, if not all, of the wool in the new liner. Your samples look to be gas phase anyway so you probably don't need the vaporization boost that wool provides. Can you give us the part number for the liner?

Best regards.

[DocHollywood]

Hey,
Here is the part #19251-60540 liner with deactivated glass wool. The old liner had glass wool in as well. i will post new pics of the methane peak as to better illustrate the problem.
Thanks
AD

[AICMM]

DocHollywood,

I would still pull out most of the wool. I think that liner has way too much wool anyway.

The other thing this kind of looks like is that the purge valve is off (meaning the injection is splitless instead of split.) Did someone check of splitless injection in the method?

Best regards.

[DocHollywood]

Thanks AICMM
I have tried taking out some of the quartz wool like you mentioned earlier.
Also, the split mode is working as i can see the same flow as I saw before the problem coming out of the split port. If you notice the peak heights are different, which indicates to me that the same amount is being injected into the columns. If it was because, somehow, i was running in splitless and too much sample was being injected into the column i would expect the peaks heights to be at the very least the same (or the bigger amount higher), with the bigger amount much broader than the other.

Thanks again
AD

[AICMM]

DocHollywood,

A comment and a question. Comment: don't be bashful about taking out the wool. You are shooting gas phase components so you don't really need it. Now the question. Are you using a 5890? If so, the flow out the split vent is the same whether it is running in split or splitless mode because of the way the 5890 is plumbed. So, when you make the injection what is the status of purge A (or B)?

Best regards.

[DocHollywood]

Hey,
Purge A comes out at around 2ml/min.
Thanks
AD

[Schmitty]

I think AICMM would like to know wether the purge is "ON" or "OFF" at the start of the run, and the times at which they do turn on and off. Also, if the purge is set to turn off at 25 min, and your run is only 20 minutes long, the firmware for the 5890 will not ever set the purge back to an "OFF" state.

[DocHollywood]

Thanks for the clarification

Here are the settings:
Purge valve on: 1:00
Purge valve off: 0:00

I assume the purge valve is always on in this case.

Valve 1 - Off
Valve 2 - Off
Valve 3 - On
Valve 4 - On

Thanks for helping a novice such as myself!
AD

[Schmitty]

I see. I usually have the purge turn off before the end of the run (0.1 min or so). In my experience, it's better for consistency at the injector port.

I cannot see how this would contribute to the dramatic change in chromatography, since this setting is theoretically the same as before the liner change. If you have time, it is somethig to try next!

[Peter Apps]

The bad peak has a classic splitless profile, while the good one is classic split (with methane there will be minimal focussing on the column so what you see at the end of the column is the same profile as left the inlet).

The real puzzle is that with a split purge flow of only 2 ml/min you got a sharp peak with the old liner.

I suspect that you had a leak with the old liner that was givng you a higher split ratio than you had set and measured at the split flow outlet. You have now cured the leak, and you are getting a poor profile because the real split ratio is now 2:1, and this is too low to flush the liner quickly.

Try increasing the set (and measured) split flow to about 10 ml/min.

Peter

[DocHollywood]

Peter,
Thanks for the reply,
I think i did not clarify the problem well enough.
I have a 5890 series 2 with an EPC inlet system (not the manual).

From the purge vent i measure 2 ml/min.
From the split/splitless vent I have a flow of 59 ml/min.

I thought AAICM was asking for the purge vent flow, he mentioned with how the 5890 is plumbed you don't know if the split valve might be stuck in splitless because there would still be flow coming out of the split/splitless vent.

My split ratio is around 55:1 (which, with the new liner still might be too high?).

Thanks a ton guys, this problem has already downed my research a few weeks.

AD

[AICMM]

DocHollywood,

Some other things I would try. First, try your current run conditions except cut the Purge A on time down to 0.25 minutes. Second, try a split run and post that for us. Meaning, purge A on at 0.00 and never off. Third, I would still say remove ALL the wool.

By the way, if the splitless/split valve fails it fails in the split position.

Best regards.

[Peter Apps]

OK, now I see that you are doing a splitless injection, which would give the problem profile. My suspicion that there was a leak before still remains - you were then doing a split injection during the splitless tile, and increasing the split ratio when the split valve opened. Now the leak is fixed you are really doing a splitless injection, and becuase there is no focussing the peak has a tail.

Go with AICMM's second suggestion - a split injection. If the peak is too small reduce the ratio to 10:1 from the current approx 60:1. And yes, you should get rid of the wool.

Peter