Purge and Trap Mystery

Discussions about GC-MS, LC-MS, LC-FTIR, and other "coupled" analytical techniques.

15 posts Page 1 of 1
Hi All,

I'm having an issue with one of our 524.2 instruments. Earlier this week the recoveries suddenly dropped to ~15% of the calibration. I've tried everything i can think of to locate the issue and can't.

I'm running an Archon connected to an OI 4560 to 6890/5973. 11 minute purge at 40 mL/min 1 minute dry purge, desorb pre-heat to 135 and 4 minute desorb at 190 (we can't change the desorb time) and a 10 minute bake at 210. so far i've run an electronic leak detector over every line and joint and found nothing. I measured the flow from the Archon (we use soil mode for everything) and it's right about 40 mL/min. i then took the trap off and put the flow meter at the water management were the trap connects and it measures ~ 40 mL/min. From the transfer line to the GC only measures only 5-8 mL/min. So far i've tried changing the P&T-GC transfer line, 6-port valve, and the lower valve on the manifold. None of that changed anything, i'm concerned it might be the EPC on the 6890 but don't have one to swap out and check. Does anyone have any ideas?

quick edit: i did do a direct inject and everything was fine.
Have you checked trap by swaping it with new one ?
What is your column flow rate and split ratio? A flow through the transfer line of 8ml/minute and a column flow of 0.8 ml/min would be just about right if using 10:1 split ratio, if using 50:1 then it would be very low.
The past is there to guide us into the future, not to dwell in.
You may want to change your soil probe and see if this takes care of it (this was our problem). If not, carefully check the fittings connected to the probe (BE VERY CAREFUL, BECAUSE THE SURFACE IS HOT). Good luck.
James_Ball wrote:
What is your column flow rate and split ratio? A flow through the transfer line of 8ml/minute and a column flow of 0.8 ml/min would be just about right if using 10:1 split ratio, if using 50:1 then it would be very low.


My column flow is 1.2 mL/min at a 40:1 split ratio.
rv3 wrote:
James_Ball wrote:
What is your column flow rate and split ratio? A flow through the transfer line of 8ml/minute and a column flow of 0.8 ml/min would be just about right if using 10:1 split ratio, if using 50:1 then it would be very low.


My column flow is 1.2 mL/min at a 40:1 split ratio.


This should give you a flow of about 48-50ml/min out of the transfer line. But, the 6890 will be setting flow based on the measured flow through the split vent and the head pressure so you need to leave the transfer line attached to the injection port and measure the flow at the split vent on the back of the instrument top panel. If you disconnect the transfer line to measure the flow the instrument won't know how to control the flow and you never know what it will be sending through that line at that point.

What trap are you running that you are using such low temperatures for desorb and bake? Is it the default Tenax, Charcoal, Silica Gel listed in the method?
The past is there to guide us into the future, not to dwell in.
James_Ball wrote:
This should give you a flow of about 48-50ml/min out of the transfer line. But, the 6890 will be setting flow based on the measured flow through the split vent and the head pressure so you need to leave the transfer line attached to the injection port.

What trap are you running that you are using such low temperatures for desorb and bake? Is it the default Tenax, Charcoal, Silica Gel listed in the method?


I usually plug the connector to the injection port when to keep pressure when measuring through the line, i did however switch to checking the split vent. While in desorb the flow drops from 28->15 mL/min then goes back up to 28 once it switches to bake. I'm guessing that implies the EPC?

We use OI #10 traps, i don't remember the mix ratios but it is the standard composition. I have personally never done anything with the Desorb and Bake times or temps, that what it was when i started at the lab. I have noticed most people use higher temps but from what i've seen these seem to be the default temps from the manufacturer.
Looks like the #10 is the old Tenax, Silica gel, Charcoal which would use the low temperatures due to the thermal limit of the silica gel and tenax. It is much better to use a Vocarb3000 if you can get it and then desorb at 245c and bake at 260c, must faster to transfer the analytes to the instrument.

With that drop in flow it seems more related to the trap or the valve in the P&T. If there is too much restriction to flow through the trap, the EPC can't increase the pressure enough to compensate and the flow drops. The other problem could be a leak in the trap side of the connections, for example from the switching valve to the trap and back to the switching valve. When the traps desorb, it switches the normal column flow from bypass to going through the actual trap. It could also be a problem if the switching valve is not turning completely and causing flow restriction, I have had that happen before when a valve gets dirty.
The past is there to guide us into the future, not to dwell in.
Try putting a surrogate and spike in the exit of the trap and then close it up and go for a regular desorb. That will tell you something about flow. If the signal is good, your flow is fine. Then maybe its the trap af fault; perhaps ruined by a sample frothing over.
James_Ball wrote:
With that drop in flow it seems more related to the trap or the valve in the P&T. If there is too much restriction to flow through the trap, the EPC can't increase the pressure enough to compensate and the flow drops. The other problem could be a leak in the trap side of the connections, for example from the switching valve to the trap and back to the switching valve. When the traps desorb, it switches the normal column flow from bypass to going through the actual trap. It could also be a problem if the switching valve is not turning completely and causing flow restriction, I have had that happen before when a valve gets dirty.


The drop in flow rate is fairly typical of the system itself, it's just that it never seems to reach the set rate. I've used 2 different switching valves with the same result. I turned off the gas saver, in case it was giving false readings, and at idle it still shows ~28 mL/min. I made sure that the valves were clean by removing and cleaning/inspecting the rotors, careful to retain the correct orientation, after having run a few blanks through each. At this point i'm fairly certain i've done everything to the P&T except replace the switcher motor. Unfortunately i do not have a spare P&T or EPC to replace enmasse and check. Does anyone know if there is a way to check the EPC itself and not just symptoms around it?
In an attempt to cut the P&T out of the equation i hooked up the carrier gas line that would normally hook into the back of the P&T directly into inlet where the transfer line would go. The total flow stayed the same at ~26 mL/min. I'm fairly convinved at this point that it has something to do with the EPC, does that seem right to anyone else?
rv3 wrote:
In an attempt to cut the P&T out of the equation i hooked up the carrier gas line that would normally hook into the back of the P&T directly into inlet where the transfer line would go. The total flow stayed the same at ~26 mL/min. I'm fairly convinved at this point that it has something to do with the EPC, does that seem right to anyone else?


That would eliminate the P&T as the problem.

Make sure the inlet is configured to use Helium and not Hydrogen. If you are using Helium and it is configured as Hydrogen it will calculate the flow rate wrong.
The past is there to guide us into the future, not to dwell in.
Ok so i know it's been a while but i've got some more info, unfortunately not an answer though. I sent out the EPC for an inspection, it's not perfect but it's good enough. So i changed the weldment (?) the block with three lines coming out of the EPC. i've also changed the split flow trap. None of this has done anything. I can't find a leak anywhere but it seems like there is a crack or something in the injection housing, does that sound reasonable?
Do you have an electronic leak detector to search for a leak in the injection port?

If the gold seal is not deforming enough or there is a scratch in the bottom of the inlet it can cause a minor leak there. Another think to look for is the ferrule and column nut. If you happen to get a column nut made for a normal ferrule instead of the short ferrules that Agilent uses it could be that it isn't sealing. I have also had problems with the red predrilled septs getting leaks after not to many injections, if you are doing manual injections for trouble shooting.
The past is there to guide us into the future, not to dwell in.
Little off topic, but I feel you will or have an issue with water. With silica gel in your trap I doubt that the dry purge is effective. If you can switch to a VOCARB 3000, do it.
Years ago I switched my 524 from a 5890/5970 system to a 6890/5973. I was using the vocarb trap. Everything was ducky for a while, then everything tanked! It got so bad that we lost certification. I was convinced it was the P&T, bought a new one but the problem persisted.
Come to find out that the 5973 offers so much sensitivity, that all the water I was pumping into it was the problem.
My solution was working with Agilent engineers who eventually came up with a "water kit". We are under contract so we got the kit without any charge.
The kit includes a 6mm draw out plate, 20MX0.18 column, running a 150:1 split, and a BFB_autotune. I altered the desorb time as my auditor accepted that 3 min is about 4 min (from method), and at 1.5 min into the desorb I use the gas saver (?) to boost the flow to 150 ml/min. This limits the water going to the ms.
Check out Agilent publication # 5991-0029EN.
15 posts Page 1 of 1

Who is online

In total there are 4 users online :: 0 registered, 0 hidden and 4 guests (based on users active over the past 5 minutes)
Most users ever online was 234 on Tue Feb 06, 2018 7:33 am

Users browsing this forum: No registered users and 4 guests