Loss of sensitivity on 5975 Bad Repeller?

[MSCHemist]

Hi all I could use some advice

I started running a SIM method on my 6890-5975 that I haven't run in a month and had about 1/5 the sensitivity I usually have. I checked my reagents they are fine. Autosampler is also fine.

I started out ripping apart the source and injection port and cleaning the heck out of that:
Source gets Barkeepers friends on the 7 cleanable parts followed by sonication in water, MeOH, Acetone, MeCl2 then hexanes. INlet got a new sept liner gold seal and scrubbed with a brush and MeCl2, Acetone, and methanol. Then I did bakeout.mac to bake out the source at 300 and the quads at 200.

No change so I checked the flows. Vacuum is good the flow goes to 0 for a min when doing the spitless injection (someone said it shouldn't but mine seems configured a bit differently It has no screws holding the purge trap in just 1/8" Cu tubing and the compartment area with the compression nuts fitting in a groove).

Then I dipped the quads. It didn't really need it though the min voltage values were a bit lower than the range given for minimum voltage by Agilent when doing it. I think it went down maybe 1 unit when I made the ajustments.

Today I ran the OFN checkout. The report says my S/N is 17.9

I spoke with Agilent. They suggested try cleaning the source again and get a new repeller as it might be pitted/worn out.

Tune looks fine though my 502 has been running a bit low ~3%

[cleh]

Usuallly the ramp on the repeller will look bad if the repeller is bad. Is the autotune setting the repeller value to a similar setting as in previous good tunes? Is your EM voltage similar to previous good tunes or has it increased a lot? How old is your column? Have you clipped both ends of the column? May want to take about the length of the transfer line off the back end of the column to see if that is the problem. It's true, you're flow shouldn't go to 0 when doing a splitless injection.

[MSCHemist]

It hasn't changed significantly

On Sept 13th The repeller was 28.62 and EMvolts were 1812
This morning the repeller was 34.81 and EMvolts were 2000 even

I've tried 2 columns no difference.

[MSCHemist]

The only other change I made was I added a belleville spring washer and 2 spring washers to the source. Orginally our 5975b had just had the one washer before the nut and no washers on the allen bolts that hold the heater/sensor to the source body.

The manual shows they were not listed on our instrument. All the 5975 EI source diagrams online show there should be the 3 additional washers. We've also frequently broken repeller insulators no matter how careful we were not to overtighten the nut so I ordered them from Agilent and added them to put a stop to that. That should not make any difference though?

[cleh]

When you cleaned the injector, did you clean the split line and connector on the weldment? I assume the split and purge flows are OK because you said you checked the flows. One other thing to check is the method to see if the EM voltage is set the way you had it originally (relative or absolute and if relative the value is the same).

[MSCHemist]

When you cleaned the injector, did you clean the split line and connector on the weldment? I assume the split and purge flows are OK because you said you checked the flows. One other thing to check is the method to see if the EM voltage is set the way you had it originally (relative or absolute and if relative the value is the same).

I didn't clean the split line but all the flows seem to be ok. I have the flip top inlet system.

[labstephan]

Hello,

you could try to shorten your column 1m ore more.
I observed on different analytes we are searching for (residual analysis) some problems after injecting 50-100 samples. BTW. Our samples are very dirty (urine and organe extracts).
After trimming the column and injecting some solvent blanks we got back our sensitivity.
Now we are using columns with integrated precolumn. So you can cut of precolumn without affecting your separation. Of course, you have to adjust the column length in your software.

Hope this helps.

Stephan

[cleh]

Sounds like you've covered everything. I don't think it's the repeller. I would run your standard using full scan to check your retention times. Maybe they've shifted a bit and you need to update the retention time windows because you're not aquiring the whole peak when you do SIM. You still might want to check the split valve. I've found a lot of junk in them even though the flows are OK. At some point you may want to figure out why you go to 0 when doing split injections, but if that's always happened before you experienced the sensitivity loss it probably has nothing to do with this issue. Please keep us updated.

[MSCHemist]

I mentioned it to Agilent and they didn't seem to think it was an issue. My GC looks a bit different it doesn't have that thing with a screw that holds the purge trap to the instrument. Perhaps it has something to do with the ODP, flip top system, or replacement EPC.

The only other thing is the EPC never displays 0. Whenever I turn it off it displays 152 psi. WHen it turns on it goes to where it should and delivers the proper flows.

The retention times are spot on. After I started this I removed a turn of column from the head and the retention time went down .5min but it is stable and the resolution is good. The resolution on the method has little tolerance. (3-MCPD HFBI derivitized DB-5ml .25 .25 30m).

[cleh]

Try tuning on the other filament and running on it to see if your sensitivity returns.

[MSCHemist]

Does this repeller ramp look ok I don't remember it having a shoulder like that.

I cleaned the heck out of the source it on Friday and did another bakeout of the source. I will try filament 1 today.

[MSCHemist]

Back in business.

I switched to filament 1 and the sensitivity was actually better than last month.

[cleh]

Nice!!!

[tlahren]

I switched to filament 1 and the sensitivity was actually better than last month.

I am assuming the problem was occurring while using Filament 2?

Just a little note, I have heard "reports" of the performance of the Filament 2 position being poor and irreproducible compared to the Filament 1 position. The explanation given to me was that as the Filament 2 heats up it "Sags" down and away from the ion volume region whereas Filament 1 would tend to Sag down closer to the ion volume region. I'm not sure if this is accurate (as we are currently using Filament 2 on a 5975C with no noticeable differences) but I still usually try to run on filament 1 only.

[MSCHemist]

Hmm We tend to favor position 2 because if we use 1 crap from the top falls on the bottom filament whereas if we use position 2 crap from it falls to the bottom of the source.

In any case the bottom filament looked a bit charred.